NMR Imaging of 3-D Snowpack Structures

积雪的微观结构影响着雪的机械和物理特性，因此我们应该更好地了解它。目前使用薄片和平面切片进行研究。然而，它们只能提供二维信息；因此，需要进行大量的操作来重建三维结构。在这项研究中，我们开发了一种核磁共振成像技术来可视化和量化积雪的三维结构。由于来自冰的核磁共振信号非常微弱，雪中的空气空间被掺杂了乙酰乙酸铁C12H26的十二烷填充，具有良好的渗透性能和低冰点（约-12℃）。因此，核磁共振成像的是十二烷所占据的空间，而不是冰粒。利用自制的4.74 t超导磁体核磁共振成像系统获得三维图像。十二烷的测试成像结果表明，获得一张三维图像需要0.5 ~ 2小时，因此我们开发了一种样品冷却系统，以保持温度低于0℃。该系统有一个外径40毫米，内径20毫米的丙烯酸塑料制成的双管圆筒，冷空气通过该圆筒。制备了两种类型的雪：直径约3毫米的冰球和深白色晶体。特别是，由于冰球足够大且尺寸均匀，它们是测试成像的理想选择。成功获得三维显微图像；图像矩阵为128^3，体素大小为0.2 mm^3。通过不同角度和截面的观察，我们可以清晰地识别出冰球的三维结构。

Microstructure of the snow pack should be better understood because it affects the mechanical and physical properties of snow. Thin sections and plane-surface sections are currently used to investigate. However, they give only two-dimensional information ; therefore, a large number of operations are necessary to reconstruct the three-dimensional structure. In this study we have developed a NMR imaging technique to visualize and quantify the 3D structure of snow pack.Because the NMR signal from the ice was very weak, the air spaces in the snow were filled with dodecane doped with iron acetylaetonate C12H26, which has a good infiltration property and a low freezing point (about -12℃). Thus, the NMR images the space occupied with dodecane instead of ice particles. The 3D images were obtained with a homebuilt NMR imaging system using a 4.74-T superconducting magnet. The result of test imaging of dodecane showed 0.5 to 2 hours were needed to obtain the one 3D image, and hence we developed a specimen cooling system to keep the temperature below 0℃. This system has a double-pipe cylinder made of acrylic plastic with outer diameter of 40 mm and inner diameter of 20 mm through which cold air was passed. Two types of snow were prepared : ice spheres of about 3-mm diameter and depth-hoar crystals. In particular, because the ice spheres are sufficiently large and uniformly sized, they were ideal for test imaging. 3D microscopic images were obtained successfully ; the image matrix was 128^3 and voxel size was 0.2 mm^3. We could recognize the 3D structure of the packed ice spheres clearly by viewing from various angles and cross sections.