The Materials Analyst, Part 20: Finding problems in hot runners

Hot runners can be wonderful things. At best, they eliminate the need to recycle sprues and runners that can account for more than 50 percent of the shot. They cut down on molded-in stress and pressure losses by keeping the material at its stock temperature right up to the point where it enters the cavity, and they reduce cycle time. At worst, they become a haven for degraded material and obstructions in the flow path.

A significant part of the work we do focuses on identifying the composition of metal and polymer contamination clients find in their hot runner systems. Without analytical tools, this task can be nothing but a frustrating guessing game. Two recent cases illustrate how useful it can be to know the cause of the problem.

The Case of the Sticky Residue
The first client brought us a large piece of sticky, dark-colored residue that had been removed from a hot runner system. The system was handling natural polypropylene and producing small parts in a multiple-cavity tool. Periodically, discolored material would begin to contaminate the molded parts. When the system was pulled apart, these chunks of material would be found.

It was hard to believe all we were dealing with was degraded polypropylene, particularly given the relatively mild temperatures the molder was using. The one red flag was that the problem was a new one but the mold was not. Recently, the molder had switched to a new grade of polypropylene in order to enhance the performance of the product in its end use. We were also provided with some of the raw material being run.

We began by trying to identify the tar-like material. We started by running DSC tests on both the residue and a pellet of the virgin material. Because the residue was soft and sticky, we suspected it might contain oil or wax that would be a liquid at room temperature.

In order to confirm this, we started our scan below room temperature, at -50C. Figure 1 shows a comparison of the virgin material and the residue. The virgin material looks very normal. A close examination of the low-temperature region shows some small transitions that occur near 0C. These are associated with the glass transition of the polypropylene. Nothing much happens after that until we begin to melt the material. The melting process looks very typical for polypropylene, and the melting point just below 160C suggests the material is a copolymer. The residue, however, is another matter.

Contamination or Degradation?
In the residue, the intensity of the melting point is drastically reduced. The depth of the peaks is about half of what we observe in the raw material. This would be expected if the polypropylene was being replaced by some foreign material. What we did not expect was the drastic reduction in the melting point.

Although the melting event begins at approximately the same temperature, the melting process is split into three overlapping steps, none of which comes within 10 deg C of the melting point for the virgin material. While melting point reduction can be caused by contamination with another polymer, it can also be a sign of severe degradation. In addition, the low-temperature region associated with the polypropylene glass transition shows some unusually vigorous activity that looks almost like a second set of melting points.

Figure 2 shows the comparison of the cooling portion of the DSC test. Although it is harder to see in this graph, the phase change for the sticky residue is much less energetic, and the recrystallization temperature has also been suppressed by 16 deg C. The reheating step essentially revealed the same features we saw on first heat. The low-temperature region was two to four times more energetic in the residue than in the good material, and the energy associated with the melting point had been reduced to about 60 percent of what we normally see for poly-propylene.

Figure 3 shows comparative TGA results for the polypropylene resin and the hot runner residue. Here again, there were vague similarities indicating that at least some of this material was the polymer. But something else was clearly in the mix.

Normally, in the quiet, inert atmosphere environment of the TGA, polypropylene will remain stable until the material approaches 400C (752F). This may seem ridiculous based on our processing experience, because most of us have seen polypropylene turn to wax if left in a barrel for any length of time at temperatures above 290C (550F). But the TGA instrument is a much more hospitable environment than the injection molding machine barrel. There is no shear generated by a large feed screw. There is no large mass of material; our samples are barely the size of a single pellet. And, finally, in the initial stages of our tests, we use pure nitrogen as an atmosphere, so we can observe the weight loss process without the interference of water or oxygen absorption.

All of this maximizes the apparent stability of the polymer, so the numbers from a TGA test always look better than what happens in real life when we are processing. Tests like this are designed to reflect relative thermal stability, not an absolute correlation with real-world behavior.

Even as a relative indication of stability, the TGA clearly shows we have a problem. Our mystery material begins to lose mass at 250C. By the time we reach the point where polypropylene normally starts to decompose, we have already lost 16 percent of our sample. The rest of the material burns off as a typical polypropylene.

As with the DSC, the picture was emerging of polypropylene mixed with an oily, sticky material that behaves more like an adhesive. It seemed likely that the vigorous low-temperature transitions in the DSC were associated with the early weight losses in the TGA.

The Infrared Spectrum Revisited
While we get a lot of mileage from the combined tools of DSC and TGA in making identifications of unknown materials, there are times when more is needed. In these cases, the best tool for analyzing organic materials is the infrared spectrum.

As mentioned in last month's article, all organic compounds absorb infrared radiation selectively in a way that produces a pattern distinct to the compound being studied. By matching up spectral patterns from unknown materials with a good library, a skilled interpreter can tell a great deal about a material, particularly if it has undergone significant chemical changes.

Figure 4 shows infrared spectra from our residue matched up to a good polypropylene and a purposely oxidized polyethylene. Chemically, polyethylene and polypropylene are very similar and, therefore, produce similar infrared spectra. In addition, they undergo certain key reactions in similar ways.

What we saw was surprising. Our mystery material was nothing other than polypropylene. However, it had undergone a massive modification. The large, broad peak between 1775 and 1700 cm-1 does not belong in a healthy polypropylene. It is the result of oxygen combining with the polymer to form a host of chemical bonds that signify degraded material.

All polymers, polypropylene included, are compounded with small amounts of antioxidant to protect the polymer from such degradation during processing. Some products require additional protection from conditions that the product might encounter in the field. Changes in the molded parts will not occur as rapidly as they do in the melt, but the results are no less catastrophic.

Anyone who has ever performed the automotive oven aging tests for polypropylene knows how suddenly the ductile, flexible sample plaques can turn to dust. And more than one coffee pot component has turned brittle before its time. These failures are related to the rapid unzipping of the polymer once the antioxidant is consumed.

Two generations ago, the wire and cable industry developed a test to determine the relative degree of antioxidant protection present in a polyolefin. It consists of placing a sample into a DSC cell that can be pressurized. Pure oxygen is then pumped into the chamber, and the temperature is elevated. The pressurized oxygen will rapidly consume the antioxidant in the material.

Once this protection is gone, the polypropylene oxidizes rapidly, and the event is measured as a strong exothermic reaction. The time to onset for this reaction is a relative measure of oxidative stability. The test is designed to be complete in approximately one to two hours, but materials with excellent stability may last up to four hours, and others with almost no stabilizer may go off in a matter of minutes.

As with many of these tests, interpretation is helped by experience. Having tested hundreds of compounds in this way, we had a substantial library to fall back on, so we decided to test the virgin material and the residue by this method.

Figure 5 shows the result. The virgin material is stable at the test conditions for about 22 minutes and then oxidizes rapidly. This is a very short time to failure for a virgin material and indicates the level of protection incorporated into this material is minimal. The result for the residue shows no region of stability; the material begins to react with the oxygen immediately.

This explained the deterioration we were seeing in the hot runner. The raw material contained minimal protection against oxidative degradation. The combination of the additional thermal load of the hot runner along with the possibility of dead spots in the system caused the material to break down rapidly. The reduced melting point and the strong activity in the glass transition were signs that as the polymer lost molecular weight it was being reduced to a waxy adhesive. Although we never had the opportunity to test the original molding material for oxidative stability, it is very likely it would have come out much better.

The molder also needed to look closely at the tool to be certain dead spots and holdup zones were not contributing to the material breakdown. Nothing in the end use application dictated this material be especially stable once the parts were molded. The part was used in packaging and had a relatively short shelf life at room temperature. If antioxidant was being added to cover for problems in hot runner design, this would be raising the cost of the raw material to compensate for a mold design problem.

The Case of the Plugged Gate
Our second case came from a client who was experiencing problems with plugged gates. This can come from a variety of sources, including high-melting polymer contaminants, degraded char from hold up spots, and, of course, metals. Determining the composition of the obstruction can help pinpoint the source and prevent further problems.

In this case, we were presented with some plastic material in the familiar shape of a torpedo. In the middle of this material was a hard mass that glowed when touched with the flame from a propane torch.

Inorganic materials require a different set of analytical techniques. They do not melt or degrade at the conditions used in normal thermal analysis. A very useful technique for analyzing metals combines a scanning electron microscope (SEM) with a technique known as energy dispersive x-ray spectroscopy (EDS or EDX).

The SEM gives the analyst a direct look at the contaminant while the EDS produces a breakdown of the elements in the contaminant. Table I gives a partial list of what was found in the metal blockage.

Experienced metallurgists may recognize the ingredients for stainless steel. More importantly, it was what metallurgists refer to as an austenitic alloy (translation: not magnetic). The real bonus came from the actual image produced by the SEM (Figure 6). It is clearly a piece of a screen.

The shape of the piece and the gauge of the wire was distinctive and allowed the client to readily identify the problem as coming from the compounding process.

These two cases show the versatility of analysis to handle not only the polymer side of processing problems but also to delve into the state of the additive package and to identify foreign metallic materials.