Novel Methods for High Resolution NMR in Inhomogeneous Fields

在本文的工作中，我们计划在非均匀静态和射频场中开发高分辨率核磁共振波谱的新方法。高分辨率核磁共振实验需要非常均匀的磁场才能获得原子核的化学位移和J耦合信息，因为这些相互作用比主要的外场B0的塞曼耦合小几个数量级。实现这种同质性并非易事，尤其是在相对较大的容量上。这通常是以核磁共振系统的价格和核磁共振磁体所需的尺寸为代价的。当处理空间异质性组织或采用远程核磁共振排列时，这种均质性是无法实现的。在这里，我们提出了一种基于时变三维梯度和空间非选择性射频脉冲的非常通用的方法，用于学习非均匀磁场的分布，然后校正由于这些非均匀性而由自旋获得的相位。通过这种校正，可以提取原子核的化学位移信息，获得高分辨率的光谱。这些方法有可能取代廉价磁铁中的专用垫片线圈，用于传统核磁共振或成像实验的专门应用。所提出的方法还将推动移动和非原位核磁共振的方法，并将其应用于材料科学，化学工程和地球科学，以及过程，产品和质量控制。作为所提出工作的第二个主要组成部分，我们计划开发基于最优控制和傅立叶合成方法的新型脉冲序列，这些脉冲序列对自旋的拉莫尔色散和rf不均匀性具有鲁棒性，并且比用于相同目的的传统设计短得多。随着高达21特斯拉的高场核磁共振波谱技术逐渐成为常规技术，新出现的应用给脉冲设计带来了新的挑战。这些应用包括设计宽带激励或反转脉冲、波段选择激励、在带宽与可用射频功率相当或大得多的情况下解耦脉冲序列。在这些情况下，自然的选择是使用绝热脉冲序列，这种脉冲序列往往很长，在存在大的弛豫率时是不可取的。这就需要开发比传统方法短得多的脉冲序列设计，并提供相当的补偿性能。在这里，我们提出了研究这种设计的方法。本工作还将基于这些新方法开发蛋白质核磁共振光谱的灵敏度增强实验。

In the proposed work, we plan to develop novel methods for high resolution NMR spectroscopy in inhomogeneous static and radio-frequency fields. High resolution NMR experiments require extremely homogeneous magnetic fields to obtain chemical shift and J coupling information of nuclei, as these interactions are orders of magnitude smaller than the main Zeeman coupling to the external field B0. Achieving such homogeneity is not trivial, especially over relatively large volumes. This usually comes at the cost both in price of NMR systems and in size demanded by NMR magnets. Such homogeneities are not achievable when dealing with spatially heterogeneous tissues or employing remote NMR arrangements. Here we propose to develop very general methods based on time varying 3D gradients and spatially non-selective radio-frequency pulses for learning the distribution of inhomogeneous magnetic fields and then correcting for the phase acquired by the spins due to these inhomogeneities. With this correction, the chemical shift information of the nuclei can be extracted and high resolution spectra can be obtained. These methods can potentially replace dedicated shim coils in inexpensive magnets for specialized applications of traditional NMR or imaging experiments. The proposed methods will also advance the methodology of mobile and ex-situ NMR with applications to materials science, chemical engineering and geosciences, and in process, product and quality control. As a second major component of the proposed work, we plan to develop novel pulse sequences based on methods of optimal control and Fourier synthesis which are robust to Larmor dispersion of spins and rf-inhomogeneities and are significantly shorter than conventional designs used for the same purpose. As NMR spectroscopy at high fields up to 21 Tesla is becoming routine, there are emerging applications that pose new pulse design challenges. These applications include design of broadband excitation or inversion pulses, band selective excitations, decoupling pulse sequences in situations when the bandwidth to be covered is comparable or much larger than the rf power available. A natural choice in these settings is the use of adiabatic pulse sequences which tend to be very long and are not desirable in the presence of large relaxation rates. This makes it desirable to develop pulse sequence designs that are significantly shorter than the conventional methods and offer comparable compensation performance. Here we propose methodologies for the study of such designs. The proposed work will also develop sensitivity enhanced experiments in protein NMR spectroscopy based on these new methods.