Sponsored By

September 24, 1998

7 Min Read
The Materials Analyst, Part 5: 'It must be the process'

This series of articles is designed to help molders understand how a few analytical tools can help diagnose a part failure problem. 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. He has provided analytical services to material suppliers, molders, and end users for the last 10 years. He can be reached at (414) 369-5555, Ext. 572.

Often when a molder encounters a difficult quality problem with an uncertain cause, he is caught between an end user who is demanding a rapid resolution to the problem and a material supplier who is too busy to devote time and resources to a small account. In such cases the molder is likely to hear the refrain from both sides, "It must be the process." It's a popular response that immediately shifts responsibility to the molder. Given the complexity of the injection molding process, it is a difficult statement to refute in the absence of concrete evidence to the contrary. The problem solving becomes even more convoluted when multiple material suppliers are involved.

This is the kind of case that was presented to us one day when a molder sent us a package containing a good and a bad part, two lots of natural polypropylene, and two lots of a blue color concentrate. The good part had been made with one lot of polypropylene and one batch of concentrate; the bad part had been produced from the second lot of raw material and colorant. And the molding problem was a strange one. The bad parts came out of the mold looking fine. Hours later, several areas of the part began to exhibit a color change. The change was similar to the effect observed when a toughened material is placed under stress and the surface of the part whitens. But these parts had just been sitting in the box.

The molder had reviewed the processing conditions thoroughly and was now looking for some help with an analysis of the material. In the past few months we have highlighted the use of two analytical techniques, differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA) to solve quality problems. In this case we will show how the two techniques are used together to assemble the pieces of the puzzle.

Working on the Problem

The first thing we noticed when we examined the parts was that the stress whitening occurred only in areas where two walls met to produce a thick section, an area where cooling would take longer than in the rest of the part. We decided to check the composition of the two polypropylenes and the two lots of colorant. The polypropylene was tested with DSC, a technique that measures key transitions such as crystal melting and the glass transition temperature. The two lots proved to be identical; Figure 1 shows the results for one of the lots. It shows a melting point at 165C. A closer look at the curve shows a small dip near 120C. This comes from the melting of polyethylene and is the signature that we look for in identifying a copolymer. There was no hint of contamination or any other defect. Even the melt flow rate of the material in the two parts was essentially the same. The concentrates, however, were a different story.

Figure 2 shows a DSC scan for both lots of color concentrate. It doesn't take an analyst to see that these concentrates are not based on polypropylene. A preliminary check of our library suggested that the carrier resin was low-density polyethylene, but the shoulders and knees in the curve indicated that the resin contained other ingredients, including a small amount of high-density polyethylene. This is a common occurrence in the industry. As suppliers seek to cut costs and reduce inventory, so-called "universal" colorants are becoming increasingly popular with concentrate manufacturers. Sometimes they work, but in other applications, this type of substitution can produce less-than-satisfactory results.

All this doesn't explain why one of these batches of concentrate produced good parts and the other one did not. Maybe it was the process. But a closer look at the curves in Figure 2 made us suspicious. While the curve shapes and the various melting points are nearly the same, the concentrate used to mold the bad parts displays a more energetic melting process. Notice how the melting peak dips down farther and the whole area under the curve is greater. Since the actual pigments used in the colorant system do not soften or melt, they are invisible to the DSC. Therefore, the difference in melting energies indicates a possible difference in the resin-to-pigment ratio between batches of concentrate.

TGA tests on the concentrates confirmed this. TGA examines weight loss as a function of temperature. The manner in which a material decomposes and the amount of material that is left at the end of a test provide valuable information about composition. Figure 3 shows the weight loss vs. temperature plots for the two batches of concentrate. The ash content for the good batch was 44 percent, while the residue left from the bad batch was only 25 percent. The amount of residue or ash is directly related to the pigment loading in the concentrate. This meant that the good batch of concentrate contained almost twice the pigment of the bad batch. As the amount of pigment decreases, the level of carrier resin increases. At a 20:1 letdown, each 100 lb of material will contain 1 lb less of pigment and 1 lb more of carrier resin in the bad lot.

And what about the carrier resin? The TGA provided some additional information. Figure 4 shows the weight loss and weight loss derivative for one of the concentrate lots. If we look closely at the curves, we can see that almost 10 percent of the material decomposes before the main polymer ingredient starts to degrade. If the concentrate were based strictly on polyethylene, this early weight loss would not be present. Instead, the weight loss would remain near zero until the polyethylene began to burn off.

This early weight loss is typical of a material we discussed in last month's article, ethylene-vinyl acetate (EVA) copolymer. Ethylene-vinyl acetate is a popular carrier resin used in many color concentrates. It is inexpensive, melts at a low temperature, and is thought of as being compatible with a number of materials, especially polyolefins. However, because of its low melting point and even lower glass transition temperature, EVA has the ability to crystallize even below room temperature. This can result in residual shrinkage long after the part appears to be stable. This explained the delayed development of stress whitening in the molded parts and also accounted for the fact that the flaw was only

evident in the thicker sections of the part where complete cooling would take longer.

So why did the EVA work in one case and not in the other? Two factors combined to produce the negative result. When the concentrate was almost 45 percent pigment and 55 percent carrier resin, the effect of the carrier resin was minimized and shrinkage was kept to an acceptable level. In addition, the extra colorant probably served to cover the visual effects of any differential shrinkage that might be produced by the EVA. When the colorant level fell to 25 percent, the EVA concentration increased and the residual shrinkage became more of a factor. The reduced amount of pigment failed to hide the visual effects of shrinkage-induced stress.

It should be noted that in spite of the large difference in pigment loading, the parts did not display any difference in color. From a color-matching standpoint, the expensive pigments were not needed to produce the target color. However, with an incompatible carrier, problems arose that could only be eliminated by using costly excess pigments. A concentrate based on polypropylene would have prevented the problems with nonuniform shrinkage and the extra cost of using polypropylene as a carrier would have been more than offset by a reduction in pigment loading. This was one instance where the conclusion that "it must be the process" just didn't hold up, and the molder could respond with certainty that "it must be the material."

Sign up for the PlasticsToday NewsFeed newsletter.

You May Also Like