Sponsored By

The Materials Analyst, Part 54: Where does the moisture go? (Part 4)

August 20, 2002

9 Min Read
The Materials Analyst, Part 54: Where does the moisture go? (Part 4)

In this last segment of the series, we will use the Karl Fischer moisture measurement method to follow moisture from the pellet to the part using an unfilled nylon 6/6 that was intentionally conditioned to different moisture levels from very dry to very wet. This enables us to understand the interaction between the polymer and the water. It also illustrates the lack of correlation between the moisture content of raw material and a related part days or weeks after it was molded.

To perform this experiment, we created batches of raw material with moisture content values from a low of 280 ppm (.028 percent) to a high of 10,900 ppm (1.09 percent). Depending upon which literature you read, the maximum allowable moisture content for unfilled nylon raw material is between 2000 ppm (.20 percent) and 2500 ppm (.25 percent). Material introduced into the molding machine barrel at levels greater than these recommended values enter into a reaction with the moisture known as hydrolysis.

During hydrolysis water breaks bonds in the polymer chains, reducing the length of the chains and with it the properties of the material. The degree of damage depends upon the moisture content, the melt temperature, and the residence time of the melt in the barrel. A low melt temperature and fast throughput can reduce or prevent degradation of a material where a higher melt temperature and a long residence time may break down a material of the same moisture content. In this experiment, we kept the molding conditions constant so that the changes we observed were only due to differences in moisture content.

We prepared nine different batches from the same lot of raw material. Three of the batches were very dry, three were near the maximum allowable moisture content, and the last three were very wet. The moisture content of each sample was measured just prior to molding sample tensile bars. The tensile bars were then brought immediately to the lab and tested for moisture content. Table 1 shows the results.

The trend is fairly obvious. When the material going into the process is extremely dry, the molded parts actually come out of the mold with a higher moisture content than the resin from which they were produced. As the raw material approaches the allowable limit, this moisture gain declines. Parts molded from wet material contain less moisture than the pellets, and the higher the moisture content the greater the decrease. So what is happening here?

Condensation Polymers
At the high moisture end of the spectrum the process is fairly straightforward. Nylon polymer with a high water content that is heated above the melting point reacts with the water. In the process, some of the water is actually consumed and becomes part of the degraded nylon chains. It is obvious that not all of the available water reacts. However, as the water content in the raw material increases, the amount of water consumed by the reaction also increases. This increased consumption should translate to a greater degree of degradation.

This particular molding process involved a very small injection unit, a simple part geometry that allowed for a fast cycle, and the use of approximately 40 percent of the barrel shot capacity. In addition, the melt temperature was only 520F, which is at the low end of the recommended processing range. If we were to expand this experiment to include higher melt temperatures and longer residence times, we would very likely observe that for any given initial moisture content more of the water would be consumed. The final moisture content in the molded part would be lower, and the polymer in the part would show increased degradation.

The moisture increase in the very dry materials is not as well documented, and there may not be a definitive explanation. However, the best theory attributes the moisture content increase to continued polymerization of the nylon. First, it's important to know that nylon 6/6 is a condensation polymer. This means that the chemical reaction that creates nylon polymer also produces a low-molecular-weight byproduct, which just happens to be water.

Because of their chemistry, most condensation polymers are capable of undergoing continued polymerization if they are exposed to elevated temperatures in an inert atmosphere, which means no water and no oxygen. Many nylon suppliers actually employ this route to increase the molecular weight of nylon compounds. The process is called solid-state polymerization because typically it is conducted at a temperature below the melting point of the material. The process is run in dry nitrogen, and because additional polymerization is taking place, additional water is created that must be removed so that it does not interfere with the polymerization. If the water were not removed, the moisture content in the reactor would gradually increase until an equilibrium condition was reached and no further polymerization would occur.

If you read ASTM D 789 carefully, the section that discusses the selection of an appropriate test temperature for moisture analysis deals with this phenomenon. We also referred to it briefly in the second installment of this series (May 2002 IMM, pp. 42-45). Now consider a very dry batch of nylon entering the molding machine. The temperature of the cylinder is obviously higher than that used in the solid-state polymerization process, and it is reasonable to propose that continued polymerization might occur even more rapidly in a controlled melt than it would in the solid state. With this polymerization would come the release of water, thus accounting for the increased moisture content of the molded part.

As the moisture content of the raw material increases, the tendency for continued polymerization decreases. As shown in Table 1, when we reach a moisture content close to the recommended maximum for good processing, the net change in moisture content is nearly zero. If we graph the results in the table, we can actually see this effect in Figure 1.

Melt Flow Rate
Part of the problem with using the moisture content of the part as a measurement of the moisture content of the raw material is that it changes, particularly when the material is wet. And while Table 1 makes the change look orderly and predictable, different process conditions change these relationships.

But the real problem has to do with the time between the molding date and the moisture test date. When we receive a request to check part moisture content to show that the part was molded with wet material, the part has usually been sitting in an ambient atmosphere for days or even weeks. During that time the part is absorbing additional moisture. In nylon this process is well documented, but it occurs in other hygroscopic materials as well. Therefore, the moisture content of the molded part is not connected to the moisture content of the pellets used to make the part.

The best determination of whether a material has been properly dried is a measurement of the molecular weight of the polymer in the molded part compared to the molecular weight of the polymer in the raw material. We have discussed the various techniques for this test in previous articles. The simplest of these methods is the melt flow rate test.

For unfilled materials, the melt flow rate of the material in the molded part should not increase by more than 40 percent over the melt flow rate of the pellets used to mold the part. Once the parts exceed that limit, problems with part performance may begin to occur, and these problems increase as the level of degradation increases.

Figure 2 shows the results of melt flow rate tests performed on each of the nine sets of molded parts produced in our experiment. The line representing a 40 percent increase is shown to mark the transition from good molding to poor molding. The correlation between degradation and pellet moisture content is obvious. Some analysts prefer to use intrinsic or relative viscosity measurements to verify degradation, but the objective of documenting a change in average molecular weight is the same.

Unfortunately, the viscosity tests that check for degradation do not necessarily point to wet raw material. High moisture content in raw material is only one possible cause of degradation, and in some materials, it is difficult to sort out the effects of moisture from those of elevated melt temperatures and long barrel residence times.

All polymers are susceptible to thermal degradation, but the list of materials that are prone to hydrolysis is much shorter. For those materials that can be degraded by both heat and moisture, the exact root cause can sometimes be determined by infrared spectroscopy or a more sophisticated chemical technique called end group analysis. However, it is often more productive to return to the production floor to find the root cause. By intentionally varying different process conditions and performing the melt flow rate test, a processor can get a firm grasp on the root cause of polymer degradation. This also helps reinforce that very necessary link between the analysis lab and the manufacturing floor.

Next time, we will see how a processor used the MFR test to optimize the strength of a part molded in one of the most challenging of the condensation polymers, PET polyester.


Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like