Undoubtlely the largest scientific achievemnt of this projct is the development and implementation of a new protocol for Fast 14N NQR Resonance Scanning. The 14N NQR resonances are typically narrow and sparsed on a large frequency interval. At each measurement only a fraction of the spectrum is inspected, so that many steps are required to inspect the whole frequency region. On top of that, the 14N NQR signal is weak and requires extensive averaging. But since the relaxation parameters are not known, averaging is suboptimal. Because of these, most of the scans result in no resonance, even if this is being inspected. Our method overcomes the nedd to know the relaxation parametrs. In essence it reverses signal averaging and frequency scanning. This is being possible if the hardware is fast enough to switch from one frequency to the other. We have managed to built such a system. The maximum time to find the resonance with our protocol is limited only by the amount of sample. Typical times are few hours, whereas in favourable circumstances this can be also only few tens of minitues.
During the course of this project, we have relised that from the applicational point of view of our spectrometer, it is very important to precisely know the relationship between the NQR signal and the amount of sample. In this article we present both theorethical and experimental results for three cases where different pulse sequences are used. The relationship is different in all three cases in addition, in some cases the relationship depends also on the shape of the resonance, which is very unwanted, but one should deinitely be aware of.
We have presented a new pulse sequence, the WURST-QCPMG, for the acquisition of 14N NQR spectra. This sequence combines the advatages of WURST pulses and the "spin-lock" effect. WURST pulses are composite pulses, which allow a very broadband excitation of the spectrum. This is very wellcome when a broad spectrum is being composed of narrower, patial spectra (as in NQR) as it allows to significantly reduce the number of required steps. In addition, WURST pulses also allow to reduce the required RF power, and therefore also to reduce the size of equipement, especially the RF amplifier. The "spin-lock" effect is a phenomenom which slows down the decay of signal during multipulse sequences in solids. Because of this, the signal can be averaged many more times in a given amount of time as compared to when this effect is not present. These averages allow to significantly increase the S/N ratio, which for 14N NQR can be 10 times and even more. Because the 14N NQR signal is always small, the "spin-lock" effect is almost always necessary to obtain acceptable S/N ratios. Before owr work, it was not clear wheter the WURST pulse are compatible with "spin-lock". This pulse sequence is being routinely used in our spectrometer.
Characterization of polymorphizem is one important area where our spectrometer finds applicational use. In this article, we have done an 14N NQR research of two known polymorphs of famotidine. At room temperature, seven quadrupolar sets of transition frequencies v+, v-, and v0 corresponding to seven different nitrogen sites in the crystal structure of each of the two polymorphs were found. This confirms the expected ability of NQR to distinguish polymorph B from its analog A. NQR can also measure their ratio in a solid mixture and in the final dosage form, that is, a tablet. The NQR frequencies, line shapes, and tentative assignation to all seven molecular 14N atoms were obtained. Unravelment of these two entangled NQR spectra presents a valuable contribution to the NQR database and enables studies of some possible correlations therein. Moreover, nondestructive 14N NQR studies of commercial famotidine tablets can reveal some details of the drug fabrication process connected with compression.