The small thermal polarization of nuclear spins currently limits the capabilities of nuclear spin based technologies such as nuclear magnetic resonance spectroscopy (NMR) and magnetic resonance imaging (MRI). Existing techniques for polarizing nuclear spins beyond their thermal equilibrium, called dynamic nuclear polarization (DNP), utilize cryogenic temperatures and expensive microwave technologies to transfer the larger thermal polarization of electron spins to targeted nuclei. Ubiquitous access to polarized nuclei through a room temperature, microwave-free alternative to DNP would revolutionize the capabilities of NMR and MRI.

Optically pumping nitrogen-vacancy (NV) defects in diamond can generate room temperature, microwave-free 13C nuclear polarization at the high magnetic fields used in NMR (7.05T, 9.4T). The mechanism of NV center electronic polarization is well understood, and 13C polarization has been observed, but the mechanism for polarization transfer from NV to 13C remains unknown. Here we present NMR and EPR results characterizing the polarization dependence of 13C and NV in diamond, as well as a quantum mechanical model describing a possible polarization transfer mechanism.

The sign and magnitude of the 13C polarization sensitively depends on the orientation of the diamond with respect to the directions of the applied magnetic field and laser polarization. The polarization magnitude further depends on the defect concentrations, magnitude of the applied magnetic field, temperature, and the illumination conditions: wavelength, power, and exposure time.

To better understand the source of polarization, the NV defects were characterized with EPR to determine relaxation times, concentrations, and homogeneity. EPR was also used to determine the orientation dependence of NV polarization. The NV polarization is constant in the defect frame, which, when rotated into the laboratory frame, results in highest polarization when aligned with the field, zero polarization at 54 degrees, and inverted polarization at higher angles. These EPR insights into the NV physics were incorporated into models for 13C polarization mechanisms.

Dipolar coupled pairs of NV centers are proposed as the source for 13C polarization in NV- diamonds at high magnetic fields. Our model shows these dipolar-coupled manifolds have transitions matching the frequency of the 13C nuclei, making them a feasible source of spontaneous polarization transfer. The model also qualitatively captures the observed polarization sign changes as a function of crystal orientation.

This thesis applies magnetic resonance techniques to novel and traditional proton conducting

materials in order to gain a better understanding of the molecular aspects of their

performance. An array of experiments and techniques are used. One material is a morphologically

designed block copolymer/ionic liquid system. Simple 1-D variable temperature

(VT) 1H magic angle spinning (MAS) NMR is used to catalogue the dynamic chemical

shifts, which relates to the prevalence of hydrogen bonding. Relaxation data are used

to measure the relative mobilities of conduction protons, and these data are related to

polymer physics phenomena. The effect of morphology is investigated by comparing block

copolymer data to a series of homopolymer analogues with no morphological structure.

The role of entropy in these systems is discussed as well as the effect of a non-symmetric

ionic liquids.

NMR techniques were also applied to more traditional materials, namely perfluorosulfonic

acid membranes modified with amphoteric imidazole compounds. The chemical

environments of the imidazole as well as the dynamics of proton transfer are measured

with Solid-state 1H NMR. The effect of imidazole concentration is also considered. 1H

-13C cross polarization (CP) MAS NMR is used to reveal the presence of both fast and

slow moving imidazole in the membranes. Pulsed field gradient (PFG) NMR is used to

quantify the diffusion of protons and methanol through the material.

The goal of replacing gaseous hydrogen with organic virtual carrier molecules as a

proton source for proton exchange membranes (PEM) fuel cells is investigate by testing

the effect of model organic liquids in contact with typical membrane materials. A suite

of data, including 19F NMR relaxation data of the polymer backbone and sidechains, are

used to explain bulk phenomena and membrane performance.

Nuclear spins are harnessed in many important technologies, including the well established fields of magnetic resonance imaging for medical diagnostics, magnetic resonance spectroscopy in analytical chemistry as well as emerging technologies in quantum information and spintronics. All of these technologies either harness, or are subject to, the behavior of a nuclear spin ensemble. To achieve the most desirable behavior, (large spectroscopic signal or reduction of unwanted fluctuations) the nuclear spin ensemble should be prepared in a pure quantum state. In practice, this ``polarization" is typically created by allowing the energy levels of the spins in an applied magnetic field achieve thermal equilibrium. Unfortunately, even with the largest magnets available with fields greater than 20 Tesla, the separation between energy levels is much smaller than kT for all but extreme refrigerated systems. It is then desirable to achieve pure nuclear spin states which are not at thermal equilibrium with the environment. In order to do this, it is necessary to create a situation in which a pure quantum state can be created in a system other than the nuclear spin which then interacts with the nucleus to create a more pure nuclear spin state. In this work we harness the pure photon spin state of circularly polarized light as well as spin transition selection rules of a deep electronic defect in diamond to polarize nuclei.

In the first case, we use circularly polarized photons to excite spin polarized electrons in the semiconductor gallium arsenide which equilibrate with bound electronic states at recombination centers. These bound states then polarize nearby nuclear spins through the magnetic hyperfine interaction. While this hyperfine mechanism of nuclear spin polarization was previously known, we have identified a new regime of low optical absorption where the coupling of nuclear quadrupole moments to electric field gradients near recombination centers is the dominant mechanism of nuclear spin polarization. Through a combination of experiment and theory, we determine relative rates of these two mechanisms depending on the rate of optical absorption. Since optical absorption varies as a function of depth in a sample, we predicted that control of these two mechanisms is possible as a function of position in the sample. Using the stray field of a superconducting magnet to supply the gradient field for magnetic resonance imaging, we were able to directly observe patterns of nuclear magnetization on a micron length scale. When combined with in-plane control of the laser and NMR pulse sequences, this technique will give rise to fully 3-dimensional patterns of nuclear magnetization. These patterns may be created in bulk gallium arsenide without the need for lithography or other microfabrication techniques. These regions of magnetized nuclei will enable magnetic control over drifting electrons in future spintronics devices.

The paramagnetic nitrogen-vacancy defect in diamond provides a different tool to control nuclei. The ground state spin triplet of this defect may be easily polarized into the S_z=0 state with visible optical illumination. The polarization is due to the symmetry and selection rules within the defect itself and does not require polarized photons. We discovered that, with a sufficient density of defects, the ¹³C nuclei in the diamond lattice are spontaneously polarized upon illumination of the sample. We attribute this polarization to a highly refrigerated ``spin temperature" among the energy levels created by the magnetic dipole interaction of the many spins in the defect ensemble. This energy reservoir is in thermal contact with the <sup>13</sup>C nuclei, which are driven to highly athermal spin states. We theoretically investigate the thermodynamics of the defect spin ensemble, first with a two-spin ``toy model" and more recently have begun a many-spin theoretical approach. The polarization of nuclei in diamond has application in the quenching of nuclear fluctuations in quantum information systems and as a platform for signal enhancement in magnetic resonance imaging and spectroscopy.

This thesis develops magnetic resonance techniques to characterize the structure and proton dynamics of rare earth phosphates. Structural characterization is accomplished primarily through the use of ³¹P magic angle spinning (MAS) nuclear magnetic resonance (NMR). Spectroscopic characterization is used to develop chemical shift references on calcium and lanthanum phosphate materials in order to better understand the phases and phosphate environments in a calcium lanthanum phosphate glass-ceramic material. Similar methods are used to characterize cerium orthopohsphate, a material that has often been called "NMR invisible" in the literature, and the phosphate structures that are formed when two differing synthetic approaches are performed.

Structural characterization is enhanced through the use of double resonance techniques, such as ¹H -³¹P cross polarization (CP) MAS NMR and heteronuclear correlation (HETCOR) spectroscopy. Such techniques are a powerful way to associate protons present in the material with phosphate hosting sites in the material. In a strontium cerium metaphosphate glass-ceramic, ³¹P MAS NMR spectroscopic, relaxation, and variable temperature techniques are used to characterize and identify the structure of this partially crystalline material, while ¹H -³¹P cross polarization (CP) MAS NMR allows the preferred proton hosting site in the material to be identified.

In order to measure proton dynamics in the rare earth phosphates, field gradient NMR

techniques are adapted to the particular challenges of rare earth phosphates. Relatively low proton conductivities in the materials being studied suggest that proton motion in these materials will be slow. The combination of slow proton motion, the known short ¹H relaxation times of rare earth phosphates, and the required high temperatures at which proton diffusion takes place rule out standard field gradient techniques that have been developed for liquids. Small scale pulsed field gradient NMR probe designs and stray field steady gradient NMR have been investigated, and methods by which one can make such measurements more straightforward are discussed. It is demonstrated that high temperature stray field steady gradient NMR methods can measure proton self diffusion in a relatively proton-rich barium lanthanum metaphosphate glass.