Sponsored By

August 1, 2003

10 Min Read
The Materials Analyst, Part 59: Frequently asked questions (Part 2)—Glass and the glass transition

Troubled_blob (5K)

This series of articles is designed to help molders understand how a few analytical tools can help diagnose a part failure. Michael Sepe is our analyst and author. He is the technical director at Dickten & Masch Mfg., a molder of thermoset and thermoplastic materials in Nashotah, WI. Mike has provided analytical services to material suppliers, molders, and end users for 15-plus years.

Often we are asked to examine the temperature- and time-dependent behavior of a material in order to determine its fitness for use in a particular application. This often leads to an assessment of a very important transition in the material known as the glass transition. When the client receives our report we are often asked to define how much glass we found in the material.

Before you polymer chemists and physicists (I am assuming that two or three of you occasionally read this column) begin to chuckle, consider what a poor job we do of explaining the glass transition to the design, engineering, and manufacturing community. Countless lectures on polymer structure start with the statement, “Below the glass transition a polymer is hard and glassy while above the glass transition it is soft and rubbery.” Some texts even refer to this event as the glass-rubber transition. To an engineer hard and glassy implies brittle while the phrase “soft and rubbery” conjures up images of a structure that is incapable of bearing the slightest load.

But a material like polycarbonate, when in the solid state, is by definition below its glass transition temperature (Tg). And while it is certainly a hard material it is anything but brittle. By the same token, a 50 percent glass-fiber-reinforced nylon 6/6 at 110C (230F) is above its Tg and yet has a tensile strength at yield of more than 20,000 psi and a modulus of approximately 1 million psi. This does not exactly fit the description of a rubber band.

Look further for sensible definitions and you come to statements like, “The glass transition represents the onset of coordinated motion in the main chain backbone structure.” Succinct enough for a physicist but not exactly designed to clarify matters for an engineer. So let’s see if we can translate this into practical terms.

Begin at the Beginning
We start with the concept of crystalline solids. Most classical materials that are solids at room temperature owe their solid state to a well-organized crystal structure. In this crystal structure all of the atoms and molecules are in well-defined locations and the strength of the bonds in this crystalline structure is reflected in the melting point of the material. Melting represents the transition from solid to liquid. Material scientists refer to this as a first-order transition because it involves a fundamental change in the long-range order within the structure of the material. The liquid-to-gas transition is also considered a first-order transition.

There are some materials, however, that appear to be solid but actually contain no crystal structure. One of the most familiar of these materials is glass. Glass has many of the bulk physical characteristics of a crystalline solid like aluminum or sodium chloride, but if we evaluate glass using the tools that can actually detect crystals, we find that glass contains none. Glass is a substance with thousands of years of history, and often these materials provide the archetype for a whole class of compounds that exhibit the same type of behavior.

Consequently, materials that appear to be solid but contain no crystalline structure have become known as amorphous solids or glasses. To material scientists, who are very picky about such things, these materials are not even classified as true solids. Instead they are treated as supercooled fluids, liquids with extremely high viscosities. Most of us have heard the stories about panes of glass that are slightly wider at the bottom than at the top after years of use. This is physical evidence that materials like glass actually do flow, albeit very, very slowly.

The Chains that Bind
In polymers the molecules are very large and the long, chain-like structures allow for a wide variety of interactions. Consequently, crystallinity in polymers is not the all-or-nothing proposition that it is in classical materials. Crystallinity arises primarily from a process known as chain folding. As a polymer cools, the individual chains may be able to fold up into well-organized regions that take on the properties of a crystalline solid. The ease with which this folding occurs depends upon the shape of the polymer molecule. A very symmetrical and streamlined material like polyethylene can potentially achieve a crystalline structure over 90 to 95 percent of its volume. Materials with these properties are classified as crystalline or, more appropriately, semicrystalline.

Other materials like polycarbonate and polysulfone have very bulky backbone structures. The chains that make up these materials do not fold readily and they produce almost no well-ordered regions. Most of the structure in materials like these remains random and disorganized and these materials derive their properties almost exclusively from chain entanglements. Materials like polycarbonate are classified as amorphous since they do not naturally possess any significant degree of crystallinity.

The structure of some materials allows them to straddle the fence that divides these two classes of materials. PET polyester is a good example. PET polyester in its amorphous state is tough and transparent and is familiar to all consumers as the preferred material for soft drink bottles and some food containers. However, with some help from additive chemistry, PET polyester can also form enough of a crystal structure to develop high-heat capabilities and be used in an automobile engine compartment.

No crystalline polymer is 100 percent crystalline and no amorphous materials are completely free of crystal structure. But when approximately

35 percent of a polymer structure achieves crystallinity, it takes on the bulk characteristics of a crystalline material. The primary evidence of crystalline structure is a melting point, a temperature at which the energy in the material produces sufficient molecular motion to break up the crystal structure and change the material from solid to liquid.

The Transitions
The melting point defines the phase change from solid to liquid for those areas of the polymer that form crystals. But what about those areas that do not crystallize? As it turns out, these regions also undergo a transition of their own. However, it is not treated as a phase change such as melting or evaporation because under the strict definition of material structure, amorphous materials are already fluids. They just exhibit extremely high viscosities that make them appear to be solid.

Despite this lack of crystallinity, it is an observable fact that you can pump a significant amount of heat into an amorphous material and nothing much appears to happen until a particular temperature is attained. Then, suddenly, the material loses all of its load-bearing properties. Material scientists call this a second-order transition; in this case it represents a change from a very high-viscosity liquid to a much lower-viscosity liquid. This is the glass transition.

Figure 1 shows an artistic rendering of how we imagine an amorphous polymer looks at the molecular level. The chains are long, random coils that are loosely associated with one another. At room temperature, useful amorphous polymers appear to be solid by virtue of this loose association. Scientists learned long ago that while everything appears to be fixed and motionless in solid materials, there are subtle vibratory movements within the material that increase in magnitude as temperature increases. In fact, temperature is nothing more than a convenient construct for measuring molecular motion.

At some temperature, which is primarily dictated by the stiffness of the individual chains, this motion becomes significant enough to loosen up large segments of each chain from its associations with neighboring chains. This is the coordinated, or cooperative, motion of which polymer scientists speak. When it happens, amorphous materials soften, declining in modulus by two to three orders of magnitude.

If your process of choice is thermoforming, you can begin to work with the material at this point. Those of us in injection molding, however, must put additional work into the material in order to reduce the viscosity to the point where we can achieve the flow that our process demands. But if you are looking for the upper limit of useful load-bearing properties, the glass transition is it for amorphous materials. In amorphous materials, the familiar heat deflection temperature and Vicat softening temperature are closely related to, and in fact are driven by, the glass transition.

Figure 2 shows a schematic for a semicrystalline polymer. Through a significant portion of the structure we see the same loose association we observed in the amorphous material. But embedded in this disordered soup are regions of relative organization. These are the crystals, and they develop as the material cools from the melt; the regular and streamlined shape of the chains allows for close packing.

We have already established that these crystals represent a true solid and therefore have a melting point. But what about the areas surrounding these crystals? These areas are amorphous and therefore will also exhibit a glass transition. For most materials that do form crystalline regions, the glass transition temperature can be found 150 to 175 deg C (270 to 315 deg F) below the crystalline melting point.

Additionally, we already have said that when an amorphous polymer reaches its glass transition it softens and becomes useless as a load-bearing material. No matter how much we reinforce an amorphous material, the glass transition temperature defines the upper limit of useful structural properties. But in a semicrystalline material, the glass transition will not result in complete softening because the crystals are not affected by the increased mobility of the surrounding amorphous regions. Instead, unfilled semicrystalline materials lose 70 to 90 percent of their modulus as they pass through the glass transition and then retain the remainder until the crystal regions start to melt. Figure 3 shows a comparative plot of elastic modulus vs. temperature for an unfilled polycarbonate (amorphous) and nylon 6 (semicrystalline).

To Shrink or Not to Shrink
An interesting consequence of the difference between the first-order transition of solid-to-liquid and the second-order transition of liquid-to-liquid can be observed when the materials return to their rigid, or solid, state. This consequence is shrinkage. Because amorphous materials undergo no fundamental change in structural order, the volume they occupy above and below the glass transition is nearly the same. Consequently, unfilled amorphous materials shrink only about .5 percent and this volumetric contraction tends to be uniform across the entire mass.

However, it is apparent from Figure 2 that crystalline regions occupy less space than the surrounding amorphous material. Consequently, as semicrystalline materials solidify, the change in volume is much greater. In addition, a higher degree of crystallinity will result in more shrinkage. Since the degree of crystallinity can be influenced by part geometry and processing considerations, shrinkage in semicrystalline materials is not only greater, it is also harder to predict. So when a manufacturer of polycarbonate publishes a shrinkage range of .005 to .007 in/in and the maker of polypropylene gives .010 to .020 in/in, this is not a reflection of different measuring tools or process controls, but is the nature of things.

So that is the short course on polymer structure. It is necessarily simplified; we could dedicate an entire issue of this magazine to the implications of the glass transition. But for now the essential point is that the term refers to the softening of the noncrystalline regions in a polymer and has nothing to do with the presence or absence of glass fillers or reinforcement. If a plastic compound does contain glass, that glass will also have a glass transition. But we need not be concerned about it, because the glass transition temperature of glass occurs at a point where all of our polymers have long since turned to gas or char.

Contact information
Dickten & Masch Mfg. Co.
Nashotah, WI
Mike Sepe
(262) 369-5555, ext. 572
[email protected]

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like