The Materials Analyst, Part 19: Stained PVC

Most of us in the industry involved in making products for outdoor use know it can be a rough environment for many plastic materials. Cold temperatures can cause ductile materials to become brittle, heat can cause warpage and excessive deformation, and sunlight often changes the color of a product. Add the physical stress of driving winds and rain and the impact of an occasional hailstorm and you have the makings of a very inhospitable place.

Over the years, the industry has learned that some materials do particularly well in the great outdoors while others are best kept inside. But when one of the materials with traditional success has a problem, it is time to look for irregularities that might explain the failures.

This particular case involved some flexible PVC parts that were developing unsightly black stains after a relatively short period of time in the field. The stains were small patches distributed randomly around the surface of the part, but they grew larger with time. The product was a high-end item, and the appearance of the stains was of great concern to our client. We were provided with parts that had been returned from the field and new product that had never been exposed to the elements. The parts had a fairly large surface area. While the field returns were badly stained, they contained large areas where the material still looked normal.

Our first inclination was to blame the problem on the effects of sunlight. In use, about half of the surface of this product was almost always exposed to light while the other half faced the ground. A preliminary evaluation revealed all of the stains appeared on the side facing the sun.

Our initial checks for correct composition verified that the material was free from contamination. However, as we began to make careful comparisons between the new product and the field returns, signs of degradation became evident. In examining the field returns, we tested areas that still looked normal and compared those results with tests on the areas that were stained.

The Onset of Oxidation
One of the first evaluations we performed was a test for oxidation. This is done by simply heating a sample in the DSC in the presence of oxygen and watching for the strong exothermic reaction. Figure 1 shows a comparison of a fresh sample with a stained and an unstained region of a field return. The new material displays excellent stability until the temperature reaches 200C. Oxidation sets in rapidly beyond this point.

The samples from the field show a much earlier onset temperature. Even the area that is not stained begins to oxidize at 165C. The stained sections show comparable behavior. This result shows the material that has been in the field has lost some of the stabilizers needed to protect against environmental degradation. However, the stained areas did not look appreciably worse than the rest of the part that had been in the field.

TGA tests were also run on new and returned product. These tests were run in nitrogen, so they represented a comparison of thermal stability without the influence of oxygen. Here again, the results from the stained and unstained field product were comparable.

However, the new material and the field material showed considerable differences in how they degraded. Figure 2 shows the weight loss process from room temperature to 250C for new and returned samples. The weight loss starts at a lower temperature and proceeds more rapidly in the field return. This is a confirmation of the DSC results and again shows the material that had been in use has either lost or never had the amount of stabilizer incorporated into the new product.

While this was good information, we had not yet identified anything distinctive about the stained areas of the product. We next turned to a technique we have not discussed in any of the previous articles in this series-dynamic mechanical analysis (DMA).

DMA instruments perform a variety of functions, and a thorough treatment of all the possible uses could, literally, fill a large book. For our purposes in this case, the primary use of DMA is to provide a picture of how temperature affects modulus.

Plasticizers and Flexibility
PVC is a very versatile material, and its properties can be altered considerably through the use of various fillers, impact modifiers, and plasticizers. It was the plasticizers that interested us in this case. In its pure form, PVC is a rigid material at room temperature. When it is heated, it reaches an important point known as the glass transition when the temperature approaches 80C. The glass transition marks the point where the uncrystallized regions of the polymer structure become mobile.

For our purposes, we can think of it as a melting point for the amorphous portion of a polymer matrix. Because PVC is amorphous, this transition results in the softening of the material. Anyone who has tried to run very hot water through a PVC pipe has an intuitive understanding of this event.

Plasticizers lower the glass transition of a polymer. If we add enough plasticizer to a material, the glass transition temperature (Tg) falls below room temperature, and we perceive the material as flexible rather than rigid. Elastomers owe much of their unique property profile to the fact that the glass transition temperature is well below room temperature in these materials, but there is just enough structure left above the glass transition to prevent complete softening.

PVC can accept large amounts of plasticizer, and flexible PVC results when enough plasticizer is incorporated to reduce the Tg to a point where the material has a very low modulus at room temperature. The lower the Tg, the more plasticizer is present in the compound. Therefore, a measurement of Tg will indirectly tell us about the plasticizer content.

Figure 3 shows a modulus vs. temperature plot for a sample of the PVC that had never been in the field. The test is started at a very low temperature of -170C (-275F) and taken up to 85C (185F). At very cold temperatures, the flexible PVC is not flexible at all. Below the glass transition, it is a rigid polymer with a modulus comparable to that of a 15 percent glass-reinforced nylon.

As the temperature increases, the modulus declines. The decline is gradual at first, but the rate increases as the material passes through the glass transition. Above the Tg, the material has a modulus of about 6000 psi or about 1 percent of the stiffness it displayed at the beginning of the test. The effect of the glass transition is easily observed, but this plot alone makes it difficult to pin down the exact glass transition temperature. Another measurement can give us more information.

The Loss Modulus
Figure 4 shows the same plot with a second measurement added. Now we are seeing the modulus and something called the loss modulus plotted at the same time as a function of temperature. A thorough explanation of the loss modulus will have to wait, but, for now, it can be thought of as a measure of how compliant the material is.

When the loss modulus is low, a material will exhibit good recovery from an applied stress. When the loss modulus is high, it means a material will be susceptible to permanent deformation. All polymers exhibit a rise in the loss modulus as the material passes through the glass transition, and convention dictates that the Tg be identified as the peak of the loss modulus curve. For the new vinyl, this value is -47C (-53F). This is our baseline for evaluating the material from the field.

Figure 5 shows a comparison of the modulus for the new material, field material that was stained, and field material that still looked normal. The field material that had not developed any staining had the same Tg as the fresh material, and at room temperature, it was still very flexible. However, the stained material had gone through some key changes. First, the low-temperature modulus of the stained material was significantly reduced for some undetermined reason. More importantly, the manner in which the modulus declined had changed.

The softening process had broadened and had been shifted to a higher temperature. The shift was enough to increase the room-temperature modulus to 20,000 psi. Figure 6 shows the comparisons of the loss modulus values. The glass transition of the stained material had increased from -47 to -10C (14F). This told us the stained region had lost a lot of plasticizer while adjacent areas without staining were as good as when the material was new. Some of these stained areas were noticeably more rigid and were starting to show signs of cracking.

This is a familiar process in plasticized PVC. While it is a very useful material, the downside is the plasticizer can migrate through the PVC and leach out over time. As the plasticizer leaves the system, the Tg increases. Finally, the material loses its flexibility and becomes rigid and brittle.

Anyone who has owned an older car has observed this process, particularly in the rubber boot around the stick shift. The flexible accordion shape that moves with the stick begins to crack. Older dashboards also used plasticized PVC, and they too will show signs of embrittlement after years of exposure to bright sunlight and heat. In fact, the new car smell that car owners love so much is usually due to airborne plasticizer already leaving the vinyl moldings inside the car. These compounds also cause the problems with window fogging.

Infrared Spectroscopy
With the DMA tests, we had succeeded in finding a difference between the stained and unstained areas of the vinyl that had been in the field. Plasticizer was clearly being removed from select areas of the product. As a final step, we dissolved some of the new vinyl and the stained vinyl in acetone and performed infrared spectroscopy.

This is another technique we have not profiled in our previous articles. However, it is an extremely valuable tool in identifying the composition of organic materials. The principle involves passing infrared radiation through a sample. Chemical bonds in the sample that vibrate at frequencies in this infrared region will absorb the incident radiation at these particular frequencies and allow the rest of the radiation to pass through unobstructed. The result is a spectrum like the one in Figure 7. With a good knowledge of organic chemistry, or with a good computer library of standard spectra, the test result can be matched to a known material.

Our particular results gave us two key pieces of information. First, the infrared scans confirmed that the amount of plasticizer in the stained material was approximately 50 percent of that contained in the virgin material. We were able to determine this by the relative strength of the absorptions of infrared radiation in the two samples. In addition, the new material dissolved completely in the acetone while the stained material left a residue. This gave evidence indicating the vinyl was beginning to crosslink in the areas where staining occurred.

Usually, crosslinking is a late stage of degradation that begins with molecular weight reduction. This helped to explain the sharp reduction in low-temperature stiffness we saw in the DMA tests. Finally, the infrared results allowed us to identify the plasticizer as the commonly used diethylhexyl phthalate (DEHP).

So where did this leave us? Our initial DSC and TGA tests showed the outdoor environment was beginning to take its toll on the antioxidants and UV stabilizers that had been incorporated into the material. This was the first step in a process that led to degradation of the PVC and ended with the partial crosslinking of the material. But the real culprit in the staining had nothing to do with that process.

Instead, critical plasticizer was being extracted from the vinyl in selected areas, leaving the material without the required flexibility. This process may be promoted by the initial degradation of the polymer or caused by microbial attack because microorganisms actually gain nutritive value from some phthalate esters.

Because the product was from Asia, it was difficult to get definitive information about the types and quantities of stabilizers being added to the PVC. However, the short period of time from product release to the appearance of the problem strongly suggested the additive packages for UV protection were less than optimal. In addition, more stable plasticizers that would resist both migration and microbial attack were available, and we recommended they be incorporated into the formulation.

Unfortunately, these higher performance plasticizers, which were only marginally more expensive in the U.S., sold for a significant premium in Asia at that time. The familiar struggle between cost and performance was very evident in this application.