close

Вход

Забыли?

вход по аккаунту

?

Microwave-enhanced Processing of Transition Metal oxides, Silicates, and Aluminosilicates and Their Catalytic Studies

код для вставкиСкачать
Microwave-enhanced Processing of Transition Metal
oxides, Silicates, and Aluminosilicates and Their
Catalytic Studies
Naftali Nandera Opembe, Ph.D.
University of Connecticut, 2012
This research work delves into various aspects of heterogeneous catalysis, notably, the
synthesis and catalytic applications of inorganic catalytic materials using emerging
technologies. Selected transition metal oxide and aluminosilicate materials have been
synthesized under continuous microwave heating as well as microwave hydrothermal
processes and applied to selected catalytic reactions.
The first part presents the development of a continuous microwave technique for the
synthesis of transition metal oxides. Two classes of manganese oxides have been
synthesized through this technique; cryptomelane-type manganese oxide octahedral
molecular sieves (K-OMS-2) and gamma manganese oxide. Microwave power, flow rate
(thus residence time); temperature, co-solvent volume, and nature of reactor are some of
the parameters that have been optimized for this system. Through this system, materials
with small particle sizes have been synthesized and tested for catalytic activity.
The second and third parts of this thesis present catalyzed oxidation reactions in the
liquid phase using K-OMS-2. K-OMS-2 showed high activity for the oxidation of 9HFluorene to 9-Fluorenone. This reaction is important since 9-Fluorenone is a compound
that is showing promise in its use as a component of organic light emitting diodes, among
other uses. The kinetics as well as a proposed mechanism of 9H-Fluorene oxidation has
been reported. It has been found that 9H-Fluorene oxidizes through a first order kinetics
where the breaking of the C-H bond is the rate determining step. It has also been found
that this oxidation reaction is dependent on the nature of the K-OMS-2 used with KOMS-2 synthesized under acidic buffer conditions giving the best performance. Another
important industrial reaction that K-OMS-2 was exposed to was the oxidation of 2,3,6trimethyl phenol (TMP). TMP oxidizes catalyzed by K-OMS-2 to 2,3,6-trimethyl-4hydroquinone (TMQ) is a short time period of 30 mins. This reaction requires the
presence of tert-butyl hydroperoxide (TBHP) as the oxidant in a 2:1 ratio of TBHP to
substrate.
K-OMS-2 has also been used in the gas phase oxidation of benzyl alcohol. This had
hitherto this not been done and so provides an impetus to use K-OMS-2 for gas phase
catalyzed reactions. Benzyl alcohol flow rates, oxidant flow rate, temperature, and
catalyst loading are the variables that have been optimized for >90% conversion of
benzyl alcohol to benzaldehyde.
Lastly, microwave irradiation has been used to synthesize zeolite Y in a short time period
of 1 h under hydrothermal conditions. Further, the synthesized zeolite was exchanged for
nickel and also impregnated with 10% nickel and reduced in a hydrogen atmosphere for
use as a catalyst for steam reforming of glycerol. Steam reforming of glycerol gave good
yields of hydrogen gas.
Microwave-enhanced Processing of Transition Metal
oxides, Silicates, and Aluminosilicates and Their
Catalytic Studies
Naftali Nandera Opembe
B.S., University of Nairobi, 2004
A Dissertation
Submitted in Partial Fulfillment of the
Requirements for the Degree of
Doctor of Philosophy
At
The University of Connecticut
2012
UMI Number: 3529370
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
OiSi«Wior» Ftattlisttlfl
UMI 3529370
Published by ProQuest LLC 2012. Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code.
ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346
APPROVAL PAGE
Doctor of Philosophy Dissertation
Microwave-enhanced Processing of Transition Metal
•?
oxides, Silicates, and Aluminosilicates and Their
Catalytic Studies
Presented by
Naftali Nandera Opembe, B.S.
Major Advisor
Steven L. Suib
Associate Advisor
Raymond Joesten
Associate Advisor
Amy Howell
Associate Advisor
Xudong Yao
University of Connecticut
2012
ii
Dedicated to my son
Eli A. Opembe
iii
ACKNOWLEDGEMENTS
I would like to thank my major advisor Dr. Steven L. Suib for all the help and advice he
accorded to me as his graduate student. I would also like to personally thank him for his
understanding and friendship. I have learnt a lot from him as a mentor and advisor.
Secondly, many thanks to my son Eli for the many happy moments and not forgetting
Eli's mother, Verona. My son is the inspiration behind my work.
Thirdly, many thanks to the following for a fruitful and fulfilling relationship; Cecil,
Nashaat, Eric, Saminda, Dr Jin, Dr Son, Dr Nyutu, and anyone whom I may not have
personally acknowledged here and with whom we had a cordial relationship.
Many thanks Dr Joesten, Dr Galasso, and my advisory committee.
iv
TABLE OF CONTENTS
CHAPTER 1.
INTRODUCTION
1
1.1
Overview
1
1.2
Microwave continuous synthesis of inorganic materials
5
1.3
Liquid phase oxidation of organic molecules using manganese oxide
octahedral molecular sieves (OMS-2) catalysts
1.4
Selective partial oxidation of benzyl alcohol using manganese oxide
octahedral molecular sieves (OMS-2) catalysts in the gas phase
1.5
8
10
Steam reforming of glycerol for hydrogen generation using nickel-zeolite
Y catalysts
11
1.6
References
12
CHAPTER 2.
MICROWAVE CONTINUOUS SYNTHESIS OF INORGANIC
MATERIALS
14
2.1
Introduction
14
2.2
Principles of MW dielectric heating
14
2.3
Types of apparatus used in microwave continuous synthesis of inorganic
materials
19
2.4
Microwave techniques used in the synthesis of inorganic materials
v
24
2.5
Microwave-assisted continuous synthesis of manganese oxide materials .
28
2.6
Experimental
32
2.6.1
Materials
32
2.6.2
Microwave set-up
32
2.6.3
Microwave Treatments
33
2.6.4
Continuous synthesis of K-OMS-2
35
2.6.5
Continuous synthesis of y-Mn02
37
2.7
Characterization methods
38
2.7.1
Powder X-ray Diffraction studies
38
2.7.2
Crystallite Particle Size
38
2.7.3
Transmission Electron Microscope (TEM)
38
2.7.4
Scanning Electron Microscope
38
2.7.5
Thermal Stability
39
2.7.6
Potentiometric Titrations
39
2.7.7
Surface Area and Pore Size Distribution
39
2.7.8
Surface and Functional Group Studies
39
2.8
Results
39
2.8.1
Continuous Formation of K-OMS-2 Nanomaterials
39
2.8.2
Crystallite Size
45
vi
2.8.3
Morphology
45
2.8.4
Functional groups
49
2.8.5
Thermal Stability
49
2.8.6
Potentiometric Titrations
52
2.8.7
Surface Area and Pore Size Distribution
52
2.9
Discussion
54
2.9.1
Continuous Crystallization of K-OMS-2 Nanomaterials
54
2.9.2
Morphology and Crystal Structure
59
2.9.3
Mixed valency, Thermal stability, and Textural properties
59
2.9.4
Continuous synthesis of y-MnCh
60
2.10
Catalytic application of synthesized materials
69
2.10.1
Catalytic oxidation of 2,3,6-Trimethyl phenol
69
2.10.2
Vapor phase oxidation of Toluene
69
2.11
Conclusions
73
2.12
References
76
CHAPTER 3.
KINETICS AND MECHANISM OF 9H-FLUORENE OXIDATION
CATALYZED BY MANGANESE OXIDE OCTAHEDRAL MOLECULAR SIEVES 79
3.1
Introduction
79
3.2
Experimental Section
82
3.2.1
Reagents
82
vii
3.2.2
Catalyst Synthesis
82
3.3
Characterization Methods
83
3.3.1
X-ray diffraction
83
3.3.2
Average oxidation state (AOS)
84
3.3.3
Surface area Measurements
84
3.3.4
Morphology
84
3.4
Catalytic Oxidation Studies
85
3.4.1
Oxidation of 9H-Fluorene
85
3.4.2
Solvent study
85
3.4.3
Kinetics of 9H-Fluorene oxidation
85
3.4.4
Kinetics of 9H-Fluorene oxidation in presence of carbon tetrachloride. 87
3.4.5
Kinetic Isotope Effect (K.l.E)
87
3.4.6
Oxidation of 9H-FIuorene under inert conditions
87
3.4.7
Recylability of the catalyst
88
3.4.8
9H-Fluorene oxidation in absence of catalyst
88
3.4.9
Product identification and analysis
88
3.5
Results
88
3.6
Discussion
94
3.7
Conclusions
99
vni
3.8
References
100
CHAPTER 4.
EFFICIENT OXIDATION OF 2, 3, 6-TRIMETHYL PHENOL USING
NON-EXCHANGED AND H + EXCHANGED MANGANESE OXIDE OCTAHEDRAL
MOLECULAR SIEVES (K-OMS-2 AND H-K-OMS-2) AS CATALYSTS
102
4.1
Introduction
102
4.2
Aim of study
105
4.3
Experimental
105
4.3.1
Reagents
105
4.3.2
Catalyst syntheses
105
4.3.3
Catalytic reactions
106
4.3.4
Catalyst leaching test
106
4.4
Characterization of the catalysts
106
4.4.1
X-ray diffraction
107
4.4.2
Scanning Electron Microscope
107
4.4.3
BET Surface Area and Pore Size Distribution
107
4.5
Characterization Results
108
4.5.1
Catalyst Characterization
108
4.6
Catalytic Results
111
4.6.1
Effects of the Type of Catalyst and Oxidant on Oxidation of TMP
111
ix
4.6.2
Temperature and Reaction Time Effect
111
4.6.3
Leaching test results
111
4.7
Discussion
115
4.7.1
Catalyst Characterization
115
4.7.2
Effect of Catalyst Nature on Oxidation of TMP
115
4.7.3
Role of Oxidant on Oxidation of TMP
116
4.7.4
Temperature and Reaction Time Effect on Oxidation of TMP
117
4.8
Conclusion
118
4.9
References
119
CHAPTER 5.
SELECTIVE PARTIAL OXIDATION OF BENZYL ALCOHOL ON
MANGANESE OXIDE OCTAHEDRAL MOLECULAR SIEVES (OMS-2)
CATALYSTS IN THE GAS PHASE
121
5.1
Introduction
121
5.2
Experimental
123
5.2.1
Materials
123
5.2.2
Catalyst synthesis
123
5.2.3
Ion-exchange
124
5.2.4
Gas phase oxidation of benzyl alcohol
124
5.3
Characterization
126
5.3.1
Powder X-ray Diffraction
126
x
5.3.2
Morphology
126
5.3.3
Nitrogen sorption experiments
126
5.3.4
Oxygen evolution
127
5.3.5
Functional groups and acidity
127
5.4
Results
127
5.4.1
Identity of the synthesized materials
127
5.4.2
Morphology, Surface area and Pore size distribution
128
5.4.3
Oxygen evolution
128
5.4.4
Functional groups and acidity measurements
129
5.5
Catalytic results
134
5.6
Discussion
135
5.6.1
Catalyst characterization
135
5.6.2
Catalytic results
137
5.7
Conclusion
139
5.8
References
140
CHAPTER 6.
STEAM REFORMING GLYCEROL FOR HYDROGEN
GENERATION USING NI-ZEOLITE Y
142
6.1
Introduction
142
6.2
Experimental
146
6.2.1
Materials
146
xi
6.2.2
Synthesis of Zeolite Y
146
6.2.3
Ni-exchanged Zeolite Y and Ni-loaded Zeolite Y
146
6.2.4
Glycerol steam reforming
147
6.3
Characterization
149
6.3.1
Powder X-ray diffration
149
6.3.2
Elemental analysis
149
6.4
Results
150
6.4.1
Catalyst characterization
150
6.4.2
Catalytic results
158
6.5
Discussion
162
6.5.1
Synthesis and characterization of Zeolite Y
162
6.5.2
Glycerol steam reforming
163
6.6
Conclusion
166
6.7
References
167
xii
LIST OF FIGURES
Figure 2-1. Microwave electromagnetic field interaction and heating mechanisms, (a)
dipolar polar polarization and (b) ionic conduction mechanisms
17
Figure 2-2. Pictorial representation of hydrothermal microwave FLOWSYNTH that is a
newer version of the old model MLS ETHOS CFR microwave (obtained from
www.milestonesrl.com on April 3, 2012)
22
Figure 2-3. (a) Nozzle microwave and (b) in-situ nozzle microwave apparatus
(Reproduced with permission from: Espinal, L., Malinger, K. A., Espinal, A. E., Gaffney,
A. M., Suib. S. L. (2007) Adv. Fund. Mater. 17, 2572. Copyright: W1LEY-VCH Verlag
GmbH & Co. KGaA, Weinheim)
23
Figure 2-4. (a) Schematic diagram for the microwave plasma jet reactor showing the
components, and (b) the actual plasma generated. (Reproduced with permission from:
Kumar, V., Kim, J. H., Pendyala, C., Chernomordik, B., Sunkara M. K. (2008) J. Phys.
Chem. C, 112, 17750. Copyright: America Chemical Society)
23
Figure 2-5. Crystal structure representation of K-OMS-2 with potassium (K + ) counter
cations shown as maroon spheres while Mn06 octahedra are shown as blue octahedral.
The tetragonal unit cell (a=b=9.82
A. c=2.85 A) is also shown with blue lines. Crystal
structures drawn using Crystal Maker™
31
Figure 2-6. Crystal structural diagrams of (a) pyrolusite (P-MnCh); (b) ramsdellite and (c)
De Wolff s model fory -MnC >2 showing intergrowth between pyrolusite and ramsdellite
domains. Crystal structures drawn using Crystal Maker™
Xlll
31
Figure 2-7. Schematic drawing of the continuous flow synthesis set-up showing a coiled
Teflon reactor inside the microwave cavity
34
Figure 2-8. A photo of the actual microwave set-up used in the continuous flow synthesis.
34
Figure 2-9. XRD patterns of K-OMS-2 showing the effect of reactant flow rates at
constant microwave power (300 W) and constant volume of DMSO (10% v/v). Reactant
flow rate was varied from 40 mL/min down to 5 mL/min
42
Figure 2-10. XRD patterns of K-OMS-2 prepared using different percentages of the cosolvent (DMSO) at two flow rates - 10 mL/min and 20 mL/min. The phases were
matched to the Q-phase of cryptomelane (KMn8016; JCPDS 29-1020)
43
Figure 2-11. X-ray diffraction patterns of OMS-2 materials synthesized batch-wise with
temperature measurements determined as follows; a, b, c, and d as 90 °C, (e) as 85 °C and
(Oas 100°C
44
Figure 2-12. X-ray diffraction patterns of OMS-2 materials synthesized at different power
levels (150 W- 600 W) with 10 % DMSO and a flow rate of 10 mL/min
44
Figure 2-13. FESEM of K-OMS-2 materials MC-0-10 mL/min (0% DMSO), (b) MC-1040 mL/min (10% DMSO), (c) MC-25-20 mL/min (25%DMSO) and (d) MC-50-20
mL/min (50% DMSO). The scale is 500 nm
47
Figure 2-14. Low resolution (a) and high-resolution (b) TEM images of K-OMS-2
materials prepared by continuous flow microwave technique at 20 mL/min in 25%
DMSO (MC-25-20 mL/min)
47
xiv
Figure 2-15. FT1R spectra of OMS-2 materials: (a) MC-10-40 mL/min (b) MC-10-20
mL/min and (c) MC-10-10 mL/min. Samples were synthesized at a microwave power of
300 W.
50
Figure 2-16. TGA profiles of K-OMS-2 materials prepared continuously at different
flow rates (10 mL/min and 20 mL/min) with different volumes of DMSO (10% and
50%).
50
Figure 2-17. DSC profiles of K-OMS-2 materials prepared continuously at different flow
rates (10 mL/min and 20 mL/min) with different volumes of DMSO (10% and 50%)
51
Figure 2-18. (a) Nitrogen sorption isotherm of MC-5-10 mL/min K-OMS-2 sample and
(b) The Barrett-Joyner-Halenda (BJH) pore size distribution
53
Figure 2-19. XRD patterns of materials continuously synthesized at Mn 2+ /Mn 7+ = 2.8,
with 10 mL nitric acid, (a) without and (b) with DMSO
64
Figure 2-20. XRD patterns of y-Mn02 continuously synthesized at different flows: (a) 5
mL/min, (b) 10 mL/min and (c) 20 mL/min, constant power of 300 W and 25% DMSO,
compared to materials made by (d) oil bath treatment of the same recipe for 24 h at 85 °C
with impurity phase shown as (*)
64
Figure 2-21. XRD patterns of y-Mn02 continuously synthesized with different volumes
of DMSO: (a) 0%, (b) 1%, and (c) 25%, at constant power of 300 W and flow rate of 10
mL/min compared to materials made by oil bath treatment of the same recipe for 24 h at
85 °C (d) with impurity peaks shown as (*)
65
xv
Figure 2-22. SEM micrographs of y-Mn0 2 continuously synthesized at (a) 20 mL/min,
25% DMSO and (b) 10 mL/min, 25% DMSO and(c) 10 mL/min, 1% DMSO
65
Figure 2-23. XRD patterns of y-MnCh synthesized with Mn 2+ /Mn 7+ = 2.3 left at room
temperature for (a) 2 and (b) 24 hrs
66
Figure 2-24. SEM micrographs of y-MnCh room temperature synthesized for: (a) 2 h and
(c) and 24 h and their higher magnifications (b) and (d) respectively
66
Figure 2-25. TGA patterns of y-MnC>2 continuously synthesized at different flow rates at
constant power of 300 W and compared to materials made by oil bath treatment of the
same recipe for 24 h at 85 °C
67
Figure 2-26. TGA patterns of y-MnO^ continuously synthesized with different volumes of
DMSO at constant power of 300 W and compared to materials made by oil bath
treatment of the same recipe for 24 h at 85 °C
67
Figure 2-27. Nitrogen sorption material synthesized with 1% DMSO at 10 mL/min. The
surface area of the material was 112 m 2 /g
68
Figure 3-1. XRD patterns of K-OMS-2 materials synthesized under different reaction
conditions and methods. OMS-2B is synthesized under acidic buffer, OMS-2-Reg is
regenerated OMS-2, OMS-2R is reflux method, OMS-2S is solvent free, OMS-2H is
hydrothermal, while OMS-2-MW is under microwave heating
88
Figure 3-2. SEM micrographs of OMS-2B materials
88
Figure 3-3. Kinetics of 9H-fluorene oxidation with octahedral molecular sieves under
different conditions: in isooctane (•), in toluene at 110 °C (x), and in toluene at 80 °C
( A ) ; oxidation of fluorene-dlO in toluene at 110 °C (+); oxidation of 9H-fluorene in
toluene in the presence of 0.2 equiv CCU (x) and in toluene under nitrogen atmosphere
(*).
89
Figure 3-4. Comparison of the selectivities to 9-Fluorenone in toluene (•) and in
isooctane (•)
89
Figure 3-5. Kinetic isotope effect (KIE) in the oxidation of 9H-fluorene (kH) and [D10]fluorene (kD) based on data for the oxidation of 9H-fluorene and [D10]- fluorene in
90
Figure 3-6. Proposed mechanism for the OMS-2-catalyzed oxidation of 9Hfluorene
91
Figure 4-1. Scheme showing steps involved in the current industrial production of
vitamin E viz: (i) oxidation of TMP to TMQ and its subsequent reduction to TMHQ, and
(ii) condensation of TMHQ with isophytol to form vitamin E
104
Figure 4-2. X-ray diffraction patterns of as synthesized K-OMS-2 (bottom), H-K-OMS-2
(middle), and regenerated K-OMS-2 (top). All materials were indexed to the Q-phase of
cryptomelane (JCPDS card no. 29-1020)
109
Figure 4-3. FE-SEM micrograph images of the as-synthesized K-OMS-2 catalyst
109
Figure 4-4. Nitrogen sorption isotherms of the as-synthesized K-OMS-2 materials.
Horvath-Kawazoe pore size distribution curve (inset) Corresponding BET surface area
was 98 m 2 /g
110
xvii
Figure 4-6. Oxidation of TMP (1 mmol) at 40 °C with 2 mmol TBHP in 10 mL
acetonitrile and 50 mgof K-OMS-2
113
Figure 4-7. Oxidation of TMP (1 mmol) at 65 °C with 2 mmol TBHP in 10 mL
acetonitrile and 50 mg of K-OMS-2
114
Figure 5-1. Schematic diagram of the vapor phase oxidation of benzyl alcohol using
OMS-2 materials. Benzyl alcohol is pumped using a syringe pump through T1 is the
maintained at 220 °C where it meets the oxidant/carrier gasses for onward transport to the
furnace maintained at a temperature T2. Cold trap contents are analyzed using GC-MS
while gaseous products are analyzed using an online GC as shown in the scheme
125
Figure 5-2. X-ray diffraction results, (a) K-OMS-2, (b) H-K-OMS-2, (c) H-K-OMS-2
post-benzyl alcohol oxidation at N2/O2 (10/30) seem, (d) H-K-OMS-2 post-benzyl
alcohol oxidation in pure N2(40 seem) reaction, and (e) H-K-OMS-2 after treatment
under pure N2 (40 seem) at 220 °C without reaction
130
Figure 5-3. Scanning electron microscope images of the as-synthesized K-OMS-2
materials
130
Figure 5-4. Sorption isotherms of K-OMS-2 materials with the corresponding BET
surface area of 77m 2 /g
131
Figure 5-5. Oxygen evolution by temperature programmed desorption measured for KOMS-2
131
Figure 5-6. Oxygen evolution by temperature programmed desorption measured for H-KOMS-2
132
Figure 5-7. Functional groups study using FT1R
132
xviii
Figure 5-8. Acidity measurements by ammonia IR
133
Figure 6-1. Crystal structure representation of faujasites (FAU framework) viewed in the
(111) direction. The vertices of the structure represent sodalite units similar to carbon
atoms in the diamond structure. The tetrahedra represent either
S1O4
or AKV units.
Patterns generated using Crystal Maker™
145
Figure 6-2. Schematic diagram for the glycerol steam reforming reaction. Glycerol/water
mix is evaporated at T1 (set at 220 °C) while T2; the furnace temperature is varied from
500 to 650 °C
148
Figure 6-3. XRD patterns of microwave hydrothermally synthesized NaY materials at
different temperatures after 16 h ageing where (a) represents materials synthesized at 90
0,C, (b) 120 °C, (c) 150 °C, and (d) 180 °C
151
Figure 6-4. XRD patterns of microwave hydrothermally synthesized NaY materials at 90
O,C after 16 h ageing, indexed using JCPDS file no. 39-1380
151
Figure 6-5. XRD patterns of microwave hydrothermally synthesized NaY materials at
180 °C after 16 h ageing, indexed using JCPDS file no. 39-219
152
Figure 6-6. XRD patterns of (a) 90 °C synthesized NaY, (b) a triple ion-exchange with
Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are for
152
nickel metal
Figure 6-7. Enlarged view of (a) 90 °C synthesized NaY, (b) a triple ion-exchange with
Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are for
nickel metal
153
xix
Figure 6-8. XRD patterns of (a) 120 °C synthesized NaY, (b) after impregnation with
10% Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are
154
for nickel metal
Figure 6-10. The EDS results of as-synthesized NaY (aged 16 h and synthesized at 90
°C)
156
Figure 6-11. The EDS results of triply ion exchanged NaY (aged for 16 h and synthesized
at 90 °C)
156
Figure 6-12. Transmission electron micrograph (TEM) images of reduced Ni
impregnated NaY samples prepared at 120 °C
157
Figure 6-13. Matching XRD patterns of (a) NaY and (b)NaNi 2+ -Y against (c) a simulated
pattern obtained by generating a diffraction pattern from atomic positions and site
occupancies of hydrated NiY from ref. 21. Patterns generated using Crystal Maker IM
157
Figure 6-14. Hydrogen yield in the glycerol steam reforming at 550 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY
159
Figure 6-15. Composition of dry product stream in the glycerol steam reforming at 550
°C with glycerol/water liquid flow rate of 0.02 mL/min, N 2 flow rate of 40 seem, 0.5 g
10% Ni-impregnated NaY
159
Figure 6-16. Hydrogen yield in the glycerol steam reforming at 600 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY
160
xx
Figure 6-17 Composition of dry product stream in the glycerol steam reforming at 600 °C
with glycerol/water liquid flow rate of 0.02 mL/min. N2 flow rate of 40 seem, 0.5 g 10%
Ni-impregnated NaY
160
Figure 6-18. Hydrogen yield in the glycerol steam reforming at 650 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY
161
Figure 6-19. Composition of dry product stream in the glycerol steam reforming at 650
°C with glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g
10% Ni-impregnated NaY
161
xxi
LIST OF TABLES
Table 2-1. Loss tangents (tan 8) of selected solvents (2.45 GHz, 20 °C)
18
Table 2-2. Microwave penetration depth (2.45 GHz) in selected materials
48
Table 2-3. Crystallite sizes of K-OMS-2 materials prepared under MACS
48
Table 2-5. Oxidation of TMP to TMQ using continuously synthesized K-OMS-2
nanofibers.
71
Table 2-6. Results of the vapor phase oxidation of Toluene using gamma manganese
oxide synthesized via a microwave continuous process
72
Table 3-1. Oxidation of 9H-Fluorene under different conditions in the presence of air' d| ..
92
Table 3-2. Oxidation of 9H-fluorene catalyzed by various OMS-2 catalysts in the
presence of air.[a]
93
Table 4-1. Effects of selected parameters on the oxidation of TMP
112
Table 5-1. Results of the gas phase oxidation of benzyl alcohol in the gas phase using KOMS-2 materials
134
Table 6-1. Details of X-ray diffraction Pattern of material synthesized at 90 °C
153
Table 6-2. Peak shift (d-spacing) on exchanging sodium for nickel (Na-VNi 2+ ) and also
on reduction of nickel-exchanged NaY (Ni 2+ —•Ni). Basis for comparison is the assynthesized NaY in both cases
154
xxii
Table 6-3. Slight peak shift on impregnating Ni 2+ onto NaY ([Ni 2+ ]NaY] synthesized at
120 °C and also on reduction to nickel metal ([Ni]NaY. The basis of comparison is the
as-synthesized NaY
I
xxiii
CHAPTER 1.
1.1
INTRODUCTION
Overview
The work presented in this thesis broadly centers on the synthesis, characterizations, and
catalytic applications of various transition metal oxide and aluminosilicate materials
using emerging technologies. The uniqueness of the processes reported herein involves
the methods applied in the development of said synthetic processes as well as processes
for catalytic applications of the final materials. Thus, the work delves into the various
aspects of inorganic material synthesis and their catalytic applications using emerging
technologies as well as use of these materials for various applications, which include
renewable energy applications, mainly the production of hydrogen from renewable
resources.
Catalysts have been touted as the workhorses of the chemical industry'. They are
substances that increase the overall rates of chemical transformations without being
consumed in the process. Chemical catalysis affects our lives in many different ways.
Through catalysis, an understanding of the processes involved in changing the rates of
chemical reactions and controlling the overall purities and yields can be established. Thus
catalysis lies at the heart of chemistry as a science. Catalysis has a direct impact on the
quality of life. This is for example through; reduced undesirable emissions from
automobiles 2 , increased production of food materials, improved health through improved
pharmaceutical processes, among others.
Catalysis is also essential to a thriving economy. For example, the petroleum, chemical,
and pharmaceutical industries, which are a bigger contributor to the global economy, rely
on catalysts in one-way or another to achieve their goals. These industries are vital since
1
they produce products to meet the fuel and energy, synthetic chemicals, and drugs to be
used in the healthcare industry of a country.
Presently, there are various challenges facing humanity. These challenges include
creating alternative fuels, reducing harmful by-products in manufacturing, cleaning up
the environment and preventing future pollution, dealing with the causes of global
warming, and protecting living organisms from the release of toxic substances and
infectious agents, creating safe pharmaceutical products to meet emerging healthcare
challenges for instance cancer, among others. To meet these challenges, catalysts are
needed to be more effective thus demanding a revolution in the way they are designed
and applied. This revolution can become a reality through the application of new
approaches to synthesizing and characterizing catalytic materials.
One of the century's key developments is the area of nanotechnology in general and
nanomaterials in particular. These are materials with particle sizes confined in the
nanometer range. These developments have offered extraordinary opportunities in
various technological fields such as electronics, energy management, structural materials,
functional surfaces, construction, and information technology, and in the pharmaceutical
and the healthcare sectors. The reduction of a material's size to the nanometer range has
been characterized by different physical and chemical properties enabling the material to
be used in ways that were not possible before.
The use of emerging technologies has led to the possibilities of confining the synthesized
materials to the nanometer range by enabling their rapid synthesis. One such technology
is microwave dielectric heating 3 . Microwave-assisted synthesis as a tool for chemical
synthesis has been embraced in academia since this method is a fast synthesis technique
2
that offers unique capabilities. High product yields and purities are some of the unique
features found in materials synthesized using microwave heating. Additionally, uniform
structural, textural, and novel physical properties have been obtained. Microwaves as a
synthesis technique has the potential of contributing immensely to all areas of synthetic
chemistry
including Inorganic Chemistry. The synthesis of nanomaterials and
nanostructures is bound to benefit more from the uniform and rapid heating provided by
microwaves.
Microwave chemistry relies on the unique ability of the reaction mixture to efficiently
absorb microwave energy. This ability is derived from the phenomenon of 'microwave
dielectic heating' 3 . What happens to a material when exposed to microwaves is a
consequence of the material's intrinsic properties in relation to how the material interacts
with microwaves (MW). Based on this interaction, materials can be classified into three
classes. Generally, materials will either be MW-reflectors, transmitters, or absorbers.
MW-reflectors are useful as core components of MW waveguides since they have the
ability to reflect and therefore guide MW to a specific region. MW-transmitters on the
other hand allow passage of microwaves without interference and as such are useful in
making MW cookware or reaction vessels e.g. Teflon is a good MW transmitter. MW
transmitters are also known as MW transparent materials. MW-absorbers are useful in
chemistry. These materials can absorb microwaves causing them to be heated.
The work presented in chapters 2 and 6 of this thesis utilizes microwave dielectric
heating as a synthesis tool and the catalytic applications of the synthesized materials.
Chapters 3, 4, and 5 involve catalytic applications of materials synthesized using
conventional means.
3
Microwave heating has been applied in different ways in the synthesis of different
materials. Microwave radiation has been used in the synthesis of cryptomelane-type
manganese oxide (a-MnCb) octahedral molecular sieves (OMS-2) and Nsutite-type
manganese oxide (y-Mn02) on a continuous basis. Further to this, Zeolite Y (NaY)
molecular sieve has been crystallized in a microwave hydrothermal process on a batch
mode with or without ageing at short periods of time with unique physical and chemical
characteristics. The synthesized OMS-2 materials have been extensively studied in
various catalytic applications alongside their counterparts synthesized conventionally.
Also studied and presented here is the use of nickel (Ni) exchanged and nickel loaded
Zeolite Y catalysts for the steam reforming of glycerol reaction to produce hydrogen gas
(Hz).
1.2
Microwave continuous synthesis of inorganic materials
The continuous synthesis of materials is attractive to the chemical industry. This is
because continuous methods come with advantages otherwise not present in batch
syntheses. Continuous methods reduce synthesis to a seamless process involving the
feeding of reagents to a feeding vessel and removing of products from the receiving
vessel. Continuous modes also in most cases lead to a high throughput process that is
scalable. With continuous synthesis, kilogram-scale synthesis is possible once all
parameters are optimized. This is hardly the case in batch synthesis where the size of the
reaction vessel is a limiting factor.
When applied to continuous synthesis, microwaves impart their unique features to enable
optimal performance. Due to their nature of interaction with reactant systems,
microwaves speed up certain chemical reactions. This is due to the fast heating involved
4
(superheating). Superheating of a reaction increases its overall kinetics by providing
means to overcome the energy barrier (activation energy).
The use of microwaves in the continuous synthesis of inorganic materials is an area of
research still in its infancy. Going by the number of publications on this subject in the last
10 to 20 years,
410
continuous synthesis is an area that is showing a growing interest as
well as offers opportunities for continued research.
A microwave continuous technique for the synthesis of cryptomelane-type manganese
oxide (a-MnCh) octahedral molecular sieves (OMS-2)
10
and Nsutite-type manganese
oxide (y-Mn02) have been developed and optimized.
The synthesis of K-OMS-2 materials was prior-to-this realized through different routes.""
13, 14 15
These studies involved batch synthesis and also involved systematic studies to
among others things confine the particle size to the nanometer range. Villegas et al.
11
successfully prepared nano-sized crystalline K-OMS-2 fibers with crystallite sizes as low
as 6 nm by reducing KMn04 with H2O2 in an acidic buffer medium in about 15 h, under
conventional reflux. Recently, Nyutu et al.
16
reported a method for synthesizing K-OMS-
2 in mixed aqueous and non-aqueous solvents by use of microwave reflux, obtaining KOMS-2 nanomaterials with crystallite sizes as low as 4 nm. This method required about
10 minutes to start crystallizing the K-OMS-2 phase and about 90 minutes to obtain a
fully crystallized Q phase. Crisostomo
17
has reported the development of a continuous-
flow method using microwaves for synthesizing s-Mn02, Ce02, CoOOH, and FeOOH
using microwave irradiation.
Nyutu's microwave method that involved the use of organic co-solvent systems provided
the impetus for the development of the continuous process for the synthesis of K-OMS-2.
5
Relied on their own, the various K-OMS-2 syntheses recipes in the absence of organic
co-solvent are unable to lead to the continuous crystallization of the Q phase under
microwave conditions. At best they lead to an amorphous product and the overall process
is characterized by clogging of the tube reactor.
To successfully develop this system, Nyutu's recipe of involving organic co-solvent
systems was used. DMSO was one such solvent. DMSO shows excellent microwave
absorption properties and thus can be used to provide superheating conditions. However,
in this case, microwave absorption cannot account for the crystallization per se as other
equally and even better absorbers e.g. ethylene glycol led to crystallization of a
manganite phase (MnOOH) under batch conditions. Thus a suitable organic compound is
needed. Suitability in this case is chosen on compound-by-compound based on actual
results. This is because the molecular level interactions of said organic systems with
nucleating species are still not well understood.
This research involved a systematic study to find the best experimental conditions for the
synthesis of both K-OMS-2 and y-Mn02. This systematic approach involved studying
microwave power levels, reactant flow rates, nature of tube reactor, and percentage
volume of co-solvent used. Flow rates impacted on residence time in the microwave
oven. Faster flow rates e.g. 40 mL/min led to residence time of less than one minute. This
condition was not suitable for the crystallization of K-OMS-2. When the flow rate was
reduced to 20 mL/min, 10 mL/min and even further down to 5 mL/min, crystallization
was affected. With no DMSO no crystallization could be affected at all flow rates. Lower
volumes of DSMO e.g. 5 % DMSO required longer exposure times (slow flow rates) for
the Cryptomelane phase to be detected. 10 % DMSO level could affect crystallization
6
without impurities at both 5 mL/min and 10 mL/min, but higher volumes e.g. 25% to
crystallize the correct phase without impurities at any flow rates including higher flow
rates.
The presence of DMSO had a role to play in y-MnCh morphology control in the y-Mn02
system. Absence of DMSO still led to crystallization of y-MnCh but with a different
morphology compared to the morphology obtained when different volumes of DMSO
were incorporated. Nucleation was being achieved with the help of or aided by DMSO
for the morphology to be controlled by this absorber. Since DMSO containing systems
could lead to nanoparticles, then the presence of DMSO hindered particle growth through
aggregation.
1.3
Liquid phase oxidation of organic molecules using manganese oxide
octahedral molecular sieves (OMS-2) catalysts
OMS-2 has successfully been used in the oxidation of different organic compounds in the
liquid phase. The oxidation of various allylic and benzyllic alcohols in the liquid phase
was precedent-setting. 18 Oxidation of other organic compounds in the liquid phase has
also been reported.
(l9 22)
With this report, we extend the application of OMS-2 in liquid
phase oxidation by developing, studying, and optimizing the liquid phase oxidation of
some industrially important compounds.
Fundamentally, we have utilized OMS-2 in the oxidation of 9H-Fluorene to 9Fluorenone. 23 By doing so we have identified some features of OMS-2 that are crucial to
this oxidation reaction, report the best solvent system as well as have studied the kinetics
and proposed a mechanism of 9H- Fluorene oxidation in the liquid phase. 9-Fluorenone is
an important compound that is finding promising uses as a component in organic solar
7
cells and display devices. This reagent also represents an interesting class of compounds
for biomedical applications.' 24-
23)
9-Fluorenone has been synthesized by the oxidation of
9H-fluorene in the presence of alkali-metal hydroxides, 26 by using triton B as a catalyst
in the presence of pyridine, 27 and by using oxygen in the presence of cobalt bromide. 28
Laboratory-scale
preparation
involves
palladium-catalyzed
cyclization
of
o-
iodobenzophenones 29 and carbonylation of o-halobiaryls 30 . The most useful syntheses of
9H-fluorenone include: Friedel-Crafts closures of biarylcarboxylic acids and their
• • 3 1
• •
•
3-)
derivatives, intramolecular [4 + 2] cycloaddition reactions of conjugated enynes, '
remote metalation of 2-biphenylcarboxyamides or 2-biphenyloxazolines. 33 Oxidation of
9H-Fluorene using Ru (111) montmorillonite K 10 with tert-butylhydroperoxide (TBHP)
as the oxidant has also been performed. 34
The use of palladium as catalysts in palladium catalyzed methods is strongly depended on
use of ligands and specifically on the nature of these ligands and also utilizes carbon
monoxide (CO) and cesium pivalate, a base taking up-to 7 h of reaction time while
oxidation with the use of Ru (III) montmorillonite K 10 is insignificant in the absence of
TBHP.
Crystallite sizes as well as average oxidation states (AOS) of OMS-2 are the dominant
players in this reaction. Since OMS-2 can be synthesized conventionally through
different routes, the synthesized materials though retaining the same crystal structure
have different physical and chemical properties. Among these properties is crystallite size
calculated by applying the Scherrer Equation to the X-ray diffraction (XRD) patterns and
the average oxidation state of the manganese in the materials. The oxidation of 9HFluorene was favored by using OMS-2 materials synthesized under acidic buffer
8
conditions (known as OMS-2 buffer or OMS-2B) and OMS-2 solvent-free (OMS-2s).
These two methods led to K-OMS-2 that showed similar smaller crystallites and
comparable average oxidation state (AOS) properties. Mechanistically we have
demonstrated that indeed lattice oxygen could be playing a role together with introduced
oxygen and that this reaction most likely goes through a free-radical route.
The other reaction studied in-depth is the oxidation of 2,4,6-Trimethyl phenol (TMP)
using OMS-2 catalysts synthesized through different routes. 2,4,6-Trimethyl-l,4hydroquinone. product of this oxidation reaction, is important as a starting material in the
synthesis of a-tocophenol; a form of vitamin E that is suitable for human and animal
consumption. Virtually all of TMP used in the process comes from the oxidation of TMP.
Thus the oxidation of TMP represents an industrially important process as far as synthetic
vitamin E is concerned.
Current industrial production of vitamin E involves condensation of isophytol (IP) with 2,
3, 6-trimethyl-l,4-hydroquinone (TMHQ)
35
obtained by hydrogenating TMQ. The
industrial production is carried out via oxidation of TMP with molecular oxygen or air in
the presence of copper halides as catalysts.
oxidation of TMP. (39 '
40)
(36 38)
Generally, TMQ is obtained from
The order of reactions involves para-sulfonation of TMP
followed by its oxidation using stoichiometric amounts of commercial manganese
dioxide. Several researchers have studied the oxidation step of the reaction scheme with
the aim of reducing the overall steps involved as well as moving toward processes that
are more environmentally friendly.
9
OMS-2 materials synthesized through different routes show excellent activity towards the
oxidation of TMP using TBHP as oxidant. This has been achieved in only 30 minutes of
reaction time.
1.4
Selective partial oxidation of benzyl alcohol using manganese oxide
octahedral molecular sieves (OMS-2) catalysts in the gas phase
Selective oxidation of organic molecules, more specifically, the oxidation of benzyl
alcohol serves as a fundamental reaction for both laboratory and commercial processes.
The products of the alcohol oxidation process e.g. benzaldehyde, are valuable both as
intermediates to other compounds and also as end products in the chemical and
perfumery industries. Benzyl alcohol is currently produced through the use of
stoichiometric amounts of commercial manganese oxide (Pyrolusite) or chromium salts
in the laboratory and commercially through the hydrolysis of benzal chloride and the airoxidation of toluene. 41 Disadvantages occasioned by using stoichiometric reagents and
toxic salts have necessitated research into alternative approaches. Both liquid and vapor
phase processes are possible with vapor phase being placed at an advantage due to the
lack of down-stream processing operations that the liquid phase process faces e.g. solvent
separation or extraction. This coupled with the use of a truly heterogeneous catalyst
places gas phase oxidation a pedestal higher than liquid phase oxidation.
An optimized OMS-2 catalyzed liquid phase oxidation of benzyl alcohol has been
reported by Son et. a/. 18 The success of this process led us to study the same reaction in
the gas phase in order to move the process from the liquid to the vapor phase and thus
improve the catalysis by providing the advantages of a gas phase heterogeneous system.
This involved vaporizing liquid benzyl alcohol by passing it through a heated zone and
10
simultaneously introducing the oxidant/carrier mixture that mixes with vaporized benzyl
alcohol for on-ward transport to the reactor containing the fixed bed catalyst. The
oxidation of benzyl alcohol was achieved at various temperatures but optimized at 210
°C. Benzyl alcohol was oxidized with more than 90 % conversion and with good
selectivity to benzaldehyde using both K-OMS-2 and H-K-OMS-2 materials.
1.5
Steam reforming of glycerol for hydrogen generation using nickel-Zeolite Y
catalysts.
Currently, hydrogen is generated through four basic methods; water electrolysis,
gasification of hydrocarbons, partial oxidation of heavy oils, dry or steam reforming of
hydrocarbons. Steam reforming of renewable resources represents a promising alternative
due to environmental concerns and the eventual depletion of hydrocarbon sources.
In this study Zeolite Y was used as the support upon which nickel was loaded and after
the reduction of nickel ions to metallic nickel, used in the study of the glycerol steam
reforming reaction. Glycerol was steam reformed with good yields of hydrogen gas.
11
1.6
1.
2.
References
19.
Weckhuysen, B. M. Nature, 2006, 439, 548-548
Oliveira, M. L. M, Moreno-Tost, R., Silva, C. M., Farias, T. L., RodriguezCastellon, E., Jimenez-Lopez, A. International Journal on Energy for a Clean
Environment, 2009, 10, 103-120.
Kappe C. 0., Chem. Soc. Rev., 2008, 37, 1127-1139
Kumar, V., Kim, J. H., Pendyala, C., Chernomordik, B., Sunkara M. K. J. Phys.
Chem. C, 2008, 112, 17750
Braun, I., Schulz-Ekloff, G., Wohrle, D., Lautenschlager, W. Micropor. Mesopor.
Mater., 1998,23,79
Park, S-E., Chang, J-S., Hwang, Y. K„ Kim, D. S„ Jhung, S. H„ Hwang, J. S.
Catal. Surv. Asia, 2008, 8, 91
Park, S-E, Kim, D. S, Chang, J-S, Kim, J. M, US Patent 2001-0054549-A1
Corbin D. R„ Sacco A. J., Suib, S. L„ U.S Patent 7,014,838, 2006
Bonaccorsi, L., Proverbio, E., Micropor. Mesopor. Mater., 2008, 112, 481
Opembe, N. N., King'ondu, C. K., Espinal, A. E., Chen, C-H., Nyutu, E. K.,
Crisostomo, V. M„ Suib, S. L. J. Phys. Chem. C, 2010, 114, 14417
Villegas, J. C., Garces, L. J., Gomez, S., Durand, J. P., and Suib, S. L. Chem.
Mater., 2005,17, 1910-1918.
Post, J. E. Proc. Natl. Acad. Sci. V. S. A., 1999, 96, 3447-3454.
Post, J. E., and Burnham, C. W. Am. Mineral., 1986, 71, 1178-1185.
Feng, Q., Kanoh, H., Miyai, Y., and Ooi, K., Chem. Mater., 1995, 7, 148-153
Chen, X., Shen, Y.-F., Suib, S. L., and O'Young, C. L., J. Catal., 2001, 197, 292302.
Nyutu, E. K., Chen, C.-H., Sithambaram, S., Crisostomo, V. M. B., and Suib, S.
L. J. Phys. Chem. C, 2008, 112, 6786-6793.
Crisostomo, V. M. B. University of Connecticut Ph.D., University of Connecticut,
2008.
Son, Y.-C., Makwana, V. D., Howell, A. R. and Suib, S. L. Angew. Chem., Int.
Ed, 2001 ,40, 4280-4283.
Kumar, R., Sithambaram, S., Suib, S. L. J. Catal. 2009, 262, 304
20.
Ghosh, R., Son, Y-C., Makwana, V. D., Suib, S. L. J. Catal. 2004, 224, 288
21.
Kumar, R., Garces,
Catal. 2005, 236, 387
22.
Sriskandakumar, T., Opembe, N., Chen, C-H., Morey, A., King'ondu, C., Suib, S.
L. J. Phys. Chem. A. 2009, 113, 1523
23.
Opembe, N. N.. Son, Y. -C., Sriskandakumar, T., Suib, S. L. ChemSusChem
2008,1, 182
24.
Y. Han, A. Bisello, C. Nakamoto, M. Rosenblatt and M. Chorev, J. Pept. Res.,
2000, 55, 230-239.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
L. J., Son, Y-C., Suib, S.
L.,
Malz,
R.
E. ,/.
12
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Greenlee, M. L., Laub, J. B., Rouen, G. P., DiNinno, F., Hammond, M. L., Huber,
J. L., Sundelof, J. G., and Hammond, G. G. Bioorg. Med. Chem. Lett., 1999, 9,
3225-3230.
Carl, F.,DE 4218400, 1978.
Sprinzak, Y., J. Am. Chem. Soc. 1958, 80, 5449-5455.
Fujibayashi, R., Konuma, F., (Jpn. Kokai Tokkyo. Koho), JP 2006199627,
2006.
Qabaja, G., Jones, G. B., Tetrahedron Lett. 2000, 41, 5317.
Marino, A. C„ Larock, R. C„ Org. Lett. 2000, 2, 3675.
Olah, G. A., Mathew, T., Farnia, M., and Prakash, G. K. S. Synlett, 1999, 1999,
1067 ,1068.
Danheiser, R. L., Gould, A. E., de la Pradilla, R. F., and Helgason, A. L. The J.
Org. Chem. 1994, 59, 5514-5515.
Ciske, F. L., and Jones, W. D. J., Chemlnform, 1998, 29, 1195-1198
Hashemi, M. M„ and Beni, Y. A., J. Chem. Res. (S), 1999,434-435
Bonrath, W., Dittel, C., Giraudi, L., Netscher, T., and Pabst, T. Catal. Today
2007, 121, 65.
Rucker, R. B., Suttie, J. W., McCormick, D. B., Handbook of vitamins (Mercel
Dekker, New York, 2001)
Mercier, C., Charbardes, P. Pure Appl. Chem. 1994, 66, 1509
Bonrath, W., Netscher, T. Appl. Catal. A General 2005, 280, 55
Kholdeeva, O. A., Golovin, A. V., Maksimovskaya, R. I., and Kozhevnikov, 1. V.
J. Mol. Catal. 1992, 75, 235.
Meng, X., Sun, Z., Lin, S., Yang, M., Yang, X., Sun, J., Jiang, D., Xiao, F.-S., and
Chen, S. Appl. Catal. A 2002, 236, 17.
Kirk-Othmer Encyclopedia of Chemical Technology, 4th Edition, Vols. 4 and 5,
Wiley, New York, 1992
13
CHAPTER 2.
MICROWAVE
CONTINUOUS
SYNTHESIS
OF
INORGANIC MATERIALS
2.1
Introduction
In synthetic chemistry, the development of new routes and techniques for the synthesis of
inorganic materials is an on-going process. 1 This is because of the need for a fast energy efficient - process that improves yields, leads to purer products and is
reproducible. Microwave (MW)-assisted synthesis is a technique that scores well in terms
of providing fast, reproducible processes. Microwave-assisted chemistry has matured
from a curiosity to an established synthesis technology that finds use in both industry and
academia. Microwave-assisted synthesis has led to dramatic lowering of reaction times,
2*4,
higher product yields, 5 '
the use of microwave ovens,
synthesis.
916
6
and even much purer products.
7 8
6
Since the first reports on
there has been extensive literature on their use in organic
This enabling technology has slowly been adapted to the synthesis of
inorganic materials with tremendous success. Almost all classes of inorganic materials
have been synthesized under microwave irradiation, including but not limited to: oxides,
17 20
chalcogenides,
2126
metals,
27 31
phosphates,
32 33 34
and Zeolites
35,36
among other
materials.
2.2
Principles of MW dielectric heating
What results when a material is exposed to microwaves is a consequence of the material's
properties in relation to how it interacts with microwaves. Based on this interaction,
materials can be classified into three classes based on how they interact with microwaves.
Generally, materials will either be MW-reflectors, transmitters, or absorbers. MW-
14
reflectors are useful as core components of MW waveguides since they have the ability to
reflect and therefore guide MW to a specific region. MW-transmitters are useful in
making MW cookware or reaction vessels e.g. Teflon is a good MW transmitter. MW
transmitters are also known as MW transparent materials. MW-absorbers are useful in
chemistry. These materials can absorb microwaves and this causes them to be heated.
Microwave chemistry relies on efficient dielectric heating of a material. This efficiency is
dictated by the materials' ability to absorb microwaves and generate heat.
GHz, the energy of a photon of microwave radiation (1J mol" 1 )
37
1216
At 2.45
is hardly enough to
even cleave a chemical bond, meaning microwave heating does not induce any chemical
reactions but rather provides an efficient heating means thus enhancing said chemical
reaction. The major heating mechanisms involved in microwave heating are dipolar
polarization or ionic conduction,
38,39
due to the interactions of dielectric materials (both
liquids and solids) with the microwave electromagnetic field.
12
When a polar medium is
heated, this leads to the alignment of dipoles in the same direction as the applied
electromagnetic field. However, the applied electromagnetic field is an oscillating field,
which causes these dipoles to continuously seek to re-align themselves along the
direction of the applied field. Depending of the frequency of this phenomenon, heating
occurs. But if the response time is much slower than the oscillation frequency then no
heating occurs.
38
So is the case for dipolar polarization mechanism of microwave
heating. In the ionic conduction mechanism, dissolved ionic species will oscillate back
and forth bumping and knocking into one another thereby generating heat in the process.
If the irradiated material is an electrical conductor, charge carriers are caused to move
through the material under the influence of the applied field, which causes polarization.
15
Heating then occurs as a result of electrical resistance. Figure 2-1 shows an illustration of
these two heating mechanisms.
The dielectric properties of a material are described based on: (i) dielectric constant, e'
which describes the ability of a material to be polarized by the applied electromagnetic
field, and (ii) dielectric loss factor, E" which describes the extent of conversion of the
electromagnetic field into heat.
38
These two properties are related to each other by
Equation 2-1. Equation 2-1, also known as the energy dissipation or loss factor gives us a
direct means of comparing any two materials in terms of their microwave absorption
properties. Most reaction media (water and organic solvents) are classified for the sake of
comparing their microwave suitability as reaction media based on this parameter. Table
2-1 shows some tan 8 values for selected solvents. These are classified as strong
absorbers (tan 8 > 0.5), medium absorbers (0.1 < tan 8 < 0.5), and poor absorber (tan 8 <
0.1). Solvents with high loss factors include ethylene glycol (1.350), ethanol (0.941) or
dimethyl sulfoxide (DMSO) (0.825), The above compounds have a permanent dipole
moment, whereas solvents without a permanent dipole moment (e.g. hexane) are
basically microwave transparent.
16
Figure 2-1. Microwave electromagnetic field interaction and heating mechanisms, (a)
dipolar polar polarization and (b) ionic conduction mechanisms.
tan 6 = e"/8'
Equation 2-1. Microwaves energy dissipation factor, E" is the dielectric constant and s' is
the energy loss factor.
17
Table 2-1. Loss tangents (tan 8) of selected solvents (2.45 GHz, 20 °C)
Solvent
Dielectric
tan 5
constant
Ethylene glycol
37.00
1.350
Ethanol
24.30
0.941
DMSO
46.68
0.825
2-propanol
19.92
0.799
Formic acid
58.00
0.722
Methanol
32.70
0.659
NMP
32.20
0.275
DMF
36.71
0.161
Water
80.10
0.123
Toluene
2.38
0.040
Hexane
1.88
0.020
Note: DMSO refers to dimethyl sulfoxide, NMP refers to N-methylpyrrolidone and DMF
refers to dimethyl formamide
18
2.3
Types of apparatus used in microwave continuous synthesis of inorganic
materials
MW-assisted synthesis of inorganic nanomaterials utilizes a wide variety of apparatus.
Most of these apparatus are commercially available but others are custom-built.
Pioneering work utilized domestic (kitchen-type) microwave ovens, which have the
disadvantages in the lack of control of: irradiation power, reaction temperature, and
pressure inside the reaction vessels. Commercially available apparatus on the other hand
feature built in temperature and pressure sensors as well as magnetic stirrers among other
features. Most of the commercially available apparatus have provisions for adaptation to
computer systems for control and monitoring of the reaction parameters. Most of these
apparatus operate at a standard frequency of 2.45 GHz. 2.45 GHz is a good compromise
for
a
number
of
reasons.
Firstly,
this
frequency
avoids
interference
with
telecommunication, wireless or cellular phone networks, among other technologies
relying on the use of electromagnetic frequencies. Secondly, the corresponding
wavelength of 12.25 cm produced by this frequency is optimal for microwave penetration
into foods. Lastly, production of magnetrons that produce this frequency is cheaper than
would be done at other frequencies.
To be useful in continuous synthesis, microwave apparatus have to be adapted to receive
reagents into the microwave cavity and discharge products from the cavity on a
continuous basis. This can be achieved by customizing them to have entry and exit ports
through which the charging and discharging can be affected. Nowadays, several
manufacturers in addition to making apparatus that are dedicated to batch synthesis also
19
offer apparatus that have continuous synthesis capabilities. For most, the presence of an
entry/exit port is itself enough to enable the continuous synthesis of nanomaterials.
Continuous microwave synthesis has been performed under hydrothermal conditions in a
commercial microwave hydrothermal apparatus (Figure 2-1). This was achieved using a
microwave-heated flow vertical tube reactor (FLOWSYNTH Continuous Flow Reactor).
This system has a constant frequency of 2.45 GHz and is pressure controlled. The system
can attain a maximum pressure of 30 bar. A Teflon reaction vessel is connected to a
pressure transducer that monitors and controls the pressure during synthesis. Charging of
reactants into the reaction vessel is pump controlled while the flow rate, pressure, and
power are all computer controlled. The reactor is attached to a pump at the inlet end and
passes into a wall-cooled heat exchanger at the effluent end. Reaction mixtures are
rapidly cooled as they exit the irradiation zone.
A single-mode custom-built microwave apparatus that involves mixing reactants then
spraying them into a microwave chamber has been reported. 35
48 49 As
can be seen
schematically in Figure 2-3, the apparatus is made up of three component parts all
working to affect the continuous synthesis of inorganic nanomaterials. More importantly,
this apparatus can be used to control purity levels and confine the size of the synthesized
materials to the nanometer range. Two methods can be used. One involves just spraying
reactants into a microwave-heated zone after passing through a nozzle and is termed
nozzle microwave (NMW) and the other involves mixing of two reactant mixtures first
and is termed in-situ nozzle microwave (INM). These are shown in Figure 2-3 (a) and (b)
respectively. INM consists of three components. The first component is the in-situ mixing
setup, which relies on syringe pumps to continuously pump reactants. In this setup, two
20
solutions can be mixed essentially simultaneously at a T-connector before they arrive at
the second component, which is an atomizer. The atomizer plays the role of nebulizing
said reactants through the ultrasonic cavitation method into fine droplets. These misty
reactants then pass down a quartz or Teflon tube under gravity and reach the third
component, which is the reaction chamber. The chamber is under controlled microwave
power, which controls the resultant temperature.
Continuous microwave synthesis using a plasma jet reactor has also been reported.
8I
The
plasma reactor can be operated starting from atmospheric pressure to moderate pressure
(a few Torr). Power can be controlled from 300 W to 3 KW. A plasma zone of about 3045 cm is produced in a quartz tube of 3-5 cm diameter. The plasma is ignited with a
pointed metallic rod. This apparatus has been used to accomplish syntheses where
metallic powders were injected into the plasma cavity by means of mechanical dispensers
or carried into the plasma zone by means of a carrier gas. The powders interact with MW
radiation as they pass through the plasma zone and the resultant products are collected at
the bottom of the tube (Figure 2-4).
The work presented in this thesis is based on the use the CEM™ Mars 5 microwave
apparatus that has an opening that enables tube reactors to be inserted into the cavity (see
Figure 2-7 for a schematic). In this apparatus, a Teflon or quartz tubing can be used to
charge and discharge materials from the apparatus at atmospheric pressure.
21
• * _g-fTi.ii
*
Figure 2-2. Pictorial representation of hydrothermal microwave FLOWSYNTH that is a
newer version of the old model MLS ETHOS CFR microwave (obtained from
www.milestonesrl.com on April 3, 2012)
22
Injoctton and
» <n*Au mixture
of roctant*
gfrfuftffn
• Formation of
draptata
Uqukl Droplets
N, ,>n
•*-*-.
>V*
H .• ••
Colloid
* fornatfonand
V I .*
(A)NMW
(B) MM
Figure 2-3. (a) Nozzle microwave and (b) in-situ nozzle microwave apparatus
(Reproduced with permission from: Espinal, L., Malinger, K. A., Espinal, A. E., Gaffney,
A. M., Suib, S. L. (2007) Adv. Funct. Mater. 17, 2572. Copyright: W1LEY-VCH Verlag
GmbH & Co. KGaA, Weinheim)
Guffaw
(Ar,Atr,fK2,OZ)
3MMK
PIlWIUMl
cwplcr.
—
2.4 GHl
DX. fvmx
P—l M
Hwndhttwii
QaMtawadtl
lute
ClflCCiiMI CSp
Figure 2-4. (a) Schematic diagram for the microwave plasma jet reactor showing the
components, and (b) the actual plasma generated. (Reproduced with permission from:
Kumar, V., Kim, J. H., Pendyala, C., Chernomordik, B., Sunkara M. K. (2008) J. Phys.
Chem. C, 112, 17750. Copyright: America Chemical Society)
23
2.4
Microwave techniques used in the synthesis of inorganic materials
Most MW-assisted techniques for synthesizing inorganic materials have focused on the
use of batch techniques. However, batch operations have inherent disadvantages such as
limited scale up possibilities, time and effort in changing reagents in each batch
processes, among others. Scale-up in batch synthesis is made even harder due to limited
penetration depths of microwaves into most reaction media. The problem emanates from
establishing safe and reliable processes where penetration depths, control of the
temperature, and reactor design are crucial parameters. In a batch scale-up, microwaves
lack penetration power thereby heating only the solution at the periphery of the reaction
vessel.
40
Therefore, the bulk of the reaction mixture is heated through a conventional
heat transfer process and not by microwaves. This renders microwave chemistry
ineffective in performing multi-kilogram syntheses on a batch mode. The solution to this
drawback lays in the use of a continuous flow process
41
whereby reactant flow rates and
microwave residence times dictate the scale of the synthesis. The need for big reaction
vessels is replaced with tubular reactors with an inner diameter of a few millimeters and
capable of producing multi-kilogram quantities of materials. Reactant streams are
continuously pumped into the reactor while product streams continuously leave the
reactor at the discharge end. As can be seen in Table 2-2, the penetration depths of
microwaves are limited to about 1.4 cm at room temperature increasing slightly to 5.7 cm
when heated to 95 °C. However, based on the penetration depths of some materials e.g.
Teflon or quartz which have excellent penetration to microwaves, a continuous process
can be developed that utilizes these two reaction vessels (microwave transparent) and if
24
using water as reaction media, restrict the penetration depth to a few centimeters (i.e.
vessel diameters of a few centimeters).
Several researchers have focused their research on inorganic syntheses using batch
techniques. Their work has been captured in several journal articles and reviews.
27 42 43
Microwave-assisted continuous-flow synthesis is a relatively unexplored area going by
the volume of publications on the subject.
The increasing demand in the use of nanoparticles necessitates the industrial manufacture
of nanomaterials, which is still difficult. The fundamental challenge lies in producing
uniform nanomaterials in bulk quantities and with good reproducibility. In the recent
past, continuous techniques have been explored to scale-up nanomaterial synthesis for the
advancement of nanotechnology. Continuous microwave synthesis provides a low-cost
and easy-to-control method for such processes. Metallic nanoparticles,
nanoparticles,
46 47
multiple metallic oxide nanoparticles,
48 49
44 45
and zeolitic
metal oxides
50
materials
have been successfully synthesized via the use of continuous synthesis techniques.
Microwave assisted techniques have proved to be very powerful a tool in the synthesis of
metal oxide nanoparticles. Monodispersed zirconia colloidal spherical nanoparticles were
synthesized by Bondioli et al.
nanoparticles
were formed
46
using a continuous microwave synthesis process. ZrCh
from
the
hydrolysis and
condensation
of tetra-n-
propylzirconate under microwave irradiation. The flow rate was optimized to obtain nonagglomerated Zr02 nanoparticles. When the flow rate was 50 mL/min, the particles were
spherical with a mean particle size of about 100 nm. Nanocrystalline tin dioxide was
prepared by Subramanian et al.
47
via a continuous microwave irradiation of a tin-
containing gel precursor. After the gel matrix was burnt off in the presence of microwave
25
irradiation, this resulted in the formation of materials with high surface areas (15 - 18
nm). The flaky powder was further aged to form SnCh nanocrystals.
Multicomponent metal oxide (MMO) crystallites have also been continuously prepared
by Espinal et al.
48
via spraying a reactant solution into a receiving solution or into air
under microwave radiation at atmospheric pressure. A series of single/multiple
metal/metal oxide alloys were prepared; including, Mo-V-0 system, Mo-Nb-0 system,
Mo-V-Te system, Mo-V-Te-Nb-Pd system, Mo-V-Te system, and Mo-V-Te-Nb system.
These materials are important catalysts for highly catalytic partial oxidation of propane to
acrylic acid. This technique allows for two different mixing methods of the reagents
before the reaction in the microwave cavity. One is called the nozzle-spray/microwave
(NMW) method and the other is called the w-s/Yw/nozzle-spray/microwave (INM). The
NMW method produced particles with rod-like morphologies that were different from
those obtained using conventional hydrothermal (CH) methods. The INM method
produced amorphous materials but could be further crystallized into small (ca. 200 nm)
particles with interesting morphologies after calcination. The materials prepared by this
combined microwave assisted continuous technique and calcination method showed
smaller particle sizes, larger surface areas, and unique morphologies compared to those
prepared under conventional hydrothermal methods. The INM method was also explored
by Nyutu et al. 49 to prepare spinel-type metal oxide materials. Small sized (6-20 nm),
nanocrystalline spinel-type nickel ferrite and zinc aluminate particles with average
particle sizes about 9 nm were prepared. The syntheses were carried out at ambient
pressure (1 atm), microwave power (0-600 W), and ultrasonic nozzle with resonant
frequency of 48 or 120 kHz. The surface area of nickel ferrite ranged from 57 to 72 m 2 /g.
26
The use of a lower-frequency ultrasonic nozzle in-situ mixing (48 kHz) resulted in
marginally higher surface area of nickel ferrite nanoparticles than with a 120-kHz nozzle.
Microporous Zeolites have also been synthesized by the continuous microwave synthesis
technique.
50
Molecular sieve Linde-type Zeolite A (LTA), one of the most important
shape-selective zeolitic catalysts was recently synthesized by Bonaccorsi et al. 50 In this
work, Zeolite LTA was obtained by a microwave continuous synthesis process. A better
understanding of this microwave process was however called to further study since the
feasibility of this method pointed to some advantages over conventional processes. This
could be achieved by a careful study of the parameters involved. Some of these were
investigated, such as the effect of the seeds addition in the gel, the varying of the
microwave output power and the use of different reactor geometries. Results
demonstrated that the preliminary gel nucleation was a fundamental step for the
microwave Zeolite crystallization as well as the output power management. A coiled pipe
and a spherical vessel used as continuous reactors showed a different behavior in similar
synthesis conditions, confirming the relevance of the geometrical distribution of matter in
the microwave cavity and the generated electromagnetic field distribution. The relative
influence of microwaves, of seeding and of fluidodynamic conditions was not only
evident in terms of final product crystallinity but also in particle size and distribution, as
confirmed by the morphological analysis.
The continuous-flow concept was also extended to the synthesis of metallic
nanoparticles. Lin et al.
44
designed a continuous-flow tubular microreactor for the
synthesis of Ag nanoparticles. The silver pentafluoropropionate was used as a singlephase reactant precursor, which was thermally reduced in isoamyl ether to form silver
27
nanoparticles in the presence of trioctylamine. The produced Ag nanoparticles were
narrowly distributed in size and had an average diameter of 8.7±0.9 nm by size
distribution. A moderate temperature for the formation of Ag nanoparticles is the key to
this continuous-flow synthesis process. Gold nanoparticles were later on synthesized by
Wagner et al.
45
with the use of a microfluidic setup. The Au nanoparticles were directly
synthesized via a redox reaction of a gold salt (HAuCU) and a reducing agent (ascorbic
acid). The sizes of the as-synthesized nanoparticles ranged from 5 to 50 nm. The size
distribution was on order of magnitude two times narrower than that obtained in a
conventional synthesis when experimental parameters, such as pH, flow rate, the use of
reducing agent and the use of surfactant were optimized.
2.5
Microwave-assisted continuous synthesis of manganese oxide materials
Porous materials show lots of potential and are very interesting materials due to their
specific properties and diversity of structures. Some of their applications are: energy
storage, separations, gas sensing, and catalysis among others. Manganese oxide materials
with tunnel structures constitute a large family of porous materials both microporous and
mesoporous.
51
Two of these materials are cryptomelane-type manganese oxide (a-
Mn02) octahedral molecular sieves (OMS-2) and gamma-manganese oxide (y-Mn02).
OMS-2 belongs to a group of manganese oxides having a 2 x 2 tunnel structure formed
by edge and corner sharing of manganese oxide (Mn06) octahedra resulting in a onedimensional (1-D) channels as illustrated in Figure 2-5.
52 57
Positive cations for instance
potassium (and hence K-OMS-2) and water molecules occupy the tunnels. The tunnel
cations can be partially or fully ion-exchanged with other ions. For instance, exchange
with protons (H + ) gives its acidic form - H-K-OMS-2. These materials have been used as
28
catalysts in numerous studies such as selective oxidation; of alcohols,
58
9H-fluorene,
cyclohexene, 60 epoxidation of styrene and other olefinic substrates,
61
synthesis of 2-
aminodiphenylamine,
ethane
64
62
decomposition of dyestuffs,
63
59
oxidative dehydrogenation of
among other reactions. Alternatively, they can be used as molecular sieves in
separation processes
65
or in energy storage applications.
66
K-OMS-2 is mixed valent
with manganese found in oxidation states of Mn 2+ , Mn 3+ , and Mn 4+ . 67
The synthesis of K-OMS-2 materials has been realized through different routes.
52 54 - 56 - 68
Systematic studies have been done to among others reduce the particle size to the
nanometer range. Villegas et al. 52 successfully prepared nano-sized crystalline K-OMS-2
fibers with crystallite sizes as low as 6 nm by reducing KMn0 4 with H2O2 in an acidic
buffer medium in about 15 h, under conventional reflux. Recently, Nyutu et al. 69 reported
a method for synthesizing K-OMS-2 in mixed aqueous and non-aqueous solvents by use
of microwave reflux, obtaining K-OMS-2 nanomaterials with crystallite sizes as low as 4
nm. This method required about 10 minutes to start crystallizing the K-OMS-2 phase and
about 90 minutes to fully crystallize the phase. Crisostomo
70
developed a continuous-
flow method for synthesizing s-Mn02, Ce O2, CoOOH, and FeOOH using microwave
irradiation.
This research focuses on developing a microwave continuous synthesis technique and
reports forthwith a systematic study of the continuous-flow synthesis of K-OMS-2
materials using microwave irradiation utilizing a process similar to the one developed by
Crisostomo, 70 and using a co-solvent system from Nyutu. 69 This was accomplished by
investigating the effect of important parameters such as microwave irradiation power, cosolvent amount (% DMSO) and reactant flow rates (and hence microwave residence
29
times). By continuously flowing a reactant mixture through a microwave cavity and by a
correct choice of power, DMSO amount, and flow rates, it was possible to establish
optimal conditions that allowed the continuous mass production of pure nano-fibrous KOMS-2. This technique can be adopted for potential large-scale industrial synthesis of
this and other nanomaterials.
An irregular growth of (1 x 1) tunnels (pyrolusite) and (1x2) tunnels (ramsdellite)
results in y-Mn02 (Nsutite).
71
The crystal structures of pyrolusite, ramsdellite and
Nsutite can be seen in Figure 2-6. Ramsdellite and pyrolusite consist of hexagonal closepacked (hep) array of oxide ions in which half the octahedral sites are filled with Mn 4+
forming tunnels along the c-direction. Structures of pyrolusite, ramsdellite and gammamanganese oxide are shown in Figure 2-6. y-Mn02 was synthesized using a similar set-up
as that used for the synthesis of K-OMS-2 as will be described in the sections to follow.
30
Figure 2-5. Crystal structure representation of K-OMS-2 with potassium (K+) counter
cations shown as maroon spheres while Mn06 octahedra are shown as blue octahedral.
The tetragonal unit cell (a=b=9.82
A,
c=2.85
A) is
also shown with blue lines. Crystal
structures drawn using Crystal Maker™.
vC
(a)
(b)
(c)
Figure 2-6. Crystal structural diagrams of (a) pyrolusite (P-Mn02); (b) ramsdellite and
(c) De Wolffs model for y -Mn02 showing intergrowth between pyrolusite and
ramsdellite domains. Crystal structures drawn using Crystal Maker™.
31
2.6
Experimental
2.6.1
Materials
Potassium permanganate (KMn04), manganese sulfate monohydrate (MnSCVhhO),
2,3,6-trimethylphenol (TMP), Dimethyl sulfoxide (DMSO) (99.9%, Acros), acetonitrile
(MeCN), potassium persulfate (K2S2O8) and tert-Butyl hydroperoxide (70% in water
TBHP) was purchased from Sigma-Aldrich while concentrated nitric acid
(HNO3)
was
obtained from Alfa Aesar. All reagents were used without any further purification.
2.6.2
Microwave set-up
Continuous-flow syntheses were carried out in a CEM Mars 5 multimode microwave.
The microwave is equipped with one magnetron, fixed frequency of 2.45 GHz and a
three-level power output of 300 W, 600 W, and 1,200 W, where power at each level can
be controlled within a 0 - 100% range. The system can use a temperature-sensing device
(fiber-optic) for in-situ measurements of temperature inside open or sealed reaction
vessels and also contains an in-board pressure control system to monitor and control
reaction pressures inside sealed reaction vessels. The apparatus has overall dimensions of
25" x 20" x 23" (D x W x H). The microwave can be operated in any of the following
five control modes: standard control, ramp to temperature, ramp to pressure, power/time
control, and microvap. Temperature and pressure can be set and monitored in all except
the power/time control mode where power and time are the only variables. The
power/time control mode was used since the reactor configuration does not allow for
adaptation of temperature or pressure-monitoring devices. Inside the microwave cavity a
coiled tubular reactor made of Teflon and with dimensions of Oj nner = 1/8", <J>outer
=
1/4",
32
Ocoii
=
1-1/2", and L=72" was used. The coiled reactor was continuously fed with
feedstock by use of a peristaltic pump connected to the coiled tubular reactor inlet and
operated from outside the microwave. Only the coiled part of the reactor was directly
irradiated with the microwaves. The peristaltic pump controlled the feed flow rates. All
reactions were nominally run at atmospheric pressure since both the feeding and
sampling ends were open. Effects of feed flow rates, microwave residence times, and
influence of microwave power on the synthesis were investigated. A schematic diagram
of the setup can be seen in Figure 2-7 while the actual pictorial representation is shown in
Figure 2-8.
2.6.3
Microwave Treatments
The microwave was run in the power/time mode with power being varied from 150 W,
300 W, 450 W, and 600 W by changing percentage power at 600 W by 25%, 50%, 75%,
and 100% respectively. In the power/time mode, temperature can neither be controlled
nor monitored.
33
Product
Figure 2-7. Schematic drawing of the continuous flow synthesis set-up showing a coiled
Teflon reactor inside the microwave cavity.
I'op inl«coutl« to MW cavity
MW cavity
Coiled Teflon tmm
Figure 2-8. A photo of the actual microwave set-up used in the continuous flow
synthesis.
34
2.6.4
Continuous synthesis of K-OMS-2
The synthesis K-OMS-2 has been done using different approaches. We chose a recently
reported method 27 , which involves the use of microwave irradiation in aqueous and non­
aqueous systems. The chosen method utilizes the conproportionation by Mn 2+ and Mn 7+
ions present in manganese sulfate and potassium permanganate respectively. In a typical
reaction, 42 mmol (6.65 g) KMn0 4 was added to 100 mL distilled-deionized water
(DDW) to make mixture A. In another flask, 59 mmol (9.9 g) of MnSO^LhO was added
to 33 mL DDW to make mixture B. Both A and B were stirred separately until complete
dissolution of the reagents. To B, 3.4 mL of concentrated
HNO3
was added and further
stirred. Solution A was transferred to a dropping funnel and drop-wise added to solution
B under vigorous stirring. The resultant mixture (C) was further stirred at room
temperature for a further 15 minutes before addition of different volumes of DMSO to
make up various percentages (v/v).
For instance, to make a 10% (v/v) DMSO in C, 15 mL of DMSO was added to C.
DMSO-containing mixture C was kept stirring in a flask throughout the duration of the
microwave treatment. Rubber tubing was connected to this flask and to a peristaltic pump
and further connected to the inlet of a coiled reactor that was maintained in the
microwave cavity. Reactants were continuously pumped through the coiled reactor in the
microwave cavity at different flow rates, where they interacted with microwave
irradiation for a given duration of time. Products were thereafter collected at the
discharge end of the reactor into a collecting flask, washed with distilled-deionized water
(DDW), centrifuged and dried overnight at 120 °C in an oven.
35
As a starting point, power was fixed at 300 W and DMSO at 10%. With these parameters
fixed we studied the effect of flow rates by varying the flow by: 40 mL/min (MC-10-40
mL/min), 20 mL/min (MC-10-20 mL/min), 10 mL/min (MC-10-10 mL/min), and 5
mL/min (MC-10-5 mL/min). Next, power at 300 W and the flow rate at 10 mL/min were
held constant and the amount of DMSO varied as follows: 0% (MC-0-10 mL/min), 5%
(MC-5-10 mL/min), and 25% (MC-25-10 mL/min). These were compared to the 10%
DMSO above i.e. MC-10-10 mL/min. A similar study was done at a higher flow rate of
20 mL/min but the amount of DMSO kept at: 5% (MC-5-20 mL/min), 25% (MC-25-20
mL/min), and 50% (MC-50-20 mL/min); comparing the results to MC-10-20 mL/min
sample. Effects of the reactor configuration in the synthesis and varying the microwave
power from 150 W to 600 W were also studied. As for the reactor configuration study,
the coiled tubular reactor was replaced with a non-coiled reactor of similar dimensions.
2.6.5
Continuous synthesis of y-Mn02
Two recipes were used in the synthesis of y-Mn02 with the first involving use of
78
manganese sulfate monohydrate, nitric acid, and potassium persulfate (K2S2O8) . The
procedure involved mixing 6.5 g in 200 mL DDW that contained 2 mL
HNO3,
and stirred
until complete dissolution. The solution turned clear at this point. To this solution was
added 2.7 g MnSO^LhO with stirring until complete dissolution. The resulting clear
solution is fed into the MW at different flow rates at two different power levels (300 W
or 600 W), with or without addition of DMSO. The second recipe involved a variation of
the reported procedure.
72
We approached this by varying the Mn 2+ /Mn 7+ ratios in the
redox reaction mixture similar to that occurring in K-OMS-2 synthesis. We used ratios of
1.4,2.3, and 2.8.
2.7
Characterization methods
2.7.1
Powder X-ray Diffraction studies
The powder X-ray diffraction studies were performed on a Scintag XDS-2000
diffractometer using Cu Ka (k = 0.15406 nm) radiation and operating at a beam voltage
of 45 kV and a current of 40 mA. Data were collected continuously in the 20 range of
5-75° at a scan rate of 1.0 deg/min and the phase identified using a JCPDS database card
number 29-1020. The XRD patterns of samples were collected on either a glass or an
aluminum sample holder.
2.7.2
Crystallite Particle Size
The crystallite particle sizes of the prepared K-OMS-2 materials were determined by
applying the Scherrer Equation to the reflections at (310), and (211) of the XRD patterns
with the integral widths corrected using a LaB & standard.
2.7.3
Transmission Electron Microscope (TEM)
TEM studies were carried out with a JEOL 2010 UHR FasTEM, operating at an
accelerating voltage of 200 kV and equipped with an energy dispersive X-ray analysis
(EDS) system. The samples were prepared by dispersing the powder material in 2propanol. A drop of the dispersion was placed onto a holey carbon-coated copper grid
and allowed to dry.
2.7.4
Scanning Electron Microscope
Field Emission Scanning Electron Microscopy (FE-SEM) was performed on a Zeiss
DSM 982 Gemini instrument with a Schottky emitter at an accelerating voltage of 2 kV
37
and a beam current of 1 mA. The carbon tape mount method was used where powder
samples were dispersed in iso-propanol in a glass vial and ultra-sonicated prior to being
dispersed on Au-Pd-coated silicon glass chips previously mounted onto aluminum stabs
with a two-sided carbon tape and dried by vacuum desiccation prior to SEM studies.
2.7.5
Thermal Stability
Thermogravimetric analysis (TGA) and differential scanning caiorimetry (DSC) were
employed to study the thermal behavior of the samples. The TGA experiments were
performed with a Hi-Res TA instrument Model 2950, while differential scanning
caiorimetry (DSC) experiments were performed on a DSC Model Q20. In both
experiments, temperature was increased at a ramp rate of 20 deg/min in nitrogen
atmosphere.
2.7.6
Potentiometric Titrations
Potentiometric titrations
52
were performed to determine the average oxidation state
(AOS) of manganese in the synthesized materials.
2.7.7
Surface Area and Pore Size Distribution
These were performed using nitrogen sorption on a Micrometrics ASAP 2010 accelerated
surface area system. Samples were degassed at 120 °C for 12 h prior to the pore size
distribution experiments which were carried out at 77 K. The specific surface area of the
material was determined with the Brunauer-Emmett-Teller (BET) method while pore
size distribution (PSD) determined using Barrett-Joyner-Halenda (BJH) methods.
38
2.7.8
Surface and Functional Group Studies
The surface and functional group properties of the prepared materials were studied by
Fourier Transform Infrared (FTIR), using a Nicolet Magna-IR Model 560 in the range
-i
4000-400 cm with a DTGS detector. K-OMS-2 powders were diluted in a ratio of 1:100
with KBr and then pressed into pellets at about 15 000 psi.
2.8
Results
2.8.1
Continuous Formation of K-OMS-2 Nanomaterials.
The conproportionation of Mn 7+ and Mn 2+ in the mixed aqueous-organic solvent system
is responsible the formation of the Cryptomelane-type K-OMS-2. When KMnC >4 solution
was added to the MnSO^J-bO solution that contained nitric acid, the mixture turned light
brown in color and changed to a darker brown shade with the addition of more KMnC > 4
solution. This color did not change significantly upon addition of DMSO. The low
viscosity of the resultant slurry allowed for its flow in the coiled reactor without clogging
it. This precipitate was not analyzed since XRD results of a similar precipitate
69
showed
a poorly ordered K-OMS-2 phase. We chose 10% DMSO and 300 W power level as the
starting point for our study. The % DMSO choice was based on results from a previous
report 69 which used a mixed aqueous and non-aqueous system in a batch process. A 10%
DMSO mixture produced a crystalline phase of K-OMS-2 after 90 minutes reaction time
in the same microwave apparatus. In this study, a peristaltic pump controlled the flow
rates. The mixture was continuously fed at 40, 20, 10, and 5 mL/min flow rates with
corresponding residence times of: less than a minute (-50 seconds), 2 minutes, 4 minutes,
and 8 minutes, respectively and at a constant microwave power of 300 W. X-ray
39
diffraction was used to confirm the crystal phase. At 40 mL/min poorly ordered
manganese oxide phase was obtained as shown in Figure 2-9 (MC-10-40 mL/min). At 20
mL/min the phase was less poorly ordered (MC-10-20 mL/min) with peaks characteristic
of K-OMS-2 (JCPDS 29-1020) evident albeit with an impurity phase (20 = 23°). A
slower rate of 10 mL/min led to a much purer and more crystalline phase of K-OMS-2. A
similar pure phase was obtained at 5 mL/min.
We sought to study the effects of % volume of DMSO at different flow rates on the
synthesis of K-OMS-2. Figure 2-10 shows the resultant XRD patterns. At 10 mL/min,
when no DMSO is added to the synthesis mixture, transformation to the K-OMS-2 phase
does not occur. Similar results are to be expected at 20 mL/min and higher rates. At 5 %
DMSO, evolution of broad K-OMS-2 peaks was observed indicating the formation of
small crystallites (see MC-5-10 mL/min and MC-5-20 mL/min in Figure 2-10). However,
at this DMSO level phase impurity was observed.
Similar peak broadening was also observed at both the 10% DMSO levels at both 10
mL/min and 20 mL/min flow rates (see MC-10-10 mL/min and MC-10-20 mL/min) but
the latter sample had an impurity phase that was absent from the former. Syntheses with
25% DMSO at both flow rates (see MC-25-10 mL/min and MC-25-20 mL/min) led to
pure K-OMS-2. At 20 mL/min, both the 25% and 50% DMSO level samples showed
consistently broader peaks with no impurities (see MC-25-20 mL/min and MC-50-20
mL/min). The Scherrer Equation was applied to the (310) and (211) XRD reflections to
estimate the crystallite sizes of the materials in Figure 2-10 with the exception of the
materials that did not transform to K-OMS-2 or had pronounced impurities. We sought to
find out what the exact temperatures at the various power levels used were, since in the
40
power/time mode used we could not monitor this parameter. While in the power/time
mode, we fixed power at 300 W, and carried out batch reactions on small volumes (10
mL) of the reaction mixture for time periods similar to retention times in the flow process
while using a thermometer to measure the reaction temperature. We also studied the
temperature at different power levels at a fixed residence time of 4 minutes. Figure 2-11
shows the XRD patterns as well as temperatures measured for these batch syntheses.
Figure 2-11 (a-d) shows that decreasing the residence times at a constant power of 300 W
while Figure 2-11 (e) and (f) show the effect of different power settings at a fixed
residence time. The temperatures obtained are also indicated. At 300 W at different
residence times (Figure 2-11 a-d) the reaction temperature was constant at 90 °C.
However, when power was changed to 150 W and to 600 W, the temperature changed to
85 °C and 100 °C, respectively.
The effect of microwave power on the syntheses is shown in Figure 2-11. K-OMS-2
nanomaterials synthesized at 150 W, 450 W, and 600 W at 10 mL/min and with 10%
DMSO had impurity phases present. Only at 300 W was a pure phase obtained.
Finally, we studied the effect of reactor configuration on the syntheses. For this study, the
coiled-tubular reactor was replaced with a non-coiled one made from a similar material
and with similar dimensions. Upon replacing the coiled reactor with a non-coiled one, the
residence time changed significantly. For instance, at 10% DMSO and 10 mL/min flow
rate, the residence time was only 1.5 minutes as compared to 4 minutes for the coiled
reactor. This difference led to a phase that could largely be ascribed to y-Mn02. Figure 212 shows the results obtained. At a slightly slower rate of 5 mL/min, the phase started
transforming to K-OMS-2 with a 3-minute residence time.
41
js(2e)=-^
L cos#
Equation 2-2. Scherrer Equation showing that Peak width (B) is inversely proportional
to crystallite size (L). K is the shape factor, X is the wavelength of the X-rays, and 0 is the
Bragg angle.
S
f
I
10
20
30
40
Two Th*t» (D*g)
50
60
70
Figure 2-9. XRD patterns of K-OMS-2 showing the effect of reactant flow rates at
constant microwave power (300 W) and constant volume of DMSO (10 % v/v). Reactant
flow rate was varied from 40 mL/min down to 5 mL/min.
42
Itw
3
i
I
c
i
MC.2MM.Mb
?
K
1U
?0
Two-Theta (dM|
iij
Figure 2-10. XRD patterns of K-OMS-2 prepared using different percentages of the cosolvent (DMSO) at two flow rates - 10 mL/min and 20 mL/min. The phases were
matched to the Q-phase of cryptomelane (KMn8016; JCPDS 29-1020).
43
o
o
4 mm, 600 W
I
4 mm, 150 W
<1 min, 300 W
10
20
30
40
70
60
50
Two Thata (d*g)
Figure 2-11. X-ray diffraction patterns of OMS-2 materials synthesized batch-wise with
temperature measurements determined as follows; a, b, c, and d as 90 °C, (e) as 85 °C and
(f) as 100 °C.
tf « impurity phas«
*-AJ holder
3
i
S
t
m
»
DC
U I\
_l
to
J
20
,
3C
k
1
40
Two-Thetn (d«g)
/I*
i
50
)
60
[
70
Figure 2-12. X-ray diffraction patterns of OMS-2 materials synthesized at different
power levels (150 W- 600 W) with 10 % DMSO and a flow rate of 10 mL/min.
44
2.8.2
Crystallite Size
The crystallite particle sizes of the prepared K-OMS-2 materials were determined by
applying the Scherer Equation (Equation 2-2) to the reflections at (310), and (211) of the
XRD patterns with the integral widths corrected using a LaB 6 standard. The calculated
crystallite size results are part of Figure 2-10 and also in Table 2-3. All the materials
synthesized at 300 W at either 10 mL/min or 20 mL/min and with different percentages
of DMSO, had almost similar crystallite sizes ranging between 1.6 - 1.8 nm.
2.8.3
Morphology
Figure 2-13 (a) shows the morphology of MC-0-10 mL/min sample prepared in the
absence of DMSO. The sample shows agglomerated nanospheres of diameters averaging
about 200 nm. This is an amorphous phase as can be confirmed by the XRD pattern.
Figure 2-13 (b) is for the sample MC-5-10 mL/min (5% DMSO). Both 10 (a) and 10 (b)
show similar features (agglomerated nanospheres), but unlike 10 (a), 10 (b) shows
isolated fibers and also the presence of a third phase of manganese oxide. Material in
Figure 2-13 (b) unlike 10 (a) were synthesized in the presence of 10% DMSO and at a
higher flow rate of 40 mL/min. With the flow rate set at 20 mL/min and % DMSO
increased to 25%, a single pure phase of K-OMS-2 nanoparticles with fibrous
morphology was obtained as shown in 10 (c). Increasing the % DMSO to 50% (MC-5020 mL/min) and holding the flow rate at 20 mL/min, a similar fibrous morphology of KOMS-2 nanoparticles was obtained. Synthesis of K-OMS-2 under conventional reflux
yields materials with similar fibrous morphology but longer fiber lengths, and is normally
achieved after prolonged heating of 24 h. Figure 2-14 shows HRTEM images of the
samples prepared at a flow rate of 20 mL/min and 25% DMSO (MC-25-20 mL/min).
45
Figure 2-14 (a) taken at a lower magnification reveals K-OMS-2 images with much
smaller K-OMS-2 nanofibers. The diameter of the smaller of the particles is about 5 nm
and fibers with lengths as small as about 20 nm can be observed. HRTEM micrograph
Figure 2-14 (b) reveals lattice fringes of 0.48 nm, which can be indexed to the (200)
lattice planes. This also indicates that the orientation of K-OMS-2 is aligned with the
tunnel along the c-axis of the fibers.
46
Figure 2-13. FESEM of K-OMS-2 materials MC-0-10 mL/min (0% DMSO), (b) MC-1040 mL/min (10% DMSO), (c) MC-25-20 mL/min (25%DMSO) and (d) MC-50-20
mL/min (50% DMSO). The scale is 500 nm
IBBMBMl
MNHNllf
ill
ippiipi
lililitl
—
itsf
Figure 2-14.
llSlfei
"
*
h -j jjui
Low resolution (a) and high-resolution (b) TEM images of K-OMS-2
materials prepared by continuous flow microwave technique at 20 mL/min in 25%
DMSO (MC-25-20 mL/min).
47
Table 2-2. Microwave penetration depth (2.45 GHz) in selected materials.
Material
Penetration depth
Temperature
(mm)
(°C)
Water
14
25
Water
57
95
Glass (borosilicate)
350
25
Poly (vinyl chloride)
2100
20
Teflon
92,000
25
Quartz glass
160,000
25
Table 2-3. Crystallite sizes of K-OMS-2 materials prepared under MACS
Material ID
Crystallite size
Crystallite size
d(310)
d(2H)
MC-5- 10 mL/min
1.7
1.9
MC-10- 10 mL/min
1.7
1.9
MC-25-10 mL/min
1.9
1.9
MC-5-20 mL/min
1.6
1.8
MC-10-20 mL/min
N/D
N/D
MC-25-20 mL/min
1.9
1.7
MC-50-20 mL/min
2.0
1.8
Note: Crystallite sizes were calculated using the Scherrer Equation to the XRD
reflections at (310) and (211). N/D - Not determined due to impure sample.
48
2.8.4
Functional groups
The FT1R spectra of K-OMS-2 materials prepared under microwave continuous-flow
conditions are shown in Figure 2-15. The materials prepared exhibited the typical KOMS-2 infrared spectra corresponding to Mn-0 stretching modes
73 74
in the region 800 -
400 cm" 1 . Additional bands were observed in the region J150 - 950 cm" 1 . These bands
appear at: ~ 1120 cm" 1 , -1050 cm" 1 , and -950 cm" 1 . Another band can be observed at
1633 cm" 1 .
2.8.5
Thermal Stability
TGA and DSC were used to study the thermal stability of the synthesized materials.
Three samples were used for this study as shown in Figure 2-16 and Figure 2-17. These
samples were: MC-10-10 mL/min, MC-10-20 mL/min and MC-50-20 mL/min. All the
samples showed similar TGA and DSC weight loss profiles with slight variations in the
temperatures and weight losses. Three major TGA weight loss events occurred between
(40 - 250) °C, (350 - 600) °C and (650 - 820) °C at a ramp rate of 20 °C/min ramp and
in an inert (nitrogen) atmosphere. For MC-10-10 mL/min the percent weight losses were:
6%, 7%, and 1%, respectively, for the MC-10-20 mL/min the weight losses were: 5%,
9%, and 1%, respectively while for the MC-50-20 mL/min the losses were: 5%, 8%, and
2%, respectively. Almost similar TGA profiles have been reported and the species
evolved probed using temperature programmed desorption (TPD) for K-OMS-2 materials
synthesized conventionally without DMSO. 32
75
DSC profile showed four thermal events
occurring for these materials with slight variations. These were: endothermic peaks in the
range 50 - 150 °C, and exothermic peaks in the ranges; 200 - 250 °C, 280 - 320 °C and
340-380 °C.
49
(») MC.1t.1t nLMn
(b) MC-1t-2t al/nbi
<o) MC-1t4t nL/ain
HjO
DMSO(S-O)
SO
DMSO(CHj)
3
4,
1400
600
Figure 2-15. FTIR spectra of OMS-2 materials: (a) MC-10-40 mL/min (b) MC-10-20
mL/min and (c) MC-10-10 mL/min. Samples were synthesized at a microwave power of
300 W.
Temperature rump• 2»'C mln
Ktawsfhw* •N, tt mLmiri
96-
94-
Z
s
s
i 90-
86-
200
400
600
800
Temperature (*c/mln)
Figure 2-16. TGA profiles of K-OMS-2 materials prepared continuously at different
flow rates (10 mL/min and 20 mL/min) with different volumes of DMSO (10% and
50%).
50
-0 1
•02 -
MC-KMG ROUNW
MC-50-20 MUMM
MC-10-20 mUrrun
•04
-06-
100
150
TimyiUMW (*C)
300
Figure 2-17. DSC profiles of K-OMS-2 materials prepared continuously at different flow
rates (10 mL/min and 20 mL/min) with different volumes of DMSO (10% and 50%)
51
2.8.6
Potentiometric Titrations
These titrations were performed on sample MC-25-20 mL/min giving a value of 3.9 as
the AOS for manganese in this sample. This value is in agreement with AOS
measurements from other reports.
2.8.7
c-y
Surface Area and Pore Size Distribution
Table 2-4 shows the surface areas of the synthesized materials. The surface area ranges
from 153 to 213 m 2 /g while the representative N2 sorption isotherms are shown in Figure
2-18. Sample MC-5-10 mL/min was used to obtain representative nitrogen sorption data.
52
250
* 200 -
150-
50-
02
0.4
06
R •!<*(«• rnun (M>*)
Figure 2-18. (a) Nitrogen sorption isotherm of MC-5-10 mL/min K-OMS-2 sample and
(b) The Barrett-Joyner-Halenda (BJH) pore size distribution
Table 2-4. BET surface areas (SBET) of K-OMS-2 materials synthesized by microwave
continuous method. a b and c refer to references 69, 75, and 52 respectively
Sample
SBET
MC-5-10 mL/min
176
MC-10-10 mL/min
153
MC-25-10 mL/min
189
MC-25-20 mL/min
191
MC-50-10 mL/min
213
MR-90-25%DMS0 3
227
Conventional reflux b
91
Reflux (1% H202)C
62
±
1 (m 2 /g)
53
2.9
Discussion
2.9.1
Continuous Crystallization of K-OMS-2 Nanomaterials.
The synthesis of K-OMS-2 was achieved using a continuous-flow microwave technique.
The crystallization of K-OMS-2 was obtained under specific conditions of reactant flow
rate, volume of DMSO, and microwave power using a coiled Teflon reactor. The organic
component of the solvent system is responsible for the rapid formation of the K-OMS-2
since in its absence the K-OMS-2 phase is not obtained. This can partly be due to the
excellent coupling of microwaves with DMSO. When KMn04 solution was added to the
MnSO^HhO solution containing
HNO3,
the mixture turned light brown in color
darkening with the addition of more KMn04 solution due to formation of a precipitate.
This further addition resulted in a dark brown mixture, which did not change in color
significantly upon addition of DMSO. The dark brown precipitate is a precursor that
transforms to K-OMS-2 during the synthesis. The conproportionation of Mn 7+ and Mn 2+
is responsible for the formation of ordered K-OMS-2. Portehault et al. 7A that showed
using volumetric titrations that about -99% of KMn04 had already reacted with about
-91% of MnS04 within a very short period of mixing the reactant solutions. At the start
of the experiment, the MnS04 solution is lightly pink, almost clear. Drop-wise addition of
KMn04 solution immediately leads to formation of a brown mixture. This is attributed to
the oxidation of Mn 2+ by the Mn04~. The reduction potentials of Mn02/Mn 2+ and Mn04 _
/Mn 4+ in acidic environments are 1.23 and 1.68 volts respectively
76
and thus support this
premise. However, the final mixture was thin enough to allow flow in a coiled reactor
without clogging the system like would a thick precipitate. A batch microwave system
69
earlier developed explored the types of non-aqueous co-solvents suitable for the synthesis
54
of K-OMS-2. DMSO was identified as a suitable candidate to use in this continuous-flow
process. In making this choice, we relied on tan 5 values in Table 2-1 and the fact that a
similar batch system utilized DMSO as a co-solvent leading to the K-OMS-2 phase in as
little as 10 minutes. 69 Solvents in Table 2-1 can be categorized into three classes namely:
high microwave absorbers (tan 8 >0.5), medium absorbers (0.1 < tan 8 > 0.5) and poor
absorbers tan 8 <0.1.
Water as a standard of comparison has a tan 8 of 0.123 and so is a medium absorber of
microwaves. Water has been used extensively as a medium for many microwave-assisted
syntheses. However, compounds above water in Table 2-1 are good absorbers of
microwave radiation resulting in superheating due to their high tan 8 values. For instance
DMSO with a tan 8 = 0.825 is capable of being superheated.
The penetration depths of microwaves into reaction mixtures are usually limited as shown
in Table 2-2. Due to this, batch scale-up is usually limited in producing multi-kilogram
quantities of material. This process was developed based on the penetration depths of
water, which are limited to 1.4 cm and 5.7 cm at 25 °C and 95 °C, respectively, as shown
in Table 2-2. This means that for any meaningful scale-up in a reaction that utilizes water
as a medium this limitation has to be overcome. Using a material that is almost
transparent to microwaves and having a small diameter can overcome the penetration
limitation of water. The Teflon coiled reactor (<&inner = 1/8") overcomes the depth
limitation and enables microwaves to directly interact with the bulk of the reaction
mixture. A coiled quartz reactor with similar dimensions may also be used, based on its
penetration depth (see Table 2-2).
55
The effect of reactant flow rates that directly impacts the microwave residence time was
investigated by comparing XRD patterns of samples prepared at different flow rates. At
40 mL/min the reactant mixture resides for only 50 seconds in the microwave cavity. The
XRD pattern for this sample reveals a poorly ordered manganese oxide phase (Figure 29). Transformation to a well-ordered crystalline K-OMS-2 phase occurs as the flow rates
is slowed to 10 mL/min and even further slowed at 5 mL/min (Figure 2-9). Slower flow
rates led to longer microwave residence times of 4 minutes at 10 mL/min and 8 minutes
at 5 mL/min.
From the foregoing, transformation to crystalline K-OMS-2 occurs as the residence time
is increased. With longer exposure times to microwaves, there is ample interaction time
between the solvent and the microwaves leading to better coupling and heating of the
reactant
materials. Co-solvent concentration also plays a crucial
role in the
transformation. When no DMSO is added to the synthesis mixture, transformation to KOMS-2 does not take place. When mixed in water, DMSO with tan 8 of 0.825 produces a
solvent system capable of achieving superheating within a short time and is the reason the
transformation to K-OMS-2 occurs even with a volume as low as 5%. Higher
concentrations of DMSO will lead to a better heating system and a 50% DMSO system is
expected to have an even better coupling and hence superheating. Materials prepared
with any volume of DMSO from 5% to 50% led to XRD patterns that showed
consistently broad peaks, an indication of the formation of small crystallites.
Microwave exposure time, power, and percent volume of DMSO act together to realize
crystallization of K-OMS-2. At a faster flow rate of 20 mL/min the reaction requires
more DMSO of up-to 25% to obtain a much purer phase of K-OMS-2 as compared to
56
only about 10% at 10 mL/min. The shorter residence times contribute to shorter fiber
lengths by restricting the crystal growth time and events that lead to lengthening of the
fibers. The calculated crystallite sizes for the materials synthesized with various volumes
of DMSO and at two different flow rates all show materials with average crystallite sizes
in the 1.6 - 1.8 nm range. The exposure times under continuous-flow conditions are
believed to achieve a temperature not more than 100 °C.
Small batch reactions were performed to study reaction temperatures. When power is
increased from 150 W to 600 W, there is an increase in temperature from about 85 °C to
100 °C (see Figure 2-11). The temperature increases directly with power. Kinetically, the
temperature increase has an effect on the reaction rate based on the Arrhenius Equation
(Equation 2-3). At 100 °C, the rate is expected to be higher than at 85 °C. As can be seen
from Figure 2-11, at 150 W a poorly ordered manganese oxide phase is obtained which
transforms to the correct phase at 300 W while an increase to 450 W and 600 W leads to
impurity phases. The temperature measured by the thermocouple is the bulk temperature
(Tb)
and not the instantaneous temperature
(Tj)
resulting from the direct coupling of
microwave field and solvent's dipoles. The instantaneous temperature should be much
higher than the bulk temperature hence leading to faster reaction rates. At 300 W an
optimal instantaneous temperature is realized to propel the reactant species to the
transition state residing there for an optimal time period and then transforming to the
correct phase. At 150 W the instantaneous temperature is less than optimal and thus does
not achieve this result while at 450 W and 600 W the heating is too rapid and propels the
reacting species to the transition state too rapidly leading to side and incomplete reactions
resulting in impurities.
57
A coiled reactor led to synthesis of a pure K-OMS-2 phase. With a non-coiled reactor
reactant flow through the microwave cavity was too fast since the coiling helped increase
the residence time. A non-coiled reactor failed to lead to the formation of pure K-OMS-2
at the chosen flow rates. The samples that did not transform to pure K-OMS-2 phase had
impurity phase that could be traced to y-Mn02.
2.9.2
Morphology and Crystal Structure.
The synthesized samples showed a homogeneous fibrous morphology typical of
cryptomelane-type
K-OMS-2
materials.
However,
unlike
K-OMS-2
materials
synthesized conventionally the morphology of these materials revealed much narrower
fiber diameters. FESEM and HRTEM images of materials synthesized via this
continuous-flow route confirmed this feature. TEM shows the presence of nanofibers of
varied lengths with lengths as small as about 20 nm with diameters of about 5 nm on
average (Figure 2-14). Such ultra-fine nanofibers contribute to the large surface area of
these samples as compared to K-OMS-2 prepared under conventional reflux. The
observation of lattice fringes also confirms the good crystallinity of this material.
When no DMSO is added to the synthesis mixture transformation to K-OMS-2 is
hampered. FESEM of this material showed a largely amorphous phase of manganese
oxide. Transformation to a well-ordered K-OMS-2 and eventually to crystalline K-OMS2 was tracked using X-ray diffraction eventually forming K-OMS-2 with small fibers as
is evident in FESEM images (Figure 2-13).
58
2.9.3
Mixed valency, Thermal stability, and Textural properties
Potentiometric titrations of the synthesized material reveal the existence of a mixed
valent framework. Average oxidation state of 3.9 points to the existence of manganese in
a mixture of oxidation states (+3 and +4). The mixed valent nature of these materials has
been studied before.
67
TGA and DSC results reveal that K-OMS-2 synthesized by the continuous-flow
technique compares well to the K-OMS-2 achieved via conventional means.
75
The first
TGA weight loss can be attributed to physically adsorbed water being desorbed from the
surface of the material. The second weight loss can be attributed to evolution of
chemically adsorbed water from the materials. The last weight loss can be assigned to
evolution of structural oxygen from the materials. Carbon compounds from the presence
of DMSO could also be evolved in these TGAs. The first peak of the DSC could be due
to desorption of water from the materials and can possibly be linked to the first weight
loss in the TGA profiles. The first exothermic peak starting at around 200 °C could be as
a result of physisorbed DMSO being driven off the materials. The last two events could
be as a result of decomposition of other compounds from the material e.g. residual carbon
compounds or species. The exact nature of or the species evolved and their evolution
temperatures in this synthesis technique are subjects of further study using temperature
programmed desorption (TPD) coupled with mass spectrometry (MS) techniques.
The Brunauer-Emmett-Teller (BET) surface area of the samples are on average more than
an order of magnitude (x 2) higher than the average obtained for samples synthesized
under conventional reflux techniques 15 and on average similar to K-OMS-2 synthesized
via microwave reflux (batch) using a similar DMSO co-solvent system.
69
Based on
59
IUPAC classification, the representative sample MC-5-10 mL/min followed type II
adsorption isotherms where a monolayer of N2 molecules prevail at low relative pressures
(P/P 0 ) and capillary condensation at high P/P„. The sorption isotherms are similar to those
59
obtained for materials synthesized through either conventional reflux " or microwave
reflux. 69 The average BJH pore diameter are on average less than 10 nm.
FTIR spectra showed bands typical of K-OMS-2 and also revealed the presence of some
extra bands. These bands can be attributed to: the S-0 stretch (~ 1120 cm" 1 ), the S=0
stretch (-1050 cm" 1 ) and the DMSO -CH3 rocking frequency (-950 cm" 1 ).
tunnel water species
74
can be observed at 1633 cm" 1 . Nyutu
69
77
K-OMS-2
observed similar bands at
higher DMSO levels. These bands could be attributed to residual DMSO on the surface
of the synthesized samples. The presence of the DMSO bands on the solid samples
indicates residual DMSO either physisorbed or chemisorbed. These data also suggest a
secondary role played by DMSO, which could be a direct role in fiber growth processes.
2.9.4
Continuous synthesis of y-MnOj
The continuous synthesis of K-OMS-2 nanomaterials was achieved successfully by
utilizing a microwave-absorbing media as can be seen from the above results. This
finding provided an impetus to explore the continuous synthesis of y-Mn02 using a
similar route. In the synthesis of y-Mn02 the synthesis recipe chosen proved to be very
• 78
important. When the synthesis was done using a reported recipe, the synthesis was
unsuccessful. This recipe involved use of manganese sulfate and potassium persulfate in
an acidic environment
(H2SO4).
Even the addition of microwave absorbing solvents
(DMSO) did not result in the crystallization of any phase. The clear liquid that was fed
into the reactor led to emergence of a clear product without any crystallization occurring.
60
In the reported synthesis,
78
the persulfate ions are able to photolyze in the presence of
UV radiation under acidic conditions creating powerful oxidizing radical species. Sulfate
radicals oxidized Mn 2+ to Mn 4+ and thus crystallizing an MnC >2 phase (y-MnCh) in the
process. The use of microwaves was futile pointing to the limited role microwaves play
in the creation of radical species. This can be explained by the energy involved.
Microwave radiation is less energetic than UV radiation and thus cannot be able to cleave
the chemical bonds involved to create free radicals but UV achieves this easily.
Switching the recipe to a similar one used in the literature 72 led to precipitation of yMnC >2. However, the approach was a little different here. This approach involved varying
the Mn 2 7Mn 7+ ratios. When the recipe was exactly 1.4 OMS-2 (a-Mn02 phase) was
obtained. Adjusting the recipe to 2.8 achieved by cutting the amount of Mn 4+ in the first
synthesis by half resulted in materials whose XRD patterns shown in Figure 2-19. The
phase was still largely an a-MnCh phase.
When this ratio was re-adjusted to 2.3 by slightly increasing the amount of Mn 7+ , the
XRD patterns in Figure 2-20 were obtained. The different flow rates that were studied
could crystallize the desired phase. Based on the (110) reflection, the widening of this
reflection relative to the other reveals that the crystallites reduce in size. The width of this
reflection is increasing (a-c) as the flow rate is increased. When the same recipe is left at
85 °C for 24 hours under conventional heating, the materials with the pattern in Figure 220 (d) are obtained. This pattern shows emergence of some foreign peaks not present in
the MW-prepared patterns. These could represent an impurity phase in the y-MnCh
material with a high possibility of being a-Mn0 2 . Picking a flow rate of 10 mL/min,
effect of DMSO on the synthesis was studied (Figure 2-21). However, DMSO might not
61
be playing as dominant a role in the crystallization of y-MnCh as in the case of
crystallization of OMS-2 since y-Mn02 could be crystallized even in the absence of
DMSO (Figure 2-21 (a)). However, the contribution of DMSO becomes apparent when
we look at the SEM micrograph images (Figure 2-22) of materials prepared with various
amounts of DMSO. Here, the materials prepared with higher amounts of DMSO all lead
to nanoparticles (a and b) while reducing the DMSO amount leads to spheres with
flower-like morphology (Figure 2-22 (c)). We were also interested to see whether this
recipe can lead to crystallization of y-Mn02 phase in the absence of microwaves. For this,
after mixing all the reagents, synthesis mixture (without DMSO) was left at room
temperature and samples collected at 2 and 24 h. Thereafter, XRD and SEM were
performed on these samples after wash and dry procedures. The results are shown in
Figures 2-23 and 2-24 respectively. At both times crystallization of the correct phase is
observed without any impurities (Figure 2-23). Also, the phase does not change
significantly as can be seen in the SEM images. Compared to the materials synthesized
under MW, differences in the SEM images are noted, which points to the role played by
both microwaves and DMSO in the synthesis. From this study other than providing a
microwave-absorbing environment, DMSO also plays a direct role in controlling the
morphology of the materials synthesized. DMSO provides an environment for nucleating
species to individually crystallize thus forming individual nanoparticles rather than
aggregating to form spheres. This could be due to nucleating species being bound by
DMSO and the resultant environment created repulsing other nucleating species from
agglomerating, purely a surface phenomenon.
62
Thermal stability of the synthesized materials was studied by comparing their
Thermogravimetric analysis (TGA) profiles. These are shown in Figures 2-25 and 2-26.
As can be seen in both Figures all the TGA profiles follow a similar trend for all the
materials studied. The materials lose in total 15% of its original weight when the TGAs
are done in a nitrogen environment. All the materials had 3 weight loss profiles: 5% to
250 °C, 8% between (250 - 550) °C, and 2% between 550 - 800) °C.
Nitrogen sorption isotherms reveal type II adsorption isotherms (Figure 2-27) with
corresponding surface area of 112 m 2 /g.
63
(») •% DMSO, 11 niLiain
ft) 1% BMJO, «• •*/•*»
«
Figure 2-19. XRD patterns of materials continuously synthesized at Mn 2 VMn 7+ = 2.8,
with 10 mL nitric acid, (a) without and (b) with DMSO
I
20
40
Two TtMKa (d*d)
Figure 2-20. XRD patterns of y-MnCh continuously synthesized at different flows: (a) 5
mL/min, (b) 10 mL/min and (c) 20 mL/min, constant power of 300 W and 25% DMSO,
compared to materials made by (d) oil bath treatment of the same recipe for 24 h at 85 °C
with impurity phase shown as (*).
64
I
Jl
A
80
Figure 2-21. XRD patterns of y-Mn02 continuously synthesized with different volumes
of DMSO: (a) 0%, (b) 1%, and (c) 25%, at constant power of 300 W and flow rate of 10
mL/min compared to materials made by oil bath treatment of the same recipe for 24 h at
85 °C (d) with impurity peaks shown as (*).
(a)
(b)
(c)
Figure 2-22. SEM micrographs of y-Mn02 continuously synthesized at (a) 20 mL/min,
25% DMSO and (b) 10 mL/min, 25% DMSO and(c) 10 mL/min, 1% DMSO
65
40
m
m
Figure 2-23. XRD patterns of y-Mn02 synthesized with Mn 2+ /Mn 7+ = 2.3 left at room
temperature for (a) 2 and (b) 24 hrs
(c)
<d)
Figure 2-24. SEM micrographs of y -MnC^ room temperature synthesized for: (a) 2 h
and (c) and 24 h and their higher magnifications (b) and (d) respectively.
66
100 —f
Nitrogen atmosphere (60 seem)
—- 25% DMSO-2G mL/min
- - 25% OMSO -10 mL/min
— - 25% DMSO-5mUmin
0% DMSO, CW bath 24ft
i
i
J 90-
200
600
400
BOO
Figure 2-25. TGA patterns of X -Mn02 continuously synthesized at different flow rates
at constant power of 300 W and compared to materials made by oil bath treatment of the
same recipe for 24 h at 85 °C
100
Nitrogen almmptme (60 teemt
mDMS0-1OmU*»sn
1% 0MS0-1O mUmm
24h
0%M»SCMDll
«5 -
i
U
85-
200
600
TmpmMM (*C)
Figure 2-26. TGA patterns of Y -Mn02 continuously synthesized with different volumes
of DMSO at constant power of 300 W and compared to materials made by oil bath
treatment of the same recipe for 24 h at 85 °C
67
200
*i' 'I "I "I ' 1
1
< T
r
|i i
0.2
i
I | % i
r " f " j "T r i i t"'i
04
r
< i [
i
0.6
i
K*tiNv« Pratsiii* (WP,)
t i
; r ' r ' r i ' j ' i
it
r
i
08
Figure 2-27. Nitrogen sorption material synthesized with 1% DMSO at 10 mL/min. The
surface area of the material was 112 m /g
68
2.10
Catalytic application of synthesized materials
2.10.1 Catalytic oxidation of 2,3»6-Trimethyl phenol
The synthesized K-OMS-2 and y-Mn0 2 materials were used in the catalytic oxidation of
2,3,6-trimethylphenol in the presence of TBHP and vapor phase oxidation of toluene
respectively. The oxidation of TMP to 2,3,6-trimethyl benzoquinone (TMQ) is important
due to its use as an intermediate in the synthesis of vitamin E. The synthesized
nanomaterials showed excellent activity in the oxidation of TMP with 100% selectivity to
the desired product. These results are displayed in Table 2-5.
The K-OMS-2 materials performed exemplarily in the catalytic oxidation of TMP to
TMQ. As shown in Table 2-5, all tested samples showed excellent conversions and
selectivity to the desired product. The catalytic role of the material was confirmed by
using a blank reaction (absence of catalyst) with 2 mmol of TBHP that resulted in trace
amounts of TMQ. With 1 mmol TBHP the conversions were lower than using twice as
much TBHP. Samples with either 5% DMSO or 10% DMSO but synthesized at lower
flow rates (5 and 10 mL/min) showed high activity (conversion of 100%) with
concomitant high selectivity (100%) and turnover number of 16. By comparison, TMP
oxidation is achieved using various materials/catalysts. For instance, using copper (II)
chloride in ionic liquids and 10 bar oxygen pressure yielded 98% TMQ.
70
•
•
Ti02-Si02
aerogel using H2O2 as oxidant achieves a conversion of 94% with a TMQ yield of 89%
after 1 h. 80 Based on the foregoing, the oxidation of TMP using the synthesized catalyst
represents a milder and cheaper approach by comparison and still achieves similar
results.
69
2.10.2 Vapor phase oxidation of Toluene
Gamma manganese oxide synthesized continuously was used in the oxidation of toluene
in the gas phase. Table 2-6 lists the results conducted at various temperatures. The gas
phase oxidation of toluene in the gas phase does not lead to significant yields of
benzaldehyde at the temperature regimes studied.
70
Table 2-5. Oxidation of TMP to TMQ using continuously synthesized K-OMS-2
nanofibers.
Catalyst
Conversion
Selectivity
(%)
(%)
TON
Blank
Trace
MC-5-10 mL/min
58
82
8
90 b
100 b
14
48
75
6
MC-10-5 mL/min
MC-10-10 mL/min
47
MC-10-20 mL/min
MC-50-20 mL/min
73
c_
O
O
86 b
o
o
I00 b
16
5
14
46
85
6
83 b
94 b
13
40
66
4
91 b
96 b
14
Quantification was based on GC-MS peak areas. Catalytic tests
performed in the presence of 1 mmol of TBHP and 2 mmol b . Turnover
number (TON) calculated by: (moles of product
moles of catalyst).
Blank reaction was carried out in the absence of catalyst and with 2 mmol
of TBHP. (-) were not determined due to trace amount of product.
71
Table 2-6. Results of the vapor phase oxidation of Toluene using gamma manganese
oxide synthesized via a microwave continuous process.
Catalyst
Temperature
Conv.
Selectivity (%)
(°C)
(%)
a
b
c
d
y -MnCh
120
1
100
0
0
0
y-Mn02
150
1
100
0
0
0
y-MnCh
200
1
100
0
0
0
y-MnCh
450
2
44
21
12
23
Calculations based on liquid products only. Online GC detected significant CO2 levels but
were not quantified due to small liquid phase quantification. Selectivity refers to: (a)
benzaldehyde (b) phenol (c) acetophenone (d) bibenzyl
-Ea
k = A.ex p RT
Equation 3-3. Arrhenius Equation where k is the rate constant, T is the absolute
temperature, E a is the activation energy, R is the gas constant, and A is the preexponential factor.
72
2.11
Conclusions
Microwave continuous synthesis of inorganic materials is a research area that is yet to be
explored to full potential. However, the unique physical properties such as smaller
particle sizes and high surface areas exhibited by materials synthesized under microwave
continuous techniques that could also be a pointer to the potential uniqueness in chemical
properties provides a much needed impetus to the furtherance of this technology.
Microwave continuous synthesis offers several advantages that microwave batch
synthesis lack such as; opportunity for scaled synthesis where the only impediment is the
size of both the feeding and receiving vessels, and more rapid heating due to the smaller
sizes of reaction vessels (tubes) as opposed to lack of microwave penetration in larger
batch vessels.
With examples, we have clearly reviewed the types of microwave apparatus used in
continuous synthesis as well as given an overview of the types of materials synthesized
using microwave continuous synthesis technique. Microwave continuous synthesis is still
an emerging technology that is slowly gaining use in the synthesis of inorganic
nanomaterials. This technology has successfully been employed in the synthesis of
microporous and mesoporous aluminosilicate nanomaterials, metal oxides and mixed
metal oxide systems, as well as metallic systems. These syntheses have been
accomplished using commercially available as well as custom-made apparatus. Both
types of apparatus require a means of introducing the reactant materials into the
microwave cavity. This is achieved using pumping devices e.g. peristaltic pumps.
The only drawback to this technology may come in the form of nanomaterials that require
extreme reaction conditions to crystallize. For instance, nanomaterials that requires long
73
ageing periods for nucleation and crystallization and will take longer than say about six
hours to crystallize in a microwave environment may not benefit from continuous
synthesis.
Those nanomaterials that require extreme pressures may also offer a
challenge, not forgetting nanomaterials that tend to solidify (form thick gels) in the
course of the reaction. If the above challenges are overcome, then there is great potential
in using microwave continuous techniques in nanomaterials synthesis.
A continuous flow microwave technique has been developed for the synthesis of KOMS-2 nanofibers. Process parameters such as flow rates, microwave power, volume of
co-solvent (DMSO), and reactor type are crucial parameters for the final outcome. Slow
flow rates (10 mL/min and 5 mL/min) led to crystallization of the K-OMS-2 phase with
even lower volumes of DMSO (e.g. 5 %). At higher flow rates (20 mL/min and higher)
the crystallization of K-OMS-2 into a pure phase is also partly dependent on the amount
of DMSO present. Lower volumes of DMSO at higher flow rates resulted in an impurity
phase (y-Mn02). High microwave power levels (450 W and over) are not ideal, the
optimal level for this synthesis is about 300 W. The synthesized materials performed well
in the oxidation of TMP to TMQ achieving high conversions (100%) and selectivity
(100%). With this technique, the synthesis of K-OMS-2 on a multi-kilogram level is
possible since the only limitation to the synthesis is the size of the reagent reservoir.
When a small reservoir is used then the reservoir replenishment rate is the limiting factor.
Similarly, microwave continuous synthesis successfully yielded y-Mn02. From this study
it was concluded that other than providing a microwave-absorbing environment, DMSO
also plays a role in controlling the morphology of the materials synthesized. DMSO
provides an environment for nucleating species to individually crystallize thus forming
74
individual nanoparticles rather than aggregating to form spheres. This could be due to
nucleating species being bound by DMSO and the resultant environment created
repulsing other nucleating species from aggregating, purely a surface phenomenon.
75
2.12
References
1.
C. N. R. Rao, Chemical Approaches to the synthesis of inorganic materials,
Wiley Eastern Limited, New Delhi, 1994.
J. Bohnemann, R. Libanori, M. L. Moreira and E. Longo, Chem. Eng. J.
(Amsterdam, Neth.), 2009, 155, 905-909.
S. M. Garringer, A. J. Hesse, J. R. Magers, K. R. Pugh, S. A. O'Reilly and A. M.
Wilson, Organometallies. 2009, 28, 6841-6844.
M. Gharibeh, G. A. Tompsett, W. C. Conner and K. S. Yngvesson,
ChemPhysChem, 2008, 9, 2580-2591.
N. L. Campbell, R. Clowes, L. K. Ritchie and A. 1. Cooper, Chem. Mater., 2009,
21,204-206.
A. Dandia, R. Singh, S. Khaturia, C. Merienne, G. Morgant and A. Loupy,
Bioorg. Med. Chem., 2006, 14, 2409-2417.
R. Gedye, F. Smith, K. Westaway, H. Ali, L. Baldisera, L. Laberge and J.
Rousell, Tetrahedron Lett., 1986, 27, 279-282.
R. J. Giguere, T. L. Bray, S. M. Duncan and G. Majetich, Tetrahedron Lett.,
1986, 27, 4945-4948.
A. K. Bose, M. S. Manhas, B. K. Banik and E. W. Robb, Res. Chem. Intermedial.,
1994, 20, 1-11.
C. Strauss and R. Trainor, Aust. J. Chem., 1995, 48, 1665-1692.
L. Perreux and A. Loupy, Tetrahedron, 2001, 57, 9199-9223.
C. O. Kappe, Angew. Chem., Int. Ed., 2004, 43, 6250-6284.
A. de la Hoz, A. Diaz-Ortiz and A. Moreno, Chem.l Soc. Reviews, 2005, 34, 164178.
O. C. Kappe, CHIMIA International Journal for Chemistry, 2006, 60, 308-312.
C. Kappe and D. Dallinger, Mol. Divers., 2009, 13, 71-193.
C. O. Kappe, D. Dallinger and S. S. Murphree, in Practical Microwave Synthesis
for Organic Chemists, Wiley-VCH Verlag GmbH & Co. KGaA, 2009, pp. 11-44.
S. Komarneni, R. Roy and Q. H. Li, Mater. Res. Bull., 1992, 27, 1393-1405.
S. Komarneni, Q.H.Li and R. Roy, J. Mater. Chem., 1994, 4, 1903-1906.
S. Komarneni, R. Pidugu, Q. H. Li and R. Roy, J. Mater. Res., 1995, 10, 16871692.
Q. Li and Y. Wei, Mater. Res. Bull., 1998, 33, 779-782.
Y.-w. Jun, J.-s. Choi and J. Cheon, Angew. Chem., Int. Ed., 2006, 45, 3414-3439.
H. Qian, X. Qiu, L. Li and J. Ren, J. Phys. Chem. B, 2006, 110, 9034-9040.
Y. He, H.-T. Lu, L.-M. Sai, W.-Y. Lai, Q.-L. Fan, L.-H. Wang and W. Huang, J.
Phys. Chem. B, 2006, 110, 13352-13356.
J. Ziegler, A. Merkulov, M. Grabolle, U. Resch-Genger and T. Nann, Langmuir,
2007, 23,7751-7759.
Y. Zhao, J. M. Hong and J. J. Zhu, J. Cryst. Growth , 2004, 270, 438-445.
H. Qian, L. Li and J. Ren, Mater. Res. Bull., 2005, 40, 1726-1736.
V. Polshettiwar, M. N. Nadagouda and R. S. Varma, Aust. J. Chem., 2009, 62, 1626.
A. G. Whittaker and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 2000, 15211526.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
76
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
A. G. Whittaker and D. M. P. Mingos, J. Chem. Soc., Dalton Trans., 2002, 39673970.
Y. Luo, Mater. Lett., 2007, 61, 1873-1875.
S. Kundu, L. Peng and H. Liang, Inorg. Chem., 2008, 47, 6344-6352.
C. R. Patra, G. Alexandra, S. Patra, D. S. Jacob, A. Gedanken, A. Landau and Y.
Gofer, New J. Chem., 2005, 29, 733-739.
G. Buhler and C. Feldmann, Angew. Chem., Int. Ed., 2006, 45, 4864-4867.
S. H. Jhung, J. W. Yoon, J.-S. Hwang, A. K. Cheetham and J.-S. Chang, Chem.
Mater., 2005, 17,4455-4460.
G. A. Tompsett, W. C. Conner and K. S. Yngvesson, ChemPhysChem, 2006, 7,
296-319.
A. Arafat, J. C. Jansen, A. R. Ebaid and H. van Bekkum, Zeolites, 1993, 13, 162165.
M. Nuchter, B. Ondruschka, W. Bonrath and A. Gum, Green Chem., 2004, 6,
128-141.
D. M. P. Mingos and D. R. Baghurst, Chem. Soc. Reviews, 1991, 20, 1-47.
C. Gabriel, S. Gabriel, E. H. Grant, B. S. J. Halstead and D. M. P. Mingos, Chem.
Soc. Rev., 1998, 27,213-224.
B. L. Hayes, Aldrichimica Acta, 2004, 37, 66-77.
G. Jas and A. Kirschning, Chem.--Eur. J., 2003, 9, 5708-5723.
A. S. Vanetsev and Y. D. Tretyakov, Russ. Chem. Reviews, 2007, 76, 397.
I. Bilecka and M. Niederberger, Nanoscale, 2010, 2, 1358-1374.
X. Z. Lin, A. D. Terepka and H. Yang, Nano Lett., 2004, 4, 2227-2232.
J. Wagner and J. M. KVShler, Nano Lett., 2005, 5, 685-691.
F. Bondioli, A. Bonamartini Corradi, A. M. Ferrari and C. Leonelli, J. Amer.
Ceram. Soc., 2008, 91, 3746-3748.
V. Subramanian, W. W. Burke, H. Zhu and B. Wei, J. Phys. Chem. C, 2008, 112,
4550-4556.
L. Espinal, K. A. Malinger, A. E. Espinal, A. M. Gaffney and S. L. Suib, Adv.
Funct. Mater., 2007, 17, 2572-2579.
E. K. Nyutu, W. C. Conner, S. M. Auerbach, C. H. Chen and S. L. Suib, J. Phys.
Chem.C, 2008, 112, 1407-1414.
L. Bonaccorsi and E. Proverbio, Micropor. Mesopor. Mater., 2008, 112, 481-493.
S. L. Brock, N. Duan, Z. R. Tian, O. Giraldo, H. Zhou and S. L. Suib, Chem.
Mater., 1998. 10, 2619-2628.
J. C. Villegas, L. J. Garces, S. Gomez, J. P. Durand and S. L. Suib, Chem. Mater.,
2005, 17, 1910-1918.
J. E. Post, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 3447-3454.
J. E. Post and C. W. Burnham, ,4m. Mineral., 1986, 71, 1178-1185.
Q. Feng, H. Kanoh, Y. Miyai and K. Ooi, Chem. Mater., 1995, 7, 148-153.
R. N. DeGuzman, Y.-F. Shen, E. J. Neth, S. L. Suib, C.-L. O'Young, S. Levine
and J. M. Newsam, Chem. Mater., 1994, 6, 815-821.
Y. F. Shen, R. P. Zerger, R. N. DeGuzman, S. L. Suib, L. McCurdy, D. 1. Potter
and C. L. O'Young, Science (Washington, D. C, 1883-), 1993, 260, 511-515.
Y.-C. Son, V. D. Makwana, A. R. Howell and S. L. Suib, Angew. Chem., Int. Ed.,
2001,40, 4280-4283.
77
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
N. N. Opembe, Y.-C. Son, T. Sriskandakumar and S. L. Suib, ChemSusChem,
2008, 1, 182-185.
R. Kumar, S. Sithambaram and S. L. Suib, J. Catal., 2009, 262, 304-313.
R. Ghosh, Y.-C. Son, V. D. Makwana and S. L. Suib, J. Catal., 2004, 224, 288296.
R. Kumar, L. J. Garces, Y.-C. Son, S. L. Suib and R. E. Malz, J. Catal., 2005,
236,387-391.
T. Sriskandakumar, N. Opembe, C.-H. Chen, A. Morey, C. King'ondu and S. L.
Suib, J. Phys. Chem. A, 2009, 113, 1523-1530.
L. Jin, J. Reutenauer, N. Opembe. M. Lai, D. J. Martenak, S. Han and S. L. Suib,
ChemCatChem, 2009, 1, 441-444.
US5340562A, 1994.
J. Yuan, K. Laubernds, J. Villegas, S. Gomez and S. L. Suib, Adv. Mater.
(Weinheim, Ger.), 2004, 16, 1729-1732.
J. E. Post, R. B. VonDreele and P. R. Buseck, Acta Crystallogr., Sect. B, 1982,
B38, 1056-1065.
X. Chen, Y.-F. Shen, S. L. Suib and C. L. O'Young, J. Catal., 2001, 197, 292302.
E. K. Nyutu, C.-H. Chen, S. Sithambaram, V. M. B. Crisostomo and S. L. Suib,,/.
Phys. Chem. C, 2008, 112, 6786-6793.
V. M. B. Crisostomo, University of Connecticut Ph.D., University of Connecticut,
2008.
P. de, Acta Crystallographica, 1959, 12, 341-345.
L. Jin, C.-h. Chen, V. M. B. Crisostomo, L. Xu, Y.-C. Son and S. L. Suib, Applied
Catalysis A: General, 2009, 355, 169-175.
R. M. Potter and G. R. Rossman, Am. Mineral., 1979, 64, 1199-1218.
T. Gao, M. Glerup, F. Krumeich, R. Nesper, H. Fjellvag and P. Norby, J. Phys.
Chem. C, 2008, 112, 13134-13140.
Y.-S. Ding, X.-F. Shen, S. Sithambaram, S. Gomez, R. Kumar, V. M. B.
Crisostomo, S. L. Suib and M. Aindow, Chem. Mater., 2005, 17, 5382-5389.
D. Portehault, S. Cassaignon, E. Baudrin and J.-P. Jolivet, Chem. Mater., 2007,
19, 5410-5417.
R. Elbaum, S. Vega and G. Hodes. Chem. Mater., 2001, 13, 2272-2280.
C. K. King'ondu, A. Iyer, E. C. Njagi, N. Opembe, H. Genuino, H. Huang, R. A.
Ristau and S. L. Suib, J. Am. Chem. Soc, 2011, 133, 4186-4189.
H. Sun, K. Harms and J. Sundermeyer, J. Am. Chem. Soc., 2004, 126, 9550-9551.
N. N. Trukhan and O. A. Kholdeeva, Kinet. Catal., 2003, 44, 347-352.
Kumar, V., Kim, J. H., Pendyala, C., Chernomordik, B., Sunkara M. K. (2008)
J. Phys. Chem. C, 112, 17750
78
CHAPTER 3.
OXIDATION
KINETICS
CATALYZED
AND
BY
MECHANISM
MANGANESE
OF
9H-FLUORENE
OXIDE
OCTAHEDRAL
MOLECULAR SIEVES
3.1
Introduction
Catalysts are substances that increase the rate of a chemical reaction. However, product
selectivity, kinetics and mechanism of reactions on catalyst surfaces provide further
information vital to a deeper understanding of the catalytic process. Catalysts are the
"workhorses" of the chemical industry - more than 80% of all modern chemicals come
into contact with at least one catalyst during their manufacture.
1
Catalysis accounts for
about a quarter of the world's gross domestic products.
9H-Fluorene falls under the polycyclic alkyl-arene family of compounds, the oxidation of
which has elicited much attention lately. For instance, polycene derivatives can be
applied to organic semiconductors and transistors. 2-4 However, many oxidation reactions
have relied on stoichiometric amounts of Cr vl , Mn v ", and Os vl " that generate large
quantities of toxic metal wastes
5 6
and are performed under relatively forced conditions.
Additionally, from the standpoint of environmental friendliness, economy, and atom
efficiency, attention has shifted to the development of methods that employ clean
oxidants such as molecular oxygen.
7-9
9-Fluorenone is an important compound that is finding promising uses as a component in
organic solar cells and display devices. This material also represents an interesting class
of compounds for biomedical applications.
i0 11
9-Fluorenone has been synthesized by
the oxidation of 9H-fluorene in the presence of alkali-metal hydroxides, by using triton
B as a catalyst in the presence of pyridine,
12
and by using oxygen in the presence of
79
cobalt bromide.
13
Laboratory-scale preparation involves palladium-catalyzed cyclization
of o-iodobenzophenones
14
and carbonylation of o-halobiaryls.
The most useful syntheses of 9H-fluorenone include: Friedel-Crafts closures of
biarylcarboxylic acids and their derivatives,
reactions of conjugated enynes,
biphenyloxazolines.
17
16
15
intramolecular [4 + 2] cycloaddition
remote metalation of 2-biphenylcarboxyamides or 2-
Oxidation of 9H-Fluorene using Ru (III) montmorillonite K 10
with tert-butylhydroperoxide (TBHP) as the oxidant has also been performed.
18
The use of palladium as catalysts in palladium catalyzed methods is strongly dependent
on use of ligands and specifically on the nature of these ligands and also utilizes carbon
monoxide (CO) and cesium pivalate, a base taking up-to 7 h of reaction time while
oxidation with the use of Ru (III) montmorillonite K 10 is insignificant in the absence of
TBHP.
Here, we explore the detailed oxidation of 9H-fluorene to 9-fluorenone with manganese
oxide octahedral molecular sieves (OMS-2) as a selective catalyst and using air as an
environmentally friendly oxidant. The oxidation is carried out efficiently without use of
any additives, promoters, or radical initiators, hence representing a benign process.
Emphasis is laid on the kinetics of 9H-fluorene oxidation using the initial rates and
kinetic isotope effect (KIE). We have also effectively identified OMS-2 catalysts that
show excellent performance in this oxidation process and a solvent that is 100 %
selective to the desired product.
OMS-2 is built from six-coordinate metal atoms surrounded by an octahedral array of
anions (oxygen in this case), forming infinite 3-D frameworks with molecule-sized
tunnels similar to those found in naturally occurring Zeolites. OMS-2 materials have been
80
synthesized by use of different synthetic routes and characterized.
19
OMS-2 is mixed
valent with manganese in oxidation states Mn 2+ Mn 3+ , and Mn 4+ . Literature suggests that
only a small amount of Mn 3+ exists and that the majority of manganese is in the Mn 4+
state.
20
Son et al. 21 have recently demonstrated a selective, acid catalyzed alcohol
oxidation using OMS-2. OMS-2 has also been reported in the partial oxidation of butene,
22
2-propanol to acetone and propylene,
23
and partial epoxidation of olefins.
24
Further
work on the alcohol oxidation using OMS-2 pointed towards a Mars-van Krevelen
mechanism where the involvement of the lattice oxygen was implicated as a factor for its
excellent selectivity. 25
Mechanism of heterogeneous catalytic reaction are usually assumed to proceed through
the following categories; the so-called Langmuir-Hinshelwood-Hougen-Watson (LHHW)
26
mechanism in which adsorbed molecules react among themselves; the Rideal-Eley-
type
26
mechanism where a reactant molecule in the gas-phase reacts with another in
adsorbed state and in the case of oxidation reactions, the Mars-van Krevelen
27
or redox
mechanism has widely been accepted where the oxidized metal surface oxidizes the
substrate and is re-oxidized by gas phase O2 in a separate step. The dominance of anyone
of these mechanisms is highly depended on reaction conditions (gas phase or liquid
phase, reaction temperature, pressure), on the nature of the metal and on the oxidizing
agent used. 28 Thomas et al. have provided evidence of the involvement of free radicals in
oxidation reactions. 29
Elucidation of a mechanism in heterogeneous catalysis involves the identification and
characterization of active centers, the activation process of the reactants, surface
reactions, the study of catalyst deactivation and the intermediate species, in addition to
81
Kinetic investigations, in-situ spectroscopic methods give insights to surface reactions
and the nature of intermediate species.
This report details the kinetics of 9H-Fluorene oxidation and of it's deuteriated form
fluorene-d 10 using OMS-2 as catalysts. The kinetics of 9-hydroxyfluorene oxidation have
also been studied and together with surface techniques used to characterize OMS-2
surface, a mechanism for 9H-Fluorene oxidation has been proposed. This is the first
report on oxidation of 9H-Fluorene with OMS-2 under ambient conditions.
3.2
Experimental Section
3.2.1
Reagents
All reagents used were of analytical grade and purchased from Sigma-Aldrich.
Laboratory air acted as the oxidant and was passed through a column of anhydrous
calcium sulfate before being introduced into the reaction set-up.
3.2.2
Catalyst Synthesis
The catalyst used in this oxidation study belongs to a class of synthetic manganese oxides
with a tunnel structure, called octahedral molecular sieves (OMS). Synthetic OMS
catalyst has a one-dimensional tunnel structure formed by a 2 x 2 edge shared Mn0 6
octahedral chains, christened OMS-2 (Figure 2-5). The tunnels have dimensions of 4.6 A
x 4.6
A, and the overall composition is KMn80i6 «H20.
OMS-2 was prepared by reflux method.
30
In this synthesis, 40 mL of 1% hydrogen
peroxide (H2O2) solution was added to a buffer solution made by adding 5mL acetic acid
to a
solution of 5 g potassium acetate in 40 mL distilled de-ionized water (DDW). To
this solution, a solution of 6.5 g potassium permanganate in 15 mL DDW was added
82
drop-wise while stirring. The resulting mixture was refluxed for 24 hrs and the product
filtered, washed thoroughly and dried overnight at 80 °C in air then further dried for 2 h.
in air at 120 °C. The OMS-2 was labeled OMS-2 B .
OMS-2B was ion exchanged to give the H + form. 25 The exchanged material was labeled
fi OMS -2B.
The strategy employed towards this end utilized NH/ followed by
thermolysis to reduce to NH/to H + . OMS-2 was added to a 25 mL solution of 20% (w/v)
solution of ammonium nitrate and stirred at room temperature for 3 h. After the slurry
was allowed to settle, the clear liquid phase was decanted and another portion of 25 mL
ammonium nitrate was added and further stirred for 3 h. The slurry was filtered and
washed thoroughly until neutral to a pH paper. The residue was dried at room
temperature before being calcined at 400 °C in air for 8 h.
For comparison purposes, other methods were employed towards synthesis of OMS-2
materials, and used for fluorine oxidation. They include, synthesis under hydrothermal
conditions (OMS-2H), 31 reflux method (OMS-2R), 32 solvent free method (OMS-2s), 33 and
the constant frequency microwave technique (OMS-2MW)-
34
These materials have also
been characterized. 35
3.3
Characterization Methods
3.3.1
X-ray diffraction
The as-synthesized OMS-2B and other OMS-2 catalysts were characterized using powder Xray diffraction (XRD) methods. A Scintag XDS-2000 diffractometer with a Cu Ka radiation
and a beam voltage of 45 kV and 40 mA of beam current was used. The data were collected
in the 20 range of 5-75 °C. The pattern thus obtained was identified using the JCPDS
83
database and the materials were found to be pure (Figure 3-1). Crystallites sizes were
calculated applying Scherrer Equation (Equation 2-2) to the diffraction peaks at 20 =28.8°
and 49.8°. Instrumental line broadening was corrected using a LaEk standard.
3.3.2
Average oxidation state (AOS)
The average oxidation state of the as-synthesized OMS-2B in the states Mn 2+ , Mn 3+ , and Mn 4 '
was determined by the potentiometric titration method. The catalyst was dissolved in
concentrated hydrochloric acid so as to convert all the manganese into Mn 2 \ and titrated to
Mn 3 ' complex by sodium pyrophosphate versus potassium permanganate. This gave the total
Mn content, based on which AOS was determined by reducing the solid to Mn 2 ' by ferrous
ammonium sulfate and back titrating the excess Fe 2+ by the permanganate standard.
3.3.3
Surface area Measurements
Specific surface area was determined by nitrogen sorption. This was done at 77K with a
Micrometrics ASAP 2010 instrument. The samples had to be degassed first at 120 °C for
slightly more than 2 h. The surface area of the samples was determined by the BrunauerEmmett-Teller (BET) method.
3.3.4
Morphology
The morphology of the catalysts was studied by scanning electron micrographs taken on
a Zeiss DSM 982 Gemini field emission scanning electron microscope (FESEM) with a
Schottky Emitter at an accelerating voltage of 2 kV and a beam current of 1 mA. The
SEM images in Figure 3-2 show a needle-like morphology with particles sizes of a few
tens of nanometers.
84
3.4
Catalytic Oxidation Studies
3.4.1
Oxidation of 9H-Fluorene
The following standard procedure was used for all oxidation reactions. 1 mmol 9HFluorene was dissolved in 15 mL of a solvent and 50 mg (0.5 equiv.) OMS-2 added to the
mixture in a 25 mL round bottomed flask. The flask was immersed in an isothermal (±0.5
°C) paraffin oil bath. To this set-up, air at 1 atm. was introduced at an optimum flow rate.
The reactions were carried out, with vigorous stirring, for up-to 4 h. upon which the
catalyst was filtered off and liquid products analyzed by GC-MS and quantified based on
relative areas under GC and using an internal standard.
3.4.2
Solvent study
The solvent dependency of 9H-Fluorene oxidation in the presence of OMS-2 was studied
using the same reaction set-up as earlier described with the only variable being the type
of solvent used. The following solvents were studied: toluene, iso-octane, acetonitrile,
ethylbenzene, hexane, tetrahydrofuran (THF), octane, and dioxane. These solvents were
used without any further purification.
3.4.3
Kinetics of 9H-Fluorene oxidation
Kinetics studies were performed using a three-necked round-bottomed flask. 1 mmol 9HFluorene was dissolved in 15 mL toluene and 50 mg OMS-2 added. Air was introduced
through one of the necks and the flask immersed in an isothermal paraffin oil bath.
Periodic sampling using 0.3 mL via a gas tight syringefollowed the reaction's progress.
The syringe needle was fitted with an in-line filter to exclude any catalyst particles and
85
the product analyzed and quantified. Time t = 0 was defined as the time just after
immersing the three-necked flask in the isothermal oil bath.
3.4.4
Kinetics of 9H-Fluorene oxidation in presence of carbon tetrachloride.
The same procedure was repeated in 9H-Fluorene oxidation in the presence of an
equivalent amount of carbon tetrachloride
3.4.5
(CCI4).
Kinetic Isotope Effect (K.I.E)
To study the kinetic isotope effect ratio (kn/ko), pseudo first order conditions were
mimicked. Two sets of experiments were performed one using 9H-Fluorene and retaining
the same set-up as the one used in the kinetics experiment with the only difference being
the use of higher catalyst loading (150 mg). The second involved oxidation of fluorenedlO, a deuteriated form of 9H-Fluorene. In both cases, drawing 0.3 mL samples through
the free port on the three-necked flask did periodic samplings.
3.4.6
Oxidation of 9H-Fluorene under inert conditions
For control, oxidation of 9H-Fluorene was done under nitrogen atmosphere instead of air.
Nitrogen was flown in the reaction mixture for 30 min. before initiating the reaction and
maintained for the entire duration of the reaction. The product was analyzed after 4 h and
the filtered catalyst washed, dried and it's structure determined by X-ray diffraction.
3.4.7
Recylability of the catalyst
To test whether OMS-2 could be re-used after a reaction and hence a true catalyst, we
regenerated the catalyst and used this treated material in a second reaction. After the first
oxidation reaction, OMS-2 was filtered and washed several times using methanol. The
86
washed catalyst was dried overnight in the oven at 300 °C. An X-ray diffraction pattern
was taken thereafter in an attempt to find out any change in the structure. The regenerated
catalyst was used in a second oxidation of 9H-Fluorene reaction employing the same
reaction conditions as for as the first oxidation. The product was analyzed and quantified.
3.4.8
9H-Fluorene oxidation in absence of catalyst
In a 25 mL round-bottomed flask, 1 mmol 9H-Fluorene and 15 mL toluene were charged
before air was introduced. The set-up was immersed into an isothermal paraffin oil bath
at the solvent reflux and the reaction carried out at the solvent reflux. The product after 4
h of reaction was filtered and analyzed.
3.4.9
Product identification and analysis
The products were identified using an HP 5970 GC/MS with HP 5890 Series II GC
having a DB-1 column and dimethyl polysiloxane as the stationary phase suitable for
separating non-polar compounds while the HP 5970 GC with a flame ionization detector
had a 20% diphenyl and 80% polysiloxane stationary phase suitable for both non-polar
and polar compounds. Due to both 9-hydroxyfluorene and 9H-fluorenone having the
same retention times, the polar column gave good separation of the two compounds
during the kinetics studies.
3.5
Results
The results of the various oxidation studies are presented in Figures 3-3 to 3-5 and in
Table 3-1.
87
s
I
i
c
0MS-2B
10
ao
30
SO
Two TheU fftotttoM)
SO
7C
Figure 3-1. XRD patterns of K-OMS-2 materials synthesized under different reaction
conditions and methods. OMS-2B is synthesized under acidic buffer, OMS-2-Reg is
regenerated OMS-2, 0MS-2R is reflux method, OMS-2S is solvent free, OMS-2H is
hydrothermal, while OMS-2-MW is under microwave heating.
Figure 3-2. SEM micrographs of OMS-2B materials.
88
IIMI
90
80
70
£ 60
.2 50
- 40
20
10
0
10
0
20
15
30
25
35
Time(min)
Figure 3-3. Kinetics of 9H-fluorene oxidation with octahedral molecular sieves under
different conditions: in isooctane (•), in toluene at 110 °C (x), and in toluene at 80 °C
(A); oxidation of fluorene-dlO in toluene at 110 °C (+); oxidation of 9H-fluorene in
toluene in the presence of 0.2 equiv CCI4 (x) and in toluene under nitrogen atmosphere
(*)•
44»
2ft
O
4
12
16
24
2*
Figure 3-4. Comparison of the selectivities to 9-Fluorenone in toluene (•) and in
isooctane (•).
89
0.5
0.45
..9752
0.4
0.35
•K
0.£
*= 0.9222
0&
ime (min)
Figure 3-5. Kinetic isotope effect (KIE) in the oxidation of 9H-fluorene (kH) and [D10]fluorene (kD) based on data for the oxidation of 9H-fluorene and [D10]- fluorene in
toluene at 110 °C shown in Figure 25. Xo corresponds to the original concentration, and
X represents the concentration at any particular time.
90
OMS-2
OMS-2
HO
HO-
OOH
00
Figure 3-6. Proposed mechanism for the OMS-2-catalyzed oxidation of 9Hfluorene.
91
Table 3-1. Oxidation of 9H-Fluorene under different conditions in the presence of air' al
Entry
OMS-2 (mg)
Solvent
Conversion
Selectivity
(%)|b|
(%)|c|
1
0 ld!
Toluene
0
0
2
50
Isooctane |e]
99
100
3
50
Toluene
98
96
4
50 |fl
Toluene
57
95
5
50
Octane
74
95
6
50
Acetonitrile
12
98
7
50
Tetrahydrofuran
4
98
8
50
Hexane
38
96
9
50
Dioxane
0
0
[a] Reactions were performed under semi-batch conditions at solvent reflux under air
for 4 h and using OMS-2B unless otherwise noted.
[b] Conversions based on relative areas under gas chromatograms with 2,4,7trinitrofluorenone as internal standard, [c] Selectivity to 9-fluorenone. [d] Carried out
in the absence of catalyst, [e] Carried out in isooctane at reflux, [f] Ion-exchanged
HOMS-2 with H + as the tunnel cation was used.
92
Table 3-2. Oxidation of 9H-fluorene catalyzed by various OMS-2 catalysts in the
presence of air.[a]
Entry
OMS-2
AOS|b|
Crystallite
Conversion
size (nm)|c| (%)|d|
Selectivity
("/of1
1
OMS-2 b
3.7
8
98
96
2
OMS-2 r
3.9
18
51
98
3
OMS-2 h
3.7
17
6.3
98
4
OMS-2 s
3.7
10
95
97
5
OMS-2 MW
3.9
18
37
50
[a] Reactions were performed under semi-batch conditions in toluene at reflux under
air for 4 h. OMS-2 subscripts: B refers to reflux under acidic buffer conditions, R
refers to acidic reflux method, H refers to hydrothermal methods, S refers to solventfree, and W refers to microwave field, [b] Average oxidation state determined by
potentiometric titrations, [c] Crystallite sizes calculated with the Scherrer Equation,
[d] Conversions calculated based on GC areas and quantified using an internal
standard, [e] Selectivity to the desired product.
93
3.6
Discussion
The oxidation of 9H-fluorene goes to completion in 4 h. 9H-fluorene is not oxidized in
the absence of OMS-2, implying that OMS-2 catalyzes this reaction (Table 3-1, entry
1). The best solvent for the reaction was also studied. Having established that 9Hfluorene oxidation in toluene led to the formation of 9,9'-bifluorene, a side product, at
selectivities of 4 and 5 % (Table 3-1, entries 3 and 4), other solvents were chosen to
study the same reaction and determine their effect on selectivity. Non-polar solvents led
to better conversions than polar solvents. These solvents are either aliphatic or aromatic
hydrocarbons, and ail of them led to formation of a side product except isooctane, a
branched aliphatic that was 100 % selective to 9-fluorenone at a lower temperature.
Isooctane suppressed the intermediates that lead to formation of 9,9'-bifluorene as a side
product by not interacting adversely with the intermediate species, thus availing them to
the active sites of the catalyst for completion of the oxidation cycle. However, other noncoordinating solvents such as octane led to formation of 9,9'-bifluorene. Thus, isooctane
behaves quite uniquely in this reaction. Branched hydrocarbons are more stable than
unbranched hydrocarbons as a result of the interplay between attractive and repulsive
forces within the molecule (intramolecular forces). As a further test of this hypothesis,
2,4-dimethylpentane and 2,3,4-trimethylpentane were also used in fluorene oxidation.
2,3,4-trimethylpentane was 100 % selective, however, 2,4-dimethylpentane led to
formation of the side product. Not only is branching important but so is the number of
branches. Attractive forces within a branched hydrocarbon increase as the structure
becomes more compact. The proton-exchanged catalyst, H-K-OMS-2, led to a lower
94
conversion of 9H-fluorene (Table 3-1, entry 4), thus indicating that the reaction is
inhibited by the presence of Bransted acid sites.
The OMS-2 catalyst was washed thoroughly with methanol and dried following the
oxidation reaction. The X-ray diffraction pattern of the regenerated catalyst revealed that
the OMS-2 structure was maintained after reaction, indicating that it was not a
stoichiometric oxidant. Used in a second reaction, the regenerated catalyst led to a
decreased conversion of 30 % after 4 h. A turnover number (TON) of 16 after the first
reaction was obtained, and a total TON of 20 was observed for both reactions. The
lowering in activity may be attributed to the blocking of active sites by the organics after
the first reaction.
The kinetics of the oxidation of 9H-fluorene was followed at the very beginning when
conversions were relatively low and hence could be relied on as being representative of
the true kinetics of the reaction (Figure 3-3). 9H-Fluorene is oxidized selectively and at a
faster rate in isooctane than in toluene (Figures 3-3 and 3-4). When the reaction is
performed at 80 °C in toluene, the rate was lower than that for the reaction heated at
reflux, indicative of the thermodynamic dependency of this reaction in toluene. After
thermally treating the catalyst at 300 °C under nitrogen atmosphere and by using
nitrogen instead of air in the oxidation reaction, no conversion of the starting material
was observed implying that dissolved oxygen is involved in the oxidation cycle. To probe
the outcome of the reaction when the catalyst was calcined at a lower temperature, OMS2 was heated in air at 170 °C and the reaction was carried out under nitrogen
atmosphere. Interestingly, there was activity as 9-fluorenone was obtained in 30 min but
95
at a lower conversion of 10 %. The activity pointed to the involvement of lattice oxygen
species in the oxidation cycle. Temperature-programmed desorption mass spectrometry
and thermogravimetric analyses of this catalyst show that at 170 °C the only loss of
weight from the catalyst is by evolution of water as no other species are detected with the
mass spectrometer, but above 300 °C oxygen species are detected.
30
This observation
implies that surface oxygen species are still present at a lower temperature. There was a
drastic drop in the rates of generation of 9-fluorenone when carbon tetrachloride (CCI4)
was added into the reaction vessel. CCI4 lowered the rate of formation of the reactive
intermediates by either quenching them, thus hindering the oxidation cycle, or by
blocking the active sites thus making them unavailable for any further oxidation of 9Hfluorene.
Kinetically relevant elementary steps in the oxidation of 9H-fluorene, measured by
replacing H with D in 9H-fluorene and its effect on the rates of reaction, gave a KIE
value of 5.38 (in the absence of tunneling effects). From a pseudo-first-order plot of In
[9H-fluorene] o/[9H-fluorene] t versus time (InXo/X vs. time; Figure 3-5), the slopes,
with correlations of 0.9752 for the oxidation of 9H-fluorene and 0.9222 for oxidation of
fluorine-d 10, represent the rate constants kH and kD for 9H-fluorene and fluorine-dlO,
respectively. Early transition states for elementary C-H bond-activation steps would give
KIE values of approximately 1, and late transition states in which the relevant C-H bond
is significantly cleaved would give values as high as around 7.
36 37
Oxidation of 9H-
fluorene on OMS-2 thus represents late transition states, in which the breaking of the C-H
bond is the rate-determining step.
96
On the basis of our findings, we propose a series of steps that plausibly constitute the
reaction mechanism of OMS-2-catalyzed oxidation of 9H-fluorene. Mn n+ (n=2, 3, or 4),
2
3
1
the lattice oxygen species (O , O2 , or O ), and dissolved oxygen ( 0 2 or O2) are
involved in the oxidation cycle. The initiation step involves either dissolved oxygen in H
abstraction to generate hydroperoxy radicals or C-H bond cleavage, attributed to H
abstraction by lattice metal-oxo or hydroxy species of OMS-2. The high KIE value is
attributed to this step. Homolytic cleavage generates reactive species: free fluorene
radicals. Once formed, the free radicals quickly react with dissolved molecular oxygen to
form peroxy radicals. In an alternative route, they may also react with hydroperoxy
radicals to generate hydroperoxides. The peroxy radicals are stabilized by reaction with
neutral 9H-fluorene molecules to form hydroperoxides and generate further 9H-fIuorene
radicals. This mechanism is captured schematically in Figure 3-6. Manganese active sites
are responsible for catalyzing the homolytic decomposition of the intermediate
hydroperoxides to yield the ketone product with evolution of a molecule of water. GCMS and 1H NMR analyses did not reveal 9H-fluorenol as a product. As OMS-2
catalyzes the oxidation of alcohols 22 and hydrocarbons, then the presence of an alcohol
product is a priori not excluded, albeit, its existence may be curtailed by its short lifetime
which results from its simultaneous oxidation on OMS-2. The side product 9,9'bifluorene can be accounted for by the termination of the cycle involving two 9Hfluorene radicals. OMS-2 is regenerated at the end of the cycle.
A systematic investigation of the oxidation of 9H-fluorene using different OMS-2
catalysts (Table 3-2) led to the isolation of the two most active catalysts for this reaction.
OMS-2B and OMS-2s, prepared differently, were identified as the most active catalysts
97
for 9H-fluorene oxidation. These were followed by OMS-2R, OMS-2MW, and OMS-2H,
respectively. These catalysts have differing average oxidation states (AOS), crystallite
sizes, surface areas, Lewis acidities, and mesopore volumes.
The most active catalysts showed the lowest crystallite sizes as compared to the others
(Table 3-2). They also exhibited low AOS values (3.7) and relatively higher mesopore
volumes at 98 % and 92 % of total pore volumes, respectively. Small crystallites are the
most apparent features of both and a reason for the high activity exhibited by both
catalysts. At lower crystallite sizes, the material can fracture leading to disjointing of the
tunnels, which creates more defects and exposes active sites. Other than crystallite sizes,
a combination of these features may also be responsible for their high activity. A trend
cannot effectively be tied to Lewis acidity.
98
3.7
Conclusions
In summary, a mild, green, and efficient OMS-2 catalyzed oxidation of 9H-fluorene to 9fluorenone was successfully performed. Our study effectively identified the most active
mixed-valent OMS-2 catalysts for this reaction. The use of isooctane as solvent makes
this reaction very selective and sustainable. In terms of kinetics the breaking of the C-H
bond is rate controlling, and mechanistically the involvement of lattice oxygen species
has been implicated in a free-radical chain mechanism. This report with OMS-2 catalysts
is seminal and markedly different than those for the selective oxidation of alcohols. There
is a need to explore the nature of lattice oxygen species, the nature of dissolved oxygen,
and the exact state of the manganese centers that act as active sites. We also hope to
extend these findings to cover oxidation trends of other substrates in this family of
hydrocarbons using solvents with a higher degree of branching.
99
3.8
References
1.
2.
B. M. Weckhuysen, Nature, 2006, 439, 548-548.
J. H. Schon, C. Kloc, E. Bucher and B. Batlogg, Nature (London), 2000, 403,
408-410.
C. D. Dimitrakopoulos, S. Purushothaman, J. Kymissis, A. Callegari and J. M.
Shaw, Science (Washington, D. C.), 1999, 283, 822-824.
A. Afzali, C. D. Dimitrakopoulos and T. L. Breen, J. Am. Chem. Soc., 2002, 124,
8812-8813.
J. M. Thomas and R. Raja. Chem. Commun. (Cambridge, U. K.), 2001, 675-687.
R. Anwander, Chem. Mater., 2001, 13, 4419-4438.
B. M. Trost, Angew. Chem., Int. Ed. Engl, 1995, 34, 259-281.
B. M. Trost, Science (Washington, D. C, 1883-), 1991, 254, 1471-1477.
B. M. Choudary, M. L. Kantam, A. Rahman, C. V. Reddy and K. K. Rao, Angew.
Chem., Int. Ed., 2001, 40, 763-766.
Y. Han, A. Bisello, C. Nakamoto, M. Rosenblatt and M. Chorev, J. Pept. Res.,
2000, 55, 230-239.
M. L. Greenlee, J. B. Laub, G. P. Rouen, F. DiNinno, M. L. Hammond, J. L.
Huber, J. G. Sundelof and G. G. Hammond, Bioorg. Med. Chem. Lett., 1999, 9,
3225-3230.
Y. Sprinzak, J. Am. Chem. Soc., 1958, 80, 5449-5455.
JP2006199627A, 2006.
G. Qabaja and G. B. Jones, Tetrahedron Lett., 2000, 41, 5317-5320.
G. A. Olah, T. Mathew, M. Farnia and G. K. S. Prakash. Synlett, 1999, 1999,
1067,1068.
R. L. Danheiser, A. E. Gould, R. F. de la Pradilla and A. L. Helgason, The
Journal of Organic Chemistry, 1994, 59, 5514-5515.
F. L. Ciske and W. D. J. Jones, Chemlnform, 1998, 29, no-no.
M. M. Hashemi and Y. A. Beni, Chemlnform, 1999, 30, no-no.
S. L. Suib, Curr. Opin. Solid State Mater. Sci., 1998, 3, 63-70.
J. E. Post, R. B. VonDreele and P. R. Buseck, Acta Crystallogr., Sect. B, 1982,
B38, 1056-1065.
Y.-C. Son, V. D. Makwana, A. R. Howell and S. L. Suib, Angew. Chem., Int. Ed.,
2001 ,40, 4280-4283.
V. V. Krishnan and S. L. Suib, J. Catal, 1999, 184, 305-315.
X. Chen, Y.-F. Shen, S. L. Suib and C. L. O'Young, J. Catal., 2001, 197, 292302.
R. Ghosh, Y.-C. Son, V. D. Makwana and S. L. Suib, J. Catal., 2004, 224, 288296.
V. D. Makwana, Y.-C. Son, A. R. Howell and S. L. Suib, J. Catal., 2002, 210, 4652.
G. W. Roberts and C. N. Satterfield, Industrial & Engineering Chemistry
Fundamentals, 1966, 5, 317-325.
M. A. Vannice, Catalysis Today, 2007,123, 18-22.
I. W. C. E. Arends, R. A. Sheldon, M. Wallau and U. Schuchardt, Chemlnform,
1997, 28, no-no.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
100
29.
30.
31.
32.
33.
34.
35.
36.
37.
J. M. Thomas, R. Raja, G. Sankar and R. G. Bell, Nature, 1999, 398, 227-230.
J. C. Villegas, L. J. Garces, S. Gomez, J. P. Durand and S. L. Suib, Chem. Mater
2005,17, 1910-1918.
V. D. Makwana, L. J. Garces, J. Liu, J. Cai, Y.-C. Son and S. L. Suib, Catalysis
Today, 2003, 85, 225-233.
R. N. DeGuzman, Y.-F. Shen, E. J. Neth, S. L. Suib, C.-L. O'Young, S. Levine
and J. M. Newsam, Chem, Mater., 1994, 6, 815-821.
J. Yuan, K. Laubernds, Q. Zhang and S. L. Suib, J. Am. Chem. Soc., 2003, 125,
4966-4967.
K. A. Malinger, Y.-S. Ding, S. Sithambaram, L. Espinal, S. Gomez and S. L.
Suib, J. Catai, 2006, 239. 290-298.
R. Ghosh, X. Shen, J. C. Villegas, Y. Ding, K. Malinger and S. L. Suib, J. Phys.
Chem. B, 2006, 110, 7592-7599.
T. H. Lowry, K. S. Richardson, Mechanism and Theory in Organic Chemistry,
Harper and Row, New York, 1987.
Y.-C. Son, V. D. Makwana, A. R. Howell, S. L. Suib, Angew. Chem. 2001,
113, 4410-4413; Angew. Chem. Int. Ed. 2001, 40, 4280-4283.
101
4
EFFICIENT OXIDATION OF 2, 3, 6-TRIMETHYL PHENOL USING
NON-EXCHANGED AND H+ EXCHANGED MANGANESE OXIDE
OCTAHEDRAL MOLECULAR SIEVES (K-OMS-2 AND H-K-OMS-2) AS
CATALYSTS
4.1
Introduction
Vitamin E finds extensive use in food, medicine, and healthcare products, a-tocophenol is
the form of vitamin E known to meet human needs''' and plays a vital role in protecting
human cells from the damaging effects of free radicals thus acting to prevent various
diseases, for instance, cardiovascular diseases and cancer.' 2 ' In addition to its antioxidant
role, vitamin E also plays an important role in the regulation of gene expression, cell
signaling, and other metabolic processes."'
Current industrial production of vitamin E involves condensation of isophytol (IP) with 2,
3, 6-trimethyl-l,4-hydroquinone (TMHQ)' 3> obtained by hydrogenating TMQ as shown
schematically in Figure 4-1. The industrial production is carried out via oxidation of TMP
with molecular oxygen or air in the presence of copper halides as catalysts.' 4 " 6 ' Generally,
TMQ is obtained from oxidation of TMP.' 7
8)
The order of reactions involves para-
sulfonation of TMP followed by its oxidation using stoichiometric amounts of
commercial manganese dioxide. Several researchers have studied the oxidation step of
the reaction scheme with the aim of reducing the overall steps involved as well as moving
toward processes that are more environmentally friendly. One-step reactions have been
explored employing different oxidants (molecular oxygen (O2), hydrogen peroxide
(H2O2) or tertiary butyl hydroperoxides (TBHP).' 9 " 13 ' Sun et al.' 9
describe an approach
utilizing copper (II) chloride (CuCh) as homogenous catalysts in ionic liquid medium of
102
l-butyI-3-methylimidazolium chloride (BMIM-Cl) with n-butanol as a co-solvent. In that
study, the yield of TMQ reached 98% with use of higher amounts of CuCh gradually
decreasing with reduction in the catalyst amount. Although molecular oxygen was
utilized, this method required closed vessel, pressurized conditions (up-to 10 bars) to
achieve the reported results. Baiker et al. (12) utilized CuC^ in the presence of
NH2OH .HCI co-catalyst with TBHP as the oxidant in place of molecular oxygen. In this
approach the amount of CuCh could substantially be reduced without substantial
reduction in product yield, however, catalyst recovery is still a challenge. Interestingly,
the solubility of CuC^ in the CuCh system is not mentioned. CuCh is highly soluble in
aqueous media and has been reported elsewhere to have a solubility of 164 mg/100 g
solution in a system involving l-butyl-3-methylimidazolium hexafluorophosphate [BMIM] PF6. (,4) These solubility data imply that a similar trend is possible in the
reported catalytic syntheses utilizing [BMIM]CI, and thus homogeneous as opposed to
heterogeneous species could be the active components of the afore-mentioned systems.
On the other hand, the use of molecular oxygen (under pressure) resulted in
comparatively poor yields. Fenton's reagent (10) (FeS04/ 5% H2O2) system afforded 100
% conversion with 99.9% selectivity to TMQ after a 3 h reaction period. Ti-based
catalysts have also been studied for the synthesis of TMQ in a one-step reaction with high
conversion and selectivities after a 30 min reaction period.* I5 20)
Herein, we report the benign oxidation of TMP to TMQ using K-OMS-2 and H-K-OMS2 catalysts. These materials have previously been studied as selective catalysts for the
oxidation of different organic compounds.' 21 " 26 ' These materials showed excellent
selectivity in the alcohol oxidation as well as in other catalytic systems.
103
Figure 4-1. Scheme showing steps involved in the current industrial production of
vitamin E viz: (i) oxidation of TMP to TMQ and its subsequent reduction to TMHQ, and
(ii) condensation of TMHQ with isophytol to form vitamin E.
104
4.2
Aim of study
Given that K-OMS-2 materials have performed exemplarily in the oxidation of reported
organic molecules, their performance in the oxidation of this important industrial
chemical and the determination of the best reaction conditions are the driving forces of
our current study. These aims are successfully realized, thus extending the applications of
OMS-2 to include this important oxidation reaction.
4.3
Experimental
4.3.1
Reagents
Potassium permanganate (KMn0 4 ), manganese sulfate monohydrate (MnS0 4 H 2 0), 2, 3,
6-trimethyl phenol, acetonitrile (MeCN). and tert-butyl hydroperoxide (70% in water)
were purchased from Sigma-Aldrich while concentrated nitric acid (HNO3) was obtained
from Alfa Aesar. All reagents were used without any further purification.
4.3.2
Catalyst syntheses
Potassium containing OMS-2 (K-OMS-2) was synthesized based on earlier reported
procedures/ 27 " 29 ' In a typical reaction, 42 mmol (6.65 g) KMn0 4 was added to 100 mL
distilled-deionized water (DDW) to make mixture A. In another flask, 59 mmol (9.9 g) of
MnS04.H20 was added to 33 mL DDW to make mixture B. Both A and B were stirred
separately until complete dissolution of the reagents. To B, 3.4 mL of concentrated HNO3
was added and further stirred. Solution A was transferred to a dropping funnel and dropwise added to B under vigorous stirring. The resultant mixture was refluxed for 24 hrs in
an oil bath maintained at 110 °C upon which the product was filtered, washed until
neutral, dried in air at 120 °C for 12 hrs, and ground to fine powder. This powder (5 g)
105
was ion-exchanged for H + by dispersing in 100 mL of 1 M HNO3 and stirring at 70 °C for
6 hrs followed by a similar washing and drying procedure to form a strongly protonated
H-K-OMS-2.
4.3.3
Catalytic reactions
Catalytic oxidation reactions were performed by adding 1 mmol of TMP to 10 mL of
acetonitrile that contained 50 mg catalyst and a given amount of TBHP as oxidant.
Reactions were carried out at different temperatures for 30 minutes and samples
withdrawn using a syringe fitted with an inline (0.45 fim) filter. The composition was
analyzed and quantified using GC-MS based on their relative peak areas. Time and
temperature studies were performed by withdrawing samples for analysis at different
time intervals at given set temperatures.
4.3.4
Catalyst leaching test
To test whether the reaction was truly heterogeneous we performed the hot filtration
test. <30) This test was performed at 65 °C with all the other reaction conditions similar to
those of catalytic reactions tests as reported above. At exactly 5 minutes after start of
reaction, all the liquid was drawn into a syringe and then released into a clean flask that
had been maintained in a drying oven. An in-line filter was used to exclude solid
materials and the hot filtrate reacted for a further 30 minutes at the same reaction
conditions.
4.4
Characterization of the catalysts
The catalysts were characterized using: X-ray diffraction (XRD), scanning electron
microscope (FE-SEM), and Brunauer-Emmett-Teller (BET) surface area and pore size
106
distribution measurements. Results of these characterizations were compared to those in
literature reports.
4.4.1
X-ray diffraction
Powder X-ray diffraction studies were performed using a Scintag XDS-2000
diffractometer having Cu Ka (A. = 0.15406 nm) radiation and operating at a beam voltage
of 45 kV and a current of 40 mA. XRD data were collected continuously in the 20 range
of 5-75° at a scan rate of 1.0 deg/min and the crystalline phase identified using a JCPDS
database card number 29-1020. X-ray diffraction studies were performed on three
catalyst materials: freshly synthesized K-OMS-2, H + -exchanged H-K-OMS-2, and
recycled K-OMS-2 that had previously been used in one reaction after a washing and
drying procedure.
4.4.2
Scanning Electron Microscope
FE-SEM was performed on a Zeiss DSM 982 Gemini instrument with a Schottky emitter
at an accelerating voltage of 2 kV and a beam current of 1 mA. The carbon tape mount
method was used in sample mounting. Powder samples were finely ground and dispersed
in absolute ethanol in a glass vial then ultra-sonicated prior to being dispersed on Au-Pdcoated silicon glass chips previously mounted onto aluminum stubs with a two-sided
carbon tape and dried by vacuum desiccation prior to SEM studies.
4.4.3
BET Surface Area and Pore Size Distribution
These were performed using nitrogen sorption on a Micrometrics ASAP 2010 accelerated
surface area system. Samples were degassed at 200 °C for 12 h prior to pore size
107
distribution experiments which were carried out at 77 K. The specific surface area of the
material was determined by the Brunauer-Emmett-Teller (BET) method.
4.5
Characterization Results
4.5.1
Catalyst Characterization
K-OMS-2, H-K-OMS-2, and regenerated K-OMS-2 were subjected to X-ray diffraction
studies to identify their crystal structure and crystallinity. Figure 4-2 shows the results
from these diffraction studies. The XRD peaks could be indexed to the Q-phase of
Cryptomelane-type manganese oxide (JCPDS card no. 29-1020) for all of the three
samples. Electron microscopy experiments were performed on K-OMS-2 and are shown
in Figure 4-3 while nitrogen sorption isotherms and BET surface areas of K-OMS-2 are
shown in Figure 4-4. FE-SEM micrographs revealed a fibrous morphology while the
BET surface area was 98 m 2 /g well in line with previously reported results.
108
I
!
c
i
JL A
w
A
^Aa
• I I 1 I » I I ' I ' ' I 1 t ' ' 1 ' 1 ' 1 i ' I rr\ ' I ' I I ' ' » I I
10
20
30
40
Two Ttrata (ctog)
50
60
70
Figure 4-2. X-ray diffraction patterns of as synthesized K-OMS-2 (bottom), H-K-OMS-2
(middle), and regenerated K-OMS-2 (top). All materials were indexed to the Q-phase of
cryptomelane (JCPDS card no. 29-1020).
Figure 4-3. FE-SEM micrograph images of the as-synthesized K-OMS-2 catalyst
109
350-
300-
n
3.7 nm
I
200-
<150-
I
i too50-
02
04
06
itaMfv* fr»l*ur« (P/P*)
08
Figure 4-4. Nitrogen sorption isotherms of the as-synthesized K-OMS-2 materials.
Horvath-Kawazoe pore size distribution curve (inset) Corresponding BET surface area
was 98 m 2 /g.
110
4.6
Catalytic Results
4.6.1
Effects of the Type of Catalyst and Oxidant on Oxidation of TMP
The effect of the type of catalyst was studied by comparing a similar amount of H-KOMS-2 to K-OMS-2 in the oxidation reaction. The effect of oxidant type was studied by
comparing the results obtained by using molecular oxygen (present in air) or TBHP as
oxidants. Also studied were: the effect of performing the catalytic tests under nitrogen
atmosphere.The effect of the amount of TBHP was also studied. The results of these
studies are summarized in Table 4-1.
4.6.2
Temperature and Reaction Time Effect
Temperature effect on the oxidation of TMP to TMQ was studied by performing the
reactions at the best conditions experimentally determined of: 1 mmol substrate, 2 mmol
TBHP in 10 mL of acetonitrile and 50 mg K-OMS-2 for 30 minutes. Different
temperatures (room temperature, 40 °C, 65 °C, and 82 °C) were picked and the reactions
monitored by analyzing the composition of samples withdrawn at short intervals over a
30 min duration using a syringe with an inline filter. The results of these studies are
shown graphically in Figures 4-5 to 4-8.
4.6.3
Leaching test results
The results from the leaching test are shown in Table 4-1, entry 11. GC results of first
filtrate had a conversion of 57% and after a 30 minute filtrate reactions, stayed at 59%
which is close to the first results and so can be assumed to be a constant.
Ill
Table 4-1. Effects of selected parameters on the oxidation of TMP.
Catalyst
Oxidant
TMP
Temp.
Conv.
Sele.
loading
amount
amount
(°C)
(%)
(%)
(mg)
(mmol)
(mmol)
1
none
none
1
65
0
0
0
2
none
2.00 (a)
1
65
Trace
-
-
3
50
Nitrogen
1
65
6
0
1
4
50
Air
1
65
15
0
2
5
50
0.25
1
65
46 (35)
42(45)
7(6)
6
50
0.50
1
65
53
48
9
7
50
1.00
1
65
70
56
11
8
50
2.00
1
65
99
99
16
9
50 (b)
2.00
1
65
98
99
16
10
40
2.00
1
65
70
99
11
IjW
50
2.00
1
65
57 (59)
85 (87)
9(9)
Entry
TON
Notes: Percentage quantification (conversion and selectivity) was based on relative GCMS peak areas (averaged for three injections) and confirmed using toluene as an internal
standard to around 3% difference. Entry 1 corresponds to "blank" i.e., performed in the
absence of catalyst. (-) was not determined due to the product peak being below the
quantification range. All reactions were performed for 30 minutes except (a) for 60
minutes and (c) initially for 5 minutes. Similarly, all reactions were performed in the
presence K-OMS-2 except (b) which was performed in the presence of H-K-OMS-2.
Entry 5 (0.25 equivalent TBHP), a reaction under nitrogen purge gave results shown in
the parenthesis. Hot catalyst filtration test results are shown in entry 11 with the final
reaction results in parenthesis. Turnover number (TON) is the ratio of moles of substrate
converted to moles of catalyst used. Unless noted, the oxidant used was TBHP.
112
100
80
60
11 'ouvemon
Stlectivilv
40
20
Time <ntln)
Figure 4-5. Oxidation of TMP (1 mmol) at room temperature (25 °C) with 2 mmol
TBHP in 10 mL acetonitrile and 50 mg of K-OMS-2.
11. onvcmon
Selectivity
Time (nun)
Figure 4-6. Oxidation of TMP (1 mmol) at 40 °C with 2 mmol TBHP in 10 mL
acetonitrile and 50 mg of K-OMS-2.
113
Selecting
o
:
4
f-
8
io
»:
!<
Tttnr(itUn}
Figure 4-7. Oxidation of TMP (1 mmol) at 65 °C with 2 mmol TBHP in 10 mL
acetonitrile and 50 mg of K-OMS-2.
Sdcclivii>
0
2
-I
>S
ID
1!
I*
Time (mill)
Figure 4-8. Oxidation of TMP (1 mmol) at 82 °C with 2 mmol TBHP in 10 mL
acetonitrile and 50 mg of K-OMS-2.
114
4.7
Discussion
4.7.1
Catalyst Characterization
Pure, crystalline K-OMS-2 materials were successfully synthesized as evidenced by the
XRD patterns that were indexed to the Q-phase of Cryptomelane (Figure 4-3) and FESEM micrographs (Figure 4-4).
Ion-exchanging to prepare H-K-OMS-2 materials
appears not to have a significant effect on the crystal structure (change in d-spacing or in
the unit cell size) as identified from XRD patterns whose peak positions remain
unchanged (not shifted) from those of K-OMS-2. This may indicate a lack of framework
substitution and a possible exchange of the tunnel K + cations for H + ions. A regenerated
K-OMS-2 retained the crystalline structure albeit with decreased peak intensities.
Nitrogen sorption experiments reveal type-II isotherms due to an indefinite multilayer
formation at higher relative pressure after complete formation of monolayer at lower
relative pressure. Pore size distribution curves obtained using the Horvath-Kawazoe
method indicate that the pore sizes are narrowly distributed at about 3.7 nm,
4.7.2
Effect of Catalyst Nature on Oxidation of TMP
There is no significant difference between the performances of K-OMS-2 and protonated
H-K-OMS-2 in the oxidation process, thus Bronsted acidity alone has no significant
contribution to this oxidation reaction (Table 4-1, entries 8, 9) but rather, the active sites
(manganese in different oxidation states) present on the catalyst and the oxidizing species
(radical species from TBHP) present in solution play a more significant role. No
oxidation occurred in the absence of the catalyst lending credence to the fact that this is
indeed a reaction facilitated by the presence of K-OMS-2 (Table 4-1, entries 1 and 2).
115
Furthermore, hot filtration of the catalyst and reacting the filtrate alone did not
significantly change the conversion (Table 4-1, entry 11). Similar leaching studies using
this catalyst for other oxidation studies have been reported.
The amount of K-OMS-2
used (50 mg, 0.0625 mmol) when compared to 1 mmol of substrate also suggests that this
reaction is not stoichiometric either. Calculated turnover numbers (TON) are also above
the threshold of 1 in stoichiometric reactions.
4.7.3
Role of Oxidant on Oxidation of TMP
The presence of oxidant plays a crucial role. In the absence of the oxidant, the reaction
does not occur (Table 4-1, entry 1). Use of inert atmosphere (nitrogen) gave a conversion
of 6% (Table 4-1, entry 3) while use of air gave a slightly higher conversion of 15%
(entry 4). When TBHP was used, conversions of up-to 99 % were obtained (Table 4-1,
entries 5-9). At levels of 0.25 to 1 equivalent TBHP the conversion was low (46 % no
purge, 35% under nitrogen purge) and only reached the highest (99 %) at 2 equivalent
(Table 4-1, entries 5-8). This implies that dissolved oxygen also plays a role in the
oxidation process. Conversion and selectivity obtained were highest when 50 mg of
catalyst is used, 2 equivalent TBHP and the reaction carried out for 30 min. Thus the
foregoing are the best-determined reaction conditions for this system. Recycled K-OMS2 achieved a conversion of 70% when just 40 mg of the recycled catalyst is used in
reaction (Table 4-1, entry 10). The reuse was achieved by scrapping the filtered catalyst
off the filter paper, washing with ethanol and water then drying overnight at 120 °C.
116
4.7.4
Temperature and Reaction Time Effect on Oxidation of TMP
Conversion increased when temperature was increased from room temperature to 82 °C
(Figure 4-5 to 4-8). As reaction time was increased at any given temperature, the
conversion also increased. Time increase was limited to the first 30 min only. At room
temperature, conversion increases from slightly below 5 % at the start of the reaction to
about 33 % after 30 min (Figure 4-5). At this temperature selectivity towards TMQ is
poor. The other two products had a molecular mass of 270 but occurred at different
retention times. 2,6-Dimethyl-4-[(2,4,6-trimethylphenoxy)methyl]phenol was a given
possibility (from the database) . At 40 °C, conversions steadily increase from 5 % to 80
% in the first 15 min reaching a high of 95 % after 30 min (Figure 4-6). At 65 °C, in as
little as 10 min, the conversions reached over 95 % and increased to 99 % in 30 min
(Figure 4-7). At this temperature, selectivity rose to over 90 % in the first 10 min. At 82
°C, the conversion was at 97 % in 10 min and reached 99 % after 30 min (Figure 4-8).
However, unlike at 65 °C where conversions were at 99 %, at 82 °C the conversion or
selectivity is constant at 90 % by the end of the reaction.
117
4.8
Conclusion
From the foregoing, 65 °C is the optimum temperature for carrying out the oxidation of
2,3,6-trimethyl phenol using K-OMS-2 as benign catalysts. This temperature gave
conversions of 99 % and selectivity of 99 % in 30 min of reaction time. Other optimum
conditions are: 50 mg K-OMS-2, and 2 mmol TBHP for 1 mmol of TMP. These results
are comparable to or better than reported (9~'4> results and are achieved using much milder
and relatively inexpensive catalyst materials compared to literature reports. The
heterogeneity of the catalyst towards this reaction was also confirmed to be truly
heterogeneous.
118
4.9
References
1.
2.
Traber, M. G. (2007) Amu. Rev. Nutr. 27, 347.
Verhagen, H., Buijsse, B., Jansen, E., and Bueno-de-Mesquita, B. (2006) Nutr
Today 41 ,244.
Bonrath, W., Dittel, C., Giraudi, L., Netscher, T., and Pabst, T. (2007) Catal
Today 121, 65.
R. B. Rucker, J. W. Suttie, D. B. McCormick, Handbook of vitamins (Mercel
Dekker, New York, 2001)
Mercier, C., Charbardes, P. (1994) Pure Appl Chem 66, 1509
Bonrath, W., Netscher, T. (2005) Appl Catal A General 280, 55
Kholdeeva, O. A., Golovin, A. V., Maksimovskaya, R. I., and Kozhevnikov, 1. V.
(1992) J Mol Catal IS, 215.
Meng, X., Sun, Z., Lin, S., Yang, M., Yang, X., Sun, J., Jiang, D., Xiao, F.-S., and
Chen, S. (2002) Appl Catal A 236, 17.
Sun, H., Harms, K., and Sundermeyer, J. (2004) JAmer Chem Soc 126, 9550.
Li, Y., Zhang, P., Wu, M., Liu, W., Yi, Z., Yang, M., Zhang, J., Zhang, G., and
Bai, Z. (2009) Chem Eng J146, 270.
Sun, H., Li, X., and Sundermeyer, J. (2005) JMol Catal A Chem 240, 119.
Bodnar, Z., Mallat, T., and Baiker, A. (1996) J Mol Catal A Chem 110, 55.
Zhou, J., Hua, Z., Cui, X., Ye, Z., Cui, F„ and Shi, J. (2010) Chem Comm 46,
4994.
Yu, L„ Sun, H., He, J., Wang, D„ Jin, X., Hu, X., and Chen, G. Z. (2007)
Electrochem Commun 9, 1374.
Kholdeeva, O. A., Ivanchikova, I. D., Guidotti, M., Pirovano, C., Ravasio, N.,
Barmatova, M. V., Chesalov, Y. A. (2009) Adv Synth Catal 351, 1877
Kholdeeva, O. A., Invanchikova, I. D., Guidotti, M., Ravasio, N., Sgobba, M„
Barmatova, M. V. (2009) Catal. Today 141, 330
Kholdeeva, O. A., Ivanchikova, I. D., Guidotti, M., Ravasio, N. (2007) Green
Chem. 9, 731
Kholdeeva, O. A., Trukhan, N. N., Vanina, M. P., Romannikov, V. N., Parmon,
V. N., Mrowiec-Bialon, J., Jarz^bski, A. B. (2002) Catal Today 75, 203
Trukhan, N. N„ Romannikov, V. N., Paukshtis, E. A., Shmakov, A. N.,
Kholdeeva, O. A. (2001) J Catal 202, 110
Trubitsyna, T., and Kholdeeva, O. A. (2008) Kinet Catal 49, 371.
Trukhan, N. N., and Kholdeeva, O. A. (2003) Kinet Catal 44, 347.
Kumar, R., Sithambaram, S., and Suib, S. L. (2009) J Catal 262, 304.
Sithambaram, S., Kumar, R., Son, Y.-C., and Suib, S. L. (2008) J Catal 253, 269.
Malinger, K. A., Ding, Y.-S., Sithambaram, S., Espinal, L., Gomez, S., and Suib,
S. L. (2006) J Catal 239, 290.
Chen, X., Shen, Y.-F., Suib, S. L., and O'Young, C. L. (2001) J Catal 197, 292.
Son, Y.-C., Makwana, V. D., Howell, A. R., and Suib, S. L. (2001) Angew Chem
IntEd 40, 4280.
Opembe, N. N., Son, Y.-C., Sriskandakumar, T., and Suib, S. L. (2008)
ChemSusChem 1, 182.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
119
28.
29.
30.
Shen, Y.-F., Zerger, R. P., Suib, S. L., McCurdy, L., Potter, D. I., and O'Young,
C.-L. (1992) J Chem Soc Chem Commun 1213.
Opembe, N. N., King'ondu, C. K., Espinal, A. E., Chen, C.-H., Nyutu, E. K.,
Crisostomo, V. M., and Suib, S. L. (2010) JPhys Chem C 114, 14417.
Sheldon, R. A., Wallau, M., Arends, 1. W. C. E., Schuchardt, U. (1998) A c c C h e m
Res 31 ,485
120
CHAPTER 5.
SELECTIVE
PARTIAL
OXIDATION
OF
BENZYL
ALCOHOL ON MANGANESE OXIDE OCTAHEDRAL MOLECULAR SIEVES
(OMS-2) CATALYSTS IN THE GAS PHASE
5. Introduction
Selective oxidation is a fundamentally important topic in research owing to the
importance of oxidation as a unit operation in the manufacture of chemicals and chemical
intermediates. 1 The products of the alcohol oxidation process e.g. benzaldehyde, are
valuable both as intermediates to other compounds and also as end products in the
chemical and perfumery industries. 2 Selective oxidation of organic molecules, more
specifically, the oxidation of benzyl alcohol serves as a fundamental reaction for both
laboratory and commercial processes. 3
Understanding of this process will enable the
development of oxidation processes for other organic compounds. Benzyl alcohol is
currently produced through the use of stoichiometric amounts of commercial manganese
oxide (Pyrolusite) or chromium salts in the laboratory and commercially through the
chlorination and subsequent oxidation of toluene. 4 Disadvantages occasioned by using
stoichiometric reagents and toxic salts have necessitated research into alternatives. Both
liquid and vapor phase processes are possible with vapor phase being placed at an
advantage due to the lack of down-stream operations that the liquid phase process will
face e.g. solvent separation. This coupled with the use of a truly heterogeneous catalyst
places gas phase oxidation a pedestal higher than the liquid phase oxidation.
Gas phase selective oxidation of benzyl alcohol has been achieved using silvercontaining hexagonal mesoporous silicas (HMS) with a silver loading of between 0.55
and 3.50 wt%. 5
Nanoporous gold nanoparticles have been employed in the vapor phase
121
oxidation of benzyl alcohol with high selectivities (98.2%) and conversions of 61 %
using molecular oxygen as oxidant at 240 °C. 6 Mesoporous K-Cu-Ti02 catalyst has also
been successfully used to oxidize benzyl alcohol at the benzyl alcohol boiling point (203
°C) with a TOF of up-to 108 h' 1 . 7 A thin sheet of Ag/Ni-fiber catalyst has been employed
in the gas phase selective oxidation of benzyl alcohol with molecular oxygen giving 85 %
conversion and 94 % selectivity. 8 Benzyl alcohol has been catalytically oxidized using
Ag/SiC>2 impregnated with alkaline earth metals. 9 This was performed at 240 °C yielding
benzaldehyde as the major product with traces of benzene, toluene and CO2. The
performance of Cobalt-exchanged NaY Zeolite and ultra highly stable Y (UHY) and their
alkali metal counterpart in the oxidation of benzyl alcohol has been carried out at 350 °C
giving benzaldehyde yield of up-to 45 %. 10 Other reported catalyzed benzyl alcohol
reaction have been achieved at temperatures higher than 240 °C such as; using
K/Ag/SiCh," Au/Si0 2 , 12 Cu/Na/ZSM-5 13 and Au-Cu-Si0 2 . 14
Cryptomelane-type octahedral molecular sieves, a type of manganese oxide with tunnel
structures, have been extensively been studied via different synthetic routes and
characterized. Christened K-OMS-2, this material is made up of corner and edge shared
manganese oxide octahedral units (MnOg) resulting in the formation of tunnels with
dimensions of 4.6
A
by 4.6
A
(Figure 2-5). Potassium ions balance the charge on the
framework and also support the tunnels from collapsing.
K-OMS-2 has been studied extensively in the liquid phase oxidation of benzyl alcohol
Optimal conditions for the liquid phase oxidation were found to be: a catalyst loading of
50 mg, 10 mL of toluene as solvent, air as oxidant, and reaction time of 4 h. These
122
conditions gave conversions above 90% with excellent selectivity. The protonated
catalyst (H-K-OMS-2) gave slightly higher conversions compared to K.-OMS-2.
The success of K-OMS-2 in benzyl alcohol oxidation led to numerous other catalytic
reactions . l5 " 19 All these reactions were performed in the liquid phase.
We have developed a gas phase process for the oxidation of benzyl alcohol. This process
uses air or oxygen diluted in nitrogen as oxidants. The benzyl alcohol is first vaporized in
a heated chamber where it's met with a stream of carrier gas containing the oxidant for
onward transport to the catalytic chamber. Products are collected and analyzed
downstream.
5.2
Experimental
5.2.1
Materials
Potassium permanganate (KMn04), manganese sulfate monohydrate (MnSCVLhO), and
benzyl alcohol (99.99%, anhydrous) were purchased from Sigma-Aldrich while
concentrated nitric acid (HNO3) was obtained from Alfa Aesar. All reagents were used
without any further purification.
5.2.2
Catalyst synthesis
K-OMS-2 was synthesized based on earlier reported methods.' 20 " 23 ' In a typical reaction.
6.65 g (42 mmol) KMn04 was added to 100 mL distilled-deionized water (DDW) to
make mixture A. In another flask, 9.9 g (59 mmol) of MnS04.H20 was added to 33 mL
DDW to make mixture B. Both A and B were stirred separately until complete
dissolution of the reagents. Then 3.4 mL of concentrated HNO3 was added to flask B and
further stirred. Solution A was drop-wise added to B under vigorous stirring. The
123
resultant mixture was refluxed for 24 hrs in an oil bath maintained at 110 °C. The product
was thereafter filtered, washed until neutral to litmus, air-dried at 120 °C for 12 hrs, and
ground to a fine powder.
5.2.3
Ion-exchange
K-OMS-2 powder (5 g) was ion-exchanged (K + for H + ) by dispersing in 100 mL of 1 M
HNO3 and stirring at 70 °C for 6 hrs followed by a similar washing and drying procedure
to form protonated H-K-OMS-2.
5.2.4
Gas phase oxidation of benzyl alcohol
Zero grade air or UHP grade oxygen mixed with UHP grade nitrogen were employed as
oxidants. The oxidation reactions were performed using a conventional tubular furnace
that contained a fixed-bed glass vertical reactor (h = 1.5 ft, i.d. = 9 mm), fitted with
quartz wool supporting 0.35 g of catalyst. Prior to oxidation reactions, the catalysts were
preheated at 120 °C overnight in an oven under air and just before each reaction the
packed catalyst was further heated at the reaction temperature in pure oxygen for about 1
hour. The reaction was started by charging benzyl alcohol into a preheated zone (220 °C)
where it was vaporized before a stream of oxidant and carrier gas mixture were
introduced to mix with vaporized benzyl alcohol and transported further to the reactor.
The charging rates of benzyl alcohol were varied between 0.002 to 0.5 mL/min while that
of oxidant/carrier gas was varied between 40 and 80 seem. A syringe pump controlled the
benzyl alcohol flow rate and a mass flow meter controlled the oxidant/carrier gas flow
rate. A typical setup is represented diagrammatically in Figure 5-1. The liquid products
were collected using a cold trap and analyzed using gas chromatography. Gaseous
124
products were fed to an online GC (SRI multi component analyzer) equipped with a
thermoconductivity detector (TCD) and a molecular sieve (type 13X) column for the
analysis and quantification of gaseous products.
Mass Flow Controller
Furnace
II
||j
Heating
tape
Hci
(Tl)
Cold trap
Catalyst
<T2)
Syringe pump
Air cylinder
Figure 5-1. Schematic diagram of the vapor phase oxidation of benzyl alcohol using
OMS-2 materials. Benzyl alcohol is pumped using a syringe pump through Tl is the
maintained at 220 °C where it meets the oxidant/carrier gasses for onward transport to the
furnace maintained at a temperature T2. Cold trap contents are analyzed using GC-MS
while gaseous products are analyzed using an online GC as shown in the scheme.
125
5.3
Characterization
5.3.1
Powder X-ray Diffraction
The identity of the synthesized materials was confirmed using powder X-ray diffraction
(XRD). XRD studies were performed on a Scintag XDS-2000 diffractometer using Cu
Ka (k = 0.15406 nm) radiation and operating at a beam voltage of 45 kV and a current of
40 mA. Diffraction data were collected continuously in the 20 range of 5-75° at a scan
rate of 1.0 deg/min and the phase identified using a JCPDS database card number 291020. The XRD patterns of samples were collected on a glass sample holder. XRD
patterns of used catalysts were also taken and compared with patterns of the starting
materials.
5.3.2
Morphology
Field Emission Scanning Electron Microscopy (FE-SEM) was used to study the
morphology of the synthesized particles. This was performed on a Zeiss DSM 982
Gemini instrument with a Schottky emitter at an accelerating voltage of 2 kV and a beam
current of 1 mA. The carbon tape mount method was used where powder samples were
dispersed in iso-propanol in a glass vial and ultra-sonicated prior to being dispersed on
Au/Pd-coated silicon glass chips previously mounted onto aluminum stabs with a twosided carbon tape and dried by vacuum dessication prior to SEM studies.
5.3.3
Nitrogen sorption experiments
These were performed using on a Micrometrics ASAP 2010 accelerated surface area
system. Samples were degassed at 150 °C for 12 h prior to the sorption experiments
126
which were carried out at liquid nitrogen temperature (77 K). The specific surface area of
the material was determined with the Brunauer-Emmett-Teller (BET) method.
5.3.4
Oxygen evolution
Oxygen evolution from the synthesized K-OMS-2 was studied using temperature
programmed desorption (TPD) methods. This was achieved using a programmable
tubular furnace that housed a quartz tube (h = 2 ft, id = 9mm) with 60 mg of catalyst
supported in the tube by use of quartz wool. Masses desorbing from the catalyst were
analyzed using an MK.S quadrupole mass spectrometer. Prior to the analysis, the catalyst
bed was pretreated in stream of UHP nitrogen at 100 °C to remove any adsorbed species.
Thereafter, analyses were performed by ramping the temperature at the rate of 10 °C/min
from room temperature to 700 °C and evolved species analyzed using the mass
spectrometer.
5.3.5
Functional groups and acidity
Acidity (Lewis and Bronsted) measurements were performed by absorbing ammonia onto
the catalyst and subsequently analyzing using FTIR. Blanks were also recorded to
ascertain the validity of the results.
5.4
Results
5.4.1
Identity of the synthesized materials
The identity of the synthesized materials as well as materials recovered after gas phase
oxidation reactions are presented in Figure 5-2. The as-synthesized as well as ionexchanged OMS-2 materials display a similar XRD pattern as can be observed in Figure
127
5-2 (a) and (b). However, after performing the gas phase catalytic oxidation of benzyl
alcohol the crystal structure that the material assumes is that presented in Figure 5-2 (c).
To see why the structure changes from the crystal structure of Q-phase Cryptomelane
one, a benzyl alcohol gas phase oxidation reaction was performed using pure nitrogen as
carrier without any oxidant in the system. After this reaction an XRD pattern of the used
catalyst was obtained. These patterns are presented in Figure 5-2 (d). Finally, the assynthesized OMS-2 materials were heated at the gas phase reaction temperature (220 °C)
in the presence of pure nitrogen only under no reaction. An XRD pattern was obtained
post-treatment. The pattern obtained is presented in Figure 5-2 (e).
5.4.2
Morphology, Surface area and Pore size distribution
The as-synthesized K-OMS-2 was subjected to field emission scanning microscopy
inorder to obtain information about the morphology of the material. The results are
presented in Figure 5-3. The surface area and pore size distribution were obtained using
the Brunaeur Emmett Teller (BET) method. Sorption isotherms of the materials are
presented in Figure 5-4.
5.4.3
Oxygen evolution
Oxygen evolution was studied to determine the temperature regimes at which oxygen
evolves from the materials. The pre-treated catalyst was programmatically heated from
room temperature to 700 °C at a ramp rate of 10 °C/min with simultaneous monitoring of
the evolved species using a mass spectrometer. The results are of these analyses are
shown in Figures 5-5 and 5-6.
128
5.4.4
Functional groups and acidity measurements
The functional groups present on the synthesized materials as well as catalyst acidity
measurements were characterized using Fourier transform infrared (FT1R). The results of
functional groups and acidity measurements are shown in Figures 5-7 and 5-8
respectively.
129
111,
10
A^ RF
20
40
50
60
70
Two Tlitta (D«g)
Figure 5-2. X-ray diffraction results, (a) K-OMS-2, (b) H-K-OMS-2, (c) H-K-OMS-2
post-benzyl alcohol oxidation at N2/O2 (10/30) seem, (d) H-K-OMS-2 post-benzyl
alcohol oxidation in pure
(40 seem) reaction, and (e) H-K-OMS-2 after treatment
under pure N2 (40 seem) at 220 °C without reaction.
Figure 5-3. Scanning electron microscope images of the as-synthesized K-OMS-2
materials
130
350-
•+»Adsorption
Desorption
300 -
250-
"J 200-
150-
£ ioo50-
02
0.4
0.6
OB
fMitiv* Prassur* (F/P*)
Figure 5 -4. Sorption isotherms of K-OMS-2 materials with the corresponding BET
2
surface area of 77m /g.
Mass 16
o«yt#n)
6-
Mass 32
Mass 16
100
200
300
400
503
6(30
(DAG)
Figure 5-5. Oxygen evolution by temperature programmed desorption measured for KOMS-2
131
Mais 16 (rtoiTK o*J9»n)
32 (motoculti o«yg8tij
U -
12-
6 -
Mass 32
4-
Mass 16
Xil
100
T»>»p«nitur« {(lag)
Figure 5-6. Oxygen evolution by temperature programmed desorption measured for HK-OMS-2
3
1638
1089
2000
1500
1000
500
Wivtnumlwr (cm )
Figure 5-7. Functional groups study using FTIR
I32
1384.24
Bronsted acid site
(MS)
Lewis acid sites
S
s
8
1
1510.57
1454.13
IM
Figure 5-8. Acidity measurements by ammonia IR.
133
5.5
Catalytic results
Results of the benzyl alcohol oxidation in the vapor phase are summarized in Table 5-
Table 5-1. Results of the gas phase oxidation of benzyl alcohol in the gas phase using KOMS-2 materials.
Entry
Alcohol
Gas
flow rate flow
Conv.
(%)
(mL/min) rate
Selectivity (%)
Aldehyde Acid
Ester
TON
(seem)
1
0.002
40
93
88
12
0
255
2
0.02
40
92
99
1
0
252
3a
0.02
40
92
96
4
0
252
4h
0.02
40
91
99
1
0
250
5C
0.02
40
95
7
93
0
260
6
0.02
80
81
93
1
6
222
7
0.10
40
64
98
1
1
175
8
0.50
40
58
93
1
6
160
Results of the gas phase oxidation of benzyl alcohol in the vapor phase under reactions
conditions of: 0.35 g (-0.35 mmol) catalyst loading, temperature of 210 °C, nitrogen flow
of 15 seem and oxygen flow of 25 seem. In (a), K-OMS-2 was used while the rest were
performed using H-K-OMS-2. In (b), reaction performed using compressed air. (c) was
done using pure nitrogen gas. Turnover number (TON) is calculated as moles of product
divided by moles of catalyst. 10 mLs (0.096 mol) fed during the reaction is used as our
basis for calculating moles.
134
5.6
Discussion
5.6.1
Catalyst characterization
The wet chemical synthesis of K-OMS-2 through precipitation has been widely studied
•y
and reported. Portehault et al.
•>
have reported that this process is favorable under acidic
conditions and have shown that in a matter of minutes of mixing the synthesis mixture,
the oxidation process proceeds to about 99%. Based on the standard electrode potentials
of both Mn 2+ and Mn(V, the redox process is favorable.
Liquid phase oxidation reactions using K-OMS-2 materials have also been extensively
studied and reported 15 " 19 . Son et al. 15 pioneered the oxidation of benzyl alcohol in the
liquid phase thereby opening the pool of application of K-OMS-2 to other organic
compounds later on. However, there is no information on the use of K-OMS-2 in the
selective partial oxidation of organic compounds in the gas phase.
With this report, we have opened up a new frontier in the application of K-OMS-2 in the
partial oxidation of organic compounds in the gas phase. This has been achieved through
systematic studies to determine the best experimental conditions for the gas phase
reactions. These conditions are catalyst loading, benzyl alcohol flow rate, oxidant flow
rate, and reaction temperature.
An XRD analysis of both K-OMS-2 and H-K-OMS-2 reveal similar patterns with no
noticeable shifts in the peaks. Both materials were subjected to benzyl alcohol oxidation
and their structure re-analyzed after the reaction using XRD. As can be seen in Figure 5-2
(c), the patterns reveal a mixture of two phases. The predominant phase is K-OMS-2 with
emergence of a secondary phase (highlighted in grey) that can be traced to that of the
mineral hausmannite (Mn304 also written as MnO.M^Oj). The oxidation states of
135
manganese in hausmannite are +2 and +3. In an attempt to understand the contribution of
oxygen (in the carrier) mixture, the same reaction was done using pure nitrogen instead
of the mixture earlier used. The post-reaction catalyst was subjected to XRD with the
resultant patterns shown in Figure 5-2 (d). This is a pure phase of manganosite (MnO,
oxidation state Mn is +2). Further to understand the contribution of the reaction to the
patterns, this treatment was also carried out on the material but in the absence of a
reaction. The resultant XRD patterns are shown in Figure 5-2 (e). In K-OMS-2,
manganese exists in the oxidation states +2, +3, and +4. Used in the benzyl alcohol
oxidation in the liquid phase, the post-reaction XRD patterns showed no change. 15 In the
gas phase reaction, some of the manganese in the +4 state is reduced to +3 and +2
oxidation states, and some of the manganese in +3 reduced to +2 state. This leads to the
formation of hausmannite. Since not all K-OMS-2 is converted to hausmannite in the
presence of oxygen means oxygen is re-oxidizing the manganese in the lower oxidation
states back to the higher ones. This was confirmed by carrying the reaction in the
presence of pure nitrogen in which the post-reaction phase is pure MnO meaning all the
manganese have been reduced to +2 oxidation state. This reduction is only possible
during benzyl alcohol oxidation and not when pure nitrogen alone is used.
The synthesized materials retained a fibrous morphology, sorption isotherms, and BET
surface area that is characteristic to most K-OMS-2 materials. Oxygen evolved from both
K-OMS-2 and H-K-OMS-2 materials in the same temperature regimes of 500-550 °C. HK-OMS-2 materials showed increased oxygen evolution in the second temperature
regime of 650- 700 °C as opposed to K-OMS-2.
136
FTIR to study functional groups showed the normal vibrational spectra of K-OMS-2 with
Mn-0 vibrations in the range of 500 to 750 cm" 1 . Tunnel water vibrations are centered at
1638 cm" 1 . Vibrations at 1089 cm" 1 are attributed to Mn-O-H structural vibrations.
Acidity measurements using ammonia IR reveal the dominance of Bronsted acid sites
over Lewis acid sites. The Bronsted acid sites are more predominant on the protonated
(H-K-OMS-2) materials than the non-protonated one.
5.6.2
Catalytic results
As can be seen in Table 5-1, the analysis of liquid phase products reveal good
composition of oxidized products in the liquid phase. Both K-OMS-2 and H-K-OMS-2
were subjected to the gas phase oxidation of benzyl alcohol and both showed good
activity for the partial oxidation. H-K-OMS-2 is slightly better than K-OMS-2 since it
leads to better selectivity to benzaldehyde. K-OMS-2 leads to the presence of benzoic
acid at about 4% (Table 5-1, entry 3) as compared to H-K-OMS-2 at 1% (Table 5-1, entry
2) and so the rest of the study utilized H-K-OMS-2 as catalyst. Performing the reaction in
the presence of compressed air instead of the mixed oxygen/nitrogen leads to roughly the
same results (Table 5-1, entry 4).
The flow rate of benzyl alcohol is very crucial to the reaction. At low flow rates e.g.
0.002 mL/min ((Table 5-1, entry 1), the conversion is just as high as at 0.02 mL/min
(Table 5-1, entry 2) but the selectivity is compromised. Selectivity drops to 88%
compared to about 99% at 0.02 mL/min. When increased to 0.1 mL/min (Table 5-1, entry
7), conversion drops to 64% and drops even further to 58% at 0.5 mL/min (Table 5-1,
entry 8). However, faster flow rates don't seem to have a big swing on the selectivity as
compared to very slow flow rates.
137
The flow rate of the carrier gas is also crucial. Comparing two flow rates; 40 seem vs. 80
seem at the same flow rate of the liquid reactant, the conversion lowers at 80 seem (Table
5-1, entry 6) to about 81% down from 92% at 40 seem. This increase in the flow rate was
achieved by increasing nitrogen flow in the mixture and leaving oxygen at the initial flow
of 25 seem. This means the oxidant is diluted in this case and the residence time is also
affected which could be the reasons for the change in the conversion.
The XRD results of the used catalyst revealed the presence of hausmannite phase, which
meant that the manganese was being reduced as the alcohol was being oxidized. This
prompted us to study the effect of performing this reaction without any oxidant. The
reaction was performed in the presence of pure nitrogen. The results (Table 5-1, entry 5)
reveal that instead of selective oxidation to benzaldehyde, the reaction favors the
oxidation to benzoic acid as the major product.
Gas phase analyses of the gaseous products reveal total oxidation is occurring as well.
However, the extent to which this is occurring cannot be determined accurately and so is
excluded from these results.
138
5.7
Conclusion
Oxidation of benzyl alcohol proceeds with good conversions in the gas phase.
Benzaldehyde is the major product when the reaction is carried out at the best determined
conditions of benzyl alcohol flow rate of 0.02 mL/min, and carrier/oxidant flow rate of 40
seem. The oxidation of benzyl alcohol also produces total oxidation products i.e. carbon
dioxide to a lesser extent. This process is also a reducing reaction on the catalyst as the
presence of identified phases of hausmannite and manganosite in control experiments
reveal this. The oxidation of benzyl alcohol opens up the application of manganese oxide
(K -OMS-2) to a host of other reactions that can be attempted and optimized.
139
5.7.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
References
R. A. Sheldon and J. K. Kochi, Metal-Catalyzed Oxidations of Organic
Compounds (Academic Press, New York, 1981).
M. Beller and C. Bolm, Transition Metals for Organic Synthesis, 2nd ed., (WileyVCH, 2004).
G. Brink, I. W. C. E. Arends and R. A. Sheldon, Science 287 (2000) 1636.
(a) Kroschwitz, J. 1. Kirk Othmer Encyclopeida of Chemical Technology, 4th ed.;
Wiley-Interscience Publications: New York, 1992; Vol. 4. (b) Centi, G.; Cavani,
F.; Trifiro, F. Selective Oxidation by Heterogeneous Catalysis; Kluwer
Academic/Plenum Publishers: New York, 2001
Jia, L„ Zhang, S., Gu, F., Ping, Y., Guo, X., Zhong, Z., Su, F. Microporous
Me soporous Mater. 2012, 149, 158-165
Han, D., Xu, T., Su, J., Xu, X., Yi Ding, Yi., ChemCatChem, 2010, 2, 383-386
Fan, J., Dai, Y., Li, Y., Zheng, N., Guo, J., Yan, X., and Stucky G. D. J. Am.
Chem. Soc2009, 131, 15568-15569
Mao, J., Deng, M., Xue, Q., Chen, L., Lu, Y., Catal. Commun. 2009, 10, 1376—
1379
Sawayama, Y., Shibahara, H., Ichihashi, Y„ Nishiyama, S., Tsuruya S. Ind. Eng.
Chem. Res., 2006, 45, 8837-8845
Nakashima, D., Ichihashi, Y., Nishiyama, S., Tsuruya, S. J. Mol. Catal. A: Chem.
2006, 259, 108-115
Yamamoto, R., Sawayama, Y., Shibahara, H., Ichihashi, Y., Nishiyama, S. and
Tsuruya, S. J. Catal. 2005, 234, 308-317
Biella, S. and Rossi, M. Chem. Commun. 2003, (3), 378- 379
Hayashibara, H., Nishiyama, S., Tsuruya, S. and Masai, M. J. Catal. 1995, 153,
254- 264
Pina, C. D., Falleta, E. and Rossi, M. J. Catal. 2008, 260, 384- 396
Son, Y.-C. Makwana, V. D. Howell, A. R. and Suib, S. L. Angew. Chem., Int.
Ed., 2001 ,40. 4280-4283.
Opembe, N. N. Son, Y.-C. Sriskandakumar, T. and Suib, S. L. ChemSusChem,
2008, 1, 182-185.
Kumar, R. Sithambaram, S. and Suib, S. L. J. Catal., 2009, 262, 304-313.
Ghosh, R. Son, Y.-C. Makwana, V. D. and Suib, S. L. J. Catal., 2004, 224, 288296.
Kumar, R., Garces, L. J. Son, Y.-C. Suib, S. L. and Malz, R. E. J. Catal., 2005,
236 ,387-391.
Nyutu, E.; K.; Chen, C-H.; Sithambaram, S.; Crisostomo, V. M. B.; Suib, S. L.; J.
Phys. Chem. C, 2008, 112, 6786.
Crisostomo, V. M. B. New synthetic routes to catalytically active manganite, KOMS-2 and K-OMS-2/silicon dioxide and a preliminary study on the use of a
continuous flow microwave technique in the synthesis of nanosized manganese
and cerium oxides and cobalt and iron oxyhydroxides. Ph.D Thesis, University of
Connecticut, Storrs, CT, 2008
140
22
Nyutu, E. K. Processing and optimization of functional ceramic coatings and
inorganic nanomaterials. Ph.D Thesis, University of Connecticut, Storrs, CT,
2008.
23
D. Portehault, S. Cassaignon, E. Baudrin and J.-P. Jolivet, Chem. Mater., 2007,
19, 5410-5417.
141
CHAPTER 6.
STEAM REFORMING GLYCEROL FOR HYDROGEN
GENERATION USING NI-ZEOLITE Y
6.1
Introduction
Steam reforming (SR) is an endothermic process in which an organic compound e.g.
ethanol is heated with steam in the presence of a catalyst to produce a mixture of products
that include hydrogen, and carbon dioxide (Equation 6-2). SR is one of the ways that
hydrogen (H2) gas can be generated.
Molecular hydrogen is almost non-existent in nature and has to be produced from
available hydrogen-containing sources. Currently, hydrogen is generated through four
basic methods; water electrolysis, gasification of hydrocarbons, partial oxidation of heavy
oils, and dry or steam reforming of hydrocarbons. 1 Steam reforming of renewable
resources represents a promising alternative due to environmental concerns and the
eventual depletion of hydrocarbon sources.
Hydrogen can be utilized in fuel cell systems. Fuel cells have been considered as
effective generators of energy since they employ high conversion efficiencies from
chemical to electrical energy. 2 Hydrogen is also a clean burning fuel since the only
products of combustion are heat and water.
Proton exchange membrane fuel cell (PEMFC) is a type of fuel cell that uses hydrogen as
the fuel source. 3 PEMFC can be used as a generator of electricity for various uses or used
directly to power electric vehicles. The hydrogen used in PEMFCs is produced largely
from reforming of hydrocarbons, which can be done onboard or can be supplied from a
hydrogen cylinder. Due to safety and storage issues encountered with hydrogen, ondemand, and onboard generation is preferred.
142
Hydrogen production from steam reforming of ethanol (SRE) is a topic that has received
a lot of research focus in the recent past. 418 This is because ethanol can be obtained from
renewable sources like the fermentation of carbohydrates. However, ethanol is also
valuable as a chemical in the chemical and pharmaceutical as well as food processing
industries. Absolute ethanol is a fuel and is directly blended with gasoline to power
vehicles. On the other hand glycerol is a resource that is yet to be explored as a viable
hydrogen store. Glycerol is obtained as a bi-product of the biodiesel production process
and has very limited uses in the chemical industry. Additionally, glycerol is a poor fuel
and cannot be burned directly in internal combustion engines. Based on Equation 6-3,
glycerol appears to be a viable candidate for steam reforming to generate hydrogen.
Noble metal catalysts, such as Ir, Au, Ru, Pt and Rh
47
have been investigated as high
performance catalysts for the steam reforming of ethanol. However, they are expensive
and thus un-attractive. Among the non-noble metal catalysts studied so far, cobalt (Co)based catalysts have been reported to show the best performance, giving high hydrogen
yield in the SRE reaction. 8 " 11 However, the Co-based catalysts suffer from deactivation
after a short period of time. This deactivation is due to coke formation on the catalyst
surface. Another good candidate tolerant to deactivation better than Co is nickel. I2 ~ ,s
Nickel supported on rare earth oxides (CeC > 2 , Ye2C>3, La203) shows good catalytic
16 18
activities. " The nature of the support appears to play a very important role. In the
studies of nickel on rare earths, the excellent redox nature and good oxygen storage
capacities of the supports are touted for their good catalytic activities. Ni has also shown
good SRE performance when loaded on AI2O3 and Si0 2 . These supports offer higher
surface areas and good dispersion of the nickel metal.
143
Based on the foregoing, we hereby report our findings on the steam reforming of glycerol
(SRG) using nickel (Ni) on Zeolite Y. We chose to use glycerol as our reforming
substrate due to the advantages of using glycerol discussed earlier and zeolite Y due to
the fact that the Zeolite has high surface area and can therefore disperse nickel well.
Zeolite Y has also compositional characteristics intermediate between the two extremes
of AI2O3 and Si0 2 thus offers a good compromise of the two in terms of chemical
properties.
Zeolites are microporous aluminosilicates having pores of molecular dimensions 19
forming a subset of a larger class of molecular sieves. Pure siliceous materials have a
neutral framework. The presence of alumina in the framework gives these materials an
overall negative charge, which is normally balanced by extraframework cations e.g.
sodium ions (Na + ). The extraframework cations can be exchanged for other ions e.g. Ni 2+
which are less favorable due to charge imbalance.
Faujasites (structural type FAU, Figure 6-1) crystallize in the cubic space group Fd"3m
with lattice constant ranging from about 24.2-25.1 A depending on the composition of the
framework. 19 The structure is composed of: 24-tetrahedra cuboctahedral units (sodalite
cages), joined through hexagonal prisms (also known as double 6-rings), forming 12 ring
units known as the supercages. The two most commonly encountered faujasites are
zeolite X (with higher A1 content) and zeolite Y (with lower A1 content).
144
4
\
Figure 6-1. Crystal structure representation of faujasites (FAU framework) viewed in the
(111) direction. The vertices of the structure represent sodalite units similar to carbon
atoms in the diamond structure. The tetrahedra represent either SiC >4 or
AIO4"
units.
Patterns generated using Crystal Maker™
145
6.2
Experimental
6.2.1
Materials
The following reagents were used in this study. They were used as purchased from Sigma
Aldrich with no further purification. Colloidal silica (Ludox 40 %), sodium aluminate,
sodium hydroxide, nickel nitrate hexahydrate, and glycerol.
6.2.2
Synthesis of Zeolite Y
Zeolite Y (NaY) was synthesized through a microwave-assisted process. The synthesis
recipe used was adopted from the one used by Panzarela et al. 20 but the microwave
conditions used in our study was arrived at through our optimization processes. The
synthesis involved completely dissolving starting precursors individually in a given
amount of water followed by mixing them together while vigorously stirring. The
composition of the mixture was such as to correspond to the formula, 6 Na20:
AI2O3:
8SiC>2: 24OH2O. This mixture after 16 h ageing at room temperature, was put in a Teflon
liner and hydrothermally reacted in a CEM Mar 5 microwave with the following
parameters: 300 W power, (90, 120, 150 and 180 °C) temperature for 1 h. Pressure
corresponding to each temperature level was also directly determined using a pressure
sensor. The product was filtered and washed through a quartz sintered filtering flask until
neutral in pH, then dried overnight at 100 °C.
6.2.3
Ni-exchanged Zeolite Y and Ni-loaded Zeolite Y
For ion-exchange purposes, a 0.5 M solution of nickel nitrate was prepared. NaY (2 g)
was dispersed in 50 mL of the 0.5 M solution in a round bottom flask and kept in an oil
bath at 90 °C for 24 h. After filtering and air-drying, the same material was subjected to a
146
second and a third ion-exchange using new 0.5 M solutions in each case. This represented
a triple ion-exchange process. For nickel loading, 10 % (wt) nickel nitrate was
impregnated onto 5 g NaY, and then dried at 120 °C in air. Both ion-exchanged NaY and
Ni-loaded NaY were reduced in a stream of 5 % hydrogen in nitrogen at 550 °C for 6 h.
6.2.4
Glycerol steam reforming
Catalytic tests were performed in a fixed-bed quartz reactor (i.d. 9 mm) at atmospheric
pressure. 0.2 g of catalyst (bed volume -0.6 mL) was loaded into the reactor for each run.
The temperature was controlled using a K-type thermocouple. Prior to each catalytic
reaction, the catalyst was flushed using UHP N 2 for about 10 minutes. The catalysts were
tested at a set temperature in the range of 550-650 °C. The mixed liquid reactants with
the volume ratio of H20:Et0H=3:l was fed into the vaporizer with a flow rate of 0.02
mL/min using a peristaltic pump. The liquid reactant was vaporized at 220 °C in the
vaporizer and diluted with a stream of pure N2 flowing at 40 seem before entering the
reaction chamber. A gas-liquid separator (cold-trap) was used to separate any liquid
products from the dry product stream. The dry product stream was analyzed using an on­
line GC (SRI multi-component analyzer) equipped with a helium ionization detector
(HID) and both molecular sieve (type 13X) and Hysep columns. The liquid products
collected in the cold-trap were analyzed off-line on HP 5890 Series II gas chromatograph
equipped with an HP 5971 mass selective detector coupled with a TCD detector.
Separation was carried out in a non-polar column (HP-1). The catalytic set-up can be
seen diagrammatically in Figure 6-2.
147
Mass Flow Controller
Furnace
Cold trap
Heating tape
Hei
CT1J
Catafyst
(T2)
Syringe pump
Nitrogen cylinder
Figure 6-2. Schematic diagram for the glycerol steam reforming reaction. Glycerol/water
mix is evaporated at T1 (set at 220 °C) while T2; the furnace temperature is varied from
500 to 650 °C.
148
6.3
Characterization
6.3.1
Powder X-ray diffration
The identity of the synthesized materials was analyzed using X-ray powder diffraction
(XRD). The ion-exchanged NaY was also subjected to XRD analysis and a shift in the
peaks analyzed. XRD studies were performed on a Rigaku Ultima IV diffractometer
using Cu Ka (X = 0.15406 nm) radiation. Beam voltage and beam current of 40 kV and
44 mA were used respectively. Diffraction data were collected continuously in the 20
range of 5-75° at a scan rate of 2.0°/min, and the phases identified using the Joint
Committee on Powder Diffraction Society (JCPDS) database. For the synthesized
materials, the cell parameter was calculated using Formula 6-1. 21
a2=(A,2/4sin20) *(h2+k2+l2)
Equation 6-1. Formula for calculating lattice parameter in a cubic crystal system, where
a refers to a parameter, and hkl are the miller indeces.
6.3.2
Elemental analysis
Elemental analysis was done on both zeolite Y and the triply Ni-exchanged NaY
catalysts. The analysis was performed using two methods. The first involved using
energy dispersive X-ray analysis (EDX). The second was using inductively coupled
plasma (ICP) spectroscopy. EDX analyses were performed on an Amray model 1810
scanning electron microscope equipped with this option, while ICP was performed using
an ICP-AES instrument. ICP involved dissolving 50 mg of the materials individually in a
10 mL ofHCI:HN0 3 :HF (4:1:5 v/v). The analysis was made against calibration standards
of Na + and Ni 2+
149
6.4
Results
6.4.1
Catalyst characterization
XRD patterns of the as-synthesized materials are shown in Figure 6-3 and are indexed to
faujasite crystal structure in Figures 6-4 representing material synthesized at 90 °C and to
gismondine (GIS) structure representing material synthesized at 180 °C in Figure 6-5.
The faujasite structure has been indexed using JCPDS card no. 39-1380, 21 while the
gismondine structure has been indexed using JCPDS no. 39-219. 22 Card no. 39-1380 is
for faujasite with the chemical formula of Na2Al 2 Si 4 0i2.8H20 while card no. 39-219 is
for a sodium aluminum silicate hydrate with formula Na6Al6Siio032-12H20 (zeolite NaPl). Materials synthesized at 120 °C and 150 °C in Figure 6-3 can be indexed based on
the already indexed materials synthesized at either 90 °C or 180 °C. Calculation of the
lattice parameters of the material synthesized at 90 °C is displayed in Table 6-1.
Triple ion exchange of NaY with 0.5 M Ni 2+ yielded XRD patterns displayed in Figures
6-6 and 6-7.
XRD patterns of 10% by weight impregnated NaY before and after
reduction in hydrogen are displayed in Figures 6-8 and 6-9. Data corresponding to the
changes in the peak positions in the XRD patterns that accompany the above treatments
can be found in Tables 6-2 and 6-3.
Figures 6-10 and 6-11 show the results of elemental analysis using energy dispersive Xray analysis. 1CP analysis of the two samples in Figures 6-10 and 6-11 gave the sodium
amount to be 17.81 ppm in pure NaY with no nickel detected, while in triply Ni 2+ exchanged NaY (Figure-6-11) the sodium was present at 13.06 ppm while nickel was
present at 5.90 ppm.
TEM images of reduced nickel impregnated sample are displayed in Figure 6-12.
150
3
s
Material synthestxad at 180 C
£
C
Material synthesized at 150 C
t
iAi iifl m rtMiAi*«i * nA i.dniA ^ hi >Aiii>'iirrfinf»^i/>i in\ini«fA ,r.
I
OC
Material »ynth«»iz«d at 120 C
Material synthesized at 90 C
|1 I
20
40
30
50
1* p. '•I I"
60
70
Two Theta (dag)
Figure 6-3. XRD patterns of microwave hydrothermally synthesized NaY materials at
different temperatures after 16 h ageing where (a) represents materials synthesized at 90
°C, (b) 120 °C, (c) 150 °C, and (d) 180 °C.
Index (JCPDS: 39-1380)
IA
S? N
•
I.,L,J, ti, It.,1.1,1,it >«..
Two Theta (d«g)
Figure 6-4. XRD patterns of microwave hydrothermally synthesized NaY materials at 90
°C after 16 h ageing, indexed using JCPDS file no. 39-1380.
151
Index (JCPDS: 39-219)
3
4
«
E
|
is 3
1
« N
&
'f n n
| » I*I
10
i { i in r |"T*r-rr f-T"T-f i j ^
20
30
; «rt
ft p f i fyv ? > |
i
40
50
Two TH»tft (d*9)
Figure 6-5. XRD patterns of microwave hydrothermally synthesized NaY materials at
180 °C after 16 h ageing, indexed using JCPDS file no. 39-219.
3
T
I
m
10
20
30
50
60
70
Figure 6-6. XRD patterns of (a) 90 °C synthesized NaY, (b) a triple ion-exchange with
Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are for
nickel metal.
152
90 C miliaria is
I
£
I
£
10
20
Two That*(A»0)
30
40
Figure 6-7. Enlarged view of (a) 90 °C synthesized NaY, (b) a triple ion-exchange with
Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are for
nickel metal.
Table 6-1. Details of X-ray diffraction Pattern of material synthesized at 90 °C.
2-Theta
6.4381
10.3692
12.0646
15.8546
18.8465
20.5285
22.9446
23.7509
27.141
27.8212
29.7387
30.7858
Sin2© h2+k2+l2
0.003
3
0.008
8
0.011
11
0.019
19
0.026
27
0.031
32
0.039
38
0.042
43
0.055
56
0.057
59
0.065
67
0.07
72
hkl
111
220
311
331
511
440
620
533
642
731
733
822
X2
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
0.0237
a2
5.933
5.933
5.933
5.933
6.161
6.125
5.781
6.075
6.041
6.141
6.116
6.102
a (nm)
2.435
2.436
2.436
2.436
2.482
2.475
2.404
2.465
2.458
2.478
2.473
2.470
153
Table 6-2. Peak shift (d-spacing) on exchanging sodium for nickel (Na-»Ni 2+ ) and also
on reduction of nickel-exchanged NaY (Ni 2+ -»Ni). Basis for comparison is the assynthesized NaY in both cases.
d-spacin g
(hkl)
(111)
(220)
(311)
(331)
(511)
(440)
(533)
(642)
NaNi2+Y
NaY
13.72
8.52
7.32
5.59
4.71
4.32
3.74
3.28
14.17
8.71
7.47
5.68
4.76
4.37
3.77
3.3
A(hkl)
2+
NaNiY
14.08
8.68
7.39
5.63
4.72
4.35
3.74
3.29
A(hkl)
2+
Na->Ni
Ni
0.45
0.19
0.15
0.09
0.05
0.05
0.03
0.02
0.36
0.16
0.07
0.04
0.01
0.03
0
0.01
—>Ni
120 "C materials
?
a
§
n
i •
1
!
JlJLji
4
r n r f i
T
10
20
i
r r r i i
30
40
Two Thata (tftg)
50
60
70
Figure 6-8. XRD patterns of (a) 120 °C synthesized NaY, (b) after impregnation with
10% Ni 2+ solution, and (c) after reduction in a hydrogen atmosphere. Labeled peaks are
for nickel metal.
154
120 °C
—,—i—,—,—,—,—,—i—|—i—i—,—i—r
10
20
30
40
Two Thttft (dig)
Figure 6-9. Enlarged view of (a) 120 °C synthesized NaY, (b) after impregnation with
10% Ni24" solution, and (c) after reduction in a hydrogen atmosphere.
Table 6-3. Slight peak shift on impregnating Ni 2+ onto NaY ([Ni 2+ ]NaY] synthesized at
120 °C and also on reduction to nickel metal ([NiJNaY. The basis of comparison is the
as-synthesized NaY.
d-spacing
(hkl)
(111)
(220)
(311)
(331)
(511)
(440)
(533)
(642)
NaY
14.18
8.72
7.43
5.66
4.75
4.38
3.77
3.3
[Ni2+lNaY
14.05
8.67
7.39
5.63
4.74
4.34
3.76
3.29
[NiJNaY
14.02
8.63
7.38
5.63
4.74
4.34
3.75
3.21
A(hkl)
2+
A(hkl)
2+
INi 1
Ni -*Ni
0.13
0.05
0.04
0.03
0.01
0.04
0.01
0.01
0.16
0.09
0.05
0.03
0.01
0.04
0.02
0.09
155
Stxectnmi-t
4
A
5
CursorVert»5"64
10
IV
Window 0 005 • 40 955- 115484 cnt
*|L
Na
AtftinHr %b
16.357
24.722
58.921
100
Ka
Ka
Ka
Total
Al
Si
Figure 6-10. The EDS results of as-synthesized NaY (aged 16 h and synthesized at 90
°C)
SfMtctrau:
K|
Ni
N,i
HiUl
N,ff
•
_t
_
Nt
"" '
5
•
_
t
10
l*
CUMOf*
Waulow »<M»< -
OL
Um
t>~1<U tut
AIMMI*
16
Al
Si
Ni
Ka
Ka
Ka
Total
27.15
56.064
16.787
100
Figure 6-11. The EDS results of triply ion exchanged NaY (aged for 16 h and
synthesized at 90 °C).
156
Figure 6-12. Transmission electron micrograph (TEM) images of reduced Ni
impregnated NaY samples prepared at 120 °C.
..LU.
(C)
Ill
\
»I
ji
•I/'IIL_I^.LI
it
l'i F-
' i. Iiri..^'i I** '< !• r*
; 1
(b)
J
i:
;i ' V ! u i U ,
A
Jw VW** ^ V > V ''U VkA1
h
, «
WI/vv^VmwVV/'Aw'
Figure 6-13. Matching XRD patterns of (a) NaY and (b) NaNi 2+ -Y against (c) a
simulated pattern obtained by generating a diffraction pattern from atomic positions and
site occupancies of hydrated NiY from ref. 21. Patterns generated using Crystal Maker™
157
6.4.2
Catalytic results
Steam reforming of glycerol was conducted at a fixed flow rate of both the liquid
reactants and the carrier gas. The only variable studied under these conditions was the
reaction temperature. The amount of catalyst used was also held constant. The results
were interpreted as follows:
Hydrogen yield = actual moles of H2 produced/ theoretical moles of H2
Dry product stream composition quantified on mole percentage of quantified gaseous
products basis.
The results of these studies are presented in Figures 6-14 to 6-19.
158
0.5
0
,
15
,
55
,
95
,
135
,
175
,
215
time on stream (min)
Figure 6-14. Hydrogen yield in the glycerol steam reforming at 550 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY.
70 -
C2H4
Figure 6-15. Composition of dry product stream in the glycerol steam reforming at 550
°C with glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g
10% Ni-impregnated NaY.
159
>
M
XI
15
55
95
135
215
175
time on stream (min)
Figure 6-16. Hydrogen yield in the glycerol steam reforming at 600 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY.
55
95
135
175
215
Time on stream (min)
Figure 6-17 Composition of dry product stream in the glycerol steam reforming at 600
°C with glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g
10% Ni-impregnated NaY.
160
XI
15
55
95
135
175
215
time on stream (min)
Figure 6-18. Hydrogen yield in the glycerol steam reforming at 650 °C with
glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g 10% Niimpregnated NaY.
60 n
15
55
95
135
1/5
215
Time on straam (min)
Figure 6-19. Composition of dry product stream in the glycerol steam reforming at 650
°C with glycerol/water liquid flow rate of 0.02 mL/min, N2 flow rate of 40 seem, 0.5 g
10% Ni-impregnated NaY.
161
6.5
Discussion
6.5.1
Synthesis and characterization of Zeolite Y
Zeolite Y was successfully synthesized with a 1 h microwave crystallization time in after
ageing the synthesis gel for 16 h at room temperature. XRD patterns confirm the
successful crystallization of zeolite Y at 90 °C as indexed using JCPDS no. 39-1380
which is derived from pure faujasite material.
22
Ageing the synthesis gel at room
temperature for 16 h leads to crystallization of zeolite Y even at 90 °C (Figure 6-4).
zeolite Y is also obtained at 120 °C and 150 °C as can be seen in Figures 6-3 both of
which are indexed by comparing with the material synthesized at 90 °C which has been
indexed using JCPDS no. 39-1380. However at 150 °C there is appearance of few nonzeolite Y peaks, which on comparing to material synthesized at 180 °C. shows they are as
a result of crystallization of zeolite Na-Pl. Zeolite Na-Pl is the synthetic form of the
mineral gismondine. 23 Heating at 180 °C crystallizes the gismondine structure with
impurities of faujasite. Penzarela et al. have also synthesized zeolite Y using the chosen
recipe under microwave irradiation. 20
The approximate Si/Al ratio of the synthesized material determined by calculating the a
parameter of the cubic crystal system and compared to literature values 24 was close to
3.00 based on our calculated a parameters that ranged from 24.4 -24.8A (Table 6-1).
Using this basis and the Si+Al = 192 per 384 oxygen atoms in an ideal system give the
composition of this material as: Na48Al4gSi|440384-240H 2 0.
Successful ion-exchange was confirmed by XRD and elemental analysis studies. XRD of
the ion-exchanged sample reveals a change in the d-spacing as can be seen in Figures 6-9
and 6-10 and the peak analyses in Tables 6-1 and 6-2. Reduction of the nickel ions to
162
elemental nickel has an effect of slightly changing the d-spacing (Figure 6-10 and Table
6-2). The nickel-loaded materials also show the same effect of change in d-spacing but
only to a small extent. In both cases the d-spacing does not revert to original values
confirming the effects of ion exchange and dehydration on the structure.
Elemental analyses confirm the presence of nickel ions in the exchanged materials.
The EDX analysis (Figures 6-13 and 6-14) reveals the presence of nickel in the nickelexchanged samples and not in as-synthesized samples. Based on the EDX results, the
Si/Al ratio is approximated to be 2.50 which is a big variation to the value close to 3.00
given by relying on the a parameter.
Dooryhee et al.25 have done extensive studies on nickel exchanged and reduced NaY.
From their reported structural information on atomic positions and site occupancies, we
generated a simulated XRD pattern which we used to compare with our exchanged
materials. Both patterns have an almost perfect match indicating similar environments in
both. They have also identified using EXAFS the location of nickel in the variously
treated zeolite Y with good precision. This work can be relied on to predict the location
of nickel in our materials. In addition to the above, we performed TEM on the reduced
NaNiY with the resultant micrographs indicating the well dispersion of nickel metal on
the surface of the zeolite.
6.6
Glycerol steam reforming
Steam reforming of glycerol was studied at three different temperatures with all other
variables held constant. The results of these studies are displayed in Figures 6-14 to 6-19.
The hydrogen yield at 550 °C increases progressively from a yield of slightly less than
l(mol/mol) at 15 min and levels off at about 3.5 in a time period of 3.5 h. At 600 °C,
163
hydrogen yield increases from about 1 (mol/mol) to 5 in the first 3.5 h. At 650 °C, the
yield increases from about 1 (mol/mol) to about 7 in the first 3.5 h. The dry products
stream composition was calculated from quantifiable gaseous products at every
temperature studied. At 550 °C, the hydrogen composition starts off at about 60% of the
dry stream after 15 min increasing slightly to above 60% in the second analysis and
tending to stay at this level for the duration of the study. This is the trend for CH4 also but
at a lower percentage of less than 5%. C0 2 increases slightly from about 22% to 30% and
maintains around this level throughout. CO on the other hand decreases from 17% at 15
min to less than 10% by the end of the study. There is no detectable ethylene at this
temperature level. At 600 °C, the hydrogen composition starts off at about 60% of the dry
stream after 15 min and stays constant for the duration of the study. This is the case for
CH4. CO2
increases slightly from about 22% to 30% and maintains around this level
throughout. CO on the other hand decreases from 17% at 15 min to less than 10% by the
end of the study. There is no detectable ethylene at this temperature level.
At 650 °C, the hydrogen composition starts off at about 52% of the dry stream after 15
min and stays constant for the duration of study. This is the trend for
(-17%), and
CO2
CH4
(3-5%), CO
(25 - 29%). However, unlike the previous two temperature levels, there
is emergence of C2H4 after 2 h 15 min of reaction time at about 1% of the dry products
stream, which was not detected before.
Steam reforming of glycerol is often accompanied by glycerol decomposition or water
gas shift reaction, Equations 6-4 and 6-5. 26 Other possible reactions are shown in
Equations 6-6 to 6-13 which may also proceed together with the glycerol steam
reforming process.
164
6H 2(g ) + 2C0 2 , g )
C2H50H(g) + 3H20(g)
Equation 6-2. Ethanol steam reforming
C3H803(g) + 3H20(g)
-
7H2(g) + 3C02(g)
Equation 6-3. Glycerol steam reforming
C3H803(g) -> 3 CO(g) + 4 H2(g)
Equation 6-4. Glycerol decomposition
CO(g) + H20(g) -> C02(g)+ H2(g)
Equation 6-5. Water-gas-shift reaction
Equation 6-6.
2C3H803(g) + H2tg) -> 3CH4(g) + 3CO(g> + 3H20(g)
Equation 6-7.
C3H803,g) + 2H2(g) -> 2CH4(g) + CO(g) + 2H20(g)
Equation 6-8.
CO(g)+ 3H2 (g) ^ CH (g) + H20(g)
Equation 6-9.
C0 2(g) + 4H 2 {g) -» CH 4(g ) + 2H 2 0 (g)
Equation 6-10.
C0 2(g) + CH 4 ( g ) -> 2CO (g ) + 2H 2(g)
Equation 6-11.
CH4(g)+ -> 2 H2(g)+ C(S)
Equation 6-12.
C(s) + H20(g)-^ CO(g) + H2(g)
Equation 6-13.
2CO( g) -> C(S) + C0 2(g)
4
165
6.7
Conclusion
Zeolite Y was successfully synthesized using a microwave hydrothermal process. Our
synthesis conditions optimized by ageing the synthesis gel for 16 h prior to the synthesis
and carrying out the synthesis for 1 h in the microwave led to successful crystallization of
NaY at 90 °C, 120 °C and 150°C.
Glycerol steam reforming was successfully performed using Ni-loaded NaY as catalysts.
Nickel loaded 10 % zeolite Y showed good performance with good yields of hydrogen
gas.
166
6.8
References
1.
Sun, J., Qiu, X.-P.,Wu, F„ & Zhu,W.-T. Int. J. Hydrogen Energ., 2005, 30,
437.
2.
Costamagna P, Srinivasan S. J. Power Sources, 2001, 102, (1-2), 253.
3.
Song S Q, Tsiakaras P. Appl. Catal. B: Environmental, 2006, 63 (3-4): 187
4.
Galvita V V, Semin G L, Belyaev V D, Semikolenov V A, Tsiakaras P,
Sobyanin V A. Appl. Catal. A: General, 2001, 220, 123.
5.
Breen J., P., Burch R., Coleman H. M. Appl. Catal. B: Environmental, 2002,
39, 65.
6.
Zhang B. C„ Tang X, L„ Li Y„ Cai W. J., Xu Y. D„ Shen W. J. Catal.
Commun., 2006, 7, 367.
7.
Wen, G. D„ Xu, Y. P., Wei, Y„ Pei, R. Y„ Li, K. D„ Xu, Z. S„ Tian, Z. J. Chin.
J. Catal., 2008, 30, 830.
8.
Haga. F., Nakajima, T., Miya, H., Mishima, S. Catal. Lett., 1997, 48, 223.
9.
Llorca, J., Horns, N., Sales, J., Piscina, P. R. J. Catal., 2002, 209, 306.
10.
Llorca, J., Horns, N., Sales, J., Fierro, J. G., Piscina, P. R. J. Catal., 2004, 222,
470.
11.
Sahoo, D. R., Shilpi, V., Sanjay, P., Pant, K. K. Chem. Eng. J., 2007, 25, 139.
12.
Athanasios, N. F., Xenophon, E, V. J.Catal., 2004, 225, 439.
13.
Frusteri, F., Freni, S., Chiodo, V., Spadaro, L., Bonura, G., Cavallaro, S.
J. Power Sources, 2004, 132, 139.
14.
Jose, C., Fernando, M., Miguel, L., Norma, A. Chem. Eng. J., 2004, 98, 61.
15.
Andre, L. A., Mariana, M. V. M. S., Martin, S. Catal. Today, 2007, 123, 257.
16.
Mamontov, E., Egami, T. J. Phys. Chemistry Solids, 2000, 61, 1345.
17.
Gargi, D„ Umesh, V. W„ Tinku, B„ Hegde, M. S„ Priolkar, K. R„ Sarode, P.
R. Chem. Mater., 2006, 18, 3249.
18.
Frusteri, F., Frenia, S., Chiodo, V., Donato, S., Bonura, G., Cavallaro, S. Int.
J. Hydrogen Energy, 2006, 31, 2193.
19.
Meier, W. M„ and Olson, D. H., Atlas of Zeolite Structure Types, ButterworthHeinemann, London, 1992
20.
Panzarella, B., Tompsett, G., Conner, W. C., Jones, K. ChemPhysChem, 2007,
8 ,357
21.
Suryanarayana. C., Norton. G., M., X-Ray Diffraction: A Practical Approach,
New York: Plenum Press, 1998
22.
Gottardi, G., Galli, E., Natural Zeolites, 1985, 336
23.
Barlocher, C., Meier, W. Z., Kristallogr., Kristallgeom., Kristallphys.,
Kristallchem., 1972, 135,339
24.
Dempsey, E., Kuhl, H., and Olson, H., The Journal of Physical Chemistry, 73,
1969,387
25.
Dooryhee, E., Catlow, C. R. A., Couves, J. W., Maddox, P. J., Thomas, J. M.
Greaves, G. N., Steel, A. T. and Townsend, R. P. J. Phys. Chem. 1991, 95, 4514
26.
Cheng, C. K., Foo, S. Y., and Adesina, A. A., Ind. Eng. Chem. Res., 2010, 49,
10804
167
FUTURE WORK
There are opportunities for continued research in the areas presented in this thesis.
Firstly, microwave continuous synthesis can be explored further to widen the type of
materials that can be produced though the technique. Various approaches can be
investigated but the main guiding criteria should be whether the materials can be
crystallized in short time periods and whether they lead to clogging or not.
Secondly, the area of gas phase oxidation of inorganic molecules using K-OMS-2 is an
area that just opened up and so can be pursued further. In this work we have developed a
process for doing this where the best conditions have been determined. In the course of
our studies we realized that the catalyst structure is changing as the reaction proceeds.
This phenomenon can be studied using insitu techniques with the aim of reversing the
phenomenon. The material can be heated in the presence of oxygen to try to reverse this
or the reaction can be followed by one which oxidizes the manganese in the lower states.
Lastly, the use of nickel loaded zeolite Y in steam reforming of glycerol has scope for
further improvement and refinement. There are various parameters that need to be
studied, for instance; the location of nickel in the structure upon ion-exchange and on
impregnation and subsequent reduction processes, precise composition of the material
through running analytical tests e.g. ICP-AES for all the metals present. Further, the
steam reforming reactions need to be reproduced under the conditions chosen and time
studies done. This will answer questions like catalyst lifetime and stability. Furthermore,
when the above has been determined, regeneration studied can be designed. Another
fascinating aspect that can be attempted is the water splitting at elevated temperatures
using nickel loaded zeolite Y.
168
All these studies have preliminarily been reported and so only further studies are needed.
This means that anyone who furthers this work must give due credit by way of my name
appearing on any publication as the second author in the work.
169
Документ
Категория
Без категории
Просмотров
0
Размер файла
6 899 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа