close

Вход

Забыли?

вход по аккаунту

?

Characterization of Local Structures in Layered Niobates Using Solid-State NMR and Rapid Microwave-Assisted Synthesis and Exfoliation of Their Alcohol Grafted Derivatives

код для вставкиСкачать
Characterization of Local Structures in Layered Niobates
Using Solid-State NMR and Rapid Microwave-Assisted
Synthesis and Exfoliation of Their Alcohol Grafted
Derivatives
JOSHUA ROSS BOYKIN
DECEMBER 2017
A DISSERTATION
Submitted to the faculty of Clark University,
Worcester, Massachusetts,
in partial fulfillment of
the requirements for
the degree of Doctor of Philosophy
in the Department of Chemistry
ProQuest Number: 10683811
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
ProQuest 10683811
Published by ProQuest LLC (2018 ). Copyright of the Dissertation is held by the Author.
All rights reserved.
This work is protected against unauthorized copying under Title 17, United States Code
Microform Edition © ProQuest LLC.
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, MI 48106 - 1346
DISSERTATION COMMITTEE
Luis Smith, Ph.D.
Chief Instructor
Mark Turnbull, Ph.D.
Committee Member
Noel Lazo, Ph.D.
Committee Member
ABSTRACT
The research described in this dissertation spans nearly six years and can be
broken into three heavily related research projects:
1. Exploration of local structural changes in mixed alkali layered perovskites.
2. Determination of B-site preference in mixed Nb/Ta systems using solid-state
NMR
3. Microwave assisted grafting and exfoliation of layered perovskites using
long chain alcohols.
Exploration of local structural changes in mixed alkali layered perovskites.
A series of three-layered Dion-Jacobson (D-J) RbSrxCa2-xNb3O10 (0 < x < 2)
and RbBaxSr2-xNb3O10 (0 < x < 0.8) compounds were synthesized using microwave
and molten salt synthetic routes, respectively, and analyzed using powder X-Ray
diffraction (XRD), scanning electron microscopy (SEM), energy dispersive
spectroscopy (EDS), attenuated total reflectance-infrared spectroscopy (ATR-IR),
and wideline uniform rate smooth truncation-quadrupolar Carr-Purcell-MeiboomGill (WURST-QCPMG) nuclear magnetic resonance (NMR).
An unexpected
dichotomy regarding cation distribution was observed with distribution considered
homogeneous on a long-range scale and heterogeneous on a local scale.
Determination of B-site preference in mixed Nb/Ta systems using solid-state NMR
D-J phase layered perovskites can readily undergo proton exchange under
mild conditions leading to the formation of a solid acid capable of intercalating
various organic molecules. The B-site atom in this class of compounds is often
Ti4+, Nb5+, or Ta5+.
It has been shown that changing the B-site atom can
drastically alter the intercalation behavior and catalytic ability of these materials.
While Nb5+ and Ta5+ possess the same charge and approximately equal ionic radii
(0.64 Ǻ), Nb is more electronegative than Ta (1.6 vs. 1.5 Pauling Units,
respectively) leading to Nb-O bonds being more covalent in character than Ta-O
bonds and changes in the energy states of Ta5+ and Nb5+ compounds. As this
work shows, in mixed Nb5+/Ta5+ systems, electronegativity is a strong driver of
site ordering in D-J niobates.
Microwave assisted grafting and exfoliation of layered perovskites using long chain
alcohols.
Current grafting methods of layered perovskites involve high pressure
heating at 80 to 150 oC for days to weeks, depending on the alcohol involved. The
long reaction time involved inhibits the ability to study grafted samples in a timely
fashion. This work will present a new microwave assisted grafting technique which
decreases reaction time from 17 days to 12 hours. Furthermore, a microwavesonication based method of exfoliation of grafted niobates in organic solvents was
developed, further expanding the capabilities and potential of niobate nanosheets.
© 2017
JOSHUA ROSS BOYKIN
ALL RIGHTS RESERVED
Academic History
Name (in Full): Joshua Ross Boykin
Date: AUG 2017
Baccalaureate Degree: Bachelor of Science (Chemistry)
Source: Marlboro College, Marlboro, VT
Date: May, 2008
Occupation and Academic Connection since date of baccalaureate
degree:
Organic Chemistry Laboratory Instructor, Clark University: 2016-2017
Introductory Chemistry Instructor, Clark University: 2016
Teaching Assistant, Clark University: 2010-2016
DEDICATION
To the steadfast rejection of ignoramus et ignorabimus and all it
embodies.
Also, to my friends and family who will never read this dissertation, but
will still tell me it’s great.
vii
ACKNOWLEDGEMENTS
I have spent the past seven years completing my Ph.D. at Clark University.
I would like to extend my sincerest gratitude to everyone in the chemistry
department for helping me to reach this point. There were many times I became
unsure of myself and seriously doubted whether I could accomplish this feat, and
it was only through the strength of those around me that I find myself at the end
of this long journey. To name every person and every act that has lead me here
would result in a tome that would likely rival in length this dissertation, but I would
like to take a little time to give credit to a few.
First, I would like to thank my advisor Professor Luis Smith. When I started
at Clark I knew nothing beyond the very basics of solid-state NMR, X-ray
diffraction, or inorganic materials. He has shown great patience as I explored the
worlds of spectroscopy and perovskites, giving me guidance, encouragement, and
when need be criticism. He has served as an excellent example during my time
at Clark and I thank him for it. He has been the best mentor I could have asked
for.
Second, I would like to thank my other committee members, Professors
Mark Turnbull and Noel Lazo. During my very first semester at Clark I took courses
from both and it was through them that I came to realize the hard work that would
be required of me during graduate school. Both have helped me greatly since
then by asking me difficult questions about my work that have forced me to
consider different perspectives that I would not have otherwise considered.
Thirdly, I would like to thank Dr. Lin, Clark’s NMR manager. Reading about
new pulse sequences or doing theoretical calculations don’t mean much if you
can’t operate the instrument. He has helped me learn the hardware and software
considerations that are not discussed, but nonetheless necessary for success.
Last, but by no stretch of the imagination least, I would like to thank my
family for their love and support. Since I was a child my parents always
encouraged me in my desire to be a scientist and at my lowest points were there
to help me rise. My lovely wife, Jodi, has made me happier than I ever thought I
could be, and I know without a shadow of a doubt that I wouldn’t be here without
her. To those I have mentioned, and the many more unmentioned, I am infinitely
indebted.
viii
Table of Contents
List of Tables ............................................................................................................................. xii
List of Figures ........................................................................................................................... xv
List of Abbreviations ........................................................................................................... xxxi
Chapter 1: Background of Layered Perovskites ......................................................................... 1
1.1 Introduction ............................................................................................................................ 1
1.2 RbCaxSr2-xNb3O10 and RbCa2NaNb4-xTaxO13 Crystal Structures ........................................... 6
1.3 Soft-chemical Exchange Reactions in Layered Perovskites .................................................. 9
1.4 Niobate Nanosheets ............................................................................................................. 11
1.5 Research Presented Herein .................................................................................................. 14
1.5.1 Exploration of local structural changes in mixed alkali layered perovskites. ................... 14
1.5.2 Determination of B-site preference in mixed Nb/Ta systems using solid-state NMR ...... 15
1.5.3 Microwave assisted grafting and exfoliation of layered perovskites using long chain
alcohols. ..................................................................................................................................... 16
Chapter 2: NMR Background .................................................................................................... 18
2.1 Solid-State NMR.................................................................................................................. 18
2.2 Quadrupolar Nuclei .............................................................................................................. 20
2.3 93Nb Quadrupolar Interactions ............................................................................................. 30
2.4 Variable Offset Cumulative Spectra (VOCS) ...................................................................... 31
2.5 Magic Angle Spinning ......................................................................................................... 33
2.6 Wideline Uniform Rate Smooth Truncations (WURST) Pulses.......................................... 36
2.7 Multiple-Quantum Magic Angle Spinning (MQMAS) ....................................................... 42
Chapter 3: Experimental Overview ........................................................................................... 49
3.1 Materials .............................................................................................................................. 49
3.2 Synthetic Procedures*.......................................................................................................... 49
3.2.1 Three-layer Niobates Using MSS .................................................................................. 49
3.2.2 Three-layer Niobates Using Microwave Heating ......................................................... 50
3.2.3 Alkyl Grafting with Conventional Heating ................................................................... 50
3.2.4 Alkyl Grafting with Microwave Heating ....................................................................... 51
3.2.5 RbCa2NaNb4-xTaxO13 and HCa2NaNb4-xTaxO13 ............................................................. 52
3.2.6 Nanosheet synthesis using TBAOH ............................................................................... 53
3.2.7 Nanosheet Synthesis Using Microwave Assisted Grafting............................................ 53
3.3 Characterization Methods .................................................................................................... 54
ix
3.3.1 Powder XRD ................................................................................................................. 54
3.3.2 ATR-IR .......................................................................................................................... 54
3.3.3 SEM and EDX ............................................................................................................... 55
3.3.4 UV-visible Diffuse Reflectance ..................................................................................... 55
3.3.5 Solid-State NMR Experiments ....................................................................................... 55
Chapter 4. Exploration of Local Structural Changes in Layered Niobates due to
Compositional Changes at the A-Site. ........................................................................................ 58
4.1 Introduction .......................................................................................................................... 58
4.2 Experimental ........................................................................................................................ 59
4.3 XRD Results and Discussion ............................................................................................... 61
4.4 EDS and SEM Results and Discussion ................................................................................ 72
4.5 ATR-IR Results and Discussion .......................................................................................... 75
4.6 NMR Results and Discussion .............................................................................................. 77
4.7 Conclusions ........................................................................................................................ 109
Chapter 5: Determination of Cation Site Ordering in Four-layer HCa2NaNb4-xTaxO13
Compounds Using Solid State 93Nb and 23Na NMR ................................................................ 111
5.1 Introduction ........................................................................................................................ 111
5.3 Results and Discussion ...................................................................................................... 115
5.3.1 XRD, SEM, and Elemental Analysis ........................................................................... 115
5.3.2 93Nb NMR .................................................................................................................... 119
5.3.3 23Na MAS NMR ........................................................................................................... 131
5.4 Conclusions ........................................................................................................................ 135
Chapter 6: Microwave Assisted Grafting of Mixed Cation HCaxSr2-xNb3O10
Compounds with n-Alcohols* ................................................................................................... 136
6.1 Introduction ........................................................................................................................ 136
6.2 Experimental ...................................................................................................................... 137
6.3 XRD Results and Discussion ............................................................................................. 139
6.4 TGA and 1H NMR Results and Discussion ....................................................................... 145
6.5 Grafting of Mixed Ca/Sr Compounds ................................................................................ 147
6.6 UV-visible Results and Discussion .................................................................................... 152
6.7 Conclusions ........................................................................................................................ 155
Chapter 7: Exfoliation of Grafted Niobates Using Organic Solvents................................... 156
7.1 Introduction ........................................................................................................................ 156
x
7.2 Experimental ...................................................................................................................... 158
7.2.1 Sample Overview ........................................................................................................ 158
7.2.2 Solvent Choice ............................................................................................................ 161
7.2.4 Sonication ................................................................................................................... 162
7.2.5 Sedimentation.............................................................................................................. 162
7.2.6 Extracting Nanosheets Out of Solution ....................................................................... 163
7.3 Visual Observations and Generation of Colored Suspensions........................................... 163
7.4 XRD Results ...................................................................................................................... 173
7.2.3 Heating........................................................................................................................ 174
7.4.1 Sediments of Suspensions ............................................................................................ 175
7.4.2 Dried Suspensions ....................................................................................................... 178
7.4.3 Solid From SN ............................................................................................................. 183
7.5 SEM and EDS Results ....................................................................................................... 187
7.6 TGA Results and Exfoliation Efficiency ........................................................................... 193
7.7 Discussion .......................................................................................................................... 194
7.7.1 Role of Solvent and Parent Compound ....................................................................... 194
7.7.2 Role of Sonication ....................................................................................................... 195
7.7.3 Role of Microwave Heating ........................................................................................ 195
7.7.4 Treatment of Supernatant ........................................................................................... 196
7.7.5 Fate of the Grafted Alkyl Groups ................................................................................ 196
7.8 Conclusions ........................................................................................................................ 198
Chapter 8: Summary and Future Directions .......................................................................... 200
References ................................................................................................................................... 204
Appendix A: Indexed Lists of XRD Pattern hkl Reflections.................................................. 219
Appendix B: C1-6/CaxSr2-xNb3O10 TGA Data ......................................................................... 233
xi
List of Tables
Table 1.1. Tetragonal unit cell parameters of RbCaxSr2-xNb3O10 phases collected by
Geselbracht et al.31 ................................................................................................... 7
Table 1.2 Nb-O bond lengths in RbSr2Nb3O10. .......................................................................... 8
Table 3.1. General grafting procedure for CaxSr2-xNb3O10 compounds. .................................. 52
Table 4.1. Synthetic conditions for RbCaxSr2-xNb3O10 compounds using dry microwave
heating or MSS. ..................................................................................................... 60
Table 4.2. Synthetic conditions for RbBaxSr2-xNb3O10 compounds using dry microwave
synthesis................................................................................................................. 61
Table 4.3. Lattice parameters for RbCaxSr2-xNb3O10 compounds obtained from Le Bail
fitting using TOPAS software. .............................................................................. 63
Table 4.4. Lattice constants for RbBaxSr2-xNb3O10 compounds............................................... 63
Table 4.5. Atom percent results for RbCaxSr2-xNb3O10 obtained using EDS. Ca content was
estimated using ratio of Nb:Ca due to clean separation of peaks in EDS spectra.
Ca content error was calculated using Fieller’s method of estimated confidence
intervals between ratios. ........................................................................................ 72
Table 4.6. Atom percent results for RbBaxSr2-xNb3O10 obtained using EDS. Ba content was
estimated using ratio of Nb:Ca due to clean separation of peaks in EDS spectra.
Ba content error was calculated using Fieller’s method of estimated confidence
intervals between ratios. ........................................................................................ 73
Table 4.7. First moment analysis for RbCaxSr2-xNb3O10 compounds. The obtained CQ and
η values were calculated using Equation 2.42-45. ................................................. 89
Table 4.9. RbCaxSr2-xNb3O10 NMR Fit parameters. Errors in the last digit are shown in
parentheses and were calculated for population and CQ values using chi-square
minimization. ......................................................................................................... 93
Table 4.10. RbBaxSr2-xNb3O10 93Nb Static NMR Fit parameters. Errors in the last digit are
shown in parentheses and were calculated for population and CQ values using
chi-square minimization. ..................................................................................... 103
xii
Table 5.1. Le Bail method fit lattice parameters, particle size, zero correction, χ2, and Rwp*
for XRD patterns shown in Figures 3 and 4 ........................................................ 119
Table 5.2. NMR fit parameters for RbCa2NaNb4O13 and RbCa2NaNb2.8Ta1.2O13 obtained
from VOCS spectra. Error in the last digit is shown in parentheses................... 124
Table 5.3. 93Nb NMR fit parameters for HCa2NaNb4O13 and HCa2NaNb2.8Ta1.2O13 obtained
from MQMAS and VOCS spectra.
Error in the last digit is shown in
parentheses........................................................................................................... 130
Table 5.4 93Nb NMR fit parameters for HCa2NaNb4O13 and HCa2NaNb2.8Ta1.2O13 BRAINCP spectra. ........................................................................................................... 130
Table 5.5. 23Na NMR fit parameters obtained from MQMAS and MAS spectra. ................. 134
Table 5.6 Results of EFG tensor calculations using a point charge model for
HCa2NaNb4O13. ................................................................................................... 135
Table 6.1. Heating Cycles and Reaction Times for Microwave and Conventional Heating
Methods ............................................................................................................... 139
Table 6.2. Le Bail method fit lattice parameters, zero correction, χ2, particle size in stacking
direction, and Rwpa from Le Bail fits for hydrated and C6 samples. Error in the
last digit is represented by parentheses. ............................................................... 144
Table 6.3. Alkyl coverage percentages for C1-C6/CaxSr2-xNb3O10 compounds using
microwave grafting. ............................................................................................. 148
Table 7.1 Summary of exfoliation procedure by sample. More detailed descriptions of the
exfoliation procedures are also in the text. .......................................................... 159
Table 7.1 Cont’d. ................................................................................................................... 160
Table 7.2 Hansen Solubility Parameters for exfoliation solvents and alcohols used during
grafting of parent compounds. Values are given in units of MPa1/2. Values
taken from Hansen.197 .......................................................................................... 161
Table 7.3. XRD Stacking Reflections Summary ................................................................... 174
Table 7.4. Atomic composition of samples 14C and 15C using copper tape as substrate.
The letter in parentheses indicates to which image in Figure 7.28 above the data
xiii
correspond.
Error was automatically calculated by the Quantax EDS
application. .......................................................................................................... 191
xiv
List of Figures
Figure 1.1. Schematic representation of the ideal perovskite structure based on the cubic
SrTiO3 crystal structure. Left- Ball and stick model of unit cell where blue
spheres = Ti, red spheres = O, and green sphere = Sr. Right- polyhedral model
showing TiO6 octahedra in blue and Sr occupying interstitial 12-coordinate
sites in green. All crystallographic representations have been generated using
the VESTA software21 unless specified otherwise. ................................................. 2
Figure 1.2. Schematic representation of the D-J n=3 RbSr2Nb3O10 crystal structure with a
ball and stick model (a) and polyhedral model (b). Pink spheres represent Rb,
Light Blue spheres represent Sr, Dark Blue and Green octahedra represent the
exterior and interior NbO6 octahedra, respectively.................................................. 4
Figure 1.3. Schematic representation of the Aurivillius phase Bi2O2(Bi2Ti3O10) crystal
structure. Red spheres = Oxygen, pink spheres = Bi, and blue octahedra = TiO6
................................................................................................................................. 5
Figure 1.4. Schematic representation of the RbCa2NaNb4O13 crystal structure. Green
spheres = Nb, Red spheres = O, Pink spheres = Rb, Blue spheres = 2/3 Ca and
1/3 Na. ..................................................................................................................... 9
Figure 1.5. Ion-exchange and intercalation behavior of layered perovskites involving
modification to interlayer gallery cations. Illustration taken from Schaak and
Mallouk (Figure 2, Page 1458).40 .......................................................................... 10
Figure 1.6. Schematic assembly processes for nanosheets through flocculation,
electrostatic sequential absorption, and finally the Langmuir-Blodgett method.
Illustration taken from Ma and Sasaki (Figure 7, Page 5091).54............................ 12
Figure 1.7. Schematic illustration of the exfoliation process using HSr2Nb3O10 and TBAOH
as the intercalating base. ........................................................................................ 13
Figure 2.1. Illustration of 27Al CQ values (in MHz) with varying degrees of symmetry and
coordination numbers. Taken from Kentgens.90 ................................................... 23
Figure 2.2. Euler angles defining the direction of B0 in the principal axis system (PAS) of
the EFG under static conditions. ............................................................................ 25
xv
Figure 2.3. Energy level diagram91 of spin I=9/2 nucleus and first- and second-order
quadrupolar contribution where ωL is the Larmor frequency. ............................... 27
Figure 2.4. a) Simulated powder pattern using only first-order quadrupolar effects
displaying all transitions for an I=9/2 nucleus with CQ = 20 MHz and η = 0 at a
magnetic field strength of 9.4 T. b) Simulated powder patterns using both firstand second-order effects of the central transition for
CQ = 20MHz with η
= 0, 0.5, and 1.0.92.................................................................................................. 28
Figure 2.5. Parameters of the Herzfeld-Berger convention as they relate to the NMR
spectrum................................................................................................................. 30
Figure 2.6. 93Nb NMR chemical shifts for NbOx polyhedral.94 ............................................... 31
Figure 2.7. 93Nb VOCS spectra for RbSr2Nb3O10 at 14.1 T with offsets of 62.5 kHz between
neighboring slices.98 The summation of the slices is given with and without
baseline correction. ................................................................................................ 32
Figure 2.8. 93Nb static and MAS spectra at 9.4 T with varying spin rates (CQ = 20 MHz, η
= 0.2, δiso = 0ppm, Ω = 500 ppm, κ = 0.5, α = β = γ = 0). Spectra were simulated
using Simpson software.110 .................................................................................... 35
Figure 2.9. WURST-N Amplitude profiles for WURST-2, WURST-4, WURST-40, and
WURST-80 pulses. ................................................................................................ 37
Figure 2.10. Schematic illustrations of WURST excitation and refocusing pulses
reproduced from Schurko et al.115 A - Single excitation pulse showing signal
attenuation during the excitation pulse. B – A refocusing pulse with 2τref = τexc
leads to removal of the second-order phase distortion. C – A refocusing pulse
with τref = τexc does not eliminate second-order phase distortions as illustrated
by the displaced echoes shown in blue and red. .................................................... 38
Figure 2.11. QCPMG Pulse Sequence.120 ................................................................................ 39
Figure 2.12. 91Zr NMR powder patterns obtained from a ZrO2 MAS rotor at 9.4 T using (a)
WURST echo and (b) WCPMG pulse sequences. Values shown on the right are
the relative signal intensities, normalized to account for the differing number
of scans. The simulation in (c) includes central transition lineshapes for the
xvi
tetragonal (dotted) and orthorhombic (dashed) phases.
Illustration was
obtained from O’Dell.116 ........................................................................................ 40
Figure 2.13. BRAIN-CP pulse sequence. ................................................................................ 42
Figure 2.14. Schematic representation of the DOR sample setup requiring the spinning
about two axes simultaneously. ............................................................................. 43
Figure 2.15. Shifted-echo 3QMAS sequence with a FAM-II train of conversion pulses and
the corresponding coherence transfer pathway. ..................................................... 46
Figure 2.16. MQMAS-t2-REDOR pulse sequence. The I and S channels refer to the 1H and
93
Nb channels, respectively. An R3 recoupling sequence was used.139-140 ............ 48
Figure 4.1. XRD pattern for RbCa2Nb3O10. Le Bail fit parameters are shown in Table 4.2.
............................................................................................................................... 63
Figure 4.2. XRD pattern for RbCa1.9Sr0.1Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 64
-
64
Figure 4.3. XRD pattern for RbCa1.5Sr0.5Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 64
Figure 4.4. XRD pattern for RbCa1.4Sr0.6Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 65
Figure 4.5. XRD pattern for RbCa1.3Sr0.7Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 65
Figure 4.6. XRD pattern for RbCa1.1Sr0.9Nb3O10. Le Bail fit parameters are shown in Table
X. ........................................................................................................................... 66
Figure 4.7. XRD pattern for RbCa0.9Sr1.1Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 66
Figure 4.8. XRD pattern for RbCa0.7Sr1.3Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 67
Figure 4.9. XRD pattern for RbCa0.4Sr1.6Nb3O10. Le Bail fit parameters are shown in Table
4.2. ......................................................................................................................... 67
xvii
Figure 4.10. XRD pattern for RbSr2Nb3O10. Le Bail fit parameters are shown in Table 4.2.
............................................................................................................................... 68
Figure 4.11. XRD pattern for RbBa0.15Sr1.85Nb3O10. Le Bail fit parameters are shown in
Table 4.3. ............................................................................................................... 68
Figure 4.12. XRD pattern for RbBa0.3Sr1.7Nb3O10. Le Bail fit parameters are shown in
Table 4.3. ............................................................................................................... 69
Figure 4.13. XRD pattern for RbBa0.4Sr1.6Nb3O10. Le Bail fit parameters are shown in
Table 4.3. ............................................................................................................... 69
Figure 4.14. XRD pattern for RbBa0.5Sr1.5Nb3O10. Le Bail fit parameters are shown in
Table 4.3. ............................................................................................................... 70
Figure 4.15. XRD pattern for RbBa0.6Sr1.4Nb3O10. Le Bail fit parameters are shown in
Table 4.3. ............................................................................................................... 70
Figure 4.16. Graph of Unit Cell Volume vs. Ca content in RbCaxSr2-xNb3O10 compounds
calculated using lattice parameters from Le Bail fits. Note the gradual decrease
in unit cell volume with increasing Ca content due to the smaller ionic radius
of Ca2+ compared to Sr2+........................................................................................ 71
Figure 4.17. Graph of stacking distance in RbCaxSr2-xNb3O10 compounds vs. Ca content.
The stacking distance was obtained using the position of the (002) reflection.
Note the gradual decrease in stacking distance with increasing Ca content due
to the smaller ionic radius of Ca2+ vs. Sr2+............................................................. 71
Figure 4.18. EDS spectrum for RbCaSrNb3O10. Note the overlapping Rb and Sr signals. .. 73
Figure 4.19. SEM images of RbCa2Nb3O10 synthesized under dry microwave (a) and
molten salt (b) conditions and RbSr2Nb3O10 synthesized under dry microwave
(c) and molten salt (d) conditions. Note the increased platelet size seen in dry
microwave samples compared to molten salt samples. ......................................... 74
Figure 4.20. SEM images for attempted RbBa1.75Sr0.25Nb3O10 sample.
Note the
combination of expected platelets and the presence of rod-like impurities in the
left image resulting from formation of the BaNb2O6 phase.
xviii
The higher
magnification image on the right shows the rod-like nature of the impurity
particles.................................................................................................................. 75
Figure 4.23. Graphs of IR peak position vs. Ca content (Top) and Ba content (Bottom). ....... 76
Figure 4.24.
93
Nb Static WURST spectra for RbSr2Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site ;and Red - Interior Site. ................................................................................. 78
Figure 4.25. 93Nb Static WURST spectra for RbCa0.4Sr1.6Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 79
Figure 4.26. 93Nb Static WURST spectra for RbCa0.5Sr1.5Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); Yellow – Impurity; and Red
- Interior Site......................................................................................................... 80
Figure 4.27. 93Nb Static WURST spectra for RbCa0.7Sr1.3Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 81
Figure 4.28. 93Nb Static WURST spectra for RbCa0.9Sr1.1Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 82
Figure 4.29. 93Nb Static WURST spectra for RbCa1.1Sr0.9Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 83
Figure 4.30. 93Nb Static WURST spectra for RbCa1.4Sr0.6Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 84
Figure 4.31. 93Nb Static WURST spectra for RbCa1.5Sr0.5Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 85
xix
Figure 4.32. 93Nb Static WURST spectra for RbCa1.7Sr0.3Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior
Site 1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ........ 86
Figure 4.33. 93Nb Static WURST spectra for RbCa1.9Sr0.1Nb3O10 at 9.4 T (left) and 14.1 T
(right). Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site
1 (Sr-like); Purple - Exterior Site 2 (Ca-like); and Red - Interior Site. ............... 87
Figure 4.34.
93
Nb Static WURST spectra for RbCa2Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). Blue - Experimental Spectrum; Black - Total Fit; Purple - Exterior
Site and Red - Interior Site. .................................................................................. 88
Figure 4.35. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.4Sr1.6Nb3O10 with double
shearing scheme. The horizontal axis corresponds to the isotropic dimension
and vertical axis to the anisotropic dimension. Note the observed signal
positions along F2 are not equal to the isotropic chemical shifts in static spectra,
as illustrated in Equation 2.42. .............................................................................. 90
Figure 4.36. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.5Sr1.5Nb3O10 with double
shearing scheme. The horizontal axis corresponds to the isotropic dimension
and vertical axis to the anisotropic dimension. Note the observed signal
positions along F2 are not equal to the isotropic chemical shifts in static spectra,
as illustrated in Equation 2.42. .............................................................................. 90
Figure 4.37. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.7Sr1.3Nb3O10 with double
shearing scheme. The horizontal axis corresponds to the isotropic dimension
and vertical axis to the anisotropic dimension. Note the observed signal
positions along F2 are not equal to the isotropic chemical shifts in static spectra,
as illustrated in Equation 2.42. .............................................................................. 91
Figure 4.38. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.9Sr1.1Nb3O10 with double
shearing scheme. The horizontal axis corresponds to the isotropic dimension
and vertical axis to the anisotropic dimension. Note the observed signal
positions along F2 are not equal to the isotropic chemical shifts in static spectra,
as illustrated in Equation 2.42. .............................................................................. 91
xx
Figure 4.39. Graph of 93Nb CQ (in MHz) vs Ca content for Exterior Site 1 (blue), Exterior
Site 2 (orange), and Interior Site (grey). ................................................................ 94
Figure 4.40. Graph of 93Nb Population vs Ca content for Exterior Site 1 (blue), Exterior
Site 2 (orange), and Interior Site (grey). ................................................................ 95
Figure 4.41. Graph of 93Nb CSA span (Ω, ppm) vs Ca content for Exterior Site 1 (blue),
Exterior Site 2 (orange), and Interior Site (grey). .................................................. 95
Figure 4.42. Graph of 93Nb asymmetry parameter (η) value vs Ca content for Exterior Site
1 (blue), Exterior Site 2 (orange), and Interior Site (grey). ................................... 96
Figure 4.43. Graph of 93Nb skew (κ) value vs Ca content for Exterior Site 1 (blue), Exterior
Site 2 (orange), and Interior Site (grey). ................................................................ 96
Figure 4.44. 93Nb static WURST spectra for RbBa0.15Sr1.85Nb3O10 at 9.4 T (Top) and 14.1
T (Bottom). See Table 4.8 for detailed fit parameters. Experimental spectrum
is shown in blue and overall fit is shown in red..................................................... 99
Figure 4.45. 93Nb static WURST spectra for RbBa0.3Sr1.7Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). See Table 4.8 for detailed fit parameters. Experimental spectrum is
shown in blue and overall fit is shown in red. ..................................................... 100
Figure 4.46. 93Nb static WURST spectra for RbBa0.4Sr1.6Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). See Table 4.8 for detailed fit parameters. Experimental spectrum is
shown in blue and overall fit is shown in red. ..................................................... 101
Figure 4.47. 93Nb static WURST spectra for RbBa0.6Sr1.4Nb3O10 at 9.4 T (Top) and 14.1 T
(Bottom). See Table 4.8 for detailed fit parameters. Experimental spectrum is
shown in blue and overall fit is shown in red. ..................................................... 102
Figure 4.48. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2
(orange), Interior Site 1 (grey), and Interior Site 2 (yellow). .............................. 105
Figure 4.49. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2
(orange), Interior Site 1 (grey), and Interior Site 2 (yellow). .............................. 106
Figure 4.50. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2
(orange), Interior Site 1 (grey), and Interior Site 2 (yellow). .............................. 106
xxi
Figure 4.51. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2
(orange), Interior Site 1 (grey), and Interior Site 2 (yellow). .............................. 107
Figure 4.52. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2
(orange), Interior Site 1 (grey), and Interior Site 2 (yellow). .............................. 107
Figure 5.1. Tetragonal crystal structure for RbCa2NaNb4O13 ................................................ 113
Figure 5.2. XRD patterns for RbCa2NaNb4O13 (above) and RbCa2NaNb2.8Ta1.2O13 (below).
Note that the pattern for the Ta compound is fit with both the four and threelayer structures represented by blue and black ticks, respectively. ..................... 117
Figure 5.3. XRD patterns for HCa2NaNb4O13 (above) and HCa2NaNb2.8Ta1.2O13 (below).
Note that the pattern for the Ta compound is fit with both the four and threelayer structures represented by blue and black ticks, respectively. ..................... 118
Figure 5.4. SEM images of RbCa2NaNb4O13 (top) and RbCa2NaNb2.76Ta1.24O13 (bottom)
showing platelets with lateral dimension of 1-3 μm. ........................................... 119
Figure 5.5. 93Nb Static VOCS NMR spectra with fits for HCa2NaNb4O13 at 14.1 (a) and 9.4
T (c) along with HCa2NaNb2.8Ta1.2O13 at 14.1 (b) and 9.4 T (d). The black line
corresponds to the experimental spectrum and the blue line corresponds to the
full calculated spectrum. The other colored lines correspond to individual fits
for each Nb environment. HCa2NaNb4O13 CQ values for individual fits are as
follows: purple = 22.5 MHz; red = 28.7 MHz; green = 29.5 MHz; and orange
= 52.1 MHz. For HCa2NaNb2.8Ta1.2O13 CQ values for individual fits are as
follows: purple = 22.3 MHz; red = 25.9 MHz; green = 27.1 MHz; and orange
= 51.1 MHz. ......................................................................................................... 121
Figure 5.6. 93Nb Static VOCS NMR spectra with fits for RbCa2NaNb4O13 at 9.4 (top left)
and 14.1 T (bottom left) along with RbCa2NaNb2.76Ta1.24O13 at 9.4 (top right)
and 14.1 T (bottom right). The black line corresponds to the experimental
spectrum and the blue line corresponds to the full calculated spectrum. The
other colored lines correspond to individual fits for each Nb environment. For
RbCa2NaNb4O13 CQ values, purple = 33.5 MHz, red = 35.2 MHz, green = 18.8
MHz, and orange = 53.5 MHz. For RbCa2NaNb2.8Ta1.2O13 CQ values, purple =
28 MHz, red = 33 MHz, green = 21 MHz, and orange = 51 MHz. ..................... 123
xxii
Figure
93
5.7.
Nb
3QMAS
2D
spectrum
of
RbCa2NaNb4O13
(left)
and
RbCa2NaNb2.8Ta1.2O13 (right) with double shearing scheme. The horizontal
axis corresponds to the isotropic dimension and vertical axis to the anisotropic
dimension............................................................................................................. 126
Figure 5.8. 1H-93Nb BRAIN-CP spectrum with calculated fit. The black spectrum is the
experimental, blue the full calculated fit, red is the fit of site 3 (28.7 MHz), and
green is the fit of site 2 (29.5 MHz)..................................................................... 129
Figure 5.9. MQMAS-t2-REDOR spectra for HCa2NaNb4O13 without (left) and with (right)
REDOR sequence active. Numbers correspond to MQMAS sidebands for
appropriate sites. Site numbers are the same as those listed in Table 1.............. 131
Figure 5.9.
23
Na MAS and MQMAS spectra for HCa2NaNb4O13 (a and b) and
HCa2NaNb2.8Ta1.2O13 (c and d).
The black line corresponds to the full
experimental spectrum, the blue line is the full calculated spectrum, the red line
is the individual fit for site 1, the green line is the individual fit for site 2, and
the orange line is the individual fit for site 3. The fit parameters for each site
are listed in Table 2. ............................................................................................ 134
Figure 6.1. Illustration of stepwise microwave grafting. ....................................................... 137
Figure 6.2. (a) XRD patterns for HSr2Nb3O10 (black), C1/Sr2Nb3O10 (red), C3/Sr2Nb3O10
(green), and C6/Sr2Nb3O10 (blue) synthesized using the microwave irradiation
method. (b) XRD patterns for HCa2Nb3O10 (black), C1/Ca2Nb3O10 (red),
C3/Ca2Nb3O10 (green), and C6/Ca2Nb3O10 (blue) synthesized using a
conventional heating method. .............................................................................. 139
Figure 6.3. XRD patterns for HCa2Nb3O10 (Black), C1/Ca2Nb3O10 (red), and C3/Ca2Nb3O10
(green) synthesized using microwave grafting. Note the presence of an intense
reflection associated with the HCa2Nb3O10 phase. .............................................. 141
Figure 6.4. Long range XRD patterns for HCa2Nb3O10 (top) and HSr2Nb3O10 (bottom).
Experimental patterns are shown in blue. Le Bail fits are shown in red. Blue
ticks represent hkl reflections. The difference pattern is shown in gray. ........... 143
Figure 6.5. Long range XRD patterns for C6/Ca2Nb3O10 (top) and C6/Sr2Nb3O10 (bottom).
Experimental patterns are shown in blue. Le Bail fits are shown in red. Blue
xxiii
ticks represent hkl reflections for the C6 phase. Black ticks represent hkl
reflections for the hydrated phase. The difference pattern is shown in gray.
Listed hkl reflections refer to the C6 phase unless specified otherwise. *Refers
to the minority hydrated phase. ........................................................................... 144
Figure 6.6. (a) 1H MAS Hahn Echo NMR spectra for C6/Sr2Nb3O10 with varying echo
delay times and (b) results of peak integration as a function of the echo delay
time. ..................................................................................................................... 145
Figure 6.7. TGA data for C6/Sr2Nb3O10 (above) and C6/Ca2Nb3O10 (below) with
temperatures ranging from 100 to 800 °C. Based on theoretical values of
11.84% and 13.84% for 100% alkoxyl coverage of the strontium and calcium
samples, the alkoxyl coverage was found to be 39.1% and 45.0%, respectively.
............................................................................................................................. 146
Figure 6.8. Alkyl coverage percentages for C1-C6/CaxSr2-xNb3O10 compounds using
microwave grafting. ............................................................................................. 148
Figure 6.9. XRD patterns for H-C6/CaxSr2-xNb3O10 compounds grafted using the
microwave heating method. ................................................................................. 150
Figure 6.10. TGA graphs for C1 and C3/Sr2Nb3O10 samples. All mixed Ca/Sr samples
show the same general shape of graph with the only changes being the
magnitude of mass loss. ....................................................................................... 151
Figure 6.11. Tauc plots for H-C6/Sr2Nb3O10 (a-d, respectively) microwave grafted samples
showing substantial shift in band gap upon grafting. .......................................... 154
Figure 7.1. Image of C6/Ca2Nb3O10 in 5M2H (Sample 4) after sitting unmoved for one
week. .................................................................................................................... 164
Figure 7.2. Image of C6/Ca0.25Sr1.75Nb3O10 in 2P (Left – Sample 11), and in 5M2H (Right
– Sample 10) 30 minutes after initial sonication. Note that particles in Sample
11 are settling at a much faster rate than those in Sample 10. ............................. 165
Figure 7.3. Image of C6/Sr2Nb3O10 in 5M2H (Sample 5) shortly after microwave heating
(left) and after sitting untouched for 1 week (right). The observed Tyndall
xxiv
Effect is indicative of a colloidal suspension that does not settle out with
gravitational force alone. ..................................................................................... 166
Figure 7.4. Images for C3/Sr2Nb3O10 in 2P (Sample 8) after sitting untouched for 2 weeks
(Left) and 2 months (Right). The colloidal suspension is clearly stable after 2
months even despite approximately 50% of the solvent having evaporated. A
slight yellow tint can be observed in the sample after 2 weeks, but is ambiguous
after 2 months. ..................................................................................................... 166
Figure 7.5. Image of C6/Sr2Nb3O10 in 5M2H (Sample 15a) having sit for 5 hr after initial
sonication. Note the yellow color seen in Sample 12 from which it was
prepared is barely present, if present at all. ......................................................... 167
Figure 7.6. Image of C6/Sr2Nb3O10 in 5M2H after centrifugation (Sample 15a – Left) and
after mixing with ethanol and H2O (Sample 15b – Right). The organic layer in
Sample 15b is on top and aqueous layer on bottom. Note the substantial
cloudiness present in the aqueous layer in Sample 15b. ...................................... 168
Figure 7.7. Image of C3/Sr2Nb3O10 in 2P 2nd exfoliation SN (Sample 14c) and
C6/Sr2Nb3O10 in 5M2H 2nd exfoliation (Sample 15c). Note the yellow color in
Sample 15c returned after microwave heating .................................................... 169
Figure 7.8. Image of C6/Sr2Nb3O10 in 5M2H SN (Sample 15c) with laser beam passing
through. Observance of the Tyndall effect is evidence of a colloidal suspension.
This is the first instance in which the Tyndall Effect was observed after
centrifugation. ...................................................................................................... 169
Figure 7.9. UV-vis spectra for untreated 5M2H (Blue), 5M2H control sample (Orange),
and Nanosheet suspension (yellow)..................................................................... 170
Figure 7.10. UV-vis spectrum of C6/Sr2Nb3O10 parent compound. ...................................... 171
Figure 7.11. IR spectra for 5M2H/t-BuOH mixture untreated (Blue), 2P untreated (orange),
C6/Sr2Nb3O10 in 5M2H/t-BuOH mixture (grey), and C3/Sr2Nb3O10 in 2P
(yellow)................................................................................................................ 172
Figure 7.12.
1
H NMR spectra for 5M2H untreated, 5M2H control sample, and
C6/Sr2Nb3O10 in 5M2H SN. ................................................................................ 172
xxv
Figure 7.13. XRD pattern for C6/Ca2Nb3O10 in 5M2H sediment collected from the bottom
of the sample container (Sample 4). .................................................................... 176
Figure 7.14. XRD patterns for C6/Sr2Nb3O10 in 5M2H (Sample 5) slow air-dried
supernatant (top) and the powder sediment (bottom). ......................................... 176
Figure 7.15. XRD pattern of C6/Sr2Nb3O10 in 5M2H (Sample 6) after it had sat on the
benchtop untouched for over 1 year. All solvent had evaporated leaving solid
powder on the bottom of the container. Not the C6 phase is still present. ......... 177
Figure 7.16. XRD pattern for C3/Sr2Nb3O10 in 2P SN sediment (Sample 14c). Note the
presence of both C3 and hydrated phases at 4.8 and 5.7o, respectively. .............. 177
Figure 7.17. XRD pattern for C6/Sr2Nb3O10 in 5M2H (Sample 15c) Sediment. The
stacking reflection at 5.8o is consistent with a partially hydrated phase. The
slight plateau between 3.5 and 4.7o is possibly indicative of remaining
C6/Sr2Nb3O10. ...................................................................................................... 178
Figure 7.18. XRD pattern for C6/Ca2Nb3O10 in 5M2H suspension (Sample 4) and
C6/Sr2Nb3O10 in 5M2H suspensions (Sample 5) obtained by air-drying several
drops of the suspension (before centrifugation) onto a glass plate.
The
reflections at ~3.6 and ~5.8 degrees correspond to C6 grafted and hydrated
phases, respectively. Note the absence of a strong hydrated phase reflection in
Sample 5 is possibly due to differences in the suspension preparation
(sonication vs. microwave heating), differences in the procedure employed for
synthesis of the parent compound (conventional heating for Sample 4 vs.
microwave heating for Sample 5), or differing composition of the parent
compound. ........................................................................................................... 179
Figure 7.19. XRD patterns for C6/Sr2Nb3O10 in 5M2H (Sample 7) from the slow settling
phase (Top) and moderate settling phase (Bottom). ............................................ 181
Figure 7.20. XRD pattern for C3/CaSrNb3O10 in 2P (Sample 9) dried suspension. The
pattern on top is from a dried suspension prepared two weeks after the original
synthesis while the bottom pattern is from the day of the synthesis. The lack of
a grafted stacking reflection, in the dried suspension prepared after sitting two
weeks, and increased intensity of the hydrated stacking reflection suggest
xxvi
sheets in the suspension are reverting from the grafted state to the hydrated
state. ..................................................................................................................... 182
Figure 7.21. XRD pattern for C6/Sr2Nb3O10 in 5M2H centrifuge supernatant (Sample 5)
after drying. The sample was rapidly dried by placing drops of the supernatant
onto a glass plate heated to 80 oC such that solvent evaporated within seconds
until a significant amount of solid remained. The broad hump in the pattern is
due to the background signal of the glass plate (blank plate shown in blue). The
grey pattern is a simulation of (00l) stacking reflections assuming an average
number of 2 hydrated sheets stacked together. .................................................... 184
Figure 7.22. Long-range (Top) and short-range (Bottom) XRD patterns for C3/Sr2Nb3O10
in 2P (Sample 8) supernatant after sitting for 2 weeks. Only the hydrated phase
is observed with fewer higher order reflections compared to parent compound.
Note the slight hump at ~3o is due to the glass slide being place in a position
such that it blocked the X-Ray beam below 3o. ................................................... 185
Figure 7.23. Long-range XRD pattern of C6/Sr2Nb3O10 in 5M2H (Sample 13) supernatant
dried in a N2 atmosphere.
The lack of many higher order reflections is
consistent with irregular restacking of sheets, though the stacking reflection
position is consistent with a hydrated phase rather than the desired grafted
phase. ................................................................................................................... 186
Figure 7.24. Long-range XRD pattern for C6/Sr2Nb3O10 after freeze-drying in a 5M2H/tBuOH mixture (Sample 12). The presence of a sharp stacking reflection and
higher order reflections is likely due to the frozen solution melting several times
during the freeze-drying process, essentially resulting in a slow evaporation of
solvent and thus allowed for ordered restacking of sheets. ................................. 187
Figure 7.25. SEM image of C6/Sr2Nb3O10 in 5M2H (Sample 3) on TESCAN instrument.
While the particles have the platelet shape seen in parent compounds, the
particle size is substantially smaller with 681 nm being the largest lateral
dimension seen in the sample. Plate thickness of 67 and 154 nm correspond to
~40 and ~100 sheets stacked together, respectively. ........................................... 188
xxvii
Figure 7.26. C6/Ca2Nb3O10 in 5M2H SN (Sample 4 sitting unmoved for three weeks, Left)
and C6/Sr2Nb3O10 in 5M2H SN (Sample 6, Right) SEM images. Sample was
prepared by placing several drops onto SEM sample holder and letting air dry.
Sample 6 on the right was prepared over a much longer period with ~1mL of
SN in total dried onto the Si wafer. Note the substantially increased irregularity
and decreased size of the particles compared to parent compounds. ................... 188
Figure 7.27. SEM image (top) and EDS spectrum (bottom) for C6/Sr2Nb3O10 in 5M2H
(Sample 5) supernatant after evaporated solvent, re-dissolving, centrifuging,
and final evaporation of solvent. The small and irregular shape of the particles
combined with the presence of Nb in the EDS spectrum is strong evidence for
successful exfoliation of nanosheets. Note that there is no Sr signal assigned in
the EDS spectrum due to the Si signal from the wafer overlapping. ................... 189
Figure 7.28. SEM images for (a) C3/Sr2Nb3O10 in 2P (sample 14c) SN, (b) C6/Sr2Nb3O10
in 5M2H (sample 15c) SN. Note the horizontal lines in image (a) are to due
scratches on the copper tape background created during removal of the adhesive
layer. .................................................................................................................... 190
Figure 7.30. SEM image of C6/Sr2Nb3O10 nanosheets deposited onto a Cu2O nanocube.
The small (less than 500 nm), irregularly-shaped particles are the
exfoliated/restacked nanosheets. Note that there is a wide distribution of
nanosheet agglomeration. .................................................................................... 192
Figure 7.31. SEM image of C6/Sr2Nb3O10 nanosheets deposited onto a Cu2O nanocube.
The small (less than 500 nm), irregularly-shaped particles are the
exfoliated/restacked nanosheets. Note that the almost transparent quality of
nanosheets on the cube surface is consistent with the electron beam passing
through the thin sheet. ......................................................................................... 193
Figure A.1. List of indexed hkl reflections obtained from XRD pattern of RbSr2Nb3O10
microwave sample. .............................................................................................. 219
Figure A.2. List of indexed hkl reflections obtained from XRD pattern of RbSr2Nb3O10
molten salt sample ............................................................................................... 220
xxviii
Figure A.3. List of indexed hkl reflections obtained from XRD pattern of
RbCa0.3Sr1.7Nb3O10............................................................................................... 221
Figure A.4. List of indexed hkl reflections obtained from XRD pattern of
RbCa0.4Sr1.6Nb3O10............................................................................................... 221
Figure A.5. List of indexed hkl reflections obtained from XRD pattern of
RbCa0.5Sr1.5Nb3O10............................................................................................... 222
Figure A.7. List of indexed hkl reflections obtained from XRD pattern of
RbCa0.7Sr1.3Nb3O10............................................................................................... 223
Figure A.8. List of indexed hkl reflections obtained from XRD pattern of
RbCa1.1Sr0.9Nb3O10............................................................................................... 224
Figure A.9. List of indexed hkl reflections obtained from XRD pattern of
RbCa1.4Sr0.6Nb3O10............................................................................................... 226
Figure A.10. List of indexed hkl reflections obtained from XRD pattern of
RbCa1.9Sr0.1Nb3O10............................................................................................... 227
Figure A.11. List of indexed hkl reflections obtained from XRD pattern of RbCa2Nb3O10 .. 228
Figure A.12. List of indexed hkl reflections obtained from XRD pattern of
RbBa0.15Sr1.85Nb3O10 ............................................................................................ 229
Figure A.13. List of indexed hkl reflections obtained from XRD pattern of
RbBa0.3Sr1.7Nb3O10............................................................................................... 230
Figure A.14. List of indexed hkl reflections obtained from XRD pattern of
RbBa0.6Sr1.4Nb3O10............................................................................................... 231
Figure B.1. TGA data for C1/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 233
Figure B.2. TGA data for C3/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 233
Figure B.3. TGA data for C6/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 234
xxix
Figure B.4. TGA data for C1/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 234
Figure B.5. TGA data for C3/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 235
Figure B.6. TGA data for C6/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 235
Figure B.7. TGA data for C1/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 236
Figure B.8. TGA data for C3/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 236
Figure B.9. TGA data for C6/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 237
Figure B.10. TGA data for C1/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 237
Figure B.11. TGA data for C3/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 238
Figure B.12. TGA data for C6/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 238
Figure B.13. TGA data for C1/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 239
Figure B.14. TGA data for C3/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 239
Figure B.15. TGA data for C6/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800
°C. See Table B.1 for mass loss % and coverage %. ........................................... 240
Figure B.16. TGA data for C6/Ca2Nb3O10 with temperatures ranging from 150 to 800 °C.
See Table B.1 for mass loss % and coverage %. ................................................. 240
xxx
List of Abbreviations
TBAOH:
Tetrabutylammonium hydroxide
5M2H:
5-methyl-2-hexanone
2P:
2-pentanone
C1:
methanol grafted
C3:
n-propanol grafted
C6:
n-hexanol grafted compound
NMR:
Nuclear Magnetic Resonance
SSNMR:
Solid-State Nuclear Magnetic Resonance
MQ:
Multiple Quantum
PAS:
Principle Axis System
MAS:
Magic Angle Spinning
WURST:
Wideline Uniform Rate Smooth Truncation
BRAIN:
Broadband Adiabatic Inversion
CP:
Cross-Polarization
REDOR:
Rotational Echo Double Resonance
VOCS:
Variable Offset Cumulative Spectra
FAM:
Fast Amplitude Modulated
SN:
Supernatant
RF:
Radiofrequency
IR:
Infrared
xxxi
TGA:
Thermal Gravimetric Analysis
XRD:
X-Ray Diffraction
EFG:
Electric Field Gradient
CSA:
Chemical Shift Anisotropy
D-J:
Dion-Jacobson
R-P:
Ruddlesden-Popper
SEM:
Scanning Electron Microscopy
EDS:
Energy Dispersive Spectroscopy
CTC:
Charge-Transfer Complex
MSS:
Molten Salt Synthesis
CPMG:
Carr-Purcell-Meiboom-Gill
xxxii
Chapter 1: Background of Layered Perovskites
1.1 Introduction
The perovskite structure can be synthesized in such variety as to produce a wide
array of phases with vastly differing functions such as capacitance,1-3 piezoelectricity,4-6
insulation,7-9 metallic conduction,10-12 catalysis,13-15 and superconduction16-17 to name a
few. Furthermore, in just the past five years, hybrid organic-inorganic perovskites have
rapidly gained stardom to become the most studied semiconductors for use in photovoltaic
cells.18-19 In 1839, CaTiO3 was discovered in the Ural Mountains and subsequently named
Perovskite after Russian mineralogist Lev A. Perovski. 177 years later the term perovskite
is used to describe the large class of materials consisting of the same ABX3 structure.
Goldschmidt et al.20 performed pioneering work in the 1920’s on the perovskite structure
that laid the foundation for further exploration of the perovskite family which has bloomed
into arguably the most diverse class of ceramic materials. Even after over 100 years of
active research into perovskite compounds, there is no apparent slowdown in active
research as illustrated by 522 ACS published articles, in 2016 alone, containing the word
perovskite in the title. Such is the beauty of scientific research that the lone seedling of
discovery gives rise to a great tree of knowledge and understanding with branches reaching
out far beyond what possibly could have been envisioned when planted.
As previously mentioned, the chemical formula for perovskites is given as ABX3
with A being a Group I or II cation, B is a metal, and X is typically oxygen, though other
ions such as F- or Cl- are possible. SrTiO3 represents an idealized cubic structure with Sr2+
and O2- forming a closed packed cubic lattice with Ti4+ occupying the octahedral holes and
1
resulting in a three-dimensional structure of corner sharing TiO6 octahedra with Sr2+ ions
occupying the cavities between octahedra as shown in Figure 1.1. Both A and B cations
can be varied to a great degree resulting in a myriad of compounds with perovskite
structures.
Figure 1.1. Schematic representation of the ideal perovskite structure based on the cubic SrTiO 3
crystal structure. Left- Ball and stick model of unit cell where blue spheres = Ti, red spheres = O, and
green sphere = Sr. Right- polyhedral model showing TiO6 octahedra in blue and Sr occupying
interstitial 12-coordinate sites in green. All crystallographic representations have been generated
using the VESTA software21 unless specified otherwise.
The fundamental lattice parameter, a, in an ideal cubic perovskite can be calculated
using ionic radii rA, rB, and rO as given by the following equation:
Equation 1.1: = √2( +  ) = 2( +  )
It is possible to estimate the degree of distortion by taking the ratio of the two above
expressions to give Goldschmidt’s tolerance factor, t, calculated using Eq. 1.2. While this
tolerance factor assumes purely ionic bonding, it serves as a good approximation for
compounds with high degrees of ionic bonding.
2
Equation 1.2:  =
 +
√2( + )
The cubic structure is maintained for values of 0.89 < t < 1 with BO6 octahedra
tilting in order to maintain connectivity and minimize void space.22 For lower values of t,
the resulting crystal structure is stable in a lower symmetry environment. At the other end
of the spectrum, if t is greater than 1, hexagonal variants of the perovskite structure are
stable resulting in face sharing BO6 octahedra. It should be noted again that t serves as an
approximation and there are exceptions to these rules. Furthermore, distortions of the
perovskite structure can be caused by varying chemical composition and Jahn-Teller
distortions.
Among the class of perovskite materials, layered perovskites have arguably been
the most heavily studied in recent years due to a myriad of properties. There are three main
classes of layered perovskites, each consisting of the common motif of two-dimensional
anionic perovskite slabs stacked together with interleaving cations or cationic structural
units between each slab. These three phases of layered perovskite are the Aurivillius,
Ruddlesden-Popper (R-P), and Dion-Jacobson (D-J) phases. The D-J series of layered
perovskites, A’[An-1BnO3n+1] where n corresponds to the number of B-site octahedra, is
typified by the n = 3 triple layer RbSr2Nb3O10 (Figure 1.2) with one Rb interlayer cation
per Sr2Nb3O10 formula unit. A’ corresponds to the interlayer cation (in this case a group I
cation), A is the interstitial site (typically a group II element, although other +2 cations can
be suitable if they have an appropriate ionic radius size match), and B is the metal center
for the BO6 octahedra (typically a +5 metal for D-Js). R-P phases, A’2[An-1BnO3n+1], such
3
as K2La2Ti3O10 have two interlayer cations per formula unit and possess twice the
interlayer charge density of the D-J phases. The B site in the R-P phase is typically a +4
metal such as Ti4+.
Aurivillius phases (see Figure 1.3), including Bi2W2O9, are
intergrowths of perovskite and bismuth oxide with a covalent network of Bi2O2+ between
the two-dimensional perovskite slabs.
a)
b)
Figure 1.2. Schematic representation of the D-J n=3 RbSr2Nb3O10 crystal structure with a ball and
stick model (a) and polyhedral model (b). Pink spheres represent Rb, Light Blue spheres represent Sr,
Dark Blue and Green octahedra represent the exterior and interior NbO 6 octahedra, respectively.
4
Figure 1.3. Schematic representation of the Aurivillius phase Bi2O2(Bi2Ti3O10) crystal structure. Red
spheres = Oxygen, pink spheres = Bi, and blue octahedra = TiO6
Layered niobates have been of particular interest to researchers, due to the
versatility of Nb occupying the B-site. Different oxidation states (the most common and
stable being Nb(V)), along with high charges and small ionic radii creating a strong oxygen
affinity for Nb, has led to an extensive number of different niobate structures, with
elemental composition leading to a wide variety of interesting properties. The layered
niobates studied in this work belong entirely to the D-J class of layered perovskites.
Perhaps the most useful property of layered niobates is their inherent ability to undergo
soft-chemical structural modifications, specifically within the interlayer gallery separating
the perovskite blocks.
5
1.2 RbCaxSr2-xNb3O10 and RbCa2NaNb4-xTaxO13 Crystal Structures
A significant portion of this dissertation involves the n=3 RbCaxSr2-xNb3O10 and
n=4 RbCa2NaNb4-xTaxO13 versions of D-J perovskites, and thus it makes logical sense to
devote some time to an in-depth discussion of the synthesis and crystal structures for these
compounds derived from previous work before diving into the research conducted here. It
should be noted that while a great amount of study has been done into RbCa2Nb3O10,
RbSr2Nb3O10, and RbCa2NaNb4O13 structures, little information is available on the mixed
cation structures outside of the work presented herein.
RbCa2Nb3O10 was originally synthesized in 1981 by M. Dion23 and was initially
thought to form in the tetragonal space group P4212 with lattice constants a = 7.725(2) Å
and c = 14.909(5) Å. The original synthesis called for the heating of the carbonate and
oxide precursors Rb2CO3, CaCO3, and Nb2O5 in a 1.25:2:3 molar ratio for Rb:Ca:Nb (the
slight excess of Rb2CO3 is due to its volatility) at 1300 oC in a furnace for 24 hours, which
is the standard method of perovskite synthesis. Liang et al.24 later refined the crystal
structure from X-ray powder diffraction data using Rietveld refinement and found it to
exist in the higher symmetry tetragonal P4/mmm space group with a=3.85865(6) Å and
c=14.9108(3) Å. An alternative method involving substantially lower temperatures known
as molten salt synthesis (MSS)25-28 (discussed further in Chapter 3) has been explored for
many layered perovskites with success. Geselbracht et al.29 applied this method to
RbCa2Nb3O10 and found it could be readily synthesized in a RbCl flux at 800 oC in only 1
hour while maintaining the same tetragonal symmetry observed in conventional syntheses.
6
The NbO6 octahedra have substantial distortions resulting in a tilting effect due to the ionic
radii of Ca2+ not completely filling the void space at the interstitial site.
Thangadurai et al.30 reported the crystal structure for RbSr2Nb3O10 in the tetragonal
space group P4/mmm with a = 3.8944(2) Å and c = 15.2710(8) Å. Sr2+ has a larger ionic
radius than Ca2+ thus resulting in less octahedral tilting and an expansion of the crystal
lattice that is most dramatically seen in the difference in the c lattice parameters in
RbCa2Nb3O10 and RbSr2Nb3O10.
As with its Ca counterpart, RbSr2Nb3O10 can be
synthesized using either conventional high temperature synthesis or lower temperature
MSS. Geselbracht et al.31 synthesized a series of RbCaxSr2-xNb3O10 compounds using the
molten salt method and found the lattice parameters in the tetragonal unit cell progressed
in a linear fashion (see Table 1.1).
Table 1.1. Tetragonal unit cell parameters of RbCaxSr2-xNb3O10 phases collected by Geselbracht et al.31
X Value
a (Å)
c (Å)
V (Å3)
2
3.8551(8)
14.871(5)
221.01(9)
1.8
3.8598(9)
14.906(5)
222.1(1)
1.6
3.8632(8)
14.952(6)
223.2(1)
1.4
3.8664(9)
14.995(5)
224.2(1)
1.2
3.8704(9)
15.040(5)
225.3(1)
1.0
3.874(1)
15.078(5)
226.3(1)
0.8
3.8771(8)
15.116(4)
227.2(1)
0.6
3.8801(8)
15.149(4)
228.1(1)
0.4
3.8834(9)
15.178(5)
228.9(1)
0.2
3.8864(9)
15.208(5)
229.7(1)
0
3.8897(9)
15.233(5)
230.4(1)
The Nb-O bond lengths for RbSr2Nb3O10 as determined by Wang et al.32are shown
in Table 1.2, with site assignments illustrated in Figure 1.2a. The axial terminal Nb2-O4
bond length shown in Table 1.2 is substantially shortened at 1.7525 Å, to the point where
this bond can be considered essentially a Nb-O double bond with Nb in a pseudo five
coordinate environment. Complementing this contraction of the terminal Nb2-O4 bond,
7
the Nb2-O2 bond bridging the exterior and interior octahedra is substantially elongated to
2.4539 Å from the ideal bond length of 2.06 Å calculated based on Shannon-Prewitt Crystal
Radii.33 Elongation of the axial bridging B-O bond along with shortening of the axial
terminal B-O bond is characteristic of all alkali layered D-J perovskites.
Table 1.2 Nb-O bond lengths in RbSr2Nb3O10.
Bond
Nb1(Interior)-O2(Axial)
Nb1(Interior)-O1(Equatorial)
Nb2(Exterior)-O2(Axial)
Nb2(Exterior)-O3(Equatorial)
Nb2(Exterior)-O4(Axial)
Length (Angstroms)
1.95202
2.00852
2.45386
1.98825
1.75245
KCa2NaNb4O13 was first synthesized by Jacobson et al.34 through the incorporation
of an additional perovskite layer of NbO6 octahedra onto the parent compound
KCa2Nb3O10.
RbCa2NaNb4O13 was synthesized by mixing equimolar amounts of
RbCa2Nb3O10 and NaNbO3 followed by heating at 1200 oC for 3 hours. Sato et al.35 solved
the crystal structure for this compound, shown in Figure 1.4, using Rietveld analysis of
powder X-ray diffraction data and found it exists in the tetragonal P4/mmm space group
with a=3.8702(1) Å and c=18.8935(6) Å where Na+ and Ca2+ are uniformly distributed
among the interstitial sites. A thorough literature search has found no mention of mixed
Nb/Ta RbCa2NaNb4-xTaxO13 compounds such as those studied in this work. As will be
discussed at length in later chapters, the crystal structure proposed by Sato et al.35 does not
correspond to a stable structure as evidenced by bond valence calculations nor does it agree
with results from NMR experiments.
8
Figure 1.4. Schematic representation of the RbCa2NaNb4O13 crystal structure. Green spheres = Nb,
Red spheres = O, Pink spheres = Rb, Blue spheres = 2/3 Ca and 1/3 Na.
1.3 Soft-chemical Exchange Reactions in Layered Perovskites
A particularly desirable property of layered perovskites is the ability to undergo ion
exchange to prepare a wide variety of novel materials using low-temperature reaction
conditions. The most common form of ion exchange is the acid exchange reaction where
the layered perovskite is treated with a strong acid thus exchanging the interlayer cations
with acidic protons. Acid exchanging has led to many innovations36-39 regarding the
catalytic ability of layered perovskites due to the generation of a solid acid after successful
exchange. Of course, for these active acid sites to be any of use, the reactant must first
reach the active site and in the layered form; unless the reactant can enter the interlayer
gallery these sites are useless. Therefore, it is highly desirable to exfoliate layered niobates
9
into two-dimensional nanosheets.
The most common route towards exfoliation into
nanosheets is through intercalation of bulky organic bases such as tetrabutylammonium
hydroxide (TBAOH) into the protonated interlayer gallery. Schaak and Mallouk40 have
created an illustrated and informative summary of ion-exchange and intercalation behavior
of layered perovskites which is shown in Figure 1.5.
Figure 1.5. Ion-exchange and intercalation behavior of layered perovskites involving modification to
interlayer gallery cations. Illustration taken from Schaak and Mallouk (Figure 2, Page 1458).40
10
1.4 Niobate Nanosheets
Along with the obvious benefit of increased surface area and access to catalytically
active acid sites, nanosheets can also serve as building blocks for novel materials through
controlled reassembly of sheets.41-42
An excellent example of the potential for 2D
nanosheets as building blocks comes from Sasaki et al.43 where it was found that through
a layer-by-layer assembly approach using Langmuir-Blodgett depositions, superlattices of
(LaNb2O7/Ca2Nb3O10)5 and (Sr2Nb3O10/Ca2Nb3O10)5 possessed properties such as
increased dielectric constants and piezoelectric responses not seen in the bulk materials. A
visual guide to the Langmuir-Blodgett assembly of nanosheets is shown in Figure 1.6.
Furthermore, nanosheets have shown incredible promise in areas of photovoltaic cells,44-46
nanodielectrics,47-48 photocatalysis,49 solid acid catalysis,50-52 and flexible memory
devices53 among others. Unfortunately, it is especially difficult to characterize nanosheets
(an essential step in the development of new materials and new applications for old
materials) and it is thus of great importance to successfully characterize the parent
compounds to elucidate the structure and behavior of the resulting nanosheets.
11
Figure 1.6. Schematic assembly processes for nanosheets through flocculation, electrostatic sequential
absorption, and finally the Langmuir-Blodgett method. Illustration taken from Ma and Sasaki (Figure
7, Page 5091).54
The typical exfoliation method requires further exchanges or intercalations to be
performed on the protonated niobates to expand the interlayer gallery to a point that the
perovskite slabs break free from each other and single nanosheets are produced. These
acid-exchanged perovskites can intercalate a variety of organic bases such as alkylamines.
Intercalation with a bulky base such as tetrabutyl ammonium hydroxide (TBAOH) expands
the interlayer gallery to the point of exfoliation into single sheets. While this method of
exfoliation has been tried and true, it is decidedly lacking in many respects. The reaction
time required (typically 6 days minimum for effective exfoliation) severely inhibits the rate
at which novel compounds can be studied. Furthermore, this method is limited to using
aqueous solvents and the removal of the TBAOH cations with acid to regenerate the acid
12
sites without restacking of the sheets is an incredibly sensitive procedure that requires the
near constant presence of the investigator over the course of several hours to ensure acid
addition does not go above a certain rate. As trivial as complaints of procedural tediousness
may appear, this labor-intensive approach acts as an inhibitor to research pursuits. A
schematic illustration of the exfoliation process using TBAOH is shown in Figure 1.7.
Figure 1.7. Schematic illustration of the exfoliation process using HSr2Nb3O10 and TBAOH as the
intercalating base.
An alternative to the traditional exfoliation procedure will be discussed in great
depth in this dissertation. This new method of exfoliation involves first the grafting of long
chain alcohols into the interlayer gallery followed by treatment with an appropriate organic
solvent to break sheets apart. As will be shown, this new exfoliation method diversifies
potential reaction conditions as well as substantially shortening overall synthetic time.
While X-ray diffraction (XRD) has been used extensively in the study of layered
perovskite structure, it provides only information about long range order and is not well
suited for the study of small structural changes. Nuclear Magnetic Resonance (NMR),
however, is sensitive to the local environments surrounding each nucleus of interest and is
13
an excellent tool for the study of slight structural modifications that do not exhibit longrange order. Signals arising from unique nuclear environments within the crystal structure
further make NMR an appropriate technique for determining the relationship between
structure and composition. In this regard, this dissertation will detail the extensive solid
state
93
Nb NMR study of mixed alkaline cation layered perovskites Rb[AxSr2-xNb3O10]
where A is either Ca or Ba.
1.5 Research Presented Herein
The research described in this dissertation spans nearly six years and can be broken
into three heavily related research projects:
4. Exploration of local structural changes in mixed alkali layered perovskites.
5. Determination of B-site preference in mixed Nb/Ta systems using solid-state NMR
6. Microwave assisted grafting and exfoliation of layered perovskites using long chain
alcohols.
1.5.1 Exploration of local structural changes in mixed alkali layered perovskites.
A series of three-layered D-J RbSrxCa2-xNb3O10 (0 < x < 2) and RbBaxSr2-xNb3O10
(0 < x < 0.8) compounds were synthesized using microwave and molten salt synthetic
routes, respectively, and analyzed using powder x-ray diffraction (XRD), scanning electron
microscopy (SEM), energy dispersive spectroscopy (EDS), attenuated total reflectanceinfrared spectroscopy (ATR-IR), and wideline uniform rate smooth truncation quadrupolar
Carr-Purcell-Meiboom-Gill (WURST-QCPMG) NMR.
XRD patterns yielded lattice
parameters consistent with the expansion of the crystal lattice with increasing Sr or Ba
14
content due to the size of the Sr and Ba cations. ATR-IR spectra indicated a linear
progression in the band shift from 768.9 to 738.2 cm-1 going from the pure calcium
compound to the pure strontium compound and from 738.2 to 691.5 cm-1 going from the
pure strontium compound to the x = 0.6 barium containing compound. NMR studies were
used to obtain electric field gradient (EFG) and chemical shift anisotropy (CSA)
information for the 93Nb sites in the compounds to correlate change in the local structure
with change in the A-site cations. The EFG value was found to decrease with increasing Sr
content in the Ca/Sr series and increasing Ba content in the Sr/Ba series indicating a
reduction in NbO6 octahedral distortions, specifically octahedral tilt.
1.5.2 Determination of B-site preference in mixed Nb/Ta systems using solid-state NMR
D-J phase layered perovskites can readily undergo proton exchange under mild
conditions leading to the formation of a solid acid capable of intercalating various organic
molecules.40 The B-site atom in this class of compounds is often Ti4+, Nb5+, or Ta5+. It
has been shown that changing the B-site atom can drastically alter the intercalation
behavior and catalytic ability of these materials.31, 55-56 Geselbracht et al.31 studied the
acidity of HCa2NbxTa3-xO10 compounds through intercalation and found that the pyridine
intercalation reaction did not work for x = 2.5 and 3.0 compounds suggesting a significant
decrease in solid acid strength. For x = 0.5-2.0, however, the level of pyridine intercalation
was greater than in the x=0 compound suggesting the ordering of Ta5+ within the structure
has an effect as to whether the solid acid strength is decreased or increased. In work on
the water splitting ability of SrTa2O7 and SrNb2O7, Kudo et al.57 found the water splitting
15
activity and band gaps of the SrTa2O7 and SrNb2O7 compounds to be vastly different (4.6
eV vs. 3.9 eV, respectively). In their work, no research was done on mixed Nb5+/Ta5+
systems.
While Nb5+ and Ta5+ possess the same charge and approximately equal ionic radii58
(0.64 Ǻ), Nb is more electronegative than Ta (1.6 vs. 1.5, respectively)59 leading to Nb-O
bonds, which are more covalent in character than Ta-O bonds and changes in the energy
states of Ta5+ and Nb5+ compounds.57, 60 Studies on site ordering in mixed Nb5+/Ti4+ layered
perovskites61-62 have shown site preference based on the different charges and ionic radii
(0.64 vs. 0.605 Ǻ for Nb5+ and Ti4+, respectively) of Nb5+ and Ti4+. As subsequent chapters
will show, in mixed Nb5+/Ta5+ systems, electronegativity is a strong driver of site ordering.
As illustrated in Figure 1.4, there are two distinct crystallographic B-sites in the
four-layer RbCa2NaNb4-xTaxO13 system, two interfacial octahedra, Nb(2), and two interior
octahedra, Nb(1). X-ray powder diffraction is unable to give quantitative results as to the
populations of Nb5+ and Ta5+ at each crystallographic site due to the disordered nature of
these compounds. NMR, however, allows for the study of the local structure and is ideal
for systems of this sort. Cation site-ordering in these systems will be discussed in detail in
Chapter 5.
1.5.3 Microwave assisted grafting and exfoliation of layered perovskites using long chain
alcohols.
Effective catalytic behavior in these compounds requires access to the protonated
environments typically achieved through exfoliation into single nanosheets.63 Sugahara et
al.64-69 has pioneered grafting techniques for attachment of various organic groups into the
16
interlayer of layered perovskites. In particular, Sugahara has demonstrated a grafting
technique for protonated D-J perovskites to produce derivatives with n-alcohols as large as
n-octadecanol. Current grafting methods of layered perovskites involve high pressure
heating at 80 to 150 oC for days to weeks, depending on the alcohol involved. The long
reaction time involved inhibits the ability to study grafted samples in a timely fashion. In
addition to providing more detailed background, Chapters 6 and 7 in this dissertation will
present new microwave-assisted grafting and exfoliation techniques, along with detailed
comparisons to conventional methods, designed to expand and improve current
methodologies.
17
Chapter 2: NMR Background
2.1 Solid-State NMR
In 1946, NMR was first reported for use in bulk materials by Bloch, Hansen, and
Packard70 and by Purcell, Torrey, and Pound.71 In just six years, this discovery was
recognized with the Nobel Prize in Physics being awarded to Block and Purcell. In the
subsequent seven decades, NMR has evolved into arguably the most important
spectroscopic characterization technique especially after the implementation of
superconducting magnetics which have significantly reduced the problems of magnetic
field inhomogeneity along with the development of Fourier Transform NMR (FT-NMR)
by Richard Ernst72 which resulted in yet another NMR related Nobel Prize in 1991 for
Chemistry. Furthermore, NMR was adapted for use in biological imaging resulting in the
robust diagnostic method of Magnetic Resonance Imaging (MRI) which itself resulted in
the 2003 Nobel Prize in Physiology or Medicine being awarded to Lauterbur73 and
Mansfield.74
Liquid samples generally yield sharp NMR signals due to rapid isotropic molecular
tumbling averaging the anisotropic NMR interactions to zero. In solid samples, however,
the effects of anisotropic interactions are not averaged out thus resulting in substantial
broadening of the observed signals.
While this signal broadening does complicate
extraction of individual NMR interactions, solid-state NMR (SSNMR) is still an incredibly
useful tool for studying local structure (as opposed to the observation of long range periodic
structure in X-Ray diffraction). The development of numerous techniques to overcome the
issues of line broadening have allowed for SSNMR to be utilized in the studies of a variety
18
of materials such as crystalline,75 non-crystalline,76 and porous materials.77 SSNMR has
been used in the examination of complex organic and inorganic systems,78 biomolecular
systems,79 and polymeric systems.80
When NMR active nuclei are placed in a magnetic field, B0, their energy levels split
and shift due to the Zeeman effect. Subsequent application of a radiofrequency (rf) pulse
induces transitions between energy levels. The Zeeman interaction is treated as the
dominant interaction and is described by the Hamiltonian:
̂0 = −̂ 0
Equation 2.1: 
where ̂ is the nuclear magnetic moment operator. Since B0 is generally applied along the
z-axis, the Hamiltonian can be rewritten in terms of the spin operator ̂ :
̂0 = −ħ̂ 0 = −ħ̂ 
Equation 2.2: 
where ωL is the Larmor frequency (i.e. the rate of precession of the magnetic moment
around the external magnetic field), ωL = γB0, in units of Hz, ħ is the reduced Planck
constant, and γ is the gyromagnetic ratio.
̂ of the spin system is a
According to perturbation theory, the total Hamiltonian 
summation of the Zeeman interaction along with chemical shift, dipolar coupling, and
quadrupolar coupling and can be described as:
̂=
̂0 + 
̂ + 
̂ + 
̂
Equation 2.3: 
19
Depending on the magnitude of the various interactions, the perturbation terms may require
higher-order corrections. As will be shown in later sections, the quadrupolar Hamiltonian
̂(2) contribution for the 93Nb spin systems
requires the consideration of the second order 
studied in this work.
The magnetic moments of nuclear spins can interact through space resulting in the
dipolar coupling interaction. Dipolar coupling is a major contributor to line-broadening in
the solid-state. The Hamiltonian for dipolar coupling between two spins I and S (given in
angular frequency units, rad/s) is given by the following equation:

ℎ
̂
Equation 2.4: 
= − (40 )
  ħ
̂ ̂ (3 2 
3
− 1)
where r3 is the internuclear distance between spins I and S.
As will be discussed further in later sections, under Magic Angle Spinning (MAS)
conditions the dipolar interaction is reduced to zero.
However, many solid-state
techniques81-85 have been developed which allow for dipolar coupling to be reintroduced
under MAS by applying rf pulses to modify subsequent Hamiltonians thus allowing for the
extraction of information regarding internuclear distances.
2.2 Quadrupolar Nuclei
While most are familiar with NMR only using spin ½ nuclei, 78% of NMR active
nuclei have integer (6%) or half-integer (72%) quadrupolar spins. Quadrupolar nuclei
possess an electric quadrupole moment, eQ, that interacts with the electric field gradient
(EFG). The EFG is created by the distribution of charges surrounding the nucleus being
20
studied and is directly proportional to the asymmetry of the charges, with a more
asymmetric charge distribution creating a larger EFG. This interaction leads to broadening
of NMR powder patterns and complicates analysis such that more advanced experiments
are needed, although it also provides useful information relating to the local environments
of nuclei.
The quantitative description of the EFG is presented as a 3x3 tensor, V, shown in
Equation 2.5, where the components Vij relate to the three-dimensional distribution of
charges around the nucleus:

Equation 2.5:  = [





 ]

Components Vij can be approximated using a rudimentary point charge model using
Equation 2.6, where qe is nearby charge, r is the distance from the nucleus to the charge,
and x, y, and z are the bond distance components along the various axes:
32 −2
)
5
∑  (
Equation 2.6:  =  ∑  (
[
3 
)
3
3 
)
3
∑  (
3 
)
3
∑  (
32 −2
)
5
∑  (
∑  (
∑  (
3 
)
3
∑  (
3 
)
3
3 
)
3
32 −2
)
5
]
∑  (
The EFG tensor is described in its own principal axis system (PAS) through
diagonalization:

Equation 2.7:  = [ 0
0
0

0
0
0 ]

21
The quadrupolar Hamiltonian for spin I in applied magnetic field B 0 is given by:8688
Equation 2.8: Ĥ =

∙
6(−1)ħ
∙
The terms are typically defined such that |Vxx| < |Vyy| < |Vzz| and Vxx + Vyy + Vzz =
0.89 The quadrupolar coupling constant (CQ) is directly related to the EFG as shown
in Equation 2.9. CQ refers to the interaction of the quadrupole moment, eQ, and the
electric field gradient, eVzz. Another EFG related term involved in quadrupolar
nuclei is the asymmetry parameter, η, shown in Equation 2.10 with values ranging
from 0 to 1.
Equation 2.9:  =  2  /ℎ
Equation 2.10:  = ( −  )/
The relationship between charge distribution symmetry and CQ is most readily illustrated
in Figure 2.1, showing increasing asymmetry resulting in larger CQ values.
22
Figure 2.1. Illustration of 27Al CQ values (in MHz) with varying degrees of symmetry and coordination
numbers. Taken from Kentgens.90
According to perturbation theory and Hamiltonian theory, it is typically the firstand second-order quadrupolar interactions which are accounted for given by the
Hamiltonians: T̂
(1)
Equation 2.11: Ĥ = √6
 2
 
4(2−1) 20 20
23
 2
(2)
2
Equation 2.12: Ĥ = − (4(2−1))
1 2
 5


 {(−3√10̂̂30 + ̂̂10 (3 − 4( + 1))) 00 +


(−12√10̂̂
30 − ̂̂10 (3 − 4( + 1))) 20 + (−34√10̂̂30 +

3̂̂10
(3 − 4( + 1)))40 }

where ̂̂0 terms are spherical tensors defined by:

Equation 2.13: ̂̂10
= 
1
2
Equation 2.14: ̂̂
20 = √6 (3 − ( + 1))
1
2
Equation 2.15: ̂̂
30 = √10 (5 − 3( + 1) + 1)
Geometrical terms are contained in Vk0:

Equation 2.16: 0 = ∑ 0
(, , )
 (,

Equation 2.17: 0
, ) = (− ′ ) (−)′
()

Where 0
(, , ) are Wigner rotational matrices through which the EFG tensor is
transformed from its PAS frame to the laboratory frame, (, , ) are the Euler
angles associated with such rotation as illustrated in Figure 2.2, and dkq′q (β) is a
reduced Wigner function. Akn terms are given by:
24
1
5
Equation 2.18: 00 = − (3 + 2 )
1
1
3
Equation 2.19: 20 = 14 (2 − 3), 2±2 = 7 √2 
1
3
5
Equation 2.20: 40 = 140 (18 + 2 ), 4±2 = 70 √2  , 4±4 = 4
1
2
√70 
Figure 2.2. Euler angles defining the direction of B 0 in the principal axis system (PAS) of the EFG
under static conditions.
The first-order quadrupole effect is independent of the Larmor frequency, ωL,
meaning that the central transition (m = +1/2 ↔ -1/2) and symmetric multiple quantum
transitions (i.e. +3/2 ↔ -3/2, +5/2 ↔ -5/2, etc…) are unaffected by the strength of the
magnetic field. The second-order quadrupole effect, however, is inversely proportional to
ωL such that the effects of quadrupolar coupling are decreased with increasing magnetic
field strength. Using the above equations, the transition frequencies from the quadrupolar
coupling associated with first and second-order contributions can be given by:
(1)
3 2 
Equation 2.21: ∆ = 4(2−1) (2 + 1)20
25
(2)
 2 
2
Equation 2.22:  = − (4(2−1))
2 2
 5
∗ {(( + 1) − 9( + 1) − 3)00 +
(8( + 1) − 36( + 1) − 15)20 + 3(6( + 1) −
34( + 1) − 13)40 }
As will discussed in further detail, the quadrupolar nucleus primarily studied in this
work is spin I=9/2
93
Nb. As shown in Figure 2.3, in a B0 magnetic field, there are ten
(2I+1) Zeeman energy levels for the 93Nb nucleus with different precession frequencies at
equilibrium magnetization. Transitions between consecutive energy states are referred to
as single-quantum (1Q) transitions while transitions between non-consecutive states are
referred to as multiple-quantum (MQ) transitions.
The transition between Zeeman
eigenstates +1/2 and -1/2 is the central transition (CT) while any other single quantum
9
7
7
5
5
transitions between other eigenstates such as (± 2 ↔ ± 2) , (± 2 ↔ ± 2) , (± 2 ↔
3
3
1
± 2) ,  (± 2 ↔ ± 2) are considered satellite transitions (STs). An example of peaks
due to the CT and ST’s is shown in Figure 2.4a where satellite transitions cover a much
wider frequency range and have substantially lower intensities than the CT. It should be
noted that often if ST peaks are not well resolved they manifest as a broad baseline
overlapping with the CT which can be easily subtracted from the resulting powder pattern.
The powder pattern is the result of lines from different molecular orientations overlapping
over a range of frequencies thus forming a continuous lineshape. As the above equations
indicate, the CT frequency is unaffected by first-order quadrupolar effects. The CT
26
lineshape under static conditions also exhibits easily observable characteristic appearances
depending on the asymmetry parameter as seen in Figure 2.4b. This significantly improves
the ease of extracting the EFG tensor values under certain experimental conditions
(discussed further in subsequent sections) because the first term in the second-order
quadrupolar interactions has no PAS orientation dependence.
Figure 2.3. Energy level diagram91 of spin I=9/2 nucleus and first- and second-order quadrupolar
contribution where ωL is the Larmor frequency.
27
a)
b)
Figure 2.4. a) Simulated powder pattern using only first-order quadrupolar effects displaying all
transitions for an I=9/2 nucleus with CQ = 20 MHz and η = 0 at a magnetic field strength of 9.4 T. b)
Simulated powder patterns using both first- and second-order effects of the central transition for
CQ = 20MHz with η = 0, 0.5, and 1.0.92
Other interactions related to the local bonding electrons result in the
chemical shift anisotropy (CSA), which is dependent on the local symmetry of the metal
28
site.86 The CSA tensor is described by three diagonal elements in its respective PAS, δ11,
δ22, and δ33. Three relevant parameters arise from combinations of the tensor components,
the isotropic chemical shift, δiso, the span (related to the magnitude of chemical shift
anisotropy), Ω, and the skew (related to the symmetry of CSA), κ, defined in Equations
2.23-2.25 using the Herzfeld-Berger Convention.93 As llustrated in Figure 2.5, δiso is
commonly regarded as the center of gravity for the spectrum while Ω describes the
magnitude of the anisotropy. The skew describes the asymmetry of the chemical shift
tensor with values for the skew ranging from -1 to 1.86 When the skew is equal to +1 or 1 then δ22 must be equal to δ11 or δ33 thus implying the chemical shift tensor exhibits axial
symmetry. This serves as a quick test as to the veracity of the crystal structure being used.
As will be discussed in later chapters, departure of the skew from axial symmetry is
indicative of presence of an orthorhombic space group as opposed to a tetragonal space
group. A set of Euler angles (see Figure 2.1), α, β, and γ are used to define the relative
orientation of one tensor in terms of another.
Equation 2.23:  = 11 − 33
Equation 2.24:  = 3( − 22 )/
0
Equation 2.25: 
= (11 + 22 + 33 )/3
29
Figure 2.5. Parameters of the Herzfeld-Berger convention as they relate to the NMR spectrum.
2.3 93Nb Quadrupolar Interactions
The NbO6 octahedra are key components of the layered perovskites studied in this
work and it has been found that the EFG of the
93
Nb nuclei provide incredibly useful
information regarding the local octahedral structures.
93
Nb has a gyromagnetic ratio of γ
= 6.56 x 107 rad/sT on par with the more commonly studied 13C nucleus, but with 100%
natural abundance making it a desirable nucleus to study. It does however, have a spin of
I = 9/2 and a substantial quadrupole moment Q = -32 x 10-32 m2 resulting in large secondorder quadrupolar contributions to the lineshape of the CT spectrum. As will become
abundantly clear, 93Nb spectra are dominated by second-order quadrupolar coupling which
can be >100 MHz for Nb(V) complexes resulting in signals spanning several hundred kHz.
30
The parameters of the EFG tensor for 93Nb in layered perovskites is dependent on
the octahedral structure with quadrupolar coupling typically found in the wide range of 10120 MHz and isotropic chemical shift values from -650 ppm to -1600 ppm, as shown in
Figure 2.6, depending on the coordination of the surrounding oxygen environments.94
Traditionally, the analysis of
93
Nb NMR spectra has been limited to highly crytalline
phases, but in recent years, novel techniques have been developed and optimized such that
the highly disordered materials studied in this work, containing multiple overlapping 93Nb
signals, can be successfully characterized.
NbO4
NbO5
NbO6
NbO7
NbO8
500
750
1000
1250
1500
1750
2000
δiso, ppm
Figure 2.6. 93Nb NMR chemical shifts for NbOx polyhedral.94
2.4 Variable Offset Cumulative Spectra (VOCS)
Collecting static 93Nb spectra is done through the use of the Hahn Echo sequence95
rather than a single pulse. This allows for refocusing of magnetization lost in the dead time
and distorted by ringing signals from the probe. Unfortunately, the static 93Nb spectra with
large quadrupolar coupling can span up to several MHz making it impossible to irradiate
31
uniformly at one time using the standard hard pulse. Massiot96-97 developed the VOCS
technique as a method of collecting the full static powder pattern without spectral
distortion. The VOCS method requires the collection of a series of Hahn Echo spectra with
gradually changing offsets such that the final spectrum can be reconstructed by the
summation of the series as illustrated in Figure 2.7.
Figure 2.7. 93Nb VOCS spectra for RbSr2Nb3O10 at 14.1 T with offsets of 62.5 kHz between neighboring
slices.98 The summation of the slices is given with and without baseline correction.
32
The power of the rf pulses used in the Hahn echo sequence allows for the
calculation of the irradiation profiles for each echo experiment. From this, the frequency
offset needed can be calculated such that the total irradiation profile is even across the
whole spectrum. The major drawback of the VOCS method is the long experiment time
due to the collection of many spectra (sometimes upwards of 20 individual spectra) making
it an impractical method for the analysis of a large series of samples.
2.5 Magic Angle Spinning
When quadrupolar nuclei are put into an applied magnetic field Bo, they experience
chemical shift anisotropy (CSA), dipolar coupling and quadrupolar coupling.99 These three
main interactions give rise to static spectra with incredibly wide powder patterns spanning
several hundreds of kilohertz (kHz).96,
100-101
Even in ideal samples with only one
quadrupolar environment present, this breadth produces difficulty regarding extracting the
CSA and EFG parameters because the static signal is affected by all three interactions.
Multiple quadrupolar environments, a characteristic of the compounds studied in this work,
further complicates matters making it nearly impossible to extract the physically correct
CSA and EFG parameters rather than simply a mathematically correct fit. A way to
compensate for these complex interactions is to use magic angle spinning (MAS)
techniques.102-105
In an MAS experiment, the sample is spun about an axis inclined at an angle θR to
the applied field B0. The CSA contribution to spectral frequency, for axially symmetric
systems, is given by:
33
1
 ()

Equation 2.26: 
= −0 (
−  ) 2 (3 2  − 1)
while the heteronuclear dipolar coupling Hamiltonian is given by:

ℎ
̂
Equation 2.27: 
= − (40 )
  ħ
̂ ̂ (3 2 
3  
− 1)
When considering the relationship between θ and θR through the expression:
1
Equation 2.28: 〈3 2  − 1〉 = 2 (3 2  − 1)(3 2  − 1)
where β is the angle between the principle z-axis of the chemical shielding tensor and the
spinning axis, setting θR to 54.74o and spinning at a fast-enough rate brings the (3 2  −
1) term to zero, thus eliminating the effects of CSA and dipolar coupling in the MAS
spectrum.
When considering the quadrupolar contributions to the spectral signal, the V20
terms in the first- and second-order contributions will be averaged to zero. However, a
large enough quadrupolar interaction cannot be completely averaged out under fast MAS
(typically 20-30 kHz in the work performed here) thus resulting in the presence of spinning
sidebands. While the CT will not be effected by MAS to the first order, STs will be broken
into low intensity spinning sidebands. Second-order broadening effects can only be
partially reduced under MAS since the V40 term does not average to a nonzero value.
Figure 2.8 illustrates both the benefits and limitations of MAS alone. While a
spectrum for a 93Nb nucleus with CQ = 20 MHz (on the lower end of typical couplings seen
in this work) does exhibit substantial reduction in CSA contributions at the maximum
34
spinning speeds attainable with our instrumentation, they are not completely removed and
thus the spectrum is not highly resolved. Spinning at much higher speeds approaching 100
kHz does produce high resolution spectra, but unfortunately these spinning speeds are by
no means routine and in fact only a handful of solid-state NMR laboratories can perform
such experiments.106-109 Therefore, more specialized MAS based experiments, to be
discussed further, are required to accurately isolate and extract the different EFG
parameters without producing a mathematical fit that is not necessarily physically correct.
Figure 2.8. 93Nb static and MAS spectra at 9.4 T with varying spin rates (C Q = 20 MHz, η = 0.2, δiso =
0ppm, Ω = 500 ppm, κ = 0.5, α = β = γ = 0). Spectra were simulated using Simpson software.110
35
2.6 Wideline Uniform Rate Smooth Truncations (WURST) Pulses
The use of the VOCS method for obtaining wideline NMR spectra, while effective,
is substantially inefficient considering the time needed to collect a full spectrum. The
VOCS method can obtain slices of the full spectrum in increments of only 72.5 kHz
meaning a signal spanning 700 kHz (a range very common in the compounds studied in
this work) would require ten spectra be obtained. Further complicating matters, if the range
of the signal coverage is unknown (again, a very common occurrence in our compounds)
then slices extending substantially outside of the signal range are required to determine if
the full spectrum has been obtained. WURST broadband excitation pulses111-116 overcome
this limitation by drastically increasing the excitation bandwidth to the point that a 700 kHz
spectrum can be obtained in a single experiment. It should be noted that uniform excitation
is only when the excitation bandwidth is limited to 300 kHz using our instrumentation, but
the use of broader excitation profiles is still useful in determining the range of signal
present.
The time-dependent amplitude profile for WURST pulses is given by the equation:

Equation 2.29: 1 () =  (1−∣  ( ) ∣ )

Where ωmax is the maximum RF amplitude, τw is the pulse duration, and N determines the
extent to which the edges of the pulse are rounded off. As can be seen in Figure 2.9, higher
N values give more rectangular amplitude profiles.
36
Amplitude
WURST-N Amplitude Profiles
Time
WURST-2
WURST-4
WURST-40
WURST-80
Figure 2.9. WURST-N Amplitude profiles for WURST-2, WURST-4, WURST-40, and WURST-80
pulses.
Kupče and Freeman originally designed WURST pulses for adiabatic inversion of
magnetization.117 Adiabatic NMR experiments require pulses satisfying the adiabatic
condition (|ωeff(t)| >> |dα/dt|, where dα/dt is the instantaneous angular velocity) such that
magnetization will rotate at a constant flip angle, gradually following Beff during the
frequency sweep.118 It has been found that under nonadiabatic conditions comparable to
those employed in this work, WURST pulses can still be utilized in obtaining wideline
NMR spectra of quadrupolar nuclei.119 In the non-adiabatic regime, the WURST pulse can
be treated as a train of π/2 pulses whose individual frequencies vary linearly with time and
thus excite the signal with matching frequency, but having no effect on others. Signal
excitation then becomes time-dependent and the total spectrum will develop a frequencydependent phase difference, Φ(ν), given by the equation:119
3
)
2
Equation 2.30: () = (


− ( 2
) 2
37
Furthermore, since acquisition does not begin until the end of the pulse, signal excited at
the beginning of the pulse will be more attenuated due to dephasing and T2 relaxation than
signal excited at the end of the pulse as shown in Figure 2.10a. To overcome this problem
of phasing, application of a second WURST pulse to refocus magnetization can be used.
Assuming the refocusing pulse has an identical sweep range, Δ, the time-dependent phase
distribution then becomes:
Equation 2.31: (, ) = ( −

2

2
−  )  + (
−


) 2
where τd is the delay between pulses. Therefore, if τref is set such that 2τref = τexc then all
signals will refocus at the same point in time thus negating the second order phase
distortions (Figure 2.10b,c).
Figure 2.10. Schematic illustrations of WURST excitation and refocusing pulses reproduced from
Schurko et al.115 A - Single excitation pulse showing signal attenuation during the excitation pulse. B –
A refocusing pulse with 2τref = τexc leads to removal of the second-order phase distortion. C – A
refocusing pulse with τref = τexc does not eliminate second-order phase distortions as illustrated by the
displaced echoes shown in blue and red.
38
To further improve the efficiency, the WURST echo scheme was developed into
the WURST-QCPMG pulse sequence. As depicted in Figure 2.11, in the quadrupolar
Carr-Purcell-Meiboom-Gill (QCPMG120) pulse sequence, an off shoot of the CPMG
sequence,121 magnetization is refocused and detected N times in the cycle. Despite signal
attenuation due to relaxation as the train of echoes progresses, a substantial increase in
signal to noise is obtained. The use of WURST pulses in the CPMG cycle follows this
trend. The significant S/N improvements and greatly increased excitation bandwidth has
made WURST-QCPMG a desirable pulse sequence for use with quadrupolar nuclei.112-113
Figure 2.11. QCPMG Pulse Sequence.120
Traditionally, the processing of CPMG sequences produces a spikelet pattern after
Fourier transformation (as shown in Figure 2.12b) with equally spaced peaks whose
intensities resemble the conventional powder pattern. To assist with spectrum fitting and
analysis, co-addition of each echo in the CPMG train before Fourier transformation
produces a conventional powder pattern spectrum.
39
Figure 2.12. 91Zr NMR powder patterns obtained from a ZrO2 MAS rotor at 9.4 T using (a) WURST
echo and (b) WCPMG pulse sequences. Values shown on the right are the relative signal intensities,
normalized to account for the differing number of scans. The simulation in (c) includes central
transition lineshapes for the tetragonal (dotted) and orthorhombic (dashed) phases. Illustration was
obtained from O’Dell.116
Recently, the suite of WURST related pulse sequences has grown to include the
Broadband Adiabatic Inversion Cross-Polarization122 (BRAIN-CP) sequence shown in
Figure 2.13. Cross Polarization (CP) is used extensively in solid-state NMR as a method
40
of signal boosting where polarization is transferred from a higher frequency nucleus
(commonly 1H) to a lower frequency nucleus (such as
13
C).123-125 This transfer can be
achieved by spin locking the transverse 1H magnetization and irradiating the low frequency
nucleus during the contact period such that the Hartman-Hahn match,123 ν1(1H) = ν1(X), is
satisfied. Unfortunately, when dealing with large anisotropic interactions, CP using
traditional spin-locking fields is limited due to the bandwidth being determined by the
strength of the spin-locking fields and thus only a small fraction of the total powder pattern
is excited.
BRAIN-CP overcomes this excitation limitation using a standard pulse to maintain
the 1H spin lock, but with a WURST pulse on the X channel sweeping magnetization
adiabatically from orientation along the +Z axis to along the –Z axis over a wide
bandwidth. Under these sweeping conditions, each isochromat meets the Hartmann-Hahn
matching condition at different points in time during the sweep. CP can then be uniformly
performed over a broad range of frequencies with hardware considerations being the
limiting factor. The time interval during which CP occurs when ω1,I = ω1,S for each
isochromat can be estimated using the equation:
3
Equation 2.32:  = √
241,
2
where R is the sweep rate. Under a typical experiment with a 10 ms, 20 kHz WURST
pulse sweeping over 150 kHz, the approximate cross polarization time for each isochromat
comes to be 1.0 ms. A final π/2 pulse orients the magnetization into the x-y plane (a
necessary condition for signal observation).
41
Figure 2.13. BRAIN-CP pulse sequence.
2.7 Multiple-Quantum Magic Angle Spinning (MQMAS)
As discussed above, it is possible in principle to remove anisotropic broadening
under sufficiently fast MAS conditions, but in practice it is often impossible to reach speeds
sufficiently fast to completely remove such interactions. Take for instance a case where
the quadrupolar interaction is very strong (a recurring motif for samples studied in this
work). While MAS helps to narrow the lineshapes, the second order broadening is too
strong to be completely averaged to zero by MAS alone. Fortunately, this is by no means
an insurmountable obstacle. Various high-resolution MAS based NMR techniques have
been created to further reduce or separate anisotropic interactions. Dynamic Angle
Spinning (DAS126) and Double Rotation (DOR127) are two early developments, but both
are encumbered by mechanical complexities (see Figure 2.14). Both methods operate
through spinning about two different axes with DAS requiring switching of angles and
DOR requiring simultaneous spinning of two rotors positioned inside one another with
each spinning about a separate axis. The technique which has really taken off and become
42
a mainstay of solid-state NMR is MQMAS, first created in 1995 by Frydman and
Harwood.128
Figure 2.14. Schematic representation of the DOR sample setup requiring the spinning about two axes
simultaneously.
Under MAS conditions, the second-order quadrupolar frequency term is given by:
(2)
Equation 2.33:  (, ) = 0 0 () + 2 (, )2 ()2 () +
4 (, )4 ()4 ()
where {νQl}l=0,2,4 corresponds to the isotropic, second, and fourth-rank quadrupolar
contributions, respectively. {CIl(m)}l=0,2,4 are polynomials which depend on spin I and
coherence order m.
Equation 2.34: 0 () = 2[( + 1) − 32 ]
43
Equation 2.35: 2 () = 2[8( + 1) − 122 − 3]
Equation 2.36: 4 () = 2[18( + 1) − 342 − 5]
P2(cosβ) and P4(cosβ) terms are the second and fourth-rank Legèndre polynomials given
by
Equation 2.37: 2 () =
32 −1
2
Equation 2.38: 4 () =
354 −302 +3
8
At the magic angle, 3cos2β-1 equals zero, and thus the second-rank term disappears while
the fourth-rank term remains. The second-order quadrupolar shift for the central line in the
fast rotation regime is now given by:
Equation 2.39: 
(2)
11
− ,
22
= −
2
1
3
[
]
[(
60 2(2−1)
3
+ 1) − ]  [(, ) 4  +
4
(, ) 2  + (, )]
with
(, ) =
Equation 2.40:
21 7
7
− 2 + (2)2
16 8
48
9
1
7
(, ) = − 8 − 12  2 +   2 − 24 (2)2
(, ) =
5 1
7
− 2 + (2)2
16 8
48
44
Hence, MAS conditions alone only partially reduce the second-order quadrupolar
interaction.
Therefore, to accurately study nuclei containing large quadrupolar
interactions, other experiments such as MQMAS are employed.
The basic structure of the MQMAS experiment involves excitation of unobservable
multiple-quantum coherences (3Q, 5Q, 7Q, or 9Q for I = 9/2 spins) followed by conversion
to the observable single-quantum coherence which allows for processing of the resulting
NMR signals such that interactions can be separated. While a variety of MQMAS pulse
sequences such as two pulse,128 z-filter,129 split-t1,130 and soft pulse added mixing131
(SPAM) have been developed, the research discussed in this work utilizes the shiftedecho132-134 version of the 3QMAS pulse sequence. The 3QMAS pulse sequence is shown
in Figure 2.15, with an initial pulse generating the +3 coherence, followed by a FAM-II135137
train of conversion pulses generating the +1 coherence, and finally followed by a π pulse
to select the central transition thus generating a shifted echo such that the whole echo can
be collected.
45
Figure 2.15. Shifted-echo 3QMAS sequence with a FAM-II train of conversion pulses and the
corresponding coherence transfer pathway.
Processing of the 3QMAS data was performed using a double shearing method32
that successfully separates interactions such that only the quadrupolar chemical shift
(containing information about the EFG tensor) remains in the F1 dimension while the F2
dimension consists of both isotropic chemical shift and quadrupolar shift interactions. The
frequencies in each dimension following the double shearing can be described as:
Equation 2.41: 1 = 0 1 + 4 (, )4 ()41 (−1)
10
Equation 2.42: 2 = 0 − 27 〈1 〉 + 
where 0 is the isotropic chemical shielding, n is an integer, vR is the sample spinning
frequency, and
46
Equation 2.43: 0 = −
Equation 2.44: 1 =
2

(3+2 )
2
10 (2(2−1))
648

17
 = 9/2
and since 4 (, ) ≈ 0 the first moment of the powder pattern simplifies to:
Equation 2.45: 〈1 〉 = 0 1
The quadrupolar coupling (CQ) and asymmetry (η) parameters can then be readily extracted
from the resulting sheared spectrum. Due to this shearing, the dwell time in the indirect
1
dimension must be scaled by − (,) where (, ) = -72/34 for I=9/2 and the coherence
order p = +3
Fernandez et al.138 quite cleverly combined the MQMAS experiment with Sobserve, I-dephase Rotational Echo Double Resonance (REDOR), herein referred to as
MQMAS-t2-REDOR (Figure 2.16), to create a new tool for the study of internuclear
distances in complex systems. The REDOR sequence81-82 utilizes rotor-synchronized rf
pulses to reintroduce heteronuclear dipolar coupling between I spins and the observed S
spins resulting in signal attenuation for the observed S spins. As will be discussed further
in Chapter 5, this allows for easy assignment of 93Nb signals to either interior or exterior
NbO6 octahedra in the layered structure since the
93
Nb-1H internuclear distances for the
interior and exterior NbO6 octahedra are such that only signal due to exterior NbO6
octahedra will undergo observable attenuation.
47
Figure 2.16. MQMAS-t2-REDOR pulse sequence. The I and S channels refer to the 1H and
channels, respectively. An R3 recoupling sequence was used.139-140
93
Nb
As the following chapters will show, these SSNMR techniques have been
invaluable in the characterization of the layered materials studied in this work. Without
SSNMR, a substantial portion of the conclusions drawn from this work would lack the
substantial experimental evidence needed to be accepted as accurate and should serve as a
great example of the potential for using SSNMR in the study of layered perovskites.
48
Chapter 3: Experimental Overview
3.1 Materials
Raw materials, RbNO3, K2CO3, CaCO3, SrCO3, and Ba(NO3)2 salts were purchased
from Sigma-Aldrich and used without further purification.
RbCl and Nb2O5 were
purchased from Alfa Aesar and used without further purification. Alcohols were purchased
from Alfa Aesar and diluted with distilled water.
6M HNO3(aq) was diluted from
concentrated HNO3 using distilled water. Grafting reactions were performed using a Parr
23 mL microwave acid digestion bomb and a 1200 W domestic microwave oven.
3.2 Synthetic Procedures*
3.2.1 Three-layer Niobates Using MSS
RbSrxCa2-xNb3O10 and RbSrxBa2-xNb3O10 samples were synthesized using molten
salt heating31, 141-142 of the carbonate and oxide precursors in excess RbCl (the salt in molten
salt synthesis). K2CO3, SrCO3/CaCO3/Ba(NO3)2, and Nb2O5 were added in a 1.25:2:3 ratio
of K : Sr/Ca/Ba : Nb and ground together for five minutes before light mixing with 20 fold
molar excess RbCl. K2CO3 was added in a slight excess to account for its volatile nature
(or rather the volatile nature of K2O produced during the synthesis143). The samples were
then heated at 800-900 oC for 30-45 minutes followed by washing with warm distilled
water to remove the excess RbCl. The acid exchanged samples were obtained by mixing
the parent Rb forms with 6 M HNO3 for 3 days at 50 oC. As a rule of thumb, 1g of sample
is treated with ~75 mL of 6 M HNO3.
*The procedures presented in this chapter are generalized. Details for individual samples are presented in
subsequent chapters.
49
3.2.2 Three-layer Niobates Using Microwave Heating
Some RbSrxCa2-xNb3O10 and RbSrxBa2-xNb3O10 samples were prepared using
microwave synthesis.144 Stoichiometric amounts of Rb2CO3, CaCO3/SrCO3/Ba(NO3)2,
and Nb2O5 were ground together and placed in a thermally insulated box containing silicon
carbide microwave susceptors. The box was then heated using an 1100 W domestic
microwave oven for 5-10 min on high power followed by 10-15 minutes at 50% power to
reach an internal temperature of 1000-1150 oC. The samples were then washed with
distilled water to remove excess Rb2CO3 and dried at 110 oC. The usual acid exchange
procedure was employed to obtain the protonated forms.
3.2.3 Alkyl Grafting with Conventional Heating
C1/Ca2Nb3O10 was synthesized by mixing ~0.2 g HCa2Nb3O10 with 12.4 mL
methanol and 1.1 mL distilled water for a 90% w/w methanol solution. The sample was
then sealed in a general-purpose acid digestion bomb and heated at 100 oC for 3 days
followed by filtration and washing with acetone. C3/Ca2Nb3O10 was synthesized by
mixing the methoxylated sample with 9.1 mL n-propanol and 0.4 mL distilled water for a
95% w/w n-propanol solution followed by heating in an acid digestion bomb at 150 oC for
7 days followed by filtration and washing with acetone. C6/Ca2Nb3O10 was synthesized
by mixing the C3 sample with 10.0 mL n-hexanol followed by heating in an acid digestion
bomb at 150 oC for 7 days followed by filtration and washing with acetone.
50
3.2.4 Alkyl Grafting with Microwave Heating
C1/Sr2Nb3O10 was synthesized by mixing ~0.2 g HSr2Nb3O10 with 12.4 mL
methanol and 1.1 mL distilled water for a 90% w/w methanol solution. The sample was
then sealed in a microwave acid digestion bomb and heated in a domestic 1200W
microwave oven using a cycled heating scheme to prevent overheating of the acid digestion
bomb. The sample was heated at 100% power for 20s followed by a 40s cooling period
repeated five times in succession. The sample was left to cool for 15-20 min, then the
process was repeated a total of ten times. The sample was then filtered and washed with
acetone. C3/Sr2Nb3O10 was synthesized by mixing the methoxylated sample with 9.1 mL
n-propanol and 0.4 mL distilled water for a 95% w/w n-propanol solution followed by
heating at 100% power for 40s followed by a 20s cooling period repeated five times in
succession. The sample was left to cool for 15-20 min, then the process was repeated a
total of ten times. The sample was then filtered and washed with acetone. C6/Sr2Nb3O10
was synthesized by mixing the C3 sample with 10.0 mL n-hexanol followed by heating at
100% power for 60s followed by a 15s cooling period repeated four times in succession.
The sample was left to cool for 20-25 min, then the process was repeated a total of six
times. The sample was then filtered and washed with acetone. The microwave heating
reaction cycles are shown in Table 3.1.
51
Table 3.1. General grafting procedure for Ca xSr2-xNb3O10 compounds.
Alcohol
Heating Cycle (100% # of Cycles
Cooling Period
power)
Between Cycles
Methanol
5x (20s on 40s off)
10
15-20 min
(microwave)
Methanol
100o C 3 days
----(conventional)
Propanol
5x (40s on 20s off)
10
15-20 min
(microwave)
Propanol
150o C 7 days
----(conventional)
Hexanol (microwave) 4x (60s on 15s off)
76
20-25 min
Hexanol
150o C 7 days
----(conventional)
Total Reaction
Time
~4 hours
3 days
~4 hours
7 days
~3.5 hours
7 days
3.2.5 RbCa2NaNb4-xTaxO13 and HCa2NaNb4-xTaxO13
RbCa2Nb3O10 was synthesized using the molten salt technique described
previously. NaNbO3 and NaTaO3 were synthesized using the molten salt heating method
as well. Na2CO3 and Nb2O5 (for NaNbO3) or Ta2O5 (for NaTaO3) were ground together in
a 1:1 ratio for Na:Nb/Ta. The sample was then mixed with 10 molar excess of a
K2SO4/Na2SO4 1:1 mixture and heated at 825 oC for 15 minutes. It was then washed with
warm distilled water to remove excess K2SO4/Na2SO4. It should be noted that the 15minute reaction time was used to ensure small particle formation, which is essential to the
four-layer synthesis
RbCa2Nb3O10 and NaNbO3 or NaTaO3 were ground together for 5 minutes in a 1:3
molar ratio followed by rapidly heating in a thermally insulated box containing silicon
carbide microwave susceptors at 1200 oC for 20 minutes in a ThermWave microwave oven
with a platinum thermocouple from Research Microwave Systems, LLC. To obtain the
protonated form of the compound, the Rb parent compound was mixed with 6M HNO3 and
heated at 80 oC for 3 days with acid replacement every day.
52
3.2.6 Nanosheet synthesis using TBAOH
HCa2Nb3O10 and HSr2Nb3O10 exfoliation into nanosheets was performed using the
traditional method of intercalation of TBA+ using a TBAOH aqueous solution. Typically,
a 10:1 molar ratio of TBA+:Nb was employed using a 0.5 wt % TBAOH solution. The
mixture was then gently stirred at room temperature for 7 days. The mixture was then
centrifuged at 6000 rpm for 45 minutes to separate the unexfoliated sediment and the
colloidal suspension consisting of nanosheets. Adding a 0.1 M HNO3 solution dropwise
over the course of several hours until the pH of the solution reaches ~6-7 results in the
flocculation of nanosheets. Freezing of the flocculent followed by exposure to vacuum (~
0.1 Torr) generates solid high surface area nanosheets (though it should be noted the
presence of stacking reflections in the XRD powder patterns indicates some restacking has
occurred).
3.2.7 Nanosheet Synthesis Using Microwave Assisted Grafting
Approximately 0.1g of the alcohol grafted samples was mixed with either 2pentanone (2P) or 5-methyl-2-hexanone (5M2H in a glass vial and sonicated in a VWR
Symphony Ultrasonic Cleaner with an operating frequency of 35 kHz for four hours with
the exchange of the water in the sonication bath every 30 minutes to prevent overheating.
The colloidal suspension generated was then sealed in a microwave acid digestion bomb
and heated over a period of approximately four hours using cycled heating similar to that
employed in the grafting procedure with heating cycles of 10x[5x(45s on 15s off)] with 15-
53
20 minute cooling intervals between 5-minute heating cycles. Following the microwave
heating the sample was again placed in a glass vial and sonicated for four hours. The
solutions were then centrifuged at 6000 rpm for 45 minutes and the supernatants containing
the nanosheets were collected. Methods of collecting nanosheets out of solution will be
discussed in detail in Chapter 7.
3.3 Characterization Methods
3.3.1 Powder XRD
All XRD powder patterns were collected using a Bruker AXS D8 Focus
diffractometer, using a graphite monochromator and Cu Kα irradiation.
Data were
collected in a θ/2θ mode with a scanning 2θ range of 4-60 degrees or 2.5-14 degrees for
short range grafted samples where the first and second stacking reflections of the perovskite
were of greatest importance. Le Bail145 fits were performed on the XRD patterns using the
TOPAS146 software package from Bruker, and using a Chebychev polynomial background,
lattice parameters, particle size, zero corrections, χ2, and the background subtracted Rfactor were calculated.
3.3.2 ATR-IR
IR spectra were collected using a Perkin Elmer Infrared Spectrophotometer
(Spectrum 100) with an attached Pike Technologies GladiATR accessory featuring a
monolithic diamond crystal. Scans were performed from 450-4000 cm-1.
54
3.3.3 SEM and EDX
A table-top scanning electron microscope (SEM) TM-3000 coupled with an
electron dispersive spectrometer (EDX) was used for imaging and elemental analysis. 15
kV was used for all EDX measurement with 300 seconds used for signal averaging. Higher
resolution SEM images were obtained using a Tescan Vega-3 SEM instrument
3.3.4 UV-visible Diffuse Reflectance
UV-visible spectra were collected on a Perkin Elmer Lambda 35 UV/Vis
spectrophotometer with an attached LabSphere RSA-PE-20 diffuse reflectance and
transmittance accessory using a 160 nm/min scan rate from 200 – 1000 nm. Barium sulfate
was used for background collection.
3.3.5 Solid-State NMR Experiments
All 93Nb SSNR data were collected at magnetic field strengths of 9.4 and 14.1 T on
Varian INOVA spectrometers with resonance frequencies of 97.85 and 146.61 MHz,
respectively. All 1H,
23
Na, and
87
Rb NMR experiments were performed at 9.4 T with
resonance frequencies of 399.76, 104.61, and 67.96 MHz, respectively. 2.5 mm, 5.0mm,
and 7.0mm Chemagnetics double resonance MAS probes were used for experiments at 9.4
T. A 5 mm Varian triple resonance liquid state probe was used for all experiments on the
14.1 T instrument.
55
VOCS
VOCS experiments were performed using 15-21 Hahn Echo spectra with offset step
sizes of 78.125 and 62.5 kHz for the 9.4 and 14.1 T instruments, respectively. The π/2
pulse lengths were 2.4 and 4.0 μs at 9.4 and 14.1 T, respectively. The rf field strengths
were 20.8 and 12.5 kHz, respectively.
WURST
Experiments using WURST pulses were carried out at both 9.4 and 14.1 T fields.
Pulse parameters were optimized using a powder sample of RbSr2Nb3O10. The static 93Nb
NMR spectra for RbSr2Nb3O10 were found to have well defined features making it an ideal
test compound for the WURST pulse sequences through the comparison with VOCS
spectra. Using the intensity of one singularity in the RbSr2Nb3O10 spectrum and
incrementally changing the offset of the WURST pulse, the shape vs. offset profile was
optimized and it was found that an excitation bandwidth of 300 kHz gave an optimal
excitation profile. 50 μs pulses were used with 4 kHz and 8 kHz rf field strengths on the
initial excitation pulse and subsequent refocusing pulses, respectively.
Triple-Quantum Magic Angle Spinning (3QMAS)
3QMAS experiments used FAM-II conversion pulses under spinning rates of 25
kHz for 93Nb experiments, 20 kHz for 23Na experiments, and 10 kHz for 1H experiments.
A 1.7 μs π/2 initial excitation pulse was followed by the four-pulse FAM-II sequence with
pulse lengths of 0.6 μs at rf field strengths of 85 kHz. The final π pulse used to convert to
56
the -1 coherence order was 2.0 μs with an rf field strength of 50 kHz. Rotor synchronization
was applied on the anisotropic dimension with 8 to 16 data points depending on the
relaxation of each sample.
Multi-Nuclear Experiments
BRAIN-CP experiments were collected at 9.4 T with
93
Nb WURST pulses set to
an rf power of 3 kHz and a sweep width of 300 kHz and 1H pulses set to an rf power of
62.5 kHz. The contact time used in the CP experiments was 10.0 ms with a recycle delay
of 3.0 s. The MQMAS-t2-REDOR sequence utilized 93Nb rf field strengths of 85 kHz with
1
H RF power set to 80 kHz and an R3 recoupling sequence139-140 on the 1H channel.
NMR Data Processing and Spectral Fitting
NMR data were processed using RMN147 and fit using WSOLIDS,148 DMFIT,149
and QUADFIT.150 WURST-QCPMG spectra were obtained through the co-addition of all
individual echoes within the CPMG echo train to produce a single FID signal for
subsequent processing.
Whole-echo collection combined with magnitude spectral
transformation overcame the problem of second-order phase distortions. The SIMPSON
NMR110 simulation program was used to optimize several pulse sequences. The alternative
double shearing scheme described in Chapter 2 was used to process all MQMAS spectra.
57
Chapter 4. Exploration of Local Structural Changes in Layered
Niobates due to Compositional Changes at the A-Site.
4.1 Introduction
As the previous chapters have discussed, the composition of A’[An-1BnO3n+1]
layered perovskites leads to a myriad of different structures and properties. The work in
this chapter is focused on the effect of modifying the A site in RbA2Nb3O10 when A is a
mixture of alkaline earth metal pairs: Ca and Sr, or Ba and Sr. The structure and properties
of the end members RbCa2Nb3O1024,
29, 151-154
and RbSr2Nb3O1030,
151, 154-155
have been
extensively studied while the compound RbBa2Nb3O10 cannot be synthesized using readily
available synthetic routes as will be discussed later. As discussed in Chapter 3 and shown
by previous work within our research group,32, 98 NMR is an invaluable tool when studying
the structure of layered niobates as it provides information regarding local structure not
available with the conventional XRD structural characterization. In particular, the study
of the 93Nb nuclei allows for characterization of both exterior and interior NbO6 octahedra.
A thorough search of the literature has found only one study on the structure of mixed
cation RbCaxSr2-xNb3O10 compounds from Geselbracht et al.31 In that work, the presence
of a single phase in XRD patterns was taken as an indication of homogeneous distribution
of Sr and Ca within the perovskite structure. While the work presented here also indicated
long-range homogeneity based on XRD patterns, NMR spectra clearly show a
heterogeneity regarding the distribution of Ca and Sr within local structural ranges.
58
4.2 Experimental
RbSrxCa2-xNb3O10 and RbSrxBa2-xNb3O10 samples were synthesized using the
molten salt or microwave heating methods described in Chapter 3. Specific heating
temperatures and times for each compound are shown in Table 4.1 and 4.2. A comparison
of samples with identical composition, but with different synthetic techniques, showed no
significant changes in structure based on XRD patterns and NMR spectra. The only
noticeable difference between samples was particle size as observed in SEM images.
Samples synthesized with the molten salt method exhibited noticeably smaller particle
dimensions. The molten salt acts as a solvent-like medium during the reaction, thus
allowing for product formation under milder conditions due to the solubility of oxide
particles in the salt. Decreased particle size compared to dry synthesis may be the result
of more numerous points of nucleation growing simultaneously.
59
Table 4.1. Synthetic conditions for RbCaxSr2-xNb3O10 compounds using dry microwave heating or MSS.
60
Theor. Ca
Content
0
Exper. Ca
Content
0
0.25
0.4
0.5
0.9
0.75
0.7
1
1.1
1.25
1.3
1.5
1.5
1.75
1.9
2
2
0
0
Synthesis
Type
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Microwave
Dry
Molten Salt
0.5
0.5
Molten Salt
1
1.1
Molten Salt
1.5
1.7
Molten Salt
2
2
Molten Salt
Mass
RbCl
0
Mass
Rb2CO3
0.1913
Mass
K2CO3
0
Mass
CaCO3
0
Mass
SrCO3
0.3727
Mass
Nb2O5
0.491
Rb:K:Ca:Sr:Nb
Molar Ratio
1.25:0:0:2.03:2.97
0
0.1818
0
0.0322
0.3261
0.4994
0
0.185
0
0.0616
0.284
0.4991
0
0.1841
0
0.0949
0.2319
0.5038
0
0.1821
0
0.1248
0.1859
0.5058
0
0.1866
0
0.1619
0.1453
0.5158
0
0.1833
0
0.1893
0.0916
0.4982
0
0.185
0
0.2219
0.0461
0.5036
0
0.189
0
0.2496
0
0.491
1.25:0:0.26:1.75:2.
98
1.25:0:0.48:1.50:2.
93
1.25:0:0.74:1.23:2.
97
1.25:0:0.99:1.01:3.
02
1.25:0:1.25:0.76:3.
00
1.25:0:1.49:0.50:2.
95
1.25:0:1.73:0.22:2.
96
1.25:0:1.99:0:2.95
3.021
3
3.041
0
0.0851
0
0.3683
0.4978
20.6:1:0:1.96:2.95
0
0.0887
0.0628
0.285
0.5025
3.013
7
3.018
2
3.013
7
0
0.0852
0.1227
0.1843
0.5
0
0.0847
0.19
0.0906
0.5027
0
0.0865
0.2536
0
0.5008
19.6:1:0.49:1.50:2.
95
19.6:1:1.01:0.98:2.
99
20.4:1:1.55:0.50:3.
09
19.9:1:2.02:0:3.01
Heating Time
100% 8 min,
50% 12 min
100% 7 min,
50% 12 min
100% 6.5 min,
50% 12 min
100% 6.5 min,
50% 12 min
100% 6.5 min,
50% 12 min
100% 6.5 min,
50% 12 min
100% 5 min,
50% 11 min
100% 6 min,
50% 11 min
100% 7 min,
50% 5 min
45 min
Temp
(OC)
995*
1049*
972*
1031*
967*
1039*
952*
999*
918*
900
30 min
900
30 min
900
30 min
825
25 min
800
* Temperature was recorded shortly after removal from the microwave and thus reflects an approximation of reaction
temperature.
Table 4.2. Synthetic conditions for RbBaxSr2-xNb3O10 compounds using dry microwave synthesis.
Theor. Ba Experim. Mass (g) Mass (g) Mass (g) Mass (g)
Rb : Ba : Sr : Nb
Heating Time
content Ba Content Ba(NO3)2 SrCO3
Nb2O5 Rb2CO3
Molar Ratio
and Power %
8 min at 100%,
0
2
0
0.2223 0.3018
0.1761
1.01 : 0 : 2.00 : 3.02
12 min at 50%
8 min at 100%,
0.1
0.15
0.0196
0.2124
0.302
0.1802 1.03 : 0.09 : 1.91 : 3.02
12 min at 50%
8 min at 100%,
0.25
0.3
0.0488
0.1917 0.2998
0.174
1.00 : 0.25 : 1.73 : 3.00
12 min at 50%
8 min at 100%,
0.5
0.4
0.0991
0.1677 0.3006
0.1744 1.00 : 0.50 : 1.51 : 3.00
12 min at 50%
8 min at 100%,
0.75
0.6
0.1459
0.1399 0.2983
0.1771 1.02 : 0.74 : 1.26 : 2.98
12 min at 50%
4.3 XRD Results and Discussion
Figures 4.1-4.10 and 4.11-15 show XRD patterns for RbCaxSr2-xNb3O10 and
RbSrxBa2-xNb3O10, respectively. Le Bail fits were performed in the TOPAS program using
the Pnma space group and fit parameters are shown in Table 4.3 and 4.4. Lists of indexed
hkl reflections are provided in Appendix A. Increasing Sr content in the Ca/Sr series and
increasing Ba content in the Sr/Ba series results in expansion of the crystal lattice due to
increasing cation size. This is most clearly observed in Figures 4.16 and 4.17 showing
changes in unit cell volume and stacking distance (designated as the c-axis in the unit cell
dimensions), respectively, with changing A-site content. All of the patterns in the Ca/Sr
series showed the presence of a single phase, consistent with the work of Geselbracht et
al.31 and indicative of a homogeneous distribution of cations at the A-site. It should be
noted that the progression in unit cell dimensions in this work does not follow as uniform
a trend as that seen in Geselbracht’s work. Samples in the Ba/Sr series up to a Ba content
of 0.6 formula units also showed the presence of a single phase consistent with the threelayer structure. Ba content above 0.6 formula units, however, consistently exhibited
61
reflections in the XRD patterns that could not be accounted for using only the three-layer
structure. It is believed the larger Ba2+ ionic radius inhibits formation of a three-layer
structure without substantial Sr content acting to stabilize the structure. CsBa2Nb3O10,
however, can be synthesized successfully owing to the larger Cs+ ionic radius compared to
Rb+. The unaccounted for reflections in the XRD patterns can be fit using a unit cell
consistent with formation of a new phase BaNb2O6.156
62
Table 4.3. Lattice parameters for RbCaxSr2-xNb3O10 compounds obtained from Le Bail fitting using
TOPAS software.
Ca Content
Space Group
a(Å)
b(Å)
c(Å)
Unit
Cell
Volume (Å3)
0*
Pnma
7.763(1)
7.860(1)
30.267(3)
1847
0.4*
Pnma
7.852(1)
7.744(1)
29.822(7)
1813
0.7*
Pnma
7.826(1)
7.724(1)
29.986(4)
1813
0.9*
Pnma
7.713(2)
7.787(1)
30.166(4)
1812
1.3*
Pnma
7.720(3)
7.779(2)
29.95(1)
1799
1.4*
Pnma
7.790(1)
7.711(2)
29.81(1)
1791
1.9*
Pnma
7.712(1)
7.730(1)
29.78(1)
1775
2*
Pnma
7.690(2)
7.717(3)
29.832(3)
1770
2**
Pmna
7.722(1)
7.762(1)
29.768(4)
1784
0**
Pmna
7.765(1)
7.781(1)
30.419(5)
1838
0.5**
Pmna
7.843(2)
7.771(1)
30.190(3)
1840
1.1**
Pmna
7.813(3)
7.771(2)
30.094(8)
1827
1.7**
Pmna
7.783(1)
7.746(1)
29.685(4)
1790
* Samples synthesized using the microwave heating method
**Samples synthesized using the molten salt method
Table 4.4. Lattice constants for RbBaxSr2-xNb3O10 compounds.
X
Space Group
a (Å)
b (Å)
0
Pnma
7.763(1)
7.860(1)
0.15
Pnma
7.784(1)
7.863(2)
0.3
Pnma
7.882(4)
7.770(4)
0.5
Pnma
7.85(1)
7.80(1)
0.6
Pnma
7.843(3)
7.775(3)
c (Å)
30.267(3)
30.274(9)
30.292(8)
30.27(2)
30.557(9)
RbCa2Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.1. XRD pattern for RbCa2Nb3O10. Le Bail fit parameters are shown in Table 4.2.
63
60
RbCa1.9Sr0.1Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.2. XRD pattern for RbCa1.9Sr0.1Nb3O10. Le Bail fit parameters are shown in Table 4.2.
-
RbCa1.5Sr0.5Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.3. XRD pattern for RbCa1.5Sr0.5Nb3O10. Le Bail fit parameters are shown in Table 4.2.
64
60
RbCa1.4Sr0.6Nb3O10 MSS
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.4. XRD pattern for RbCa1.4Sr0.6Nb3O10. Le Bail fit parameters are shown in Table 4.2.
RbCa1.3Sr0.7Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.5. XRD pattern for RbCa1.3Sr0.7Nb3O10. Le Bail fit parameters are shown in Table 4.2.
65
60
RbCa1.1Sr0.9Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.6. XRD pattern for RbCa1.1Sr0.9Nb3O10. Le Bail fit parameters are shown in Table X.
RbCa0.9Sr1.1Nb3O10 MSS
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.7. XRD pattern for RbCa0.9Sr1.1Nb3O10. Le Bail fit parameters are shown in Table 4.2.
66
60
RbCa0.7Sr1.3Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.8. XRD pattern for RbCa0.7Sr1.3Nb3O10. Le Bail fit parameters are shown in Table 4.2.
RbCa0.4Sr1.6Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.9. XRD pattern for RbCa0.4Sr1.6Nb3O10. Le Bail fit parameters are shown in Table 4.2.
67
60
RbSr2Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.10. XRD pattern for RbSr2Nb3O10. Le Bail fit parameters are shown in Table 4.2.
RbBa0.15Sr1.85Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.11. XRD pattern for RbBa0.15Sr1.85Nb3O10. Le Bail fit parameters are shown in Table 4.3.
68
RbBa0.3Sr1.7Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
2θ (Degrees)
Figure 4.12. XRD pattern for RbBa0.3Sr1.7Nb3O10. Le Bail fit parameters are shown in Table 4.3.
RbBa0.4Sr1.6Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.13. XRD pattern for RbBa0.4Sr1.6Nb3O10. Le Bail fit parameters are shown in Table 4.3.
69
60
RbBa0.5Sr1.5Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
60
Figure 4.14. XRD pattern for RbBa0.5Sr1.5Nb3O10. Le Bail fit parameters are shown in Table 4.3.
RbBa0.6Sr1.4Nb3O10
4
8
12
16
20
24
28
32
36
40
44
48
52
56
2θ (Degrees)
Figure 4.15. XRD pattern for RbBa0.6Sr1.4Nb3O10. Le Bail fit parameters are shown in Table 4.3.
70
60
Unit Cell Volume vs. Ca content
1860
Unit Cell Volume ,(Å^3)
1850
1840
Microwave Samples
1830
Molten Salt Samples
1820
1810
1800
1790
1780
1770
1760
0
0.5
1
1.5
2
2.5
Ca content, formula units
Figure 4.16. Graph of Unit Cell Volume vs. Ca content in RbCa xSr2-xNb3O10 compounds calculated
using lattice parameters from Le Bail fits. Note the gradual decrease in unit cell volume with
increasing Ca content due to the smaller ionic radius of Ca 2+ compared to Sr2+.
c(Å) first stacking peak
Stacking Distance vs. Ca content
30.3
30.2
30.1
30
29.9
29.8
29.7
29.6
29.5
29.4
Microwave Samples
Molten Salt Samples
0
0.5
1
1.5
Experimental Ca content
2
2.5
Figure 4.17. Graph of stacking distance in RbCa xSr2-xNb3O10 compounds vs. Ca content. The stacking
distance was obtained using the position of the (002) reflection. Note the gradual decrease in stacking
distance with increasing Ca content due to the smaller ionic radius of Ca 2+ vs. Sr2+.
71
4.4 EDS and SEM Results and Discussion
EDS data for the Ca, Ba, and Nb contents shown in Tables 4.5 and 4.6 were used
to determine the composition of each sample in the Ca/Sr and Ba/Sr series.
The
composition was calculated using average ratios of Ca/Ba : Nb since Nb content within the
three-layer structure is constant across all samples. The measured Sr content was not used
due to overlap of Rb and Sr in EDS spectra (see Figure 4.18) leading to highly distorted
values. Errors for the calculated composition were determined using Fieller’s method157
of calculating confidence intervals for the ratio of two means. The experimental sample
compositions were in close alignment with theoretical compositions, though
experimentally calculated Ca content was consistently slightly higher than theoretical Ca
content. It is believed this is the result of an inherent error in atomic compositional
measurement on the instrument as evidenced by the experimental value of 2.1 for the pure
RbCa2Nb3O10 sample. This inherent imprecision was considered in subsequent analysis.
Table 4.5. Atom percent results for RbCaxSr2-xNb3O10 obtained using EDS. Ca content was estimated
using ratio of Nb:Ca due to clean separation of peaks in EDS spectra. Ca content error was calculated
using Fieller’s method of estimated confidence intervals between ratios.
Intended Ca Rb (%) error Ca (%) error Sr (%) error Nb (%) error Ca units error
2
7
1
5.6
0.4
0
0
8
1
2.1
0.3
1.75
6
1
7.4
0.4
2
1
12
2
1.9
0.3
1.5
5
0.9
5.6
0.3
3
1
11
2
1.5
0.3
1.25
7
1
4.9
0.3
4
2
11
2
1.3
0.3
1
4.7
0.8
2.1
0.2
6
2
10
2
0.7
0.2
0.75
5.1
0.9
2.9
0.2
7
2
12
2
0.7
0.2
0.5
4.3
0.8
1.7
0.2
5
2
10
2
0.9
0.2
0.25
2.2
0.9
0.6
0.1
5
2
5
1
0.4
0.1
0*
3.8
0.8
0
0
9
2
10
2
NA
NA
*Performing the same analysis using Sr content in place of Ca content leads to a Sr unit value of 2.7
(compared to theoretically 2.0) illustrating the distorting effect from the overlapping Rb signal in the EDS
spectra.
72
Table 4.6. Atom percent results for RbBaxSr2-xNb3O10 obtained using EDS. Ba content was estimated
using ratio of Nb:Ca due to clean separation of peaks in EDS spectra. Ba content error was calculated
using Fieller’s method of estimated confidence intervals between ratios.
Intended Ba Content
% Ba error % Nb
error % Sr error
Ba Units
error
0
0
0
8
1
6
1
0
0
0.1
0.2
0.1
4
1
3
1
0.15
0.08
0.25
0.8
0.2
7
1
5
2
0.3
0.1
0.4
0.5
0.2
4
1
3
1
0.4
0.1
0.5
1.8
0.3
11
2
6
2
0.5
0.1
0.75
1.2
0.3
6
1
3
1
0.6
0.2
cps/eV
RbCa1Sr1Nb3O10 1
1.8
1.6
1.4
1.2
1.0
C
Nb O
Sr
Rb
Nb
Ca
Rb
Sr
Ca
0.8
0.6
0.4
0.2
0.0
2
4
6
keV
8
10
12
14
Figure 4.18. EDS spectrum for RbCaSrNb3O10. Note the overlapping Rb and Sr signals.
SEM images in Figure 4.19 clearly show that the triple layer compounds form thin
platelets with varying lateral sizes up to several micrometers. This is consistent with
literature reports on the morphology of three-layered perovskites.158-159 Comparisons of
samples synthesized using dry microwave heating and those using molten salts show dry
73
Nb
samples exhibit particle sizes substantially larger than molten salt samples. Dry samples
produced particles up to ~5 μm while molten salt samples produced particle sizes up to ~23 μm. This is believed to be the result of increased particle diffusion within molten salts
which allow for reactants to come into contact with greater ease thus forming the desired
layered niobate, but concurrently inhibiting formation of large particles. SEM images of
attempted RbBaxSr2-xNb3O10 syntheses (Figure 4.20) show the presence of expected
platelets, but also rod-like impurities thought to be from formation of BaNb2O6 seen in
XRD patterns and consistent with SEM images for BaNb2O6 published previously.160
These impure samples were not used in subsequent analysis.
a)
b)
c)
d)
Figure 4.19. SEM images of RbCa2Nb3O10 synthesized under dry microwave (a) and molten salt (b)
conditions and RbSr2Nb3O10 synthesized under dry microwave (c) and molten salt (d) conditions. Note
the increased platelet size seen in dry microwave samples compared to molten salt samples.
74
Figure 4.20. SEM images for attempted RbBa 1.75Sr0.25Nb3O10 sample. Note the combination of
expected platelets and the presence of rod-like impurities in the left image resulting from formation of
the BaNb2O6 phase. The higher magnification image on the right shows the rod-like nature of the
impurity particles.
4.5 ATR-IR Results and Discussion
IR spectra gathered for the mixed A-site samples exhibit two absorption bands
characteristic of the perovskite structure. The strongest is a sharp peak in the 910-920 cm1
range due to the asymmetric stretching mode of the terminal Nb-O bond with double bond
character while the second peak occurs in the 690-770 cm-1 range and is associated with
the interior NbO6 octahedral Nb—O—Nb asymmetric stretching model.161-162
The
position of the interior NbO6 peak was found to be highly dependent on composition with
RbCa2Nb3O10 having a peak at 768.9 cm-1, RbSr2Nb3O10 at 738.4 cm-1, and
RbBa0.6Sr1.4Nb3O10 at 691.2 cm-1 while the peak at 910-920 cm-1 shows no significant shift.
Mixed cation samples show a near linear progression in peak position with increasing
cation size seen in both the Ca/Sr and Sr/Ba series of samples as seen in Figure 4.23.
Raman spectral analysis of various niobium oxides (including D-J KCa2Nb3O10 and
HCa2Nb3O10) from Jehng et al.153 clearly illustrated the correlation in stretching mode
75
wavenumber to NbO6 octahedral distortion with bands in the 600-800 cm-1 corresponding
to lightly distorted octahedra and bands in the 850-1000 cm-1 range corresponding to highly
distorted octahedra. Therefore, the IR results here indicate A-site modification does not
significantly alter the distortion of the exterior NbO6 octahedra, but increasing cation size
in the progression from Ca2+ to Sr2+ to Ba2+ leads to reduced octahedral distortions in the
interior NbO6 octahedra. This is further confirmed by the 93Nb NMR results to be discussed
in section 4.6.
Wavenumber, cm-1
RbCaxSr2-xNb3O10 IR vs. Ca Content
775
765
755
y = 14.586x + 741.25
745
735
0
0.5
1
1.5
2
2.5
Ca Content, formula units
Wavenumber, cm-1
RbBaxSr2-xNb3O10 IR vs Ba Content
750
740
y = -80.921x + 745.65
730
720
710
700
690
0
0.1
0.2
0.3
0.4
0.5
0.6
Ba Content, formula units
0.7
0.8
Figure 4.23. Graphs of IR peak position vs. Ca content (Top) and Ba content (Bottom).
76
0.9
4.6 NMR Results and Discussion
The results presented up to this point all indicate the distribution of cations in mixed
Ca/Sr and Ba/Sr compounds to be homogeneous with gradual progressions seen in XRD
and IR parameters. These characterization techniques, however, are lacking with respect
to characterizing disorder on the local scale. NMR is unique in its ability to characterize
local environments and, as will be shown, indicates an interesting dichotomy in which
despite evidence of global homogeneity, cation distribution on the local scale exhibits clear
heterogeneity. Figures 4.24-4.34 and 4.35-4.38 show the
93
Nb WURST-QCPMG static
spectra (at both 9.4 and 14.1 T) and MQMAS spectra, respectively, for samples in the Ca/Sr
and Ba/Sr series. Static NMR data were used as the primary approach for analyzing the
different 93Nb environments when such environments could be easily distinguished in the
spectra.
In cases where the spectral resolution needed for distinguishing multiple
environments was not present, 3QMAS was employed as a supplementary technique.
77
Figure 4.24. 93Nb Static WURST spectra for RbSr2Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom). Blue Experimental Spectrum; Black - Total Fit; Green - Exterior Site ;and Red - Interior Site.
78
Figure 4.25. 93Nb Static WURST spectra for RbCa0.4Sr1.6Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
79
Figure 4.26. 93Nb Static WURST spectra for RbCa0.5Sr1.5Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); Yellow – Impurity; and Red - Interior Site.
80
Figure 4.27. 93Nb Static WURST spectra for RbCa0.7Sr1.3Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
81
Figure 4.28. 93Nb Static WURST spectra for RbCa0.9Sr1.1Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
82
Figure 4.29. 93Nb Static WURST spectra for RbCa1.1Sr0.9Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
83
Figure 4.30. 93Nb Static WURST spectra for RbCa1.4Sr0.6Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
84
Figure 4.31. 93Nb Static WURST spectra for RbCa1.5Sr0.5Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
85
Figure 4.32. 93Nb Static WURST spectra for RbCa1.7Sr0.3Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
Blue - Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior
Site 2 (Ca-like); and Red - Interior Site.
86
Figure 4.33. 93Nb Static WURST spectra for RbCa1.9Sr0.1Nb3O10 at 9.4 T (left) and 14.1 T (right). Blue
- Experimental Spectrum; Black - Total Fit; Green - Exterior Site 1 (Sr-like); Purple - Exterior Site
2 (Ca-like); and Red - Interior Site.
87
Figure 4.34. 93Nb Static WURST spectra for RbCa2Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom). Blue Experimental Spectrum; Black - Total Fit; Purple - Exterior Site and Red - Interior Site.
The double shearing transformation used in our MQMAS processing allows for
the separation of signals based on CQ values and was thus used to confirm the presence of
multiple unique environments associated with exterior Nb octahedra. First moment
analysis was employed to extract CQ values for several samples, as shown in Table 4.7.
These values were consistent with those obtained from static fits of spectra with better
88
resolved features. Another useful feature of the obtained MQMAS spectra is the
consistent slope and width in the 93Nb signals. The observed slope is indicative of a
distribution of CQ, consistent with both the disordered nature of these materials and the
static NMR spectra. Using this combination of NMR experiments, fits were obtained
leading to unique observations regarding the structure of mixed cation perovskites.
MQMAS lineshape analysis was employed to determine CQ and η, then subsequently
used to extract appropriate static 93Nb fit parameters in Table 4.8.
Table 4.7. First moment analysis for RbCa xSr2-xNb3O10 compounds. The obtained CQ and η values
were calculated using Equation 2.42-45.
Sample
CQ, MHz
η
<v1>, ppm
RbCa0.9Sr1.1Nb3O10
29.2
0.32
-197
RbCa0.7Sr1.3Nb3O10
30.5
0.34
-215
44.4
0.06
-382
RbCa0.5Sr1.5Nb3O10
28
0.24
-181
RbCa0.4Sr1.6Nb3O10
35.9
0.21
-297
44.9
0.01
-388
89
Figure 4.35. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.4Sr1.6Nb3O10 with double shearing
scheme. The horizontal axis corresponds to the isotropic dimension and vertical axis to the anisotropic
dimension. Note the observed signal positions along F2 are not equal to the isotropic chemical shifts
in static spectra, as illustrated in Equation 2.42.
Figure 4.36. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.5Sr1.5Nb3O10 with double shearing
scheme. The horizontal axis corresponds to the isotropic dimension and vertical axis to the anisotropic
dimension. Note the observed signal positions along F2 are not equal to the isotropic chemical shifts in
static spectra, as illustrated in Equation 2.42.
90
Figure 4.37. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.7Sr1.3Nb3O10 with double shearing
scheme. The horizontal axis corresponds to the isotropic dimension and vertical axis to the anisotropic
dimension. Note the observed signal positions along F2 are not equal to the isotropic chemical shifts in
static spectra, as illustrated in Equation 2.42.
Figure 4.38. Contour plot of 93Nb 3QMAS 2D spectrum of RbCa0.9Sr1.1Nb3O10 with double shearing
scheme. The horizontal axis corresponds to the isotropic dimension and vertical axis to the anisotropic
dimension. Note the observed signal positions along F2 are not equal to the isotropic chemical shifts in
static spectra, as illustrated in Equation 2.42.
91
Assuming the homogeneous cation distribution observed in long-range and bulk
properties, it would be expected that the 93Nb spectra would contain one signal from the
exterior NbO6 octahedra and a second from the interior NbO6 octahedra with a gradual
change in CQ values as composition changes. However, as mentioned previously, this is
not the case. Rather, spectra consistently exhibit multiple signals associated with exterior
octahedra while there is one identifiable signal associated with interior octahedra. It should
be noted that the observation of a single interior 93Nb signal does not negate the absence
of a second. The substantially larger CQ value in the interior site results in signal intensity
being spread over a much larger range such that low population signals may not be readily
distinguished.
Another consequence of the low intensity signal arising from the interior
environments is a lack of signal to noise necessary for more in-depth analysis such as that
which will be presented for the exterior environments.
92
93
Table 4.9. RbCaxSr2-xNb3O10 NMR Fit parameters. Errors in the last digit are shown in parentheses and were calculated for population and CQ
values using chi-square minimization.
X
Site
Pop. (%)
δiso (ppm)
CQ (MHz)
η
Ω (ppm)
κ
α (Deg)
β (Deg)
γ (Deg)
0
Ext 1
62.8(3)
-963
45.1(4)
0.04
831
0.83
0
0
0
Int
37.2(4)
-1043
93(1)
0.01
490
0.97
0
0
0
0.4
Ext 1
27(1)
-953
35.9(8)
0.21
1091
0.65
44.1
9
24.9
Ext 2
41(1)
-955
44.9(7)
0.01
851
0.9
0
0
0
Int
33(2)
-1051
92(2)
0.04
875
0.53
38.7
6.3
33.9
0.5
Ext 1
33(3)
-931
32(1)
0.27
1118
0.59
44.8
1.9
7.9
Ext 2
4.2(7)
-954
28(2)
0.24
1105
0.7
92.2
12.9
16.3
Ext 3
35(1)
-955
44.5(6)
0.07
852
0.91
0
0
0
Int
27.2(9)
-1068
93(3)
0.06
984
0.27
-14.6
15.9
62.1
0.7
Ext 1
24.5(8)
-976
30.5(7)
0.34
1114
0.6
100
7.2
87.4
Ext 2
50.5(5)
-950
44.4(5)
0.06
849
0.92
0
0
0
Int
25(4)
-1068
92(2)
0.06
978
0.29
72.8
15.9
61
0.9
Ext 1
21(3)
-981
29.2(4)
0.32
956
0.51
44.6
3.64
19.1
Ext 2
45(2)
-953
45(1)
0.06
817
0.91
31.7
0.73
45.3
Int
33(2)
-1069
93(2)
0.06
544
0.87
0
0
0
1.1
Ext 1
60(1)
-954
33(1)
0.27
978
0.51
-1.7
3.3
15.2
Ext 2
18.6(7)
-958
44(1)
0.12
861
0.87
51
5.2
-8
Int
21(5)
-1071
105(3)
0.18
683
0.73
42.4
17
41.1
1.4
Ext 1
40(2)
-937
32.6(7)
0.29
931
0.51
20.9
2
26.4
Ext 2
38(2)
-905
44(1)
0.18
870
0.82
10.3
0.3
16
Int
21(7)
-1071
104(4)
0.19
757
0.61
14.3
14.7
48.8
1.5
Ext 1
49(1)
-962
33.7(5)
0.26
872
0.63
0
2
26.4
Ext 2
29.2(9)
-953
43.5(6)
0.18
765
0.81
10.3
0.3
16.1
Int
21(6)
-1034
111(2)
0.16
816
0.4
13.9
13.9
54.3
1.7
Ext 1
53(2)
-974
31(1)
0.28
904
0.67
33.1
2.2
44.5
Ext 2
15(4)
-926
44.5(4)
0.06
842
0.92
12.2
0.2
10
Int
33(3)
-940
111(5)
0.17
905
0.1
1.1
16.4
43.5
1.9
Ext 1
64(1)
-981
30.1(9)
0.21
892
0.74
0.9
1
10.6
Int
35(2)
-992
110(2)
0.17
795
0.66
14.5
7.1
44.1
2
Ext 1
59(1)
-987
28.8(7)
0.19
910
0.64
45
4.6
27
Int
40(3)
-1102
109(4)
0.11
1060
0.86
0
0
0
For easier discussion and analysis of the numerous parameters for the mixed Ca/Sr
series, Figures 4.39-4.43 show graphs of Ca content vs CQ, Population, Ω, η, and κ values,
respectively.
93Nb
CQ vs. Ca Content
50
115
40
105
35
95
30
85
25
Interior CQ, MHz
Exterior CQ, MHz
125
45
75
0
0.5
1
1.5
2
Ca Content, formula units
Exterior 1
Exterior 2
Interior
Figure 4.39. Graph of 93Nb CQ (in MHz) vs Ca content for Exterior Site 1 (blue), Exterior Site 2
(orange), and Interior Site (grey).
94
Site Population vs. Ca Content
70
Population, %
60
50
40
30
20
10
0
0
0.5
1
1.5
2
Ca Content, formula units
Exterior 1
Exterior 2
Interior
Figure 4.40. Graph of 93Nb Population vs Ca content for Exterior Site 1 (blue), Exterior Site 2 (orange),
and Interior Site (grey).
CSA Span vs. Ca Content
CSA Span (Ω), ppm
1200
1000
800
600
400
200
0
0
0.5
1
1.5
2
Ca Content, formula units
Exterior 1
Exterior 2
Interior
Figure 4.41. Graph of 93Nb CSA span (Ω, ppm) vs Ca content for Exterior Site 1 (blue), Exterior Site
2 (orange), and Interior Site (grey).
95
Asymmetry Parameter vs. Ca Content
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0
0.5
1
1.5
2
Ca Content, formula units
Exterior 1
Exterior 2
Interior
Figure 4.42. Graph of 93Nb asymmetry parameter (η) value vs Ca content for Exterior Site 1 (blue),
Exterior Site 2 (orange), and Interior Site (grey).
Skew vs. Ca Content
1.2
Skew (κ) value
1
0.8
0.6
0.4
0.2
0
0
0.5
1
1.5
2
Ca Content, formula units
Exterior 1
Exterior 2
Interior
Figure 4.43. Graph of 93Nb skew (κ) value vs Ca content for Exterior Site 1 (blue), Exterior Site 2
(orange), and Interior Site (grey).
96
Fits of exterior site signals in the Ca/Sr series consistently show one environment
with a CQ value close to that of the pure Ca form (~28 MHz) and a second environment
with a CQ value close to that of pure Sr form (~45 MHz). The Ca-like environment does
see significant fluctuations in CQ value going as high as 35.9 MHz in the Ca0.4 sample and
consistently being larger than that seen in the pure Ca form. The Sr-like environment,
however, sees less than 5% fluctuation in CQ values up to Ca1.9 (at which point no Sr-like
environment can be distinguished). Fits of the interior 93Nb signals, while being assigned
to single 93Nb environments, still do not exhibit the same gradual changes observed in longrange measurements, but instead can be treated as either a Ca-like or Sr-like environment.
Interestingly, despite clear evidence of heterogenous distribution of A-site cations,
populations of the observed environments do not correlate to the composition determined
from EDS data suggesting that the A-site cation effects the EFG tensor of octahedra beyond
those directly adjacent to that particular A-site. This is illustrated in Figure 4.40 showing
a comparison of site populations for varying Ca contents. Looking at the exterior sites,
while there is a general trend where the population of Ca-like environments increases with
increasing Ca content, individual sample analysis does not correspond to composition
exactly. For instance, RbCa0.5Sr1.5Nb3O10 would theoretically have two exterior sites with
a population ratio of 1:3 for Ca-like:Sr-like, but experimentally this ratio is approximately
1:1.
The degree of octahedral tilting in both the xy plane and along the z-axis for the
interior NbO6 octahedra is correlated to CQ values with larger tilting leading to larger EFG
97
values. Thus Ca-like environments correspond to more highly distorted octahedra than Srlike environments. Considering the smaller ionic radius of Ca2+, tilting of octahedra is
expected in order for Ca2+ to coordinate stably to interior octahedral O2-. EFG values in
the exterior site are instead the result of out of center displacement of the Nb atom in the
octahedra. Considering the double bond character of the terminal Nb=O bond, this is to be
expected. As evidenced by the lack of shifting in the terminal Nb=O IR absorption band,
A-site does not lead to significant changes in Nb=O bond length. Changing EFG values
are then attributed to shifting of the equatorial oxygen atom position and thus less out center
displacement of the Nb atom with decreasing A-site cation size. Spectra collected for
samples in the Ba/Sr series were analyzed in a similar fashion to the Ca/Sr series.
98
Figure 4.44. 93Nb static WURST spectra for RbBa0.15Sr1.85Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
See Table 4.8 for detailed fit parameters. Experimental spectrum is shown in blue and overall fit is
shown in red.
99
Figure 4.45. 93Nb static WURST spectra for RbBa0.3Sr1.7Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
See Table 4.8 for detailed fit parameters. Experimental spectrum is shown in blue and overall fit is
shown in red.
100
Figure 4.46. 93Nb static WURST spectra for RbBa0.4Sr1.6Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
See Table 4.8 for detailed fit parameters. Experimental spectrum is shown in blue and overall fit is
shown in red.
101
Figure 4.47. 93Nb static WURST spectra for RbBa0.6Sr1.4Nb3O10 at 9.4 T (Top) and 14.1 T (Bottom).
See Table 4.8 for detailed fit parameters. Experimental spectrum is shown in blue and overall fit is
shown in red.
102
Table 4.10. RbBaxSr2-xNb3O10 93Nb Static NMR Fit parameters. Errors in the last digit are shown in parentheses and were calculated for
population and CQ values using chi-square minimization.
X
Site
Pop. (%) δiso (ppm)
CQ (MHz)
η
Ω (ppm)
κ
α (Deg)
β (Deg)
γ (Deg)
0
0.15
0.3
103
0.4
0.6
Ext 1
Int
Ext 1
Ext 2
Int 1
Int 2
Impurity
Ext 1
Ext 2
Int 1
Int 2
Impurity
Ext 1
Ext 2
Int 1
Int 2
Impurity
Ext 1
Ext 2
Int 1
Int 2
Impurity
62.8(3)
37.2(4)
61(2)
8.8(6)
22(3)
5.4(9)
2(1)
50(3)
17(1)
16(1)
10(1)
6.9(7)
49(4)
18(2)
16(3)
11(1)
5.3(5)
39(2)
31(3)
5.8(6)
19(1)
5.8(7)
-963
-1043
-959
-970
-1045
-1051
-945
-960
-965
-1056
-1068
-969
-958
-950
-1060
-1066
-980
-960
-945
-1075
-1039
-963
45.1(4)
93(1)
45.7(9)
53(1)
92(3)
90(2)
21(1)
45(2)
55(2)
93(3)
87(2)
22(1)
45.3(8)
53.5(9)
93(4)
88(3)
20(1)
43(1)
55(1)
93(3)
85(2)
21.4(8)
0.04
0.01
0.08
0.14
0.05
0.02
0.8
0.14
0.17
0.06
0.05
0.83
0.19
0.11
0.06
0.03
0.82
0.21
0.2
0.07
0.09
0.85
831
490
800
900
586
350
110
795
905
560
410
110
805
890
600
445
110
810
880
560
500
120
0.83
0.97
0.86
0.92
0.46
0.79
0.08
0.88
0.92
0.54
0.70
0.08
0.90
0.90
0.62
0.72
0.16
0.86
0.80
0.84
0.56
0.13
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
As seen in Figures 4.44-4.47 and Table 4.8, RbBaxSr2-xNb3O10 93Nb spectra and the
corresponding fit spectra indicate a clear heterogeneity between Sr-like environments and
Ba-like environments. This is further illustrated in graphs of CQ, population, CSA, skew,
and asymmetry parameter vs. Ba content in Figures 4.48-4.52, respectively. Using work
regarding changes in the 93Nb EFG of CsSr2Nb3O10 and CsBa2Nb3O10, the samples in this
series were fit using five total 93Nb environments.98 The sites designated Ext 1 and Ext 2
in Table 4.8 correspond to Sr-like and Ba-like environments, respectively, with CQ values
typically 44-46 MHz in Ext 1 and 53-55 MHz in Ext 2. As discussed in the Ca/Sr analysis,
this increase in CQ corresponds to decreasing degrees of octahedral tilting and Nb
displacement in the exterior octahedra. Unlike the exterior octahedra, decreasing degrees
of octahedral tilting in interior octahedra results in decreasing CQ values. The sites
designated Int 1 and Int 2 in Table 4.9 correspond to Sr-like and Ba-like environments,
respectively, with CQ values typically 92-94 MHz in Int 1 and 86-88 MHz in Int 2
indicating the Int 2 octahedra have decreased octahedral tilting compared to Int 1.
It
should be noted these ranges are larger than those seen in pure Sr or Ba compounds, and is
a result of the disordered nature of cation distribution. The substantially smaller CQ value
of the interior Ba-like octahedra compared to Ca-like octahedra lead to greater signal
intensity and thus the two interior sites can be clearly identified in the Ba/Sr series, but not
in the Ca/Sr series. The final site included in the calculated spectra has been designated an
impurity due to its low CQ (~20 MHz), CSA (~110 ppm), and population (~5%) being
inconsistent with the layered perovskite structure. This impurity could possibly be due to
104
a BaNb2O6 phase. This phase is seen in XRD patterns for attempted syntheses of higher
Ba content samples, and if it is amorphous and/or present in small quantities in low Ba
content samples it would not be observed in XRD, but would be present in the 93Nb spectra.
93Nb
CQ vs. Ba Content
100
CQ, MHz
90
80
Ext 1
70
Ext 2
60
Int 1
50
Int 2
40
0
0.1
0.2
0.3
0.4
0.5
0.6
Ba Content, formula units
0.7
0.8
Figure 4.48. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2 (orange),
Interior Site 1 (grey), and Interior Site 2 (yellow).
105
Site Population vs. Ba Content
70
Population, %
60
50
40
Ext 1
30
Ext 2
20
Int 1
10
Int 2
0
0
0.1
0.2
0.3
0.4
0.5
0.6
Ba Content, formula units
0.7
0.8
Figure 4.49. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2 (orange),
Interior Site 1 (grey), and Interior Site 2 (yellow).
CSA Span vs. Ba Content
1000
CSA Span (Ω), ppm
900
800
700
Ext 1
600
Ext 2
Int 1
500
Int 2
400
300
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Ba Content, formula units
Figure 4.50. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2 (orange),
Interior Site 1 (grey), and Interior Site 2 (yellow).
106
Skew vs. Ba Content
1
Skew (κ) value
0.9
0.8
Ext 1
0.7
Ext 2
0.6
Int 1
0.5
Int 2
0.4
0
0.1
0.2
0.3
0.4
0.5
0.6
Ba Content, formula units
0.7
0.8
Figure 4.51. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2 (orange),
Interior Site 1 (grey), and Interior Site 2 (yellow).
Asymmetry Parameter (η) value
Asymmetry Parameter vs. Ba Content
0.25
0.2
0.15
Ext 1
Ext 2
0.1
Int 1
0.05
Int 2
0
0
0.1
0.2
0.3
0.4
0.5
0.6
Ba Content, formula units
0.7
0.8
Figure 4.52. Graph of 93Nb CQ vs. Ba content for Exterior Site 1 (blue), Exterior Site 2 (orange),
Interior Site 1 (grey), and Interior Site 2 (yellow).
107
As will be discussed in Chapter 5, heterogeneous cation distribution can often be
attributed to preference for unique crystallographic sites. In the three-layer structure
studied here, though, there is only one unique crystallographic A-site and thus
heterogeneity observed on the local scale cannot be explained based on preference on the
global scale. Considering the rapid nature of these compounds’ syntheses, it is possible
that longer reaction time such as that used in conventional methods would allow for greater
diffusion of cations and thus heterogeneity would not be observed. While this possibility
most certainly deserves further study, the following chapters will show local-scale
heterogeneity to be a potentially huge advantage in the compositional design of layered
perovskites.
As mentioned previously, the resolution of
93
Nb signals arising from interior
octahedra is not at a level necessary to easily distinguish the effects of cation distribution.
Therefore, while this analysis is primarily focused on cation distribution in terms of
exterior octahedral effects, the relative magnitude of effects on exterior and interior sites
cannot be accurately compared using current data. Considering Figures 4.39-4.43, clear
heterogeneity is observed most readily in analysis of CQ and η values. Fluctuations in
various fit parameter values may be the result of WURST pulse inhomogeneity, disordered
nature of the compounds, and rapid synthetic methods.
Based on the structural
relationships with these parameters discussed in Chapter 2, it is expected they would
exhibit the most readily distinguished effects of heterogeneous distribution as they are
related to both symmetry about the nucleus and internuclear distances. The presence of
108
both Sr-like and Ca-like exterior
93
Nb environments suggests there is a heterogeneous
distribution of exterior octahedral distortions, with localized areas of low-distortion (Srlike) and other areas with high-distortion (Ca-like). As Chapter 6 will discuss in further
detail, less distorted octahedra lead to weakening of O-H bonds in acid exchanged samples
(i.e. more reactive acid sites). Mixed cation compounds then can exhibit substantially
increased reactivity from these localized regions without altering the majority of the
structure.
4.7 Conclusions
RbSrxCa2-xNb3O10
and
RbBaxSr2-xNb3O10
compounds
were
successfully
synthesized and analyzed using XRD, SEM, EDS, ATR-IR, and NMR. XRD patterns
confirmed formation of the triple layered compounds while yielding lattice parameters
consistent with lattice expansion due to increased cation size. SEM showed the desired
platelet formation with varying uniformity and size depending on the synthetic route while
EDS data were used to calculate the Ca, Sr, or Ba content in each sample. IR spectra
showed an absorption band that shifts linearly with increasing Sr content in the Ca/Sr series
and increasing Ba content in the Ba/Sr series. This band is likely due to changes regarding
the interior Nb octahedra.
NMR experiments utilizing the WURST-QCPMG pulse
sequence clearly show the presence of uniquely Ca-like and Sr-like environments in the
Ca/Sr series, and uniquely Ba-like and Sr-like environments in the Ba/Sr series.
93
Nb NMR
analysis clearly indicates heterogeneity on the local-scale while bulk properties can be
109
attributed to global-scale homogeneity. As the following chapters will show, local-scale
heterogeneous cation distribution is potentially a huge advantage in the design of new
compounds.
110
Chapter 5: Determination of Cation Site Ordering in Fourlayer HCa2NaNb4-xTaxO13 Compounds Using Solid State 93Nb
and 23Na NMR
5.1 Introduction
Building off the success of studying cation distribution in mixed A-site DionJacobson niobates, our attention then turned towards the study of cation distribution in
compounds with both mixed A- and B-sites, specifically HCa2NaNb4-xTaxO13. Unlike the
Rb(Ca/Ba)xSr2-xNb3O10 compounds investigated in Chapter 4, there are distinct
crystallographic A- and B-sites in HCa2NaNb4-xTaxO13 that can be accurately studied using
XRD. Unfortunately, due to the rapid synthetic procedure employed and inherent disorder
of these samples, conventional PXRD is insufficient and only a synchrotron source would
provide the signal to noise necessary for Rietveld refinement.163. To further reiterate, NMR
is ideal for disordered systems and was thus employed to determine site-ordering through
observations of 93Nb and 23Na nuclei.
It has been shown that changing the B-site atom can drastically alter the
intercalation behavior and catalytic ability of these materials.31, 55-56 Geselbracht et al.31
studied the acidity of HCa2NbxTa3-xO10 compounds through intercalation and found that
the pyridine intercalation reaction did not work for x=2.5 and 3.0 compounds suggesting a
significant decrease in solid acid strength. For x=0.5-2.0, however, the level of pyridine
intercalation was greater than in the x=0 compound suggesting the ordering of Ta5+ within
the structure has an effect as to whether the solid acid strength is decreased or increased.
111
In work on the water splitting ability of Sr2Ta2O7 and Sr2Nb2O7, Kudo et al.57 found the
water splitting activity and band gaps of the SrTa2O7 and SrNb2O7 compounds to be
significantly different (4.6 eV vs. 3.9 eV, respectively).
Subsequent study of
Sr2(Ta1-xNbx)2O7 compounds found photocatalytic activity does not progress linearly with
changing composition.164
While Nb5+ and Ta5+ possess the same charge and approximately equal ionic radii58
(0.64 Ǻ), Nb is more electronegative than Ta (1.6 vs. 1.5 PU, respectively)59 leading to
Nb-O bonds more covalent in character than Ta-O bonds and changes in the energy states
of Ta5+ and Nb5+ compounds.57, 60 Differences in both charge and ionic radius have been
shown to be strong drivers of site ordering in mixed cation compound such as mixed
Nb5+/Ti4+ layered perovskites.61-62 In mixed Nb5+/Ta5+ systems165-166, however, ionic
character has been shown to be the driver of site preference, with the Ta5+ cation occupying
more symmetrical sites due to the more ionic character of Ta-O bonds.
112
Figure 5.1. Tetragonal crystal structure for RbCa2NaNb4O13
In the present study, we have attempted to determine any site preference of Ta in
four layer Rb/HCa2NaNb4-xTaxO13 systems using solid-state NMR techniques.
As
illustrated in Figure 5.1, there are two distinct crystallographic B-sites in the four-layer
compound, two interfacial octahedra, Nb(1), and two interior octahedra, Nb(2). X-ray
powder diffraction is unable to give quantitative results as to the populations of Nb5+ and
Ta5+ at each crystallographic site due to the disordered nature of these compounds. NMR,
however, allows for the study of the local structure and is ideal for systems of this sort. To
the best of our knowledge, this is the first reported evidence of site ordering in mixed
Nb5+/Ta5+ D-J perovskites.
Previous reports of the crystal structure of RbCa2NaNb4O1335, 167 suggested Na and
Ca were randomly distributed throughout the perovskite slabs, but to our knowledge no
research has been performed to confirm this. It has been shown that the composition of
113
the alkali cations in layered perovskites has a significant effect on structure and
properties.63 It is logical to assume that distribution of cations in the four-layer systems
would also influence structure and properties. Determining whether the cations are
randomly distributed or if some site preference exists is important information for any
attempt at predicting structure and properties based on composition.
To study the
distribution of interstitial cations in the compounds, we performed 23Na NMR experiments
to extract chemical shift and EFG data, and analyzed relative populations to determine
distribution. This new information regarding the distribution of A and B site cations should
help in predicting the behaviors of mixed metal layered perovskites, especially regarding
solid acid behavior.
As this report will show, solid state NMR serves a vital purpose in the determination
of cation ordering within mixed cation compounds. By obtaining static spectra at multiple
fields as well as using Multiple Quantum Magic Angle Spinning (MQMAS) experiments
we could accurately determine the different EFG environments in the compound. Crosspolarization (CP) and recoupling experiments allowed for the linking of the EFG
environment to the crystallographic position of the Nb octahedra. Using the relative
populations of the static spectra, quantitative determination of ordering within the
compounds was successful and points to a non-random arrangement for both sets of cations
in the mixed Nb/Ta systems.
114
5.2 Experimental
Relevant details of synthetic procedures and characterization techniques are
provided in Chapter 3. Elemental Analysis using ICP-OES was performed by RobertsonMicrolit Laboratories.
5.3 Results and Discussion
5.3.1 XRD, SEM, and Elemental Analysis
XRD patterns for the four-layer Rb samples (Figure 5.2) and protonated samples
(Figure 5.3) confirm formation of the desired structures with the orthorhombic Pnma space
groups, along with a small amount of precursor impurities. Earlier work on the crystal
structure of RbCa2NaNb4O13 determined the structure to be tetragonal35 (P4/mmm space
group, a = 3.8702(1) Ǻ, c = 18.8936(6) Ǻ), however bond valence calculations of this
structure results a global instability index of 0.34. A value above 0.2 indicates an unstable
structure,22 and thus an orthorhombic unit cell is more appropriate than tetragonal. Due to
the similar size of Nb5+ and Ta5+, there is no significant difference in lattice parameters
between the pure Nb5+ and Ta5+ incorporated samples. Both samples’ XRD patterns were
fit using a Pnma space group with lattice parameters doubled compared to the tetragonal
unit cell reported in the literature and lattice parameters a = 7.9162(7) Ǻ, b = 7.7364(6) Ǻ,
and c = 37.729(3) Ǻ for RbCa2NaNb4O13 and a = 7.7328(7) Ǻ b = 7.7178(7) Ǻ, and c =
37.628(3) Ǻ for RbCa2NaNb2.8Ta1.2O13. Ta5+, however, has a stronger x-ray scattering
factor and analysis of the ratio of the reflection intensities for the (100) and (110) hkl
115
reflections indicated successful incorporation of Ta5+ into the crystal lattice. Powder
diffraction, however, was unable to give information on the position of Ta5+ in the
structure. After the acid-exchange was performed, the (00l) stacking reflections were
shifted to a slightly higher 2θ angle because of the decreased size of the interfacial cation.
Detailed Le Bail fit information is given in Table 5.1. It should be noted small stacking
reflections related to the RbCa2Nb3O10 precursor are present in XRD patterns, though the
absence of 93Nb signals associated with this phase in subsequent NMR spectra indicate the
amount of precursor is too low to significantly affect the analysis.
SEM images for the acid exchanged samples (Figure 5.4) show large platelet
formation with lateral dimensions of several micrometers, consistent with literature reports
for other layered perovskites.52 Platelet size does appear to be somewhat smaller in the
Ta5+ containing sample, possibly due to the necessary diffusion of Ta5+ into the crystal
structure acting as a platelet growth inhibitor. The syntheses of Ta containing compounds
in general require a higher reaction temperature than pure Nb compounds suggesting that
a reaction temperature higher than the 1200 oC used may have promoted better platelet
growth. Elemental analysis was performed using ICP-OES from Robertson-Microlit
Laboratories.
The weight percentage composition for Nb5+ and Ta5+ in the mixed
Nb5+/Ta5+ sample was 26.26 and 22.98%, respectively.
These weight percentages
correspond to an empirical formula of RbCa2NaNb2.8Ta1.2O13, slightly higher than the
theoretical formula of RbCa2NaNb3TaO13 based on the stoichiometry of the precursors
during synthesis.
116
RbCa2NaNb4O13
2θ
RbCa2NaNb2.8Ta1.2O13
2θ
Figure 5.2. XRD patterns for RbCa2NaNb4O13 (above) and RbCa2NaNb2.8Ta1.2O13 (below). Note that
the pattern for the Ta compound is fit with both the four and three-layer structures represented by
blue and black ticks, respectively.
117
HCa2NaNb4O13
2θ
HCa2NaNb2.8Ta1.2O13
2θ
Figure 5.3. XRD patterns for HCa2NaNb4O13 (above) and HCa2NaNb2.8Ta1.2O13 (below). Note that the
pattern for the Ta compound is fit with both the four and three-layer structures represented by blue
and black ticks, respectively.
118
Table 5.1. Le Bail method fit lattice parameters, particle size, zero correction, χ2, and Rwp* for XRD
patterns shown in Figures 3 and 4
Sample
Figure 5.1
RbCa2NaNb4O13
RbCa2Nb3O10
RbCa2NaNb2.8Ta1.2O13
RbCa2Nb3O10
Figure 5.2
HCa2NaNb4O13
HCa2Nb3O10
HCa2NaNb2.8Ta1.2O13
HCa2Nb3O10
Space
Group
a (Ǻ)
b (Ǻ)
c (Ǻ)
Particle
Size (Ǻ)
Zero
Correction
χ2
Rwp
(%)
Pnma
Pnma
Pnma
Pnma
7.9162(7)
7.8710(8)
7.7328(7)
7.7935(5)
7.7364(6)
7.7643(8)
7.7178(7)
7.8440(8)
37.729(3)
32.119(3)
37.628(3)
29.845(2)
123.4(4)
85.0(4)
140.0(8)
153(10)
0.1202(9)
0.1202(9)
0.0096(1)
0.0096(1)
1.13
1.13
1.08
1.08
16.36
16.36
20.85
20.85
Pnma
Pnma
Pnma
Pnma
7.7259(5)
7.869(1)
7.7296(9)
7.7853(5)
7.7319(9)
7.8353(7)
7.978(1)
7.8612(6)
36.818(2)
28.866(2)
36.703(3)
29.124(2)
80.2(1)
87.3(4)
50.0(1)
114.7(6)
-0.0193(8)
-0.0193(8)
0.009(1)
0.009(1)
1.43
1.43
3.28
3.28
15.17
15.17
9.37
9.37
Figure 5.4. SEM images of RbCa2NaNb4O13 (top) and RbCa2NaNb2.76Ta1.24O13 (bottom) showing
platelets with lateral dimension of 1-3 μm.
5.3.2 93Nb NMR
Static 93Nb VOCS NMR spectra for HCa2NaNb4O13 and HCa2NaNb2.8Ta1.2O13 at
9.4 T and 14.1 T shown in Figure 5.5 were fit using 4 sites for both the pure Nb5+ and Ta5+
incorporated compounds (detailed site parameters shown in Table 5.3). Three sites had
moderate quadrupolar coupling values (CQ) in the range of 20-30 MHz. These values are
consistent with the interlayer 93Nb environments based on previous work on the three-layer
RbCa2Nb3O10 compound which had an interlayer 93Nb CQ value of 30 MHz. The fourth
site CQ value was substantially larger at approximately 50 MHz. In the three-layer
compound the interior site CQ value is traditionally much larger than the interlayer site (110
119
MHz vs. 30 MHz). Assuming the fourth site is associated with the interior Nb octahedra,
the large difference between the four-layer and three-layer EFG values is consistent with
the different crystal structures of the compounds. In the three-layer compound there is only
one interior Nb(2) octahedra per three-layer slab, whereas in the four-layer compound there
are two interior Nb(2) octahedra. Having two interior octahedra creates a bridging O atom
between the two interior Nb(2) octahedra not seen in the three-layer compound. The axial
interior Nb-O bonds in the three-layer system have been found to be 1.876 Ǻ.168 Assuming
the Nb-O bond lengths presented in previous literature35 for RbCa2NaNb4O13 are not
significantly different from our compounds, the axial interior bridging Nb(2)-O bond
length at 1.99424 Ǻ is approximately 0.15 Ǻ longer than the Nb-O bond in the interior
octahedra of the three-layer compound and close to equatorial Nb-O bond lengths. Based
on the longer Nb-O bond, the CQ value of the interior Nb site would be expected to decrease
substantially due to the site now having a more symmetric environment.
120
HCa2NaNb4O13 14.1 T
0
-100
-200
-300
HCa2NaNb2.8Ta1.2O13
14.1 T
0
-400
-100
kHz
a)
-300
-400
b)
HCa2NaNb4O13 9.4 T
0
-200
HCa2NaNb2.8Ta1.2O13
9.4 T
0
-400
-200
-400
kHz
kHz
c)
-200
kHz
d)
Figure 5.5. 93Nb Static VOCS NMR spectra with fits for HCa2NaNb4O13 at 14.1 (a) and 9.4 T (c) along
with HCa2NaNb2.8Ta1.2O13 at 14.1 (b) and 9.4 T (d). The black line corresponds to the experimental
spectrum and the blue line corresponds to the full calculated spectrum. The other colored lines
correspond to individual fits for each Nb environment. HCa2NaNb4O13 CQ values for individual fits are
as follows: purple = 22.5 MHz; red = 28.7 MHz; green = 29.5 MHz; and orange = 52.1 MHz. For
HCa2NaNb2.8Ta1.2O13 CQ values for individual fits are as follows: purple = 22.3 MHz; red = 25.9 MHz;
green = 27.1 MHz; and orange = 51.1 MHz.
121
Static 93Nb VOCS NMR spectra for RbCa2NaNb4O13 and RbCa2NaNb2.8Ta1.2O13
at 9.4 and 14.1 T shown in Figure 5.6 were fit using 4 sites for both the pure Nb5+ and Ta5+
incorporated compounds. The fit parameters shown in Table 5.2 are consistent with those
seen in the protonated compounds. Three sites showed moderate CQ values in the range of
30-40 MHz and fourth site with a much larger CQ of approximately 50 MHz. These
environments correspond to the exterior and interior Nb octahedra. Furthermore, the
population changes upon incorporation of Ta5+ are consistent with those seen in the
protonated form. As shown in Table 5.2, the population of the interior site drops from
48.2% to 34% indicating a strong preference for Ta5+ at the interior site.
122
RbCa2NaNb2.76Ta1.24O13 9.4 T
RbCa2NaNb4O13 9.4 T
0
-200
0
-400
kHz
-100
-200
-400
kHz
RbCa2NaNb2.76Ta1.24O13 14.1
T
RbCa2NaNb4O13 14.1 T
0
-200
-300
-400
0
kHz
-100
-200
-300
-400
kHz
Figure 5.6. 93Nb Static VOCS NMR spectra with fits for RbCa 2NaNb4O13 at 9.4 (top left) and 14.1 T
(bottom left) along with RbCa2NaNb2.76Ta1.24O13 at 9.4 (top right) and 14.1 T (bottom right). The black
line corresponds to the experimental spectrum and the blue line corresponds to the full calculated
spectrum. The other colored lines correspond to individual fits for each Nb environment. For
RbCa2NaNb4O13 CQ values, purple = 33.5 MHz, red = 35.2 MHz, green = 18.8 MHz, and orange = 53.5
MHz. For RbCa2NaNb2.8Ta1.2O13 CQ values, purple = 28 MHz, red = 33 MHz, green = 21 MHz, and
orange = 51 MHz.
123
Table 5.2. NMR fit parameters for RbCa2NaNb4O13 and RbCa2NaNb2.8Ta1.2O13 obtained from VOCS spectra. Error in the last digit is shown in
parentheses.
Compound
Site
Population
CQ
η
δis0 (ppm)
κ
Ω (ppm)
α (o)
β (o)
γ (o)
(%)
(MHz)
RbCa2NaNb4O13
RbCa2NaNb2.8Ta1.2O13
124
1
8.2(5)
33.5(1)
0.35(1)
-985(1)
0.60(1)
968(2)
9.8(6)
0.0(4)
18.9(6)
2
36.6(9)
35.2(3)
0.38(2)
-990(4)
0.62(3)
990(10)
17(3)
3.3(7)
24(3)
3
7.0(4)
18.8(6)
0.40(4)
-1040(5)
0.63(2)
640(30)
43(6)
9(3)
11(7)
4
48.2(9)
53.5(4)
0.45(2)
-1050(10)
0.06(2)
370(50)
6(1)
11.7(6)
17(4)
1
36(2)
28(1)
0.49(4)
-985(20)
0.7(1)
950(30)
3(3)
6(2)
50(20)
2
20(3)
33(1)
0.41(4)
-904(10)
1.0(3)
1100(50)
0(10)
14(4)
20(30)
3
10(2)
21(1)
0.6(1)
-1000(30)
-0.7(2)
290(50)
0(15)
0(10)
5(10)
4
34(3)
51(2)
0.41(5)
-1060(30)
0.0(2)
390(50)
0(15)
24(3)
3(10)
To ensure physically correct fits were obtained, rather than simply mathematically
correct fits, several methods were employed. Collecting static spectra at multiple fields
ensures fidelity of fits due to varying influence of EFG and CSA parameters at multiple
fields. Furthermore, MQMAS and BRAIN-CP spectra were used to extract various
parameters for several
93
Nb environments.
The reduced number of interactions of
unknown strength help to ensure the accuracy of static spectra fits. MQMAS spectra for
RbCa2NaNb4O13 and RbCa2NaNb2.8Ta1.2O13 are shown in Figure 5.7. The MQMAS
spectrum for RbCa2NaNb4O13 shows two 93Nb signals. Moment analysis of the two signals
(<v1> = -286 and -259 ppm for sites 1 and 2, respectively) resulted in CQ values of 35.2
and 33.5 MHz, and η values equal to 0.38 and 0.35 for sites 1 and 2, respectively. The
MQMAS spectrum for RbCa2NaNb2.8Ta1.2O13 shows three 93Nb signals. Moment analysis
(<v1> = -182, -251, and -102 ppm for sites 1, 2, and 3, respectively) of the two signals
resulted in CQ values of 28, 33 and 21 MHz, and η values equal to 0.49, 0.41, and 0.6 for
sites 1, 2, and 3, respectively.
125
Figure 5.7. 93Nb 3QMAS 2D spectrum of RbCa2NaNb4O13 (left) and RbCa2NaNb2.8Ta1.2O13 (right)
with double shearing scheme. The horizontal axis corresponds to the isotropic dimension and vertical
axis to the anisotropic dimension.
If Ta5+ had no preference for the interlayer or interior octahedra, the relative 93Nb
populations (Table 5.3) would show no significant change when Ta5+ was present. As seen
in Table 5.3 and Figure 5.5, the relative population of site 4 dropped from 44.6% in the
pure Nb5+ compound down to 32.4% in the Ta5+ containing compound while the other three
93
Nb environments’ relative populations remained unchanged. It should be noted that the
44.6% population is slightly lower than the theoretical 50% population for the interior Nb
environment, but the slight distributions of EFG and CSA with such large magnitudes in
these systems makes exact fitting difficult. If 100% of Ta5+ in the system was in the interior
octahedra, based on the elemental composition the population of site 4 in the 93Nb spectra
should decrease to 28.6%. The population drop to 32.4% indicates Ta5+ has an almost
completely exclusive site preference for the octahedral site associated with the interior
Nb(2) site 4 environment with nearly all Ta5+ in the structure located in this site. The 93Nb
static spectra for the parent compounds RbCa2NaNb4O13 and RbCa2NaNb2.8Ta1.2O13
shown in Figure 5.6 also show substantial population reduction of the larger EFG
126
environment upon Ta5+ incorporation. It should also be noted that the high asymmetry
parameters, η, in the range of approximately 0.4-0.7 support our XRD pattern fits using
orthorhombic unit cells. If the compounds indeed had a tetragonal structure as suggested
originally, the 4-fold symmetry about each Nb site would lead to an η value close to zero
as seen in 93Nb NMR studies of tetragonal RbSr2Nb3O10.169
To confidently assign the structural site associated with the ~50 MHz Nb
environment,
93
Nb-1H cross polarization and recoupling NMR experiments were
performed. In the BRAIN-CP experiment, polarization from 1H is transferred to nearby
93
Nb environments. The pulse sequence utilizes WURST pulses that allow for broadband
excitation large enough to cover the Nb environments seen in the static spectra. The
strength of the polarization transfer is inversely proportional to distance, thus only the
interfacial Nb environments would be polarized while the interior Nb sites would remain
unchanged. As seen in Figure 5.8 and Table 5.4, the BRAIN-CP spectrum for the pure
Nb5+ sample shows only signals with moderate CQ values suggesting the 52.1 MHz
environment is indeed associated with the interior octahedral site.
Complementing the cross-polarization data, MQMAS-t2-REDOR spectra also
indicate the moderate CQ environments are associated with the interlayer Nb octahedra. In
recoupling experiments, nonzero dipolar dephasing causes attenuation of
93
Nb signals in
close proximity to 1H nuclei. The original MQMAS spectrum (Figure 8a) shows the three
moderate CQ environments as expected, and the signal intensity for all three is distinctly
attenuated when the dephasing pulse on 1H is introduced (Figure 5.9b). The degree of
signal attenuation appears
to
vary
somewhat
127
between
the
three
93
Nb
environments suggesting varying Nb-H distances among the exterior octahedra which
implies a degree of disorder is present throughout the structure.
Based on the static
93
Nb wideline and 1H-93Nb cross polarization and recoupling
experiments, Ta5+ must have strong preference for the interior octahedral site in the fourlayer compound. Due to Nb5+ and Ta5+ having the same charge and similar size, it was
assumed Ta5+ would either prefer the interlayer octahedral site due to kinetic limitations of
the rapid synthesis or would be randomly distributed throughout the compound. A likely
explanation for the site preference is the difference in electronegativity of Nb 5+ and Ta5+.
As previous studies of Nb5+ and Ta5+ systems have shown, the more ionic Ta-O bonds
inhibit the degree to which octahedra undergo second order Jahn-Teller effects resulting in
TaO6 octahedra exhibiting higher symmetry than their NbO6 counterparts in isostructural
compounds.166, 170-172 In mixed Nb5+/Ta5+ systems, including this one, this manifests itself
as Ta5+ preferentially occupying more symmetric octahedral sites if such sites are
available.165
As mentioned previously, the increased electronegativity of Nb5+ leads to more
covalent Nb-O bonds relative to the Ta-O bonds. In a study of three-layer perovskites
RbLa2-xCaxTi2-xNb1+xO10 (0 < x <2) compounds showing site ordering of -(Ti0.5Nb0.5)O6TiO6-(Ti0.5Nb0.5)O6- for x=1 samples, Byeon et al.62 found, using Raman spectroscopy,
that reducing the covalent character of the interior octahedral sites in turn increased the
covalent character of the interlayer octahedral sites. If this behavior is consistent with
mixed Nb5+/Ta5+ systems, the interior octahedral preference of Ta5+ would lead to
increased covalent character of the interfacial Nb(1) site. This increased covalent character
128
would then lead to increased solid acid strength as the increased covalent Nb-O bond would
reduce the charge on the O thus weakening the O-H bond in layered perovskites where
Ta5+ is located exclusively in the interior octahedra compared to pure Nb5+ compounds.
HCa2NaNb4O13 1H-93Nb
BRAIN-CP 9.4 T
0
-50
-100
-150
-200
-250
kHz
Figure 5.8. 1H-93Nb BRAIN-CP spectrum with calculated fit. The black spectrum is the experimental,
blue the full calculated fit, red is the fit of site 3 (28.7 MHz), and green is the fit of site 2 (29.5 MHz).
129
Table 5.3. 93Nb NMR fit parameters for HCa2NaNb4O13 and HCa2NaNb2.8Ta1.2O13 obtained from MQMAS and VOCS spectra. Error in the last
digit is shown in parentheses
Sample
Site
Population
CQ (MHz)
η
δiso (ppm)
κ
Ω (ppm)
α (o)
β (o)
γ (o)
(%)
HCa2NaNb4O13
1
6.0(2)
22.5(1)
0.56(2)
-1081(5)
0.9(1)
290(10)
73(4)
73(2)
159(4)
HCa2NaNb2.8Ta1.2O13
130
2
13.9(4)
28.7(2)
0.65(1)
-1113(3)
0.1(1)
875(5)
68(1)
13.4(4)
89(8)
3
35.5(4)
29.5(2)
0.23(2)
-1099(2)
0.42(3)
580(10)
12(5)
2.4(9)
71(5)
4
44.6(3)
52.1(3)
0.47(1)
-1070(3)
0.37(3)
620(10)
16(2)
9.3(6)
18(3)
1
13.7(2)
22.3(2)
0.55(6)
-1087(4)
0.91(1)
180(7)
80(15)
74(3)
103(3)
2
31.4(5)
25.9(1)
0.75(1)
-1108(3)
0.03(1)
850(4)
73(1)
19.6(7)
77(1)
3
22.5(3)
27.1(3)
0.62(2)
-1106(3)
0.63(2)
620(7)
4(2)
3.3(5)
90(2)
4
32.4(2)
51.1(3)
0.46(1)
-1057(5)
0.25(2)
650(10)
4(4)
14.8(6)
54(4)
Table 5.4 93Nb NMR fit parameters for HCa2NaNb4O13 and HCa2NaNb2.8Ta1.2O13 BRAIN-CP spectra.
Sample
Population (%)
CQ (MHz)
η
δiso (ppm)
κ
HCa2NaNb4O13
HCa2NaNb3.3Ta0.7O13
Ω (ppm)
α (o)
β (o)
γ (o)
33.3
28.7
0.65
-1100
0.25
850
95.0
18.0
100.0
66.7
29.5
0.23
-1080
0.55
720
10.0
6.0
10.0
20.0
25.9
0.62
-1100
0.20
850
95.0
15.0
50.0
80.0
27.1
0.40
-1080
0.0.50
675
10.0
6.0
10.0
Figure 5.9. MQMAS-t2-REDOR spectra for HCa2NaNb4O13 without (left) and with (right) REDOR
sequence active. Numbers correspond to MQMAS sidebands for appropriate sites. Site numbers are
the same as those listed in Table 1.
5.3.3 23Na MAS NMR
The 23Na MQMAS spectrum for the pure Nb compound (Figure 5.10a) indicates
two 23Na environments, consistent with an interior Na site situated between interior Nb(2)
octahedra as well as an exterior Na site situated between interior Nb(2) octahedra and
exterior Nb(1) octahedra. As mentioned previously, if Na is randomly distributed as
suggested in literature reports, there should be a 2 to 1 population for the Na sites. Using
MQMAS fitting and moment analysis as well as fitting the MAS spectra (Figure 5.10b and
Table 5.5), however, revealed two sites with a population ratio of approximately 3 to 1
suggesting Na has a site preference. A rudimentary point charge model developed by
Koller et al.,173 estimating the charge on the various oxygen cations using the calculated
bond valences of the M-O bonds, was implemented to determine the appropriate ratio of
the EFG of the exterior 23Na to the interior 23Na environment. The bond valences, sij, were
calculated using Equation 5.1 where ro is the empirically derived ideal bond length, rij is
131
the bond length in the crystal structure, and B=0.37 is a constant, as demonstrated by
Shannon and Brown174 with bond lengths obtained from Sato et al.35 The charges, qi, on
the oxygen atoms were calculated using Equation 5.2, where a and M are empirically
determined parameters dependent on the number of core electrons of the cation,174 and used
in Equation 5.3 to calculate the resulting EFG tensor, where qe is the charge on the O in
the Na-O bond, x, y, and z are the bond distance components along the various axes, and r
is the bond distance. Finally, the EFG tensor was diagonalized in order to relate Vzz to CQ.
The calculated values of qi for each oxygen atom and resulting V zz for both exterior and
interior 23Na sites are presented in Table 5.6.
0 − 
Equation 5.1:
 =  [
Equation 5.2:

 = (∑ 
)−2
Equation 5.3:
 =  ∑  (

]
32 −2
) , 
5
3 
) ,  …
3
=  ∑  (
Using the tetragonal crystal structure proposed by Sato et al. as a guide, values for
Vzz were calculated for 23Na environments. Vzz is proportional to CQ and thus the ratio of
Vzz between 23Na environments is equal to the ratio of CQ values between environments.
It was found that the 23Na environment at the interstitial site closer to the interface should
have a CQ value approximately 40% larger than the interior interstitial 23Na environment.
The CQ ratio of the two experimental environments reasonably matches the model
suggesting that the Na+ cation prefers the exterior interstitial site. The MQMAS spectrum
132
and MAS spectrum for the Ta5+ containing compound (Figure 5.10c and 5.10d) show three
23
Na environments with relative populations of 12.8, 19.2, and 68.0 %. Comparing the CQ
and δiso values of HCa2NaNb2.8Ta1.2O13 with the pure Nb compound it appears that Na is
predominately or entirely at the exterior interstitial site based on the increased C Q of all
three 23Na environments in the Ta containing sample. Considering the more ionic nature
of the Ta-O bond, it should be expected that Ca2+ would be preferred over Na+ at the interior
to better compensate for the more negative oxygen environments compared to the exterior
interstitial site. Furthermore, considering the interior interstitial sites are now adjacent to
a mixture of NbO6 and TaO6 octahedra, the observation of two sites is consistent with the
previous determination of interior site preference for Ta5+.
133
HCa2NaNb4O13 23Na
MAS
-1000
-3000
-5000
Hz
a)
b)
HCa2NaNb2.8Ta1.2O13
23Na MAS
0
-1000 -2000 -3000 -4000 -5000
Hz
c)
d)
Figure 5.9. 23Na MAS and MQMAS spectra for HCa2NaNb4O13 (a and b) and HCa2NaNb2.8Ta1.2O13 (c
and d). The black line corresponds to the full experimental spectrum, the blue line is the full calculated
spectrum, the red line is the individual fit for site 1, the green line is the individual fit for site 2, and
the orange line is the individual fit for site 3. The fit parameters for each site are listed in Table 2.
Table 5.5. 23Na NMR fit parameters obtained from MQMAS and MAS spectra.
Sample
Site Pop (%) CQ (MHz) δiso (ppm) η
HCa2NaNb4O13
HCa2NaNb2.8Ta1.2O13
1
74.1(1.5)
1.93(2)
-14.3(2)
0.72(4)
2
25.9(0.7)
1.27(3)
-20.8(2)
0.9(1)
1
12.8 (4.2)
2.1(1)
-14.4(3)
0.78(6)
2
19.2(0.3)
1.9(1)
-20.6(2)
0.9(1)
3
68.0(1.2)
1.7(1)
-15.5(2)
0.9(1)
134
Table 5.6 Results of EFG tensor calculations using a point charge model for HCa 2NaNb4O13.
Na Site
Vzz* (x1024 V/m2)
O site
Description
Bond Length (Å)
# of Bonds
qi
Interior
0.8
1
Axial Nb(2)-O(1)-Nb(2)
2.73
4
-1.24e
2
Equatorial to Nb(2)
2.75
8
-1.06e
2
Equatorial to Nb(2)
2.98
4
-1.06e
3
Axial Nb(2)-O(3)-Nb(1)
2.76
4
-1.21e
4
Equatorial to Nb(1)
2.5
4
-1.17e
Exterior
1.11
5.4 Conclusions
Our 93Nb and 23Na NMR studies have shown Ta5+ to have a strong site preference
for the interior octahedra in the four-layer RbCa2NaNb4-xTaxO13 systems while Na+ is
believed to be favored at the exterior interstitial site. The use of quadrupolar solid-state
NMR to determine site ordering is a viable alternative to XRD, especially in materials in
which such conclusions cannot be readily made without access to a synchrotron source.
This determination of site preference based on octahedral distortions and electronegativity
can be quite useful in analyzing the results of other work on mixed Nb5+/Ta5+ systems.31,
164
Furthermore, since composition has been shown to have drastic impacts on the
properties of layered perovskites, this observation of site affinity can provide invaluable
insight into the nature of these behaviors and potentially allow for specialized tuning of
properties based on composition.
135
Chapter 6: Microwave Assisted Grafting of Mixed Cation
HCaxSr2-xNb3O10 Compounds with n-Alcohols*
6.1 Introduction
Effective catalytic behavior in the layered perovskite class of compounds requires
access to the protonated environments, typically done through exfoliation into single
nanosheets.63 Sugahara et al.64-69 have pioneered grafting techniques for attachment of
various organic groups into the interlayer of layered perovskites. Sugahara demonstrated
a grafting technique for protonated D-J perovskites to produce derivatives with n-alcohols
up to chain length n=18. Current grafting methods of layered perovskites involve high
pressure heating at 80-150 oC for days to weeks depending on the alcohol involved. The
long reaction time involved inhibits the ability to study organically grafted samples in a
timely fashion. In this current work, a novel microwave irradiation method (Figure 6.1 and
Table 6.1) was developed to reduce the grafting reaction time by 94-97% while maintaining
coverage levels consistent with traditional heating methods. As will be discussed, grafting
leads to a pillaring effect where the interlayer space has greatly increased while potentially
active protonated sites remain.
*Chapter adapted from “Boykin, J. R. and L. J. Smith (2015). "Rapid Microwave-Assisted
Grafting of Layered Perovskites with n-Alcohols." Inorganic Chemistry 54(9): 41774179.”
136
Proton (not to scale)
Sr
NbO6
Alkyl Chain
Microwave Heating
90% Methanol
Microwave Heating
95% Propanol
Microwave Heating
100% Hexanol
Figure 6.1. Illustration of stepwise microwave grafting.
6.2 Experimental
Parent compounds RbCa2Nb3O10 and RbSr2Nb3O10 were synthesized using molten
salt heating of the carbonate and oxide precursors in RbCl salt at 800 oC for 30-45
minutes.29 The hydrated protonated forms were obtained by treating samples with 6M
137
HNO3(aq) at 50 oC for 3 days. The hydrated HSr2Nb3O10 samples (~0.2g for each sample)
were then heated in a domestic microwave oven using a Parr 23 mL microwave acid
digestion bomb with 13.5 ml of 90% methanol using the cycled heating times shown in
Table 6.1. Microwave heating was cycled to prevent overheating of the acid digestion
bomb. After successful methyl grafting, the samples were filtered and washed with
acetone, heated in 9.5 ml of 95% n-propanol, and collected by filtration and washing with
acetone. Samples were then heated in 10.0 ml of 100% n-hexanol with detailed cycles
times shown in Table 6.1, followed by filtration and washing with acetone. HCa2Nb3O10
samples were grafted using the conventional heating method as a way of comparing the
level of grafting with published reports66 as well as with microwave grafted samples. The
HCa2Nb3O10 samples were heated in 90% methanol at 100 oC for 3 days in an acid digestion
bomb then filtered and washed with acetone. This was followed by heating in 100% npropanol at 150 oC for 7 days, then filtering and washing with acetone. This was finally
followed by heating in 100% hexanol at 150 oC for 7 days, then filtering and washing with
acetone. Instrumental details of data collection are given in Chapter 3. It should be noted
that while no precise measurement of pressure was obtained, the heating times were chosen
such that pressure reaches a maximum of approximately 500 psi.
138
Table 6.1. Heating Cycles and Reaction Times for Microwave and Conventional Heating Methods
Alcohol
Heating Cycle (100%
# of
Cooling Period
Total Reaction
power)
Cycles
Between Cycles
Time
Methanol
5x (20s on 40s off)
10
15-20 min
~4 hours
(microwave)
Methanol
100o C 3 days
----3 days
(conventional)
Propanol
5x (40s on 20s off)
10
15-20 min
~4 hours
(microwave)
Propanol
150o C 7 days
----7 days
(conventional)
Hexanol (microwave)
4x (60s on 15s off)
7
20-25 min
~3.5 hours
Hexanol
150o C 7 days
----7 days
(conventional)
6.3 XRD Results and Discussion
XRD powder patterns were collected for a 2θ range of 2.5-14o and 2.5-60o. The
crystal structure for fully hydrated HSr2Nb3O10 has an orthorhombic space group with
lattice parameters a = 7.808(5), b= 7.82(1), and c = 32.49(2) Ǻ. Upon grafting with nalcohols, the changes in the interlayer distance result in changes in the c lattice parameter
as shown in Figure 6.2. The 2.5-14o range contains the reflections for the (001) and (002)
planes associated with the stacking direction in the grafted and hydrated forms,
respectively.
C6/Sr2Nb3O10
C6/Ca2Nb3O10
C1/Sr2Nb3O10
C3/Ca2Nb3O10
C3/Sr2Nb3O10
C1/Ca2Nb3O10
HSr2Nb3O10
HCa2Nb3O10
2.5
4
5.5
7
8.5
2θ (o)
10
11.5
13
2.5
4
5.5
7
8.5
2θ (o)
10
11.5
13
Figure 6.2. (a) XRD patterns for HSr2Nb3O10 (black), C1/Sr2Nb3O10 (red), C3/Sr2Nb3O10 (green), and
C6/Sr2Nb3O10 (blue) synthesized using the microwave irradiation method. (b) XRD patterns for
HCa2Nb3O10 (black), C1/Ca2Nb3O10 (red), C3/Ca2Nb3O10 (green), and C6/Ca2Nb3O10 (blue) synthesized
using a conventional heating method.
139
Despite repeated attempts with modification to all synthetic parameters (including
water content, heating time, number of cycles, and solvent volume), the HCa2Nb3O10
samples never achieved successful grafting beyond the methyl stage. As Figure 6.3
illustrates, XRD patterns for attempted grafting with n-propanol consistently contained
strong reflections associated with the hydrated HCa2Nb3O10 phase and less intense
reflections associated with the C3/Ca2Nb3O10 phase. Furthermore, TGA data show only
20.1 % methyl coverage for C1/Ca2Nb3O10, compared to 35-40 % coverage seen in
C1/Sr2Nb3O10 samples. It was therefore concluded that the Ca form of the three-layer
niobate contains protons so tightly bound that the favorable reaction conditions of
microwave heating are unable to overcome the kinetic barrier of microwave-based grafting.
After the ultimately unsuccessful attempts with microwave grafting of HCa2Nb3O10, the
HSr2Nb3O10 compound was used for attempted grafting.
A large motivation for the potential of HSr2Nb3O10 to be more grafting susceptible
came from the observations of the difficulty in obtaining a fully hydrated HCa2Nb3O10
sample. It has been seen that HCa2Nb3O10 requires a narrow temperature range of ~50-60
o
C during the acid exchanging step in order to achieve a fully acid-exchanged hydrated
state. Furthermore, if HCa2Nb3O10 is not acid exchanged within this necessary temperature
range it is difficult to revert to a hydrated state even with treatment with distilled water
under various conditions including submersion for several days, treatment in a high
humidity environment, or microwave heating with distilled water. The HSr2Nb3O10
compounds however revert to a fully hydrated state simply by being exposed to
atmospheric conditions. In fact, several attempts, made through heating of the HSr2Nb3O10
140
compound to obtain the anhydrous form, have shown that reversion to the hydrated state
occurs within 1-3 hours after the sample has been exposed to the atmosphere even in
situations in which humidity has been measured to be below 25%. Therefore, it was
believed the interlayer gallery of HSr2Nb3O10 is more susceptible to modification than the
HCa2Nb3O10 form. We were quite satisfied to find this assumption was correct and
microwave grafting was incredibly successful in the pure Sr state.
HCa2Nb3O10 Microwave Grafting
3
4
5
6
7
8
9
10
11
12
13
14
2θ (Degrees)
Figure 6.3. XRD patterns for HCa2Nb3O10 (Black), C1/Ca2Nb3O10 (red), and C3/Ca2Nb3O10 (green)
synthesized using microwave grafting. Note the presence of an intense reflection associated with the
HCa2Nb3O10 phase.
The successfully methoxylated sample showed a slight shift of the stacking
reflection to a smaller distance consistent with the decreased size of the interlayer when
water molecules from the hydrated phase are replaced by methyl groups in the C1 stage.
The C3 and C6 samples showed increased stacking distances from a starting point of ~16.2
Ǻ to ~19.8 and ~24.5 Ǻ, respectively. This increase in stacking distance is consistent with
the longer alkyl chains now present in the interlayer gallery. The C3 and C6 samples’ XRD
patterns also show low intensity broad reflections at 2θ = ~5.5O, consistent with a small
141
amount of hydrated phase still being present. Any hydrated phase that remains in the
sample is a small fraction of the final composition.
The changing position of stacking reflections was also seen in the HCa2Nb3O10
samples grafted using conventional heating. Long range XRD patterns were obtained for
C6/Sr2Nb3O10 and C6/Ca2Nb3O10 (Figures 6.4 and 6.5) to confirm three-dimensional
ordering of the compounds remained intact upon alkoxylation. Le Bail175 fits of the long
range patterns performed with the Bruker TOPAS software package showed elongation of
the c-axis consistent with successful grafting while maintaining crystallinity in three
dimensions. Calculated lattice parameters with errors along with particle sizes based on
Scherrer analysis are given in Table 6.2. It should be noted that the grafted samples’ XRD
patterns were fit using a tetragonal P4 unit cell that provides a satisfactory simulated pattern
based on the small number of reflections present and confirms the elongations of the caxis, but does not necessarily reflect the true space group of the compounds.
142
Figure 6.4. Long range XRD patterns for HCa2Nb3O10 (top) and HSr2Nb3O10 (bottom). Experimental
patterns are shown in blue. Le Bail fits are shown in red. Blue ticks represent hkl reflections. The
difference pattern is shown in gray.
143
Figure 6.5. Long range XRD patterns for C6/Ca2Nb3O10 (top) and C6/Sr2Nb3O10 (bottom).
Experimental patterns are shown in blue. Le Bail fits are shown in red. Blue ticks represent hkl
reflections for the C6 phase. Black ticks represent hkl reflections for the hydrated phase. The
difference pattern is shown in gray. Listed hkl reflections refer to the C6 phase unless specified
otherwise. *Refers to the minority hydrated phase.
Table 6.2. Le Bail method fit lattice parameters, zero correction, χ2, particle size in stacking direction,
and Rwpa from Le Bail fits for hydrated and C6 samples. Error in the last digit is represented by
parentheses.
Sample
Space
a (Ǻ)
b (Ǻ)
c (Ǻ)
Particle
Zero
χ2
Rwp
Group
Size (Ǻ)b
Corr.
(%)
Figure 3
HCa2Nb3O10
Pnma
7.675(1)
7.917(1) 32.333(7)
203(9)
-0.0074 1.24 10.49
HSr2Nb3O10
Pnma
7.808(5)
7.82(1)
32.49(2)
350(25)
-0.0005 1.18 11.11
Figure 4
C6/Ca2Nb3O10
P4
3.8569(5) 3.8569(5) 24.800(9)
400(11)
-0.0032 1.71 7.67
HCa2Nb3O10
P4
7.712(9)
7.716(9)
32.19(4)
80(20)
-0.0032 1.71 7.67
C6/Sr2Nb3O10
P4
3.9044(8) 3.9044(8) 24.77(1)
151(5)
-0.0042 1.46 9.77
HSr2Nb3O10
P4
7.702(5)
7.702(5)
31.39(3)
140(10)
-0.0042 1.46 9.77
a. Rwp refers to the background subtracted R-factor
b. Particle size refers to stacking direction based on (001) and (002) reflections for the grafted and hydrated
compounds, respectively.
144
6.4 TGA and 1H NMR Results and Discussion
While the XRD data indicate the majority of the samples have been successfully
grafted, they unfortunately do not provide quantitative information as to the alkoxyl
coverage levels. To obtain quantitative information about the alkoxyl coverage, 1H NMR
and TGA experiments were performed on the grafted samples. MAS NMR spectra of the
grafted samples contained peaks associated with residual water in the sample, unreacted
acid sites from the Nb-O-H groups, and alkyl protons on the alkoxyl chains as shown in
Figure 6.6a. Peak integration of
1
H Hahn Echo176 spectra with varying relaxation
times were used to estimate alkoxyl coverage for each sample by comparing the integral of
the unreacted acid sites to the integral of the alkyl peak as shown in Figure 6.6b. Varying
relaxation times were used to extrapolate back to an echo time of zero and negate the effects
of differing relaxation rates.
Methoxylated and C3 samples did not show uniform
exponential decay due to a broad overlapping peak at short relaxation times resulting in a
slightly skewed data set. The 1H NMR results indicate 40% coverage per perovskite unit
for C6 samples.
C6/Sr2Nb3O10 1H NMR with Varying
Echo DelayTime
100 us
800 us
11.4 ppm
H-O-NbO5
6.5 ppm
H2O
ppm 5
20
Integral Units
1.4 ppm
-(CH2)5-CH3
C6/Sr2Nb3O10 Peak Area vs. Echo
DelayTime
15
Peak at 1.4
ppm
10
Peak at
11.4 ppm
5
0
1200
(μs)800
Figure 6.6. (a) H MAS Hahn Echo NMR spectra for C6/Sr2Nb3O10 with varying echo delay times and
(b) results of peak integration as a function of the echo delay time.
15
0
-5
1
145
400tau
TGA data shown in Figure 6.7 collected from 30 – 800 oC were collected to
substantiate the NMR results. Water loss was seen up to 110 oC, followed by further mass
loss up to 800 oC. Assuming alcohol loss would occur before condensation of the
compound into a denser phase, the mass loss derivatives were used to determine the
appropriate point to consider alcohol loss complete which was found to be approximately
375 oC. The TGA results are in good agreement with the NMR data and suggest 39.1%
coverage per perovskite unit for the microwave C6 samples. TGA data were also obtained
for the C6/Ca2Nb3O10 compound synthesized using conventional heating; it was found that
the alkoxyl coverage was 45.0%.
0.1
96
0.05
94
0
92
-0.05
100
300
Temp500
(oC)
C6/Ca2Nb3O10 TGA
96
94
0.05
92
90
-0.05
100
700
0.25
0.15
Weight %
98
98
Derivative of Weight
%
100
Weight %
0.15
Derivative of Weight
%
C6/Sr2Nb3O10 TGA
100
300 Temp (500
oC)
700
Figure 6.7. TGA data for C6/Sr2Nb3O10 (above) and C6/Ca2Nb3O10 (below) with temperatures ranging
from 100 to 800 °C. Based on theoretical values of 11.84% and 13.84% for 100% alkoxyl coverage of
the strontium and calcium samples, the alkoxyl coverage was found to be 39.1% and 45.0%,
respectively.
Not only has the total reaction time to obtain the C6 phase been decreased from 17
days to 12 hours, there is no significant decrease in the total alkoxyl group coverage of the
compound. XRD, NMR, and TGA data have shown that microwave irradiation is an
enticing alternative to conventional heating for alkoxylation of layered perovskites. While
using longer chain n-alcohols such as n-octadecanol are not viable due to limited
microwave susceptibility, long chain alcohols could possibly be introduced if more polar
groups are present on the chain to increase the microwave susceptibility. Thorough
146
reviews by Kapp177 and Gabriel et al.178 on microwave heating of organic compounds
discuss in great detail the properties of molecules that affect the ability of a solvent to
absorb microwave energy (referred to as loss tangent), such as the nature of the functional
groups and volume of the molecule. These should serve as excellent resources for those
looking to adopt this microwave assisted grafting technique using other organic species.
6.5 Grafting of Mixed Ca/Sr Compounds
Building off the success of microwave grafting using the HSr2Nb3O10 samples, it
was thought that mixed Ca/Sr forms HCaxSr2-xNb3O10 would potentially exhibit successful
microwave grafting thus imbuing compounds with microwave grafting ability without
substantially altering the composition and thus any other potentially desirable properties.
As shown in Chapter 4, substitution of Ca as low as 10% results in a heterogeneous exterior
Nb site as evidenced by fits of
93
Nb NMR spectra for mixed Ca/Sr compounds. It was
observed that based on CQ values, mixed Ca/Sr compounds contained a mixture of Sr-like
environments and Ca-like environments. As discussed in Chapter 4, Ca-like environments
correspond to more highly distorted exterior octahedra when compared to Sr-like
environments. This increased distortion leads to a decreased O-H bond strength and thus
decreased reactivity of the acid site, as evidenced by comparing the catalytic activity of
HSr2Nb3O10 and the more distorted HCa2Nb3O10.52 Due to the presence of less distorted
octahedra in mixed Ca/Sr compounds, it was expected that Sr-like environments would be
able to undergo microwave-assisted grafting beyond C1 stage, even if the more distorted
Ca-like environments cannot. It was initially thought that coverage percentages would
147
gradually increase with increasing Sr content due to the presence of more reactive Sr-like
environments.
To both surprise and delight, it was clear that with as little as 10% Sr substitution,
mixed Ca/Sr samples exhibited grafting ability equal to that of the pure Sr compound. As
shown in Figure 6.8 and Table 6.3, the alkyl coverage for the C1, C3, and, C6 samples of
mixed Ca/Sr compounds exhibited equal or greater coverage than that of the pure Sr
compound.
Table 6.3. Alkyl coverage percentages for C1-C6/CaxSr2-xNb3O10 compounds using microwave
grafting.
Ca Content C1 Coverage (%) C3 Coverage (%) C6 Coverage (%)
0
36.5
29.5
39.1
0.4
43.5
33.6
45.1
0.8
39.2
31.8
43
1
39.7
32.9
43.4
1.5
36.4
33.3
42.8
1.8
36
31.1
44.3
2
20.1
NA
NA
Alkyl Coverage vs. Ca content
50
45
40
35
30
25
20
15
10
5
0
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Ca Content (Formula Units)
Hexyl Coverage
Methyl Coverage
Propyl Coverage
Figure 6.8. Alkyl coverage percentages for C1-C6/CaxSr2-xNb3O10 compounds using microwave
grafting.
148
The XRD patterns for mixed Ca/Sr grafted samples shown in Figure 6.9 are
remarkably consistent with the results of microwave grafting of the pure HSr2Nb3O10
compounds. The alkyl coverages for C1, C3, and C6 samples shown in Figure 6.8 were
calculated using TGA data alone without additional NMR data due to the consistency seen
between the two methods based on the Sr2Nb3O10 research. The TGA data collection
procedure being less time consuming and less sensitive to the presence of water was
another benefit. Figure 6.10 shows the TGA data for C1-C3/Sr2Nb3O10 samples which are
consistent among all mixed Ca/Sr samples with the exception of the Ca2Nb3O10 samples.
For simplicity, Figure 6.10 has been chosen to represent the general TGA data collected
for all samples. Individual TGA graphs for each sample studied are included in Appendix
B. These results indicate that microwave grafting ability can be imbued with minor
modifications to composition without changing other desirable properties in these D-J
layered niobates.
149
H-C6/Ca0.8Sr1.2Nb3O10
H-C6/Ca0.4Sr1.6Nb3O10
350
1200
300
C3 Stage
250
C6 Stage
1000
800
400
1400
350
1200
300
C1 Stage
250
C3 Stage
C6 Stage
1000
200
Protonated Stage
150
200
C1 Stage
150
600
Protonated Stage
100
400
50
200
600
400
100
200
50
0
0
0
2 3 4 5 6 7 8 9 10 11 12 13 14
2θ (Degrees)
2θ (Degrees)
H-C6/Ca1.0Sr1.0Nb3O10
H-C6/Ca1.2Sr0.8Nb3O10
1400
1200
500
Protonated Stage
1000
400
C1 Stage
300
C3 Stage
800
C6 Stage
600
200
400
100
200
0
0
500
450
400
350
300
250
200
150
100
50
0
2500
2000
1500
1000
500
0
1400
1200
Protonated Stage
C1 Stage
1000
C3 Stage
800
C6 Stage
600
400
200
0
2 3 4 5 6 7 8 9 10 11 12 13 14
2 3 4 5 6 7 8 9 10 11 12 13 14
2θ (Degrees)
2θ (Degrees)
H-C6/Ca1.5Sr0.5Nb3O10
3000
0
2 3 4 5 6 7 8 9 10 11 12 13 14
600
800
H-C6/Ca1.8Sr0.2Nb3O10
900
800
C1 Stage
700
600
Protonated Stage
500
C3 Stage
400
C6 Stage
300
200
100
0
2 3 4 5 6 7 8 9 10 11 12 13 14
500
450
400
350
300
250
200
150
100
50
0
2θ (Degrees)
1000
900
800
C6 Stage
700
C1 Stage
600
Protonated Stage
500
C3 Stage
400
300
200
100
0
2 3 4 5 6 7 8 9 10 11 12 13 14
2θ (Degrees)
Figure 6.9. XRD patterns for H-C6/CaxSr2-xNb3O10 compounds grafted using the microwave heating
method.
150
As mentioned previously, alkyl group mass loss in TGA data was determined using
the derivative of weight % to isolate alkyl loss from condensation of the layered structure.
As illustrated in Figure 6.8, the C3 stage for all successfully grafted samples shows the
peculiar trend of decreasing in alkyl coverage by 5-10% from the C1 stage while the
subsequent C6 stage increases by 5-10%. Further analysis of the TGA data for C3
compounds provides an explanation for this phenomenon. While TGA graphs for C1 and
C6 compounds show one distinct mass loss event in the 275-400 oC range, the C3 graphs
show two mass loss events. The mass loss event in the C6 stage occurs at ~100 oC higher
temperature than the C1 stage due to the longer alkyl chain present. Looking at the C3
stage, the first mass loss event occurs at a temperature consistent with that of the C1 stage
while the second mass loss event occurs at a temperature between the mass loss
temperatures of the C1 and C6 stages and can be assigned to propyl group loss. One
possible conclusion is that the C3 samples contain both methyl and propyl sites, which
would also explain the presence of two mass loss events in C3 samples. Unfortunately, the
current TGA data are insufficient to accurately test this hypothesis.
Figure 6.10. TGA graphs for C1 and C3/Sr2Nb3O10 samples. All mixed Ca/Sr samples show the same
general shape of graph with the only changes being the magnitude of mass loss.
151
It is believed that the presence of even small amounts of Sr allows for alteration of
exterior Nb sites beyond those immediately bonded to Sr cations creating an environment
suitable for microwave grafting. This is supported strongly by the results of grafting
coverage in mixed Ca/Sr samples. Despite varying Ca contents, coverage in all stages is
rather consistent with the expected progressions stated previously. Furthermore, this is
consistent with the NMR results of Chapter 4 which showed that the ratios of Sr-like to
Ca-like 93Nb environments was not equal to the atomic ratio of Sr to Ca. Therefore, the
presence of Sr is clearly having an effect extending to nearby sites containing only Ca,
consistent with observed changes in CQ of Ca-like environments seen in the
93
Nb NMR
data in Chapter 4.
6.6 UV-visible Results and Discussion
Having successfully applied the microwave assisted grafting technique to mixed
Ca/Sr compounds, UV-visible spectroscopy was used to study the effect of grafting on
band gap. An ISR-3100 integrating sphere attachment was used in conjunction with a
Perkin Elmer Lambda 35 UV-visible spectrophotometer to obtain diffuse reflectance
spectra. Barium sulfate was used as a standard. Tauc plots were obtained using the method
first developed by Tauc, Davis, and Mott179-181 based on the following expression:
Equation 6.1: (ℎ)1/ = (ℎ –  )
Where h = Planck’s constant, ν = frequency, α = absorption coefficient, A = proportional
constant, Eg = band gap, and n denotes the nature of the transition (equal to ½ in this work
152
for the direct allowed transition). The acquired spectra were converted to the KubelkaMunk function with the vertical axis converted to the quantity F(R∞) (F(R∞) = (1- R∞)2/2
R∞) which is proportional to the absorption coefficient. The expression in equation 2.1
thus becomes:
Equation 6.2: (hνF(R∞))2 = A(hν – Eg)
Using the Kubelka-Munk function, the value of (hν – (hνF(R∞))2) is plotted with hν (with
values in eV) on the horizontal axis and (hνF(R∞)2 on the vertical axis. To calculate the
band gap, Eg, a line is drawn tangent to the point of inflection on the curve and the hν value
at the intersection of the tangent line and the horizontal axis is thus equal to Eg.
As shown in Figure 6.11, the direct band gap undergoes a substantial shift upon
grafting with Eg = 3.28 eV for HSr2Nb3O10, 3.64 eV for C1/Sr2Nb3O10, 3.55 eV for
C3/Sr2Nb3O10, and 3.46 eV for C6/Sr2Nb3O10. The value of 3.28 eV for HSr2Nb3O10 is in
line with literature reported values182 indicating the band gap determination procedure was
applied correctly. It has been shown that compositional differences of D-J niobates lead to
changes in band gap such as HSr2Nb3O10 being reported with Eg = 3.30 eV and
HCa2Nb3O10 being reported with 3.50 eV.182 The results shown here, however, are the first
reported instances of band gap modification through organic grafting. While investigation
into band gap engineering through organic grafting methods lay outside the scope of the
work presented here, the effect of grafting on band gap of layered perovskites is most
certainly deserving of further study.
153
HSr2Nb3O10 Tauc Plot
(hνF(R∞))2
1.7
1.1
Experimental
0.5
-0.1
Fit
2
2.5
a)
3
3.5
Energy (eV)
4
4.5
C1/Sr2Nb3O10 Tauc Plot
(hνF(R∞))2
5
3
Experimental
1
-1 2
Fit
2.5
b)
3
3.5
Energy (eV)
4
4.5
C3/Sr2Nb3O10 Tauc Plot
(hνF(R∞))2
7
5
3
Experimental
1
Fit
-1 2
2.5
c)
3
3.5
Energy (eV)
4
4.5
C6/Sr2Nb3O10 Tauc Plot
(hνF(R∞))2
2.5
1.5
Experimental
-0.5 2
d)
Fit
0.5
2.5
3
3.5
Energy (eV)
4
4.5
Figure 6.11. Tauc plots for H-C6/Sr2Nb3O10 (a-d, respectively) microwave grafted samples showing
substantial shift in band gap upon grafting.
154
6.7 Conclusions
The substantial reduction in total reaction time using microwave heating compared
to conventional heating clearly increases the viability of further research on grafting of
layered perovskites.
In fact, two other research groups have already begun using
microwave assisted grafting for their work. Wiley et al.183 have used microwave assisted
grafting to successfully graft the two layer HLaNb2O7 compound while also confirming
our results of HCa2Nb3O10 microwave grafting in that the Ca form is unable to graft past
the C1 stage. Wang et al.184-185 have applied the microwave grafting method to the postsynthetic modification of the Aurivillius phase Bi2SrTa2O9 using amines and diamines
further illustrating the broad applicability of this method. Building on the success of this
grafting work, the next stage of research was to investigate the ability of these grafted
compounds to be exfoliated into nanosheets which will be discussed in the following
chapter.
155
Chapter 7: Exfoliation of Grafted Niobates Using Organic
Solvents
7.1 Introduction
As previous chapters have discussed, the ion-exchange ability of D-J layered
perovskites is one of the properties that has stimulated research interest in recent years.
This ion-exchangeability is key to currently employed exfoliation techniques. The
conventional method of D-J layered niobate exfoliation relies on the intercalation of bulky
bases such as TBAOH after interlayer proton exchange of parent compounds.36, 38, 52, 186-189
The intercalated base increased the interlayer distance thus overcoming forces holding the
layers together and resulting in exfoliation of layers into single nanosheets. This method,
while successful, requires the use of an aqueous solvent and is rather time consuming with
full intercalation only being achieved after treatment for seven days. Furthermore, removal
of TBA cations requires a dropwise addition of dilute acid over the course of several hours,
and if the rate is too fast or too much of an excess of acid is added, sheets will restack and
revert to the initial protonated state. This difficult method of exfoliation is a hindrance in
further research into the properties and applications of niobate nanosheets.
Organic solvents have been used successfully for exfoliation of many layered
materials with a graphite like structure such as metal chalcogenides190-195 (e.g. MoS2, WS2,
GaS, and GaSe to name a few). Typically, with an appropriate solvent, sonication alone
was sufficient to produce colloidal suspensions of exfoliated materials. The efficiency of
the exfoliation solvent was found to be partially dependent on its Hansen solubility
parameters (HSP), a semi-empirical explanation of dissolution behavior. Three HSP values
156
are used: δD, δP, and δH, corresponding to the strength of dispersive, polar, and hydrogenbonding forces of the solvent with itself. In work from Zhou et al.159 it was found that
having a close HSP match between the solvent and the nanomaterial for all three HSP
values was key to successful exfoliation. This method of exfoliation in organic solvents
has been found to lack the issues of structural deformations, sensitivity to environmental
conditions, and layer reaggregation that are persistent problems with ion intercalation
methods.39, 182, 196 Up to this point, however, organic solvent exfoliation has not been
applied to layered perovskites due to the strong electrostatic forces holding perovskite
layers in place. Herein, a new method of D-J layered niobate exfoliation using organic
solvents and grafted niobates is presented.
After the success of microwave grafting using n-alcohols, it was thought that the
grafted compounds could be exfoliated using a novel method of treatment with organic
compounds. As the results in Chapter 6 show, stacking distance increases 16.2 Å to ~25
Å going from HSr2Nb3O10 to C6/Sr2Nb3O10 due to substantial expansion of the interlayer
gallery. It was therefore believed that the electrostatic forces holding the sheets together
were sufficiently weakened that treatment with an appropriate solvent would allow for easy
exfoliation. Due to the presence of alkyl chains in the interlayer gallery, introducing an
organic solvent into the interlayer would lead to weakening of the dispersion forces holding
sheets together without the need for modification to the surface sites as is the case with
TBAOH exfoliation.
A variety of different solvents and synthetic techniques were
employed to test this hypothesis. As the following sections will show, grafted layered
157
niobates are well-suited for suited for organic solvent exfoliation using relatively rapid
synthetic methods.
7.2 Experimental
7.2.1 Sample Overview
Table 7.1 includes summaries of the exfoliation procedures employed for the 15
samples to be discussed in this chapter. It should be noted that each sample number is
representative of a unique procedure that was repeated multiple times to confirm results.
The synthetic variables studied were composition of the parent compound, solvent choice,
heating, solvent choice, sonication (before and after heating), sedimentation method, and
methods of extracting nanosheets out of solution. Instrumental details for synthetic
procedures and XRD, SEM, EDS, UV-visible, and TGA characterization methods can be
found in Chapter 3. Preparation methods for data collection were highly sample dependent
and are thus discussed as the data are presented.
158
Table 7.1 Summary of exfoliation procedure by sample. More detailed descriptions of the exfoliation procedures are also in the text.
159
#
Composition
Solvent
Initial
Sonication
Heatinga
Final
Sonication
Centrifuge
SNb
Treatment
Notes
1
C6/Sr2Nb3O10
Cyclohexane
30 min
X
X
X
X
No colloidal suspension
2
C6/Sr2Nb3O10
Pyridine
30 min
X
X
6000 rpm
45 min
Air Dry
Dropwise
No suspension
3
C6/Sr2Nb3O10
5M2Hc
30 min
X
X
6000 rpm
45 min
4
C6/Ca2Nb3O10
5M2H
60 mind
X
X
6000 rpm
45 min
5
C6/Sr2Nb3O10
5M2H
X
10x
X
6000 rpm
45 min
6
C6/Sr2Nb3O10
5M2H
X
8x
60 min
7
C6/Sr2Nb3O10
5M2H
X
150C 3d
Steel
Bomb
8
C3/Sr2Nb3O10
2Pe
60 min
9
C3/Ca1Sr1Nb3O10
2P
10
C6/Ca0.25Sr1.75Nb3O10
11
C6/Ca0.25Sr1.75Nb3O10
White suspension
Strongly yellow suspension
6000 rpm
45 min
Air Dry
Dropwise
Dry
Dropwise
at 80 oC
Air Dry
Dropwise
X
X
X
More pale yellow compared to
microwaved suspensions
10x
60 min
X
X
Slight yellow suspension
60 min
10x
60 min
X
X
Consistent with pure Sr form in Sample
8
5M2H
180 min
10x
180 min
6000 rpm
45 min
Air Dry
Dropwise
Consistent with pure Sr form in Sample
6
2P
180 min
10x
180 min
6000 rpm
45 min
Air Dry
Dropwise
Settled faster than 10 after initial
sonication (no Tyndall effect), slight
yellow color after heating
White suspension
Yellow suspension
Table 7.1 Cont’d.
#
Composition
Solvent
Initial
Sonication
Heatinga
Final
Sonication
Centrifuge
SN
Treatment
Notes
12
C6/Sr2Nb3O10
5M2H
60 min
10x
60 min
6000 rpm 45
min
Freeze
Dryingf
3.0 mL of SN mixed with 6.0 mL t-BuOH
before freeze drying
13
C6/Sr2Nb3O10
5M2H
60 min
10x
60 min
6000 rpm 45
min
Dried
Dropwise
under N2
SN was distilled at 140-144 oC. Distillate
was clear and colorless
Freeze
Drying
160
14a
C3/Sr2Nb3O10
2P
60 min
10x
60 min
6000 rpm 45
min
14b
C3/Sr2Nb3O10
H2O/Ethanol
X
X
X
X
Freeze
Drying
14c
C3/Sr2Nb3O10
2P
60 min
10x
60 min
6000 rpm 45
min
Air Drying
15a
C6/Sr2Nb3O10
5M2Hg
60 min
X
X
6000 rpm 45
min
Freeze
Drying
15b
C6/Sr2Nb3O10
H2O/Ethanol
X
X
X
X
Freeze
Drying
15c
C6/Sr2Nb3O10
5M2H
60 min
10x
60 min
6000 rpm 45
min
Rapid
Drying
a)
b)
c)
d)
e)
f)
g)
h)
5.0 mL SN mixed with 5.0 mL t-BuOH
then dried under air stream with regular
addition of extra t-BuOH until sample had
no scent of 2P
5.0 mL SN from 14a mixed with 2.0 mL
ethanol followed by 5.0 mL H2O (thin film
at interface). Aqueous phased was used for
freeze drying.
Sediment from 14a was used as starting
material.
5.0 mL SN mixed with 5.0 mL t-BuOH
then dried under air stream with regular
addition of extra t-BuOH until sample had
no scent of 5M2H
5.0 mL SN from 15a mixed with 2.0 mL
ethanol followed by 5.0 mL H2O.
Aqueous phase was used for freeze drying.
Sediment from 15a was used as starting
material.
Heating cycle was 5x(45s on 15s off) for both 2P and 5M2H samples.
SN = Supernatant
5M2H = 5-methyl-2-hexanone
Initially stirred for 3 days in N2 atmosphere.
2P = 2-pentanone
Suspension was brought to -25 oC and freeze dried at ~1 mmHg pressure with no additional heat added to the system
Organic phase in top layer was evaporated using a stream of air. Thin film remained on top of aqueous phase
Sample was prepared using C6/Sr2Nb3O10 in 5M2H suspension made 6 months earlier.
7.2.2 Solvent Choice
The premise of this work relies on the ability to weaken the intermolecular forces
between grafted alkyl chains, and thus the need to study the effects of differing solvents is
self-evident. Solvents in samples 1 and 2 were chosen due to their laboratory ubiquity. 5methyl-2-hexanone (5M2H) and 2-propanone were chosen based on their Hansen
Solubility Parameters197 (Table 7.2). Since the grafted alcohols were no longer in their
original form due to the removal of the C-O-H bond and the formation of the C-O-Nb bond,
using the parameter values for the initial alcohol was not ideal. However, since the
propoxyl substructure remains present, certain assumptions can be made. Modification of
the O-H bond would not significantly alter the dispersion forces or dipole-dipole
interactions, but rather it would reduce hydrogen bonding. Therefore, it was believed that
a solvent with similar δD and δP values to that of the initial alcohol, but with a low δH value
would be appropriate for exfoliation. Thus, 5M2H and 2P were the solvents of choice for
C6 and C3 compounds, respectively. It should be noted that there are readily available
solvents other than 2P with closer solubility matches to the propoxyl group such as 1,1dichloroethane and chloromethane (with δD, δp, and δH values equal to 16.6/15.3, 8.2/6.1,
and 0.4/3.9 MPa1/2, respectively197), but the similarity of structure between 2P and 5M2H
allows for easier comparisons between samples.
Table 7.2 Hansen Solubility Parameters for exfoliation solvents and alcohols used during grafting of
parent compounds. Values are given in units of MPa 1/2. Values taken from Hansen.197
Name
δD
δP
δH
1-hexanol
15.9
5.8
12.5
1-propanol
16
6.8
17.4
5M2H
16
5.7
4.1
2-pentanone
15.5
10.4
6.7
161
7.2.4 Sonication
Variable sonication steps were present both before and after heating steps. Initial
sonication was thought to reduce particle size and prepare particles for easier interlayer
solvent introduction. As subsequent sections will illustrate, the initial sonication itself
produced exfoliated materials (albeit not in substantial amounts) and is a vital step in the
optimized exfoliation procedure. The final sonication step was thought to assist in the
exfoliation of stacked sheets that may have solvent in the interlayer, but had yet to break
apart. The efficacy of this final sonication step is ambiguous, but is included due to the
effectiveness of the initial sonication and its ease of implementation.
7.2.5 Sedimentation
Sedimentation behavior is highly dependent on particle size in both lateral and
stacking dimensions.198-200 The most basic sedimentation method was to let samples sit
untouched on the benchtop letting the largest particles settle due to gravitational forces
alone. Centrifugation of samples at 6000 rpm (equivalent to 4303 g-force) for 45 minutes
is the traditional technique of separating exfoliated and unexfoliated nanosheets using the
TBAOH method198, 201 and was thus employed in this work for the same reason. The
supernatant (SN) was carefully pipetted away for further use.
The g-force during
centrifugation should be sufficient such that only exfoliated single nanosheets remain in
solution
162
7.2.6 Extracting Nanosheets Out of Solution
Lyophilizing (or freeze-drying) is the preferred method of collecting single
nanosheets out of solution using the TBAOH method. This technique involves freezing a
suspension and placing it under vacuum such that the system pressure allows for
sublimation until all solvent has been removed and only nanosheets remain. Simply
evaporating the liquid solvent inevitably leads to sheet restacking. The freezing points of
5M2H and 2P are too low (-73.9 and -76.8 oC, respectively) to be used alone in the
lyophilizing system in our lab. Tert-butanol (t-BuOH, m.p. 25.8 oC) has been applied
previously as a co-solvent202-203 to increase solution melting point and was used with
limited success in this work. Attempts at transferring sheets into an aqueous phase were
also tested and will be discussed in greater detail in subsequent sections.
Many suspensions and SN’s were treated by evaporation of the solvent to obtain
materials for characterization. As the following sections will show, solvent evaporation
inevitably leads to restacking of exfoliated sheets. This restacking, however, is disordered
in nature and can be observed in XRD patterns and SEM images, thus serving as evidence
of the presence of exfoliated sheets in solution. The method of evaporation was varied and
will be discussed in greater detail in later sections.
7.3 Visual Observations and Generation of Colored Suspensions
Figures 7.1-8 show sample images at different synthetic stages. These simple
visual observations provide great information regarding the nature of these suspensions.
Sonication alone in an appropriate solvents was the simplest method of producing distinct
163
colloidal suspensions. The image in Figure 7.1 of C6/Ca2Nb3O10 (Sample 4) after stirring
in 5M2H for 3 days in a N2 atmosphere followed by 60 minutes of sonication shows that
after one week untouched, particles are still suspended in solution. This sample showed
little change after sitting untouched for over 2 months. The SN following centrifugation
was clear and colorless with no observed Tyndall Effect. C6/Sr2Nb3O10 (Sample 3)
sonicated 30 minutes in 5M2H without prior stirring also produced a colloidal suspension,
but to a substantially lesser degree than Sample 4. Solvent choice was found to be vital for
producing colloidal suspensions after sonication as seen in Figure 7.2 showing Sample 10
(C6 phase in 5M2H) and Sample 11 (C6 phase in 2P) after 180 minutes of sonication.
Figure 7.1. Image of C6/Ca2Nb3O10 in 5M2H (Sample 4) after sitting unmoved for one week.
164
Figure 7.2. Image of C6/Ca0.25Sr1.75Nb3O10 in 2P (Left – Sample 11), and in 5M2H (Right – Sample 10)
30 minutes after initial sonication. Note that particles in Sample 11 are settling at a much faster rate
than those in Sample 10.
The addition of a microwave heating stage after initial sonication produced colored
suspensions.
Figures 7.3 and 7.4 show C6 and C3/Sr2Nb3O10 (Samples 5 and 8,
respectively) in 5M2H and 2P, respectively, show the distinct color generated after
microwave heating. C6 samples in 5M2H consistently developed strong yellow colors
after microwave heating while C3 samples in 2P samples developed slight yellow tints.
Potential causes of this dramatic color change will be discussed later in this section. The
suspensions remain colored after sitting for several weeks.
165
Figure 7.3. Image of C6/Sr2Nb3O10 in 5M2H (Sample 5) shortly after microwave heating (left) and after
sitting untouched for 1 week (right). The observed Tyndall Effect is indicative of a colloidal suspension
that does not settle out with gravitational force alone.
Figure 7.4. Images for C3/Sr2Nb3O10 in 2P (Sample 8) after sitting untouched for 2 weeks (Left) and 2
months (Right). The colloidal suspension is clearly stable after 2 months even despite approximately
50% of the solvent having evaporated. A slight yellow tint can be observed in the sample after 2 weeks,
but is ambiguous after 2 months.
Sample 15a (C6/Sr2Nb3O10 in 5M2H) was initially prepared with 60 minutes initial
sonication, microwave heating 10x, and 60 minutes sonication. The data presented herein
were collected on the suspension 6 months after initial preparation at which time the sample
was sonicated 60 minutes. 5M2H was periodically added to the suspension to prevent
complete solvent evaporation. As seen in Figure 7.5, the initial yellow color had faded
substantially in subsequent months with only a faint tint remaining. The suspension,
166
however, still clearly contained colloidal particles consistent with sheet exfoliation. The
SN after centrifugation (Figure 7.6 – left) was used to collect suspended nanosheets
through extraction into an aqueous phase. An aliquot of the SN was first thoroughly mixed
with ethanol (miscible in 5M2H) followed by mixing with water (immiscible in 5M2H)
leading to a substantial cloudiness being produced in the aqueous phase (Figure 7.6-right),
possibly due to the presence of particles extracted into the aqueous phase.
The
effectiveness of this technique is discussed in section 7.7.
Figure 7.5. Image of C6/Sr2Nb3O10 in 5M2H (Sample 15a) having sit for 5 hr after initial sonication.
Note the yellow color seen in Sample 12 from which it was prepared is barely present, if present at all.
167
Figure 7.6. Image of C6/Sr2Nb3O10 in 5M2H after centrifugation (Sample 15a – Left) and after mixing
with ethanol and H2O (Sample 15b – Right). The organic layer in Sample 15b is on top and aqueous
layer on bottom. Note the substantial cloudiness present in the aqueous layer in Sample 15b.
As section 7.4 will show, since the sediment following exfoliation treatment still
contained grafted material, it was believed that the sediment could undergo a second
exfoliation treatment to produce more nanosheets from the parent sample. Samples 14c
and 15c (C3/Sr2Nb3O10 in 2P and C6/Sr2Nb3O10 in 5M2H) were prepared using the
sediment of samples 14a and 15a, respectively. Figure 7.7 shows the SNs of 14c and 15c
after centrifugation at which point their appearance is consistent with the initially prepared
samples. It should be noted that 15c had a strong yellow color that had faded in the original
sample. Furthermore, Figure 7.8 shows that the Tyndall Effect was observed in the SN
after centrifugation. In all previous samples the Tyndall Effect was observed in the initial
suspensions, but was not observed after centrifugation due either to decreased sizes of
particles present or lower concentration of particles. This would suggest the first treatment
of the sample primed the sediment for later exfoliation
168
Figure 7.7. Image of C3/Sr2Nb3O10 in 2P 2nd exfoliation SN (Sample 14c) and C6/Sr2Nb3O10 in 5M2H
2nd exfoliation (Sample 15c). Note the yellow color in Sample 15c returned after microwave heating
Figure 7.8. Image of C6/Sr2Nb3O10 in 5M2H SN (Sample 15c) with laser beam passing through.
Observance of the Tyndall effect is evidence of a colloidal suspension. This is the first instance in which
the Tyndall Effect was observed after centrifugation.
Considering the role of the solvent in this new exfoliation method was initially
considered to be only as a medium in which to suspend nanosheets, it is tempting to simply
brush aside the emergence of color in suspensions. The origin of this color, however, is
169
clearly related to the properties of layered niobates, and the absence of any other reported
observation of such change in both published reports and informal discussions with other
researchers in the field indicate that new insights into the nature of these materials are ready
to be illuminated.
Considering the oft-touted potential of niobate nanosheets as catalytic materials,
the possibility of a chemical reaction generating a colored product needed to be addressed.
UV-visible spectra for 5M2H untreated, a control sample of 5M2H sonicated and
microwave heated, and C6/Sr2Nb3O10 in 5M2H are shown in Figure 7.9. The untreated
and control samples show no significant differences while the nanosheet sample absorbs
into the visible spectrum consistent with the observed yellow color. The diffuse reflectance
UV-visible spectrum of the C6/Sr2Nb3O10 parent compound (Figure 7.10) shows
absorption at lower wavelength than 5M2H confirming a new component in present in the
sample, though its identity is not yet clear.
5.9
Absorbance
4.9
3.9
2.9
1.9
0.9
-0.1
200
250
300
350
400
450
500
550
600
Wavelength (nm)
5M2H
Control
nano no butanol
Figure 7.9. UV-vis spectra for untreated 5M2H (Blue), 5M2H control sample (Orange), and Nanosheet
suspension (yellow).
170
C6/Sr2Nb3O10
2.5
K-M
2
1.5
1
0.5
0
250
300
350
400
450
Wavelength, nm
500
550
600
Figure 7.10. UV-vis spectrum of C6/Sr2Nb3O10 parent compound.
Figure 7.11 shows comparisons of IR spectra for untreated solvents and the
resulting colored SNs. The suspensions of exfoliated niobates produced no new IR signals
nor do solvent signals show any shifting outside of instrumental error. Furthermore,
comparisons of 1H NMR spectra (Figure 7.12) of untreated solvent, control samples
(solvent treated without the presence of niobates), and colloidal suspension SNs show no
signals except for those from the solvent. The lack of significant differences in IR and
NMR spectra of these samples, however, does not conclude the absence of a chemical
reaction due to limits of detection, but rather provides no additional evidence as the to the
nature of the yellow color.
171
IR Spectra
100
95
5M2H/t-BuOH
85
2P
80
C6 in 5M2H/t-BuOH (Sample 15a)
75
C3 in 2P (sample 14a)
70
%T
90
65
60
3800
3400
3000
2600
2200
1800
1400
1000
600
Wavenumber, cm-1
Figure 7.11. IR spectra for 5M2H/t-BuOH mixture untreated (Blue), 2P untreated (orange),
C6/Sr2Nb3O10 in 5M2H/t-BuOH mixture (grey), and C3/Sr2Nb3O10 in 2P (yellow).
Figure 7.12. 1H NMR spectra for 5M2H untreated, 5M2H control sample, and C6/Sr 2Nb3O10 in 5M2H
SN.
172
7.4 XRD Results
As previous chapters have shown, XRD is an incredibly useful tool for structural
characterization of layered perovskites. Stacking reflections seen in patterns collected with
a 2θ range of 1.5-14o are particularly useful in this work due to their shifting positions
depending on the composition of the interlayer gallery (see Chapter 6 for a more detailed
discussion) and thus relate to the stacking motif of the sample (i.e. ordered vs. disordered
stacking, particle size, presence of alkyl groups, etc.…) A summary of short-range XRD
pattern results including stacking distance, corresponding phases, and mean particle size in
the stacking dimension can be found in Table 7.3. Mean particle size relates to particle
size in the ordered domain and was calculated using Scherrer’s equation. Long range
patterns were also employed to both confirm the perovskite structure is intact after
exfoliation as well as to study the long-range order of the samples. XRD patterns were
collected for the sediment after the initial preparation of colloidal suspensions, suspensions
dried onto glass slides after settling (and before centrifugation), and for solids collected
from SNs after centrifugation. Several methods were employed for collecting solids from
SNs and will be discussed in greater detail in subsequent sections.
173
Table 7.3. XRD Stacking Reflections Summary
Sample
Reflection Position Stacking Distance (Å)
(2θ Degrees)
13 - SN
5.65
15.63
Phase
Protonated
Mean Particle
Size (Å)
140
14 - SN
5.70
15.50
Protonated
160
8 - SN
5.68
15.55
Protonated
160
5 - SN
5.64
15.66
Protonated
40
9 - Sus day of
4.30
20.54
C3
80
5.75
15.36
Protonated
140
9 - Sus 2 days
6.00
14.72
Protonated
160
7 - FS Sus
3.94
22.41
C6
110
5.74
15.39
Protonated
140
7 - SS Sus
5.74
15.39
Protonated
140
5 - Sus
3.52
25.09
C6
180
4 - Sus
3.54
24.95
C6
80
5.76
15.34
Protonated
200
15c Sed
5.80
15.23
Protonated
140
14c Sed
4.80
18.40
C3
70
5.76
15.34
Protonated
200
3.02
29.24
C6
90
5.58
15.83
Protonated
110
3.70
23.87
C6
110
5.65
15.63
Protonated
140
3.12
28.30
C6
190
6 Sed
5 Sed
4 Sed
7.2.3 Heating
The results of the grafting work discussed in Chapter 6 clearly illustrate the
important role of heating in introducing organic groups to the interlayer gallery. Therefore,
it was thought that additional thermal energy would allow solvent to enter the interlayer
gallery with greater ease. Microwave heating was employed using a heating cycle of
5x(45s on 15s off) repeated a varied number of time with 15-20 minute cooling intervals
between heating cycles.
It should be noted that while 2P has a higher microwave
susceptibility than 5M2H,177 the same heating cycle was employed with both solvents.
174
This was done to prevent overheating of the acid digestion vessel and thus 2P will reach
higher temperatures than 5M2H during microwave heating.
7.4.1 Sediments of Suspensions
Figures 7.13-17 show XRD patterns for Samples 4 (C6/Ca2Nb3O10 in 5M2H), 5
(C6/Sr2Nb3O10 in 5M2H), 6 (C6/Sr2Nb3O10 in 5M2H), 14c (C3/Sr2Nb3O10 in 5M2H), and
15c (C6/Sr2Nb3O10 in 5M2H), respectively. The presence of sharp and intense stacking
reflections associated with the C6 phase in Figures 7.13 and 7.14 strongly suggest the
suspension sediments can be reused in another exfoliation attempt because of the remaining
presence of alkyl groups in the interlayer of unexfoliated sheets. Furthermore, Figure 7.15
shows the XRD pattern for a C6/Sr2Nb3O10 in 5M2H suspension that, due to benign
neglect, sat untouched for over one year after it was prepared (the sample was never
centrifuged). During this time, the 5M2H solvent slowly evaporated until only dry powder
remained on the bottom of the container. The presence of a C6 stacking reflection is a
testament to the relative stability of the C6 phase (whereas C3 phases quickly revert to
hydrated states within days) and suggests the C6 phase is better suited for exfoliation. This
is also supported by the XRD patterns in Figures 7.16 and 7.17 for sediments of 14c and
15c, respectively. The presence of a C3 stacking reflection in 14c and the absence of a C6
stacking reflection in 15c, combined with visual observations (see Figures 7.7 and 7.8) is
indicative of a larger fraction of the total sample having been exfoliated.
175
C6/Ca2Nb3O10 Sediment Sample 4
2
4
6
8
10
12
14
2θ (Deg)
Figure 7.13. XRD pattern for C6/Ca2Nb3O10 in 5M2H sediment collected from the bottom of the sample
container (Sample 4).
C6/Sr2Nb3O10 in 5M2H Sediment Sample 5
2
8
14
20
26
32
38
44
50
56
2θ (Deg)
Figure 7.14. XRD patterns for C6/Sr2Nb3O10 in 5M2H (Sample 5) slow air-dried supernatant (top) and
the powder sediment (bottom).
176
C6/Sr2Nb3O10 in 5M2H (Sample 6) Suspension Dried Over
1 year
2
4
6
8
10
12
14
2θ (Deg)
Figure 7.15. XRD pattern of C6/Sr2Nb3O10 in 5M2H (Sample 6) after it had sat on the benchtop
untouched for over 1 year. All solvent had evaporated leaving solid powder on the bottom of the
container. Not the C6 phase is still present.
C3/Sr2Nb3O10 in 2P (Sample 14c) SN Sediment
3
5
7
9
11
13
2θ (Deg)
Figure 7.16. XRD pattern for C3/Sr2Nb3O10 in 2P SN sediment (Sample 14c). Note the presence of
both C3 and hydrated phases at 4.8 and 5.7 o, respectively.
177
C6/Sr2Nb3O10 in 5M2H (Sample 15c) SN Sediment
2
4
6
8
10
12
14
2θ (Deg)
Figure 7.17. XRD pattern for C6/Sr2Nb3O10 in 5M2H (Sample 15c) Sediment. The stacking reflection
at 5.8o is consistent with a partially hydrated phase. The slight plateau between 3.5 and 4.7o is possibly
indicative of remaining C6/Sr2Nb3O10.
7.4.2 Dried Suspensions
The fundamental premise of this exfoliation is built upon the necessity for alkyl
grafting into the interlayer gallery in order for organic solvent to be employed, and it is
naturally important to determine the behavior of these groups upon exfoliation. After
sitting for over one week, several drops of suspension from Samples 4 and 5
(C6/Ca2Nb3O10 and C6/Sr2Nb3O10, respectively, in 5M2H) were air dried onto a glass plate
and used to obtain the XRD patterns shown in Figure 7.18. Both patterns show broad
stacking reflections due to the C6 phase consistent with either reduced particle size from
the exfoliation procedure and/or disordered restacking of nanosheets during drying.
Sample 4 also showed a small stacking reflection due to the hydrated phase. As subsequent
results will show, the point at which reversion to the hydrated phase and the degree to
which this occurs are both ambiguous and difficult to accurately determine experimentally
178
C6/Ca2Nb3O10 in 5M2H Dried Suspension Sample 4
2.5
4.5
6.5
8.5
10.5
12.5
2θ (Deg)
C6/Sr2Nb3O10 in 5M2H Dry Suspension Sample 5
2.5
4.5
6.5
8.5
10.5
12.5
2θ (Deg)
Figure 7.18. XRD pattern for C6/Ca 2Nb3O10 in 5M2H suspension (Sample 4) and C6/Sr2Nb3O10 in
5M2H suspensions (Sample 5) obtained by air-drying several drops of the suspension (before
centrifugation) onto a glass plate. The reflections at ~3.6 and ~5.8 degrees correspond to C6 grafted
and hydrated phases, respectively. Note the absence of a strong hydrated phase reflection in Sample
5 is possibly due to differences in the suspension preparation (sonication vs. microwave heating),
differences in the procedure employed for synthesis of the parent compound (conventional heating for
Sample 4 vs. microwave heating for Sample 5), or differing composition of the parent compound.
Based on the success of adding a microwave heating step, a sample was prepared
using a conventional heating step to serve as a comparison. Sample 7 was prepared by
mixing C6/Sr2Nb3O10 with 5M2H and heating in a steel bomb for 3 days at 150 oC.
Immediately after transferring the sample to a glass vial post-heating, it was observed that
along with some of the powder settling out of solution within minutes, there was a distinct
179
portion of suspension settling out over the course of hours (designated moderate settling
phase) after which the suspension remained stable (designated slow settling phase). Using
a pipette, several drops were taken from the fast settling and slow settling phases and dried
on glass slides for comparisons with XRD (see Figure 7.19). The fast settling phase
contained an intense reflection related to the C6 phase and a smaller reflection related to
the hydrated phase while the slow settling phase showed only a hydrated phase reflection.
This suggests a greater degree of exfoliation results in reversion to the hydrated phase.
Considering exfoliation leads to more accessible sheet surface sites, it is logical that grafted
alkyl groups would then be more easily removed. However, the presence of C6 grafted
phase in dried suspensions prepared using microwave heating, even after sitting untouched
for long periods of time, indicates that rapid heating reduces the susceptibility to alkyl
group loss. The fate of the grafted alkyl groups will be discussed further in section 7.7.
180
C6/Sr2Nb3O10 in 5M2H Sample 7 Slow Settling Phase
2
4
6
8
10
12
14
2θ (Deg)
C6/Sr2Nb3O10 in 5M2H Sample 7 Moderate Settling Phase
1.5
3.5
5.5
7.5
9.5
11.5
13.5
2θ (Deg)
Figure 7.19. XRD patterns for C6/Sr2Nb3O10 in 5M2H (Sample 7) from the slow settling phase (Top)
and moderate settling phase (Bottom).
As the results of the initial grafting work show (see Chapter 6), the longer alkyl
chains in C6 compounds create a more hydrophobic interlayer environment leading to
substantially less reversion to the hydrated phase over time when compared to C1 and C3
compounds. This trend is also seen with exfoliated samples. Figure 7.20 shows XRD
patterns of dried suspensions of C3/CaSrNb3O10 in 2P (Sample 9) prepared the same day
the suspension was prepared (Bottom pattern) and 2 days after sitting untouched (Top
pattern). While the same day pattern shows both a C3 phase and a hydrated phase, after 2
181
days the C3 stacking reflection has been reduced almost completely while the hydrated
stacking reflection has increased intensity. This further suggests C6 grafted compounds
are better suited for efficient exfoliation.
C3/CaSrNb3O10 in 2P Sample 9 Dried Suspension
(Two Weeks)
2.5
4.5
6.5
8.5
10.5
12.5
2θ (Deg)
C3/CaSrNb3O10 in 2P Sample 9 Dried Suspension
(Day of Synthesis)
2.5
4.5
6.5
8.5
10.5
12.5
2θ (Deg)
Figure 7.20. XRD pattern for C3/CaSrNb3O10 in 2P (Sample 9) dried suspension. The pattern on top
is from a dried suspension prepared two weeks after the original synthesis while the bottom pattern is
from the day of the synthesis. The lack of a grafted stacking reflection, in the dried suspension
prepared after sitting two weeks, and increased intensity of the hydrated stacking reflection suggest
sheets in the suspension are reverting from the grafted state to the hydrated state.
182
7.4.3 Solid From SN
While the lack of particle sedimentation in these colloidal suspensions after sitting
untouched over long periods of time is consistent with successful exfoliation into single
nanosheets, partially exfoliated sheets (still stacked, but to a lesser degree) can also remain
suspended if only the force of gravity is inducing sedimation. To ensure only fully
exfoliated sheets are present in solution, samples were centrifuged at 6000 rpm for 45
minutes. The centrifugal force of 4303 g under these conditions induces sedimentation of
partially exfoliated sheets while fully exfoliated sheets remain in the supernatant. Absence
of an observed Tyndall Effect in the supernatant of all samples except Sample 15c (a 2nd
exfoliation) confirmed that partially exfoliated sheets were present in the original
suspensions. To determine whether nanosheets were still present in the supernatant at all,
the solvent was removed and the remaining solid studied with XRD.
Four methods of solvent removal were tested: (1) Rapid evaporation by dropping
SN onto a glass plate heated to 80 oC; (2) Slow evaporation by dropping SN onto a glass
plate at room temperature; (3) Slow evaporation in a N2 atmosphere; and (4) freeze-drying
the supernatant such that solvent is sublimed away. As seen in Figure 7.21, rapid drying
of the SN resulted in the observance of broad low intensity reflections in the XRD pattern
associated with the hydrated phase. Simulation of the (001) stacking reflections as a
function of the number of sheets stacked together204 (shown in grey in Figure 7.21) matches
with the observed stacking reflection assuming an average number of 2 sheets are stacked
together. The simulated pattern was calculated using the Laue function:205
183
2
Equation 7.1: 00 (ℎ) =
2 (
)
1+2 2

|00 ()|2
2
2
 
2 (
)

where N is the number of stacked sheets and F00l is the structure factor, calculated using a
tetragonal Sr2Nb3O10 crystal structure. Reflections at 23.1o (3.85 Å), 28.7o (3.10 Å), and
32.7o (2.74 Å) correspond to lateral perovskite dimensions confirming the perovskite
structure remains intact throughout the exfoliation process. No discernable stacking
reflections associated with the C6 phase were observed.
C6/Sr2Nb3O10 in 5M2H SN Sample 5 Rapid Drying
23.1O
28.7O
32.7O
Supernatant
Fit
Blank Glass Plate
0
450
2
6
10
14
18
22
26
30
34
38
42
46
2θ Deg
Figure 7.21. XRD pattern for C6/Sr2Nb3O10 in 5M2H centrifuge supernatant (Sample 5) after drying.
The sample was rapidly dried by placing drops of the supernatant onto a glass plate heated to 80 oC
such that solvent evaporated within seconds until a significant amount of solid remained. The broad
hump in the pattern is due to the background signal of the glass plate (blank plate shown in blue). The
grey pattern is a simulation of (00l) stacking reflections assuming an average number of 2 hydrated
sheets stacked together.
184
Figures 7.22 and 7.23 show XRD patterns for SNs of C3/Sr2Nb3O10 in 2P (Sample
8) slowly dried in air and C6/Sr2Nb3O10 in 5M2H (Sample 13) dried in a N2 atmosphere,
respectively. Compared to the rapid drying method, reflections in the XRD patterns of
these samples are much sharper. This indicates slow evaporation leads to more ordered
restacking of the exfoliated sheets, although they are still significantly more disordered
than the parent compounds. It was thought that drying in a N2 atmosphere would protect
sheets from reverting to the hydrated phase (assuming grafted alkyl groups are still present
in the SN), but again only the hydrated phase was observed.
C3/Sr2Nb3O10 in 2P Sample 8 SN
002
200
220
004
3
9
15
21
27
33
39
45
51
57
2θ (Deg)
C3/Sr2Nb3O10 in 2P Sample 8 SN
002
004
2
4
6
8
10
12
14
2θ (Deg)
Figure 7.22. Long-range (Top) and short-range (Bottom) XRD patterns for C3/Sr2Nb3O10 in 2P
(Sample 8) supernatant after sitting for 2 weeks. Only the hydrated phase is observed with fewer
higher order reflections compared to parent compound. Note the slight hump at ~3 o is due to the glass
slide being place in a position such that it blocked the X-Ray beam below 3o.
185
C6/Sr2Nb3O10 in 5M2H Sample 13 Dried SN
3
13
23
33
43
53
2θ (Deg)
Figure 7.23. Long-range XRD pattern of C6/Sr2Nb3O10 in 5M2H (Sample 13) supernatant dried in a
N2 atmosphere. The lack of many higher order reflections is consistent with irregular restacking of
sheets, though the stacking reflection position is consistent with a hydrated phase rather than the
desired grafted phase.
As discussed in section 7.2, freeze-drying of nanosheet suspensions has been used
in previous work to successfully extract nanosheets out of solution without restacking.
Figure 7.24 shows the resulting XRD pattern for the SN of C6/Sr2Nb3O10 in 5M2H (Sample
14), freeze-dried using t-butanol as a co-solvent. As will be discussed further in section
7.7, the freeze-drying process for these exfoliated systems presents inherent difficulties and
has not yet been optimized. As a result, at several points in the freeze-drying process, the
sample began to liquify and was refrozen. This is the likely cause of the restacking
observed in the XRD pattern. As with the other methods of solvent removal, only the
hydrated phase was observed in the final product.
186
C6/Sr2Nb3O10 Sample 14 Freeze-Dried Product
3
13
23
33
43
53
2θ (Deg)
Figure 7.24. Long-range XRD pattern for C6/Sr2Nb3O10 after freeze-drying in a 5M2H/t-BuOH
mixture (Sample 12). The presence of a sharp stacking reflection and higher order reflections is likely
due to the frozen solution melting several times during the freeze-drying process, essentially resulting
in a slow evaporation of solvent and thus allowed for ordered restacking of sheets.
7.5 SEM and EDS Results
Figure 7.25 shows an SEM image of C6/Sr2Nb3O10 in 5M2H (Sample 3) treated
with only sonication. The typical layered niobate parent compounds exist as thin platelets
with lateral dimensions of 1-3 um and platelet thickness of several hundred nanometers
(see chapter 4 for figures and discussion). After sonication, Figure 7.26 clearly shows that
particle size has been reduced dramatically such that the maximum particle size observed
became 681 and 154 nm in the lateral and stacking directions, respectively, while most
particles are too small and irregular to be clearly pictured and/or measured. This is
consistent with generation of a colloidal suspension. Furthermore, SEM data for the SN
of C6/Ca2Nb3O10 in 5M2H (Sample 4, Figure 7.26) showed that after centrifugation, only
irregularly shaped particles lacking the typical platelet shape with lateral dimensions less
than ~500 nm and a thickness too small to be measured with precision are observed. This
187
is consistent with irregular restacking of sheets observed in several other studies on niobate
nanosheets.182, 188, 198
Figure 7.25. SEM image of C6/Sr2Nb3O10 in 5M2H (Sample 3) on TESCAN instrument. While the
particles have the platelet shape seen in parent compounds, the particle size is substantially smaller
with 681 nm being the largest lateral dimension seen in the sample. Plate thickness of 67 and 154 nm
correspond to ~40 and ~100 sheets stacked together, respectively.
Figure 7.26. C6/Ca2Nb3O10 in 5M2H SN (Sample 4 sitting unmoved for three weeks, Left) and
C6/Sr2Nb3O10 in 5M2H SN (Sample 6, Right) SEM images. Sample was prepared by placing several
drops onto SEM sample holder and letting air dry. Sample 6 on the right was prepared over a much
longer period with ~1mL of SN in total dried onto the Si wafer. Note the substantially increased
irregularity and decreased size of the particles compared to parent compounds.
188
Of course, considering that even the slightest agitation after centrifugation can lead
to sedimentary particles entering the supernatant, combined with the irregular shape of
particles observed, it was necessary to use a more thorough procedure to prevent
contamination and confirm the particles were indeed niobates.
The supernatant of
C6/Sr2Nb3O10 in 5M2H (Sample 6), after centrifuging for 45 min at 6000 rpm, was rapidly
dried under heat, followed by dispersing the solid residue in 10 mL of 5M2H and
centrifuging again for 45 min at 6000 rpm. This final supernatant was placed dropwise
onto a Si wafer for SEM analysis (see Figure 7.27). The presence of Nb in the EDS
spectrum combined with a stacking reflection seen in the XRD pattern (Figure 7.21) for
the same sample is strong evidence that exfoliated nanosheets were successfully produced.
cps/eV
C6Sr2Nb3O10 5-methyl-2-hexanone supernatant 2
6
5
4
Nb
C
O
Na
Si
Nb
3
2
1
0
1
2
3
keV
4
5
Figure 7.27. SEM image (top) and EDS spectrum (bottom) for C6/Sr2Nb3O10 in 5M2H (Sample 5)
supernatant after evaporated solvent, re-dissolving, centrifuging, and final evaporation of solvent. The
small and irregular shape of the particles combined with the presence of Nb in the EDS spectrum is
strong evidence for successful exfoliation of nanosheets. Note that there is no Sr signal assigned in the
EDS spectrum due to the Si signal from the wafer overlapping.
As Figure 7.27 above shows, while Nb is observed in the EDS spectrum, Sr cannot
be observed due to the large Si signal. To overcome this, a new sample preparation method
was employed. A piece of double sided copper tape was placed on an aluminum SEM
189
6
sample holder. Acetone was used to strip away the adhesive on the copper, due to the
presence of Si in the adhesive, and the nanosheet suspension was air dried dropwise onto
the copper at room temp (~1mL added in total. As was expected, the resulting SEM images
contained irregularly shaped particles <1 micrometer in length, consistent with previous
images and the corresponding EDS spectra contain both Nb and Sr signals with the desired
compositional ratios (when error is considered – see Table 7.4).
a
b
Figure 7.28. SEM images for (a) C3/Sr2Nb3O10 in 2P (sample 14c) SN, (b) C6/Sr2Nb3O10 in 5M2H
(sample 15c) SN. Note the horizontal lines in image (a) are to due scratches on the copper tape
background created during removal of the adhesive layer.
190
cps/eV
8
C6Sr2 5M2H 106a 4
7
6
5
Nb
4
C
O
Cu
Sr
Nb
3
2
1
0
0.5
1.0
1.5
2.0
2.5
keV
3.0
3.5
4.0
4.5
5.0
Figure 7.29. EDS spectrum of C6/Sr2Nb3O10 in 5M2H SN (Sample 15c).
Table 7.4. Atomic composition of samples 14C and 15C using copper tape as substrate. The letter in
parentheses indicates to which image in Figure 7.28 above the data correspond. Error was
automatically calculated by the Quantax EDS application.
Sample
Nb%
error
Sr%
error
14c (a)
0.96
0.2
0.73
0.1
15c (b)
2.97
0.9
2.01
0.5
In an effort to overcome SEM resolution problems when imaging nanosheets as
well as expand the potential applications of these materials, sample 15c SN was treated
with a suspension of Cu2O nanocubes in order to deposit the nanosheets on the cube
surface. Approximately 1 mL of both nanosheet SN (C6/Sr2Nb3O10 in 5M2H) and
nanocube suspension were mixed together vigorously. One drop of this mixture was placed
on a Si wafer and left to air dry, followed by gold coating and SEM imaging. SEM images
(Figures 7.30 and 7.31) show the presence of nanosheets on the Cu2O nanocube surface.
The image in 7.30 was taken with a lower beam voltage leading to decreased resolution,
191
but due to the tendency for the electron beam to pass through sheets at higher power,
significant aggregation of sheets was observed with lateral dimensions less than 500 nm
and irregular shapes. Figure 7.31 shows an image using higher beam voltage and thus
improved resolution, although the sheets have an almost transparent appearance due to the
aforementioned effect of thin particles.
Figure 7.30. SEM image of C6/Sr2Nb3O10 nanosheets deposited onto a Cu2O nanocube. The small (less
than 500 nm), irregularly-shaped particles are the exfoliated/restacked nanosheets. Note that there is
a wide distribution of nanosheet agglomeration.
192
Figure 7.31. SEM image of C6/Sr2Nb3O10 nanosheets deposited onto a Cu2O nanocube. The small (less
than 500 nm), irregularly-shaped particles are the exfoliated/restacked nanosheets. Note that the
almost transparent quality of nanosheets on the cube surface is consistent with the electron beam
passing through the thin sheet.
7.6 TGA Results and Exfoliation Efficiency
To quantify the degree of exfoliation, both 5M2H and 2P samples, along with a
suspension produced using the conventional TBAOH exfoliation, underwent Thermal
Gravimetric Analysis. By measuring the initial mass of a droplet of each sample along
with the final mass after heating to 200 oC (along with an isothermal stage to ensure all
solvent had been evaporated) the w/v concentration in the original solution was calculated.
Using that concentration and the theoretical maximum concentration assuming 100%
exfoliation it was found the 5M2H sample underwent 25% exfoliation, the 2-pentanone
193
sample underwent 28% exfoliation, and the conventional TBAOH sample underwent 22%
exfoliation.
7.7 Discussion
7.7.1 Role of Solvent and Parent Compound
While both C3 and C6 parent compounds were successfully exfoliated, the
tendency for C3 compounds to revert to a hydrated state at a much faster rate suggests that
the C6 phase is better suited for successive exfoliations. As the TGA data show, initial
exfoliation at best leads to ~25% of the sample being successfully exfoliated. Thus
successive exfoliations have clear benefits. While the microwave grafting technique
developed in this work was not successful for long chain alcohols beyond 1-hexanol, longer
chain alcohols would likely lead to greater interlayer hydrophobicity and thus further
prevent hydration.
Choice of solvent for exfoliation is less clear and more dependent on the goals of
the exfoliation. While any solvent used would need to be a relatively close HSP match,
the results of the same C6 parent compound using either 2P or 5M2H (samples 10 and 11,
respectively) show that increased microwave susceptibility can overcome deviations from
ideal HSP values in terms of fraction of sample exfoliated. However, the suspension color
generated after microwave heating was substantially more dramatic in 5M2H for all
samples tested. While the origins of this change need further elucidation, 5M2H is superior
to 2P should generation of a colored suspension be proven desirable. Testing a wider
variety of solvents is undoubtedly a desirable route for future research into this technique.
194
7.7.2 Role of Sonication
Sonication was found to be very useful in the exfoliation procedure. The initial
sonication step, in and of itself, produces colloidal suspensions consistent with exfoliation
and is important to improved exfoliation efficiency. This is particularly true of samples
where the solvent is a close HSP match to the grafted parent compound. The importance
of initial sonication for samples without a close HSP match is more ambiguous and needs
further study to determine its necessity. Post-heating sonication has not been shown to
significantly affect exfoliation behavior for better or for worse. Final sonication was
included in the optimized procedure due to its simplicity, relatively short time, and the little
attention required by the operator during this step.
7.7.3 Role of Microwave Heating
As with the grafting procedure, microwave heating was found to be key to efficient
exfoliation. Exfoliated suspensions consistently developed noticeably more opaqueness
post-heating as well as producing more opaque suspensions when heated without initial
sonication than with sonication alone. Furthermore, only samples with a heating stage
produce colored suspensions. Microwave heating is also preferable to conventional
heating not only due to its substantially more rapid nature, but also because conventional
heating resulted in more susceptibility to hydration thus inhibiting the effectiveness of
repeated exfoliations. The stability of the perovskite structure after undergoing such
195
extreme conditions of high temperature and pressure highlight the broad range of systems
in which these nanosheets can be used.
7.7.4 Treatment of Supernatant
The supernatant treatment methods used in this work including air-drying, drying
under heat, drying in a N2 atmosphere, and freeze-drying have been useful in obtaining
indirect experimental confirmation of successful nanosheet exfoliation in solution.
However, collecting sheets out of solution using these techniques has inevitably led to
restacking, albeit in a disordered fashion.
Potential applications of this exfoliation
technique are by no means limited to nanosheets out of solution, but the ability to collect
sheets out of solution without restacking would certainly increase the technique’s potential.
As has been discussed previously, freeze-drying is the most widely used method of
collecting exfoliated samples, but the freezing points of the solvents used in this work are
too low to be used in our freeze-drying set up. Indeed, most freeze-drying systems are not
equipped to handle such compounds, but the use of a co-solvent has been applied with
much success to systems with similar limitations.
7.7.5 Fate of the Grafted Alkyl Groups
One effect of exfoliation of layered niobates into single nanosheets is dramatically
increased access to the perovskite surface sites. This is particularly useful in their use as
solid acid catalysts due to the highly acidic protonated environment on the surface of the
sheets. In these grafted materials, however, this means alkyl groups are more easily
196
accessed for potential removal by hydration. This is not inherently disadvantageous as
alkyl group loss produces new protonated environments that can be accessed, but if alkyl
group loss is unavoidable then it is a limitation to the controlled design of nanosheets in
the sense that initially grafted groups can be further reacted. However, it is possible that
observations of alkyl group loss in this work are the result of unoptimized reaction
conditions or collection procedures.
Evidence of alkyl group loss and subsequent generation of the protonated site in
this work is seen most clearly in XRD patterns. As shown in section 7.4, each successive
stage in the exfoliation process from the parent compound to the initial colloidal suspension
to the final centrifuged SN shows progressive loss of the grafted stacking reflection and
gain of the hydrated phase stacking reflection. However, it is not evident whether alkyl
group loss occurs during the exfoliation itself or whether it is the result of solvent removal.
While 1H and 13C NMR have been used to distinguish between grafted alkyl groups and
nongrafted alcohols by Sugahara et al.,64-66, 68 the dilute nature of the suspensions would
make observation of the relevant signals difficult. Furthermore, if a successful freezedrying method can be implemented, it may be possible to observe alkyl groups on extracted
nanosheets using the same methods used in chapter 6.
There also exists the possibility that alkyl groups are still present after solvent
removal, but are unobserved in subsequent XRD patterns. The flexibility of the alkyl
groups within the interlayer can be most clearly illustrated by the range of positions where
the C6 phase stacking reflections are observed (See Table 7.2). Should grafted sheets be
restacking in excessive disorder, a clear stacking reflection may not be readily observable
197
in XRD patterns. A test of this hypothesis requires an attempted 1-hexanol grafting of the
final product post solvent removal.
7.8 Conclusions
Using rapid microwave heating generating high temperatures and pressures, a new
method of nanosheet exfoliation from grafted parent compounds in organic solvents has
been successfully developed. Visual observations of colloidal suspensions not settling
under gravitational force is strongly indicative of the presence of small exfoliated (or
partially exfoliated at this point) particles and XRD patterns confirm the presence of the
parent layered niobates. Centrifugation at 6000 rpm for 45 minutes served to remove any
partially exfoliated sheets from solution. After removing the exfoliation solvent from the
SN, SEM images of irregularly shaped particles, the presence of Nb in EDS spectra, and
highly disordered stacking reflections in XRD patterns, along with observed Tyndall
effects in centrifuged suspensions, illustrate that layered niobates have been successfully
exfoliated into nanosheets. Furthermore, generation of a colored suspension serves as
inspiration for future studies into the behavior of nanosheets in organic suspensions.
The optimal exfoliation procedure found in this work consisted of an initial
sonication stage, a microwave heating stage using an acid digestion bomb, followed by a
final sonication stage. Varying combinations of solvent and parent grafted compound show
this method has the potential to be applied to a broad range of systems. Not only do TGA
data indicate the efficiency of this method is comparable to the conventional exfoliation
method using TBAOH, but also that unexfoliated material after initial treatment has been
198
primed for exfoliation in subsequent treatments. The rapid nature of this exfoliation
method, use of organic solvents, and exfoliation efficiency show it to be a promising
alternative to current exfoliation methods that broaden the range of systems niobate
nanosheets in which niobate nanosheets can be applied.
199
Chapter 8: Summary and Future Directions
The work detailed in this dissertation has been a synergistic combination of the
synthesis and the characterization of D-J Niobates. The compounds studied include
Rb(H)CaxSr2-xNb3O10, Rb(H)BaxSr2-xNb3O10, and (C1/C3/C6)/CaxSr2-xNb3O10 for x = 0-2,
and RbCa2NaNb4-xTaxO13 for x = 0 and 1.2. The synthetic techniques employed include
rapid microwave heating, MSS, acid exchanging, alcohol grafting (both conventional the
newly developed rapid microwave assisted method), and exfoliation (both using TBAOH
and the newly developed organic solvent method). Samples were characterized using
XRD, SEM, EDS, IR, UV-visible, and both liquid- and solid-state NMR (for 1H, 23Na, and
93
Nb nuclei). Using these synthetic and characterization methods, new and invaluable
insights into the structure and behavior of D-J Niobates have been elucidated. Should only
the most central results of this work be instilled upon the reader, they are as follows:
1.
93
Nb NMR has exhibited heterogeneous cation distribution at the interstitial A-site
in RbCaxSr2-xNb3O10 and RbBaxSr2-xNb3O10 compounds.
This contrasts the
observed homogeneity in the corresponding XRD patterns (both collected in this
work and in studies elsewhere). In fact, while many bulk/long range properties of
these compounds are consistent with homogeneous cation distribution, the local
structures observed by NMR show a clear heterogeneity. This dichotomy of cation
distribution not only explains the grafting results presented in this work, but
provides insight into the potential intelligent design of future perovskite materials.
2.
93
Nb and
23
Na NMR was employed to determine cation site ordering within the
four-layer Rb(H)Ca2NaNb4-xTaxO13 structure. Despite the multiple overlapping
200
signals in static NMR spectra, individual site parameters could be obtained using a
combination of MQMAS, MQMAS-t2-REDOR, and BRAIN-CP experiments as
well as collecting static spectra at multiple magnetic field strengths.
It was
ultimately determined that Ta has a strong site preference for interior octahedra.
The increased ionic character of Ta-O bonds compared to Nb-O bonds reduces the
ability to undergo second-order Jahn-Teller distortions and thus it is expected Ta
would prefer the more symmetrical interior octahedral site.
3. A rapid microwave-assisted method of grafting n-alcohols up to hexanol into the
perovskite interlayer gallery was developed that reduced synthesis time by 97%
when compared to the conventional heating method without substantial loss of alkyl
group coverage. It was found that while HCa2Nb3O10 does not undergo rapid
grafting beyond the methylated stage, substitution of as little as 10% Sr into the Asite imbues compounds with grafting ability equal to that of the pure HSr2Nb3O10
form. The heterogeneous cation distribution observed in 93Nb NMR of mixed Ca/Sr
compounds explains this trend in grafting ability and shows that grafting ability can
be imbued without substantially altering other bulk properties.
4. A new organic solvent based nanosheet exfoliation method was successfully
developed using grafted niobates.
The expanded interlayer gallery reduces
electrostatic forces holding sheets together and a combination of sonication and
microwave heating in an acid digestion bomb introduces organic solvent into the
interlayer gallery such that the dispersion forces of the alkyl groups are weakened
and the niobate sheets break apart until only single sheets remain. A close HSP
201
match between the grafted alcohol and organic solvent (accounting for the reduced
hydrogen bonding interactions of the alcohol post-grafting) is a strong indicator of
the solvent’s exfoliation ability, though high microwave susceptibility can
overcome a slight HSP mismatch.
A fascinating, but still not conclusively
explained, phenomenon of this exfoliation procedure was the generation of strongly
yellow colored suspensions. Further study is required to properly identify the origin
of these colored suspensions, but preliminary data in this work suggest it is likely
a charge-transfer complex between the ketone and nanosheet surface is the culprit.
While I may not be personally involved in future research directions, there remains an
abundance of questions left to answer:
•
What is the nature of the color generated after exfoliation?
•
Why does grafting effect the band gap and can it be controlled?
•
How can exfoliated sheets be collected out of solution?
•
What are the catalytic properties of these materials before and after exfoliation?
•
What other groups can be grafted into the interlayer and can they be modified
post-exfoliation without being removed from the surface?
•
How can 93Nb NMR be used in colloidal solutions?
•
Will polarization transfer-based NMR experiments help characterize H-O-Nb
environment and elucidate grafting ability?
•
Would it be possible to design a high-temperature and high-pressure sample
rotor to monitor grafting/exfoliation in situ?
202
•
Can a combination of microwave heating and conventional heating be used to
graft HCa2Nb3O10 beyond the C1 stage?
Despite the new insights into layered niobate structure and properties this work has
produced, it is but a mere glimpse of the vast unknown still hidden in the darkness, waiting
for inevitable illumination.
203
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Dimos, D.; Mueller, C., Perovskite Thin Films for High-Frequency Capacitor
Applications 1. Annual Review of Materials Science 1998, 28 (1), 397-419.
Izuha, M.; Abe, K.; Fukushima, N., Electrical properties of all-perovskite oxide
(SrRuO3/BaxSr1-xTiO3/SrRuO3) capacitors. Japanese journal of applied physics
1997, 36 (9S), 5866.
Kim, Y.-H.; Kim, H.-J.; Osada, M.; Li, B.-W.; Ebina, Y.; Sasaki, T., 2D Perovskite
Nanosheets with Thermally-Stable High-κ Response: A New Platform for HighTemperature Capacitors. ACS applied materials & interfaces 2014, 6 (22), 1951019514.
Ahn, C.-W.; Choi, J.-J.; Ryu, J.; Yoon, W.-H.; Hahn, B.-D.; Kim, J.-W.; Choi, J.H.; Park, D.-S., Composition design rule for energy harvesting devices in
piezoelectric perovskite ceramics. Materials Letters 2015, 141, 323-326.
Kim, Y.-J.; Dang, T.-V.; Choi, H.-J.; Park, B.-J.; Eom, J.-H.; Song, H.-A.; Seol, D.;
Kim, Y.; Shin, S.-H.; Nah, J., Piezoelectric properties of CH 3 NH 3 PbI 3
perovskite thin films and their applications in piezoelectric generators. Journal of
Materials Chemistry A 2016, 4 (3), 756-763.
Cheng, L.-Q.; Wang, K.; Li, J.-F., Synthesis of highly piezoelectric lead-free (K,
Na) NbO3 one-dimensional perovskite nanostructures. Chem. Commun. 2013, 49
(38), 4003-4005.
Paul, A. K.; Reehuis, M.; Ksenofontov, V.; Yan, B.; Hoser, A.; Többens, D. M.;
Abdala, P. M.; Adler, P.; Jansen, M.; Felser, C., Lattice Instability and Competing
Spin Structures in the Double Perovskite Insulator Sr 2 FeOsO 6. Physical Review
Letters 2013, 111 (16), 167205.
Torrance, J.; Lacorre, P.; Nazzal, A.; Ansaldo, E.; Niedermayer, C., Systematic
study of insulator-metal transitions in perovskites R NiO 3 (R= Pr, Nd, Sm, Eu) due
to closing of charge-transfer gap. Physical Review B 1992, 45 (14), 8209.
De Teresa, J.; Ibarra, M.; Garcia, J.; Blasco, J.; Ritter, C.; Algarabel, P.; Marquina,
C.; Del Moral, A., Spin-Glass Insulator State in (Tb-La) 2/3 C a 1/3 Mn O 3
Perovskite. Physical Review Letters 1996, 76 (18), 3392.
Michel, C.; Er-Rakho, L.; Raveau, B., The oxygen defect perovskite
BaLa4Cu5O13. 4, a metallic conductor. Materials research bulletin 1985, 20 (6),
667-671.
Mizusaki, J., Nonstoichiometry, diffusion, and electrical properties of perovskitetype oxide electrode materials. Solid State Ionics 1992, 52 (1), 79-91.
Xu, X.; Randorn, C.; Efstathiou, P.; Irvine, J. T., A red metallic oxide
photocatalyst. Nature materials 2012, 11 (7), 595-598.
Jung, J. I.; Jeong, H. Y.; Lee, J. S.; Kim, M. G.; Cho, J., A bifunctional perovskite
catalyst for oxygen reduction and evolution. Angewandte Chemie 2014, 126 (18),
4670-4674.
204
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Jin, C.; Yang, Z.; Cao, X.; Lu, F.; Yang, R., A novel bifunctional catalyst of Ba 0.9
Co 0.5 Fe 0.4 Nb 0.1 O 3− δ perovskite for lithium–air battery. International
Journal of Hydrogen Energy 2014, 39 (6), 2526-2530.
Lin, K.-H.; Wang, C.-B.; Chien, S.-H., Catalytic performance of steam reforming
of ethanol at low temperature over LaNiO 3 perovskite. International Journal of
Hydrogen Energy 2013, 38 (8), 3226-3232.
Rubel, M. H.; Miura, A.; Takei, T.; Kumada, N.; Mozahar Ali, M.; Nagao, M.;
Watauchi, S.; Tanaka, I.; Oka, K.; Azuma, M., Superconducting Double Perovskite
Bismuth Oxide Prepared by a Low‐Temperature Hydrothermal Reaction.
Angewandte Chemie International Edition 2014, 53 (14), 3599-3603.
Schoop, L. M.; Müchler, L.; Felser, C.; Cava, R., Lone pair effect, structural
distortions, and potential for superconductivity in Tl perovskites. Inorganic
chemistry 2013, 52 (9), 5479-5483.
Lee, M. M.; Teuscher, J.; Miyasaka, T.; Murakami, T. N.; Snaith, H. J., Efficient
Hybrid Solar Cells Based on Meso-Superstructured Organometal Halide
Perovskites. Science 2012, 338 (6107), 643-647.
Kim, H.-S.; Lee, C.-R.; Im, J.-H.; Lee, K.-B.; Moehl, T.; Marchioro, A.; Moon, S.J.; Humphry-Baker, R.; Yum, J.-H.; Moser, J. E.; Grätzel, M.; Park, N.-G., Lead
Iodide Perovskite Sensitized All-Solid-State Submicron Thin Film Mesoscopic
Solar Cell with Efficiency Exceeding 9%. Scientific Reports 2012, 2, 591.
Goldschmidt, V., Crystal structure and chemical constitution. Transactions of the
Faraday Society 1929, 25, 253-283.
Momma, K.; Izumi, F., VESTA 3 for three-dimensional visualization of crystal,
volumetric and morphology data. Journal of Applied Crystallography 2011, 44 (6),
1272-1276.
Brown, I. D., The Chemical Bond in Inorganic Chemistry: The Bond Valence
Model. Oxford University Press: New York, 2002; p 278.
Dion, M.; Ganne, M.; Tournoux, M., Nouvelles familles de phases
MIMII2Nb3O10 a feuillets “perovskites”. Materials research bulletin 1981, 16
(11), 1429-1435.
Liang, Z.-H.; Tang, K.-B.; Chen, Q.-W.; Zheng, H.-G., RbCa2Nb3O10 from X-ray
powder data. Acta Crystallographica Section E 2009, 65 (6), i44.
Toda, K.; Sato, M., Synthesis and structure determination of new layered
perovskite compounds, ALaTa2O7 and ACa2Ta3O10 (A= Rb, Li). J. Mater.
Chem. 1996, 6 (6), 1067-1071.
Arney, D.; Porter, B.; Greve, B.; Maggard, P. A., New molten-salt synthesis and
photocatalytic properties of La 2 Ti 2 O 7 particles. Journal of Photochemistry and
Photobiology A: Chemistry 2008, 199 (2), 230-235.
Li, L.; Deng, J.; Chen, J.; Sun, X.; Yu, R.; Liu, G.; Xing, X., Wire structure and
morphology transformation of niobium oxide and niobates by molten salt synthesis.
Chemistry of Materials 2009, 21 (7), 1207-1213.
Porob, D. G.; Maggard, P. A., Flux syntheses of La-doped NaTaO 3 and its
photocatalytic activity. Journal of Solid State Chemistry 2006, 179 (6), 1727-1732.
205
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
Geselbracht, M. J.; Walton, R. I.; Cowell, E. S.; Millange, F.; O'Hare, D., An
Investigation of the Synthesis of the Layered Perovskite RbCa2Nb3O10 Using
Time-Resolved in Situ High-Temperature Powder X-ray Diffraction. Chemistry of
Materials 2002, 14 (10), 4343-4349.
Thangadurai, V.; Schmid-Beurmann, P.; Weppner, W., Synthesis, Structure, and
Electrical Conductivity of A′[A2B3O10](A′=Rb, Cs; A=Sr, Ba; B=Nb, Ta): New
Members of Dion–Jacobson-Type Layered Perovskites. Journal of Solid State
Chemistry 2001, 158 (2), 279-289.
Geselbracht, M. J.; White, H. K.; Blaine, J. M.; Diaz, M. J.; Hubbs, J. L.; Adelstein,
N.; Kurzman, J. A., New solid acids in the triple-layer Dion–Jacobson layered
perovskite family. Materials research bulletin 2011, 46 (3), 398-406.
Wang, X.; Adhikari, J.; Smith, L. J., An Investigation of Distortions of the Dion−
Jacobson Phase RbSr2Nb3O10 and Its Acid-Exchanged Form with 93Nb Solid
State NMR and DFT Calculations. The Journal of Physical Chemistry C 2009, 113
(40), 17548-17559.
Shannon, R. t.; Prewitt, C. T., Effective ionic radii in oxides and fluorides. Acta
Crystallographica Section B: Structural Crystallography and Crystal Chemistry
1969, 25 (5), 925-946.
Jacobson, A. J.; Johnson, J. W.; Lewandowski, J. T., Interlayer chemistry between
thick transition-metal oxide layers: synthesis and intercalation reactions of
K[Ca2Nan-3NbnO3n+1] (3 .ltoreq. n .ltoreq. 7). Inorganic chemistry 1985, 24 (23),
3727-3729.
Sato, M.; Kono, Y.; Jin, T., Structural Characterization and Ion Conductivity of
MCa2NaNb4O13 (M=Rb, Na) with Four Units of Perovskite Layer. Journal of the
Ceramic Society of Japan 1993, 101 (9), 980-984.
Maeda, K.; Mallouk, T. E., Comparison of two-and three-layer restacked Dion–
Jacobson phase niobate nanosheets as catalysts for photochemical hydrogen
evolution. Journal of Materials Chemistry 2009, 19 (27), 4813-4818.
Hata, H.; Kobayashi, Y.; Bojan, V.; Youngblood, W. J.; Mallouk, T. E., Direct
deposition of trivalent rhodium hydroxide nanoparticles onto a semiconducting
layered calcium niobate for photocatalytic hydrogen evolution. Nano letters 2008,
8 (3), 794-799.
Ebina, Y.; Sasaki, T.; Harada, M.; Watanabe, M., Restacked perovskite nanosheets
and their Pt-loaded materials as photocatalysts. Chemistry of materials 2002, 14
(10), 4390-4395.
Compton, O. C.; Carroll, E. C.; Kim, J. Y.; Larsen, D. S.; Osterloh, F. E., Calcium
niobate semiconductor nanosheets as catalysts for photochemical hydrogen
evolution from water. The Journal of Physical Chemistry C 2007, 111 (40), 1458914592.
Schaak, R. E.; Mallouk, T. E., Perovskites by design: a toolbox of solid-state
reactions. Chemistry of Materials 2002, 14 (4), 1455-1471.
Sasaki, T., Fabrication of nanostructured functional materials using exfoliated
nanosheets as a building block. Journal of the Ceramic Society of Japan 2007, 115
(1337), 9-16.
206
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
Ozawa, T. C.; Onoda, M.; Iyi, N.; Ebina, Y.; Sasaki, T., Bulk Functional Materials
Design Using Oxide Nanosheets as Building Blocks: A New Upconversion
Material Fabricated by Flocculation of Ca2Nb3O10– Nanosheets with Rare-Earth
Ions. The Journal of Physical Chemistry C 2014, 118 (3), 1729-1738.
Bao-Wen, L.; Minoru, O.; Kosho, A.; Yasuo, E.; Tadashi, C. O.; Takayoshi, S.,
Solution-Based Fabrication of Perovskite Multilayers and Superlattices Using
Nanosheet Process. Japanese journal of applied physics 2011, 50 (9S2), 09NA10.
Li, C.; Li, Y.; Xing, Y.; Zhang, Z.; Zhang, X.; Li, Z.; Shi, Y.; Ma, T.; Ma, R.;
Wang, K.; Wei, J., Perovskite Solar Cell Using a Two-Dimensional Titania
Nanosheet Thin Film as the Compact Layer. ACS applied materials & interfaces
2015, 7 (28), 15117-15122.
Hu, X.; Zhang, X.; Liang, L.; Bao, J.; Li, S.; Yang, W.; Xie, Y., High-Performance
Flexible Broadband Photodetector Based on Organolead Halide Perovskite.
Advanced Functional Materials 2014, 24 (46), 7373-7380.
Rao, C. N. R.; Ramakrishna Matte, H. S. S.; Maitra, U., Graphene Analogues of
Inorganic Layered Materials. Angewandte Chemie International Edition 2013, 52
(50), 13162-13185.
Zhuang, X.; Mai, Y.; Wu, D.; Zhang, F.; Feng, X., Two-Dimensional Soft
Nanomaterials: A Fascinating World of Materials. Advanced Materials 2015, 27
(3), 403-427.
Osada, M.; Takanashi, G.; Li, B.-W.; Akatsuka, K.; Ebina, Y.; Ono, K.; Funakubo,
H.; Takada, K.; Sasaki, T., Controlled Polarizability of One-Nanometer-Thick
Oxide Nanosheets for Tailored, High-κ Nanodielectrics. Advanced Functional
Materials 2011, 21 (18), 3482-3487.
Xu, T. G.; Zhang, C.; Shao, X.; Wu, K.; Zhu, Y. F., Monomolecular-Layer
Ba5Ta4O15 Nanosheets: Synthesis and Investigation of Photocatalytic Properties.
Advanced Functional Materials 2006, 16 (12), 1599-1607.
Tagusagawa, C.; Takagaki, A.; Hayashi, S.; Domen, K., Efficient utilization of
nanospace of layered transition metal oxide HNbMoO6 as a strong, water-tolerant
solid acid catalyst. Journal of the American Chemical Society 2008, 130 (23),
7230-7231.
Kitano, M.; Nakajima, K.; Kondo, J. N.; Hayashi, S.; Hara, M., Protonated titanate
nanotubes as solid acid catalyst. Journal of the American Chemical Society 2010,
132 (19), 6622-6623.
Takagaki, A.; Sugisawa, M.; Lu, D.; Kondo, J. N.; Hara, M.; Domen, K.; Hayashi,
S., Exfoliated nanosheets as a new strong solid acid catalyst. Journal of the
American Chemical Society 2003, 125 (18), 5479-5485.
Han, S.-T.; Zhou, Y.; Roy, V. A. L., Towards the Development of Flexible NonVolatile Memories. Advanced Materials 2013, 25 (38), 5425-5449.
Ma, R.; Sasaki, T., Nanosheets of Oxides and Hydroxides: Ultimate 2D ChargeBearing Functional Crystallites. Advanced Materials 2010, 22 (45), 5082-5104.
Tagusagawa, C.; Takagaki, A.; Hayashi, S.; Domen, K., Characterization of
HNbWO6 and HTaWO6 Metal Oxide Nanosheet Aggregates As Solid Acid
Catalysts. Journal of Physical Chemistry C 2009, 113 (18), 7.
207
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
Takagaki, A.; Yoshida, T.; Lu, D.; Kondo, J. N.; Hara, M.; Domen, K.; Hayashi, S.,
Titanium Niobate and Titanium Tantalate Nanosheets as Strong Solid Acid
Catalysts. Journal of Physical Chemistry B 2004, 108 (31), 7.
Kudo, A.; Kato, H.; Nakagawa, S., Water Splitting into H2 and O2 on New
Sr2M2O7 (M = Nb and Ta) Photocatalysts with Layered Perovskite Structures: Factors Affecting the Photocatalytic Activity. The Journal of Physical Chemistry B
1999, 104 (3), 571-575.
Shannon, R. D., Revised Effective Ionic Radii and Systematic Studies of
Interatomic Distances in Halids and Chalcogenides. Acta Crystallographica 1976,
A32, 751.
Huheey, J. E., Inorganic Chemistry. Harper and Row: New York, 1983.
Wen-Long;, Z.; Qing-Li;, Z.; Jin-Yun;, G.; Wen-Peng;, L.; Li-Hua;, D.; Shao-Tang,
R., Structures and luminescence properties of Yb3+ in the double perovskites
Ba2YB;O6 (B'=Ta5+, Nb5+) phosphors. Chin. Phys. B 2011, 20 (1), 016101.
Hong, Y.-S.; Kim, S.-J.; Kim, S.-J.; Choy, J.-H., B-site cation arrangement and
crystal structure of layered perovskite compounds CsLnTiNbO (Ln = La, Pr, Nd,
Sm) and CsCaLaTiNbO. Journal of Materials Chemistry 2000, 10 (5), 1209-1214.
Kim, H. J.; Byeon, S.-H.; Yun, H., Raman Spectra of the Solid-Solution between
Rb~ 2La~ 2Ti~ 3O~ 1~ 0 and RbCa~ 2Nb~ 3O~ 1~ 0. BULLETIN-KOREAN
CHEMICAL SOCIETY 2001, 22 (3), 298-302.
Maeda, K.; Mallouk, T. E., Comparison of two- and three-layer restacked DionJacobson phase niobate nanosheets as catalysts for photochemical hydrogen
evolution. Journal of Materials Chemistry 2009, 19, 4813-4818.
Takahashi, S.; Nakato, T.; Hayashi, S.; Sugahara, Y.; Kuroda, K., Formation of
Methoxy-Modified Interlayer Surface via the Reaction between Methanol and
Layered Perovskite HLaNb2O7.cntdot.xH2O. Inorganic Chemistry 1995, 34 (20),
5065-5069.
Suzuki, H.; Notsu, K.; Takeda, Y.; Sugimoto, W.; Sugahara, Y., Reactions of
Alkoxyl Derivatives of a Layered Perovskite with Alcohols: Substitution Reactions
on the Interlayer Surface of a Layered Perovskite. Chemistry of Materials 2003, 15
(3), 636-641.
Tahara, S.; Sugahara, Y., Interlayer Surface Modification of the Protonated TripleLayered Perovskite HCa2Nb3O10·xH2O with n-Alcohols. Langmuir 2003, 19
(22), 9473-9478.
Tahara, S.; Takeda, Y.; Sugahara, Y., Preparation of Organic−Inorganic Hybrids
Possessing Nanosheets with Perovskite-Related Structures via Exfoliation during a
Sol−Gel Process. Chemistry of Materials 2005, 17 (24), 6198-6204.
Tahara, S.; Ichikawa, T.; Kajiwara, G.; Sugahara, Y., Reactivity of the RuddlesdenPopper Phase H2La2Ti3O10 with Organic Compounds: Intercalation and Grafting
Reactions. Chemistry of materials 2007, 19 (9), 2352-2358.
Shimada, A.; Yoneyama, Y.; Tahara, S.; Mutin, P. H.; Sugahara, Y., Interlayer
surface modification of the protonated ion-exchangeable layered perovskite
HLaNb2O7•xH2O with organophosphonic acids. Chemistry of Materials 2009, 21
(18), 4155-4162.
208
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
87.
Bloch, F.; Hansen, W. W.; Packard, M., Nuclear Induction. Physical Review 1946,
69 (3-4), 127-127.
Purcell, E. M.; Torrey, H. C.; Pound, R. V., Resonance Absorption by Nuclear
Magnetic Moments in a Solid. Physical Review 1946, 69 (1-2), 37-38.
Ernst, R.; Anderson, W., Application of Fourier transform spectroscopy to
magnetic resonance. Review of Scientific Instruments 1966, 37 (1), 93-102.
Lauterbur, P. C., Image Formation by Induced Local Interactions: Examples
Employing Nuclear Magnetic Resonance. Nature 1973, 242 (5394), 190-191.
Stehling, M.; Turner, R.; Mansfield, P., Echo-planar imaging: magnetic resonance
imaging in a fraction of a second. Science 1991, 254 (5028), 43-50.
Laws, D. D.; Bitter, H.-M. L.; Jerschow, A., Solid-State NMR Spectroscopic
Methods in Chemistry. Angewandte Chemie International Edition 2002, 41 (17),
3096-3129.
Eckert, H., Structural characterization of noncrystalline solids and glasses using
solid state NMR. Progress in Nuclear Magnetic Resonance Spectroscopy 1992, 24
(3), 159-293.
Bakhmutov, V. I., Strategies for Solid-State NMR Studies of Materials: From
Diamagnetic to Paramagnetic Porous Solids. Chemical Reviews 2011, 111 (2), 530562.
Hanna, J. V.; Smith, M. E., Recent technique developments and applications of
solid state NMR in characterising inorganic materials. Solid State Nuclear
Magnetic Resonance 2010, 38 (1), 1-18.
Grey, C. P.; Tycko, R., Solid-state NMR in biological and materials physics.
Physics today 2009, 62 (9), 44.
Chierotti, M. R.; Gobetto, R., Solid-state NMR studies of weak interactions in
supramolecular systems. Chemical Communications 2008, (14), 1621-1634.
Gullion, T.; Schaefer, J., Rotational-echo double-resonance NMR. Journal of
Magnetic Resonance (1969) 1989, 81 (1), 196-200.
Fyfe, C. A.; Mueller, K. T.; Grondey, H.; Wong-Moon, K. C., Dipolar dephasing
between quadrupolar and spin-12 nuclei. REDOR and TEDOR NMR experiments
on VPI-5. Chemical Physics Letters 1992, 199 (1–2), 198-204.
Hing, A. W.; Vega, S.; Schaefer, J., Transferred-echo double-resonance NMR.
Journal of Magnetic Resonance (1969) 1992, 96 (1), 205-209.
van Eck, E. R. H.; Janssen, R.; Maas, W. E. J. R.; Veeman, W. S., A novel
application of nuclear spin-echo double-resonance to aluminophosphates and
aluminosilicates. Chemical Physics Letters 1990, 174 (5), 428-432.
Gullion, T., Measurement of dipolar interactions between spin-12 and quadrupolar
nuclei by rotational-echo, adiabatic-passage, double-resonance NMR. Chemical
Physics Letters 1995, 246 (3), 325-330.
Duer, M. J., Solid state NMR spectroscopy: principles and applications. John
Wiley & Sons: 2008.
Kaupp, M.; Bühl, M.; Malkin, V. G., Calculation of NMR and EPR parameters:
theory and applications. John Wiley & Sons: 2006.
209
88.
Wasylishen, R. E.; Ashbrook, S. E.; Wimperis, S., NMR of Quadrupolar Nuclei in
Solid Materials. John Wiley & Sons: 2012.
89. Wasylishen, R. E.; Ashbrook, S. E.; Wimperis, S., NMR of Quadrupolar Nuclei in
Solid Materials. John Wiley & Sons Ltd: Chichester, UK, 2012.
90. Kentgens, A., A practical guide to solid-state NMR of half-integer quadrupolar
nuclei with some applications to disordered systems. Geoderma 1997, 80 (3-4),
271-306.
91. Lapina, O. B.; Khabibulin, D. F.; Shubin, A. A.; Terskikh, V. V., Practical aspects
of 51V and 93Nb solid-state NMR spectroscopy and applications to oxide
materials. Progress in Nuclear Magnetic Resonance Spectroscopy 2008, 53 (3),
128-191.
92. Liu, T., Investigation of Local Structures in Layered Niobates by Solid-state NMR
Spectroscopy. 2012.
93. Herzfeld, J.; Berger, A. E., Sideband intensities in NMR spectra of samples
spinning at the magic angle. The Journal of Chemical Physics 1980, 73 (12), 60216030.
94. Lapina, O. B.; Khabibulin, D. F.; Romanenko, K. V.; Gan, Z.; Zuev, M. G.;
Krasil’nikov, V. N.; Fedorov, V. E., 93Nb NMR chemical shift scale for niobia
systems. Solid State Nuclear Magnetic Resonance 2005, 28 (2–4), 204-224.
95. Hahn, E. L., Spin Echoes. Physical Review 1950, 80 (4), 580-594.
96. Massiot, D.; Farnan, I.; Gautier, N.; Trumeau, D.; Trokiner, A.; Coutures, J. P.,
71Ga and 69Ga nuclear magnetic resonance study of β-Ga2O3: resolution of fourand six-fold coordinated Ga sites in static conditions. Solid State Nuclear Magnetic
Resonance 1995, 4 (4), 241-248.
97. Bureau, B.; Silly, G.; Buzaré, J. Y.; Legein, C.; Massiot, D., From crystalline to
glassy gallium fluoride materials: an NMR study of 69Ga and 71Ga quadrupolar
nuclei. Solid State Nuclear Magnetic Resonance 1999, 14 (3–4), 181-190.
98. Wang, X., Determination of local structures of layered niobate, niobate nanosheet
and mesoporous niobium oxide by solid state NMR and PXRD. CLARK
UNIVERSITY: 2009.
99. Smith, M. E.; van Eck, E. R. H., Recent advances in experimental solid state NMR
methodology for half-integer spin quadrupolar nuclei. Progress in Nuclear
Magnetic Resonance Spectroscopy 1999, 34 (2), 159-201.
100. Du, L.-S.; Schurko, R. W.; Lim, K. H.; Grey, C. P., A Solid-State 93Nb and 19F
NMR Spectroscopy and X-ray Diffraction Study of Potassium
Heptafluoroniobate(V): Characterization of 93Nb, 19F Coupling, and Fluorine
Motion. The Journal of Physical Chemistry A 2001, 105 (4), 760-768.
101. Schurko, R. W.; Hung, I.; Macdonald, C. L. B.; Cowley, A. H., Anisotropic NMR
Interaction Tensors in the Decamethylaluminocenium Cation. Journal of the
American Chemical Society 2002, 124 (44), 13204-13214.
102. Lowe, I. J., Free Induction Decays of Rotating Solids. Physical Review Letters
1959, 2 (7), 285-287.
210
103. Andrew, E. R.; Bradbury, A.; Eades, R. G., Removal of Dipolar Broadening of
Nuclear Magnetic Resonance Spectra of Solids by Specimen Rotation. Nature
1959, 183 (4678), 1802-1803.
104. Andrew, E. R.; Eades, R. G., Possibilities for high-resolution nuclear magnetic
resonance spectra of crystals. Discussions of the Faraday Society 1962, 34 (0), 3842.
105. Andrew, E. R., The narrowing of NMR spectra of solids by high-speed specimen
rotation and the resolution of chemical shift and spin multiplet structures for solids.
Progress in Nuclear Magnetic Resonance Spectroscopy 1971, 8 (1), 1-39.
106. Laage, S.; Sachleben, J. R.; Steuernagel, S.; Pierattelli, R.; Pintacuda, G.; Emsley,
L., Fast acquisition of multi-dimensional spectra in solid-state NMR enabled by
ultra-fast MAS. Journal of Magnetic Resonance 2009, 196 (2), 133-141.
107. Bertini, I.; Emsley, L.; Lelli, M.; Luchinat, C.; Mao, J.; Pintacuda, G., Ultrafast
MAS Solid-State NMR Permits Extensive 13C and 1H Detection in Paramagnetic
Metalloproteins. Journal of the American Chemical Society 2010, 132 (16), 55585559.
108. Agarwal, V.; Penzel, S.; Szekely, K.; Cadalbert, R.; Testori, E.; Oss, A.; Past, J.;
Samoson, A.; Ernst, M.; Böckmann, A.; Meier, B. H., De Novo 3D Structure
Determination from Sub-milligram Protein Samples by Solid-State 100 kHz MAS
NMR Spectroscopy. Angewandte Chemie International Edition 2014, 53 (45),
12253-12256.
109. Miah, H. K.; Bennett, D. A.; Iuga, D.; Titman, J. J., Measuring proton shift tensors
with ultrafast MAS NMR. Journal of Magnetic Resonance 2013, 235, 1-5.
110. Bak, M.; Rasmussen, J. T.; Nielsen, N. C., SIMPSON: a general simulation
program for solid-state NMR spectroscopy. Journal of Magnetic Resonance 2011,
213 (2), 366-400.
111. O’Dell, L. A.; Rossini, A. J.; Schurko, R. W., Acquisition of ultra-wideline NMR
spectra from quadrupolar nuclei by frequency stepped WURST–QCPMG.
Chemical Physics Letters 2009, 468 (4–6), 330-335.
112. Hung, I.; Gan, Z., On the practical aspects of recording wideline QCPMG NMR
spectra. Journal of Magnetic Resonance 2010, 204 (2), 256-265.
113. O’Dell, L. A.; Schurko, R. W., QCPMG using adiabatic pulses for faster
acquisition of ultra-wideline NMR spectra. Chemical Physics Letters 2008, 464 (1),
97-102.
114. MacGregor, A. W.; O’Dell, L. A.; Schurko, R. W., New methods for the
acquisition of ultra-wideline solid-state NMR spectra of spin-1/2 nuclides. Journal
of Magnetic Resonance 2011, 208 (1), 103-113.
115. Schurko, R. W., Ultra-wideline solid-state NMR spectroscopy. Accounts of
chemical research 2013, 46 (9), 1985-1995.
116. O'Dell, L. A., The WURST kind of pulses in solid-state NMR. Solid State Nuclear
Magnetic Resonance 2013, 55, 28-41.
117. Kupce, E.; Freeman, R., Adiabatic Pulses for Wideband Inversion and Broadband
Decoupling. Journal of Magnetic Resonance, Series A 1995, 115 (2), 273-276.
211
118. Garwood, M.; DelaBarre, L., The return of the frequency sweep: designing
adiabatic pulses for contemporary NMR. Journal of magnetic resonance 2001, 153
(2), 155-177.
119. Bhattacharyya, R.; Frydman, L., Quadrupolar nuclear magnetic resonance
spectroscopy in solids using frequency-swept echoing pulses. The Journal of
chemical physics 2007, 127 (19), 194503.
120. Larsen, F. H.; Jakobsen, H. J.; Ellis, P. D.; Nielsen, N. C., Sensitivity-Enhanced
Quadrupolar-Echo NMR of Half-Integer Quadrupolar Nuclei. Magnitudes and
Relative Orientation of Chemical Shielding and Quadrupolar Coupling Tensors.
The Journal of Physical Chemistry A 1997, 101 (46), 8597-8606.
121. Meiboom, S.; Gill, D., Modified Spin‐Echo Method for Measuring Nuclear
Relaxation Times. Review of Scientific Instruments 1958, 29 (8), 688-691.
122. Harris, K. J.; Lupulescu, A.; Lucier, B. E. G.; Frydman, L.; Schurko, R. W.,
Broadband adiabatic inversion pulses for cross polarization in wideline solid-state
NMR spectroscopy. Journal of Magnetic Resonance 2012, 224, 38-47.
123. Hartmann, S. R.; Hahn, E. L., Nuclear Double Resonance in the Rotating Frame.
Physical Review 1962, 128 (5), 2042-2053.
124. Pines, A.; Gibby, M. G.; Waugh, J. S., Proton‐Enhanced Nuclear Induction
Spectroscopy. A Method for High Resolution NMR of Dilute Spins in Solids. The
Journal of chemical physics 1972, 56 (4), 1776-1777.
125. Pines, A.; Gibby, M. G.; Waugh, J. S., Proton‐enhanced NMR of dilute spins in
solids. The Journal of chemical physics 1973, 59 (2), 569-590.
126. Mueller, K. T.; Sun, B. Q.; Chingas, G. C.; Zwanziger, J. W.; Terao, T.; Pines, A.,
Dynamic-angle spinning of quadrupolar nuclei. Journal of Magnetic Resonance
(1969) 1990, 86 (3), 470-487.
127. Chmelka, B. F.; Mueller, K. T.; Pines, A.; Stebbins, J.; Wu, Y.; Zwanziger, J. W.,
Oxygen-17 NMR in solids by dynamic-angle spinning and double rotation. Nature
1989, 339 (6219), 42-43.
128. Frydman, L.; Harwood, J. S., Isotropic Spectra of Half-Integer Quadrupolar Spins
from Bidimensional Magic-Angle Spinning NMR. Journal of the American
Chemical Society 1995, 117 (19), 5367-5368.
129. Amoureux, J.-P.; Fernandez, C.; Steuernagel, S., ZFiltering in MQMAS NMR.
Journal of Magnetic Resonance, Series A 1996, 123 (1), 116-118.
130. Brown, S. P.; Wimperis, S., Two-dimensional multiple-quantum MAS NMR of
quadrupolar nuclei: a comparison of methods. Journal of Magnetic Resonance
1997, 128 (1), 42-61.
131. Gan, Z.; Kwak, H.-T., Enhancing MQMAS sensitivity using signals from multiple
coherence transfer pathways. Journal of Magnetic Resonance 2004, 168 (2), 346351.
132. Massiot, D.; Touzo, B.; Trumeau, D.; Coutures, J. P.; Virlet, J.; Florian, P.;
Grandinetti, P. J., Two-dimensional magic-angle spinning isotropic reconstruction
sequences for quadrupolar nuclei. Solid State Nuclear Magnetic Resonance 1996, 6
(1), 73-83.
212
133. Baltisberger, J. H.; Xu, Z.; Stebbins, J. F.; Wang, S. H.; Pines, A., Triple-Quantum
Two-Dimensional 27Al Magic-Angle Spinning Nuclear Magnetic Resonance
Spectroscopic Study of Aluminosilicate and Aluminate Crystals and Glasses.
Journal of the American Chemical Society 1996, 118 (30), 7209-7214.
134. Prasad, S.; Zhao, P.; Huang, J.; Fitzgerald, J. J.; Shore, J. S., Niobium-93 MQMAS
NMR Spectroscopic Study of Alkali and Lead Niobates. Solid State Nuclear
Magnetic Resonance 2001, 19 (1–2), 45-62.
135. Madhu, P. K.; Goldbourt, A.; Frydman, L.; Vega, S., Sensitivity enhancement of
the MQMAS NMR experiment by fast amplitude modulation of the pulses.
Chemical Physics Letters 1999, 307 (1–2), 41-47.
136. Goldbourt, A.; Madhu, P. K.; Vega, S., Enhanced conversion of triple to singlequantum coherence in the triple-quantum MAS NMR spectrosocopy of spin-5/2
nuclei. Chemical Physics Letters 2000, 320 (5–6), 448-456.
137. Colaux, H.; Dawson, D. M.; Ashbrook, S. E., Efficient Amplitude-Modulated
Pulses for Triple- to Single-Quantum Coherence Conversion in MQMAS NMR.
The Journal of Physical Chemistry A 2014, 118 (31), 6018-6025.
138. Fernandez, C.; Lang, D. P.; Amoureux, J. P.; Pruski, M., Measurement of
Heteronuclear Dipolar Interactions between Quadrupolar and Spin-1/2 Nuclei in
Solids by Multiple-Quantum REDOR NMR. Journal of the American Chemical
Society 1998, 120 (11), 2672-2673.
139. Oas, T. G.; Griffin, R. G.; Levitt, M. H., Rotary resonance recoupling of dipolar
interactions in solid‐state nuclear magnetic resonance spectroscopy. The Journal of
Chemical Physics 1988, 89 (2), 692-695.
140. Levitt, M. H.; Oas, T. G.; Griffin, R. G., Rotary Resonance Recoupling in
Heteronuclear Spin Pair Systems. Israel Journal of Chemistry 1988, 28 (4), 271282.
141. Yoon, K. H.; Cho, Y. S.; Kang, D. H., Molten salt synthesis of lead-based relaxors.
Journal of materials science 1998, 33 (12), 2977-2984.
142. Gopalan, S.; Mehta, K.; Virkar, A. V., Synthesis of oxide perovskite solid solutions
using the molten salt method. Journal of materials research 1996, 11 (08), 18631865.
143. Lee, Y.-h.; Cho, J.-h.; Kim, B.-i.; Choi, D.-k., Piezoelectric properties and
densification based on control of volatile mass of potassium and sodium in (K0.
5Na0. 5) NbO3 ceramics. Japanese Journal of Applied Physics 2008, 47 (6R),
4620.
144. Yang, G.; Kong, Y.; Hou, W.; Yan, Q., Heating Behavior and Crystal Growth
Mechanism in Microwave Field. The Journal of Physical Chemistry B 2005, 109
(4), 1371-1379.
145. Le Bail, A.; Duroy, H.; Fourquet, J., Ab-initio structure determination of LiSbWO
6 by X-ray powder diffraction. Materials Research Bulletin 1988, 23 (3), 447-452.
146. TOPAS, Bruker: Karlsruhe, 2003.
147. Grandinetti, P. RMN, Ohio State University: Columbus, OH, 2005.
148. Eichele, K.; Wasylishen, R. E. WSOLIDS NMR Simulation Package, University of
Tuebingen: Tuebingen, Germany, 2001.
213
149. Massiot, D.; Fayon, F.; Capron, M.; King, I.; Le Calvé, S.; Alonso, B.; Durand, J.
O.; Bujoli, B.; Gan, Z.; Hoatson, G., Modelling one‐and two‐dimensional solid‐
state NMR spectra. Magnetic Resonance in Chemistry 2002, 40 (1), 70-76.
150. Kemp, T. F.; Smith, M. E., QuadFit—a new cross-platform computer program for
simulation of NMR line shapes from solids with distributions of interaction
parameters. Solid state nuclear magnetic resonance 2009, 35 (4), 243-252.
151. Wang, X. Determination of local structures of layered niobate, niobate nanosheet
and mesoporous niobium oxide by solid state NMR and PXRD. 2009.
152. Geselbracht, M. J.; Walton, R. I.; Cowell, E. S.; Millange, F.; O’Hare, D., A
furnace for the in situ study of the formation of inorganic solids at high temperature
using time-resolved energy-dispersive x-ray diffraction. Review of Scientific
Instruments 2000, 71 (11), 4177-4181.
153. Takano, Y.; Kimishima, Y.; Yamadaya, T.; Ogawa, S.; Môri, N., High Pressure
Studies of the Non-Copper Superconductors KCa2Nb3O10 and RbCa2Nb3O10.
The Review of High Pressure Science and Technology 1998, 7, 589-591.
154. Kurzman, J. A.; Geselbracht, M. J. In Probing octahedral tilting in Dion-Jacobson
layered perovskites with neutron powder diffraction and Raman spectroscopy,
Materials Research Society Symposium Proceedings, Warrendale, Pa.; Materials
Research Society; 1999: 2007; p 74.
155. Kulischow, N.; Ladasiu, C.; Marschall, R., Layered Dion-Jacobson type niobium
oxides for photocatalytic hydrogen production prepared via molten salt synthesis.
Catalysis Today 2016, (Copyright (C) 2017 American Chemical Society (ACS).
All Rights Reserved.), Ahead of Print.
156. Sirotinkin, S.; Sirotinkin, V.; Trunov, V., Structure of low-temperature
modification of BaNb 2 O 6. Zhurnal Neorganicheskoj Khimii 1990, 35 (6), 16091611.
157. Fieller, E. C., Some problems in interval estimation. Journal of the Royal Statistical
Society. Series B (Methodological) 1954, 175-185.
158. Ziegler, C. Two-dimensional transition metal oxide nanosheets for
nanoarchitectonics. Ludwig-Maximilians-Universität, 2015.
159. Schaak, R. E.; Mallouk, T. E., Topochemical synthesis of three-dimensional
perovskites from lamellar precursors. Journal of the American Chemical Society
2000, 122 (12), 2798-2803.
160. Oral, A.; Mecartney, M., Properties of sol-gel derived strontium barium niobate
ceramics and the effect of V2O5 additive. Journal of Materials Science 2001, 36
(22), 5519-5527.
161. Jehng, J. M.; Wachs, I. E., Molecular structures of supported niobium oxide
catalysts under in situ conditions. The Journal of Physical Chemistry 1991, 95 (19),
7373-7379.
162. Jehng, J. M.; Wachs, I. E., Structural chemistry and Raman spectra of niobium
oxides. Chemistry of Materials 1991, 3 (1), 100-107.
163. Rietveld, H., A profile refinement method for nuclear and magnetic structures.
Journal of Applied Crystallography 1969, 2 (2), 65-71.
214
164. Kato, H.; Kudo, A., Energy structure and photocatalytic activity for water splitting
of Sr2(Ta1−XNbX)2O7 solid solution. Journal of Photochemistry and
Photobiology A: Chemistry 2001, 145 (1–2), 129-133.
165. Hojamberdiev, M.; Bekheet, M. F.; Zahedi, E.; Wagata, H.; Vequizo, J. J. M.;
Yamakata, A.; Yubuta, K.; Gurlo, A.; Domen, K.; Teshima, K., The contrasting
effect of the Ta/Nb ratio in (111)-layered B-site deficient hexagonal perovskite Ba
5 Nb 4− x Ta x O 15 crystals on visible-light-induced photocatalytic water
oxidation activity of their oxynitride derivatives. Dalton Transactions 2016, 45
(31), 12559-12568.
166. Weitzel, H., Kristallstrukturverfeinerung von Wolframiten und Columbiten.
Zeitschrift für Kristallographie-Crystalline Materials 1976, 144 (1-6), 238-258.
167. Sugimoto, W.; Ohkawa, H.; Naito, M.; Sugahara, Y.; Kuroda, K., Synthesis and
Structures of Reduced Niobates with Four Perovskite-like Layers and Their
Semiconducting Properties. Journal of Solid State Chemistry 1999, 148 (2), 508512.
168. Liang, Z.-H.; Tang, K.-B.; Chen, Q.-W.; Zheng, H.-G., RbCa2Nb3O10 from X-ray
powder data. Acta Crystallographica Section E: Structure Reports Online 2009, 65
(6), i44-i44.
169. Wang, X.; Adhikari, J.; Smith, L. J., An Investigation of Distortions of the
Dion−Jacobson Phase RbSr2Nb3O10 and Its Acid-Exchanged Form with 93Nb
Solid State NMR and DFT Calculations. The Journal of Physical Chemistry C
2009, 113 (40), 17548-17559.
170. Ratheesh, R.; Wöhlecke, M.; Berge, B.; Wahlbrink, T.; Haeuseler, H.; Rühl, E.;
Blachnik, R.; Balan, P.; Santha, N.; Sebastian, M., Raman study of the ordering in
Sr (B 0.5′ Nb 0.5) O 3 compounds. Journal of Applied Physics 2000, 88 (5), 28132818.
171. Lufaso, M. W., Crystal structures, modeling, and dielectric property relationships
of 2: 1 ordered Ba3MM'2O9 (M= Mg, Ni, Zn; M'= Nb, Ta) perovskites. Chemistry
of Materials 2004, 16 (11), 2148-2156.
172. Welk, M. E.; Norquist, A. J.; Arnold, F. P.; Stern, C. L.; Poeppelmeier, K. R., Outof-center distortions in d0 transition metal oxide fluoride anions. Inorganic
chemistry 2002, 41 (20), 5119-5125.
173. Koller, H.; Engelhardt, G.; Kentgens, A. P. M.; Sauer, J., 23Na NMR Spectroscopy
of Solids: Interpretation of Quadrupole Interaction Parameters and Chemical Shifts.
The Journal of Physical Chemistry 1994, 98 (6), 1544-1551.
174. Brown, I. t.; Shannon, R., Empirical bond-strength-bond-length curves for oxides.
Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and
General Crystallography 1973, 29 (3), 266-282.
175. Le Bail, A., Whole powder pattern decomposition methods and applications: a
retrospection. Powder Diffraction 2005, 20 (04), 316-326.
176. Duer, M. J., Solid-state NMR spectroscopy: Principles and Applications. Blackwell
Science: Oxford, 2002.
177. Kappe, C. O., Microwave dielectric heating in synthetic organic chemistry.
Chemical Society Reviews 2008, 37 (6), 1127-1139.
215
178. Gabriel, C.; Gabriel, S.; H. Grant, E.; S. J. Halstead, B.; Michael P. Mingos, D.,
Dielectric parameters relevant to microwave dielectric heating. Chemical Society
Reviews 1998, 27 (3), 213-224.
179. Tauc, J.; Grigorovici, R.; Vancu, A., Optical properties and electronic structure of
amorphous germanium. physica status solidi (b) 1966, 15 (2), 627-637.
180. Tauc, J., Optical Properties and Electronic Structure of Amorphous
Semiconductors. In Optical Properties of Solids: Papers from the NATO Advanced
Study Institute on Optical Properties of Solids Held August 7–20, 1966, at
Freiburg, Germany, Nudelman, S.; Mitra, S. S., Eds. Springer US: Boston, MA,
1969; pp 123-136.
181. Davis, E.; Mott, N., Conduction in non-crystalline systems V. Conductivity, optical
absorption and photoconductivity in amorphous semiconductors. Philosophical
Magazine 1970, 22 (179), 0903-0922.
182. Maeda, K.; Mallouk, T. E., Comparison of two- and three-layer restacked DionJacobson phase niobate nanosheets as catalysts for photochemical hydrogen
evolution. J Mater Chem 2009, 19 (27), 4813-4818.
183. Akbarian-Tefaghi, S.; Teixeira Veiga, E.; Amand, G.; Wiley, J. B., Rapid
Topochemical Modification of Layered Perovskites via Microwave Reactions.
Inorganic Chemistry 2016, 55 (4), 1604-1612.
184. Wang, Y.; Delahaye, E.; Leuvrey, C. d.; Leroux, F.; Rabu, P.; Rogez, G., Efficient
Microwave-Assisted Functionalization of the Aurivillius-Phase Bi2SrTa2O9.
Inorganic chemistry 2016, 55 (8), 4039-4046.
185. Wang, Y.; Delahaye, E.; Leuvrey, C.; Leroux, F.; Rabu, P.; Rogez, G., PostSynthesis Modification of the Aurivillius Phase Bi2SrTa2O9 via In Situ
Microwave-Assisted “Click Reaction”. Inorganic Chemistry 2016, 55 (19), 97909797.
186. Osada, M.; Sasaki, T., Two‐Dimensional Dielectric Nanosheets: Novel
Nanoelectronics From Nanocrystal Building Blocks. Advanced Materials 2012, 24
(2), 210-228.
187. Ziegler, C.; Werner, S.; Bugnet, M.; Wörsching, M.; Duppel, V.; Botton, G. A.;
Scheu, C.; Lotsch, B. V., Artificial solids by design: assembly and electron
microscopy study of nanosheet-derived heterostructures. Chemistry of Materials
2013, 25 (24), 4892-4900.
188. Osada, M.; Sasaki, T., Chemical nanomanipulation of two-dimensional nanosheets
and its applications. 2011.
189. Xu, F.; Ebina, Y.; Bando, Y.; Sasaki, T., Structural characterization of (TBA, H)
Ca2Nb3O10 nanosheets formed by delamination of a precursor-layered perovskite.
The Journal of Physical Chemistry B 2003, 107 (36), 9638-9645.
190. Zhou, K. G.; Mao, N. N.; Wang, H. X.; Peng, Y.; Zhang, H. L., A Mixed‐Solvent
Strategy for Efficient Exfoliation of Inorganic Graphene Analogues. Angewandte
Chemie 2011, 123 (46), 11031-11034.
191. Štengl, V.; Henych, J., Strongly luminescent monolayered MoS 2 prepared by
effective ultrasound exfoliation. Nanoscale 2013, 5 (8), 3387-3394.
216
192. Coleman, J. N.; Lotya, M.; O’Neill, A.; Bergin, S. D.; King, P. J.; Khan, U.;
Young, K.; Gaucher, A.; De, S.; Smith, R. J., Two-dimensional nanosheets
produced by liquid exfoliation of layered materials. Science 2011, 331 (6017), 568571.
193. Han, W.-Q.; Wu, L.; Zhu, Y.; Watanabe, K.; Taniguchi, T., Structure of chemically
derived mono- and few-atomic-layer boron nitride sheets. Applied Physics Letters
2008, 93 (22), 223103.
194. Lin, Y.; Williams, T. V.; Connell, J. W., Soluble, Exfoliated Hexagonal Boron
Nitride Nanosheets. The Journal of Physical Chemistry Letters 2010, 1 (1), 277283.
195. Zhi, C.; Bando, Y.; Tang, C.; Kuwahara, H.; Golberg, D., Large-Scale Fabrication
of Boron Nitride Nanosheets and Their Utilization in Polymeric Composites with
Improved Thermal and Mechanical Properties. Advanced Materials 2009, 21 (28),
2889-2893.
196. Ebina, Y.; Sakai, N.; Sasaki, T., Photocatalyst of lamellar aggregates of RuO xloaded perovskite nanosheets for overall water splitting. The Journal of Physical
Chemistry B 2005, 109 (36), 17212-17216.
197. Hansen, C. M., The three dimensional solubility parameter. Danish Technical:
Copenhagen 1967, 14.
198. Agrawal, K. V.; Topuz, B.; Jiang, Z.; Nguenkam, K.; Elyassi, B.; Francis, L. F.;
Tsapatsis, M.; Navarro, M., Solution‐processable exfoliated zeolite nanosheets
purified by density gradient centrifugation. AIChE Journal 2013, 59 (9), 34583467.
199. Keller, A. A.; Wang, H.; Zhou, D.; Lenihan, H. S.; Cherr, G.; Cardinale, B. J.;
Miller, R.; Ji, Z., Stability and aggregation of metal oxide nanoparticles in natural
aqueous matrices. Environmental science & technology 2010, 44 (6), 1962-1967.
200. Marinuc, M.; Rus, F., THE EFFECT OF SOLID PARTICLE SIZE UPON TIME
AND SEDIMENTATION RATE. Bulletin of the Transilvania University of
Brasov, Series II. Forestry, Wood Industry, Agricultural Food Engineering 2012,
(1).
201. Ida, S.; Ogata, C.; Eguchi, M.; Youngblood, W. J.; Mallouk, T. E.; Matsumoto, Y.,
Photoluminescence of Perovskite Nanosheets Prepared by Exfoliation of Layered
Oxides, K2Ln2Ti3O10, KLnNb2O7, and RbLnTa2O7 (Ln: Lanthanide Ion).
Journal of the American Chemical Society 2008, 130 (22), 7052-7059.
202. Teagarden, D. L.; Baker, D. S., Practical aspects of lyophilization using nonaqueous co-solvent systems. European Journal of Pharmaceutical Sciences 2002,
15 (2), 115-133.
203. Yong, Z.; Yingjie, D.; Xueli, W.; Jinghua, X.; Zhengqiang, L., Conformational and
bioactivity analysis of insulin: Freeze-drying TBA/water co-solvent system in the
presence of surfactant and sugar. International journal of pharmaceutics 2009, 371
(1), 71-81.
204. Ebina, Y.; Sasaki, T.; Watanabe, M., Study on exfoliation of layered perovskitetype niobates. Solid State Ionics 2002, 151 (1), 177-182.
217
205. Friedrich, W.; Knipping, P.; Laue, M., Interferenzerscheinungen bei
roentgenstrahlen. Annalen der Physik 1913, 346 (10), 971-988.
218
Appendix A: Indexed Lists of XRD Pattern hkl Reflections
Figure A.1. List of indexed hkl reflections obtained from XRD pattern of RbSr2Nb3O10 microwave
sample.
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.16297
5.82390
36.9
0
2
0
3.88017
22.90106
11.5
2
0
4
3.46316
25.70316
17
1
0
9
3.09239
28.84789
29.6
0
1
9
3.09077
28.86336
38.8
2
0
6
3.08422
28.92597
141
0
2
6
3.07780
28.98763
27.5
0
0
10
3.03259
29.42942
10.2
1
2
5
3.01376
29.61757
20.6
0
1
10
2.82458
31.65144
30.7
2
2
0
2.74824
32.55477
116
2
2
1
2.73702
32.69191
13.7
2
0
8
2.71591
32.95325
36.6
0
3
1
2.57742
34.77877
10.7
0
2
9
2.54414
35.24846
16.4
3
1
2
2.42957
36.96945
12.5
0
1
12
2.40296
37.39391
12.2
3
1
4
2.34108
38.42045
11.9
2
2
7
2.32068
38.77158
13.1
0
3
6
2.30272
39.08630
17.3
0
2
11
2.24739
40.08929
11.4
0
0
14
2.16614
41.66153
11.3
2
2
9
2.12970
42.40841
11.3
1
3
8
2.06052
43.90495
11.4
0
3
9
2.05187
44.09970
12.5
1
2
12
2.04340
44.29221
16.2
2
3
5
2.03023
44.59473
15.7
1
1
14
2.01529
44.94347
21.3
0
2
13
1.99927
45.32348
18
4
0
0
1.94653
46.62291
47.5
2
2
11
1.94635
46.62732
27.6
0
4
0
1.94008
46.78696
17.3
1
1
15
1.89744
47.90337
12.2
1
4
2
1.86818
48.70194
18.2
2
4
3
1.71135
53.50172
19.1
4
1
9
1.64710
55.76625
41.2
4
2
6
1.64513
55.83892
18.3
3
2
13
1.58383
58.20197
13.5
219
Figure A.2. List of indexed hkl reflections obtained from XRD pattern of RbSr2Nb3O10 molten salt
sample
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.26925
5.78333
87.5
0
0
4
7.63462
11.58146
7.35
2
0
0
3.89776
22.79630
8.71
0
2
0
3.88346
22.88136
16.4
0
0
8
3.81731
23.28339
8.14
1
0
7
3.80702
23.34718
10.2
2
0
4
3.47151
25.64029
8.32
0
2
4
3.46140
25.71649
7.82
2
1
1
3.46125
25.71761
7.45
2
0
5
3.28570
27.11719
8.21
1
1
8
3.13639
28.43462
15.6
1
0
9
3.11122
28.66960
26.2
0
1
9
3.10939
28.68680
6.2
2
0
6
3.09457
28.82709
51
0
2
6
3.08740
28.89550
239
0
0
10
3.05385
29.22001
69.3
2
1
5
3.02606
29.49437
7.88
1
2
5
3.02103
29.54463
10.3
1
0
10
2.84345
31.43596
6.58
2
2
0
2.75107
32.52039
200
2
2
1
2.73997
32.65576
16.5
2
0
8
2.72725
32.81241
23
0
2
8
2.72233
32.87329
5.28
2
1
7
2.72226
32.87419
5.07
1
2
7
2.71860
32.91978
12.8
2
2
2
2.70747
33.05890
9.16
0
0
12
2.54487
35.23801
9.89
0
0
14
2.18132
41.35822
29.2
2
1
11
2.17113
41.56136
6.3
0
2
12
2.12855
42.43237
10.4
2
2
10
2.04398
44.27875
10.4
2
2
11
1.95413
46.43085
12
2
1
13
1.94768
46.59378
19.6
1
2
13
1.94633
46.62786
48.2
4
0
1
1.94493
46.66362
40.9
2
0
14
1.90351
47.74113
9.93
2
2
12
1.86814
48.70299
18
4
1
7
1.73447
52.73308
7.08
2
2
14
1.70923
53.57335
17.8
1
1
17
1.70767
53.62610
10.7
3
1
13
1.70032
53.87658
6.86
1
0
18
1.65778
55.37609
10.2
4
1
9
1.65133
55.61093
7.79
4
2
6
1.64801
55.73264
11.4
1
4
9
1.64725
55.76084
20
3
2
12
1.64663
55.78342
40.2
2
3
12
1.64528
55.83332
12.2
2
4
6
1.64476
55.85242
6.12
220
4
3
2
4
2
0
0
3
2
4
10
15
13
8
8
1.64285
1.60260
1.58865
1.58467
1.58178
55.92315
57.45652
58.00872
58.16831
58.28482
8.79
6.07
6.48
14.7
8.68
Figure A.3. List of indexed hkl reflections obtained from XRD pattern of RbCa0.3Sr1.7Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.78497
5.97293
17.5
0
2
1
3.84304
23.12534
15.7
2
0
2
3.73513
23.80306
10.5
0
1
7
3.70928
23.97145
6.17
2
0
4
3.42188
26.01861
14.4
1
1
8
3.06265
29.13422
44.8
0
2
6
3.04660
29.29107
34.2
2
0
6
3.03903
29.36572
31.2
2
2
0
2.73517
32.71471
15.3
2
2
1
2.72354
32.85831
49
2
2
2
2.68953
33.28582
7.63
2
2
7
2.29592
39.20686
5.21
2
3
0
2.14730
42.04420
7.81
0
3
9
2.03107
44.57539
6.02
0
4
0
1.93796
46.84138
8.14
0
4
1
1.93381
46.94790
9.91
4
0
0
1.93018
47.04155
24.1
3
1
10
1.88314
48.29036
6.38
2
2
12
1.83076
49.76427
6.77
3
3
4
1.77038
51.58363
5.82
4
2
4
1.68245
54.49616
12.7
0
0
18
1.64277
55.92593
12.4
0
3
14
1.63532
56.20325
9.83
3
3
8
1.63528
56.20482
9.45
4
1
10
1.58228
58.26454
9.96
1
1
18
1.57336
58.62701
5.87
Figure A.4. List of indexed hkl reflections obtained from XRD pattern of RbCa0.4Sr1.6Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.92165
5.91817
96.8
1
0
1
7.59071
11.64870
10.6
0
1
1
7.49886
11.79187
9.22
0
1
3
6.11239
14.47960
6.59
1
1
1
5.42200
16.33517
6.1
0
2
0
3.87372
22.93972
24.1
2
0
2
3.79535
23.42000
30
0
2
2
3.74943
23.71097
6.75
2
0
4
3.47324
25.62732
5.25
1
2
1
3.45039
25.79992
30.9
2
1
3
3.30236
26.97778
9.29
2
1
4
3.16932
28.13302
6.61
221
1
1
2
0
2
2
1
3
2
1
3
1
3
0
4
0
4
4
4
4
1
0
4
2
4
3
2
0
1
3
2
1
0
1
2
2
2
0
2
3
2
3
2
3
0
4
1
1
0
1
4
0
2
4
2
2
1
3
2
0
4
8
6
9
1
2
8
3
6
4
1
7
5
9
2
0
0
1
4
2
2
16
2
3
4
11
16
14
16
15
3.14909
3.08971
3.08091
3.04844
2.74519
2.71100
2.54219
2.53024
2.41126
2.33037
2.16241
2.12625
2.03783
2.03746
1.94546
1.93686
1.90215
1.89830
1.89768
1.88688
1.86569
1.86521
1.73853
1.71096
1.70417
1.69371
1.64615
1.64396
1.64329
1.58366
28.31750
28.87350
28.95772
29.27301
32.59188
33.01464
35.27650
35.44863
37.26035
38.60401
41.73679
42.48063
44.41971
44.42801
46.65000
46.86955
47.77746
47.88047
47.89714
48.18850
48.77114
48.78469
52.60052
53.51479
53.74506
54.10400
55.80127
55.88210
55.90668
58.20884
5.36
10.3
351
16.4
190
45.8
10.6
6
5.74
6.21
12.9
5.31
11
10.9
102
29.2
5.4
5.24
5.29
5.25
10.6
16.8
9.11
23.2
30.4
6.11
61.3
49.1
18.3
16.6
Figure A.5. List of indexed hkl reflections obtained from XRD pattern of RbCa0.5Sr1.5Nb3O10.
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.91646
5.92023
48.9
0
2
1
3.85978
23.02366
69.4
0
2
2
3.76637
23.60279
26.9
2
0
2
3.74365
23.74810
26.3
0
1
7
3.73833
23.78244
9.8
1
0
7
3.73273
23.81859
12.7
0
0
8
3.72912
23.84204
8.12
0
2
4
3.45079
25.79686
36
2
1
2
3.37383
26.39584
11.4
2
1
3
3.27087
27.24250
81.6
1
2
4
3.15140
28.29639
15.4
2
1
4
3.14137
28.38857
9.69
1
1
8
3.08424
28.92577
15.1
0
2
6
3.06499
29.11146
109
2
0
6
3.05271
29.23118
47.6
222
0
1
2
0
0
0
2
2
0
1
1
2
2
1
0
4
1
2
2
2
1
3
2
1
0
1
0
2
1
2
2
2
3
1
3
3
1
4
0
4
2
4
3
2
3
1
9
9
5
10
7
10
0
1
8
4
12
1
2
14
0
0
1
12
4
11
16
8
16
3.04982
3.04678
2.99547
2.98329
2.87414
2.78575
2.74351
2.73198
2.69279
2.33640
2.26449
2.14926
2.13272
1.98639
1.94625
1.93372
1.88365
1.84223
1.69313
1.68715
1.64320
1.64213
1.64178
29.25951
29.28934
29.80260
29.92704
31.09181
32.10446
32.61246
32.75394
33.24436
38.50045
39.77380
42.00403
42.34539
45.63393
46.62998
46.95022
48.27636
49.43370
54.12411
54.33173
55.90999
55.94991
55.96282
74.6
90
27.8
27.8
17.4
8.34
15.3
210
25.2
37.2
9.1
11
8.55
9.56
116
16.1
9.15
16.5
11.8
8.42
38.8
57.5
9.81
Figure A.7. List of indexed hkl reflections obtained from XRD pattern of RbCa0.7Sr1.3Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.99434
5.88946
45.5
1
1
1
5.40788
16.37811
11.8
0
1
6
4.19585
21.15739
10.5
1
1
5
4.05286
21.91292
11.5
0
2
0
3.86086
23.01712
33
2
0
2
3.78808
23.46563
16.2
0
2
2
3.73891
23.77868
11.1
1
1
6
3.69833
24.04344
10.2
0
2
3
3.60157
24.69949
15
1
2
1
3.43994
25.87966
16.3
0
2
4
3.43246
25.93708
17.9
2
1
3
3.29655
27.02627
10.2
1
2
3
3.27204
27.23259
10.4
2
1
4
3.16539
28.16872
15.1
1
2
4
3.14367
28.36738
5.15
2
0
6
3.08209
28.94640
75.3
1
0
9
3.06601
29.10154
54.1
0
1
9
3.05939
29.16596
36
0
2
6
3.05543
29.20453
58.5
1
2
5
2.99887
29.76797
8.75
0
0
10
2.99887
29.76799
5.43
0
2
7
2.86803
31.15963
11.8
223
2
0
2
2
2
2
2
1
2
2
3
1
3
1
4
1
2
0
0
2
4
1
3
4
2
3
1
0
1
1
2
2
0
1
2
2
2
1
1
3
2
1
0
0
2
1
4
2
2
1
3
0
4
2
4
4
6
10
0
1
8
7
2
7
3
8
1
4
1
14
2
15
11
15
0
12
4
17
8
10
6
12
9
10
2.86249
2.79545
2.74901
2.73753
2.70760
2.70668
2.70394
2.69306
2.65061
2.55507
2.46425
2.32467
2.15671
1.99592
1.94107
1.93710
1.93574
1.93543
1.93043
1.84917
1.70044
1.67970
1.64644
1.63921
1.63602
1.63518
1.63360
1.62320
31.22149
31.99007
32.54540
32.68567
33.05736
33.06886
33.10331
33.24088
33.78920
35.09281
36.43061
38.70239
41.85227
45.40382
46.76193
46.86333
46.89815
46.90629
47.03495
49.23582
53.87247
54.59267
55.79055
56.05806
56.17729
56.20873
56.26785
56.66076
6.01
6.56
11.1
110
5.24
6.01
8.61
5.27
5.68
5.04
5.93
9.78
6.82
6.25
22.9
17.3
16.3
16.1
5.97
10.5
5.02
7.29
5.38
40.3
11.7
6.41
5.13
7.51
Figure A.8. List of indexed hkl reflections obtained from XRD pattern of RbCa1.1Sr0.9Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.09998
5.84822
115
0
1
3
6.15008
14.39038
14.9
1
1
1
5.42854
16.31535
6.33
0
1
4
5.41422
16.35882
6.31
0
2
0
3.88420
22.87697
49.8
2
0
2
3.79449
23.42541
72.6
1
2
1
3.45763
25.74497
27.7
0
2
4
3.45392
25.77310
17.8
1
2
3
3.28945
27.08564
96
1
0
9
3.08491
28.91940
137
0
1
9
3.08046
28.96207
279
0
2
6
3.07504
29.01420
60.7
2
1
5
3.02823
29.47281
39.7
1
2
5
3.01566
29.59846
12.3
0
2
7
2.88665
30.95358
7.35
1
1
9
2.86711
31.16990
5.7
1
2
6
2.86274
31.21868
11.8
0
1
10
2.81478
31.76459
5.91
2
2
1
2.74777
32.56049
312
224
2
2
2
1
2
2
0
3
0
3
1
3
2
1
3
3
1
2
2
1
3
2
0
2
4
0
4
4
4
4
2
1
2
0
2
4
2
4
1
3
1
1
2
2
0
1
2
2
2
1
2
0
0
1
1
1
2
3
2
2
2
2
3
1
2
3
3
2
0
4
1
0
1
1
2
4
2
4
4
2
4
0
4
2
4
2
3
4
8
7
2
7
3
8
9
3
12
1
11
3
7
6
1
3
12
10
5
14
6
6
10
11
2
0
0
4
1
3
12
5
13
7
3
4
6
10
9
12
10
17
13
8
2.71924
2.71799
2.71427
2.70889
2.66106
2.56654
2.53923
2.52966
2.51666
2.46880
2.45805
2.40534
2.32448
2.20931
2.16280
2.11975
2.03938
2.03703
2.03441
2.00910
1.99143
1.98546
1.96578
1.94617
1.94383
1.94210
1.90057
1.89724
1.89682
1.86758
1.85940
1.79952
1.77710
1.77094
1.71482
1.70475
1.64473
1.64418
1.64353
1.64272
1.59915
1.58228
1.58212
1.58041
225
32.91179
32.92728
32.97373
33.04106
33.65245
34.93087
35.31892
35.45696
35.64619
36.36112
36.52576
37.35542
38.70577
40.81075
41.72889
42.61725
44.38395
44.43807
44.49818
45.08950
45.51186
45.65639
46.13976
46.63189
46.69138
46.73552
47.81954
47.90871
47.92006
48.71863
48.94706
50.68870
51.37455
51.56623
53.38467
53.72543
55.85370
55.87396
55.89801
55.92795
57.59194
58.26450
58.27089
58.34016
6.84
12.7
26.5
5.79
5.26
5.51
10.3
13.9
6.01
5.82
6.77
9.46
64.5
9.96
35.6
15.3
5.63
5.97
7.81
7.11
5.18
7.97
9.72
23.6
113
35.1
23.8
17.5
14.8
11
22.2
6.36
6.02
5.71
10.6
43.2
63.7
55.7
30.3
13.4
5.54
5.79
6.46
26.6
Figure A.9. List of indexed hkl reflections obtained from XRD pattern of RbCa1.4Sr0.6Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.88440
5.93300
105
0
0
4
7.44220
11.88197
13.3
0
1
3
6.09310
14.52569
13
0
2
0
3.85994
23.02269
90.4
2
0
2
3.76066
23.63916
20.1
0
2
2
3.73635
23.79519
25
1
0
7
3.73087
23.83065
17.1
0
1
7
3.72490
23.86945
21.1
0
0
8
3.72110
23.89415
33.6
2
1
1
3.44822
25.81645
10.4
0
2
4
3.42649
25.98303
81.6
1
1
8
3.07804
28.98529
97.5
2
0
6
3.05968
29.16305
173
0
2
6
3.04655
29.29158
249
1
0
9
3.04358
29.32083
71.6
2
2
0
2.73883
32.66980
96.4
2
2
1
2.72731
32.81166
207
2
2
2
2.69360
33.23403
16.6
1
2
7
2.68260
33.37440
16.6
0
2
8
2.67896
33.42107
17.4
2
1
11
2.13436
42.31137
11.4
1
2
11
2.13100
42.38125
10.2
4
0
0
1.94338
46.70289
36.6
0
4
0
1.92997
47.04684
154
4
0
4
1.88033
48.36710
10.7
1
1
15
1.86590
48.76543
11.9
2
0
14
1.86544
48.77827
10.2
1
4
3
1.84060
49.48031
16.4
2
4
3
1.70295
53.78667
11
4
2
4
1.69042
54.21781
21.4
2
3
11
1.68132
54.53556
10.4
4
1
8
1.68126
54.53789
10.3
2
1
16
1.63989
56.03308
17.8
3
3
8
1.63918
56.05927
10.4
0
3
14
1.63915
56.06052
10.2
4
1
9
1.63745
56.12388
14.3
2
4
6
1.63236
56.31430
86.1
1
4
9
1.62990
56.40688
24.3
3
2
13
1.56794
58.84947
10.1
2
4
8
1.56770
58.85934
10.1
226
Figure A.10. List of indexed hkl reflections obtained from XRD pattern of RbCa1.9Sr0.1Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
14.86902
5.93914
65.3
0
0
4
7.43451
11.89431
11.7
0
0
6
4.95634
17.88193
7.92
2
0
0
3.85350
23.06170
83.1
0
2
2
3.74482
23.74063
10.9
2
0
2
3.73027
23.83458
47.8
0
1
7
3.72409
23.87470
21.6
1
0
7
3.72050
23.89810
11.4
2
1
1
3.42656
25.98250
47.8
2
0
4
3.42123
26.02365
15
1
2
3
3.26516
27.29107
6.19
0
2
5
3.24350
27.47686
5.03
1
2
4
3.13554
28.44252
6.46
1
1
8
3.07291
29.03481
36.6
0
2
6
3.05007
29.25700
308
2
0
6
3.04220
29.33445
62.4
0
1
9
3.03884
29.36755
5.25
0
0
10
2.97380
30.02477
6.42
0
2
7
2.86073
31.24121
7.86
1
2
6
2.83606
31.52006
7.35
0
1
10
2.77592
32.22126
5.62
2
2
0
2.73049
32.77229
215
2
2
2
2.68559
33.33614
20.5
1
2
7
2.68193
33.38287
9.59
0
2
8
2.68072
33.39844
7.27
1
1
10
2.61168
34.30833
6.5
2
2
4
2.56309
34.97942
5.37
3
0
3
2.48684
36.08823
6.37
2
2
7
2.29697
39.18816
6.2
0
2
11
2.21616
40.67893
10.2
0
3
7
2.20501
40.89394
6.69
2
3
0
2.14369
42.11846
6.95
2
3
1
2.13814
42.23298
6.21
3
2
1
2.13474
42.30341
14
1
2
11
2.12986
42.40513
6.5
2
1
11
2.12784
42.44726
7.81
3
0
9
2.02813
44.64344
9.35
2
3
5
2.01669
44.91037
8.58
2
2
10
2.01128
45.03799
8.27
0
4
0
1.93477
46.92306
29.1
0
4
1
1.93069
47.02822
109
1
4
1
1.87282
48.57349
8.78
0
4
4
1.87241
48.58491
16.8
4
1
0
1.86968
48.66042
9.26
0
3
11
1.86634
48.75309
7.65
4
1
1
1.86599
48.76274
6.03
4
0
4
1.86513
48.78672
11.2
4
2
0
1.72477
53.05276
11.1
2
4
2
1.71750
53.29495
5.73
227
1
0
2
4
2
0
0
3
2
1
0
3
2
4
4
4
2
3
0
3
3
4
4
4
0
4
7
8
4
4
11
18
14
8
6
9
11
15
8
1.71654
1.71622
1.68412
1.68015
1.67971
1.65211
1.63979
1.63483
1.63258
1.63176
1.57336
1.56952
1.56777
53.32700
53.33768
54.43734
54.57697
54.59243
55.58231
56.03667
56.22161
56.30618
56.33706
58.62689
58.78448
58.85663
12.9
17.9
31.7
9.25
10.6
6.34
15.3
20.6
52.4
33.6
7.64
13.3
8.42
Figure A.11. List of indexed hkl reflections obtained from XRD pattern of RbCa2Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.00294
5.88608
18.7
0
0
6
5.00098
17.72101
10.7
0
2
0
3.89119
22.83530
58.3
2
0
0
3.86377
22.99958
12.7
0
2
2
3.76657
23.60154
7.64
0
1
7
3.75468
23.67738
36.3
0
0
8
3.75074
23.70261
12.2
1
0
7
3.74847
23.71717
10.5
2
0
2
3.74168
23.76081
19.1
0
2
4
3.45413
25.77148
8.57
1
2
1
3.45236
25.78494
37.3
2
1
1
3.43794
25.89502
28.3
1
1
8
3.09580
28.81539
5.7
0
2
6
3.07106
29.05261
65.5
0
1
9
3.06461
29.11518
155
1
0
9
3.06123
29.14803
88.5
2
0
6
3.05753
29.18407
61.5
2
1
5
2.99795
29.77729
11.5
2
1
6
2.84578
31.40958
5.47
2
2
0
2.74174
32.63406
180
2
1
7
2.69270
33.24553
23.4
2
0
8
2.69124
33.26402
15.7
0
3
8
2.13355
42.32828
25.7
2
0
12
2.09924
43.05427
14
3
2
5
2.02226
44.78012
24.1
0
3
10
1.96242
46.22337
5.17
0
1
15
1.93741
46.85532
31.4
1
0
15
1.93656
46.87723
88.3
2
2
11
1.93376
46.94917
8.12
4
1
1
1.87133
48.61473
10.4
4
0
4
1.87084
48.62825
24.8
2
3
8
1.86772
48.71488
24.8
228
3
4
3
3
4
4
2
4
1
1
4
4
0
4
4
0
4
4
4
3
2
1
3
0
2
2
4
2
4
3
1
2
3
2
1
4
0
1
2
3
8
3
3
13
2
3
4
4
8
13
8
5
14
6
9
10
10
10
8
10
1.86390
1.84288
1.79805
1.71898
1.71897
1.70504
1.69289
1.68609
1.68549
1.68299
1.67710
1.66263
1.65229
1.63524
1.63427
1.63246
1.62434
1.59007
1.57122
1.56101
48.82121
49.41507
50.73305
53.24528
53.24583
53.71568
54.13229
54.36876
54.38977
54.47701
54.68430
55.20069
55.57582
56.20624
56.24283
56.31056
56.61754
57.95189
58.71473
59.13673
9.07
25.1
5
12.6
12.4
6.76
6.06
7.08
27
18
8.24
10.9
7.61
83.7
25.9
10.1
14.7
5.46
30.8
8.17
Figure A.12. List of indexed hkl reflections obtained from XRD pattern of RbBa0.15Sr1.85Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.18911
5.81387
888
0
2
0
3.88558
22.86871
165
0
0
8
3.79728
23.40796
95.4
2
1
1
3.46814
25.66566
118
1
2
4
3.16311
28.18939
93.8
1
1
8
3.12676
28.52399
143
1
0
9
3.09868
28.78811
92.5
0
1
9
3.09594
28.81414
422
2
0
6
3.09333
28.83892
1.22e+003
0
2
6
3.08247
28.94275
991
0
0
10
3.03782
29.37764
237
1
2
5
3.01924
29.56257
571
2
0
7
2.90381
30.76619
105
1
2
6
2.86747
31.16594
139
1
0
10
2.83142
31.57301
565
2
2
0
2.75521
32.47017
2.08e+003
2
0
8
2.72318
32.86276
248
2
1
7
2.72011
32.90089
82.9
2
2
9
2.13441
42.31036
86
2
0
12
2.12459
42.51529
97.5
3
0
9
2.06222
43.86676
280
2
3
6
1.98600
45.64322
110
3
0
10
1.97747
45.85151
116
1
0
15
1.96045
46.27242
102
2
2
11
1.95050
46.52235
340
4
0
1
1.94968
46.54321
848
229
0
1
1
1
2
2
3
1
4
4
1
1
2
0
2
2
4
4
4
2
2
3
4
3
1
4
2
4
0
2
1
2
0
4
4
4
2
0
2
1
2
3
4
3
0
10
15
2
12
3
14
16
9
6
18
9
6
10
15
17
7
10
8
13
8
10
1.94279
1.91122
1.90090
1.87104
1.86413
1.71451
1.66723
1.66663
1.65223
1.65017
1.64965
1.64602
1.64522
1.63670
1.63181
1.62508
1.61940
1.60767
1.58595
1.58579
1.58156
1.57182
46.71789
47.53663
47.81089
48.62262
48.81482
53.39529
55.03545
55.05682
55.57819
55.65345
55.67255
55.80593
55.83554
56.15166
56.33521
56.58942
56.80572
57.25861
58.11673
58.12328
58.29379
58.69020
74.2
93.1
141
217
105
121
87.7
121
85.9
233
318
219
112
170
106
100
216
122
85.4
89
125
90.8
Figure A.13. List of indexed hkl reflections obtained from XRD pattern of RbBa0.3Sr1.7Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.13804
5.83350
15.3
2
0
1
3.96155
22.42444
10.1
0
2
0
3.93215
22.59432
6.55
2
1
2
3.46769
25.66899
11.7
2
1
3
3.35927
26.51232
5.97
2
1
4
3.22326
27.65284
31.3
2
0
6
3.13263
28.46949
8.09
0
1
9
3.09292
28.84282
18.6
2
1
5
3.07045
29.05851
10.2
1
2
5
3.04847
29.27275
35
0
0
10
3.02761
29.47899
49.2
2
0
7
2.93504
30.43088
10.8
1
1
9
2.88444
30.97790
7.01
1
0
10
2.83125
31.57497
47.8
2
2
0
2.80272
31.90491
22.6
2
2
1
2.79079
32.04499
44.1
2
2
2
2.75588
32.46198
9.46
0
3
1
2.61166
34.30856
6.45
1
3
8
2.08064
43.45861
5.9
0
3
9
2.06775
43.74345
21.4
2
3
5
2.06099
43.89424
19.2
2
1
12
2.05894
43.94038
8.37
4
0
1
1.99362
45.45916
6.61
230
4
0
0
3
4
2
1
2
0
2
4
1
2
2
0
2
2
3
2
4
3
0
4
4
3
2
3
3
4
0
1
0
2
4
3
4
2
4
3
0
0
0
2
0
2
1
1
10
12
5
18
16
10
16
6
12
10
15
7
9
17
11
15
1.98078
1.96607
1.94970
1.86493
1.77813
1.77543
1.77256
1.69369
1.68200
1.67113
1.66758
1.66756
1.66527
1.65466
1.64891
1.63788
1.63346
1.63343
1.62669
1.61687
1.60878
45.77054
46.13248
46.54263
48.79237
51.34233
51.42614
51.51568
54.10474
54.51165
54.89597
55.02295
55.02349
55.10569
55.48938
55.69976
56.10775
56.27316
56.27422
56.52811
56.90280
57.21541
34.6
11.9
5.32
11
6.82
5.86
7.41
8.79
8.27
7.9
5.39
5.24
6.48
8.28
6.58
7.02
5.58
5.6
8.44
12.7
5.42
Figure A.14. List of indexed hkl reflections obtained from XRD pattern of RbBa0.6Sr1.4Nb3O10
h
k
l
d (Å)
2θ
Intensity
0
0
2
15.36151
5.74857
387
2
0
0
3.88861
22.85070
132
2
0
1
3.85783
23.03549
502
0
0
8
3.84038
23.14160
97.2
0
1
7
3.83323
23.18534
71.4
0
2
2
3.81203
23.31609
121
1
1
6
3.75778
23.65754
431
2
0
3
3.63536
24.46635
296
2
1
0
3.48628
25.52984
398
1
0
8
3.44344
25.85293
174
1
2
2
3.42296
26.01030
166
2
1
2
3.39982
26.19045
109
1
2
3
3.32144
26.81992
82.7
2
0
5
3.28589
27.11559
108
1
1
8
3.15470
28.26614
123
0
1
9
3.13176
28.47752
106
0
2
6
3.12017
28.58558
96.1
2
0
6
3.09683
28.80566
195
0
0
10
3.07230
29.04064
669
1
2
5
3.04860
29.27144
1.98e+003
2
1
5
3.03222
29.43313
819
2
0
7
2.91057
30.69297
99
1
0
10
2.85742
31.27830
84.9
2
2
0
2.76596
32.34048
149
1
2
7
2.74181
32.63329
367
2
0
8
2.73245
32.74817
502
231
2
2
1
1
3
2
2
1
2
0
3
1
2
2
3
4
1
4
3
4
4
1
1
3
2
1
1
2
0
3
0
4
2
4
1
3
1
2
3
0
4
0
3
0
1
2
1
0
1
0
3
0
3
3
1
1
0
3
2
0
3
0
1
1
1
0
4
3
2
4
0
4
2
2
0
1
1
2
0
1
1
4
3
3
2
1
1
2
7
2
10
11
7
12
3
14
4
9
8
14
13
6
6
1
10
2
10
0
2
16
4
1
13
6
17
3
16
11
18
8
16
5
18
14
18
7
9
15
8
19
15
18
2.72987
2.72218
2.68588
2.62863
2.14742
2.13838
2.12734
2.11203
2.09251
2.08011
2.07281
2.03986
2.01958
2.00171
1.99397
1.94042
1.93248
1.92891
1.92136
1.88756
1.87347
1.86421
1.85123
1.84066
1.79677
1.78747
1.76033
1.73038
1.72570
1.71105
1.70683
1.69400
1.68194
1.67697
1.66716
1.63827
1.63097
1.63005
1.62240
1.61443
1.58728
1.58392
1.57463
1.56588
232
32.77996
32.87517
33.33242
34.08026
42.04185
42.22796
42.45784
42.78058
43.19959
43.47016
43.63112
44.37309
44.84268
45.26501
45.45066
46.77837
46.98223
47.07421
47.27063
48.17017
48.55570
48.81249
49.17726
49.47860
50.77186
51.05485
51.89996
52.86740
53.02190
53.51178
53.65451
54.09402
54.51383
54.68912
55.03801
56.09338
56.36681
56.40135
56.69111
56.99654
58.06343
58.19864
58.57508
58.93454
292
695
265
127
77.2
75.6
84.3
70.5
135
75.2
70.6
74.8
77.5
126
142
143
465
378
116
86.1
179
228
78.1
76.6
84.8
158
113
96.1
91.3
95.5
73.9
246
88.7
190
206
193
533
75
205
132
78.1
86.5
72
133
Appendix B: C1-6/CaxSr2-xNb3O10 TGA Data
C1/Sr2Nb3O10 TGA
100
0.16
Weight %
Derivative Weight %
0.14
99.5
0.12
0.1
99
0.08
0.06
98.5
0.04
0.02
98
0
97.5
-0.02
100
200
300
400
500
600
700
Temperature, Degrees C
Figure B.1. TGA data for C1/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C. See Table B.1
for mass loss % and coverage %.
C3/Sr2Nb3O10
101
0.07
100
0.06
99
0.05
98
0.04
97
0.03
96
0.02
95
0.01
94
0
93
-0.01
0
100
200
300
400
500
600
700
800
Figure B.2. TGA data for C3/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C. See Table B.1
for mass loss % and coverage %.
233
100
0.16
99
0.14
98
0.12
0.1
97
0.08
96
0.06
95
0.04
94
0.02
93
0
92
Derivative Weight %
Weight %
C6/Sr2Nb3O10 TGA
-0.02
100
200
300
400
500
600
700
Temperature, Degrees C
Figure B.3. TGA data for C6/Sr2Nb3O10 with temperatures ranging from 150 to 800 °C. See Table B.1
for mass loss % and coverage %.
C1/Ca0.4Sr1.6Nb3O10 TGA
100
0.16
Weight %
0.12
99
0.1
0.08
98.5
0.06
98
0.04
0.02
97.5
Derivative Weight %
0.14
99.5
0
97
-0.02
100
200
300
400
500
600
700
Temperature, Degrees C
Figure B.4. TGA data for C1/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
234
C3/Ca0.4Sr1.6Nb3O10
101
0.05
100
0.04
99
0.03
98
0.02
97
0.01
96
0
95
-0.01
94
93
-0.02
0
100
200
300
400
500
600
700
800
900
Figure B.5. TGA data for C3/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
C6/Ca0.4Sr1.6Nb3O10
101
0.16
100
0.14
99
0.12
98
0.1
97
96
0.08
95
0.06
94
0.04
93
0.02
92
0
91
90
-0.02
0
100
200
300
400
500
600
700
800
900
Figure B.6. TGA data for C6/Ca0.4Sr1.6Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
235
C1/Ca1Sr1Nb3O10
101
0.16
0.14
100
0.12
99
0.1
98
0.08
97
0.06
0.04
96
0.02
95
0
94
-0.02
0
100
200
300
400
500
600
700
800
900
Figure B.7. TGA data for C1/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
C3/Ca1Sr1Nb3O10
101
0.25
100
0.2
99
98
0.15
97
0.1
96
95
0.05
94
0
93
92
-0.05
0
100
200
300
400
500
600
700
800
900
Figure B.8. TGA data for C3/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
236
C6/Ca1Sr1Nb3O10
102
0.25
100
0.2
98
0.15
96
94
0.1
92
0.05
90
0
88
86
-0.05
0
100
200
300
400
500
600
700
800
900
Figure B.9. TGA data for C6/Ca1Sr1Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
C1/Ca1.5Sr0.5Nb3O10
101
0.1
100
0.08
99
0.06
98
0.04
97
0.02
96
0
95
-0.02
0
100
200
300
400
500
600
700
800
900
Figure B.10. TGA data for C1/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
237
C3/Ca1.5Sr0.5Nb3O10
101
0.07
100
0.06
99
0.05
98
0.04
97
0.03
96
0.02
95
0.01
94
0
93
-0.01
0
100
200
300
400
500
600
700
800
Figure B.11. TGA data for C3/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
C6/Ca1.5Sr0.5Nb3O10
101
0.16
100
0.14
99
0.12
98
0.1
97
96
0.08
95
0.06
94
0.04
93
0.02
92
0
91
90
-0.02
0
100
200
300
400
500
600
700
800
900
Figure B.12. TGA data for C6/Ca1.5Sr0.5Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
238
C1/Ca1.8Sr0.2Nb3O10 TGA
100
0.16
Weight %
0.12
0.1
99
0.08
0.06
98.5
0.04
0.02
98
Derivative Weight %
0.14
99.5
0
97.5
-0.02
100
200
300
400
500
600
700
Temperature, Degrees C
Figure B.13. TGA data for C1/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
C3/Ca1.8Sr0.2Nb3O10
101
0.035
0.03
100
0.025
99
0.02
0.015
98
0.01
97
0.005
0
96
-0.005
95
-0.01
94
-0.015
0
100
200
300
400
500
600
700
800
900
Figure B.14. TGA data for C3/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
239
C6/Ca1.8Sr0.2Nb3O10
102
0.25
100
0.2
98
0.15
96
94
0.1
92
0.05
90
0
88
86
-0.05
0
100
200
300
400
500
600
700
800
900
Figure B.15. TGA data for C6/Ca1.8Sr0.2Nb3O10 with temperatures ranging from 150 to 800 °C. See
Table B.1 for mass loss % and coverage %.
C1/Ca2Nb3O10 Microwave Grafting TGA
0.05
Weight %
0.04
99
0.03
0.02
98
0.01
0
97
Derivative Weight %
100
-0.01
100
200
300
400
500
600
700
Temperature, Degrees C
Figure B.16. TGA data for C6/Ca2Nb3O10 with temperatures ranging from 150 to 800 °C. See Table
B.1 for mass loss % and coverage %.
240
Документ
Категория
Без категории
Просмотров
0
Размер файла
7 886 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа