Xenon-129 NMR Studies of Gas Adsorption Dynamics in Crystalline Nanotube Materials

MISSING IMAGE

Material Information

Title:
Xenon-129 NMR Studies of Gas Adsorption Dynamics in Crystalline Nanotube Materials
Physical Description:
1 online resource (76 p.)
Language:
english
Creator:
Akel, Christopher D
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Chemistry
Committee Chair:
Bowers, Clifford R
Committee Co-Chair:
Smith, Benjamin W
Committee Members:
Mareci, Thomas H

Subjects

Subjects / Keywords:
adsorption -- hyperpolarization -- hyperpolarize -- hyperpolarized -- nanotube -- nmr -- seop -- xenon
Chemistry -- Dissertations, Academic -- UF
Genre:
Chemistry thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Nanotube materials have a wide range of practical applications in the field of chemistry.  Characterizing one-dimensional (1D) nanotubular systems is vital in regards to optimizing their applications.  In channels large enough for molecules to pass by one another, normal diffusion (ND) is observed where the mean-squared displacement is proportional to the observation time.  In contrast, nanotubes having channels too small for molecules to pass one another exhibit single-file diffusion (SFD).  In this case, one may expect a square-root time dependency of the mean-squared displacement. To achieve the results of this study, a 129Xe polarizer and gas handling system was built to allow for spin-exchange optical pumping (SEOP) of rubidium metal which is the process utilized to hyperpolarize xenon gas in our lab.  This involved the design and construction of an oven for the optical pumping cell, the re-designing of the entire gas handling system previously built in our lab, and the design of a new and larger optical pumping cell. Continuous-flow hyperpolarized 129Xe nuclear magnetic resonance (NMR) was then employed to systematically investigate adsorption dynamics of xenon (Xe) occurring in two different 1D nanotubular materials: self-assembled L-phenylalanyl-L-phenylalanine (FF) and self-assembled bis-urea macrocycles.  The channel diameters of FF and the bis-urea macrocycle are ~10Å and ~4Å, respectively.  FF also possesses a large central channel with a diameter of approximately 100nm.  Upon cooling of the sample to temperatures below -75ºC, xenon adsorption into the smaller channels of FF is observed, whereas adsorption into the larger channel is present at all temperatures studied. It was demonstrated through saturation and inversion recovery experiments that a temperature dependence exists for T1 in FF nanotubes.  As the temperature is decreased, T1 gets shorter.  Alternatively, as the temperature was decreased, the chemical shift of the adsorbed phases increased.  Temperature discrepancies between the tracer-exchange and saturation recovery experiments made it difficult to directly compare the T1 values.  This was due to a warming effect of the sample in the continuous-flow experiments caused by the flow-rate and gas stream. The Xe partial-pressure dependence study performed on the bis-urea macrocycle demonstrated that the chemical shift anisotropy was axially symmetric and positive at low loading and then lost its axial symmetry with increased loading.  The tracer-exchange experiments were inconclusive in regards to the type of diffusion occurring in the channel.  This study, along with the FF experiments, demonstrates that Xe NMR can be utilized to obtain large amounts of information in regards to adsorption dynamics and characteristics of crystalline nanotubular materials.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Christopher D Akel.
Thesis:
Thesis (M.S.)--University of Florida, 2012.
Local:
Adviser: Bowers, Clifford R.
Local:
Co-adviser: Smith, Benjamin W.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044758:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 XENON 129 NMR STUDIES OF GAS ADSORPTION DYNAMICS IN CRYSTALLINE NANOTUBE MATERIALS By CHRISTOPHER AKEL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DE GREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Christopher Akel

PAGE 3

3 To my family and friends

PAGE 4

4 ACKNOWLEDGMENTS Many people have helped me succeed throughout my graduate career at the University of Florida. First and foremo st, I would like to thank my research advisor, Dr. Russ Bowers. It has been an honor and a privilege to study physical chemistry through Nuclear Magnetic Resonance under his expert guidance. His creativity, knowledge, and advice have been an inspiration throughout my time at UF, and I have gained more than merely scientific knowledge as a student in his group. Secondly, I would like to thank Dr. Muslim Dvoyashkin, the post doctoral student in our group with whom I collaborated with on all of these studi es. Many thoughtful, creative, and insightful discussions took place between the two of us, for which I am incredibly grateful. From data analysis to planning future experiments, Dr. Dvoyashkin was instrumental in my success at the graduate school level. I would also like to thank Ryan Wood, a senior graduate level student in our research group. His willingness to lend his expertise in NMR and the workings of Varian software were directly responsible for some of the results contained in this thesis. Man y intuitive discussions took place between him, me and Dr. Dvoyashkin which allowed this project to come together in a scientifically responsible fashion. Ronghui Zhou, another member of our research group with whom Dr. Dvoyashkin and I collaborated wit h on the design of the gas handling system also deserves recognition. The combining of the parahydrogen converter and xenon polarizer into a single but independently functioning apparatus would not have been possible without his help Two more group memb ers that I would like to thank are Hrishi Bhase and Turgut Sonmez for their acceptance in further carrying out the research contained in this thesis

PAGE 5

5 and for making good use of the xenon polarizer which was constructed during my time at UF. On a different n ote, I would like to thank my committee members, Drs. Ben Smith and Thomas Mareci, for their advice and scientific expertise. Dr. Linda Shimizu of the University of South Carolina deserves gratitude for providing us with the bis urea samples which were st udied using hyperpolarized 129 Xe NMR. The University of Florida Department of Physics Cryogenic Services, namely Greg Labbe and John Graham, were instrumental in the successful construction of the gas handling system/xenon polarizer. Their willingness to share their proficiency in Swagelok parts and adapters and in leak testing is greatly appreciated. In addition to this, I would like to thank the throughout my time here. Joe Shalosky and Todd Prox of chemistry really came through for us when our re circulating pump died and were directly responsible for the successful construction of the oven for the xenon polarizer. Mark Link and his colleagues in physics made many small parts for us and were more than willing to offer their expertise at any time of the day, for which I am appreciative. Finally, I would like to thank my family and friends for their support, encouragement, and understanding during my graduate career of s tudying NMR. Without them, none of this would have been possible.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 2 BACKGROUND ................................ ................................ ................................ ...... 16 Introduction ................................ ................................ ................................ ............. 16 Brief NM R Overview ................................ ................................ ............................... 16 129 Xe NMR ................................ ................................ ................................ .............. 18 Spin Exchange Optical Pumping and Hyperpolarization ................................ ........ 19 Optical Pumping ................................ ................................ ............................... 19 Spin Exchange ................................ ................................ ................................ 22 Continuous Flow Hyperpolarized 129 Xe NMR ................................ ......................... 24 3 DESIGN AND CONSTRUCTION OF THE 129 XENON HYPERPOLARIZER ....... 26 Brief Background ................................ ................................ ................................ .... 26 Polarizer Design/Construction ................................ ................................ ................ 27 Sample Holder ................................ ................................ ................................ .. 29 Hyperpolarization Chamber ................................ ................................ .............. 30 4 NANOPOROUS MATERIALS ................................ ................................ ................. 36 L Phenylalanyl L Phenylalanine ................................ ................................ ............. 36 Bis Urea Macrocycle ................................ ................................ ............................... 41 5 METHODS AND PULSE S EQUENCES ................................ ................................ 44 Spin (Hahn) Echo ................................ ................................ ................................ ... 44 Hyperpolarized 129 Xe Tracer Exchange ................................ ................................ .. 45 I nversion Recovery ................................ ................................ ................................ 48 Saturation Recovery ................................ ................................ ............................... 49 6 RESULTS AND DISCUSSION ................................ ................................ ............... 51

PAGE 7

7 L Pheynyalanyl L Pheynlalanine ................................ ................................ ............ 51 Xenon Adsorption at Various Temperatures ................................ ..................... 51 Tracer Exchange ................................ ................................ .............................. 53 T 1 Estimation ................................ ................................ ................................ .... 56 Bis Urea Macrocycle ................................ ................................ ............................... 64 Xenon partial pressure dependence ................................ ................................ 64 Tracer Exchange ................................ ................................ .............................. 69 7 CONCLUSION AND RECOMMENDATIONS ................................ ......................... 71 LIST OF REFERENCES ................................ ................................ ............................... 73 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 76

PAGE 8

8 LIST OF TABLES Table page 2 1 NMR p roperties of 1 H and 129 Xe ................................ ................................ ......... 18 6 1 Summary t able of T 1 relaxation t imes ................................ ................................ 63

PAGE 9

9 LIST OF FIGURES Figure page 2 1 The Nuclear Zeeman Effect ................................ ................................ ................ 17 2 2 Energy level splitting in 87 Rb showing optical transitions ................................ .... 20 2 3 Spin exchange optical pumping process inside of the pumping cell ................... 23 3 1 129 Xe gas handling system and p olarizer ................................ ............................ 28 3 2 Top view of the NMR probe and sample holder ................................ .................. 29 3 3 Optical pumping cell for SEOP ................................ ................................ ........... 31 3 4 O ven constructed for SEOP ................................ ................................ ............... 32 3 5 Magnetic field of Helmholtz coils ................................ ................................ ........ 33 3 6 Fiber optic coupled laser diode array ................................ ................................ 34 3 7 Fully assembled oven in the hyperpolarization chamber ................................ .... 34 3 8 129 Xe polarizer/gas handling system ................................ ................................ ... 35 4 1 Schematic s tructure of FF ................................ ................................ .................. 36 4 2 Model for the se lf assembly of the FF nanotubes ................................ ............... 38 4 3 Channel of FF Nanotubes ................................ ................................ .................. 39 4 4 Self assembled FF nanotubes ................................ ................................ ............ 40 4 5 SEM image s of FF ................................ ................................ .............................. 41 4 6 Bis urea macrocycle 1 ................................ ................................ ........................ 42 4 7 Illustration of xenon adsorption into the bis u rea macrocycle 1 nanochannels ... 43 5 1 Spin (Hahn) echo pulse sequence ................................ ................................ ...... 44 5 2 Illustration of the hyperpolarized NMR tracer exchange ................................ ..... 45 5 3 Tracer exchange/selective saturation recovery pulse sequence ........................ 46 5 4 Single and normal file diffusion in 1D channels ................................ .................. 48 5 5 Inversion recovery pulse sequence ................................ ................................ .... 49

PAGE 10

10 5 6 Saturation recovery pulse sequence ................................ ................................ .. 50 6 1 Variable temperature hyperpolarized 129 Xe study on FF nanotubes ................... 52 6 2 Tracer exchange data on FF nanotubes ................................ ............................ 54 6 3 Tracer exchange fits to the small, 1nm channels of FF ................................ ...... 55 6 4 Tracer exchange fits to the large, central channels of FF ................................ ... 55 6 5 Inversion recovery experiments on Xe gas reference sample at 25 C ............... 58 6 6 Saturation recovery experiments on Xe gas reference sample at 25 C ............. 58 6 7 Inversion recovery experiments on Xe gas reference sample at 85 C .............. 59 6 8 Saturation recovery experiments on Xe gas reference sample at 85 C ............ 59 6 9 Density and temperature dependence of 129 Xe spin relaxation times ................. 60 6 10 Satura tion recovery experiments on the small channels of FF nanotubes at 85 C ................................ ................................ ................................ ................... 61 6 11 Satura tion recovery experiments on the small channels of FF nanotubes at 95 C ................................ ................................ ................................ ................... 62 6 12 Saturation recovery experiments on t he large channels of FF nanotubes at 85 C ................................ ................................ ................................ ................... 62 6 13 Saturation recovery experiments on the large channels of FF nanotubes at 95 C ................................ ................................ ................................ ................... 63 6 14 Dependence of the 129 Xe line shape in the one dimensional pores of bis urea crystals on Xe partial pressure in a He/Xe mixture with a total pressure of 3115 Torr ................................ ................................ ................................ ............ 65 6 15 Structure and channel diameter of ALPO 1 1 ................................ ...................... 66 6 16 Structure of one channel in ALPO 11 indicating the three types of xenon si tes ................................ ................................ ................................ ................... 67 6 17 129 Xe NMR line shapes for xenon adsorbed in ALPO 11. ................................ .. 68 6 18 Tracer exchange data on bis urea macrocycle ................................ ................... 70 6 19 Tracer exchange fits for the bis urea macrocycle ................................ ............... 70

PAGE 11

11 LIST OF ABBREVIATION S 1D One d imensional FF L phenylala nyl L phenylalanine HP Hyperpolarize ND Normal Diffusion NMR Nuclear Magnetic Resonance PEEK Polyether Ether Ketone PFA Perfluoroalkoxy R B Rubidium RF Radiofrequency SEM Scanning Electron Microscopy SEOP Spin Exchange Optical Pumping SFD Single File Diffus ion X E Xenon

PAGE 12

12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements fo r the Degree of Master of Science XENON 129 NMR STUDIES OF GAS ADSORPTION DYNAMICS IN CRYSTALLI NE NANOTUBE MATERIALS By Christopher Akel August 2012 Chair: Clifford R. Bowers Co Chair: Ben jamin W. Smith Major: Chemistry Nanotube materials have a wide range of practical applications in the field of chemistry. Characterizing one dimensional (1D) n anotubular systems is vital in regards to optimizing their applications. In channels large enough for molecules to pass by one another, normal diffusion (ND) is observed where the mean squared displacement is proportional to the observation time. In cont rast, nanotubes having channels too small for molecules to pass one another exhibit single file diffusion (SFD). In this case, one may expect a square root time dependency of the mean squared displacement. To achieve the results of this study, a 129 Xe pol arizer and gas handling system was built to allow for spin exchange optical pumping (SEOP) of rubidium metal which is the process utilized to hyperpolarize xenon gas in our lab. This involved the design and construction of an oven for the optical pumping cell, the re designing of the entire gas handling system previously built in our lab, and the design of a new and larger optical pumping cell. Continuous flow hyperpolarized 129 Xe nuclear magnetic resonance (NMR) was then employed to systematically investi gate adsorption dynamics of xenon (Xe)

PAGE 13

13 occurring in two different 1D nanotubular materials: self assembled L phenylalanyl L phenylalanine (FF) and self assembled bis urea macrocycles. The channel diameters of FF and the bis urea macrocycle are ~10 and ~4 respectively. FF also possesses a large central channel with a diameter of approximately 100nm. Upon cooling of the sample to temperatures below 75 C, xenon adsorption into the smaller channels of FF is observed, whereas adsorption into the larger ch annel is present at all temperatures studied. It was demonstrated through saturation and inversion recovery experiments that a temperature dependence exists for T 1 in FF nanotubes. As the temperature is decreased, T 1 gets shorter. Alternatively, as the t emperature was decreased, the chemical shift of the adsorbed phases increased. Temperature discrepancies between the tracer exchange and saturati on recovery experiments made it dif ficult to directly compare the T 1 values. This was due to a warming effect of the sample in the continuous flow experiments caused by the flow rate and gas stream. The Xe partial pressure dependence study performed on the bis urea macrocycle demonstrated that the chemical shift anisotropy was axially symmetric and positive at lo w loading and then lost its axial symmetry with increased loading. The tracer exchange experiments were inconclusive in regards to the type of diffusion occurring in the channel This study, along with the FF experiments, demonstrates that Xe NMR can be utilized to obtain large amounts of information in regards to adsorption dynamics and characteristics of crystalline nanotubular materials.

PAGE 14

14 CHAPTER 1 INTRODUCTION In recent years, nanoporous materials with one dimensional (1D) channels have been studied intensively due to their possible applications to gas storage separation and catalysis. 1 Diffusion of gases through these 1D channels represents the primary focus of this thesis N anoporous materials possessing 1D channels with diameter s that are suffic ient to allow for particles inside of the channel to pass one another will exhibit normal diffusion, where the mean squared displacement is proportional to the observa tion time. For channel s too narrow to allow for the mutual passing of particles, single file diffusion may be observed in which the mean squared displacement is proportional to the square root of the observation time. Xenon 129 Nuclear Magnetic Resonance is a spectroscopic technique that is well suited to the study of nanoporous solids. 2 Bene fits of 129 Xe NMR include its large chemical shift range which is sensitive to the size, shape, and loading of nanopores and its long T 1 relaxation time which leads to a larger range of achievable observation times. Given these characteristics, a physisor bed phase can readily be distinguished from a bulk gas phase inside of a 1D nanochannel by a chemical shift which helps reveal the interactions of xenon atoms with the surfaces of the solid. Accordingly, 129 Xe NMR has been utilized to study the structures of solids such as zeolites 3 polymers 4 nanotubes 5 and proteins 6 Given the fact that conventional NMR suffers from inherently small signals, hyperpolarized 129 Xe NMR has been developed to improve sensitivity. 7 Hyperpolarized 129 Xe NMR signals can be up to four orders more intense in comparison to signals obtained from thermally polarized 129 Xe NMR. 8 Contained in this thesis is the

PAGE 15

15 application of continuous flow hyperpolarized 129 Xe NMR spectroscopy to nanoporous solids containing 1D channels to explo re xenon gas adsorption and diffusion within the nanochannels. Chapter 2 of this thesis will give a brief background of the essential topics necessary for understanding this thesis. Chapter 3 will give a detailed account of the 129 Xe polarizer and gas han dling system that was constructed so that these experiments could be carried out. Chapter 4 will describe the nanoporous mat erials that were studied while C hapter 5 introduces the reader to the methods and pulse sequences that were to acquire the data, wh ich is presented in C hapter 6. Finally, C hapter 7 will sum up the results through a conclusion as well as offer suggestions and recommendations for future studies.

PAGE 16

16 CHAPTER 2 BACKGROUND Introduction Chapter 2 represents a brief overview of NMR and h yperpolarized NMR that will be critical to understanding the remaining chapters of this thesis. Section 2.2 will focus on the basics of NMR whereas section 2.3 will identify the physical properties and applications of 129 Xe. Section 2.4 will discuss the technique of spin exchange optical pumping which is the method commonly utilized to hyperpolarize noble gases such as xenon to enhance the sensitivity of NMR, and finally, section 2.5 will discuss the technique of continuous flow hyperpolarized 129 Xe NMR. Brief NMR Overview Nuclear magnetic resonance is the phenomenon where electromagnetic radiation is absorbed and re emitted by nuclei in a magnetic field. This energy is at a specific resonance frequency which is dependent on the magnetic properties of the nuclei of interest and the strength of the magnetic field. Isotopes having an odd number of protons and/or neutrons (nonzero spin) have a magnetic moment and possess angular momentum and can be studied using NMR. 9 When a static external magnetic field ( B 0 ) is applied, it interacts with the magnetic dipole moment of the nuclei causing the spins to order preferentially along the magnetic field. This creates a difference in the energy levels and is referred to as the Nuclear Zeeman Effect, which is represe nted in Figure 2 1 and forms the basis for all NMR experiments. The NMR signal is proportional to the difference in the populations of the spin states. Each spin state is populated according to a Boltzmann distribution: (2 1)

PAGE 17

17 where and represent the number of spins in the and states, respectively, is the energy difference arising from the Nuclear Zeeman Effect, is the Boltzmann constant, and is the temperature. 9 Figure 2 1. The Nuclear Zeeman Effect As mentioned previously, NMR suffers from inherently low sensitivity. This can be demonstrated using standard values for a proton in the presence of a 400 MHz (9.4T) external magnetic field, wher eas the would be approximately 2.66 x 10 25 J Using this value and E quation (2 1) at standard temperature, one obtains This represents a very small difference in the spin state populations of the and states which correl ates to a small signal thus rendering NMR quite insensitive. However, a technique known as hyperpolarization, which seeks to increase the

PAGE 18

18 difference in the populations of the spin states, can be utilized to overcome this insensitivity. 129 Xe NMR Xenon has a molecular weight of 131.29 grams per mole. It contains 54 electrons and has the electron configuration [Kr] 4d 10 5s 2 5d 6 Of the many stable isotopes of xenon, two have a nonzero spin and can be studied using NMR: 129 Xe with I = 1/2 and 131 Xe with I = 3 /2. It was first applied to NMR in 1982 by Fraissard and Ito. 10 Xenon is a noble gas meaning that its external electronic orbital is completely occupied. As such, it is chemically inert. At standard temperature and pressure, xenon exists as a gas. Gi ven its standard boiling and melting points of 165.02K and 161.38K respectively, both solid and liquid phases can readily be obtained within most laboratory environments. 11 The long T 1 relaxation time of 129 Xe makes it a favorable candidate for hyperpolar ized NMR experiments. Table 2 1. NMR p roperties of 1 H and 129 Xe Nuclei 1 H 129 Xe Spin, I Magnetogyric Ratio (10 8 s 1 Tesla 1 ) 2.675 0.74 Natural Abundance, (%) 99.99 26.44 As can be seen above in T able 2 1, 129 Xe has a much lower magnetog yric ratio and natural abundance than 1 H. Together, these combine to make the NMR sensitivity for 129 Xe much worse than for 1 H at thermal equilibrium. As a result, signal averaging must be employed to increase the signal to noise ratio in NMR experiments when

PAGE 19

19 working with thermally polarized 129 Xe. As mentioned previously, however, this can be overcome through hyperpolarization of xenon which will be discussed shortly. Xenon 129 is a monatomic gas possessing a spherically symmetric electron cloud. Any d istortion to this cloud is reflected in the chemical shielding of the nucleus which in turn will be observed as a change in the NMR chemical shift. Hence, 129 Xe has a chemical shift that is very sensitive to interactions with the local environment which a gain makes it an ideal candidate for the study of nanoporous materials. 11 Spin Exchange Optical Pumping and Hyperpolarization Spin exchange optical pumping (SEOP) is the primary method used to hyperpolarize gases such as 129 Xe 83 Kr and 3 He. 12 This me thod is suitable for producing large amounts of hyperpolarized gases for NMR measurements as well as medical imaging. Optical Pumping Optical pumping begins with an alkali metal vapor. In this first step of the SEOP process, angular momentum is transfer red from circularly polarized photons of light to alkali metal electrons. This process was developed by Kastler in 1957, for which he was awarded the Nobel Prize for Physics. 13 Rubidium is often the alkali metal of choice for the creation of hyperpolariz ed gas through SEOP due to its large spin exchange cross sections, relatively high vapor pressures at moderate temperatures and the availability of high power light sources that emit electromagnetic radiation at its D line absorption wavelengths, which ca n be seen in F igure 2 2. 1 4 Two stable isotopes of rubidium exist in nature: 85 Rb with I = 5/2 and a natural abundance of 72.2%, and 87 Rb with I = 3/2 and a natural abundance of 27.2% The total electronic angular momentum is represented by where is

PAGE 20

20 the electron spin and is the orbital angular momentum. 11 The total atomic angular momentum is represented by In the presence of an external magnetic field, B 0 the Hamiltonian is: 11,12,1 5 (2 2) where A is the hyperfine coupling constant and I and S are the nuclear and electron spins of rubidium, respectively. The next two terms are representative of the magnetic dipole couplings of the electron and nuclear spins to the magnetic field, B 0 where is the g factor of an electron, or 2.00232. I corresponds to the nuclear spin quantum number. The Bohr magneton and nuclear magnetic moment are represented by and respectively. To ensure that the hyperfine interaction is domina nt, which is the first term in E quation ( 2 2 ) a weak magnetic field of approximately 10 20 gauss is commonly used to optically pump rubidium for spin exchange applications. Figure 2 2 Energy level splitting in 87 Rb showing optical transitions Figure 2 2 shows the energy levels and transition wavelengths of 87 Rb. Electrons in the ground state can be excited to higher energy states by absorbing photons at the D transitio ns. The D 1 transition is most commonly used for optical pumping of rubidium and this wavelength of light was used throughout all of our SEOP experiments.

PAGE 21

21 The rubidium atom is spherically symmetric with an orbital spin of L = 0 It possesses one optically a ctive electron having an orbital angular momentum of 0, 1, 2, etc. and an electron spin of 1/2. The spin orbit coupling results in a total momentum of As a result of this coupling, the first excited energy level with an orbital angular momentum of splits into a 2 P 1/2 and a 2 P 3/2 state. With optical absorption from the ground state of 2 S 1/2 to the 2 P 1/2 and 2 P 3/2 the doublet D 1 and D 2 transitions are observed. 12 In optical pumping a heated glass cell containing rubidium vapor is pla ced in a relatively weak magnetic field and is irradiated with circularly polarized light at the D 1 transition wavelength of approximately 795 nm T he absorption of light excites transitions between the various Zeeman levels according to the select ion rules. T he excited states decay spontaneously back to the ground state and re emit light in all spatial directions. 15 The irradiation of the atomic vapor within the pumping cell using circularly polarized light induces a polarization of the rubidium. This occurs through the process of depopulation optical pumping where the circularly polarized light passes angular momentum to the rubidium vapor atoms. Through a combination of irradiation with circularly polarized light and relaxation back to the ground state, a strong non thermal equilibrium polarization of the ground state can be induced 15 Once polarization of the rubidium has been achieved through irradiation and optical pumping, transfer of spin angular moment um from the alkali atom to the Xe nuclear spin can occur by the process of spin exchange within the pumping cell.

PAGE 22

22 Spin Exchange Inside of the pumping cell and during spin exchange collisions, the rubidium electron gets flipped due to its coupling with the nuclear spin of colliding xen on atoms, which can be seen in F igure 2 3. The hyperfine interaction between the electron spins and the nuclear spins of the noble gas is: (2 3) Where again, A is the hyperfine coupling constant that is dependent on the distance between the electron and the nucleus, and I and S are the nuclear and electron spins of rubidium, respectively. The term in parenthesis of E quation (2 3) allows spin exchange to happen 15 Buffer gases such as nitrogen and helium are often mixed with the rubidium vapor inside of the pumping cell to increase efficiency of the optical pumping process. Nitrogen prevents radiation trapping because it nonradiati vely de excites, or quences, the excited rubidium atoms before they have a chance to re radiate a photon. 12 Helium allows for pressure broadening of the D 1 absorption line. This results from collision interactions between the rubidium electron cloud and t he buffe r gases and helps to improve absorption efficiency of the light sources emitting radiation at the D 1 transition of about 795 nm. 11 When atoms of rubidium collide with buffer gases, a mixing of atomic sublevels occurs. It has been reported by Walke r and Happer that the efficiency of the optical pumping process can be enhanced by a factor of 1/3 to 1/2 due to the collision mixing of the rubidium excited states. 15

PAGE 23

23 Figure 2 3 Spin exchange optical pumping process inside of the pumping cell As seen in F igure 2 3, thermally polarized xenon, with only a slight po pulation difference between the and states which translates to a small NMR signal, enters the pumping cell. Inside of the pumping cell which is also in the presence of an external magnetic field (not shown), r ubidium vapor is optically pumped by 795 nm laser light. Xenon atoms col lide with polarized rubidium atoms and through spin exchange, xenon becomes hyperpolarized and exits the pumping cell. As can be seen on th e right side of Figure 2 3 after the process of SEOP, there exists a large difference between the populations of th e and states which correlates to a vastly improved NMR signal. This deviation from the thermal equilibrium populations of the alpha and beta states is what is referred to as hyperpolarization throughout this thesis.

PAGE 24

24 Continuous Flow Hyperpolarized 129 Xe NMR Thermally polarized 129 Xe NMR suffers from long acquisition times resulting from its long T 1 relaxation time and low spin polarization. Together, these create the need for a fairly long recycle delay and much signal averaging resulting in lengthy exp eriments. Using hyperpolarized 129 Xe through the SEOP process just discussed, NMR sensitivity can be drastically increased. Prior to the use of continuous flow, hyperpolarized xenon experiments were carried out using a batch method developed by Pines an d others. 16 During this process, a pumping cell having valves is filled with rubidium metal and a mixture of xenon and buffer gases. After the rubidium has been optically pumped and the polarization transferred to xenon, the gas mixture containing hyperp olarized xenon is expanded into the NMR probe. This process works well but the development of a continuous flow apparatus has made the entire process much more efficient. The method of continuously flowing the hyperpolarized 129 Xe throughout NMR measure ments was developed by Happer and others and represents the technique utilized throughout the studies contained in this thesis. 12 During the process of continuous flow hyperpolarized 129 Xe NMR, a gas re circulating pump constantly refreshes the xenon ins ide of the pumping cell and the NMR probe at a flow rate of approximately 75 150 ml/min. This range gives ample time for the 129 Xe gas to interact with the rubidium atoms inside of the pumping cell and become hyperpolarized. Upon exiting the pumping cell the xenon gas mixture flows through the sample that is contained in the NMR probe and the signal is acquired. In contrast to the batch method previously discussed, continuous flow experiments do not rely on any sort of gas expansion to transfer the hype rpolarized xenon to the NMR probe. Instead, it

PAGE 25

25 takes advantage of retaining the high xenon spin polarizations throughout the entire NMR experiment. Utilizing this method allows for the spin polarizations to be replenished on a time scale that is dependen t on the flow rate and not the T 1 relaxation time. Accordingly, the recycle delay for the experiments is no longer limited by the long longitudinal relaxation time of xenon. This allows for phase cycling, signal averaging, and 2D NMR experiments on a pla usible time scale. Over the years, many continuous flow hyperpolarized 129 Xe NMR apparatuses have been designed with their main goal being to maximum xenon polarization that can be transferred to NMR samples. 17,18 Many experimental variables such as laser power, gas pressure and composition, and flow rate have been analyzed to optimize xenon polarization. 19 Accordingly, using all of the available information and taking into account previous design flaws and successes, an apparatus was developed in our lab to efficiently polarize xenon in a continuously flowing operational mode and capably transfer this hyperpolarized xenon to our NMR probe and sample. Chapter 3 will discuss, in detail, the xenon polarizer that was designed, built, and utilized in our lab for NMR studies of nanoporous solids and other materials.

PAGE 26

26 CHAPTER 3 DESIGN AND CONSTRUCT ION OF THE 129 XENON HYPERPOLARIZER Brief Background In our lab, a 129 Xe polarizer was previously built by Zook, a former graduate student. 17 In 2005, this polarizer was re designed by Cheng, another former graduate student, to improve its efficiency and versatility. 11 This polarizer performed exceptionally well and provided many experimental NMR results. However, it did possess certain limitations. Br ass valves and copper tubing, over time, led to rather sever leaks, while the permanent silver soddered connections made the accompaniment of a vast array of experiments difficult. This also made the addition of parts to the system and the replacement of leaking valves a complex situation. Leaks are detrimental and harmful to a xenon polarizer. Rubidium is a highly reactive alkali metal and will ignite upon reaction with the moisture in the atmosphere of a laboratory. This poses a major safety hazard mak ing the achievement of a leak tight apparatus a necessity. Even minute amounts of moisture can deactivate the rubidium within the pumping cell resulting in the loss of experiment time, as the pumping cell will then have to be laboriously cleaned, dried, a nd pressure tested before being loaded once again with rubidium. This loading process must take place inside of a glove box that is absent of moisture, a dangerous and often times stressful task where numerous safety precautions must be taken. Consequent ly, in the name of safety and experimentation time, it is best practice to develop a system that is completely free of leaks while at the same time functional with the ability to interchange parts should the need arise.

PAGE 27

27 Polarizer Design/Construction To acc omplish this task, our polarizer was constructed solely out of stainless steel parts. Though costly, this offers the benefit of a longer life and better resistance to wear when compared to brass and copper parts. With the exception of the gas recirculati ng pump, gas reservoir, solenoid valves, and flow controller, all of the parts used to construct the 129 Xe polarizer/gas handling system were purchased new. This made the accomplishment of a leak tight system feasible. Figure 3 1 represents our design for the gas handling system and polarizer. The red line indicates the flow path of xenon. Thermally polarized xenon enters the pumping cell. Upon exiting the pumping cell, the xenon is hyperpolarized through SEOP and then travels through the NMR probe and sample that is inside of a superconducting magnet. It then travels through a flow controller which simultaneously indicates and regulates the gas flow. After that, it travels through a rotary vane gas recirculating pump that is capable of achieving both low and high pressure s It then proceeds through a moisture trap before again entering the pumping cell for re hyperpolarization. The dashed blue line indicates the aluminum table upon which the gas handling system was built. It should be noted that the pumping cell is contained inside of a hyperpolarization chamber and is not in the center of the table. Several bypasses have been built into this system to allow for the easy activation of samples and testing of flow rates. These include bypasses for the moisture trap, the pumping cell, and the NMR probe/sample.

PAGE 28

28 Figure 3 1 129 Xe gas handling system and p olarizer

PAGE 29

29 Utilizing these bypasses, we have the ability to further activate a sample by passing xenon, nitrogen, or helium gas through the NMR prob e/sample and the moisture trap while bypassing the rubidium inside of the pumping cell. When the sample has been sufficiently activated and dried, we can then pass xenon through the pumping cell and the NMR probe/sample while bypassing the moisture trap. This is done to prevent contamination within the moisture trap from reaching the rubidium. Another benefit of these bypasses is that they allow the sample and system to be tested for leaks before exposure to the pumping cell and rubidium. As will soon be discussed, our sample holder is made of a plastic and achieving a leak tight seal is not guaranteed. Accordingly, this must be checked before experiments can proceed. These bypasses offer a sizeable advantage over the previous polarizer utilized by Zoo k and Cheng, while the utilization of all stainless steel parts should provide the gas handling system with leak resilience for an indefinite period of time. Sample Holder Our custom designed sample holder is made from polyether ether ketone (PEEK). It fe gas through the sample holder while maintaining a leak tight seal utilizing PEEK flange free fer rules. This is represented in F igure 3 2. Figure 3 2 Top view of the NMR probe and sample holder (Photo courtesy of author)

PAGE 30

30 As can be seen in F igure 3 probe. It then goes into the sample holder where a leak tight seal between the tubing and the sample holder is created. From there, the xenon gas travels through the sample, which is packed into the narrower diameter of the sample holder and is surrounded by the radiofrequency (RF) coil. This RF coil attaches to two brass leads that sit atop the NMR probe. The sample is held in place by two g lass wool plugs on each side of the sample. The sample is packed into the sample holder using a custom machined tool that allows the sample to be packed around a brass rod that extends through the sample holder. This creates a path for the gas to travel through the sample and sample holder at flow rates that will allow for sufficient xenon polarization and subsequent NMR experiments. Upon exiting the sample holder, the gas travels back through the bottom of the magnet where it is re introduced to the 129 Xe polarizer. Hyperpolarization Chamber The most important and critical part of the 129 Xe polarizer is the hyperpolarization chamber. This is where the process of SEOP takes place to hy perpolarize the xenon. Accordingly, the highly reactive rubidium metal is found here contained inside of the pump ing cell, which can be seen in F igure 3 3. The inlet of the pumping cell, which is shown on the ri ght side of Figure 3 3 is fitted wit h an Ace threaded adapter allowing handling system via a Swagelok connection. The outlet, see n on the left si de of Figure 3 3 is fitted with an identical nnect the pumping cell and the NMR probe/sample. The reason for using tubing with a reduced diameter for xenon exiting the pumping cell is to transfer the hyperpolarized gas to the sample as quickly as

PAGE 31

31 possible. This prevents loss of signal from relaxati on. It should also be noted that care was taken to minimize stainless steel valves between the outlet of the pumping cell and the NMR probe, as this could also weaken the polarization. The bulb on the bottom of the pumping cell, pictured in the bottom c en ter of Figure 3 3 is where rubidium is stored. The rubidium is loaded into the pumping cell through the inlet stopcock in a nitrogen filled glove box. This is normally done about twice per year depending on the use of the 129 Xe polarizer. Figure 3 3 Optical pumping cell for SEOP (Photo courtesy of a uthor) Once filled with approximately 0.5g of rubidium, the pumping cell is placed inside of the custom d esigned oven, seen in F igure 3 4 This design allows the pumping cell to slide into the oven through the back via the two slits that were cut into the sides. Once the pumping cell is inside, two panels are screwed in place to cover the slits, as shown. Two end plates are also attached to the front and back of the oven via screws.

PAGE 32

32 tical flat directly in the center of them to allow laser light to enter and exit the oven. Copper tubing enters through the bottom of the oven and makes a closed, uniform loop throughout the oven. Small holes were drilled into this coil to allow for an e ven distribution of warm air. This air is pre heated outside of the oven to a set temperature depending on the laser power being used, as irradiation also provides additional heating The air enters the closed loop of the coil and this pressure build up evenly forces the air through the small holes, which in turn heats the oven. The holes were drilled in a manner that prevented direct heating of the pumping cell, as this could result in unneeded stress on the cell causing it to explode. The temperature of oven is normally kept between 130 and 140 C, and its legs are mounted to the laser table using two clamps. Figure 3 4 O ven constructed for SEOP (left) and the pumping cell placed inside of the oven (right) ( Photos courtesy of author) The oven is moun ted to the laser table directly between a Helmholtz pair, wh ich can be seen to the left of F igure 3 4. This pair consists of two circular magnetic coils

PAGE 33

33 that are placed symmetrically on each side of the oven along a common axis. The coils possess an iden tical electrical current which flows in the same direction for both. These coils create a magnetic field surrounding the pumping cell of approximately 20 gauss which helps to create and retain the xenon polarization during gas circulation. The magnetic field of the pair is shown in F igure 3 5, where the homogenous region around 20 G represents the placement of the pumping cell. Figure 3 5 Magnetic field of Helmholtz coils The final aspect of the hyperpolarization chamber is the fiber optic coupled laser diode array. This allows for optical pumping of the rubidium inside of the pumping cell and oven through irradiation, which is seen in F igures 3 6 and 3 7. A laser trap is placed at the back window of the oven to catch any laser light that was not absorbed by the rubidium. This laser trap is cooled using a continuous, re circulating flow of cold water to prevent excessive heating. 0 5 10 15 20 30 20 10 0 10 20 30 Magnetic Field (G) Distance From Center of the Oven (cm)

PAGE 34

34 Figure 3 6 Fiber optic coupled laser diode array (Photo courtesy of author) Figure 3 7 Fully assembled oven in the hyperpolarization chamber (Photo courtesy of author)

PAGE 35

35 Upon exiting the hyperpolarization chamber, which is encased by aluminum and flame retardant green laser curtains, the xenon quickly travels to the NMR probe/sample that is contained in the bore of the 9.4 Tesla superconducting magnet. The fully a ssembled 129 Xe polarizer/g as handling system is shown in F igure 3 8. Figure 3 8 129 Xe polarizer/gas handling system ( Photo courtesy of author) W ith this polarizer, NMR studies were conducted on two nanoporous materials which will now be discussed.

PAGE 36

36 CHAPTER 4 NANOPOROUS MATE RIALS L Phenylalanyl L Phenylalanine L phenylalanyl L phenylalanine (FF), shown in F igure 4 1, is a dipeptide that is the core recognition motif of amyloid polypeptide and is associated with diseases such as 20 It is known to be capa ble of s elf assembly upon crystallization. Due to the chemical and physical stability and rigidity of its nanotubes, FF has garnered attention in the nanotechnology world. To date, numerous promising applications of FF nanotubes have been explored and developed. Figure 4 1. Schematic s tructure of FF [ Image reproduced with permission from Gorbitz, C.H. European Journal of Chemistry 2001 7 (Page 5153). John Wiley and Sons. ] In 2003, Reches and Gazit reported the self assembly of FF through the dilution of a concentrated solution of the dipeptide in 1,1,1,3,3,3 hexafluoro 2 propanol with water. 21 Here, they showed that the stiff and discrete nanotubes that formed could serve as casts for the formation of silver nanowires possessing long, persistent lengths. T his was performed through reduction of ionic silver within the FF nanotubes followed by enzymatic degradation of the peptide backbone.

PAGE 37

37 Further investigations of FF nanotubes by Reches and Gazit have shown that they possess unique mechanical properties maki ng them some of the stiffest biological materials currently known to exist. 22 This was demonstrated and directly measured in 2005 through indentation type experiments using atomic force microscopy. These characteristics make FF attractive building blocks for the design and assembly of biocompatible nanodevices. In 2004, synthesis of peptide nanotube platinum nanoparticle composites were reported using the D phenylalanly D phenylalanine enantiomer. 23 This enantiomer is resistant to degradation by enzymes. Utilizing electron microscopy and Monte Carlo simulations, it was determined that the nanotubes were porous and were capable of forming the composites. This study also demonstrated an alternative to the crystallization method previously mentioned by Rech es and Gazit which omitted the use of the 1,1,1,3,3,3 hexafluoro 2 propanol making the procedure more environmentally friendly. This crystallization method involved dissolving the dipeptide in water at an elevated temperature, waiting for equilibration, a nd then gradually cooling the sample to room temperature. This resulted in nanotubes that were 100nm to 2 m in diameter and in excess of 100 m in length In 2006, a study was performed investigating the chemical stability of FF nanotubes. 24 This involved mixing the nanotubes with numerous organic solutions including acetone, ethanol, acetonitrile, isopropanol, and methanol for approximately 30 minutes. These mixtures were then analyzed using scanning electron microscopy which still revealed the assembled FF nanotubes. In addition to the remarkable

PAGE 38

38 chemical stability reported, it was also demonstrated that the nanotubes possess extraordinary thermal stability at temperatures up to 90 C. Gazit and Reches, in 2004, investigated the rationale behind the mole cular self assembly of FF nanotubes. 25 They hypothesized that one dimensional propagation occurs through the linking of the dipeptides via hydrogen bonding. The inter sheet stacking interaction that results between the aromatic groups of FF offers an ener getic contribution to form extended sheets. The final tubular and nanoporous structure occurs through closure of the extended sheets along one axis, which is illustrated in F igure 4 2 This hypothesis is in agreement with other evidence found in 2005 which suggested the importance of FF groups in amyloid fibril formation through stacking interactions. 26 Figure 4 2 Model for the self assembly of the FF nanotube s [ Image adapted with permission from Reches, M; Gazit, E. Nano Letters 2004, 4 (P age 584, Figure 5). American Chemical Society ] In 2001, the molecular packing of FF nanotubes was investigated by Gorbitz. 27 Here, it was determined that the nanotubes have chiral hydrophilic channels with a van or 1nm. Within these hydrophilic channels, water molecules stay well organiz ed, which is shown in F igure 4 3

PAGE 39

39 Figure 4 3 Channel of FF Nanotubes. Water positio ns inside a si ngle channel (left), and a top view of the chan nel (right ). [ Image reproduced with permission from Gorbitz, C.H. European Journal of Chemistry 2001 7 (Page 515 7, Figures 4 and 5 ). John Wiley and Sons. ] These results were consistent with other results reported in 2004 where it was found that parallel channels ran in the direction of the FF tube axis. 23 In 2009, vertical alignment of FF nanotubes was achieved using vapor deposition methods. 28 This approach allowed the length and density of the nanotubes to be adjusted by cautiously controlling the supply of the building blocks from the gas phase. It was further shown that these nanotube arrays could be used to develop high surface area electrodes for energy storage applications, microfluidic chips, and highly hydrophobic self cleaning surfaces. In another study, vertical alignment was attained by using a thin film of FF. Here, the dipeptide was dissolved in 1,1,1,3,3,3 hexafluoro 2 propanol in the absence of water to produce an amorphous peptide thin film. This film acted as a template and, under the proper conditions, FF nanorod s grew. 29 In 2006, Reches and Gazit developed methods for horizontally and vertically aligning FF nanotubes. 30 Vertical alignment was achieved through axial unidirectional

PAGE 40

40 growth of a dense array of the dipeptide. Horizontal alignment was attained throu gh a non covalent coating of the nanotubes with a ferrofluid and the application of an external magnetic field, which can be seen schematically in F igure 4 4 Figure 4 4 Self assembled FF nanotubes coated with magnetic particles (left), coated na notubes before exposure to an external magnetic f ield (center), and horizontally aligned coated nanotubes after exposure to an external magnetic field (right ). [ Image reproduced with permission from Reches, M.; Gazit, E. Nanture Nanotechnology 2006, 1 (Page 198, Figure 4). Nature Publishing Group .] For the experiments and results presented in this thesis, the lyophilized form of FF was purchased from MP Biomedicals and was used as purchased. Some attempts were made to re crystallize the FF utilizing the method of Reches and Gazit which involved an environm entally hazardous solvent, but it was ultimately determined through scanning electron microscopy (SEM) that the dipeptide, as purchased, possessed adequate nanochannels for 129 Xe NMR studies. Figure 4 5 represents SEM images of the purchased FF that was u tilized throughout these studies.

PAGE 41

41 Figure 4 5 SEM image s of FF showing the s iz e distribution (left) and the actual nanochannel (right) Bis Urea Macrocycle As mentioned previously, hollow nanotubular structures have a wide range of app lications in biology, chemistry, and material science. As was seen with FF, there is great interest in the development of new molecular building blocks that predictably self assemble into nanotubular structures. Accordingly, in 2001, a series of bis urea macrocycles that are readily synthetically accessible and self assemble into columnar nanotubes was designed. 31 Bis urea macrocycle 1, as referred to in the literature, was cho sen for the present NMR studies and can be seen in F igure 4 6 Like FF, this macrocycle self assembles into nanotubes. This self assembly process is guided primarily by hydrogen bonding and aromatic stacking interactions that yield crystals of filled host 1 acetic acid. 32 These acetic acid guest molecules are bound in the cylindr ical cavities of the bis urea crystal. They can be removed through heating to form a stable crystalline apohost 1.

PAGE 42

42 Figure 4 6 Bis urea macrocycle 1. This apohost possessed a type 1 carbon dioxide gas adsorption isotherm which is consistent with an open framework microporous material. It is able to bind a range of small molecule guests with explicit stochiometry. Similar to zeolites, apohost 1 can be reused because the formation of these inclusion complexes does little to destroy the crystal fra mework. The nanochannels formed by bis urea macrocyle 1 have an elliptical shape. The diameter of these nanochannels is approximately 3.7 x 4.8 32 129 Xe NMR has not previously been reported in the literature on any bis urea macrocycles. 129 Xe possesse s a collision diameter of approximately 4.4 2 As demonstrated in F igure 4 7 129 Xe adsorption into the channels of bis urea macrocycle 1 is not favorable given the dimensions of the nanochannels. Nevertheless, and as will be discussed shortly, xenon ads orption into the channels does occur.

PAGE 43

43 Figure 4 7 Illustration of xenon adsorption into the bis urea macrocycle 1 nanochannel s. [ Image adapted with permission from Dewal, M.B.; Lufaso, M.W.; Hughes, A.D.; Samuel, S.A.; Pellechia, P.; Shimizu, L.S. Chemisty of Materials 2006 18 (Page 4861, Figure 8) American Chemical Society. ]

PAGE 44

44 CHAPTER 5 METHODS AND PULSE SEQUENCES Spin (Hahn) Echo In NMR, the refocusing of magnetization utilizing a pulse of resonant electromagnetic radiation is referred to as a spin echo. The spin echo pulse sequence is depicted below in Figure 5 1. It was discovered in 1950 by Erwin H ahn and was further developed by Carr and Purcell. 33,34 Figure 5 1. Spin (Hahn) echo pulse sequence The observable NMR signal following an initial excitation pulse decays with time due to spin relaxation and inhomogeneous effects. These inhomogen eous effects cause different spins within a sample to precess at different rates. 35 Spin relaxation leads to a loss of magnetization that is irreversible. However, inhomogeneous effects such as a magnetic field gradient or a distribution of chemical shift s can be removed by applying a pulse which inverts the magnetization vectors. If this inversion pulse is applied after a time of dephasing, the inhomogeneous evolution will rephrase to form an echo at time 2 At this time, the chemical shift and field inhomogeneity are refocused leading to a truer representation of the actual NMR signal.

PAGE 45

45 This pulse sequence was utilized for the temperature dependence study of FF nanotubes. The experimental parameters included a recycle delay of 5s, a pressure of approximately 2300 Torr, a gas fl ow rate of about 250mL/min, and 128 scans. With the exception of the inversion and saturation recovery experiments, all of the data collected for this thesis were done on a Bruker 400MHz Avance spectrometer. Hyperpolarized 129 Xe Tracer Exchange The tr acer exchange method can be used to study the diffusion time scaling over a wide range of time scales, from approximately 10ms to 1000s. The experiment begins with the destruction of the spin polarization of xenon atoms inside of the nanochannels and is t hen followed by measurement of the recovery of the polarization as a f unction of time. A t time t = 0, there is no signal as the spin polarization of the xenon atoms inside of the channel has been saturated. As t increases, however, the hyperpolarized xe non atoms outside of the channels exchange with the unpolarized atoms inside of the channels leading to the observation of an NMR signal growing in due to atoms At sufficiently long t values, the hyperpolarization distribut ion reaches a steady state, resulting in the maximum observable NMR signal. T his is depicted graphically in F igure 5 2. Figure 5 2. Illustration of the hyperpolarized NMR tracer exchange

PAGE 46

46 To achieve saturation at time t = 0, a train of selectively satu rating pulses is applied to destroy the polarization inside of the nanochannels. A certain amount of time then passes which allows hyperpolarized atoms outside of the channels to exchange with the unpolarized atoms inside of the channel. Once that time h as passed, an acquisition pulse is applied and the NMR signal is collected. This is repres ented in the pulse sequence of F igure 5 3. Figure 5 3. Tracer exchange/selective saturation recovery pulse sequence This method was first demonstrated in 2000, w here xenon diffusion inside tris (o phenylenedioxy) cyclotriphosphazene, or TPP, nanochannels was measured with respect to time as a function of temperature and occupancy utilizing continuous flow hyperpolarized 129 Xe saturation recovery NMR. 36 These expe riments made use of non selective saturating pulses which destroyed all of the xenon magnetization inside of the sample space. The recovery of the xenon signal inside of the channels was monitored as a function of diffusion time under continuous flow cond itions. This use of non selective saturating pulses creates an ill defined initial condition for the magnetization kinetics modeling

PAGE 47

47 A well defined initial condition is obtained by using frequency selective saturating RF pulses, which was demonstrated in 2007 by Cheng and Bowers 37 This makes it possible to destroy the xenon magnetization inside of the nanochannels while leaving the magnetization of the xenon bulk gas phase unaffected. The selective saturation recovery pulse sequence has been employed throughout the studies contained in this thesis. The tracer exchange curve, (t) can be defined as: (5 1) As demonstrated in the literature, this tracer exchange curve can be used to distinguish between single file diffusion (SFD) and normal diffusion (ND) in nanotubular materials 36 38 Single file diffusion occurs when the atoms inside of the channel are unable to pass one another. When this happens, th e mean squared displacement is proportional to the squa re root of the observation time: (5 2) Here, F is the single file mobility. Normal diffusion occurs when the atoms inside of the channel can pass one another. This results in a mean squared displacement that is propor tional to the observation time, (5 3) Here, D is the diffusivity. These two modes of diffusion are illustrated in F igure 5 4. The tracer exchange fitting functions for SFD and ND are sh own in E quations ( 5 4 ) and ( 5 5 ) respectively. 37 (5 4) (5 5)

PAGE 48

48 Figure 5 4. Single and normal file diffusion in 1D channels In these functions, T 1c is a fi tting parameter and represents the spin lattice relaxation time of the tracer inside the channels Inversion Recovery In NMR, the time constant T 1 is referred to as the spin lattice relaxation time, or longitudinal relaxation time. It represents the time to reestablish the thermal equilibrium Boltzmann distribution following a perturbation The T 1 inversion recovery experiment is a common method for measuring this time constant In the inversion re covery pulse sequence, sho wn in F igure 5 5 the magnetization, M 0 is initially inverted with a pulse. Following this is a time which allows for relaxation along the +z axis. A fter this time has passed, the NMR signal is measured by applying a pulse.

PAGE 49

49 Figure 5 5. Inversion recovery pulse sequence Utilizing the method, the longitudinal relaxation time constant can be determined by fitting the magnetization curve usi ng the following function: 9 (5 6 ) Alternatively, T 1 can be directly calculated utilizing the following formula, where is the corresponding to the node at : (5 7 ) Another technique that is sometimes used to measure T 1 relaxation is the saturation recovery method. Saturation Recovery In the saturation recovery method for the measurement of the longitudinal relaxation, any initial magnetization present in the beginning of the experiment is destroyed using a train of pulses. Longitudinal magnetization then accumulates for a time It is then converted into detectable transverse magnet ization using a pulse and the NMR signal is acquired. This is shown in the pulse sequence of F igure 5 6. The NMR signal represents the recovery of the longitudinal magnetization back towards its thermal equilibrium value of as a function of

PAGE 50

50 Figure 5 6 Saturation recovery pulse sequence The saturation recovery pulse s equence is similar to the tracer exchange experiment with the exception being that all magnetization is initially saturated rather than only magnetization in a specif ic environment. Besides this initial difference in the type of pulse used to destroy the magnetization, the two are identical. A major advantage of using the saturation recovery method over the inversion recovery method to measure T 1 relaxation is that no recycle delays have to be included between scans. This makes signal averaging much more cost effective with saturation recovery by drastically decreasing the experiment time, especially in samples with rather long T 1 values. Using saturation recovery, T 1 can be determined by fitting the magnetization curve using the following function: 40 (5 8 ) The inversion and saturation recovery experiments were carried out on a Varian Inova 500MHz spectrometer.

PAGE 51

51 CHAPTER 6 RESULTS AND DISCUSSION L Pheynyalanyl L Pheynlalanine Xen on Adsorption at Various Temperatures The first study performed on FF included a variable temperature study using the spin echo pulse sequence mentioned previously. The FF nanotubes contain two channels of different sizes Each of these channels gives ri se to a distinct NMR signal as seen in the variable temperature study. The large central channel s of crystalline FF have diameters ranging from 100 to 300nm. 20 The exact nature of this inner surface is not known, but it is believed to be either mixed hy drophobic/hydrophilic, or entirely hydrophobic. T he small er individual and narrow channels of FF make up the dipeptide wall. The diameters of these channels are approximately 10 or 1nm. These channels are hydrophilic in nature and have the ability t o accommodate guest molecules of some size. The temperature study that was carried out on the FF nanotubes utilizing a Bruker Avance 400MHz spe ctrometer is shown in F igure 6 1 These were hyperpolarized 129 Xe spin echo variable temperature experiments u sing pure natural abundance xenon gas. Spectra were acquired at temperatures rang ing from room temperature to 96 C. At temperatures greater than 75 C, only two peaks are observed and those spectra have been omitted from Figure 6 1 However, as the tem perature is decreased below 75 C, a third peak begins to grown in around 240ppm. The intensity of this peak continues to increase with decreasing temperature at the expense of the intensity of the middle peak, which is centered around 60ppm. This peak a round 240ppm represents xenon adsorption into the small, 1nm channels while the broad peak close to 60ppm

PAGE 52

52 represents xenon adsorption into the large, central channel. The narrow peak close to 0ppm represents the bulk xenon gas, which was referenced to 0pp m at 25 C. Figure 6 1 Variable temperature hyperpolarized 129 Xe study on FF nanotubes utilizing the spin echo pulse sequence with 128 scans and a recycle delay of 5s. These results indicate a temperature dependence of the adsorbed xenon chemical shift T he chemical shift of both phases decreases with increasing temperature. This is in agreement with a study performed in 1995 which analyzed the temperature dependence of the xenon chemical shifts in other microporous solids. 41 Here, it was determined that the chemical shift is sensitive to the shape of the potential experienced

PAGE 53

53 by the xenon atom inside of the pore. It was found that if the pore is about the same size as the xenon atom, then the potential has a single minimum around the center of the p ore. This, in turn, leads to a chemical shift that increases with temperature. 41 However, for larger pores, as is the case with FF nanotubes, the potential has a hump near the center which is surrounded by a low energy trough. In this type of system, the xenon chemical shift decreases with increasing temperature, which represe nts the trend seen in F igure 6 1 These results also show that at decreasing temperatures, xenon redistributes from the large, central channel to the small, narrow channels makin g up the wall of FF. This is evidenced by the decreasing signal at 60ppm and the increasing signal at 240ppm as the temperature is decreased. The intensity of the bulk xenon gas peak remains unchanged as the two adsorbed peaks undergo exchange. To furth er probe the xenon diffusion through these two channels, the tracer exchange method was employed. Tracer Exchange The adsorption of Xe into the small channels of FF at 93 C was chosen for the tracer exchange experiments. The spectra for these experiments acquired as a function of time, are shown in F igure 6 2 The experiments were carried out with values ranging from 0.1ms 200s. However, after about 5s, the hyperpolarized xenon steady state was re achieved. Figure 6 2 shows the recovery of the po larization from 0 6s. The integrals of the signals arising from the adsorption into the two different channels were referenced to the bulk xenon gas peak. This was done because the bulk xenon peak fluctuated very little and remained essentially constant t hroughout the experiments. The integrals were normalized to an average of the value which

PAGE 54

54 correspon ds to the hyperpolarized steady state. These values were then fitted to the tracer exchange fitting functions described in E quations ( 5 4 ) and ( 5 5 ) Figure 6 2 Tracer exchange data on FF nanotubes As seen in F igures 6 3 and 6 4, both the small and large channels of FF show a very reasonable fit to the NMR tracer exchange function (Eq. [5 5]) for normal diffusion. This is expected given the channel diameters. The small, narrow channel possesses a diameter of 10 which allows for two xenon atoms to pass one another inside of the channel. This is even more the case for the larger diameter channel. The longitudinal relaxation times extracted from these fits for the small and large channels were 5.4s and 3.6s, respectively.

PAGE 55

55 Figure 6 3 Tracer exchange fits to the small, 1nm channels of FF Figure 6 4 Tracer exchange fits to the large, central channels of FF

PAGE 56

56 T 1 Estimation In an attempt to verify the T 1 values that were extracted from the tracer exchange fitting function s, thermally polarized 129 Xe saturation recovery experiments were performed on FF on a Varian Inova 500MHz spectrometer at a similar pressure to the tracer exchange experiments. Similar experiments were performed on a xenon gas reference sample consisting of 10 bar pure xenon gas and 1 bar oxygen. Also performed on this reference sample were inversion recovery experiments to verify the T 1 values obtained from the saturation recovery experiments. The FF sample was cooled to 85 C where xenon adsorption i nto both channels was observed. This made it difficult to verify the T 1 values extracted from the tracer exchange fitting functions, as the experiments were performed at two different temperatures. This difference in temperatures was not intentional and was realized only after a calibration of the variable temperature apparatus used for the hyperpolarized experiments was performed, as this was done after all data had been collected. Also of concern here however, is the fact that during the hyperpolariz ed experiments, warm gas is flowing through the sample and sample holder while the thermocouple measuring the temperature of the sample is outside of the sample holder barely extending out through the probe head. This investigation of the effect that the circulation of warm ga s has on the temperature is a continuing investigation. It is believed that the temperatures reported in the hyperpolarized xenon variable temperature and tracer exchange experiments may actually be higher due to the warming effects of the gas flow. Once these things are resolved and it is know with certainty what the temperature of the sample is when hyperpolarized experime nts are

PAGE 57

57 conducted, the T 1 relaxation values can be compared to the values obtained from the thermally polarized xenon inversion and satu ration recovery experiments. Because the thermally polarized experiments were carried out on sealed samples using a different spectrometer which had a different variable temperature setup that was calibrated to the splitting of me thanol t he accu racy of the temperatures for tho se experiments is unquestionable. As just mentioned T 1 saturation and inversion recovery experiments were performed on a xenon reference sample. This was done to ensure continuity between the two different methods for measuring T 1 The inversion recovery and saturation recovery experiments for the reference sample at room temp erature are shown in F igures 6 5 and 6 6, respectively. The same experiments at 85 C are shown in F igures 6 7 and 6 8, respectively. The absolute integrals of the signal were tabulated, normalized to the value corresponding to the maximum integral/signal, and plotted with respect to the recovery time As can be seen in Figures 6 5 6 8 t he two different methods for measuring the longitudinal relaxation give roughly the same values that are within the experimental error s of each other. Being that the FF experiments needed to be performed at temperatures well below room temperature, simila r experiments were performed on the xenon gas reference sample at 85 C. This was done simply to ensure consistency between the results and also to reference the bulk xenon gas peak to 0ppm at that specific temperature.

PAGE 58

58 Figure 6 5 Inversion recovery experiments on Xe gas reference sample at 25 C Figure 6 6 Satur ation recovery experiments on Xe gas reference sample at 25 C

PAGE 59

59 Figure 6 7 Inversion recovery experiments on Xe gas reference sample at 85 C Figure 6 8 Saturation recovery experiments on Xe gas reference sample at 85 C

PAGE 60

60 A s noticed in Figures 6 5 6 8 T 1 decreases with decreasing temperature. In 1988, a report was published studying the nuclear spin relaxation by intermolecular magnetic dipole coupling in the gas phase for mixtures of 129 Xe in oxygen. 4 2 This study came to the conclusion that T 1 relax ation rates are dependent on temperature, however, it was concluded that the specific temperature dependence could not be very precisely determined by the experiments. The temperature dependence from th is study is shown in F igure 6 9 As can be seen, as the temperature is decreased for a constant density of oxygen, 1/T 1 increases meaning that T 1 decreases, which is a similar trend seen in the above experimental results at 25 and 85 C. Figure 6 9 Density and temperature dependence of 129 Xe spin re laxation times [ Image reproduced with permission from Jameson, J.C.; Jameson, A.K. ; Hwang, J.K. Journal of Chemical Physics 1988, 89 (Page 4075, Figure 3). American Institute of Physics.]

PAGE 61

61 I n another study conducted in 1995, the longitudinal relaxation time of the 129 Xe nucleus of natural xenon gas dissolved in various liquids was s tudied as a function of temperature. 4 3 This study came to the conclusion that the T 1 relaxation increases with increasing temperature, which corresponds to the same trend that is seen in the xenon reference sample of F igures 6 5 6 8 This publication pr oposed that the relaxation mechanisms of the 129 Xe nucleus in gases and solutions are exclusively due to intermolecular interactions. In the gas phase, relaxation comes from the spin rotation coupling during atomic collisions or during the transient exist ence of diatomic molecules. Other possible relaxation mechanisms in xenon include diffusion through magnetic field gradients caused by a host, wall interactions, and relaxation with other xenon atoms. The saturation recovery experiments performed on FF a t 85 C and 95 C for the small and large channels are shown in F igures 6 10 6 13 Figure 6 10 Saturation recovery experiments on the small channels of FF nanotubes at 85 C

PAGE 62

62 Figure 6 11 Saturation recovery experiments on the small channels of FF nanotubes at 95 C Figure 6 12 Saturation recovery experiments on the large channels of FF nanotubes at 85 C

PAGE 63

63 Figure 6 13 Saturation recovery experiments on the large channels of FF nanotubes at 95 C Table 6 1. Summary t able of T 1 r elaxation t imes Sample Method Temperature ( C) Channel T 1 (seconds) FF Sat. Rec. 85 Small 18.61.2 Large 43.31. 5 9 5 Small 16.41.0 Large 41. 91. 9 Tracer Exchange 93 (questionable) Small 5.4 0.3 Large 3.6 0.4 Xe Reference Sat. Rec. 25 6.3 0. 5 85 3.60.2 Inv. Rec. 25 5.80.3 85 3.2 0. 4 As was the case with the xenon gas reference sample, the channels in the FF nanotubes also show a T 1 temperature dependence where the relaxation time decrea ses with decreasing temperature. It should be noted, however, that relaxation

PAGE 64

64 mechanisms in the bulk and adsorbed phases may be different. The decrease in T 1 observed here is substantially less when compared to the reference sample considering it is only over a 10 C range as opposed to the 110 C range that was used for the reference. One noticeable difference here is the trend in the T 1 values between the FF saturation recovery experiments and the FF tracer experiments. In the saturation recovery experiments, fo r the same temperature, the T 1 of the large channel is greater than that of the small channel. However, in the tracer exchange experiments, the T 1 of the large channel is smaller than that of the small channel. This investigation is ongoing and no clear explanation for this discrepancy currently exists. Bis Urea Macrocycle Xenon partial pressure dependence The first experiments performed on the bis urea macrocycle constitu t ed a xenon partial pressure dependence study, which is shown in F igure 6 14 At l ow pressure/loading, the chemical shift powder pattern shows a very large anisotropy and is opposite in sign to that observed in other single f ile materials at low loading 44 ,4 5 At higher pressures, the shielding anisotropy decreases and appears to revers e sign near the highest pressure studied due to Xe Xe interactions The chemical shift anisotropy of the xenon atom should be observed as an average over the volume sampled due to rapid movement of the xenon atom. However, if xenon were static at some s ite on the pore wall, a large anisotropy may be observed. In 1995, the anisotropic chemical shift of 129 Xe in the aluminophosphate mol ecular sieve ALPO 11, shown in F igure 6 15, was studied. 46 The dependence of the chemical shift trapped in the one dimens ional pores was measured as a function of xenon loading.

PAGE 65

65 Figure 6 14. Dependence of the 129 Xe line shape in the one dimensional pores of bis urea crystals on Xe partial pressure in a He/Xe mix ture with a total pressure of 31 15 Torr

PAGE 66

66 The anisotropy was found to be axially symmetric and positive at low loading and then it appeared to lose its axial symmetry with increased loading. However, at even higher loading, an axial line shape again was observed but the sign was reversed from the previous time with nonaxial line shapes at intermediate and high loadings. T wo principal components of the observed shielding tensor were found to vary linearly with Xe loading while a third remained invariant. This behavior of the line shape was analyzed in terms of a st atistical distribution of three types of xenon sites. Each of these sites had 0, 1, or 2 neighboring sites occupied by other Xe atoms shown in F igure 6 16 Each site also possessed its own characteristic shielding tensor, and fast exchange of the three site types at room temperature was assumed. 45 The experimental results of thi s study are shown in F igure 6 17 and provide a decent comparison to the results obtained in the bis urea macrocycle. Figure 6 15 Structure and channel diameter of ALPO 11

PAGE 67

67 Figure 6 1 6 Structure of one channel in ALPO 11 indicating the three types of xenon sites [ Image reproduced with permission from Ripmeester, J.A.; Ratcliffe, C.I. Journal of Physical Chemistry 1995 99 (Page 621, Figure 3) Americ an Chemical Society. ]

PAGE 68

68 Figure 6 17 129 Xe NMR line shapes for xenon adsorbed in ALPO 11. Loading levels: A=1.58, B=2.70, C=3.80, D=5.16, E=5.83, F=8.05, G=10.30, H=11.10, I=13.40, J=16.40 mol of Xe (x10 4 )/g of dry ALPO 11. [ Image reproduced with permission from Ripmeester, J.A.; Ratcliffe, C.I. Journal of Physical Chemistry 1995 99 (Page 62 0 Figure 1 ) Americ an Chemical Society. ] In 1989, a magic angle spinning NMR study on the silicoaluminophosphate molecular sieve SAPO 11, which is structurally similar to ALPO 11, concluded that the

PAGE 69

69 xenon static line shapes arose from a single anisotropic chemical shift interaction. 45 The variations in this anisotropy were attributed to changes /deformations in the shape of the electron cloud around the xenon atom as more and more atoms were packed into the elliptical nanochannel. Thi s could potentially be the case with the bis urea macrocycle, as deformations in the electron cloud seem even more likely given the smaller elliptically shaped nanochannels. Another possible explanation for these lineshapes is that the three components of the chemical shift shielding tensor may be different. In the case of a cylindrically symmetric nanochannel, two components of the tensor are identical with the third being different. However, being that the bis urea channels are elliptical and not cylin drically symmetric, it is possible and plausible that each of these three components is different, and an averaging of these three tensors could account for the observed lineshapes. Tracer Exchange T racer exchange experiments were performed on the bis urea sample to probe xenon dynamics upon the adsorption process The data for this collected at room temperature, is shown in F igure 6 1 8 This data was fit to the tracer exchange functions mentioned in C hapter 5 and the fits to normal and single file diff usion are shown in F igure 6 19 As seen in F igure 6 19, the bis urea data visually seems to fit the tracer exchange function (Eq. [5 4]) for single file diffusion better than for normal diffusion. However, the statistics and error between the two fits ar e too close to determine which one is actually better. Single file diffusion is expected given the channel diameter of the sample as 3.7x4.8 and the collision diameter of xenon as 4.4. Within this narrow channel, two

PAGE 70

70 xenon atoms would not be predicted to pass one another based on simple geometry. T 1 measurements through thermally polarized xenon saturation recovery experiments at room temperature could help distinguish which fit is more accurate but the amount of sample available was insufficient for these experiments. This remains an ongoing investigation. Figure 6 18 Tracer exchange data on bis urea macrocycle Figure 6 19 Tracer exchange fits for the bis urea macrocycle

PAGE 71

71 CHAPTER 7 CONCLUSION AND RECOMMENDATIONS The FF tracer exchange stu dies conducted clearly demonstrated the observance of normal diffusion within the two different sized nanochannels. An attempt was made to verify the T 1 relaxation times extracted from these fits utilizing thermally polarized xenon saturation and inversio n recovery experiments at a similar temperature and pressure. It was found that at 75 C, no xenon adsorption into the smaller channel of FF was observed using thermally polarized xenon which opposed the results of the hyperpolarized tracer exchange experiments. However, adsorption into the smaller channel was observed at 85 C. A possibl e reason for this temperature discrepancy is the heating of the sample cause by the gas stream during the continuous flow hyperpolarized 129 Xe experiments. This can be rectified through a temperature calibration. 47 In studying the xenon adsorption at vari ous temperatures, it was found that the chemical shift decreases with increasing temperature which is in agreement with similar results found throughout the literature. It was also determined that at decreasing temperatures, xenon migrates from the large, central channel to the small, narrow channel of the FF nanotubes. One possible explanation for this is that at reduced temperatures, condensation occurs in the smaller channels and a liquid phase is observed, as this would be thermodynamically favored to occur in the smaller channels a s opposed to the larger channels For the xenon gas reference sample and the FF nanotubes, the T 1 relaxation decreased with decreasing temperature, albeit the two systems may have different

PAGE 72

72 relaxation mechanisms. This was also in line with results found throughout the literature. The bis urea macrocycle pressure study demonstrated that the anisotropy was axially symmetric and positive at low loadin g and then lost its axial symmetry with increased loading. Ab initio calc ulations of 129 Xe shielding tensors could possibly account for the chemical shift powder patterns and the Xe loading dependence. The tracer exchange experiments for the bis urea were not conclusive as to which type of diffusion is occurring in the channel However, single file diffusion is expected given the channel geometry. A larger bis urea macrocycle possessing a channel diameter around 10 has recently been synthesized and is awaiting experimentation. It is expected that this sample will exhibit no rmal diff usion given its channel size. With this sample, v erification of T 1 values through thermally polarized xenon experiments w ould prove beneficial

PAGE 73

73 LIST OF REFERENCES (1) Karger, J.; Ruthven, D.M. Diffusion in Zeolites and Other Microporous Solids ; W iley, 1992. (2) Goodson, B.M. Journal of Magnetic Resonance 2002 155 157. (3) Springuel Huet, M.A.; Fraissard, J.P. Zeolites 1992 12 841. (4) Raftery, D.; Long, H.; Meersmann, T.; Grandinetti, P.J.; Reven, L.; Pines, A. Physical Review Letters 1991 66 584. (5) J ameson, C.J.; de Dios, A.C. Journal of Chemical Physics 2002 116 3805. (6) Brunner, E. Protein Science 2005 14 847. (7) Oros, A.M.; Shah, J. Physics in Medicine and Biology 2004 49 R105. (8) Zook, A.L.; Adhyaru, B.B.; Bowers, C.R. Journal of Magnetic Resonanc e 2002 159 175. (9) Levitt, M.H. Spin Dynamics: Basics of Nuclear Magnetic Resonance ; John Wiley & Sons 2008 10) Ito, T.; Fraissard, J. Journal of Chemical Physics 1982 76 5225. 11) Cheng, C.Y. PhD Dissertation, University of Florida, 2008 12) Appel t, S.; Baranga, B.A.; Erickson, C.J.; Romalis, M.V.; Young, A.R.; Happer, W. Physical Review A 1998 58 1412. 13) Kastler, A. Journal of the Optical Society of America 1957 47 460. 14 ) Whiting, N.; Eschmann, N.A.; Goodson, B.M. Physical Review A 2011 83 053428. 15 ) Walker, T.G.; Happer, W. Reviews of Modern Physics 1997 69 629. 16) Raftery, D.; Long, H.; Meersmann, T.; Grandinetti, P.J.; Reven, L.; Pines, A. Physical Review Letters 1991 66 584. 17) Zook, A.L.; Adhyaru, B.B.; Bowers, C.R. Journ al of Magnetic Resonance 2002 159 175. 18) Rosen, M.S.; Chupp, T.E.; Coulter, K.P.; Welsh, R.C.; Swanson, S.D. Review of Scientific Instruments 1999 70 1546.

PAGE 74

74 19) Ruest, I.C.; Ketel, S.; Hersman, F.W. Physical Review Letters 2006 96 053002. 20) Gor bitz, C.H. Chemical Communications 2006 2332. 21) Reches, M.; Gazit, E. Science 2003 300 625. 22) Kol, N.; Adler Abramovich, L.; Barlam, D.; Shneck, R.Z.; Gazit, E.; Rousso, I. Nano Letters 2005 5 1346. 23) Song, Y.; Challa, S.R.; Medforth, C.J.; Qiu, Y.; Watt, R.K.; Pena, D.; Miller, J.E.; Swol, F.; Shelnutt, J.A. Chemical Communications 2004 1044. 24) Adler Abramovich, L.; Reches, M.; Sedman, V.L.; Allen, S.; Tendler, S.J.; Gazit, E. Langmuir 2006 22 1313. 25) Reches, M.; Gazit, E. Nano Lett ers 2004 4 581. 26) Makin, O.S.; Atkins, E.; Sikorski, P.; Johansson, J.; Serpell, L.C. Proceedings of the National Academy of Sciences 2005 102 315. 27) Gorbitz, C.H. European Journal of Chemistr y 2001 7 5153. 28) Adler Abramovich, L.; Aronov, D. ; Beker, P.; Yevnin, M.; Stemper, S.; Buzhansky, L.; Rosenman, G.; Gazit, E. Nature Nanotechnology 2009 4 849. 29) Ryu, J.; Park, C.B. Chemistry of Materials 2008 20 4284. 30) Reches, M.; Gazit, E. Nature Nanotechnology 2006 1 195. 31) Shimizu, L .S.; Smith, M.D.; Hughes, A.D.; Shimizu, K.D. Chemical Communications 2001 1592. 32) Dewal, M.B.; Lufaso, M.W.; Hughes, A.D.; Samuel, S.A.; Pellechia, P.; Shimizu, L.S. Chemisty of Materials 2006 18 4855. 33) Hahn, E.L. Physical Review 1950 80 580. 34) Carr, H.Y.; Purcell, E.M. Physical Review 1954 94 630. 35) Aguilar, J.A.; Nilsson, M.; Bodenhausen, G.; Morris, G.A. Chemical Communications 2012 48 811. 36) Meersmann, T.; Logan, J.W.; Simonutti, R.; Caldarelli, S.; Comotti, A.; Sozzani, P.; Ka iser, L.G.; Pines, A. Journal of Physical Chemistry A 2000 104 11665.

PAGE 75

75 37) Bowers, C.; Cheng, C. Chem. Phys. Chem. 2007 8 2077. 38) Vasenknov, S.; Karger, J. Physical Review E 2002 66 052601. 39) Jeener, J.; Meier, B.H.; Bachmann, P.; Ernst, R.R. Jo urnal of Chemical Physics 1979 71 4546. 40) Blumich, B. NMR Imaging of Materials ; Oxford University Press, 2000. 41) Cheung, T.T.P. Journal of Physical Chemistry 1995 99 7089. 42 ) Jameson, J.C.; Jameson, A.K.; Hwang, J.K. Journal of Chemical Physics 1988 89 4074. 43 ) Oikarinen, K.; Jokisaari, J. Applied Magnetic Resonance 1995 8 587. 44 ) Jameson, C.J.; de Dios, A.C. Journal of Chemical Physics 2002 116 3805. 45 ) Springuel Huet, M.A.; Fraissard, J. Journal of Chemical Physics Letters 1989 15 4 299. 46 ) Ripmeester, J.A.; Ratcliffe, C.I. Journal of Physical Chemistry 1995 99 619. 47) Kneller, J.M.; Soto, R.J.; Surber, S.E.; Colomer, J.F. ; Fonseca, A.; Nagy, J.B.; Pietra T. Journal of Magnetic Resonance 2000 147 261.

PAGE 76

76 BIOGRAPHICAL SKETCH Christopher Akel was born in Jacksonville, Florida in 1987. He received his B.S. degree in chemistry with a minor in physics from Barry University located in Miami Shores, Florida, of Florida during the summer of 2009 to pursue his M.S. degree in physical chemistry. Since then, he has worked full time as a graduate student and teaching assistant. Since entering graduate school, he has been involved in several projects including solid state deuterium NMR studies of polymers, 23 Na magic angle spinning NMR of polycrystalline Na 24 Si 136 rubidium optical pumping at high magnetic fields, hyperpolarized 129 Xe NMR in nanotubular materials, and numerous crystallization projects. His current research interests include solid state NMR and hyperpolarized noble gas NMR. After graduation in 2012, he plans to enter the work force in the chemical industry business.