Amorphous and Crystalline Metals

Metals are generally crystaHinc and glasses are lunorphous. It should be noted that liquids and gases do
not show any ordering on an atomic scale. Thougli not obvious at present, this classification is useful in
the processing of materials. Amorphous m~terials are generally cooled very rapidly, while crystalline
nmx-ials are processed more slowly so tliat the atoms have sufficient time to occupy their respective
positions. So, if you want to make a niaterial amorphous, the clue is to heal lhc material till it melts and
rapidly as possible from its liquid state to then cool il as the solid stale. On the other hand, if you want to
make the material illore crystalline, then the processing should be very slow indeed. To give an example,
single crystal silicon is oftelm grown at the ralc of a few millimetres per hour (and tliis too is supposed to
be rapid for single crystal growth!), while amorphous polynlers are drawn into sheets at the rate of a
few rnelres per seco~id! liis classification does not help when we consider categorisiilg mnaterials in
terms of grouping of properties. Boh kinds of solids can show optical transparency or opaqueness; botl~
show ferroniagnelic ordering or sitnilar electrical conductivity. The only difference in the two cases is
that the properties of ,amorphous solids show greater variation with time tl~an crystalline ones. This has
enhancer1 tlie belief that Ule aiiorpli6uS-state is a metastable state and gradually, over an extcnded
time in many cases, tlie ~natcrial goes over to the crystalAine slate. And yet, today, we have been able
to use, with adv'mtage, some amorphous solids. Glass Corms a rnajor material for widespread use
ranging from test-tubes to window pancs to televisiorl screcns. In amorphous ferromagnets wc have
been able to make some really good permanent niagnets as well as some very good magnetic shielding
materials. Single cryshls (or materials wllicli show perfect atomic ordering throughout the bulk) also
have inlportant apglicaticms such u dia~nond for cutting or for measuring the hardness of materials,
optically active crystals such as calcile, laser crystals such as ruby etc. So the large scale mangcment of
atoms does show significant property differences, and we should be able to process nilaterials to take
advantage of these features. But what is the scale over which materials can be considered to be
ordered? We know that one can have perfect disorder as in glass over lenglhs or tens of metres (large
glass slieets are routinely made and polymers like polyethylene are drawn into sheets tllal may be
kilometres long !). Do we have perfect xdering over say a metre leriglh? As we shall see, this is certainly
possible but very difficult to achieve in practice. It is niore conimoii to grow single ~rystals of a
centirnetre length and in e~igi~ieeri~lg materials, the scale of cristalline order spans over distances of a
few micrunzelcrs! Can we qumlify the aniount of crystallinity in a material? This question is important
since, as mentioned above, many of today's important materials have regioris of crystalli~iity and
regions of amorphousness. A lump of carbon orc rnay be made of many large diamonds merged
togelher and the diamond artisan carefully chips this lump so tliat a large piece can ~nGrials Structure of
Materials be obtained. This piece has definite external shape with smooth faces. This is a characteristicof
the single crystal. You can see the same feature in common rock salt or sugar crystals. Each "grain" of
salt has the sodium and chlorine atoms arranged regularly in a three-dimensional array. Each grain,
weighingrabout one thousandth of a milligram, would have more than a thousand billion ions, all
regularly placed. This arrangement of ions can be "seen" using X-ray diffraction and is discussed in a
latter Unit. A property of this regular arrangement is that, when a crystal is struck sharply with a knife
edge, the c;rystal "cleaves" into two. The resulting two small crystals show the same kind of external
appearance as the larger crystal. The property of cleavage is used by diamond manufacturers to "cut"
the diamond appropriately so that the resulting entity gives out the maximum brilliance. A copper rod,
on the other hand does not have the property of cleavage. It, however, shows a crystalline nature when
subjected to X-ray diffraction analysis. So what is the difference between the atomic arrangement in
diamond and in the copper rod? The difference is that in the copper rod there are a large number of
"grains" or crystallites and these are randomly "glued7' together. A copper rod, then, is said to
polycrystdline. Most engineering n~etallic solids are polycrys talline and making single crystal metals is
difficult. What about a sheet of glass? X-ray diffraction shows that in this sheet the atoms do not show
any regular arrangement. That is why, when glass is struck sharply, it breaks up into irregular pieces. As
can be appreciated from the discussion above, using X-ray diffraction analysis, it is possible to quantify
the amount of crystallinity in a material. This is an important characteristic of polymeric materials as
their mechanical properties do show dependence on the degree of crystallinity. This can be understood
as follows. In a polycrystdline material their are many grains "stuck" together. In each grain there is
perfect ordering and so the material has high mechanical strength as it takes a lot of energy to break up
a bond. But between two crystallites there is a region of disorder called the "grain boundary" and this is
a place of weakness. Materials tend to fracture at the grain boundaries. The wehess of the grain
boundary is taken advantage of in studying the degree of crystallinity. If a solvent is applied to a surface
of the polycrystalline material, the atoms that dissolve first would be those in the grain boundary as they
would be more loosely bound. The solvent is called an etchant and the process of etching is the first step
in studying the surface of a material under a microscope. Such an analysis is called microstructural
analysis. Under a microscope one can see regions which have definite boundaries and so are most
probably crystalline and regions where there are no clear cut boundaries and so would be amorphous.
Using image processing techniques it is now possible to quantify the mount of such phases in any
engineering material. To summarise, then, a polycrystalline material shows the presence of grain
boundaries. If a material does not show grain bounduies then it is either a single crystal or it is (UI
amorphous material. Grain boundaries generally have dimensions of the order of the wavelength of
visible light and so tend to scatter light. Polycrystalline materials tend to be opaque. Single crystals as
well as amorphous materials do show transparency to lighl. In fact, if one wants to increase the
transparency of a polymer sheet, one must see that crystallisation is prevented. Also, it has been found
that most corrosion in solids starts from the grain boundaries, and single crystals or amorphous
materials show very low degradation by corrosion. Taking advantage of this, degradation of
polycrystalline materials by corrosion has been slowed down by heat-treating the surface so that an
amorphous film is formed that gives protection to the bulk. One must realise that classification is often
ambiguous. In facl, in today's world most classifications have broken down. We have been able to make
malerials which combine properties lhat did not seem to he compatible. Glass is often termed as a
super-cooled liquid. But by properly cooling the glass we have been able to make a rnixlure of glass and
a crystalline phase, the product being called a glass-ceramic! We have materials that are liquid and yet
show crystalline order, such as liquid crystals used in most of our wrist watches, Further, we have made
composite materials which combine glasses and meh In spite of these complexities, a simple
understanding of materials does emerge by this classification. Oilly, one must remember that this
categorisation is, at best, a cohenient method of grouping together properties, or methods of
processing. Let us expand this idea.