The intensive work of materials scientists and solid-state physicists has given rise to a class of solids known as amorphous metallic alloys or glassy metals. There is a growing interest among theoretical and applied researchers alike in the structural properties of these materials.

When a molten metal or metallic alloy is cooled to a solid, a crystalline structure is formed that depends on the particular alloy composition. In contrast, molten nonmetallic glass-forming materials when cooled do not assume a crystalline structure, but instead retain a structure somewhat like that of the liquid ― an amorphous structure. At room temperature the natural long-term tendency for both types of materials is to assume the crystalline structure. The difference between the two is in the kinetics or rate of formation of the crystalline structure which is controlled by factors such as the nature of the chemical bonding and the ease with which atoms move relative to each other. Thus, in metals, the kinetics favors rapid formation of a crystallines structure whereas in nonmetallic glasses the rate of formation is so slow that almost any cooling rate is sufficient to result in an amorphous structure. For glassy metals to be formed, the molten metal must be cooled extremely rapidly so that crystallization is suppressed.

The structure of glassy metals is thought to be similar to that of liquid metals. One of the first attempts to model the structure of a liquid was that by the late J. D. Bernal of the University of London, who packed hard spheres into a rubber vessel in such a way as to obtain the maximum possible density. The resulting dense, random-packed structure was the basis for many attempts to model the structure of glassy metals.

Calculations of the density of alloys based on Bernal-type models of the alloys metal component agree fairly well with the experimentally determined values from measurements on alloys consisting of a noble metal together with a metalloid such as alloys of palladium and silicon or alloys consisting of iron phosphors, and carbon, although small discrepancies remained. One difference between real alloys and the hard spheres area in Bernal models is that the components of an alloy have different size, so that models based on two sizes of spheres are more appropriate for a binary alloy for example. The smaller metalloid atoms of the alloys might fit into holes in the dense random-packed structure of the larger metal atoms.

One of the most promising properties of glassy metals is their high strength combined with high malleability. In usual materials, one finds an inverse relation between the two properties, whereas for many practical applications simultaneous presence of both properties is desirable. One residual obstacle to practical applications that is likely to be overcome is the fact that glassy metals will crystallize at relatively low temperatures when heated slightly.

[参考译文]

材料科学家和固体物理学家的深入研究已促进了一种固体物质的出现，这类固体被称为非晶体金属合金，也就是玻璃金属。理论和应用研究者对这些材料的结构特性的兴趣正与日俱增。

当一种熔化的金属和金属合金冷却成固体时，依赖于特定的合金成份将形成各种晶体结构。相比之下，熔化的非金属、玻璃类材料在冷却后将不会形成晶体结构，而是保留一点类似于液体的非晶体结构，在室温条件下，两类材料的自然的长期倾向都形成了晶体结构。它们之间的不同在于动态性，即形成晶体结构的速度。这种动态性受下述两种因素控制：化学结合的性质和分子之间相互运动的自由程度。由此，对金属而言，动态历程有利于晶体结构的快速形成;而对非金属来说，这种形成速度非常慢，以至于任何自然冷却速度都足以形成一种非晶体结构。要想形成玻璃金属，熔化的金属必须以极快的速度冷却，以抑制晶体的形成。

人们认为玻璃金属的结构与液态金属的结构类似。创建这种液体结构模型的第一次尝试是已故的伦敦大学的J. D.鲍纳尔进行的，他将坚硬的球体尽可能多地填塞进一个橡胶容器中，以便得到一种最大可能的密度。这个密度结果以及随机填塞结构以后便成为试图建立玻璃金属结构模型的基础。

基于鲍纳尔模型，由合成金属的成份组成对合金密度的计算结果与实验测得的结果相当地吻合，当然一些细微的差异仍然存在。实验结果是通过测量由一种重金属和类金属组成的合金得到的，如钯和硅的合金，或铁磷和碳组成的合金。实际的合金和鲍纳尔模型所用的球体之间的差别在于合金的成份有不同的体积大小，因此，基于两种大小的球体的模型更适合于两类物质的合金。合金中非金属的小原子可能填进由大原子随机填塞形成的紧密结构中。

玻璃金属最有前景的一个特征是高强度与高延伸性的结合。在常见的晶体材料中，这两种特性一般是成反比的，但人们渴望它们同时存在。在实际用途中可能还有一个问题急待解决，即当玻璃金属在相对的低温下慢慢加热时，它会逐渐变为晶体结构。