From crystalline to amorphous state: the role of the glass in the Dynamic Nuclear Polarization process.

The main purpose of this thesis work is the investigation of the role of the glassy state on the Dynamic Nuclear Polarization (DNP) process, paying particular attention to the application in the diagnostic field (Magnetic Resonance Imaging, MRI). So far, experimental evidences have shown that the crucial requests for a sample containing paramagnetic impurities to be polarized by DNP are its amorphous state and the homogeneity of its glassy mixture and dispersion. On the other hand, a consistent theoretical interpretation for this phenomenon is missing, as well as a deep analysis on the †œgoodness†� of the glassy state obtained after the amorphization process, intended for DNP-MRI application. An alternative preparation procedure of contrast agents containing radical molecules, hyperpolarizable by DNP, is proposed in this thesis. The novelty of these glassy samples is that they are solid at room temperature. Under these circumstances, a methodology of characterization of the amorphous solids in all the fundamental aspects (thermal, spectroscopic, structural and magnetic properties) is suggested to investigate the correlation between the glassy state and the hyperpolarization features (such as the maximum achievable value of polarization, P%). In particular, in this way it is possible to study one of the main problems in this topic, that is the effect of the presence of nano- and micro-crystalline domains on the homogeneity of the radical distribution and, then, on the efficiency of the magnetic polarization transfer. Furthermore, in order to optimize the DNP efficiency, another crucial issue is the role of the radical concentration on the polarization transfer and whether high concentration could lead to either quenching effect or to radical aggregation. For this purpose, several amorphization procedures of solids have been analyzed. This study shows that co-milling is the best procedure, that provides riproducibility, prevents degradation and allows a good control of the physical features of the glass and of the crystalline phase. A milled mixture of trehalose and TEMPO molecules has been chosen as model system, because of the high stability of trehalose and high solubility of the TEMPO radical. Chapters 1 and 2 briefly report the state of art regarding both the glassy state and the preparation of amorphous samples, addressing to the issue relevant for the hyperpolarization by DNP in chapter 2. Chapter 3 presents a discussion about the choice of the optimal combination of amorphization technique and model system. The characterization of the model system perfomed by Differential Scanning Calorimetry (DSC), Electron Paramagnetic Resonance (EPR), Solid State Nuclear Magnetic Resonance (SSNMR) and Raman spectroscopies and X-Ray Diffraction (XRD) is described in chapter 4. The effects of both the concentration and the amorphization degree on the physical properties of the samples have been highlighted. Chapter 5 reports results of DNP measurements on the model system. The effect of radical concentration on the polarization transfer has been stressed for fully amorphized samples (12 h of milling), paying attention to the physical stability of these amorphous solids. In addiction, some alternative substrates used in DNP-MRI have been tested for comparison. In the final part of this work, chapter 6 describes an ancillary study on the dehydration of solutions, carried out by means of a novel calorimetric approach to investigate the role of water (possibly absorbed from the environment) on the stability of the amorphous solids. Further investigation in this direction is needed. For example, direct comparison between EPR spectra and DNP enhancement, as well as a deep analysis of the TEMPO-trehalose and TEMPO-TEMPO interactions by mean of vibrational spectroscopy could allow to investigate aggregation processes in high concentration samples. Moreover, further studies of dynamic properties by mean SSNMR would allow to separate the behaviour of the crystalline and amorphous phases and to follow separately the two processes, amorphization and co-mixing, that simultaneously occur during the milling. Furthermore, FT analysis of the obtained XRD patterns will provide information on the spatial distribution of the molecules of trehalose (and TEMPO) by deriving the pair distribution function, g(r). Finally, the methodology of investigation above described opens a new avenue to characterize the effect of the matrix density of aged glasses or of controlled dispersion of nanocrystals into the amorphous matrix on the DNP performance, from both experimental and theoretical approches. A more systematic analysis of the DNP measurements could be carried out, in order to have more information on the correlation between the presence of crystals and the efficiency of the magnetic transfer: in particular, the maximum dimension of the crystalline domains and the maximum achievable polarization are the parameters that should be correlated.