University of Southern California Dissertations and Theses

Exploring new frontiers in catalysis: correlating crystal chemistry and activity in layered silicates

(USC Thesis Other)

Exploring new frontiers in catalysis: correlating crystal chemistry and activity in layered silicates

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content EXPLORING NEW FRONTIERS IN CATALYSIS: CORRELATING CRYSTAL
CHEMISTRY AND ACTIVITY IN LAYERED SILICATES
by
Erica S. Howard
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)
August 2019
Copyright 2019 Erica S. Howard
for my family
and all the kids who have never seen a scientist that looks
like them
ii
Acknowledgments
First and foremost, I want thank my advisor, Professor Brent C. Melot. Brent
gave me the opportunity to create something brand new, and guided my growth
as a scientist and critical thinker. Moreover, he taught me the power of self-
reliance and and self-confidence and for that I am so thankful.
I owe so much to the entire Melot group. More than a group, we were a
family and a support network; it was a pleasure to work with you all. From
group breakfast to sunsets on the beach, adventures to disneyland or just the
rose garden, trips to Santa Barbara and countless hours at the beamline, our time
together has made the past 5 years an unforgettable journey. I am so grateful for
the friendship that came with it.
A huge THANK YOU to Shiliang "Nemo" Zhou. He was my first grad school
mentor, and set the example I followed throughout my PhD. Furthermore, every
time Nemo taught me something new, he also inspired a little more confidence
in my own scientific abilities. Abbey Neer, thank you for being by day-one in LA,
your adventure-ready attitude and willingness to do anything for your friends
has been such a blessing. I’m so grateful to you for forcing me to get out of my
apartment and enjoy the city, for our cross-country roadtrips, and your guidance,
which has helped me more than you know. JoAnna Milam-Guerrero, thank you
for being my partner in crime, shoulder to lean on, and complaint secretary. Your
talks, gifs, jokes,spontaneous dance parties and out of the blue messages have
always been exactly the prescription the doctor ordered. Nick Bashian, thank
you for being a tremendous friend, go Bucks; you always know how to lighten
the mood. Believe it or not, your work and effort have also helped guide me over
the years. Laura Estergreen, thank you for having my back since the first day of
iii
grad school, always being there to teach me the hard science, and being a great
roommate.
Last but not least, many thanks to undergraduate and high school interns over
the years. Kyle Nolan, Lilly Shanahan, Sabrina Mir, Mari Rustebakke, Audrey
Franklin, Tiffany Yen, Nicole Spence, Samantha Abdel-Latif, Joseph Stiles, April
(Beatriz) Lopez-Burmudez, and Allyson Ee, I would not have been able to accom-
plish this without you guys! I would like to thank Lilly and Kyle especially for
several years in the synthetic-trenches with me, their work lead to the creation of
most of these compounds. Sabrina, it has been a joy to mentor you over the past
three years, you are always surprising me with your insight and understanding
for science and for life. Mari, your work ethic, friendship, and general awesome-
ness may be the only reason I survived my final semester; I’m so glad you chose
me to be your grad student. Watching all of their love for science grow over the
years motivated me through the hardest of days in the lab, thank you.
None of my this work would have been possible without the gracious help of
my collaborators and USC’s Chemistry Department. I am forever grateful to Judy
Fong, Magnolia Benitez, and Michele Dea for always being there to explain how
the real world work. Professor Sri Narayan, thank you for your guidance during
my Burg teaching fellowship. I would not have been able to conduct PDF studies
without the help of Dr. Kate Page and Dr. Jue Liu. Professor Raphaele Clement,
thank you so much for taking the time and energy to teach me ssNMR, it remains
a highlight of my graduate career.
Finally, to my family, thank you for always believing me and supporting me.
Having you guys behind me made each day easier. Jenny, I am so thankful to
you for always being there. Mom and Dad, I appreciate every phone call, even
though I only answer 1 in 5. Without you none of this would be possible. To
iv
my Mama and Grandma, our conversations have always guided my path. To my
aunts and uncles who never missed a chance to reach out just to tell me how
much I am loved, I am forever grateful. To my cousins and brothers, youhave to
call me Doctor now, and your friendship has been a lifelong light. Finally, saving
the best for last: Thank You to my best friend, Justin. Thank you for all of the
Disneyland trips, animal adoptions, and love and support you’ve given me over
the past years. I can’t wait to see what the future holds.
v
Curriculum Vitae
Education
2014-2019 Ph.D., Chemistry, Department of Chemistry
University of Southern California, Los Angeles, CA
2010-2014 B.S., Chemistry, Department of Chemistry and Biochemistry
The Ohio State University, Columbus, OH
US Patents Pending
1. E.S. Howard, B.C. Melot. Activated Phyllosilicate Clay Oxidation Catalyst.
United States Patent Application, 16/055,993.
2. E.S. Howard,R. Alamillo, B.C. Melot, L. Perez. Phyllosilicate Catalysts for
the Selective Reduction of Alkenes and Alkynes. United States Provisional
Patent Application.
Publications
1. E.S. Howard, R. Alamillo, L.M. Shanahan, D. Falcone, L. Perez, and B.C.
Melot; Catalytic Hydrogenation of Cyclohexene by Ni
3
−xZn
x
Si
2
O
5
(OH)
4
(inpreparation)
2. E.S. Howard, M. Rustebakke, A. Tadle, S. Mir, and B.C. Melot; Mild, Sol-
vent Free Oxidation of Benzyl Alcohol to Benzaldehyde by Molecular Oxy-
gen Using Synthetic Bimetallic Clay Catalysts (inpreparation)
vi
3. J. Milam-Guerrero, E.S. Howard, A.J. Neer, J. Stiles, N.H. Bashian, K.
Nolan, and B.C. Melot; Low Temperature Synthesis, Structure and Mag-
netic Properties of Layered Ferromagnetic Silicates (inpreparation)
4. E.S. Howard, A. Tadle, J. Liu, L.M. Shanahan, A.E. Nessl, K. Butler, A.
Regoutz, D. Payne, Z. Lu, T.J. Williams, and B.C. Melot; Structure, Reactiv-
ity and Activation of Novel Earth-abundant Oxidation Catalyst Ni
3
Si
2
O
5
(OH)
4
(inpreparation)
5. S. Zhou, E.S. Howard, J. Liu, N.H. Bashian, K. Nolan, S. Krishnamoor-
thy, G.M. Rangel, M.T. Sougrati, G.K. S. Prakash, K. Page, and B.C. Melot;
Hydrothermal Preparation, Crystal Chemistry, and Redox Properties of Iron
Muscovite Clay, ACS Appl. Mater. Interfaces, 2017, 9, 34024-34032.
vii
Exploring New Frontiers in Catalysis: Correlating Crystal Chemistry and Activity
in Layered Silicates
by
Erica S. Howard
viii
Contents
List of Tables xi
List of Figures xii
1 Catalysis, Energy and Materials 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 What is a Phyllosilicate? . . . . . . . . . . . . . . . . . . . . . . . 7
1.2.1 Crystal Structure . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.2 Preparation Methods . . . . . . . . . . . . . . . . . . . . . 14
1.2.3 Catalytic Studies of Phyllosilicates . . . . . . . . . . . . . 18
1.3 Sustainable Catalysis . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.4 Thesis overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2 Crystal Chemistry and Catalytic Properties of Synthetic
Muscovites, KFe
3
Si
4
O
10
(OH)
2
, KCo
3
Si
4
O
10
(OH)
2
and KZn
3
Si
4
O
10
(OH)
2
23
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 28
2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
3 Crystal Chemistry and Oxidation Properties of 1:1 Phyllosilicates
Ni
3
Si
2
O
5
(OH)
4
, Co
3
Si
2
O
5
(OH)
4
, and Mg
3
Si
2
O
5
(OH)
4
44
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 50
3.3.1 Preparation and Bulk Structure Analysis . . . . . . . . . . 50
ix
3.3.2 Local Structure Analysis. . . . . . . . . . . . . . . . . . . . 54
3.3.3 Reactivity studies. . . . . . . . . . . . . . . . . . . . . . . 56
3.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
4 Creation of Active sites on Ni
3
Si
2
O
5
(OH)
4
by Thermal Activation 61
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
4.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . 68
4.3.1 Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
4.3.2 Catalytic performance . . . . . . . . . . . . . . . . . . . . 79
4.3.3 Reaction pathways for oxidation of glycerol by as-prepared
and thermally activated catalyst . . . . . . . . . . . . . . . 82
4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5 Tailorable Frustrated Lewis Pair-like Active Sites on
Bimetallic 1:1 Phyllosilicate Catalysts 85
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
5.2 Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
5.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
5.3.1 Reduction of cyclohexene . . . . . . . . . . . . . . . . . . 90
5.3.2 Structure Analysis . . . . . . . . . . . . . . . . . . . . . . 90
5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6 Summary and Future Outlook 94
Reference List 97
x
List of Tables
1.1 Cations found in phyllosilicates . . . . . . . . . . . . . . . . . . . 12
1.2 Reported synthetic conditions for phyllosilicates . . . . . . . . . . 16
2.1 Correlation of Fe environments and KOH concentration . . . . . . 37
2.2 Distribution of Fe environments from PVP. . . . . . . . . . . . . . 37
3.1 Oxidation of benzyl alcohol by Ni, Co, and Mg 1:1 clays . . . . . 57
3.2 Optimization of benzyl alcohol oxidation conditions . . . . . . . . 57
4.1 Glycerol test conditions for as-prepared Ni
3
Si
2
O
5
(OH)
4
. . . . . . 80
4.2 Glycerol test conditions for activated Ni
3
Si
2
O
5
(OH)
4
. . . . . . . 80
4.3 Basic oxidation conditions for as-prepared Ni
3
Si
2
O
5
(OH)
4
. . . . 81
4.4 Basic oxidation conditions for activated Ni
3
Si
2
O
5
(OH)
4
. . . . . . 81
5.1 Reduction of cyclohexene . . . . . . . . . . . . . . . . . . . . . . 89
5.2 TPD of Ni, Mg and Zn containing clays . . . . . . . . . . . . . . . 90
xi
List of Figures
1.1 Average production and cost by industry subdivision . . . . . . . 4
1.2 Price comparison of platinum vs abundant resources . . . . . . . 5
1.3 Basic structure of the layered silicate . . . . . . . . . . . . . . . . 9
1.4 Comparison of 2:1 and 1:1 layered phyllosilicates . . . . . . . . . 10
1.5 Topology of dioctahedral and trioctahedral layers . . . . . . . . . 11
1.6 Comparison of silicate morphologies . . . . . . . . . . . . . . . . 13
1.7 Polymerization reaction between silicic acid and precursor salts . 17
2.1 Basic structure of the 2:1 silicate . . . . . . . . . . . . . . . . . . 28
2.2 Laboratory XRD of 2:1 muscovite . . . . . . . . . . . . . . . . . . 29
2.3 SEM and XRD fitting of KFe
3
Si
4
O
10
(OH)
2
. . . . . . . . . . . . . 30
2.4 Unfit PDF of 2:1 materials . . . . . . . . . . . . . . . . . . . . . . 32
2.5 Least-squares refinement of KFe
2.75
Si
3.25
O
10
(OH)
2
PDF . . . . . . 33
2.6 XRD of Fe-Phyllosilicates made with varying KOH ratios . . . . . 35
2.7 SEM of Fe-Phyllosilicates made with varying KOH ratios . . . . . 36
2.8 RT Mössbauer spectra of KFe . . . . . . . . . . . . . . . . . . . . 38
xii
2.9 1H NMR of PVP oxidation products . . . . . . . . . . . . . . . . . 39
2.10 Image of metal leached 2:1 silicates . . . . . . . . . . . . . . . . . 40
2.11 NMR of benzyl alcohol reactions . . . . . . . . . . . . . . . . . . 41
3.1 Oxidation of benzyllic alcohol scheme . . . . . . . . . . . . . . . 45
3.2 Depiction of the 1:1 phyllosilicate structure . . . . . . . . . . . . 46
3.3 TEM of Ni, Co, and Mg 1:1 clays . . . . . . . . . . . . . . . . . . 51
3.4 Laboratory XRD of Ni, Co, and Mg 1:1 clays . . . . . . . . . . . . 52
3.5 X-ray PDF fits of Ni, Co, and Mg 1:1 clays . . . . . . . . . . . . . 54
3.6 Benzyl alcohol oxidation pathway . . . . . . . . . . . . . . . . . . 59
4.1 Glycerol oxidation scheme . . . . . . . . . . . . . . . . . . . . . . 61
4.2 Depiction of 1:1 phyllosilicate building blocks . . . . . . . . . . . 62
4.3 TGA and TEM of as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
. . . 69
4.4 XRD of as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
. . . . . . . . . 71
4.5 PDF fits of as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
. . . . . . . 72
4.6 FTIR and SFG of as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
. . . . 75
4.7 XPS of the as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
. . . . . . . 76
4.8 PPMS measurements of Ni
3
Si
2
O
5
(OH)
4
. . . . . . . . . . . . . . . 77
4.9 Calculated band structure of Ni
3
Si
2
O
5
(OH)
4
. . . . . . . . . . . . 78
4.10 Calculated DOS for Ni
3
Si
2
O
5
(OH)
4
. . . . . . . . . . . . . . . . . 79
4.11 Proposed tandem reaction pathways with glycerol . . . . . . . . . 83
5.1 XRD and X-ray PDF for Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
series catalysts . . . 91
5.2 XRD and X-ray PDF for Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
series catalysts . . . 92
xiii
Exploring New Frontiers in Catalysis: Correlating Crystal Chemistry and Activity
in Layered Silicates
by
Erica S. Howard
xiv
Abstract
The rational design of catalysts was declared aHolyGrail of chemistry research
by the 1995AccountsofChemicalResearch articles by Breslow, [1] and Bard and
Fox.[2] Pursuant to this goal, the fundamental understanding of activity in het-
erogeneous catalysts is an outstanding problem in materials chemistry. Though
heterogeneous catalysts are utilized in more than 85% of industrial reactions
and account for 25% of industrial energy use, there is a lack of basic guide-
lines for rational catalyst design and synthesis. [3] This dissertation addresses a
range of topics chosen to develop of structure-property relationships using the-
ory and experimental data to understand adsorption, reactions, and desorption
processes.
To inform the understanding earth-abundant catalysts, we begin by exam-
ining 2:1 and 1:1 phyllosilicates as model frameworks. Their unique layered
structure represents one of the best opportunities for facilitating correlations
between local structure, composition, and catalytic activity on a well-defined sur-
face. Studied first, 2:1 phyllosilicates, display a highly tunable synthetic system
determined from a combination of x-ray and neutron diffraction, SEM, and Möss-
bauer spectroscopy. Despite promising characterization, these materials are poor
oxidation catalysts as a result of limited access to the surface-active site. Follow-
ing this, our investigation of several 1:1 phyllosilicates found only Ni
3
Si
2
O
5
(OH)
4
is active for the oxidation of benzyl alcohol. To our knowledge, this is the first
demonstration of activation of molecular oxygen by a 1:1 phyllosilicate.
Guided by previously reported surface dehydration mechanisms for these
compounds, we developed heat treatment techniques to tune the surface active
site. [4, 5, 6, 7, 8, 9] Next, we present the thermal treatment Ni
3
Si
2
O
5
(OH)
4
xv
and characterization of new catalytic sites formed on its surface, which are capa-
ble of oxidizing glycerol. Through local structure characterization including pair
distribution function analysis (PDF), Fourier transform spectroscopy, x-ray pho-
toelectron spectroscopy, DFT calculation and magnetic measurement, the effect
of anionic modification is elucidated. Specifically, the local structure determined
from PDF in combination with computational studies of these surfaces is used to
ascertain the role of surface coordination and oxidation state.
Replacement of aluminum by a transition metal allowed us to develop and
tune oxidation catalysis of these materials. The final study presented here investi-
gates both bimetallic substitution and thermal activation in series Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
. These phases introduce the opportunity to study coop-
erative catalytic synergy and a route to tailor the activity by systematically alter-
ing the composition. Through reduction of cyclohexene we demonstrate heat
treatment creates frustrated Lewis pair-like active sites on the surfaces capable
of binding and dissociating H
2
. This understanding from these collective stud-
ies can be used as guiding design principles to synthesize catalysts for biomass
transform, selective oxidation, and hydrogenation chemistry.
xvi
Chapter 1
Catalysis, Energy and Materials
As a result of the growing economic and environmental consequences of
excessive fossil fuel use, there is both societal and self-imposed scientific pres-
sure to develop greener industrial processes. The shift in energy trend from con-
ventional to sustainable resources emphasizes processes that minimize energy
consumption as well as use of hazardous substances. While the push to cre-
ate greener catalytic processes has resulted in the development of several highly
active precious metal catalysts [10, 11, 12] the use of earth-abundant resources
is imperative to sustainable, low-cost chemical processes. The high Earth abun-
dance of transition metals, metal oxides, and polyanionic compounds such as sil-
icates, phosphates, and aluminates prequalify them as desirable groups to func-
tionalize. [13] Likewise, transition metals such as cobalt, iron, and nickel are
promising candidates for clean energy processes, previously being utilized for
1
high impact reactions such as Fischer-Tropsch synthesis. [3, 14, 15] Unfortu-
nately, creation of similar technology is hindered by lack of fundamental under-
standing. [3, 16, 17] Though heterogeneous catalysts are utilized in more than
85% of industrial reactions and account for 25% of industrial energy consump-
tion, we lack basic guidelines for rational catalyst design.[3, 18]
There is a tremendous need for both the creation of novel catalysts and effec-
tive rules for their design. [3, 16] Designing earth-abundant catalysts is ham-
pered by the complex morphology of relevant materials. The work contained
in this thesis address topics within a theme of developing a model material to
understand what gives rise to catalytic activity, and create structure-property
relationships for solid catalysts. We have sought to describe the model system,
phyllosilicates, as a platform to explore the effects of morphology, acidity, compo-
sition, and oxidation state. The specific studies presented here aim to understand
the influence of thermal treatment and compositional variation, as well as exam-
ine a solid state frustrated Lewis pair. In light of the interest in phyllosilicates
and catalysis this chapter reviews the phyllosilicate chemistry.
1.1 Motivation
There is an exigent need to seek new ways of not only powering the planet
in a sustainable manner but also finding new routes to produce the chemicals
required for the maintenance of civilized life.[3, 16, 19, 20] Catalysis, a main-
stay of the chemicals and energy sectors, will be key in driving the transition
to sustainable industry.[3, 18, 20] The need for new catalysts and reactions
stems from the ever-increasing demand for goods and products that result from
2
their use. Processes that rely on catalysis range from the development of mod-
ern medicine [21]to innovative materials for clothing [22], construction and
packaging[23], and production of adequate food supplies [24], clean air [25],
water, and energy[26]. In fact, catalysis is so wide-reaching that it is used in
90% of industrial chemical reactions. [23] At present, the global catalyst busi-
ness represents a 20 billion-dollar per-year enterprise that is expected to grow
14 billion dollars over the next 6 years. [27] Currently, this results in a global
energy demand of 42 exajoules per year, equivalent to approximately 30% of the
total industrial energy demand worldwide. [20, 19] It is estimated improved cat-
alytic processes could reduce energy consumption by 40% , but if not curtailed
energy demand could increase by the same amount. [20, 19] Developing novel
catalysts with high efficiency and selectivity is critical for enabling clean energy
and sustainable industry. The design of economic, efficient, and reliable catalytic
processes to better the world is the ultimate objective.
The development of new catalysts is critical for supporting sustainable growth
of products such as plastics and high-grade polymers. [27] Currently, the chemi-
cals industry relies on only the seven petrochemicals listed in Figure 1.1 for the
development of nearly 90% of all bulk chemicals. [16, 20] Those are included
in a larger list of 18 chemicals that represent the highest demand (highest vol-
ume) chemicals (HVCs) as well as the most energy-intensive to produce.[20, 19]
More robust and economic catalysts are necessary to create value-added chem-
icals from viable feedstocks and make current processes more sustainable. In
recent years, the drive to create environmentally friendly routes to produce com-
modity chemicals has resulted in the development of catalytic processes that
rely on precious metals, such as platinum, palladium, and iridium. [10, 11, 12]
Platinum group metal catalysts are used frequently due to their high activity
3
Figure 1.1: Output of chemical industries divided by cost. Listed with them are
representative products of these industries. The center-most molecules are the seven
most utilized feestock chemicals. [16, 19]
and resistance to catalytic poisoning, but low natural abundance of these metals
means their use is prohibitively expensive on the industrial scale. [Figure 1.2]
[12, 28] Because platinum group metals are highly applicable their demand has
increased, while their relative production remains constant, resulting in rising
costs. [29] In comparison, silicon, the second most abundant element, forms sev-
eral naturally-occurring compositions with catalytically active metals, and costs
4
2012 2014 2016 2018
Year
10
100
1000
10000
Average $ / troy ounce
Average $ / troy ounce
Ir
Pd
Pt
Rh
Ru
2012 2014 2016 2018
Year
.01
.1
1
10
100
Average $ / lb
Average $ / lb
Co
Ni
Fe
Al
2
O
3
SiO
2
Figure 1.2: A)Average price in dollars per troy ounce of platinum group metals and
B) average price in dollars per pound of select Earth-abundant resources. One troy
ounce is equivalent to 0.0685714 pounds. Note the log scale.
only 0.02 USD per pound. [29] Even cobalt (38 USD/lb), arguably the most
highly demanded transition metal, is one hundred times less expensive than a
platinum group metal. [29] As a result, it is a foremost goal to minimize the use
of precious metals and create inexpensive catalytic processes.
Heterogeneous catalysts are preferred over their homogeneous counterparts
due to inherent practical advantages, including recovery, reusability, atom utility,
and enhanced stability. [16] Heterogeneous catalysts facilitate reactions at an
interface between their surface and the substrate at theactivesite. The nature of
the active site significantly influences the rate of reaction (catalyticactivity) and
the distribution of observed products (selectivity). Designing selective catalysts
is often limited by our capacity to understand the origin of catalytic activity in
solids and design systems that meet the previous criteria. The ability to study
5
many relevant oxide and polyanionic catalysts is limited by the ability to isolate
and study a single reactive site, or a single reactive surface. [30] As a result,
the active site for catalytic processes on oxides is still largely unknown for most
reactions.
Beyond the active site, tuning the catalyst system is an important design strat-
egy to improve activity. [31] Here, tuning refers to making small changes in
the chemical environment to institute changes in the overall reaction process.
Tuning allows catalysts to be methodically adjusted to improve performance by
means of small change in the structure or chemical configuration, resulting in
altered catalytic performance. [28, 32] In solids, the presence of corners, edges,
defects and local distortions influence both the activity and selectivity of the
catalyst. [28, 31, 32] The microstructure of the reactive site is also mechanisti-
cally deterministic. For example, the hexagonal pocket in montmorillonite clays
facilitates acid-catalyzed transformations by these structures. [17] Beyond the
microstructure, the macroscopic nature of the reaction also influences the activ-
ity and selectivity, including the bulk crystal structure, composition, use of a
cocatalyst, solvent effect, and the nearby presence of another active site possibly
resulting incatalyticsynergyordualfunctionality. [33] Currently, the governing
principle of how surface structure contributes to catalytic behavior is predomi-
nantly unknown. Instead, fine-tuning catalytic properties for a desired process
often involves high-throughput "lets see what works" methodology. [3, 30] As a
result, the development of rational design rules for solid catalysts is considered a
foremost goal. Pursuant to this goal, the fundamental understanding of activity
in heterogeneous catalysis is an outstanding problem in materials chemistry.
The development of fundamental correlations between catalytic activity, com-
position, atomic and electronic structure is imperative to catalysis by design.
6
[34, 35] To develop this knowledge we must identify model catalysts and plat-
form chemicals that may represent reactions involving a wide spectrum of cata-
lysts and substrates. [3, 30] In order to identify the active part of a catalyst, we
need to identify and study the role of the catalytically-active site. Well-defined
model catalysts allow theory to assist in the elucidation of reaction paths and aid
in the directed search for new catalytic materials. The foremost goal of this work
has been to establish a model platform that will allow this kind of study.
1.2 What is a Phyllosilicate?
Ubiquitous in nature, phyllosilicate clays have been used by humans since
prehistoric times due to their global distribution and diverse properties. [14]
Even today, they find widespread industry use as components of building mate-
rials, [36] insulation, [37] toothpaste, [38] ceramics, [39] adsorbents, [40] and
cosmetics. [41] In recent years the unique properties that arise as a result of
their nanostructure, such as plasticity, collective charge excitation, and extreme
flexibility, have caused a resurgence in phyllosilicate research. [14, 42, 43]
Such properties have resulted in studies that focus on phyllosilicates as super
capacitors[44, 45], lithium-ion battery electrodes [46, 47], and drug delivery
agents [48] across the fields of material science, chemistry, geology and geo-
chemistry.
It comes as no surprise many phyllosilicates also find use in catalysis. Subsec-
tion 1.2.3 provides a brief review of studies focused on the catalytic properties of
phyllosilicates. Previously, development of phyllosilicate catalysts has been hin-
dered by lack of synthetic methods to form highly crystalline materials and meth-
ods to describe the order of these approximately crystalline materials.[49, 50]
7
This work develops controllable hydrothermal techniques to make homogeneous
materials and describe their local environment using probes such as pair distri-
bution function analysis. This body of work shows, when well understood, their
lamellar structure represents one of the best opportunities for developing cor-
relations between binding energy, structure, and composition on a well-defined
surface.
1.2.1 Crystal Structure
The structure of phyllosilicate can be partially derived from its name, which
translates assheetsilicate. Phyllosilicates are silicate minerals composed of alter-
nating silicate and metal oxide sheets. The silicate sheet consists of SiO
4
tetra-
hedra joined at three corners to form the six-membered rings seen in Figure 1.3.
These rings connect in a hexagonal array to extend indefinitely in the ab plane,
as depicted in Figure 1.3. The apical (free) oxygen of a silicate sheet always
points in the same direction: towards the metal layer. The metal sits at the cen-
ter of a 6-coordinate octahedra, which lies on a triangular face, oblique to the the
silicate sheets. The octahedra connect at their edges to form a psuedo-hexagonal
array that matches the silicate sheet, also creating an infinitely extending sheet
in the ab plane. The sheets are joined at the apical oxygen, thus completing
the metal octahedra. At the surface the octahedra is most often coordinated by
hydroxyls; where the bottom face is not completed by silicate, at the center of the
hexagonal hole, an undistorted hydroxyl group will sit creating a three fold axis
as seen in Figure 1.4c. [51] The sheets are joined either by hydrogen bonding or
electrostatic interactions with an interlayer species forming a layered structure.
The nature of the structure dictates that [00l] facets are almost always exposed,
creating a known reactive surface. [51]
8
(b) (a)
(c)
Figure 1.3: The fundamental structural building blocks of the phyllosilicate depicted
alongtheabplane. (a)topologyofthesilicatesheetcomposedofcorner-sharingsilicate
tetrahedra (gray). (b) the metal sheet composed of edge-sharing metal octahedra
(teal). (c) a depiction of the layer connectivity and the hexagonal hole. Orange and
black spheres represent oxygen and hydrogen respectively.
Phyllosilicates are subdivided into two major groups, 1:1 phyllosilicates,
which contain 1 tetrahedral sheet for every 1 octahedral sheet, and 2:1 phyl-
losilicates, which contain 2 tetrahedral sheets for every 1 octahedral sheet. The
structure is accompanied by an interlayer space that may be filled with a neutral
species or ion if the structure needs to be charge balanced. The two main struc-
tures are built up corresponding to the following formulations: M
3
Si
2
O
5
(OH)
4
9
(a) (b)
Figure 1.4: (a) Illustration of a unit cell of 1:1 phyllosilicate as compared to that of a
(b) 2:1 unit cell. As illustrated, the metal layer of the 1:1 phyllosilicate is coordinated
by only 1 silicate sheet, while it is sandwiched between two in the 2:1 unit cell (b).
As can bee seen, the 2:1 phyllosilicate often contains an interlayer cation for charge
balance.
and AM
3
Si
4
O
10
(OH)
2
. [51] These structures, while not very intricate, boast both
a wide compositional variety as well as a large number of sites that can be altered
to tune the properties. [14, 51, 52] Additionally, the interlayer species may be
exchanged to promote activity or structural changes, and the surface exchanged
for fluorine or chlorine. [14, 51] By substituting into these sites the catalytic
properties may be tailored to study the catalytic synergy within the structure,
and a family of catalysts can be prepared and modified for various reactions.
Within these groups the phyllosilicate structure can be further subdivided
based on their layer occupancy. The pseudo-hexagonal symmetry of the octahe-
dral layer yields three equivalent sites, depicted in Figure 1.5a. Several divalent
(Mg
+2
, Fe
+2
, Ni
+2
) and trivalent metals (Al
+3
, Fe
+3
) are found to occupy the
metal sites, and Table 1.1 reviews the atoms previously incorporated into the
10
(a) (b)
(c) (d)
Figure 1.5: (a) A cartoon representation of an ideal trioctahedral layer coordination
and (b) an idea dioctahedral layer coordination. (c) depicts the bond connectivity of
a filled metal layer as compared to (d) a layer with an octahedral hole.
phyllosilicate structure. [14, 52, 53] Phyllosilicates with octahedral layers occu-
pied exclusively with divalent cations are referred to as trioctahedral materials
(Fig 1.5a). Replacing divalent metals with two trivalent metals creates a vacancy,
and are referred to as a dioctahedral phyllosilicates (Fig 1.5b). The trioctahe-
dral and dioctahedral phyllosilicates sometimes differ in the angle of the basal
hydroxyl groups due to the misfit of the tetrahedral and octahedral sheets neces-
sitated by either structure. Flexibility between trioctahedral and dioctahedral
structures allows aliovalent substitution into isomorphous structures, resulting
in the creation of exotic solid solutions.
11
Cation InterlayerA-site OctahedralM-site TetrahedralT-site
Li ×
Na × ⊗
Mg × ×
Al × ⊗
Si ×
K × ⊗
Ca × ⊗
Ti ⊗ ⊗ ⊗
Mn ⊗
Fe × ⊗
Co × ⊗
Ni ×
Cu ×
Zn × ⊗
Ge ⊗
Table 1.1: A summary of the cations reported in phyllosilicates showcases their com-
positional variability. When a× is shown it refers to the particular cation’s ability to
occupy the site in whole integers. Whereas a⊗ describes when the specific cation only
occupies this site in small fractional amounts. A blank indicates the cation has not
been reported on the sublattice. [51, 52]
Finally, phyllosilicates offer unique morphological control. Phyllosilicates can
be selectively encouraged to take a planar, rolled, helical, or nanotubular struc-
ture. [14, 54, 55] Additionally, the growth of the particle can be determined by
controlling interlayer strain. [56] Interlayer strain is intrinsic to these materials
and caused by the size mismatch between the shared oxygen of the silicate tetra-
hedra and the metal octahedra, as well as the materials natural desire to satisfy
dangling bonds. [57] We demonstrate in chapter 3 the degree and direction of
mismatch varies by the size of the octahedral layer, which is dictated by the size
of the ion filling theM position. [56] Regardless, the metal oxyhydroxide layers
are usually quite rigid, whereas the tetrahedral layer is flexible. Twisting of the
hexagonal rings formed by the silicate tetrahedra can accommodate the lattice
12
(a)
(b)
(c)
Figure 1.6: a)An illustration of a flat silicate sheet without strain. (b) A cartoon
representing collective expansion of the silicate layer, leading to a tubular or scrolled
morphology. (c) A cartoon depicting rocking of the silicate tetrahedra with respect
to one another, leading to a corrugated or helical structure. The morphology can be
controlled directly by the mismatch induced between the octahedral and tetrahedral
layers.
difference between the octahedral and tetrahedral layers. In the event the octa-
hedral layer is smaller, such is the case in Muscovite and Kaolinte, the tetrahedral
rings may shrink by rotating and tilting the individual tetrahedra with respect to
their neighbor leading to an extended rocking or puckering of the layers similar
to corrugation (Figure 1.6c). [56, 58] When the octahedral layer is larger the
opposite occurs, the tetrahedral layers expand to complete the connectivity. This
expansion sometimes results in uniform tilting (Figure 1.6b) causing the layers
13
to curl, as is seen in Asbestos. [42, 54] These defects are consistent across the
entire structure and averaged out in bulk analysis techniques such as diffraction.
[53]
The development of structure-property relationships using theory and exper-
imental data to understand the adsorption, complex reactions and desorption
processes is a prominent research goal. We aim to correlate trends between the
physical properties of a material with the way it behaves during a reaction. This
is of particular interest in catalysis as several properties can be directly related
to the activity or selectivity of a given reaction. Design of new and sophisticated
catalysts requires linking trends to current knowledge of successful systems. The
precise modification of the catalyst surface by introducing another component or
changing morphology facilitates controlled tuning of catalytic properties.
1.2.2 Preparation Methods
Given the fundamental nature of this work it is necessary to show the phyl-
losilicate can be both judiciously designed and studied. Naturally occurring min-
erals suffer from several drawbacks, such as incorporation of impurities, random
composition, and uncontrolled variations in structure. [49, 50, 59] Inconsisten-
cies among mined silicates limit the uses of these clays; however, phase pure and
homogeneous materials can be obtained through controllable synthesis. Syn-
thetic clay minerals with deliberate composition and structure serve as models
for fundamental studies that need homogeneous, well-defined samples.
Geological silicates are formed in conditions that prevail in the Earth’s man-
tle. Here, silicates grow at high temperatures and pressures, often by way of
14
hydrothermal diagenesis. [60] Mimicking this environment in a low-cost, low-
energy way has made synthetic development of these materials particularly diffi-
cult. Due to the high interest in these materials and their properties, the control-
lable synthesis of phyllosilicates has been highly sought after. [49, 61] Often,
reported conditions require high temperature, high pressure synthetic condi-
tions, and sacrifice morphology for phase-purity or vice versa. [49, 55, 62] Meth-
ods range from solid-state synthesis to exclusively solution techniques, the most
commonly used being hydrothermal.
Synthetic regimes for silicates fall into 4 general classifications, separated by
the temperature and pressure range of reaction conditions, as outlined in Table
1.2. Most successful techniques fall into the moderate regime; however, the work
contained in this thesis has targeted methods considered mild. When considering
development of sustainable technology it is important to consider the environ-
mental impact of not only the catalysis but also the production of materials. Our
goal has been to create methods that obtain highly uniform samples in the short-
est time at the lowest possible temperature. This thesis contributes several low
energy (low temperature, subcritical) solvothermal routes for highly crystalline
silicates, which include low incubation temperature and oven residence times to
achieve uniformity.
One goal was to optimize preparation methods to achieve highly crystalline,
uniform samples necessary to probe catalytic activity. Three routes to achieve
phase pure, morphologically desirable materials are identified: direct synthe-
sis, gel route, and osmotically-controlled teflon pouch. Additionally, a method
to increase the scale by a factor of twenty, facilitating industrial scale produc-
tion, was developed. In hydrothermal synthesis several factors contribute to the
15
Regime Temperature Pressure Mineralizer
Ambient 100
◦
C Ambient
Mild 100-240
◦
C Subcritical
Moderate 240-1000
◦
C <10 KBar
Extreme <1000
◦
C >10 Kbar
Toxic Any Any HF
Table 1.2: A summary of the temperature and pressure ranges of successful synthetic
regimes for phyllosilicates. All techniques noted are hydrothermal. Here, techniques
are separated by the use of HF. All techniques developed in this thesis fall into the mild
regime. [14, 49]
outcome of the reaction: precursor choice and coordination chemistry, struc-
ture directing agents, mineralizer choice, pH dependence, dwell time, reaction
temperature, order of operations, and stoichiometry. [63] Methods usually vary
based on desired structure and morphology, in addition to composition. The use
of mineralizers, pH choice, and silica source is developed from both the metal
ion of choice, as well as the conditions favorable for phyllosilicate growth.
Formation of undesired competing phases such as oxyhydroxides, amorphous
silicates, or more stable clays is often the greatest hurdle. [49] Metal precursors
should be soluble in the solvent of choice either at room temperature or sub-
critical conditions. With a wide variety of precursors meeting this criteria it is
prudent to make a choice with beneficial solution properties. For example, chlo-
rides and sulfates often form competing complexes, whereas nitrates support the
metal in solution and do not form competing phases. [64] Similarly, pH must be
controlled to support the solution state metal ions without encouraging alternate
phases. [65]
Formation of 2:1 phyllosilicates tends to be favored over their 1:1 counter-
parts because of their structure, which minimizes strain and allows for three-
dimensional connectivity. Because of this, phases such as Muscovite can be
16
Si
O O
O
O
Si
Ni
Si
Ni
O
O
O
O
O
O
OH
HSi
Ni
Si
O
OH
OH
O
O
Ni
Ni
O
HO
O
O
O
O
Si
Si
HO
OH
Ni
O O
HO
OH
Ni
NO
3
Si
Si
O
HO OH
Na
+
OH
-
Ni
O
3
N
NO
3
Na
+
OH
-
HO
OH
Si
O
HO OH
Na
+
OH
-
Na
+
OH
-
NO
3
Si
O
HO OH
Na
+
OH
-
Na
+
OH
-
Si
O
HO OH
Na
+
OH
-
Si
O
O OH
H
O
H
NO
3
Ni
NO
3
Na
+
Ni
O
Si
O
O
HO
Ni
Si
NO
3
O
OH
OH
initial reaction
polymerization
Figure 1.7: The mechanism by which silicic acid polymerizes is illustrated. The
reaction is initiated by the addition of base to a mixture of silicic acid and metal salt
in water. Initially the base deprotonates the acid, resulting in a condensation reaction
between the nucleophilic oxygen and a positive center.
17
accessed through the direct synthetic route. Different methods to achieve the
nominal composition KM
+2,+3
2
Si
4
O
10
(OH)
2
are explored in Chapter 2. The
role of a surfactant, mineralizer (KOH vs NaOH), and concentration (0, 5mm,
10mm, 20mm) were investigated in Muscovite synthesis. Synthetic conditions
that allowed for the control of ion distribution, crystallinity, and oxidation state
are discussed. Other silicate phases favored to form quickly can also be made
using thedirect hydrothermal routes
Due to strain, the growth of 1:1 phyllosilicates must be approached more
carefully. Most commonly, silica polymerization is used to create a highly-
coordinated precursor mixture of silicate and the metal ion. Pre-correlation helps
to curtail the development of impurities. To form the precursor, stoichiometric
amounts of silicic acid and metal salt are combined in water. Silica polymeriza-
tion is induced by the addition of base and proceeds following the mechanism
outlined in Figure 1.7. [66] The resulting silica gel is allowed to rest at least
50 hours before being sealed in a teflon lined stainless steel autoclave and being
placed in an oven a until crystallinity is reached. The length of gelation, oven
temperature, and dwell time are determined by the nature of the metal cation in
solution. The development of mixed phases, as well as incorporation of metals
with easily accessible oxidation states in water, requires diligence. Methods to
incorporate a variety of transition metals are discussed in chapters three and five.
1.2.3 Catalytic Studies of Phyllosilicates
Though phyllosilicate and phyllosilicate derived materials have recently
gained popularity as catalysts studies that focus on the catalytic properties of a
phyllosilicate are relatively few. [14] Earliest studies focused on aluminum based
silicate clays utilized as cracking catalysts, and found reactivity was facilitated by
18
uncompensated Lewis-acidic edge sites. [67] Additionally, the microstructure of
the 2:1 phyllosilicate hydroxyl encourages acid-catalyzed transformations. [67]
Following this, 2:1 phyllosilicates have been popularized both as catalysts and
catalytic supports, sometimes incorporating free transition metals into the inter-
layer. [68]
More recently, catalytic studies have focused on materials created when phyl-
losilicates are decomposed to a more active form. [15, 44, 69, 70, 71, 72, 73]
Phyllosilicate-derived materials have shown activity for electrochemical water
splitting, [44, 69] Suzuki coupling reactions, [74] methane reforming, [70, 71]
and hydrogenation. [75] Some examples include Jin et al’s work on hierar-
chical flower-like particles of Ni
3
Si
2
O
5
(OH)
4
, which creates an ideal support
for silver nanoparticles to perform 4-nitrophenol reduction.[73] Additionally,
cobalt-silica hybrid nanocatalysts created from the destruction of Co
3
Si
2
O
5
(OH)
4
were demonstrated by Park et al to show high conversion for Fischer-Tropsch
synthesis.[15] While these results don’t invoke the crystal structure of phyllosili-
cates they indicate these phases have potential that should be further explored.
1.3 Sustainable Catalysis
Periodically the International Energy Association (IEA) outlines several key
actions that can be taken to lower international energy usage and environmental
impact. [19] In 2013 they produced a catalysis roadmap that targets reforming
the most energy intensive reactions and highest demand products. [17, 20, 19]
The chemical and petrochemical sectors represent the most energy intensive,
and within these sectors improved processing could result in an energy savings
greater than 10.56 exajoules per year. [19] Beyond energy, there is a urgent
19
need to develop processes that reduce the use of harsh or corrosive materials,
and toxic solvents, in order to comply with ever increasing environmental reg-
ulations. [16, 17] Where possible it would be beneficial to move to renewable
feedstocks, and develop processes which rely on benign oxidants, such as molec-
ular oxygen or air. The IEA highlights the use of renewable resources, such as
biomass, syngas (H
2
and CO
2
), and air (N
2
, O
2
, CO
2
). The final goal is to enable
this technology with environmentally-friendly catalysts, which are both energet-
ically and monetarily inexpensive to produce. This thesis aims to contribute a
platform that allows the creation of a broad set of design rules that aid future
catalyst development.
Selective oxidation. Beyond petrochemical transformation and polymeriza-
tion chemistry, oxidation is the most used reaction process, comprising one step
of almost all reactions in the chemical sector. [76] Currently, industrial oxi-
dation processes suffer from the use of oxidants that are harsh, corrosive, and
oftentimes toxic, being used in stoichiometric proportions. [16, 76] The use of
molecular oxygen, or air, as a cheap and readily available oxidant in correlation
with benign heterogeneous catalysts would be hugely beneficial. Nickel contain-
ing hydrotalcites have been found to activate oxygen at nickel metal centers to
facilitate oxidation. [77] Futhermore, oxides which posses labile lattice oxygen
are capable of using structural oxygen to facilitate reactions followed by structure
reoxidation by molecular oxygen. [33] In fact, within the study of electrocataly-
sis, there is a wide volume of work that shows that metal oxyhydroxides utilize
lattice oxidation mechanism during oxygen-evolution. [33] Phyllosilicates, being
very similar to hydrotalcites and oxyhydroxides, represent a platform for directly
investigating the activation of molecular oxygen, and the chemical and structural
origin of such a lattice oxidation mechanism.
20
Clean fuels and small molecule activation. Fossil fuels supply both the
majority of energy and chemicals to the world. Outside of transportation, indus-
try represents the largest user of petroleum (24% ) and natural gas (35% ).
[19, 20] Within industry, the chemical sector represents the largest user, con-
suming 28% of fossil fuels, where 10% of natural gas and 45% of petroleum
are used exclusively as chemical feedstocks. [20] Increasing demand, limited
resources, and the growing concern for the environmental consequences of our
actions has prompted a shift from the most demanded feedstock chemicals the
towards renewable fuels and feedstocks. As it is difficult to perform the current
reactions in a more energetically favorable manner the obvious alternative is to
develop methods to harness less energetically intensive feedstocks. [20]
Using earth-abundant catalysts to enable less energy-intensive reactions is
a promising avenue for change. Most often, heterogeneous catalysts that can
activate small molecules rely on noble metals. While non-noble metals have
been used they often consist of poorly understood nanoparticles, and prompt
safety concerns. [11] Instead, we focus on creation of a solid frustrated Lewis
pair to polarize and activate small molecules on the phyllosilicate surface. There
a number of studies that support that the idea that thermal treatment of these
clays leads to the loss of surface water, resulting in a bare Lewis-basic oxygen and
a Lewis-acidic metal center. [4] Through these studies we discovered thermally
treated clays are able to activate the small molecules.
21
1.4 Thesis overview
The collection of works presented in this thesis expanded the applications and
understanding of heterogeneous catalysis in the context of creating structure-
property rules to develop Earth-abundant catalysts. Chapter 2 and 3 establish
criteria for active phyllosilicate catalysts by examining several 2:1 and 1:1 phyl-
losilicates. Chapter 2 explores the crystal chemistry of 2:1 phyllosilicates, demon-
strating synthetic methods to create highly crystalline materials, and character-
ization methods used to elucidate the nature the surface, ultimately showing
synthetic Muscovite is not active for oxidation. The next three chapters focus on
the crystal chemistry of 1:1 phyllosilicates, which demonstrate excellent poten-
tial as catalysts. Chapter 3 considers the oxidation properties and structure of
1:1 phyllosilicates Ni
3
Si
2
O
5
(OH)
4
, Co
3
Si
2
O
5
(OH)
4
, and Mg
3
Si
2
O
5
(OH)
4
. The
1:1 material Ni
3
Si
2
O
5
(OH)
4
is shown to be highly active for the oxidation of
benzyl alcohol using O
2
as the only oxidant. To the best of our knowledge this is
the first demonstration of the activation of oxygen by a phyllosilicate clay.
The effect of thermal treatment on the acidity and catalytic nature of the
surface hydroxyl in compound is probed in Chapter 4. This work provides evi-
dence that calcined phyllosilicates undergo hydrolysis to create surface active
sites which act like metal nanoparticles that are incorporated into the bulk struc-
ture. Furthermore, it shows that these compounds may utilize molecular oxygen
and produce hydrogen upon oxidation of glycerol. The ability of this material to
activate hydrogen validates their use as a frustrated Lewis pair model. Finally,
Chapter 6 discusses the future outlook of these studies. This work demonstrates
that phyllosilicates represent a family that can be modified to create a family of
catalysts using rapid, low-cost processing.
22
Chapter 2
Crystal Chemistry and Catalytic
Properties of Synthetic
Muscovites, KFe
3
Si
4
O
10
(OH)
2
,
KCo
3
Si
4
O
10
(OH)
2
and KZn
3
Si
4
O
10
(OH)
2
2.1 Introduction
Since the industrial revolution there has been an ever-present need to develop
sustainable ways to achieve chemical processes that create the molecules neces-
sary for civilized life. While several reports call attention the price and scarcitiy
23
of platinum group metals, more worrying is the lack basic design rules for sus-
tainable alternatives. [3, 16] A shift towards catalysts based on Earth-abundant
materials will not only help to address the issues of manufacturing cost, but also
ensure that production can efficiently scale to the levels needed to meet global
demand. In this regard, transition metal silicates, which in many ways are the
very definition of sustainable materials, offer clear advantages. [13]
Like many other polyanionic materials, transition metal silicates are a hugely
diverse family of structures and compositions, which allow incorporation of
many catalytically-active metals. Interest in their catalytic chemistry has pri-
marily focused on montmorillonite, (Na,Ca)
0.33
(Al,Mg)
2
Si
4
O
10
(OH)
2
, which is
utilized as an acid catalyst. [51] Building on reports of silicates being used as
battery materials, our group synthesized several transition metal silicates to test
their electrochemical performance. [47] While their battery performance was
poor, our observation of its surface redox properties and ability to liberate O
2
from hydrogen peroxide inspired us to explore its use as an oxidation catalyst.
The phyllosilicate family of minerals have great potential as a platform under-
stand oxidation chemistry due to their well-segregated layers of transition metals
and silicate groups. This planar structure allows for making correlations between
activity and the exposed surface. In the following, we describe the development
of a highly versatile hydrothermal reaction that allows for efficient control over
the particle morphology and cation distribution through the complex network of
crystallographic positions. Finally, their oxidative properties are described under
a number of conditions.
24
2.2 Experimental
Synthesis In a typical preparation, of KFe
3
Si
4
O
10
(OH)
2
0.020 mol KOH,
0.004 mol SiO
2
(fumed silica, 0.2-0.3μm average particle size), 0.002 mol
Fe(NO
3
)
3
·9H
2
O (K/Fe = 10/1), and 1.00 g polyvinylpyrrolidone (PVP) (average
molecular weight = 10,000 g/mol) were combined in 15 mL of deionized water.
The resultant red slurry was transferred into a 23 mL Teflon-lined stainless steel
autoclave, sealed and maintained at 220
◦
C for 16-18 hours. Once the autoclave
was cooled, a dark green powder was collected by vacuum filtration, washed
with distilled water, and dried at 110
◦
C for 30 mins. When varying amount
of one precursor was chosen to study its effect on the preparation, loading of all
other precursors were kept the same as stated above. The same method was used
for the synthesis of KCo
3
Si
4
O
10
(OH)
2
and KZn
3
Si
4
O
10
(OH)
2
with the following
exception: KZn
3
Si
4
O
10
(OH)
2
must be maintained at 220
◦
C for 5 days.
Physical Characterization.
Elemental analysis was performed in-house using a Thermo Scientific iCAP
7000 inductively coupled plasma-optical emission spectroscometer(ICP-OES).
Roughly 2.0 mg of sample powder was digested with 10% HNO
3
and 2% HF in
a polyethylene volumetric flask, with each sample being measured three times,
and the presented values representing the average over all measurements. Lab-
oratory X-ray diffraction patterns were collected on a Bruker D8 diffractometer
with a Co
Kα
source (λ
1
= 1.78897 Å, λ
2
= 1.79285 Å), equipped with a Lynx-
eye detector. High resolution synchrotron powder diffraction data were collected
using beamline 11-BM at the Advanced Photon Source (APS), Argonne National
Laboratory using an average wavelength of 0.413682 Å. Discrete detectors cov-
ering an angular range from -6 to 16
◦
2θ were scanned over a 34
◦
2θ range,
25
with data points collected every 0.001
◦
2θ at a scan speed of 0.01
◦
s
−1
. Neutron
pair distribution function (PDF) data were collected on the NOMAD beamline
at the Spallation Neutron Source at Oak Ridge National Laboratory at 60 Hz
setting.[78]
Le Bail fits and Rietveld refinement of the structure were carried out
using the TOPAS software suite (version 6) using the fundamental parameter
approach.[79] Anisotropic broadening of the Bragg reflections due to strain was
modeled using the Stephen’s model.[80] A numerical method was used to char-
acterize the honeycomb stacking faults in the synthesized silicates with a 30-layer
supercell. The local structure was refined using either least-square refinements
or simulated annealing of the initial Rietveld model using total scattering neu-
tron data. Structures and charge density were visualized using VESTA.[81]
57
Fe Mössbauer spectra were collected in the transmission geometry with a
source of
57
Co in Rhodium metal. During the measurements, both the source
and the absorber were kept at ambient temperature (294 K). The spectrome-
ter was operated with a triangular velocity waveform. The velocity scale was
calibrated with the magnetically split sextet spectrum of a high-purity α-Fe foil
as the reference absorber. The absorbers were made by mixing 40 mg of the
compound with 80 mg of boron nitride. The spectra of the measured samples
were fitted to an appropriate combination of Lorentzian profiles representing
quadrupole doublets by least-squares methods. In this way, spectral parameters
such as quadrupole splitting (QS), isomer shift (IS), and relative resonance areas
of the different spectral components were determined. Isomer shifts are given
relative toα-Fe metal.
26
catalytic testing. oxidation of polyvinylpyrollidone was done once, during
synthesis. Oxidation of PVP was carried out in a 23 mL teflon lined autoclave at
220
◦
C. The autoclave was filled as referenced in the synthetic section.
Oxidation of glycerol was carried out in a round bottom flask (25 mL)
equipped with a reflux condenser. In a typical reaction, 62.5 mg of catalyst
was used with 10 mL of glycerol solution in the presence of 1 mL of 30 %
hydrogen peroxide. The tests were performed at 60
◦
C, immersed in an oil bath
under constant stirring (350 rpm).
Oxidation of benzyl alcohol was performed in a modified Fisher-Porter bottle
(FPB) with a total internal volume of 15 mL (25.5mm x 10.22mm). [82] The
reactor was loaded with 50 mg of catalyst and 5mL of 1M benzyl alcohol in
acetonitrile. The tests were performed at pressures in the range of 50-90 PSI O
2
and temperatures between 60 and 120
◦
C. The FPB was immersed in a silicone
oil bath where the temperature was controlled and stirred by magnetic stirring
(350 rpm). At the end of the catalytic test the reaction was cooled to room
temperature. The reaction mixture was centrifuged at 6,000 rpm for 10 minutes
to separate the product mixture and the catalyst. The general reaction procedure
was the following: the catalyst (50mg) and benzyl alcohol (1000 mg/ml 5mL)
were transferred to FPB, and purged with 60 psi oxygen three times before the
reaction was initiated.
Organic products were identified using
1
H or
13
NMR. NMR spectra were
recorded on a Varian VNMRS 400 or VNMRS 600 spectrometer and processed
using MestroNova. All chemical shifts are reportead in units of ppm and refer-
enced to the residual
1
H or
13
C solvent peak.
27
2.3 Results and Discussion
(a) (b)
(c) (d)
Figure 2.1: An illustration of the a 2:1 silicate including (a) that contains corner
sharing silicate tetrahedra (b) which appear on either side of the MO
6
octahedra.
Potassium ions (blue) reside in the interlayer spacing between repeating metal silicate
layers. There are two metal sites in the octahedral layer, the stuffed site and the
honeycomb site shown in yellow and red respectively in (c) and (d). Orange spheres
represent oxygen while black spheres represent hydrogen.
Phyllosilicates, KFe
3
Si
4
O
10
(OH)
2
, KZn
3
Si
4
O
10
(OH)
2
, and KCo
3
Si
4
O
10
(OH)
2
were produced hydrothermally. The structure of each is the prototypical 2:1
trioctahedral phyllosilicate, as seen in Figure 2.1. Within the octahedral layer
of 2:1 phyllosilicates there are two distinct metal sites: one which forms the
hexagonal network (honeycomb site) and one which sits at the interstitial of
28
the hexagonal network (stuffed site). The honeycomb site is structurally distin-
guished by locally coordinated hydroxyl ions that coordinate the unit in the cis
positions, whereas they sit trans in the stuffed site. In this structure K
+
ions
located between the layers bind them together, compensating for the negative
charge in each sheet.
Figure 2.2: Laboratory x-ray diffraction patterns for hydrothermally prepared (a)
KFe
3
Si
4
O
10
(OH)
2
, (b) KCo
3
Si
4
O
10
(OH)
2
and KZn
3
Si
4
O
10
(OH)
2
. The dashed lines
represent peaks which appear due to the stacking faults, the origin of these peaks is
discussed.
29
The precise composition of the as-prepared materials were determined using
a combination of ICP-OES and CHNS elemental analysis, resulting in a nom-
inal stoichiometry of KFe
2
(Si
3.25
Fe
0.75
)O
10
(OH)
2
, K
0.78
Zn
2
(Si
3.4
Zn
0.6
)O
10
(OH)
2
and KCo
2
(Si
3.75
Co
0.25
)O
10
(OH)
2
. The composition of these materials are
matches well with expected composition of the clay mineral Muscovite,
K(Al)
2
(AlSi
3
)O
10
(OH)
2
. [83] Similar to the aluminum in Muscovite, while most
of the iron, cobalt and zinc are found in the metal sites, a small fraction of each
substitutes onto the tetrahedral site. These compositions are highly repeatable
for a fixed set of synthetic conditions.
Figure 2.3: SEM image of as-prepared KFe
3
Si
4
O
10
(OH)
2
and (b) results of a LeBail
fit of as-prepared sample using a C 2m space group against synchrotron XRD pattern
obtained on the 11-BM beamline at Argonne National Laboratory.
30
A typical SEM of the as-prepared samples shows that most particles adopt a
plate like shape with a thickness around 100nm and a diameter of 1μm. Platelet
morphology in combination with a layered structure often results in anisotropic
broadening of the peaks, as seen in the inset to Figure 2.3. Here, the broadening
is usually the result of discs not stacking perfectly atop one another. We also see
the [00l] peaks, which correspond to the layer spacing in the direction that the
sheets are stacked, produces significantly sharper reflections.
The complexities associated with this morphology and its scattering data
implored us to develop a structure model with which we could accurately fit
the resulting patterns. To create our model, we defined a single repeat slab as
consisting of one transition metal layer sandwiched between two silicate layers,
and one layer of potassium ions. The various way these slabs can be stacked with
respect to each other results in a number of space groups, [84, 85, 86, 87, 50]
but the most common are P3
1
12, C2/c, and C2/m.[88] Le Bail fits using the
sychrotron X-ray diffraction data show that the pattern is best described using
the monoclinic unit cell C2/m, with the cell parameters of 5.432 Å , 9.227 Å,
10.270 Å and 100.864
◦
.
Due to the unique reflections in the XRD, the compound KFe
3
Si
4
O
10
(OH)
2
was used as model to determine the exact distribution of the metal cations across
the sites and investigate the general structure of these silicates. KFe
3
Si
4
O
10
(OH)
2
was modeled using a small-box least-squares refinements performed on neutron
total scattering data [Figure 2.5]. From these refinements it is clear that iron pre-
dominantly occupies the honeycomb site and the stuffed position remains mostly
vacant other than a roughly 5% random occupancy by iron. This is consistent
with the iron Mössbauer experiment that will be discussed later in greater detail,
and confirms that the materials obtained are isomorphous to mineral Muscovite.
31
10 20
r(Å)
-2
0
2
4
6
8
10
PDF(Å
-2
)
KFe
3
Si
4
O
10
(OH)
2
KCo
3
Si
4
O
10
(OH)
2
KZn
3
Si
4
O
10
(OH)
2
Figure 2.4: Low-temperature neutron pair distribution function analysis data for (a)
KFe
3
Si
4
O
10
(OH)
2
(b) KCo
3
Si
4
O
10
(OH)
2
and (c) KZn
3
Si
4
O
10
(OH)
2
. The highlighted
region represents distances associated with stacking faults in the iron phase.
After confirming the iron was still well ordered within the ab plane, we sought
to describe the coherence between the sheets, again using KFe
3
Si
4
O
10
(OH)
2
as a
model. We estimated the degree of stacking faults by introducing an iron and a
vacancy antisite disodering at the "stuffed" position during Rietveld refinement.
This antisite mixing indicates the degree if stacking disorder between the honey-
comb layers along the stacking direction. [89, 90] The best fit using this method
indicates a stacking fault concentration of approximately 20% suggesting that
the coherence of the stacking sequence is interrupted every 5 layers on average.
32
3 6 9 12
r (Å)
-3
-2
-1
0
1
2
3
G(r) (Å
-2
)
Observed
Calculated
Difference
Figure 2.5: Least-squares refinement of the local structure of KFe
2.75
Si
3.25
O
10
(OH)
2
using neutron PDF data obtained at the NOMAD beamline at the Oak Ridge National
Laboratory (R
wp
= 14.65%). Locally, the octahedral Fe and vacancy sites are found
to be well-ordered though about 5% Fe are refined to be randomly distributed on the
vacancy site.
Though we could achieve reasonable fits using the above method, it still left unfit
intensity for may reflections, especially the strong warren peak at about 5.2
◦
.[91]
To accomodate this, a large supercell along the stacking direction was created to
account for the aperiodic stacking of sheets. A supercell with ac-axis translation
periodicity of 30 layers (30nm) was constructed within TOPAS. The three lateral
translation vectors were initial randomly distributed among these 30 layers, and
it should be noted that the in-plane components of these vectors were allowed to
vary from ideal values in order to account for the monoclinic distortion. A dra-
matic improvement could be achieved and almost all diffuse scattering intensity
33
can be successfully modeled in this way. This method yield a 25% stacking fault
concentration based on the counting scheme proposed by Liu, which is slightly
larger than that obtained from Rietveld analysis, but is expected to be a more
accurate reflection of disorder in the compound. [90]
Effect of synthetic conditions. After developing a model for the structure,
we explored the effect of varying the reaction conditions, beginning with the
influence the mineralizer, KOH. Phase-pure samples are not created when NaOH
is used. The laboratory diffraction patterns for different loadings of KOH are
shown in Figure 2.6, where the amount of metal, silica, and PVP were held con-
stant at the values discussed in the experimental section and the molar ratio of
KOH to Fe was adjusted from 2.5:1 to 10:1. As more KOH was added to the
reaction mixture, the diffracted intensity increased and was correlated with a
sharpening of the peaks, indicatig the crystallinity of the final product was sig-
nificantly improved. The corresponding FTIR show sharpening of resonances
with the increasing base content. This suggests that the reaction proceeds via
a silica-dissolution-precipitation process: higher concentrations of KOH dissolve
the silica more effectively and improve the quality of the resulting crystallites.
[92, 93]
The dissolution-precipitation mechanism is further supported by SEM images
of samples prepared from various KOH concentrations (Figure 2.7). For the
smallest amount of KOH, particles resembling the morphology of fumed silica
are clearly seen, supporting the notion that the solution is not basic enough to
fully dissolve all of the precursors. As the KOH to Fe ratio reaches 5:1, heavily
agglomerated plate-like crystallites can be found, but only after the concentra-
tion of base exceeds 7.5:1 do well-defined particles begin to appear. Most likely,
34
Figure 2.6: Laboratory X-ray diffraction patterns on hydrothermally prepared Fe-
phyllosilicateswithdfferentKOHconcentrations. Thebottom, blackcurvecorresponds
to sample prepared with a KOH to Fe ratio of 2.5:1, or a KOH concentration of 0.33
M. The blue, yellow and green curves correspond to KOH to Fe ratios of 5:1, 7.5:1
and 10:1, respectively.
low concentration of KOH prevents the correct Fe:Si ratio from existing in solu-
tion and results in off-stoichiometric particles with poor crystallinity.
Mössbauer spectroscopy was used to examine how the local environment of
Fe was affected by different concentrations of KOH and PVP. A representative
Mössbauer spectrum K-Fe synthetic end members are shown in Figure 2.8, with
the resulting fits to several samples prepared with different KOH concentrations
given in Table 2.1. Samples repeated under the same reaction conditions were
found to produce the same spectra within the error of the measurement, and
those from different conditions were clearly distinguishable from each other as
seen by comparison. As mentioned earlier, the iron appears to be distributed
35
Figure 2.7: SEM images of phyllosilicate samples prepared with different KOH con-
centrations.(a), (b), (c), and (d) correspond to samples prepared with KOH to Fe ratio
of 2.5:1, 5:1, 7.5:1, and 10:1, respectively.
between two different octahedral environments as well as on the tetrahedral
sites. Indeed, we see some signature of Fe
3+
filling the tetrahedral site for KOH
to Fe ratios exceeding 5:1 as listed in Table 2.1. Given that the (001) reflection
(Figure 2.3) and the disk-like morphology (Figure 2.7) do not evolve until KOH
to Fe ratio is greater than 5:1, the substitution of Fe
3+
onto the tetrahedral site
appears to be correlated with an increase in coherence between the (00l) planes.
Hydrothermal Oxidation of PVP
PVP is frequently utilized as a templating agent in hydrothermal reactions but
has also been noted for its slight reducing character.[32, 94] Indeed we noticed,
36
Table 2.1: Distribution of Fe environments of phyllosilicates prepared with increasing
KOH concentration. Values marked with (*) were fixed during the fitting procedure to
increase stabilize the routine.
K:Fe Environment % IS QS LW
(mol/mol) mm/s mm/s mm/s
2.5:1
3Oct(1) 88(4) 0.346(4) 0.66(2) 0.44(1)
3Oct(2) 12(4) 0.431(3) 1.28(8) 0.39(3)
5.0:1
2Oct(1) 8(2) 1.100(4) 0.97(1) 0.49(4)
3Oct(1) 39(2) 0.430(2) 0.77(1) 0.30(*)
3Tet(1) 54(2) 0.271(2) 0.43(1) 0.46(*)
7.5:1
3Oct(1) 87(3) 0.332(7) 0.58(1) 0.45(1)
3Tet(1) 13(3) 0.165(9) 0.36(3) 0.35(3)
10.0:1
2Oc(1) 4(*) 1.150(*) 2.39(6) 0.46(*)
2Oct(2) 5(1) 1.050(*) 0.78(5) 0.30(*)
3Oct(1) 76(1) 0.348(4) 0.56(1) 0.30(2)
3Tet(1) 15(1) 0.172(5) 0.36(1) 0.33(2)
Table 2.2: Distribution of Fe environments with varying PVP additions.
PVP (g) Fe
3+
Fe
3+
Fe
3+
Fe
2+
Oct,% Tet, % total, % Oct, %
0 72(1) 22(1) 94(1) 5(1)
0.1 81(2) 19(2) 100(2) 0
0.4 78(1) 12(1) 90(1) 10(1)
0.7 81(2) 15(2) 96(2) 4(2)
1.0 82(2) 14(2) 96(2) 4(2)
despite using a completely trivalent iron precursor, Fe(NO
3
)
3
·9H
2
O, samples pre-
pared in the presence of PVP often contained Fe
2+
between 5-10%. NMR spec-
tra of the solution after hydrothermal treatment showed the presence of typical
aromatic sp
2
carbon, which is most likely due to the oxidation of PVP (Figure
2.9). This change was also reflected in the color of the final product, which
was dark green when PVP was used and a goldenrod brown when no PVP was
37
Figure 2.8: Room temperature Mössbauer spectrum of the as prepared KFe prepared
with no PVP(a), and KFe prepared with 1g of PVP(b) that shows the creation of two
Fe
2+
sites when PVP is used. PVP is not linked to tetrahedral Fe
+3
used. Despite the varying colors, powder XRD and SEM images were identical
for nearly all of the products. More interesting, the Mössbauer spectra of samples
prepared with varying amounts of PVP show that increasing PVP content appears
correlated with a decreasing concentration of Fe
3+
on the tetrahedral site, staying
around 15 % when more than 0.4 g of PVP was used (Table 2.2). This change in
the concentration of tetrahedral iron likely explains the variation in colors of the
samples as there may be some kind of charge transfer process between the octa-
hedral and tetrahedral iron sites. Combining the results from elemental analysis
and Mössbauer spectroscopy, the composition of the as-prepared KFeSiO-20 sam-
ple, which was used for all of the characterization, was confirmed to be K(Fe
3+
1.75
Fe
2+
0.125
)
h
(Fe
2+
0.125
)
s
(Si
3.25
Fe
3+
0.75
)
t
O
10
(OH)
2
, withh ands indicating the octahe-
dral honeycomb and stuffed sites respectively while thet denotes the tetrahedral
position. The use PVP in the synthesis of KCo
3
Si
4
O
10
(OH)
2
follows a similar
trend.
But more importantly, in addition to the concentration of KOH, it introduces
another dimension of tuning the Fe
3+
in the tetrahedral site, which again has
38
Figure 2.9: The
1
H NMR spectrum (a) shows two peaks at 6.87 ppm and 7.76 ppm
that are shown to be splitting each other in the 2D
1
H spectra (b). This indicates the
presence of sp
2
carbon species after the hydrothermal treatment due to the oxidation
of PVP.
significant implications for catalytic studies of phyllosilicates. Furthermore, it
shows the that this series of compounds warranted further exploration as oxida-
tion catalyst, due to their oxidation of PVP during synthesis.
Oxidation in the Presence of Hydrogen Peroxide. Following recent
studies,[95, 96] the oxidative properties of the synthetic Muscovite series
KFe
3
Si
4
O
10
(OH)
2
, KZn
3
Si
4
O
10
(OH)
2
and KCo
3
Si
4
O
10
(OH)
2
were then tested for
39
(a) (b) (c)
Figure 2.10: An image of (a) KFe
3
Si
4
O
10
(OH)
2
(b) KCo
3
Si
4
O
10
(OH)
2
and (c)
KZn
3
Si
4
O
10
(OH)
2
samples after reaction with glycerol in the presence of hydrogen
peroxide. The colored solution indicates the presence of free metal ions in solution.
Note, a nickel doped KZn
3
Si
4
O
10
(OH)
2
sample was used for the purposes of this
picture, to provide the light blue coloring.
the oxidation of glycerol using hydrogen peroxide as the oxidant. Initial experi-
ments were carried out with each silicate as the catalyst using hydrogen peroxide
as the oxidant reacted with glycerol (0.3 g/L) in water at 80
◦
C. In a typical reac-
tion, about 65 mg of catalyst were mixed with glycerol solution before the hydro-
gen peroxide was added and the resulting solution was stirred. Immediately after
mixing with hydrogen peroxide, the reactions begins producing bubbles, indicat-
ing the reaction started. As the reaction proceeds the solution would be come
colored and the catalysts begin to decompose leading us to believe metal was
leached from the lattice during the reaction, resulting in free metal ions in solu-
tion. [97] In lieu of this result, the products were not analyzed, as free iron is
known to facilitate oxidation in the presence of hydrogen peroxide. [98]
In view of these results, we tested the remaining silicate series to deter-
mine their stability in hydrogen peroxide. Both KCo
3
Si
4
O
10
(OH)
2
and
40
KZn
3
Si
4
O
10
(OH)
2
showed similar instabilities, indicating they are not stable cat-
alysts for these reaction conditions. Even in a mild peroxide, tert-butyl hydrogen
peroxide, each silicate in the series suffered from metal leaching and catalyst
decomposition. Previous silicate investigators have shown that sheet silicates,
especially those that adopt planar morphologies, tend to undergo dissolution at
their edges in peroxide. [99, 100] While this limits their stability for catalysis, it
is the mechanism that allows silicate soils to release metals. [48] Under oxida-
tive conditions in the absence of any peroxide, the structures did not suffer from
leaching but also did not show any conversion of the glycerol.
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 0.5
f1 (ppm)
1
2
3
(a) KFe
3
Si
4
O
10
(OH)
2
(b)
(c)
KCo
3
Si
4
O
10
(OH)
2
KZn
3
Si
4
O
10
(OH)
2
Figure 2.11: The
1
H NMR spectrums show the lack of oxidation product from
the reaction of benzyl alcohol and 2:1 phyllosilicates (a) KFe
3
Si
4
O
10
(OH)
2
(b)
KCo
3
Si
4
O
10
(OH)
2
and (c) KZn
3
Si
4
O
10
(OH)
2
. The lack of an aldehyde peak at 10.00
ppm shows there is no benzaldehyde present.
Oxidation of Benzyl Alcohol under molecular O
2
To further explore, we
investigated the oxidation of benzyl alcohol in the presence of molecular oxygen.
Benzyl alcohol oxidation reactions were carried out in a home-built Fisher Porter
41
Bottle (FPB) under pressurized oxygen. Experiments were carried out with 50
mg of catalyst in 5 mL of neat benzyl alcohol at 100
◦
C and 60 PSI oxygen
pressure for 3 hours. Following the reaction, the catalyst was separated from the
benzyl alcohol by centrifugation, and the product was pipetted off for further
analysis. After separation, the product was analyzed by NMR in CDCl
3
. As seen in
Figure 2.11, by
1
H NMR analysis no products for the oxidation of benzyl alcohol
were observed for any KFe
3
Si
4
O
10
(OH)
2
, KCo
3
Si
4
O
10
(OH)
2
or KZn
3
Si
4
O
10
(OH)
2
.
From this we have determined these clays are only capable of true redox oxi-
dation chemistry in extreme conditions, such as those present during hydrother-
mal synthesis. In the absence of subcritical conditions, these compounds lack
outright oxidative properties. Given their structure, we assume this is due to
physical blocking of the metal sites due to the presence of two silicate sheets.
Furthermore, the interlayer potassium ion is held in place by a strong charge
transfer, making it impossible to even electrochemically exchange it, opening the
interlayer sites. The edges, though reactive in the presence of hydrogen peroxide,
deplete the stability of the structure if accessed. As a result, we do not believe
that these phyllosilicates will be a viable platform to investigate catalytic reac-
tivity. However, the results do indicate that these compounds may be judiciously
designed and studied.
2.4 Conclusions
In summary, I have presented a systematic study of the structure, composition
and morphology of three silicate clays, KFe
3
Si
4
O
10
(OH)
2
, KCo
3
Si
4
O
10
(OH)
2
, and
KZn
3
Si
4
O
10
(OH)
2
. Using a combination of spectroscopic and structural charac-
terization tools, we precisely characterized the composition of highly crystalline
42
samples made at relatively low temperatures. Following their physical character-
ization, we investigated their properties as catalysts for the oxidation of glycerol
and benzyl alcohol. We investigated activity in a range of oxidants and demon-
strated that in the presence of peroxides, these structures are unstable, resulting
in catalyst decomposition. With only molecular oxygen as the oxidant, these
compounds showed no reactivity, indicating they most likely cannot bind and
utilize O
2
. Though these materials are ill-suited as oxidation catalysts, this work
demonstrates that these silicates exhibit a highly tunable structure. Beyond that,
it shows these materials can be well characterized and very informative. These
results indicate it is likely a 1:1 phyllosilicate will likely be more reactive due to
easy surface access.
43
Chapter 3
Crystal Chemistry and Oxidation
Properties of 1:1 Phyllosilicates
Ni
3
Si
2
O
5
(OH)
4
, Co
3
Si
2
O
5
(OH)
4
, and
Mg
3
Si
2
O
5
(OH)
4
Following our catalytic study of muscovites, we examined 1:1 phyllosilicates
to ascertain their catalytic potential. Herein, we identify the most promising 1:1
phase for further exploration: Ni
3
Si
2
O
5
(OH)
4
. Furthermore, we demonstrate the
structure of 1:1 phyllosilicates may be well-defined.
44
OH
Benzyl Alcohol
O
Benzaldehyde
OH
Cinnamyl Alcohol
O
Cinnamaldehyde
M
3
Si
2
O
5
(OH)
4
O
2
(90 PSI)
M
3
Si
2
O
5
(OH)
4
O
2
(90 PSI)
M= Ni, Co, Mg
Figure 3.1: Benzyllic alcohol oxidation scheme.
3.1 Introduction
Thedevelopmentofnewlowcost sustainable catalysts is critical for decreas-
ing our reliance on fossil fuels to produce commodity chemicals. [101] Though
heterogeneous catalysts are preferred due to their inherent practical advantages,
their conception lacks basic design rules. [3, 16, 34] The fundamental under-
standing of activity in solid catalysts is often limited by our ability to probe the
active site in solids and therefore design effective systems. [10, 16, 17]
The development of fundamental correlations between catalytic activity, com-
position, atomic and electronic structure is imperative to catalysis by design.[34,
35] We must identify and study the role of a single reactive site to create design
rules. To mature this chemistry, model catalysts and platform chemicals that
represent a broad family of catalysts and substrates are ideal. Thus, to facilitate
catalytic understanding it is extremely desirable to develop a planar model struc-
ture that would limit the reactive surface to a single face. This is necessary to
eliminate the effects of structural rearrangement and complications associated
with determining multiple reactive faces. [10, 30]
45
Figure 3.2: The phyllosilicate crystal structure is comprised of alternating layers of
metal octahedra and cornersharing Si-O tetrahedra (top). The Si-O tetrahedra join
the metal layers at their apical oxygen, and the layers are held together by relatively
weak hydrogen bonding by the surface hydroxyl groups.
Phyllosilicate materials have been utilized previously as adsorbents, makeup,
and cracking catalysts due to unique properties facilitated by their lamellar struc-
ture. [67, 102] Development of synthetic phyllosilicate catalysts is hindered
by methods to form highly crystalline materials and methods to describe the
order of these approximately crystalline materials.[60, 87, 100] The structures
of 1:1 phyllosilicates are customarily described from powder x-ray diffraction
patterns consisting of broad asymmetric peaks. [53, 54] The inability to study
this complex structure has prevented the development of more detailed correla-
tions between structural features and catalytic ability. By studying this structure
46
we may develop correlations between binding energy, local structure, and com-
position on a well-defined surface.
The goal of this work has been to validate the 1:1 phyllosilicate as a suf-
ficient model to facilitate interrogation of the active site at the atomic level.
By using pair distribution function analysis (PDF) we provide a comprehensive
description of the 1:1 phyllosilicate structure. Herein, we present the synthesis
and characterization of the 1:1 phyllosilicates Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
and Co
3
Si
2
O
5
(OH)
4
, and their evaluation as catalysts for the oxidation of benzyl
alcohol. This work calls attention to a hugely earth-abundant family of minerals
possessing useful catalytic properties that warrant further exploration.
3.2 Experimental
Synthesis. Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
were prepared using
hydrothermal synthesis. In a typical preparation, stoichiometric amounts of
precursor salts either Ni(NO
3
)
2
· 7H
2
O (EMD Millipore, 99% ) or Mg(NO
3
)
2
· 7H
2
O (Sigma-Aldrich, 99% ) were combined with 2mmol H
2
SiO
3
(Sigma-
Aldrich, 99.9% ), dissolved in 15mL DI H
2
O, and polymerized by the addition
of 8.75mmol NaOH. The resulting silica gel was allowed to age for 2-10 days
before being transferred to a 23mL teflon lined autoclave. The autoclave was
sealed and heated at 200
◦
C for 2 days. Upon cooling, the resulting powder was
collected via vacuum filtration and washed with DI water. The powder was dried
at 100
◦
C and ground in an agate mortar for characterization.
To prepare Co
3
Si
2
O
5
(OH)
4
, 3mmol CoCl
2
· 6H
2
O (Sigma-Aldrich, 99 % ),
2mmol Na
2
SiO
3
(Wards Science, lab grade), 5.83mmol NaOH and approximately
1g of polyvinylpyrrolidone (PVP) (Sigma-Aldrich, Average MW 10,000) were
47
combined and dissolved in 15mL DI H
2
O. The resulting dark blue solution was
transferred to a 23mL teflon-lined stainless steel autoclave and heated at 200
◦
C
for 3 days. The resulting magenta powder was collected under vacuum filtration
and washed with copious amounts of DI H
2
O. Following collection, the powder
was dried at room temperature under vacuum and ground in an agate mortar.
Characterization techniques. Composition was determined in-house using
a Thermo Scientific iCAP 7000 inductively coupled plasma-optical emission spec-
trometer (ICP-OES). Samples were dissolved in 2% HF before being diluted with
2% Nitric Acid and measured three times with the presented values representing
the average over all measurements. Phase identification was performed with a
Bruker Advanced D8 Diffractometer in bragg geometry, using Cu-Kα radiation.
Synchrotron X-ray total scattering data were collected at 100 K on beamline 11-
ID-B at the Advanced Photon Source, Argonne National Laboratory, with an X-ray
wavelength of 0.2114nm (about 58.6 keV). [103] Total scattering samples were
first ground thoroughly in an agate mortar and then packed in Kapton tubes
sealed with epoxy. A 2D amorphous Si image-plate detector (PerkinElmer, 2048
x 2048 pixels and 200 x 200 mm pixel size) was used for two-dimensional data
collection with a sample-to-detector distance of ˜ 170mm. The data was con-
verted to 1D XRD data using Fit2D software. [104] The PDF was then created
from Fourier transform of the total scattering data in either PDFgetX3 or with
a Q range of 0.2 - 22 Å
−1
. [105] Local structure refinements and models using
X-ray PDF were carried out in PFDgui.[106] Microscopy images were acquired
using JEOL 2100F TEM at the University of Southern Californiaâ
˘
A
´
Zs Core Center
of Excellence for NanoImaging
Catalytic testing. Catalytic experiments were carried out in a home-built
modified Fisher-Porter bottle (FPB). [82] The modified FPB was created from an
48
ace glassware pressure bottle (15 mL, 25.5mm x 10.22mm) connected to a pres-
sure gauge and needle valves. In a typical experiment the bottle was loaded with
50mg of catalyst and 5mL of benzyl alcohol solution (1M in acetonitrile). Condi-
tions were optimized at pressures in the range of 50-90 psi O
2
and temperatures
between 60 and 120
◦
C. The FPB was immersed in a silicone oil bath where the
temperature was controlled and stirred by magnetic stirring (350 rpm). At the
end of the catalytic test the reaction was cooled to room temperature in air. The
reaction mixture was centrifuged at 6,000 rpm for 10 minutes to separate the
product mixture and the catalyst. In a typical reaction procedure was the follow-
ing: the catalyst (50mg) and benzyl alcohol solution (1M in acetonitrile) were
transferred to FPB, and purged with oxygen 3-5 times before the reaction was
initiated.
Organic products were identified using
1
H or
13
C NMR. NMR spectra were
recorded on a Varian VNMRS 400 or VNMRS 600 spectrometer and processed
using MestroNova. All chemical shifts are reported in units of ppm and refer-
enced to the residual
1
H or
13
C solvent peak. The concentration of the substrate
and the product were determined by high performance liquid chromatography
(HPLC). HPLC-DAD-Q-TOF analysis was performed on an Agilent series 1290
Infinity HPLC instrument (Agilent Technologies, Santa Clara, CA, USA) coupled
with an Agilent 6545b Dual AJS ESI Q-TOF mass spectrometer equipped with
a dual (AJS) and electrospray (ESI) interface. Chromatographic separation was
carried out at 30
◦
C on an Agilent Zorbax-Bonus RP (2.1 mm x 150 mm, I.D., 2.7
um).
The chromatographic conditions were as follows: flow rate of 0.4 mL/min,
sample injection volume of 0.5 uL, mobile phase consisting of a 70:20:10 mixture
of 0.1 % formic acid, acetonitrile and methanol. All solvents used were HPLC
49
grade and the condition was kept isocratic. The re-equilibration duration was
2 min between individual runs. The DAD detector scanned from 190 - 400nm
and the samples were detected at 254nm and 282nm for benzyl alcohol and
benzaldehyde respectively.
For MS detection, the operating parameters were as follows: drying gas flow
rate, 10.0 L/min; drying gas temperature, 325
◦
C; nebulizer, 35 psig; capillary,
2000 V; skimmer, 65 V; Oct RFV, 750 V; and fragmentor voltage, 75 V. All the
acquisition and analysis of data were controlled by MassHunter software (Agilent
Technologies).
3.3 Results and Discussion
3.3.1 Preparation and Bulk Structure Analysis
The structure of a classic divalent 1:1 phyllosilicate (Figure 4.2) is made up
of a metal oxyhydroxide layer bound to a silicate layer at the silicates apical
oxygen. The layers are held together by weak hydrogen bonding forces either
between layers or facilitated by interlayer water. In this structure the morphology
is directly determined by the size mismatch between the metal and the silicate
layer. [56, 49] As a result, different synthetic methods were necessary to target
the inclusion of three different metals: nickel, magnesium and cobalt.
A hydrothermal method for the reproducible synthesis of uniform
Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
and Co
3
Si
2
O
5
(OH)
4
was developed first. Both
Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
were obtained by means of amorphous
silica-gel precursor, while Co
3
Si
2
O
5
(OH)
4
can only be accessed using a direct
synthesis. To obtain both Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
stoichiometric
ratios of silicic acid and a metal-nitrate precursor were dissolved in 15mL of
50
water. Silica polymerization was then induced by the addition of sodium hydrox-
ide (1.5M). The resulting mixture aged for 24-72 hours before being transferred
to a teflon lined autoclave and heated to 200
◦
C for 50 hours.
(a) (b) (c)
Figure 3.3: TEM of Ni
3
Si
2
O
5
(OH)
4
(a), Mg
3
Si
2
O
5
(OH)
4
(b), and Co
3
Si
2
O
5
(OH)
4
(c), indicating all three phases form both platelets and nanotubes.
In contrast, Co
3
Si
2
O
5
(OH)
4
is only formed through the combination of cobalt
chloride (CoCl
2
) and waterglass (Na
2
SiO
3
) in the presence of polyvinylpyy-
rolidone (PVP). Due to the easy accessibility of the Co
+2
/Co
+3
redox cou-
ple PVP is necessary to retain Co
+2
and produce a single crystalline phase of
Co
3
Si
2
O
5
(OH)
4
. We have shown previously in hydrothermal conditions PVP acts
as a reducing agent. [32, 47, 94] The black solid Co
+3
2
Si
2
O
5
(OH)
4
is formed in
the absence of PVP.
A typical TEM image [Figure 3.3 of the as-prepared materials show that
Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
develop as both platelets and nanotubes
with an average particle size of about 50nm. The tubes are 50-200nm long and
have have a diameter of about 7 Å . In contrast, Co
3
Si
2
O
5
(OH)
4
forms very thin
single-layer nanotubes. Powder x-ray diffraction (Figure 3.4) confirms the as-
prepared Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
, and Co
3
Si
2
O
5
(OH)
4
match well with
51
the reported crystal patterns for chrysotile (JCPDS 43-662) and the nickel ana-
logue, pecoraite (JCPDS 49-1859), with no impurities observed.
2 � (deg) [ � = 1.54184Å]
Intensity (arb. units)
Ni
3
Si
2
O
5
(OH)
4
Mg
3
Si
2
O
5
(OH)
4
Co
3
Si
2
O
5
(OH)
4
Figure 3.4: Laboratory XRD of 1:1 catalysts Ni
3
Si
2
O
5
(OH)
4
(blue), Mg
3
Si
2
O
5
(OH)
4
(yellow) and Co
3
Si
2
O
5
(OH)
4
(magenta). Similarity indicates that the structures adopt
the same bulk configuration. The nickel and magnesium analogues are the most crys-
talline.
Selective broadening and overlapping reflections present in the diffraction
data make it very difficult to refine. The extent of crystallinity can often be
observed from the quality of the XRD patterns the samples produce. Prominent
peaks at ˜ 7.31 Å and ˜ 3.64 Å represent the the (002) and (004) planes, charac-
terizing the interlayer spacing, or in the event of nanotubes represents the inner
diameter of the tube. Here, the (002) peak for Ni
3
Si
2
O
5
(OH)
4
is shifted slightly
52
to the right of that for Mg
3
Si
2
O
5
(OH)
4
and Co
3
Si
2
O
5
(OH)
4
, indicating that the
spacing is slightly shorter for the nickel analogue. This is logical as nickel has a
radius of about 0.690 Å in octahedral coordination, slightly smaller than that of
cobalt (0.74 Å ) and magnesium (0.72 Å ). [107]
TEM indicates that nanotubular structures curl along the (200) axis, there-
fore the splitting in this peak also may indicate differences in the nature of the
scrolling. The (200) and (202) reflections are comprised of the octahedral layer
thus sensitive to the positions of metals atoms and possible vacancies. The
intralayer reflections at 2.64 Å (200), and 1.53 Å (060) also show subtle dif-
ferences between Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
indicating the strain is
indeed a function of the metal. Broadening along the (110) (4.52 Å ) and (202)
(2.44 Å ) planes results from the unique local distortions of each metal in the ab
plane. The large diffuse scattering components present in the background of all
samples show a varying degree of structural disorder due to local distortions and
the resulting strain. [108]
Qualitatively, it can be seen that Mg
3
Si
2
O
5
(OH)
4
and Ni
3
Si
2
O
5
(OH)
4
pro-
duce similar, very crystalline diffraction patterns. Compared to its nickel and
magnesium counterparts Co
3
Si
2
O
5
(OH)
4
is seen to have the least crystalline
diffraction pattern, likely due to small particle size. The magenta color of the
Co
3
Si
2
O
5
(OH)
4
sample indicates it likely contains both octahedral and tetra-
hedrally coordinated cobalt. [108] Tetrahedral cobalt may provide an alter-
native strain relief mechanism for layer mismatch inhibiting scrolling, result-
ing in poor crystallinity. [108] Our previous work on the 2:1 phyllosilicate,
KCo
3
Si
4
O
10
(OH)
2
, demonstrates cobalt may displace silicon in the tetrahedral
site, occupying about an eighth of the sites. Structural refinements of the unit
53
cell were carried out starting from the local structure determined from total scat-
tering data.
0
PDF(Å
-2
)
Ni
3
Si
2
O
5
(OH)
4
0
PDF(Å
-2
)
Mg
3
Si
2
O
5
(OH)
4
2 4 6
r(Å)
0
PDF(Å
-2
)
Co
3
Si
2
O
5
(OH)
4
(a)
(b)
(c)
Figure3.5: ThepairdistributionfunctionsforNi
3
Si
2
O
5
(OH)
4
(blue), Mg
3
Si
2
O
5
(OH)
4
(yellow) and Co
3
Si
2
O
5
(OH)
4
(magenta) from total scattering sychrotron x-ray data
obtained at the 11-ID-B beamline at Argonne National Lab. Locally, these structures
are very similar, indicating the local structure is well-ordered and isomorphous between
all three analogues.
3.3.2 Local Structure Analysis.
Synchrotron x-ray total scattering measurements were carried out at the 11-
ID-B beamline at Argonne National Lab using powder samples at 100 K to probe
the local structures of Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
and Co
3
Si
2
O
5
(OH)
4
. The
experimental pair distribution function was extracted from x-ray total scattering
54
data using the program PDFgetX3 and refined in PDFgui. Due to difficulty asso-
ciated with stacking faults, lack of layer coherence and interlayer strain, fits for
Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
, and Co
3
Si
2
O
5
(OH)
4
were confined to a single
unit cell in the z-axis. We defined a representative structure as an ideal tri-
octahedral 1:1 phyllosilicate composed of a single layer consisting of one metal-
octahedral sheet bound to one silicate sheet. We assigned our model space group
C
c
, based on previous studies. [42, 54, 62]
It can be seen within one unit cell that all three systems are very similar. Fits
within 8 Å confirm they share the same local structure, despite the variations
in XRD intensity. Fits for each structure are seen in Figure 3.5. Ni
3
Si
2
O
5
(OH)
4
can be described with cell parameters of 5.24 Å , 9.37 Å , 7.45 Å , and 93.7
◦
and
Rw=0.2803. Mg
3
Si
2
O
5
(OH)
4
is fit with unit cell parameters of 5.22 Å , 9.50
Å , 7.46 Å , and 92.96
◦
with a Rw=0.2489. Co
3
Si
2
O
5
(OH)
4
is fit with unit cell
parameters of 5.3Å , 9.54 Å , 7.48 Å , and 92.90
◦
, with aRw=0.3252.
Patterns for each structure maintain similarities at characteristic bond dis-
tances, such as the distance for the Si-O bond, though we observe the Si-O peak
in Co
3
Si
2
O
5
(OH)
4
is shifted to slightly longer distances, reinforcing the suppo-
sition that the cobalt material has site mixing. [109, 110, 111, 112, 113] The
relative M-O bond distances increase slightly with the size of the metal cation,
as expected. Additionally, the size of the silicate faces (O-O distances, 2.5 Å ) is
consistent across all three compounds. The silicate tetrahedra in Co
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
have bond variance around 18
◦2
making them slightly more
distorted than in Ni
3
Si
2
O
5
(OH)
4
, which has a tetrahedral bond variance around
13
◦2
. There is some variation in the size of the octahedra ((O-O) and (M-
M) distances), which shift from 2.79 Å at its onset and 3.06 Å at its peak in
Ni
3
Si
2
O
5
(OH)
4
to 2.85 Å at onset and 3.12 Å at the peak in Co
3
Si
2
O
5
(OH)
4
.
55
Likewise, the cobalt and magnesium octahedra have volumes of ˜ 12 Å
3
while
the nickel octahedra have a volume of 10.9 Å
3
. Additionally, each structure
shares an elongated surface -OH bond. This indicates the octahedra increase in
size as as the metal size increases, confirming the metal ion dictates the inter-
layer strain. These variations help us understand the effect of octahedra size and
distortion on the XRD of the ab planes.
It would require several sensitive studies beyond the scope of this investiga-
tion to associate morphological changes with its effect in the PDF.[114] How-
ever, similarities seen between Mg
3
Si
2
O
5
(OH)
4
and Ni
3
Si
2
O
5
(OH)
4
at distances
of 3.5 - 4.2 Å , corresponding to the lengths between the basal oxygen of the
silicon tetrahedra and the surface hydroxyl, do not exist for the Co
3
Si
2
O
5
(OH)
4
phase. This indicates that those distances may be associated with development
of multi-walled crystallites in more coherent phases. The most stark differences
between the three phases are seen in distances between 5.61 - 7.59 Å , indicat-
ing morphological difference at the unit cell boundary. The subtle differences
between each phase appear to be due to changes in the way the metal occupies
the octahedral site. Following this, we aim to understand how these variations
impact the catalytic ability of these compounds.
3.3.3 Reactivity studies.
The catalytic properties of the as-prepared Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
,
and Co
3
Si
2
O
5
(OH)
4
, were first tested for the oxidation of benzyl alcohol. Liquid
phase oxidation experiments were carried out in a magnetically stirred home-
built Fisher Porter reactor. Catalytic screening experiments were carried out
under batch conditions, with 50mg of catalyst in 5mL of benzyl alcohol [1M,
56
acetonitrile] at 100
◦
C under 60 psi oxygen pressure for 3 hours. Results show-
ing the liquid phase oxidation of benzyl alcohol to benzaldehyde by molecular
oxygen over Ni
3
Si
2
O
5
(OH)
4
, Mg
3
Si
2
O
5
(OH)
4
, and Co
3
Si
2
O
5
(OH)
4
are presented
in Table 3.1. In the absence of any catalyst the conversion was 2% with 100%
selectivity for benzaldehyde.
Entry Catalyst Benzyl Alcohol Conv. Benzaldehyde yield
1 Ni
3
Si
2
O
5
(OH)
4
90% 90%
2 Co
3
Si
2
O
5
(OH)
4
2% 1 %
3 Mg
3
Si
2
O
5
(OH)
4
2% 1 %
4 N/A 4% 2 %
Table 3.1: Benzyl alcohol conversions for the three materials tested. All reactions
were performed with 50mg as-prepared catalyst, 5mL benzyl alcohol (1M), 60 psi O
2
,
stirring at 355 rpm for 3 hours unless otherwise stated.
Only as-prepared Ni
3
Si
2
O
5
(OH)
4
showed activity for the oxidation of ben-
zyl alcohol to benzaldehyde. In comparison, the activity of Mg
3
Si
2
O
5
(OH)
4
and
Co
3
Si
2
O
5
(OH)
4
was very low, with almost no conversion by the catalysts. As a
result, optimization of catalytic conditions, summarized in Table 3.2, was com-
pleted using only Ni
3
Si
2
O
5
(OH)
4
.
Entry Pressure (PSI) Temperature (
◦
C) Conversion Selectivity
1 65 60 66 % 100 %
2 65 80 64 % 100 %
3 65 100 63 % 100 %
4 65 120 67 % 100 %
5 45 100 40 % 100 %
6 85 100 90 % 100 %
Table 3.2: Optimization of oxidation conditions for the conversion of Benzyl Alcohol
to Benzaldehyde by Ni
3
Si
2
O
5
(OH)
4
in the presence of molecular oxygen as the only
oxidant. All reactions were performed with 50mg as-prepared Ni
3
Si
2
O
5
(OH)
4
catalyst,
5mL benzyl alcohol (1M), stirring at 355 rpm for 3 hours unless otherwise stated.
57
Oxidation of benzyl alcohob by Ni
3
Si
2
O
5
(OH)
4
nearly 100% selective for ben-
zaldehyde with no noticeable benzoic acid as a side product. Increased temper-
atures did not lead to an increased conversion of benzyl alcohol, only improving
the yield about 1% as the temperature is increased from 60
◦
C to 120
◦
C. While
this may seem strange, similar studies suggest 70
◦
C the ideal reaction tempera-
ture to promote surface interactions that facilitate benzyl alcohol oxidation.[115]
As the effect of the oxidant:substrate ratio is a key variable in these reactions the
influence of the oxygen pressure was investigated from 40 to 90 psi. Indeed,
oxygen pressure proved the most influential factor, giving a 100% increase in
conversion when oxygen pressure is doubled. Conversion dropped by 70% in air
indicating the catalyst may be poisoned by the presence of CO
2
.
We aim to understand the variations in activity based on the structure and
compositions of these catalysts. The performance Mg
3
Si
2
O
5
(OH)
4
can be ratio-
nalized by the lack of oxidation states available for the reaction. However, low
conversion by the cobalt phase indicates conversion may be unique to the nickel,
not only correlated to the use of a redox active metal. We hypothesize the ten-
dency for the Co
3
Si
2
O
5
(OH)
4
phase to form only single-walled nanotubes may
limit its available surface area and stability. Attempts to produce a more crys-
talline Co
3
Si
2
O
5
(OH)
4
phase for testing were not successful.
The first oxidation of benzyl alcohol results in benzaldehyde, and further
oxidation results in the production of benzoic acid; though this is often inhibited
by the presence of benzyl alcohol (Figure 3.6). [116] After the initial oxidation
side reactions may also occur, such as the condensation of benzaldehyde and
benzyl alcohol, which is catalyzed by the presence of acid-base sites. [117] We
do not see evidence for any of these products by NMR or HPLC-qTOF. On the
basis of these results, we propose the oxidation mechanism proceeds similarly
58
Figure3.6: Reactionnetworkshowingtheoxidationofbenzylalcoholtobenzaldehyde,
oxidation benzaldehyde to benzoic acid and possible esterifcation pathways with the
products and benzyl alcohol.
to that proposed by Guo et al, in which the alcohol group of benzyl alcohol
is adsorbed at a Lewis-acidic center. [118] Concurrently, O
2
is adsorbed and
dissociated at the nickel, and the free oxygen adsorbs theα-hydrogen of benzyl
alcohol. The abstracted hydrogen atoms are the either reduced to form hydrogen
gas or combined with local hydroxyls to form water. Though no water is seen by
NMR, Ni
3
Si
2
O
5
(OH)
4
likely proceeds via the latter route.
From these results, the phyllosilicate Ni
3
Si
2
O
5
(OH)
4
demonstrates the abil-
ity to adsorb and dissociate oxygen to oxidize benzyl alcohol. In these experi-
ments, Ni
3
Si
2
O
5
(OH)
4
also oxidized cinnamyl alcohol to cinnamaldehyde under
the same conditions. Work to understand the origin of this catalytic ability make-
up the remaining chapters of this thesis.
59
3.4 Conclusion
In summary, it was demonstrated that the 1:1 phyllosilicate Ni
3
Si
2
O
5
(OH)
4
can be used to oxidize benzyl alcohol using O
2
as the sole oxidant. Ni
3
Si
2
O
5
(OH)
4
is also able to oxidize cinnamyl alcohol. The oxidation was far superior using
Ni
3
Si
2
O
5
(OH)
4
in comparison to Co
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
, which
both showed no conversion. Through a series of physical characterization, the
structure of these compounds was described on the atomic level for the first
time. By TEM it was shown Co
3
Si
2
O
5
(OH)
4
forms small crystallites that may be
unstable therefore unable to perform like Ni
3
Si
2
O
5
(OH)
4
.
Through a series of catalytic reactions it was demonstrated Ni
3
Si
2
O
5
(OH)
4
shows increased reactivity correlated with increased oxygen pressure, but not
temperatures. This reactivity is inhibited in the presence of air, which may indi-
cate the active site is poisoned by CO
2
or N
2
. Furthermore, it lends credence to
the possibility these clays can activate small molecules. While the active site did
not show a temperature dependence, through literature search we discovered
that is likely a result of substrate choice, not the catalyst. The focus of future
chapters will be addressing the origin of activity in Ni
3
Si
2
O
5
(OH)
4
.
60
Chapter 4
Creation of Active sites on
Ni
3
Si
2
O
5
(OH)
4
by Thermal Activa-
tion
4.1 Introduction
Acetate and Formate
1
Figure 4.1: Glycerol oxidation scheme.
61
The need for rational design rules to create nontoxic, efficient, and effective
heterogeneous catalysts is an outstanding problem in materials science. The 1:1
phyllosilicate structure is an ideal model for creating structure-property relation-
ships. In our previous investigation we found Ni
3
Si
2
O
5
(OH)
4
can utilize molec-
ular oxygen to transform benzyl alcohol to benzaldehyde. Following this we
attempted to tune the reactivity of our active site.
Figure 4.2: The phyllosilicate crystal structure is comprised of alternating layers of
metal octahedra and cornersharing Si-O tetrahedra (top). The Si-O tetrahedra join
the metal octahedra at their apical oxygen. Layers are held together by relatively weak
hydrogen bonding by the surface hydroxyl groups.
Previously reported surface dehydration mechanisms for these compounds
preclude the ability to activate oxygen containing functional groups. [4, 5, 6, 7,
8, 9] Inability to study this complex structure has prevented the development of
more detailed correlations between the surface reactivity and catalytic potential.
62
However, if well understood, this process can be utilized to form similar active
sites on a wide range of materials.
Herein, we introduce a way to alter the active site of phyllosilicate catalysts.
We discuss a method to thermally modify its surface and its evaluation as a het-
erogeneous catalysts for the oxidation of glycerol. Through a combination of
local and surface sensitive techniques we demonstrate methods to interrogate
and understand the active site created at high temperatures. Using pair distru-
bution function analysis (PDF) we provide a comprehensive description of the
changes to our active site. We believe this not only demonstrates a way to mod-
ify a platform of catalysts but also study active sites created in high temperature
conditions.
4.2 Experimental
Reagents. The following reagents: metasilicic acid (H
2
SiO
3
) (Sigma-Aldrich,
99.9% ), nickel nitrate hexahydrate (Ni(NO
3
)
2
·6H
2
O) (EMD Millipore, 99% ),
sodium hydroxide (NaOH) , Glycerol (Sigma-Aldrich, Reagent Grade), DI H
2
O ,
O
2
, were used as received.
Synthetic Methods. To prepare the starting material, stoichiometric metasili-
cic acid H
2
SiO
3
was added to 15mL distilled water, then nickel nitrate
(Ni(NO
3
)
2
·6H
2
O) was added under constant stirring until dissolved. Following
this the solution was gelled by the addition of 1.5M NaOH. This gel was allowed
to sit 72 hours before being transferred to a Teflon lined stainless steel autoclave
and heated to 200
◦
C for 50 hours. The resulting teal powder was collected by
vacuum filtration, washed with distilled water and driedinvacuo. Followng this,
the catalyst was activated by heating at 500
◦
C for 6 hours in air.
63
Catalysis. Oxidation of glycerol was performed in a modified Fisher-Porter bot-
tle (FPB) with a total internal volume of 15mL (25.5mm x 10.22mm).[82] The
reactor was loaded with 50mg of catalyst and 5mL of aqueous glycerol solution
(1000 mg/ml). Tests were performed at pressures in the range of 50-90 psi O
2
and temperatures between 60 and 120
◦
C. The FPB was immersed in a silicone
oil bath and the temperature was controlled. At the end of the catalytic test
the reaction was cooled to room temperature. The reaction mixture was cen-
trifuged at 6,000 rpm for 10 minutes to separate the product mixture and the
catalyst. The general reaction procedure was the following: the catalyst (50mg)
and glycerol (1000 mg/ml, 5mL) were transferred to FPB, and purged with 65
psi oxygen three times before the reaction was initiated. The following reaction
conditions were used for all reactions unless stated otherwise: 0.05 g of catalyst,
5mL glycerol
a
q, 65 psi O
2
.
Density Functional Theory Calculations First principles calculations were
performed in the Vienna Ab Initio Simulation Package (VASP) [119], within the
projector augmented wave formalism [120], and the PBEsol functional [121]; a
cutoff energy of 500 eV was used. To account for van der Waals interactions the
D3 correction of Grimme was applied [122]. For calculations in which defects are
introduced at the clay surface it is necessary to correct for the well-known lack
of localisation of electrons within DFT; we have employed the DFT+U approach,
where a parameter (U) penalises partial occupancy of orbitals, thereby correcting
for the self-interaction error in DFT and reinforcing charge location’s. [123] A U
value of 5 eV for Ni is employed, as in previous calculations of Ni based systems.
[124] The slab models used to simulate the surface have a macroscopic dipole
across the unit cell due to the lack of inversion symmetry of the system normal
64
to the surface, this can lead to divergence of electrostatic energy and spurious
results. To correct for this in all cases final energies of slab systems had a dipole
correction scheme applied to ensure accurate energies. [125] Model systems
were constructed by initially taking the experimentally determined unit cell of
the biotite clay. This structure was then allowed to relax fully in terms of both
unit cell parameters and ionic positions, using the settings outlined above and a
k-point mesh sampling density with a target length cutoff of 25 Å, as prescribed
by Moreno and Soler. [126] The relaxed structure was then used to create a sur-
face structure by cleaving the [001] surface, using three double layers as the slab
thickness, with a 15 Å gap, which we have previously shown to be sufficient for
obtaining converged. electronic properties [127, 128] the surface was expanded
to a (2x1) supercell. The possible effects of deprotonation and dehydroxylation
were investigated by removing one hydrogen atom from the surface hydroxyls
and one complete surface hydroxyl group respectively. The preferred adsorption
geometry of the glycerol was determined by initially considering the molecule in
either standing or lying conformation, followed by a set of calculations of these
configurations with the molecule transposed at regular spacing across the sur-
face, sampling the entire surface with a resolution of 1 Å. A short molecular
dynamics (MD) calculation, 5000 steps of 0.5 fs, at 400 K was then performed
on the preferred configurations to allow the molecules to sample additional con-
figurational space. The MD calculations were then cooled from 400 K to 0 K at
a rate of 5 x 10-13 Ks
−1
, which we have previously applied to generate accurate
structures. [129]
Physical Characterization. Simultaneous thermogravimetric analysis (TGA)
and differential scanning calorimetry (DSC) was carried out on a Netzsch STA
65
449 F3 Jupiter (Netzsch Instrument North America) under air. Samples were
placed in an alumina pan and stabilized at 35
◦
C for 5 minutes, heated to 500
◦
C
at a rate of 8
◦
C/min, held at this temperature for 6 hours, then cooled to room
temperature at a rate of 8
◦
C/min. A correction was performed prior to each run
using both the reference and sample crucibles.
Laboratory X-ray diffraction patterns were collected on a Bruker D8 diffractome-
ter with a Cu-Kα source equipped with a motorized anti-air scattter screen and
Lynxeye detector. 100K Synchrotron X-ray total scattering data were collected
on beamline 11-ID-B at the advanced photon source (APS) at Argonne national
lab (ANL). Rapid-acquisition PDF method was used with an X-ray energy of
58.6 eV at a wavelength of 0.2114 Å,.[103] A Perkin Elmer amorphous Si two-
dimensional image-plate detector (2048 x 2048 pixels and 200 x 200 mm pixel
size) was used for 2 dimensional data collection with a sample to detector dis-
tance of 170 mm. The two dimensional data were converted to one-dimensional
XRD data using the FIT2D software. [104]
Neutron pair distribution function (PDF) data were transformed from total
scattering data collected on the NOMAD beamline at the Spallation Neutron
Source (SNS) at Oak Ridge National Laboratory at 60 Hz setting. For the current
experiment, about 2g of powder was loaded into a 6 mm vanadium can. a total of
three 30-minute scans were collected for each powder sample and then summed
together to improve statistics. The detectors were calibrated using scattering
from a diamond powder standard prior to measurement. Data were normalized
against a vanadium rod, the background was subtracted and the total structure
factor were transformed into PDF data using specific IDL codes developed for
the NOMAD scattering instrument with a Q range of 0-31.4 Å and a hydrogen
66
correction factor was applied. [78] The local structure was refined in PDFgui.
[106] Structures and charge density were visualized using VESTA. [81]
FTIR spectra were collected using a Bruker Vertex 80 FTIR spectrometer with
OPUS software under vacuum. The spectrometer consisted of a silicon carbide
IR source, KBr beam splitter, and a room temperature DTGS detector. Samples
were pressed into KBr pellets and placed in IR cards. Spectra collected consisted
of 32 scans with 0.2 cm
−1
resolution. Data analysis including corrections were
done within qtgrace.
The broadband SFG spectroscopy experimental setup has been described in
previous publications in more detail.[130] The 14 W output from the Ti:Sapphire
laser operating at a repetition rate of 5 kHz was split into femtosecond IR and
picosecond visible pulses centered at 2.74μm and 804 nm, respectively. The IR
and visible pulses were focused to an 200 μm spot size. To generate the SFG
signal, the pulses were overlapped spatially and temporally at the air-catalyst
interface in the inverted reflection geometry. The SFG spectra was measured
in SSP polarization (SFG, visible, IR). Microscopy images were acquired using
JEOL 2100F TEM at the University of Southern Californiaâ
˘
A
´
Zs Core Center of
Excellence for NanoImaging.
Organic Characterization. Organic products were identified using
1
H or
13
C
NMR. NMR spectra were recorded on a Varian VNMRS 400 or VNMRS 600 spec-
trometer and processed using MestroNova. All chemical shifts are reportead in
units of ppm and referenced to the residual
1
H or
13
C solvent peak. The concen-
tration of the substrate and the product were determined by high performance
liquid chromatography (HPLC). HPLC-DAD-Q-TOF analysis was performed on an
Agilent series 1290 Infinity HPLC instrument (Agilent Technologies, Santa Clara,
67
CA, USA) coupled with an Agilent 6545b Dual AJS ESI Q-TOF mass spectrometer
equipped with a dual (AJS) and electrospray (ESI) interface. Chromatographic
separation was carried out at 30
◦
C on an Agilent Zorbax-Bonus RP (2.1 mm x
150 mm, I.D., 2.7 um). The chromatographic conditions were as follows: flow
rate of 5.0 mL/min, sample injection volume of 0.5 uL, mobile phase A (0.1%
formic acid, 0.5 mM aqueous ammonium acetate) and mobile phase B (0.1%
formic acid in methanol). The gradient profile was optimized as follows: Re-
equilibration duration was 2 min between individual runs. For MS detection, the
operating parameters were as follows: drying gas flow rate, 9.0 L/min; drying
gas temperature, 350 C; nebulizer, 45 psig; capillary, 3000 V; skimmer, 65 V; Oct
RFV, 750 V; and fragmentor voltage, 130 V. At 1.6 minutes the detector polarity
switches from positive to negative. Glycerol and lactic acid were monitored by
the characteristic ion masses of 115
M
/
z
and 89
M
/
z
. All the acquisition and
analysis of data were controlled by MassHunter software (Agilent Technologies).
4.3 Results and Discussion
In this chapter, a 2 step hydrothermal synthesis was used to obtain a mix of
tube and platelet-like crystallites of Ni
3
Si
2
O
5
(OH)
4
and a method was developed
to create new catalytic sites by thermal treatment. To prepare the starting mate-
rial, stoichiometric amounts of the precursors were dissolved and gelled by the
addition of NaOH (1.5M) as described in the experimental section. This gel was
allowed to rest 72 hours before being transferred to a teflon lined stainless steel
autoclave (23mL) and heated to 200
◦
C for 50 hours. The resulting teal pow-
der was collected by vacuum filtration, washed with distilled water, and dried
at 80
◦
C. Thermal activation of Ni
3
Si
2
O
5
(OH)
4
was carried out by heating the
68
material at 500
◦
C for 6 hours in static air. On activation, the pristine powder
changes from teal to black in color. As the catalysts are prepared with no reduc-
tion step neither metallic Ni nor Ni
1+
are observed by magnetic measurement
(Figure 4.8). The catalytic ability of the as prepared and thermally treated were
compared for the oxidation of glycerol.
TEM images of the as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
with an aver-
age size of 50-100nm can be seen in Figure 4.3. The compositional effect of
(a) (b)
Figure 4.3: TEM of as-prepared (a) and activated (b) Ni
3
Si
2
O
5
(OH)
4
showing mor-
phology is retained upon activation. TGA data (bottom panel) shows a 12% mass loss
after activation.
thermal activation was first probed using combined thermogravimetric anaylsis
69
and differential scanning calorimetry that shows a total mass loss of 12% [Fig-
ure 4.3]. Based on the mass of the as-prepared material the mass loss correlates
to removal of two waters, presumably one interlayer and one surface. [4] This
implies thermal activation causes a surface condensation-dehydration process
that likely results in dehydroxylation and deprotonation occurring at the metal
site. [4] In order to identify and ascertain the chemistry associated with the
change in structure and catalytic activity, synchrotron x-ray diffraction and neu-
tron diffraction studies were performed. The catalytic surface was investigated
by sum frequency generation spectroscopy and electronic structure was probed
by XPS, DFT calculation, and magnetic measurement.
4.3.1 Structure
Powder x-ray diffraction shows the long range order of as-prepared
Ni
3
Si
2
O
5
(OH)
4
is retained upon heating. The corresponding XRD patterns, seen
in Figure 4.4, match well with the nickel serpentine mineral, Pecoraite (JCPDS
49-1859), without evidence of impurities. The broad, asymmetric, and miss-
ing peaks in both patterns are manifestations of small particles size, turbostratic
disorder and strain, which make this data difficult to refine. [108] Selective
broadening caused by the basal spacing of the layers is reflected by the (002)
and (004) peaks. Stacking faults stemming from weak electrostatic attraction
between layers result in loss of coherence about the stacking axis. [108] This
broadness is also influenced by the incorporation interlayer water as seen by
FTIR and TGA/DSC study. The x-ray diffraction pattern for the thermally treated
shows the most prominent change is a loss of intensity on the (00l) peaks, pre-
sumably the result of delamination as residual water is removed.
70
Intensity (arb. units)
As-prepared Ni
3
Si
2
O
5
(OH)
4
20 40 60 80
2 � (deg) [ � = 1.54184Å]
Intensity (arb. units)
Activated Ni
3
Si
2
O
5
(OH)
4
Figure 4.4: Laboratory powder XRD of as-prepared (top) and thermally activated
(bottom) Ni
3
Si
2
O
5
(OH)
4
showing crystallinity is retained. Miller indices are based on
the reported monoclinic structure (Cc). The highlighted peaks, [002] and [004], reduce
in intensity upon activation.
Intralayer strain, intrinsic to these materials, is caused by a size mismatch
between the shared oxygen of the silicate tetrahedra and the metal octahedra.
[42, 56, 62] Twisting of the hexagonal rings formed by the silica tetrahedra can
accommodate the lattice difference between octahedral and tetrahedral layers.
Because this does not affect the average structure it cannot be adequately mod-
eled by diffraction. [53, 108] Distorting the silicate ring or the hydroxide layer
71
2 3 4 5 6 7
(Å) r
-6
-3
0
3
PDF(Å
-2
)
(d) (c)
-6
-3
0
3
6
PDF(Å
-2
)
-6
-3
0
3
6
PDF(Å
-2
)
(a)
-3
0
3
6
PDF(Å
-2
)
(b)
2 3 4 5 6 7
(Å) r
Figure 4.5: Least-squares fits to the (a) synchrotron X-ray and (c) spallation neutron
experimental PDF at r>1.4 for as-prepared Ni
3
Si
2
O
5
(OH)
4
and (b) synchrotron X-ray
and (d) spallation neutron and thermally activated Ni
3
Si
2
O
5
(OH)
4
. Difference curves
are shown at the bottom of each panel.
may lead to extended defects such as bending or puckering of the sheets. These
may be qualitatively described from XRD but do not allow description of the
atomic local geometry giving rise the new active site.
Due to challenges associated with this strain the local structure was deter-
mined first and the long-range structure later reconstructed. For the first time,
combined x-ray and neutron PDF data were used to identify the accurate local
structure of Ni
3
Si
2
O
5
(OH)
4
. The results of the local structure of the catalyti-
cally active site were then compared to analysis of complimentary techniques to
describe the local structure (FTIR), surface (SFG), and electronic structure (mag-
netism, XPS, and DFT calculation). It may also provide insight for the rational
design of transition metal oxides capable of oxidation catalysis.
Local Structure Determination from Pair Distribution Function Analysis In
the previous chapter, we described the as-prepared Ni
3
Si
2
O
5
(OH)
4
in the C
c
space group with unit cell parameters 5.24 Å , 9.37 Å , 7.45 Å , and 93.7
◦
.
72
From this we determined the structure change upon activation. We employed
PDF analysis to probe specific local deviations that cannot be described by the
average structure. These distortions contribute to the broadening and diffuse
nature of XRD peaks. [47, 108] This is converted via appropriate Fourier trans-
form of the total scattering data to real-space atom atom correlations in the PDF.
Experimental PDFs for as-prepared and thermally activated Ni
3
Si
2
O
5
(OH)
4
are shown as points in the four panels of Figure 4.5. While modeling the x-ray
PDF data provides accurate local environments for relatively heavy elements such
as nickel and silicon, the hydrogen atoms (hydroxyls and interlayer water) are
ignored because of the very low sensitivity of x-rays towards light elements. In
contrast, neutron scattering is isotope-dependent and very sensitive to hydrogen
(b = -3.74 fm) and deuterium (D;b = 6.67 fm). [11] To account for those light
elements x-ray and neutron PDF data were simultaneously refined. A large drop
of fit after 7.5 Å indicates the limits of layer coherence along the z axis and the
deviation of the average structure from the microstructure. [11, 114]
The refinement is initially confined to a single layer unit cell (7.5 Å ) to
deconvolute the effect of stacking faults as the data begins to include mid and
long-range order. Simultaneous refinement of x-ray and neutron PDF for as-
prepared and thermally treated Ni
3
Si
2
O
5
(OH)
4
indicates that the average Si-
O bond length is about 1.61 Å , in good agreement with our previous results.
[109, 110, 111, 112, 113] Pair distribution function indicates that both the aver-
age Si-O and Ni-O bond length (2.05 Å ) are retained upon activation. While
these bonds are maintained upon activation, there is a slight contraction of the
unit cell, seen by the slight leftward shift in the next nearest neighbor atom posi-
tions at distances greater than 2.8 Å . This is corroborated by the shift in lattice
parameters in thermally treated Ni
3
Si
2
O
5
(OH)
4
to 5.16 Å , 9.28 Å , 7.52 Å , and
73
94.04
◦
. This confirms that the particles lose coherence in their stacking direction
and sinter along theab plane indicating a rehybridization of the Ni-O bonds. The
former expansion along the z axis is consistent with diminished basal spacing
seen in XRD, implying the extended stacking structure has collapsed somewhat.
The latter supports the idea that surface H
2
O has been lost, resulting in a local
change to the active site. This change is associated with the long range atom-
atom correlations between the hydroxyls and nickel and silicon. Contractions
indicate an increase in the order of bonding between oxygen and nickel.
The pristine and activated patterns, while similar, show the inter-atomic dis-
tances at 3.41 Å , 3.7 Å and 4.4 Å are significantly affected by thermal activation.
These correspond to O-OH and Ni-OH distances indicating the loss of hydroxyls
upon the activation of Ni
3
Si
2
O
5
(OH)
4
is seen in the short-range order. It is impor-
tant to note increased peak intensity seen in Figure 4.5d at the O-OH and Ni-OH
distances in activated Ni
3
Si
2
O
5
(OH)
4
correspond to a decrease in the negatively
scattering H
+
.
Vibrational spectroscopy provides useful qualitative information about the
nature of light elements. The observed differences in the shape and position
of the IR absorption bands can be used to distinguish the result of dehydration
across the surface. The FTIR of Ni
3
Si
2
O
5
(OH)
4
has previously been reported and
analyzed. [131] Thermal treatment causes significant changes to the spectra, as
shown in Figure 4.6. The interlayer water is decreased upon activation, as seen
by the decrease in intensity of the hydrogen bonded OH region from 3400-3200
cm
−1
and the OH bend at 1632 cm
−1
. [132] The hydroxyl stretching region
of nickel phyllosilicate has been identified as an in-phase inner-surface Ni
3
-OH
stretching at 3648 cm
−1
and a doubly degenerate out-of-phase stretching at 3610
74
500 1000 1500 2000 2500 3000 3500 4000
Wavenumber (cm
-1
)
0
0.2
0.4
0.6
0.8
1
Normalized transmission
3500 3650 3800
3500 3600 3700 3800
Wavenumber (cm
-1
)
0
500
1000
1500
2000
SFG Intensity (arb. units)
3500 3600 3700 3800
Wavenumber (cm
-1
)
200
400
600
800
SFG Intensity (arb. units)
Figure 4.6: The top panel shows FTIR before (blue) and after (black) activation
demonstrating a the growth of the a high frequency peak at 3400 wavenumbers corre-
lated to the creation of a new Ni-O stretch. The corresponding SFG data (lower) also
shows the growth of a new signal
cm
−1
.[131] Activation causes a splitting of the hydroxyl region with a new moi-
ety seen at 3628 cm
−1
. This splitting is induced by a change in the hydroxyl
environments. Vibrational sum frequency generation (SFG) spectra clearly show
the new hydroxyl feature, as seen in the lower panels of Figure 4.6. Orienta-
tional studies of nickel phyllosilicate and the thermally treated phyllosilicate are
currently underway.
75
Figure 4.7: Oxygen 1S core level XPS of as-prepared (light blue) and thermally
activated (dark blue) Ni
3
Si
2
O
5
(OH)
4
showing a shift to higher energy in the activated
sample. Fits to the as prepared (a) and thermally activated (b) Ni
3
Si
2
O
5
(OH)
4
show
this shift correlates with an decrease in hydroxide environments and growth of a new
Ni-OOH region.
To further probe the electronic structure the chemical states of Ni, Si and O
were investigated using XPS. The O1s core line in the as-prepared sample could
be deconvoluted into two peaks at binding energy of 532.78 eV and 531.57 eV
assigned to Ni/Si-O environments and Ni/Si-OH environments respectively.
Upon activation the O 1s core line shifts to higher energy and a third contri-
bution can be seen at 530.15 eV, attributed to the development of a new type of
Ni-O-OH environment. [133] The overall shift of the O 1s core from 531.87 eV
to 532.59 eV upon activation reflects the decrease in number of hydroxides in the
activated Ni
3
Si
2
O
5
(OH)
4
indicating the ratio of OH:O ions on the surface shifts
from 61:39 in the pristine sample to 32:68 in the activated sample. This suggests
30% of the protons attached to the hydroxyl were lost during activation. This is
76
(a) (b)
(c) (d)
(e) (f)
Figure 4.8: Magnetic measurements of as-prepared Ni
3
Si
2
O
5
(OH)
4
(a, c, e), and
thermally activated Ni
3
Si
2
O
5
(OH)
4
(b, d, f). Measurements of both phases show
characteristic superparamagnetic wiggles in the hysteresis. As-prepared Ni
3
Si
2
O
5
(OH)
4
is 100% Ni
+2
and the thermally-activated Ni
3
Si
2
O
5
(OH)
4
is 89% Ni
+3
77
(a) (b)
Figure 4.9: Band structures of as-prepared Ni
3
Si
2
O
5
(OH)
4
(a) and thermally-
activated Ni
3
Si
2
O
5
(OH)
4
(b) shows the re-hybridization of the Ni-O bands into the
band-gap.
further reflected in the Si 2p states, which shift to high energy after activation,
indicating an increase in the prevalence of silicate groups compared to hydroxyl
corresponding to the loss of interlayer hydrogen bonding. Using the Ni 2p lines
it is difficult to discriminate the exact oxidation state, but the activated phase
does show shift to higher energy, consistent with oxidation to the trivalent state,
supporting our previously indicated dehydration mechanism. [133] Changes in
the relative intensities of the contributions to the valence band are seen upon
activation supporting band structure-rehybridization.
Magnetism shows the strong black color upon activation can be ascribed to
the oxidation of 80% of Ni
2+
to Ni
3+
upon activation [Figure 4.8]. DFT calcu-
lation shows the creation of oxygen defects on the surface results in the rehy-
dbridization of the Ni-O bonds, leading to their energy dropping into the band
gap in Figures 4.9 and 4.10. This electronic evolution is confirmed via the XPS
conducted on the as-prepared and activated samples.
78
(a)
(b)
Figure 4.10: Density of states calculation for the as-prepared Ni
3
Si
2
O
5
(OH)
4
(a)
thermally-activated Ni
3
Si
2
O
5
(OH)
4
(b) where the labels p0, p1, p2, and p3 correspond
to bands for Si, O, Ni, and H respectively.
4.3.2 Catalytic performance
The catalytic activity of the as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
were
systematically tested for the oxidation of glycerol under basic conditions. The
catalytic experiments were first performed under expected conditions for glycerol
oxidation, using 50mg of catalyst, 5mL of aqueous glycerol (100 mg/mL), and
1.5M NaOH, at 110
◦
C and 65 psi O
2
. The oxidation reactions were performed
in a modified Fischer-Porter bottle (FPB) following the procedure outlined in the
experimental section unless otherwise noted. Both as-prepared and activated
79
Table4.1: Testconditionsfortheoxidationofglycerolbyas-preparedNi
3
Si
2
O
5
(OH)
4
.
All reactions were performed with 50mg of catalyst in 5 mL glycerol (1M) and 2:1
NaOH:glycerol, stirring at 355 rpm for 3 hours unless otherwise stated.
Entry Catalyst Temp(
◦
C) Pressure (PSI) Conv LA
1 Ni
3
Si
2
O
5
(OH)
4
60 65 5 3
2 Ni
3
Si
2
O
5
(OH)
4
80 65 5 3
3 Ni
3
Si
2
O
5
(OH)
4
100 65 4 2
4 Ni
3
Si
2
O
5
(OH)
4
120 65 6 2
5 Ni
3
Si
2
O
5
(OH)
4
120 45 4 1
6 Ni
3
Si
2
O
5
(OH)
4
120 85 13 10
7 Ni
3
Si
2
O
5
(OH)
4
120 0 3 2
Table 4.2: Test conditions for the oxidation of glycerol by activated Ni
3
Si
2
O
5
(OH)
4
.
All reactions were performed with 50 mg of catalyst in 5 mL glycerol (1M) and 2:1
NaOH:Glycerol, stirring at 355 rpm for 3 hours unless otherwise stated.
Entry Catalyst Temp(
◦
C) Pressure (PSI) Conv LA
1 Ni
3
Si
2
O
5
(OH)
4
60 65 5 1
2 Ni
3
Si
2
O
5
(OH)
4
80 65 4 2
3 Ni
3
Si
2
O
5
(OH)
4
100 65 3 1
4 Ni
3
Si
2
O
5
(OH)
4
120 65 4 1
5 Ni
3
Si
2
O
5
(OH)
4
120 45 3 1
6 Ni
3
Si
2
O
5
(OH)
4
120 85 11 8
7 Ni
3
Si
2
O
5
(OH)
4
120 0 3 1
Ni
3
Si
2
O
5
(OH)
4
were active for the oxidation of glycerol under basic conditions.
Oxidation products lactic acid, acetate, formate and ethylene glycol identified
by
1
H NMR. The products were then separated and lactic acid quantified by
UPLC-DAD-Qtof mass spectroscopy. Under neutral conditions, no products were
noted. In addition, because sodium hydroxide itself can act as a catalyst for
the oxidation of glycerol a rigorous control was run at optimal conditions in
the absence of catalyst. [134] When the reaction was performed at optimal
conditions in the absence of catalyst the conversion was 5% .
After determining the as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
were effec-
tive oxidation catalysts reaction conditions were screened to optimize catalyst
performance. Tables 4.1 and 4.2 summarize glycerol conversion after 3 hours
80
Table4.3: Optimizationofbaseconcentrationforoxidationofglycerolbyas-prepared
Ni
3
Si
2
O
5
(OH)
4
.
Entry Glycerol Concentration Base Ratio Conv
1 1M NaOH 0.5:1 1.25%
2 1M NaOH 1:1 1 %
3 1M NaOH 2:1 3 %
4 1M KOH 2:1 5 %
5 1M Neutral 0 %
Table 4.4: Optimization of base concentration for oxidation of glycerol by activated
Ni
3
Si
2
O
5
(OH)
4
.
Entry Glycerol Concentration Base Ratio Conv
1 1M NaOH 0.5:1 0.09%
2 1M NaOH 1:1 0.09 %
3 1M NaOH 2:1 1.5 %
4 1M KOH 2:1 5 %
5 1M Neutral 0 %
at differing conditions for both the as-prepared and thermally activated cata-
lyst. As expected, higher temperatures favor increased reaction rates, leading
to an increase in glycerol conversion from 3 - 13% over the temperature range
of 60-120
◦
C. However, increasing the temperature also decreases the selectiv-
ity leading to more decarboxylation products such as ethylene glycol, acetate,
and formate. Overall, under optimal conditions for lactic acid production both
as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
selectively produce lactic acid at con-
versions of 13% and 11% respectively.
The effect of oxygen concentration was examined in the range of 50-90 psi,
and compared to air and nitrogen at optimal conditions. In nitrogen atmosphere,
ethylene glycol is the major product by NMR, showing absence of oxygen hinders
the oxidation pathway, resulting in ethylene glycol being formed as the major
product as a result of decorboxylation. Switching from O
2
to air results in a 50%
loss in conversion, seemingly related to the lower concentration of O
2
in air or
81
inhibition by the presence of CO
2
. As can be seen by the products, increasing
the O
2
pressure increases the oxidation and decarboxylation products, whereas
at lower pressure lactic acid is produced selectively.
Finally, the influence of the base was investigated by varying molar
NaOH:glycerol from 2:1 to 0.5:1. At NaOH:glycerol ratios larger than 2:1 the
as-prepared and activated Ni
3
Si
2
O
5
(OH)
4
become unstable and begin to decom-
pose. We found the optimal ratio of NaOH to glycerol to be 2:1 molar. To investi-
gate the electrostatic effects of the cocatalyst experiments were performed using
KOH resulting in improved yield. This implies the electrostatic effect of the cocat-
alyst may play a significant role in polarizing the active site. Tables 4.3 and
4.4 summarize the cocatalyst variations, concentration profiles and conversion.
Increased base contributes higher lactic acid yields, but decreases the stability of
the catalyst, whereas low base increases reusability but lowers conversion.
Following the reaction the catalyst was separated from the substrate by cen-
trifuging for 10 minutes at 6,000 rpm and then carefully separating the liquid
and solid by pipette. The spent catalyst were washed with deionized water,
ethanol and acetone at room temperature and drying under vacuum. From the
XRD patterns of the spent catalysts, after 3 hours of glycerol oxidation, neither
additional crystalline phases, nor large changes relative intensities or shapes are
seen. The above-mentioned data indicates that the catalyst does not undergo
significant bulk structural changes during the catalytic reaction.
4.3.3 Reaction pathways for oxidation of glycerol by as-
prepared and thermally activated catalyst
As can be seen via the tandem reaction pathways (Figure 4.11) and results,
the glycerol oxidation pathway for Ni
3
Si
2
O
5
(OH)
4
shows competition between
82
OH
OH
HO
Glycerol
O
HO OH
1,3 Dihydroxyacetone
OH
HO O
Glyceraldehyde
O
O
OH HO
Hydroxypyruvic acid
O
O HO
Hydroxypyruvaldehyde
OH
O
OH
Lactic Acid
OH
HO
O
OH
Glyceric Acid
HO
OH
Ethane 1,2-diol
O
2
O
2
O
O
Pyruvaldehyde
O
2
-H
2
O
+H
2
O
O
2 O
2
-CO
2
rearrangement
O
2
O
O
-
Acetate
O
-
O
Formate
O
2
Figure 4.11: Proposed tandem reaction pathways for the oxidation and decarboxyla-
tion of glycerol.
oxidation and decarboxylation. Lactic acid can be directly produced during the
hydrogenoylsis of glycerol in an alkaline solution and under pressure of oxygen.
Ni
3
Si
2
O
5
(OH)
4
’s ability to produce hydrogen suggests it can access this route.
This is likely because the hydroxide ions and surface hydroxyls both facilitate a
H-abstraction in the intial dehydrogenation of the hydroxyl group of glycerol and
also catalyze the 1,2 intramolecular dihydride shift. [135] For Ni
3
Si
2
O
5
(OH)
4
base is also needed to polarize the active site, which shows a preference for K
+1
.
The absence of oxygen, presence of C2 and C1 moieties indicates the strong
prevalence for these materials to decarboxylate terminal carboxyl sites.
83
4.4 Conclusions
In conclusion, we prepared an ordered, lamellar Ni
3
Si
2
O
5
(OH)
4
, thermally
activated it by heating in air at 500
◦
C, and investigated their ability to oxi-
dize glycerol in basic media. The catalyst is a cost efficient and environmentally
friendly alternative to reported nobel metal glycerol oxidation catalysts. Compar-
ison between the as-prepared and thermally activated Ni
3
Si
2
O
5
(OH)
4
shows sim-
ilar activity and stability. The oxidation of glycerol at optimal conditions yields
Lactic acid, C2 ethylene glycol and sodium acetate, and C1 sodium formate as
by-products. The conversion can be increased by tuning the oxygen pressure and
base concentration. These results suggest thermal activation results in Ni-O bond
change giving this change in catalytic activity. Further studies will be conducted
by our group in order to understand the catalytic qualities of these materials. We
feel this calls attention to a largely tunable earth-abundant model catalyst.
84
Chapter 5
Tailorable Frustrated Lewis Pair-like
Active Sites on
Bimetallic 1:1 Phyllosilicate Cata-
lysts
5.1 Introduction
The need to develop nobel-metal-free catalysts capable of activating small
molecules (e.g., H
2
, CO, CO
2
, NO
x
and many others) requires new synthetic
strategies to create active materials. [136, 137, 138, 139] In solution chemistry
homogeneous frustrated Lewis pairs (FLPs) have attracted considerable attention
by enabling metal-free activation of small molecules. [136, 137, 138, 139] These
85
systems have been explored extensively for several advanced organic reactions,
radical chemistry and polymerizations. [136, 137, 138, 139] Inspired by the
substantial progress in homogeneous systems, we sought to develop a model to
understand heterogeneous FLP systems. In this report, we highlight methods
to construct tailorable FLP-like active sites on the surface of 1:1 phyllosilicate
catalysts.
A FLP originates from the combination of a Lewis acid and a Lewis base that
are physically prevented from forming a typical Lewis acid-base adduct. [136] In
solution FLPs can be formed by following a reliable protocol: employ functional
groups that both inhibit the direct adduction of the Lewis acid and base, and
preserve the requisite interaction between the Lewis acid and base. [136] In this
way, heterogeneous catalysts do not yet have dependable design principles. Sur-
face FLP sites, analogous to homogeneous FLPs, may be effectively constructed
on solid with tailorable surface groups, such as hydroxyls, vacancies, or anionic
sites. [140, 141, 142] Semi-solid FLP catalysts, such as combinations of gold
nanoparticles and imines for the hydrogenation of nitriles under mild conditions
have been investigated. [143] Additionally, Ozin et al successfully composed
physically separated surface Lewis base and Lewis acid pairs on In
2
O
3x
(OH)
y
by
regulating oxygen vacancies.[142, 144] However, much is still unknown about
the fundamental physical properties that control fabrication of catalytic sites on
solid surfaces.
Systems with tunable surfaces are ideal for investigating the noble-metal-
free activation of small molecules on solids. [140, 141] In the previous chapter
Ni
3
Si
2
O
5
(OH)
4
exhibited tunable surface that allowed for the formation of new
active sites on a well-defined surface. Previously, we investigated a heat treat-
ment method that gives rise to an oxygen vacancy correlated to a Lewis acidic
86
metal and Lewis basic oxo-environment. Through catalytic study this new site
was correlated with the production of H
2
. To gain insight into this mechanism on
how this could be extended reactivities we sought to probe if this site can acti-
vate H
2
. Herein, we explore the creation of FLP-like active sites on the surface of
phyllosilicate catalysts.
A proximal Lewis acid-base surface site and mechanism are supported by
preliminary structure investigations. A number of representatives in the series
Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
and Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
were tested for reduction of
cyclohexene. Results show the addition of zinc leads to a marked improvement of
both activity. Through x-ray PDF and TPD we correlated this to increased acidity.
The results of this study emphasizes the importance of engineering nanostructure
surfaces to facilitate technologically relevant reactions.
5.2 Experimental
Synthesis Ni
3
Si
2
O
5
(OH)
4
, Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
, and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
,
were prepared hydrothermally. In a typical preparation, stoichiometric amounts
of precursor salts [Ni(NO
3
)
2
· 7H
2
O (EMD millipore), Mg(NO
3
)
2
· 7H
2
O (Sigma-
Aldrich) and Zn(NO
3
)
2
· 9H
2
O (Alfa Aesar)] were combined with 2mmol H
2
SiO
3
(Sigma-Aldrich), in 15mL DI H
2
O. Following dissolution silica polymerization
was induced by the addition of 8.75mmol NaOH. The resulting silica gel was
allowed to age for 2-10 days before being transferred to a 23mL teflon lined
autoclave. The autoclave was sealed and heated to 200
◦
C for 2 days. Upon
cooling, the resulting powder was collected via vacuum filtration and washed
with DI water. The powder was dried at 100
◦
C then activated in a box furnace
in static air at 500
◦
C for 6 hours.
87
Characterization techniques Powder X-ray diffraction was collected on a
Bruker Advanced D8 Diffractometer equipped with a motorized air scatter screen
and lynxeye detector, using Cu-Kα radiation. Synchrotron x-ray total scatter-
ing data were collected at 100K on beamline 11-ID-B at the Advanced Photon
Source, Argonne National Laboratory, with an X-ray wavelength of 0.2114 (about
58.6 keV). Samples were ground thorough in an agate mortar and packed in Kap-
ton tubes. A 2D amorphous Si image-plate detector (PerkinElmer, 2048 x 2048
pixels and 200 x 200 mm pixel size) was used for two-dimensional data collec-
tion with a sample-to-detector distance of ˜ 170mm. The data was the converted
to 1D XRD dtata using Fit2D software. [104] The PDF was then created from
Fourier transform of the total scattering data in PDFgetX3 with a Q range of 0.2
- 22 Å
−1
. [105]
5.3 Results
Phyllosilicate catalysts facilitate carbon cracking by activating a substrates
carboxyl bond at a strong lewis-acidic site. [67] In the previous chapter we
demonstrated upon heating Ni
3
Si
2
O
5
(OH)
4
, releases hydroxyls from its surface
in the form of H
2
O. Upon this release, a new active site is created that is capable
of oxidizing glycerol, and in the process, creating hydrogen gas. This site should
be analogous to a homogeneous FLP, in which the removal of -OH leads to a
Lewis-acid, and the removal of a single hydrogen results in a Lewis-base. Our
proposed active site is composed of a Lewis base adjacent to a Lewis acid that
facilitates the dissociation of of H
2
enabling reduction at the metal hydride.
88
Preparation. Ni
3
Si
2
O
5
(OH)
4
, Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
, and
Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
were prepared from an amorphous silica-gel precur-
sor obtained from hydrolysis of silicic acid. Stoichimetric amounts of precursor
salts were combined with silicic acid and gelled via the addition of NaOH. The
resulting gel was allowed to age at least two days before transferring to a teflon
lined autoclave. Phase pure materials were obtained after allowing the autoclave
to incubate at 200
◦
C for at least 50 hours. The use of an amorphous gel and
aging is necessary for the creation of a homogeneous material. Synthetically
the entire Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
series can be accessed, including both end
members Ni
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
though, only the first is discussed
in this chapter. In comparison, only low Zn containing Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
phases are easily accessible and the Zn end member is unattainable, as has
been established by previous studies. [145] The Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
series can
only be synthesized to include about 16 % zinc. Following incubation samples
were collected via vacuum filtration, dried at 80
◦
C, and thermally activated by
heating at 500
◦
C in air for 6 hours.
Catalyst Psuedo-1rst order rate Conversion
Ni
3
Si
2
O
5
(OH)
4−x
0.2*10
3
3.2 %
[Ni
2.3
Mg
0.7
]Si
2
O
5
(OH)
4−x
1.45 5.8 %
[Ni
2.2
Mg
0.8
]Si
2
O
5
(OH)
4−x
1.45 5.7 %
[Ni
0.5
Mg
2.5
]Si
2
O
5
(OH)
4−x
0.81 10 %
[Ni
2.6
Zn
0.4
]Si
2
O
5
(OH)
4−x
1.26 5.6 %
[Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
7.37 18.3 %
[Ni
2.1
Zn
0.9
]Si
2
O
5
(OH)
4−x
1.86 6.9 %
None N/A 1 %
Table 5.1: Reduction of clycohexene by members of phyllosilicate series
Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
with Psuedo-1rst order rate for
hydrogen listed, only values with an r
2
> .98 accepted.
89
5.3.1 Reduction of cyclohexene
To establish FLP-like character in these compounds were tested for hydro-
genation of cyclohexene. As-prepared Ni
3
Si
2
O
5
(OH)
4
showed no conversion
when tested, demonstrating the thermally-created active site is responsible for
this reactivity. In a typical experiment approximately 100mg of catalyst was com-
bined with 40mL neat cyclohexene in a Parr reactor. Reactions were carried at
70
◦
C under 150 psi H
2
for about 300 minutes. To determine the most active cata-
lyst representatives of the Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
series
were tested under these conditions. Results showing the reduction of cyclohex-
ene are summarized in Table 5.1. In the absence of any catalyst the conversion
was about 1% . [Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
shows the highest conversion for any
catalyst tested. Using structure characterization we try to understand what gives
rise to the unique reactivity of [Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
.
5.3.2 Structure Analysis
Catalyst Sites (μmole per gram)
Ni
3
Si
2
O
5
(OH)
4
240
[Ni
3−x
Mg
x
]Si
2
O
5
(OH)
4
240
[Ni
3−x
Zn
x
]Si
2
O
5
(OH)
4
410
Table 5.2: Number of Lewis-acidic active sites for catalysts Ni
3
Si
2
O
5
(OH)
4
,
Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
, and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
.
Like the former chapters, these materials suffer from broad and overlapping
XRD reflections. However, these XRD may be qualitatively analyzed and pro-
vides valuable insight about the nature of the materials. XRD of the resulting
90
2 3 4 5 6 7 8 9 10 11 12
r (Å)
G(r) (Å)
10 20 30 40 50 60 70 80 90
2 �[ �=1.5406]Å
Intensity (Arb. Units)
x=0
x=0.5
x=1
x=1.5
x=2
x=2.5
x=3
(b)
(a)
Figure 5.1: XRD and X-ray PDF for as-prepared and thermally activated
Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
series catalysts. Highlighted regions indicate structural markers
for the inclusion of nickel.
Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
series [Figure 5.1] shows a gradual change in the propor-
tion and definition of the (200) and (202) peaks as the series approaches a Ni:Mg
ratio of 1:1. These peaks are sensitive to the distortion of the octahedral sheet
indicating nickel and magnesium induce different strain. As seen in Figure 5.2,
inclusion of zinc usually results in the shifting of the (002) peak to lower angles,
indicating a slight increase in the size of the unit cell. A tetrahedral substitution
by zinc, which has a known tetrahedral site preference, leads to an increase in
the unit cell size. [146]
Upon activation, like Ni
3
Si
2
O
5
(OH)
4
, these compounds lose intensity at the
(002) and (004) planes, indicating a loss of coherence between the sheets as seen
in chapter 4. Otherwise these materials exhibit a change in long range structure
so we turned to pair distribution function analysis to understand the origin of
activity differences.
91
PairDistributionFunctionAnalysis The local structure and unit cell of these
compounds was investigated by x-ray total scattering pair distribution function
analysis (PDF). The PDF for various members of the Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
and
Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
series can be seen Figures 5.1 and 5.2. These patterns
are very similar to those presented in the previous chapters, however, distinct
variations can bee seen at certain distances. The Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
series
are very similar, aside from reduced intensity at x=2.5, as a result of decreased
x-ray scattering power. Reduction of nickel also causes a loss of splitting in the
peak at 7Å indicating a change at the phase boundary. Upon activation the PDF
2 3 4 5 6 7 8 9 10
r (Å)
-4
-3
-2
-1
0
1
2
3
4
5
6
7
8
9
10
11
12
G(r) (Å)
x=0.6
x=0.5
x=0.4
x=0.3
x=0
10 20 30 40 50 60 70 80 90
2 �[ �=1.5406]Å
Intensity (Arb. Units)
x=0.6
x=0.5
x=0.4
x=0.3
x=0
(a)
(b)
Figure 5.2: XRD and X-ray PDF for as-prepared and thermally activated
Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
series catalysts. Highlighted regions indicate structural mark-
ers for the incorporation of zinc into the tetrahedral site of [Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
.
shifts, and shows a loss of the Fourier transform effect 2.34 Å . Futhermore, like
Ni
3
Si
2
O
5
(OH)
4
, upon activation Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
, and Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
contract in theab plane and elongate along the z-axis.
Examining our most active catalyst, [Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
, the width
and distribution of the tetrahedral bond distance (about 1.61 Å ) increases
92
with increasing zinc content. Furthermore, the incorporation of more zinc in
Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
results in the loss of the Fourier transform effect at 2.34
Å , indicating a fundamental change in the nature of the silicate face (2.37 Å ).
In Co
3
Si
2
O
5
(OH)
4
and Mg
3
Si
2
O
5
(OH)
4
this correlated to an increase in tetra-
hedra bond distortion, and is directly influenced by the metal filling of sheets.
Futhermore, recently zinc incorporated into silicate structures has been demon-
strated as active for hydrogenation, [75, 147, 148] and incorporation of zinc in
aluminosilicate structures has been shown to result in a large increase in acid-
ity. [149, 150, 151] We believe that the incorporation of zinc into this structure
results in the increased activity, and tetrahedral substitution of zinc improves
Lewis acidity.
5.4 Conclusions
In conclusion, several samples in the series Ni
3−x
Mg
x
Si
2
O
5
(OH)
4
and
Ni
3−x
Zn
x
Si
2
O
5
(OH)
4
synthesized and tested for the hydrogenation of cyclohex-
ene. Through a series of test it was shown the catalyst [Ni
2.4
Zn
0.6
]Si
2
O
5
(OH)
4−x
was the most active. Studies indicate increased activity may be due to the inclu-
sion of zinc, which was shown to tremendously increase the number of Lewis-
acidic sites in the structure. Further PDF studies indicate zinc likely substitutes
silicon on the tetrahedral site, resulting in some dual site functionality. A mecha-
nistic study examining the access of the site with different substrates would help
determine if the hydride is buried.
93
Chapter 6
Summary and Future Outlook
The tunability of both the structure and the composition of phyllosilicate
materials allow systematic studies that correlate their physical properties and
catalytic activity. The work discussed focused on establishing their intrinsic cat-
alytic activity as well as techniques to interrogate their structure at the atomic
level. These studies serve as a foundation to enable a wide-array of studies
that emphasize further understanding the activity of these compounds, in the
hope this insight can be extended to other materials. Through this work, phyl-
losilicates that incorporate a variety of reactive metals have been synthesized.
Using techniques such as diffraction, pair distribution function, FTIR, TEM, and
DFT calculation, fundamental atomic correlations between the activity and local
structure of phylloslicates have been made.
Some, 2:1 phyllosilicates, such as montmorillonite have previously been used
as cracking catalysts, while others, such as muscovite, are used in fireproofing
94
and lubricants. [67, 9] By incorporating transition metals into the muscovite
structure we examined its catalytic properties. Using a combination of of diffrac-
tion and spectroscopy techniques, we elucidate tetrahedral site-mixing, the metal
order of the octahedral sheet, and the relative charge distribution. Despite pre-
viously demonstrated surface redox [47], these compounds are not active oxida-
tion catalysts, likely because majority of their surface is blocked by the silicate
sheet and potassium is electrostatically bound between the sheets. [47]
Following this, we investigated 1:1 phyllosilicates Ni
3
Si
2
O
5
(OH)
4
,
Mg
3
Si
2
O
5
(OH)
4
, and Co
3
Si
2
O
5
(OH)
4
, demonstrating that when modified
to include nickel these compounds possess the ability to activate molecular
oxygen. We further probed this by x-ray PDF find all three materials possess a
similar local structure. Differences between Ni
3
Si
2
O
5
(OH)
4
and Co
3
Si
2
O
5
(OH)
4
and believed to arise from the low structural stability of the latter, while lack
of activity by Mg
3
Si
2
O
5
(OH)
4
confirms the presence of nickel facilitates this
reaction. While this work emphasizes the versatility of these compositions and
structures, we still sought to describe methods to modify the catalytic surface.
Elucidating the impact of thermal treatment on the surface hydroxyls of
phyllosilicates and related compounds, such as oxyhydroxides, and hydrotal-
cites, remains an active area of investigation. Several studies have shown that
upon heating phyllosilicates undergo a surface dehydration processes, resulting
in the loss of hydroxyl groups. [4, 5, 6, 7, 8, 9] Our study of Ni
3
Si
2
O
5
(OH)
4
demonstrates incorporation of redox active metals also results in for oxidation
at the metal center. This is further borne out by neutron PDF study, which indi-
cates Ni
3
Si
2
O
5
(OH)
4
dehydroxylates upon heating, and magnetic measurement,
which confirms the creation of Ni
+3
after activation. FTIR, XPS, and SFG confirm
the creation of a new nickel-oxo environment correlated with heat treatment.
95
Reactivity studies indicate this new active site mimics a metal nanoparticle. Fur-
thermore, it created a platform to investigate the create of FLP-like active sites on
the surface of hydroxylated compounds. This behavior was further exemplified
by the reduction of cyclohexene by activated cataysts in the presence of H
2
as the
only reductant.
Clarifying the role of the basic oxygen site and oxidized metal centers will
benefit from further studies done in the gas phase, to further borne out the reac-
tivity of these phases. Research to correlate the morphology, surface proper-
ties, and reaction conditions to activity are currently underway. The work on
thermally activated phyllosilicates indicates zinc-doped samples are capable of
effectively utilizing hydrogen gas. A systematic study investigating zinc in the
phyllosilicate structure is needed to elucidate its role as a Lewis-acid. Future
gas-phase reactions will allow dynamic study of the active site. Because of the
studies of the last two chapters, a great deal is known about the role of this site.
Some work has been done with these materials as electrocatalysts. Previously,
the materials KFe
3
Si
4
O
10
(OH)
2
, KCo
3
Si
4
O
10
(OH)
2
, and Ni
3
Si
2
O
5
(OH)
4
, as well
as its many variations, have shown promise for electrochemical reactions. Pre-
liminary results indicate they are active for both electrochemical water splitting,
as well as electrochemical oxidation. Further studies should be done to investi-
gate their potential in this realm, given the surface electrochemistry. The range
of topics explored in this dissertation develop understanding that can be used
as guiding principles to synthesize and effectively study catalysts for biomass
transform, selective oxidation, and hydrogenation chemistry.
96
Reference List
[1] Ronald Breslow. Biomimetic chemistry and artificial enzymes: Catalysis
by design. Accounts of Chemical Research, 28(3):146–153, mar 1995.
[2] Allen J. Bard and Marye Anne Fox. Artificial photosynthesis: Solar split-
ting of water to hydrogen and oxygen. Accounts of Chemical Research,
28(3):141–145, mar 1995.
[3] Cynthia M. Friend and Bingjun Xu. Heterogeneous catalysis: A central sci-
ence for a sustainable future. Accounts of Chemical Research, 50(3):517–
521, mar 2017.
[4] A. F. Gualtieri, C. Giacobbe, and C. Viti. The dehydroxylation of serpentine
group minerals. American Mineralogist, 97(4):666–680, mar 2012.
[5] M. Zhang, S. A. T. Redfern, E. K. H. Salje, M. A. Carpenter, and L. Wang.
H2o and the dehydroxylation of phyllosilicates: An infrared spectroscopic
study. American Mineralogist, 95(11-12):1686–1693, nov 2010.
[6] Congcong Che, Timothy D. Glotch, David L. Bish, Joseph R. Michalski,
and Wenqian Xu. Spectroscopic study of the dehydration and/or dehy-
droxylation of phyllosilicate and zeolite minerals. Journal of Geophysical
Research, 116(E5), may 2011.
[7] M. Zhang. Dehydroxylation, proton migration, and structural changes
in heated talc: An infrared spectroscopic study. American Mineralogist,
91(5-6):816–825, may 2006.
[8] Roy Trittschack and Bernard Grobéty. Dehydroxylation kinetics of
lizardite. European Journal of Mineralogy, 24(1):47–57, feb 2012.
97
[9] Lisa Heller-Kallai and Isaak Lapides. Dehydroxylation of muscovite: study
of quenched samples. PhysicsandChemistryofMinerals, 42(10):835–845,
jul 2015.
[10] He Zhang, Junming Sun, Vanessa L. Dagle, Barr Halevi, Abhaya K. Datye,
and Yong Wang. Influence of ZnO facets on pd/ZnO catalysts for methanol
steam reforming. ACS Catalysis, 4(7):2379–2386, jun 2014.
[11] Lichen Liu and Avelino Corma. Metal catalysts for heterogeneous catal-
ysis: From single atoms to nanoclusters and nanoparticles. Chemical
Reviews, 118(10):4981–5079, apr 2018.
[12] Peter B. Kettler. Platinum group metals in catalysis: fabrication of cat-
alysts and catalyst precursors. Organic Process Research & Development,
7(3):342–354, may 2003.
[13] S. R. Taylor. Abundance of chemical elements in the continental crust: a
new table. Geochim. Cosmochim. Acta, 28:1273, 1964.
[14] Zhoufeng Bian and Sibudjing Kawi. Preparation, characterization and
catalytic application of phyllosilicate: A review. CatalysisToday, dec 2018.
[15] Ji Chan Park, Shin Wook Kang, Jeong-Chul Kim, Jae In Kwon, Sanha
Jang, Geun Bae Rhim, Mijong Kim, Dong Hyun Chun, Ho-Tae Lee, Heon
Jung, Hyunjoon Song, and Jung-Il Yang. Synthesis of co/SiO2 hybrid
nanocatalyst via twisted co3si2o5(OH)4 nanosheets for high-temperature
fischer–tropsch reaction. Nano Research, 10(3):1044–1055, jan 2017.
[16] W. John Thomas John Meurig Thomas. Principles and Practice of Hetero-
geneous Catalysis. John Wiley & Sons, 2014.
[17] John Meurig Thomas, Robert Raja, Gopinathan Sankar, Robert G. Bell,
and Dewi W. Lewis. Benign by design. new catalysts for an environmen-
tally conscious age. Pure and Applied Chemistry, 73(7):1087–1101, jul
2001.
[18] Makarand R. Gogate. New paradigms and future critical directions in het-
erogeneous catalysis and multifunctional reactors. Chemical Engineering
Communications, 204(1):1–27, aug 2016.
[19] International Energy Association. Technology roadmap energy and ghg
reductions in the chemical industry via catalytic processes, 2013.
[20] Peter G. Levi and Jonathan M. Cullen. Mapping global flows of chemicals:
From fossil fuel feedstocks to chemical products. EnvironmentalScience&
Technology, 52(4):1725–1734, feb 2018.
98
[21] Zhaoqianqi Feng, Tengfei Zhang, Huaimin Wang, and Bing Xu.
Supramolecular catalysis and dynamic assemblies for medicine. Chemi-
cal Society Reviews, 46(21):6470–6479, 2017.
[22] Xi Wang, Xiaogang Chen, Stephen Cowling, Libo Wang, and Xuqing Liu.
Polymer brushes tethered ZnO crystal on cotton fiber and the applica-
tion on durable and washable UV protective clothing. Advanced Materials
Interfaces, page 1900564, jun 2019.
[23] C. Richard Catlow, Matthew Davidson, Christopher Hardacre, and Gra-
ham J. Hutchings. Catalysis making the world a better place. Philosophical
TransactionsoftheRoyalSocietyA:Mathematical,PhysicalandEngineering
Sciences, 374(2061):20150089, jan 2016.
[24] Bruce E Dale. Greening the chemical industry: research and development
priorities for biobased industrial products. JournalofChemicalTechnology
& Biotechnology, 78(10):1093–1103, 2003.
[25] Masakazu Iwamoto. Air pollution abatement through heterogeneous
catalysis. InStudiesinSurfaceScienceandCatalysis, pages 23–47. Elsevier,
2000.
[26] Steven Chu, Yi Cui, and Nian Liu. The path towards sustainable energy.
Nature Materials, 16(1):16–22, jan 2017.
[27] The Catalyst Group Resources. Business shifts in the global catalytic pro-
cess industries 2017 to 2023.
[28] Ilkeun Lee, Françoise Delbecq, Ricardo Morales, Manuel A. Albiter, and
Francisco Zaera. Tuning selectivity in catalysis by controlling particle
shape. Nature Materials, 8(2):132–138, jan 2009.
[29] U.S. Geological Survey. National minerals information center, June 2019.
[30] Q. Pan, L. Li, S. Shaikhutdinov, Y. Fujimori, M. Hollerer, M. Sterrer, and H.-
J. Freund. Model systems in heterogeneous catalysis: towards the design
and understanding of structure and electronic properties. FaradayDiscus-
sions, 208:307–323, 2018.
[31] Peter Strasser, Shirlaine Koh, Toyli Anniyev, Jeff Greeley, Karren More,
Chengfei Yu, Zengcai Liu, Sarp Kaya, Dennis Nordlund, Hirohito Oga-
sawara, Michael F. Toney, and Anders Nilsson. Lattice-strain control of
the activity in dealloyed core–shell fuel cell catalysts. Nature Chemistry,
2(6):454–460, apr 2010.
99
[32] J. H. Zhang, H. Y. Liu, Z. L. Wang, N. B. Ming, Z. R. Li, and A. S.
Biris. Polyvinylpyrrolidone-directed crystallization of zno with tunable
morphology and bandgap. Advanced Functional Materials, 17(18):3897–
3905, December 2007.
[33] Alexis Grimaud, Oscar Diaz-Morales, Binghong Han, Wesley T. Hong,
Yueh-Lin Lee, Livia Giordano, Kelsey A. Stoerzinger, Marc T. M. Koper,
and Yang Shao-Horn. Activating lattice oxygen redox reactions in metal
oxides to catalyse oxygen evolution. NatureChemistry, 9(5):457–465, jan
2017.
[34] Vojislav Stamenkovic, Bongjin Simon Mun, Karl J. J. Mayrhofer, Philip N.
Ross, Nenad M. Markovic, Jan Rossmeisl, Jeff Greeley, and Jens K.
Nørskov. Changing the activity of electrocatalysts for oxygen reduction
by tuning the surface electronic structure. Angewandte Chemie Interna-
tional Edition, 45(18):2897–2901, apr 2006.
[35] J. K. Nørskov, T. Bligaard, J. Rossmeisl, and C. H. Christensen. Towards
the computational design of solid catalysts.NatureChemistry, 1(1):37–46,
apr 2009.
[36] G. M. Clarke. Phyllosilicates as industrial minerals. In Health Related
Effects of Phyllosilicates, pages 31–45. Springer Berlin Heidelberg, 1990.
[37] David W Glasspoole. Large flake reconstitued mica insulation, September.
[38] Frank Lippert. An introduction to toothpaste - its purpose, history and
ingredients. In Monographs in Oral Science, pages 1–14. S. KARGER AG,
2013.
[39] S. Mahmoudi, A. Bennour, A. Meguebli, E. Srasra, and F. Zargouni. Char-
acterization and traditional ceramic application of clays from the douiret
region in south tunisia. Applied Clay Science, 127-128:78–87, jul 2016.
[40] Timothy D. Phillips, A. Bashir Sarr, and Patrick G. Grant. Selective
chemisorption and detoxification of aflatoxins by phyllosilicate clay. Nat-
ural Toxins, 3(4):204–213, 1995.
[41] M. Isabel Carretero and Manuel Pozo. Clay and non-clay minerals in the
pharmaceutical and cosmetic industries part II. active ingredients. Applied
Clay Science, 47(3-4):171–181, feb 2010.
[42] N. Roveri, G. Falini, E. Foresti, G. Fracasso, I.G. Lesci, and P. Sabatino.
Geoinspired synthetic chrysotile nanotubes. JournalofMaterialsResearch,
21(11):2711–2725, nov 2006.
100
[43] Handbook of Clay Science. Elsevier Science, 2011.
[44] Ju Seong Kim, Inchul Park, Eun-Suk Jeong, Kyoungsuk Jin, Won Mo
Seong, Gabin Yoon, Hyunah Kim, Byunghoon Kim, Ki Tae Nam, and Kisuk
Kang. Amorphous cobalt phyllosilicate with layered crystalline motifs as
water oxidation catalyst. AdvancedMaterials, 29(21):1606893, mar 2017.
[45] Qing Rong, Lu-Lu Long, Xing Zhang, Yu-Xi Huang, and Han-Qing Yu. Lay-
ered cobalt nickel silicate hollow spheres as a highly-stable supercapacitor
material. Applied Energy, 153:63–69, sep 2015.
[46] Yan Yang, Qingqin Liang, Jinghong Li, Yuan Zhuang, Yunhua He, Bo Bai,
and Xun Wang. Ni3si2o5(OH)4 multi-walled nanotubes with tunable
magnetic properties and their application as anode materials for lithium
batteries. Nano Research, 4(9):882–890, may 2011.
[47] Shiliang Zhou, Erica S. Howard, Jue Liu, Nicholas H. Bashian, Kyle
Nolan, Sankarganesh Krishnamoorthy, Geovanni M. Rangel, Moulay-
Tahar Sougrati, G. K. Surya Prakash, Katharine Page, and Brent C. Melot.
Hydrothermal preparation, crystal chemistry, and redox properties of iron
muscovite clay. ACS Applied Materials & Interfaces, 9(39):34024–34032,
sep 2017.
[48] Yuri Lvov, Wencai Wang, Liqun Zhang, and Rawil Fakhrullin. Halloysite
clay nanotubes for loading and sustained release of functional com-
pounds. Advanced Materials, 28(6):1227–1250, oct 2015.
[49] J. Theo Kloprogge. Synthesis of smectite clay minerals: A critical review.
Clays and Clay Minerals, 47(5):529–554, 1999.
[50] Y. F. Wang and H. F. Xu. Geochemical chaos: Periodic and nonperiodic
growth of mixed-layer phyllosilicates. Geochimica Et Cosmochimica Acta,
70(8):1995–2005, April 2006.
[51] Kazuo Urabe Yusuke Izumi and Makoto Onaka. Zeolite, Clay, and Het-
eropoly Acid in Organic Reactions. VCH Publishers Inc, 1992.
[52] E. N. Korytkova and L. N. Pivovarova. Hydrothermal synthesis of nan-
otubes based on (mg,fe,co,ni)3si2o5(OH)4 hydrosilicates. Glass Physics
and Chemistry, 36(1):53–60, feb 2010.
[53] Xiang Zhou, Dong Liu, Hongling Bu, Liangliang Deng, Hongmei Liu, Peng
Yuan, Peixin Du, and Hongzhe Song. XRD-based quantitative analysis of
clay minerals using reference intensity ratios, mineral intensity factors,
rietveld, and full pattern summation methods: A critical review. Solid
Earth Sciences, 3(1):16–29, mar 2018.
101
[54] Junjian Xu, Xuemei Li, Weiwei Zhou, Lei Ding, Zhong Jin, and Yan Li. The
preparation of mg3si2o5(OH)4 nanotubes under solvothermal conditions.
Journal of Porous Materials, 13(3-4):275–279, aug 2006.
[55] Bostjan Jancar and Danilo Suvorov. The influence of hydrothermal-
reaction parameters on the formation of chrysotile nanotubes. Nanotech-
nology, 17(1):25–29, nov 2005.
[56] Mireille Richard-Plouet, Serge Vilminot, and Murielle Guillot. Synthetic
transition metal phyllosilicates and organic-inorganic related phases. New
Journal of Chemistry, 28(9):1073, 2004.
[57] Allison McDonald, Bradley Scott, and Gilles Villemure. Hydrothermal
preparation of nanotubular particles of a 1:1 nickel phyllosilicate. Micro-
porous and Mesoporous Materials, 120(3):263–266, apr 2009.
[58] Xiaoyan Zhu, Zhichao Zhu, Xinrong Lei, and Chunjie Yan. Defects in struc-
ture as the sources of the surface charges of kaolinite. AppliedClayScience,
124-125:127–136, may 2016.
[59] Di Zhang, Chun-Hui Zhou, Chun-Xiang Lin, Dong-Shen Tong, and Wei-
Hua Yu. Synthesis of clay minerals. Applied Clay Science, 50(1):1–11, sep
2010.
[60] Kathleen A Carrado. Synthetic organo- and polymer–clays: preparation,
characterization, and materials applications. Applied Clay Science, 17(1-
2):1–23, jul 2000.
[61] A. Baumgartner, C. Butterhof, S. Koch, R. Mariychuk, and J. Breu. Melt
synthesis and characterization of synthetic Mn-rich tainiolite. Clays and
Clay Minerals, 57(2):271–277, April 2009.
[62] Giuseppe Falini, Elisabetta Foresti, Massimo Gazzano, Alessandro F.
Gualtieri, Matteo Leoni, Isidoro G. Lesci, and Norberto Roveri. Tubular-
shaped stoichiometric chrysotile nanocrystals. Chemistry - A European
Journal, 10(12):3043–3049, jun 2004.
[63] William C. Sheets, Emmanuelle Mugnier, Antoine Barnabé, Tobin J.
Marks, and Kenneth R. Poeppelmeier. Hydrothermal synthesis of
delafossite-type oxides. Chemistry of Materials, 18(1):7–20, jan 2006.
[64] Pehr Aake Bergstroem, Jan Lindgren, and Olof Kristiansson. An IR study
of the hydration of perchlorate, nitrate, iodide, bromide, chloride and
sulfate anions in aqueous solution. The Journal of Physical Chemistry,
95(22):8575–8580, oct 1991.
102
[65] Uhlig’s Corrosion Handbook. John Wiley & Sons, 2011.
[66] T. L. OConnor and S. A. Greenberg. The kinetics for the solution of silica in
aqueous solutions. The Journal of Physical Chemistry, 62(10):1195–1198,
oct 1958.
[67] W D Johns. Clay mineral catalysis and petroleum generation. Annual
Review of Earth and Planetary Sciences, 7(1):183–198, may 1979.
[68] Carmen Ciotonea, Brindusa Dragoi, Adrian Ungureanu, Alexandru
Chirieac, Sabine Petit, Sébastien Royer, and Emil Dumitriu. Nanosized
transition metals in controlled environments of phyllosilicate–mesoporous
silica composites as highly thermostable and active catalysts. Chemical
Communications, 49(69):7665, 2013.
[69] Ce Qiu, Jing Jiang, and Lunhong Ai. When layered nickel–cobalt sili-
cate hydroxide nanosheets meet carbon nanotubes: A synergetic coaxial
nanocable structure for enhanced electrocatalytic water oxidation. ACS
Applied Materials & Interfaces, 8(1):945–951, dec 2015.
[70] M.V. Sivaiah, S. Petit, M.F. Beaufort, D. Eyidi, J. Barrault, C. Batiot-
Dupeyrat, and S. Valange. Nickel based catalysts derived from hydrother-
mally synthesized 1:1 and 2:1 phyllosilicates as precursors for carbon
dioxide reforming of methane. Microporous and Mesoporous Materials,
140(1-3):69–80, apr 2011.
[71] M.V. Sivaiah, S. Petit, J. Barrault, C. Batiot-Dupeyrat, and S. Valange. CO2
reforming of CH4 over ni-containing phyllosilicates as catalyst precursors.
Catalysis Today, 157(1-4):397–403, nov 2010.
[72] Tao Wu, Weiye Cai, Peng Zhang, Xuefeng Song, and Lian Gao.
Cu–ni@SiO2 alloy nanocomposites for methane dry reforming catalysis.
RSC Advances, 3(46):23976, 2013.
[73] Renxi Jin, Yan Xing, Xiaodan Yu, Shaolin Sun, Donghui Yu, Fangfang
Wang, Wenbin Wu, and Shuyan Song. Facile synthesis of well-dispersed
silver nanoparticles on hierarchical flower-like ni3si2o5(OH)4with a high
catalytic activity towards 4-nitrophenol reduction. Chemistry - An Asian
Journal, 7(12):2955–2961, oct 2012.
[74] Mijong Kim, Ji Chan Park, Aram Kim, Kang Hyun Park, and Hyunjoon
Song. Porosity control of pd@SiO2 yolk–shell nanocatalysts by the forma-
tion of nickel phyllosilicate and its influence on suzuki coupling reactions.
Langmuir, 28(15):6441–6447, apr 2012.
103
[75] Meiling Wang, Xiaoqi Qian, Liqiang Xie, Huihuang Fang, Linmin Ye, Xin-
ping Duan, and Youzhu Yuan. Synthesis of a ni phyllosilicate with con-
trolled morphology for deep hydrogenation of polycyclic aromatic hydro-
carbons. ACS Sustainable Chemistry & Engineering, 7(2):1989–1997, dec
2018.
[76] Zhen Guo, Bin Liu, Qinghong Zhang, Weiping Deng, Ye Wang, and Yanhui
Yang. Recent advances in heterogeneous selective oxidation catalysis for
sustainable chemistry. Chemical Society Reviews, 43(10):3480, 2014.
[77] Pablo V. Tuza, Robinson L. Manfro, Nielson F.P. Ribeiro, and Mari-
ana M.V.M. Souza. Production of renewable hydrogen by aqueous-phase
reforming of glycerol over ni–cu catalysts derived from hydrotalcite pre-
cursors. Renewable Energy, 50:408–414, feb 2013.
[78] JÃ
ˇ
CÂ˝ urg Neuefeind, Mikhail Feygenson, John Carruth, Ron Hoffmann,
and Kenneth K. Chipley. The nanoscale ordered materials diffractome-
ter nomad at the spallation neutron source sns. Nuclear Instruments and
Methods in Physics Research Section B: Beam Interactions with Materials
and Atoms, 287:68–75, September 2012.
[79] R. W. Cheary and A. Coelho. A fundamental parameters approach to x-
ray line-profile fitting. JournalofAppliedCrystallography, 25(2):109–121,
1992.
[80] Stephens and Peter. Phenomenological model of anisotropic peak
broadening in powder diffraction. Journal of Applied Crystallography,
32(2):281–289, 1999.
[81] K. Momma and F. Izumi. VESTA: a three-dimensional visualization system
for electronic and structural analysis. J. Appl. Cryst., 41:653–658, 2008.
[82] Yi Shu. Fischer Porter Guide. Texas A&M, April 2008.
[83] G. J. Redhammer, G. Amthauer, W. Lottermoser, M. Bernroider, G. Tippelt,
and G. Roth. X-ray powder diffraction and
57
fe-mossbauer spectroscopy of
synthetic trioctahedral micas K[Me
3
](TSi
3
)O
10
(OH)
2
),Me = Ni
2+
, Mg
2+
,
Co
2+
, Fe
2+
; T = Al
3+
, Fe
3+
. Mineralogy and Petrology, 85(1-2):89–115,
November 2005.
[84] S. B. Hendricks and M. E. Jefferson. Polymorphism of the micas with
optical measurements. Amer. Min., 24:729, 1939.
[85] J. v. Smith and H. S. Yoder. Experimental and theoretical studies of the
mica polymorphs. Mineral Mag., 31:209, 1956.
104
[86] M. Ross, H. Takeda, and D. R. Wones. Mica polytypes - systematic descrip-
tion and identification. Science, 151(3707):191, 1966.
[87] H. f. Xu and D. R. Veblen. Periodic and nonperiodic stacking in biotite
from the bingham canyon porphyry copper-deposit, utah. Clays and Clay
Minerals, 43(2):159–173, April 1995.
[88] Armel Le Bail. Whole powder pattern decomposition methods and appli-
cations: A retrospection. 20(4):316–326, 2005.
[89] Rebecca W. Smaha, John H. Roudebush, Jake T. Herb, Elizabeth M. Seibel,
Jason W. Krizan, Gary M. Fox, Qingzhen Huang, Craig B. Arnold, and
Robert J. Cava. Tuning sodium ion conductivity in the layered hon-
eycomb oxide Na
3−x
Sn
2−x
Sb
x
NaO
6
. Inorg. Chem., 54(16):7985–7991,
August 2015.
[90] Jue Liu, Liang Yin, Lijun Wu, Jianming Bai, Seong-Min Bak, Xiqian Yu,
Yimei Zhu, Xiao-Qing Yang, and Peter G. Khalifah. Quantification of hon-
eycomb number-type stacking faults: Application to na3ni2bio6 cathodes
for na-ion batteries. Inorg. Chem., 55(17):8478–8492, September 2016.
[91] C. R. Houska and B. E. Warren. X-ray study of the graphitization of carbon
black. Journal of Applied Physics, 25(12):1503–1509, December 1954.
[92] Huarui Xu and Lian Gao. New evidence of a dissolution-precipitation
mechanism in hydrothermal synthesis of barium titanate powders. Mate-
rials Letters, 57(2):490–494, December 2002.
[93] Xue Qin, Xiaohui Wang, Huimin Xiang, Jie Xie, Jingjing Li, and Yanchun
Zhou. Mechanism for hydrothermal synthesis of LiFePO
4
platelets as cath-
ode material for lithium-ion batteries. J. Phys. Chem. C, 114(39):16806–
16812, October 2010.
[94] Y. J. Zhang and J. J. Lu. A mild and efficient biomimetic synthe-
sis of rodlike hydroxyapatite particles with a high aspect ratio using
polyvinylpyrrolidone as capping agent. Crystal Growth & Design,
8(7):2101–2107, July 2008.
[95] Paul McMorn, Glyn Roberts, and Graham J. Hutchings. Catalysis Letters,
63(3/4):193–197, 1999.
[96] Artit Ausavasukhi and Tawan Sooknoi. Oxidation of tetrahydrofuran
to butyrolactone catalyzed by iron-containing clay. Green Chemistry,
17(1):435–441, 2015.
105
[97] M. L. B. Rao, A. Damjanovic, and J. O. M. Bockris. Oxygen adsorption
related to the unpaired d-electrons in transition metals. The Journal of
Physical Chemistry, 67(11):2508–2509, nov 1963.
[98] Sara Goldstein, Dan Meyerstein, and Gidon Czapski. The fenton reagents.
Free Radical Biology and Medicine, 15(4):435–445, oct 1993.
[99] Barry R. Bickmore, Dirk Bosbach, Michael F. Hochella, Laurent Charlet,
and Eric Rufe. In situ atomic force microscopy study of hectorite and non-
tronite dissolution: Implications for phyllosilicate edge surface structures
and dissolution mechanisms. American Mineralogist, 86(4):411–423, apr
2001.
[100] Barry R. Bickmore, Kevin M. Rosso, Kathryn L. Nagy, Randall T. Cygan,
and Christopher J. Tadanier. Determination of edge surface structures for
dioctahedral 2:1 phyllosilicates: implications for acid-base for reactivity.
Clays and Clay Minerals, 51(4):359–371, aug 2003.
[101] Fabrizio Cavani and Joaquim Henrique Teles. Sustainability in catalytic
oxidation: An alternative approach or a structural evolution? Chem-
SusChem, 2(6):508–534, jun 2009.
[102] Lydie Le Forestier, Fabrice Muller, Villieras, and Manuel Pelletier. Textural
and hydration properties of a synthetic montmorillonite compared with a
natural na-exchanged clay analogue. AppliedClayScience, 48(1-2):18–25,
mar 2010.
[103] Peter J. Chupas, Xiangyun Qiu, Jonathan C. Hanson, Peter L. Lee, Clare P.
Grey, and Simon J. L. Billinge. Rapid acquisition pair distribution function
(ra-pdf) analysis.
[104] A. P. Hammersley, S. O. Svensson, M. Hanfland, A. N. Fitch, and
D. Hausermann. Two-dimensional detector software: From real detec-
tor to idealised image or two-theta scan. High Pressure Research, 14(4-
6):235–248, jan 1996.
[105] Xiangyun Qiu, Jeroen W. Thompson, and Simon J. L. Billinge. PDFgetX2:
a GUI-driven program to obtain the pair distribution function from x-ray
powder diffraction data. Journal of Applied Crystallography, 37(4):678–
678, jul 2004.
[106] C L Farrow, P Juhas, J W Liu, D Bryndin, E S Božin, J Bloch, Th Prof-
fen, and S J L Billinge. PDFfit2 and PDFgui: computer programs for
studying nanostructure in crystals. Journal of Physics: Condensed Matter,
19(33):335219, jul 2007.
106
[107] R. D. Shannon and C. T. Prewitt. Effective ionic radii in oxides and flu-
orides. Acta Crystallographica Section B Structural Crystallography and
Crystal Chemistry, 25(5):925–946, may 1969.
[108] James R. Neilson, Birgit Schwenzer, Ram Seshadri, and Daniel E.
Morse. Kinetic control of intralayer cobalt coordination in layered
hydroxides: Co1-0.5xoctcoxtet(OH)2(cl)x(h2o)n. Inorganic Chemistry,
48(23):11017–11023, dec 2009.
[109] J. V. Smith and S. W. Bailey. Second review of al–o and si–o tetrahedral
distances. Acta Crystallographica, 16(8):801–811, aug 1963.
[110] M. OKeeffe and B. G. Hyde. On si—o—si configurations in silicates. Acta
Crystallographica Section B, 34(1):27–32, jan 1978.
[111] I. D. Brown. Predicting bond lengths in inorganic crystals. Acta Crys-
tallographica Section B Structural Crystallography and Crystal Chemistry,
33(5):1305–1310, may 1977.
[112] J. B. Jones. Al–o and si–o tetrahedral distances in aluminosilicate frame-
work structures. Acta Crystallographica Section B Structural Crystallogra-
phy and Crystal Chemistry, 24(3):355–358, mar 1968.
[113] Martin T. Dove, David A. Keen, Alex C. Hannon, and Ian P. Swainson.
Direct measurement of the si-o bond length and orientational disorder
in the high-temperature phase of cristobalite. Physics and Chemistry of
Minerals, 24(4):311–317, may 1997.
[114] Jue Liu, Daniel Olds, Rui Peng, Lei Yu, Guo Shiou Foo, Shuo Qian, Jong
Keum, Beth S. Guiton, Zili Wu, and Katharine Page. Quantitative analysis
of the morphology of{101} and{001} faceted anatase TiO2 nanocrys-
tals and its implication on photocatalytic activity. Chemistry of Materials,
29(13):5591–5604, jun 2017.
[115] Guang Wu, Yan Gao, Fangwei Ma, Binghui Zheng, Liguo Liu, Hongyan
Sun, and Wei Wu. Catalytic oxidation of benzyl alcohol over man-
ganese oxide supported on MCM-41 zeolite. Chemical Engineering Jour-
nal, 271:14–22, jul 2015.
[116] Meenakshisundaram Sankar, Ewa Nowicka, Emma Carter, Damien M.
Murphy, David W. Knight, Donald Bethell, and Graham J. Hutchings. The
benzaldehyde oxidation paradox explained by the interception of peroxy
radical by benzyl alcohol. Nature Communications, 5(1), feb 2014.
107
[117] Gang Li, Dan I. Enache, Jennifer Edwards, Albert F. Carley, David W.
Knight, and Graham J. Hutchings. Solvent-free oxidation of benzyl alco-
hol with oxygen using zeolite-supported au and au–pd catalysts. Catalysis
Letters, 110(1-2):7–13, aug 2006.
[118] Dapeng Guo, Yan Wang, Peng Zhao, Meifen Bai, Hui Xin, Zhi Guo, and
Jingyi Li. Selective aerobic oxidation of benzyl alcohol driven by visible
light on gold nanoparticles supported on hydrotalcite modified by nickel
ion. Catalysts, 6(5):64, apr 2016.
[119] G. Kresse and J. FurthmÃijller. Efficient iterative schemes forab initiototal-
energy calculations using a plane-wave basis set. Physical Review B,
54(16):11169–11186, oct 1996.
[120] P. E. BlÃ˝ uchl. Projector augmented-wave method. Physical Review B,
50(24):17953–17979, dec 1994.
[121] John P. Perdew, Adrienn Ruzsinszky, Gábor I. Csonka, Oleg A. Vydrov, Gus-
tavo E. Scuseria, Lucian A. Constantin, Xiaolan Zhou, and Kieron Burke.
Restoring the density-gradient expansion for exchange in solids and sur-
faces. Physical Review Letters, 100(13), apr 2008.
[122] Stefan Grimme, Jens Antony, Stephan Ehrlich, and Helge Krieg. A consis-
tent and accurate ab initio parametrization of density functional disper-
sion correction (DFT-d) for the 94 elements h-pu. TheJournalofChemical
Physics, 132(15):154104, apr 2010.
[123] S. L. Dudarev, G. A. Botton, S. Y. Savrasov, C. J. Humphreys, and A. P.
Sutton. Electron-energy-loss spectra and the structural stability of nickel
oxide. Physical Review B, 57(3):1505–1509, jan 1998.
[124] O. Bengone, M. Alouani, J. Hugel, and P. BlÃ˝ uchl. calculated electronic
and structural properties of NiO(001) and NiO(111) surfaces. Computa-
tional Materials Science, 24(1-2):192–198, may 2002.
[125] G. Makov and M. C. Payne. Periodic boundary conditions inab initiocalcu-
lations. Physical Review B, 51(7):4014–4022, feb 1995.
[126] Juana Moreno and José M. Soler. Optimal meshes for integrals in real-
and reciprocal-space unit cells. Physical Review B, 45(24):13891–13898,
jun 1992.
[127] Keith T. Butler, John Buckeridge, C. Richard A. Catlow, and Aron Walsh.
Crystal electron binding energy and surface work function control of tin
dioxide. Physical Review B, 89(11), mar 2014.
108
[128] Keith T. Butler, Rachel Crespo-Otero, John Buckeridge, David O. Scan-
lon, Edward Bovill, David Lidzey, and Aron Walsh. Band energy con-
trol of molybdenum oxide by surface hydration. Applied Physics Letters,
107(23):231605, dec 2015.
[129] Matthew J. Wahila, Keith T. Butler, Zachary W. Lebens-Higgins, Christo-
pher H. Hendon, Abhishek S. Nandur, Robert E. Treharne, Nicholas F.
Quackenbush, Shawn Sallis, Katie Mason, Hanjong Paik, Darrell G.
Schlom, Joseph C. Woicik, Jinghua Guo, Dario A. Arena, Bruce E. White,
Graeme W. Watson, Aron Walsh, and Louis F. J. Piper. Lone-pair stabiliza-
tion in transparent amorphous tin oxides: A potential route to p-type con-
duction pathways. Chemistry of Materials, 28(13):4706–4713, jun 2016.
[130] Mikhail Vinaykin and Alexander V. Benderskii. Vibrational sum-frequency
spectrum of the water bend at the air/water interface. The Journal of
Physical Chemistry Letters, 3(22):3348–3352, nov 2012.
[131] Fabien Baron and Sabine Petit. Interpretation of the infrared spectra of
the lizardite-nepouite series in the near- and mid-infrared range.American
Mineralogist, 101(2):423–430, feb 2016.
[132] H. J. Bakker and J. L. Skinner. Vibrational spectroscopy as a probe of
structure and dynamics in liquid water. Chemical Reviews, 110(3):1498–
1517, mar 2010.
[133] R SHALVOY. Characterization of coprecipitated nickel on silica methana-
tion catalysts by x-ray photoelectron spectroscopy. Journal of Catalysis,
56(3):336–348, mar 1979.
[134] Thabang Ntho, James Aluha, Pumeza Gqogqa, Mpfunzeni Raphulu, and
Gary Pattrick. Au-al2o3 catalysts for glycerol oxidation: the effect of
support acidity and gold particle size. Reaction Kinetics, Mechanisms and
Catalysis, 109(1):133–148, feb 2013.
[135] Zhiyao Lu, Ivan Demianets, Rasha Hamze, Nicholas J. Terrile, and Travis J.
Williams. A prolific catalyst for selective conversion of neat glycerol to
lactic acid. ACS Catalysis, 6(3):2014–2017, feb 2016.
[136] Douglas W. Stephan. Frustrated lewis pairs: From concept to catalysis.
Accounts of Chemical Research, 48(2):306–316, dec 2014.
[137] Douglas W. Stephan and Gerhard Erker. Frustrated lewis pair chemistry:
Development and perspectives. Angewandte Chemie International Edition,
54(22):6400–6441, may 2015.
109
[138] D. W. Stephan. The broadening reach of frustrated lewis pair chemistry.
Science, 354(6317):aaf7229–aaf7229, dec 2016.
[139] Andrew R. Jupp and Douglas W. Stephan. New directions for frustrated
lewis pair chemistry. Trends in Chemistry, 1(1):35–48, apr 2019.
[140] Yuanyuan Ma, Sai Zhang, Chun-Ran Chang, Zheng-Qing Huang,
Johnny C. Ho, and Yongquan Qu. Semi-solid and solid frustrated lewis
pair catalysts. Chemical Society Reviews, 47(15):5541–5553, 2018.
[141] Sai Zhang, Zheng-Qing Huang, Yuanyuan Ma, Wei Gao, Jing Li, Fangxian
Cao, Lin Li, Chun-Ran Chang, and Yongquan Qu. Solid frustrated-lewis-
pair catalysts constructed by regulations on surface defects of porous
nanorods of CeO2. Nature Communications, 8(1), may 2017.
[142] Kulbir Kaur Ghuman, Laura B. Hoch, Thomas E. Wood, Charles Mims,
Chandra Veer Singh, and Geoffrey A. Ozin. Surface analogues of molec-
ular frustrated lewis pairs in heterogeneous CO2 hydrogenation catalysis.
ACS Catalysis, 6(9):5764–5770, aug 2016.
[143] Stephanie R. Flynn, Owen J. Metters, Ian Manners, and Duncan F.
Wass. Zirconium-catalyzed imine hydrogenation via a frustrated lewis
pair mechanism. Organometallics, 35(6):847–850, mar 2016.
[144] Kulbir Kaur Ghuman, Thomas E. Wood, Laura B. Hoch, Charles A. Mims,
Geoffrey A. Ozin, and Chandra Veer Singh. Illuminating CO2 reduction on
frustrated lewis pair surfaces: investigating the role of surface hydroxides
and oxygen vacancies on nanocrystalline in2o3-x(OH)y. Physical Chem-
istry Chemical Physics, 17(22):14623–14635, 2015.
[145] Shoji Higashi, Kazuhiko Miki, and Sridhar Komarneni. Hydrothermal syn-
thesis OF zn-smectites.ClaysandClayMinerals, 50(3):299–305, jun 2002.
[146] Yuanzhi Tang, Helen F. Chappell, Martin T. Dove, Richard J. Reeder,
and Young J. Lee. Zinc incorporation into hydroxylapatite. Biomaterials,
30(15):2864–2872, may 2009.
[147] Neil M. Schweitzer, Bo Hu, Ujjal Das, Hacksung Kim, Jeffrey Greeley,
Larry A. Curtiss, Peter C. Stair, Jeffrey T. Miller, and Adam S. Hock. Propy-
lene hydrogenation and propane dehydrogenation by a single-site zn2 on
silica catalyst. ACS Catalysis, 4(4):1091–1098, mar 2014.
[148] Jeffrey Camacho-Bunquin, Payoli Aich, Magali Ferrandon, Andrew “Bean”
Getsoian, Ujjal Das, Fulya Dogan, Larry A. Curtiss, Jeffrey T. Miller,
Christopher L. Marshall, Adam S. Hock, and Peter C. Stair. Single-site
110
zinc on silica catalysts for propylene hydrogenation and propane dehydro-
genation: Synthesis and reactivity evaluation using an integrated atomic
layer deposition-catalysis instrument. Journal of Catalysis, 345:170–182,
jan 2017.
[149] H. Berndt, G. Lietz, B. LÃijcke, and J. VÃ˝ ulter. Zinc promoted h-ZSM-
5 catalysts for conversion of propane to aromatics i. acidity and activity.
Applied Catalysis A: General, 146(2):351–363, nov 1996.
[150] Makoto Onaka and Ryota Yamasaki. Catalysis by solid acid: Selec-
tive diels-alder reactions are promoted by mesoporous aluminosilicate
with zinc ions and nanometer honeycomb openings. Chemistry Letters,
27(3):259–260, mar 1998.
[151] Abderrahim Solhy, James H. Clark, Rachid Tahir, Said Sebti, and
Mohamed Larzek. Transesterifications catalysed by solid, reusable
apatite–zinc chloride catalysts. Green Chem., 8(10):871–874, 2006.
111

Abstract (if available)

Abstract The rational design of catalysts was declared a Holy Grail of chemistry research by the 1995 Accounts of Chemical Research articles by Breslow, [1] and Bard and Fox.[2] Pursuant to this goal, the fundamental understanding of activity in heterogeneous catalysts is an outstanding problem in materials chemistry. Though heterogeneous catalysts are utilized in more than 85% of industrial reactions and account for 25% of industrial energy use, there is a lack of basic guidelines for rational catalyst design and synthesis. [3] This dissertation addresses a range of topics chosen to develop of structure-property relationships using theory and experimental data to understand adsorption, reactions, and desorption processes. ❧ To inform the understanding earth-abundant catalysts, we begin by examining 2:1 and 1:1 phyllosilicates as model frameworks. Their unique layered structure represents one of the best opportunities for facilitating correlations between local structure, composition, and catalytic activity on a well-defined surface. Studied first, 2:1 phyllosilicates, display a highly tunable synthetic system determined from a combination of x-ray and neutron diffraction, SEM, and Mössbauer spectroscopy. Despite promising characterization, these materials are poor oxidation catalysts as a result of limited access to the surface-active site. Following this, our investigation of several 1:1 phyllosilicates found only Ni₃Si₂O₅(OH)₄ is active for the oxidation of benzyl alcohol. To our knowledge, this is the first demonstration of activation of molecular oxygen by a 1:1 phyllosilicate. ❧ Guided by previously reported surface dehydration mechanisms for these compounds, we developed heat treatment techniques to tune the surface active site. [4, 5, 6, 7, 8, 9] Next, we present the thermal treatment Ni₃Si₂O₅(OH)₄ and characterization of new catalytic sites formed on its surface, which are capable of oxidizing glycerol. Through local structure characterization including pair distribution function analysis (PDF), Fourier transform spectroscopy, x-ray photoelectron spectroscopy, DFT calculation and magnetic measurement, the effect of anionic modification is elucidated. Specifically, the local structure determined from PDF in combination with computational studies of these surfaces is used to ascertain the role of surface coordination and oxidation state. ❧ Replacement of aluminum by a transition metal allowed us to develop and tune oxidation catalysis of these materials. The final study presented here investigates both bimetallic substitution and thermal activation in series Ni₃₋ₓMgₓSi₂O₅(OH)₄ and Ni₃₋ₓZnₓSi₂O₅(OH)₄. These phases introduce the opportunity to study cooperative catalytic synergy and a route to tailor the activity by systematically altering the composition. Through reduction of cyclohexene we demonstrate heat treatment creates frustrated Lewis pair-like active sites on the surfaces capable of binding and dissociating H₂. This understanding from these collective studies can be used as guiding design principles to synthesize catalysts for biomass transform, selective oxidation, and hydrogenation chemistry.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Vibrational spectroscopy and molecular orientation at interfaces: correlating the structure of a material's surface to its properties

PDF

Understanding the structure-property relationship in electrode materials for electrochemical energy storage

PDF

Plasmonic enhancement of catalysis and solar energy conversion

PDF

Improving the sustainability of conjugated polymer synthesis via direct arylation polymerization

PDF

Design of durable and efficient catalysts for the electro-oxidation of methanol

PDF

The effects of crystal structure and composition on the magnetism of garnets & phyllosilicates

PDF

Hydrogen transfer reactions catalyzed by iridium and ruthenium complexes

PDF

New bifunctional catalysts for ammonia-borane dehydrogenation, nitrile reduction, formic acid dehydrogenation, lactic acid synthesis, and carbon dioxide reduction

PDF

Investigations in cooperative catalysis: synthesis, reactivity and metal-ligand bonding

PDF

Synthesis, characterization and reaction chemistry of polyazides and cyanometallates

PDF

Hydrogen energy system production and storage via iridium-based catalysts

PDF

Design and modification of electrocatalysts for use in fuel cells and CO₂ reduction

PDF

Understanding the mechanism of oxygen reduction and oxygen evolution on transition metal oxide electrocatalysts and applications in iron-air rechargeable battery

PDF

Molecular approaches to solve the blue problem in organic light emitting diode display and lighting applications

PDF

Novel methods for functional group interconversions in organic synthesis and structural characterization of new transition metal heterogeneous catalysts for potential carbon neutral hydrogen storage

PDF

Solution processing of chalcogenide functional materials using thiol–amine “alkahest” solvent systems

PDF

Improving the efficiency and sustainability of direct arylation polymerization for conjugated polymer synthesis

PDF

Design of nanomaterials for electrochemical applications in fuel cells and beyond

PDF

Atomistic modeling of the mechanical properties of metallic glasses, nanoglasses, and their composites

PDF

Understanding the role of electrode design in determining the electrochemical performance of high-energy/high-power lithium-ion and lithium-sulfur batteries

Asset Metadata

Creator Howard, Erica Schell (author)

Core Title Exploring new frontiers in catalysis: correlating crystal chemistry and activity in layered silicates

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 08/06/2019

Defense Date 06/21/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag catalysis,green energy,heterogeneous catalysis,materials,OAI-PMH Harvest,solid catalyst,solid-state chemistry,structure-property relationship,sustainable materials

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Melot, Brent C. (committee chair), Gupta, Malancha (committee member), Thompson, Mark E. (committee member)

Creator Email eshoward@usc.edu,rykarykarykawot@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-209258

Unique identifier UC11663552

Identifier etd-HowardEric-7749.pdf (filename),usctheses-c89-209258 (legacy record id)

Legacy Identifier etd-HowardEric-7749.pdf

Dmrecord 209258

Document Type Dissertation

Format application/pdf (imt)

Rights Howard, Erica Schell

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA