Medical Plastics Failures from Heterogeneous Contamination

Medical Plastics and Biomaterials Magazine
MPB Article Index

Originally published March 1998

Michael T. K. Ling, Stanley P. Westphal, Chuan Qin, Samuel Ding, and Lecon Woo

In the health-care industry, failure criteria are often considerably more stringent than in other plastics market sectors. This vigilance is necessary because even minor, seemingly innocuous device defects—especially those biological in origin—could have devastating consequences for the patient. As a result, many visual inspections are built into both the manufacturing process and clinical protocols. Failures detected in these inspections are frequently cosmetic in nature and have no impact on the functionality of the medical device or drug container. However, because of the industry's precautionary practices, most of the defects are deemed unacceptable, and the affected products are rejected. Given today's cost-driven health-care environment, identifying and minimizing these cosmetic defects is an important part of the overall quality process.

The source of these cosmetic defects is often contamination from external sources. To implement a truly fundamental corrective action, the root cause of the defect must be unequivocally identified. Microscopic morphological analysis is an indispensable tool for this effort.

Besides cosmetic defects, another class of failures that originate from heterogeneous contamination involves inclusions in device components. If the inclusion is of a different modulus from the matrix material, it can act as a stress concentrator and cause premature mechanical failure well below the designed stress of the device.

There is yet a third class of failures that are due to external sources: those arising from the uneven distribution of additives and modifiers in the polymer. Since many additives are designed to protect the polymer against oxidative degradation, an uneven distribution can result in part of the product being unprotected during long-term aging, which can lead to premature failures.

In the experience of the authors, failures attributable to heterogeneous external contaminants constitute a significant part of total device failures. A detailed examination of the origin of these failures can help in implementing effective countermeasures. This article will present examples of the different types of failures and propose possible preventive measures to reduce or eliminate them.

EXPERIMENTAL

In dealing with heterogeneous structures and contamination, perhaps the most powerful tool is the optical microscope. Use of a simple stereomicroscope with moderate magnification and a long working distance is invariably the preferred first step for detailed examinations. If the samples are optically transparent, equipment with polarized light capability is also very useful. Many embedded particles are surrounded by molded-in stresses from the modulus mismatch; the stress causes massive birefringence, which can aid in detection and quantization.

Contaminated samples were microtomed to expose their surface for chemical-identification analysis via scanning electron microscopy (SEM) and energy-dispersive spectroscopy (EDS). In some cases in which microtoming was not feasible due to possible cross-contamination or the risk of losing the particles, samples were exposed to SEM under higher voltage—for deeper x-ray penetration—in order to identify the elements. The samples were sputter-coated with palladium to render them conductive for SEM observation. Fracture surfaces of the suspect specimens were also examined by SEM, and the exposed contaminants identified by EDS.

RESULTS AND DISCUSSION

Polyvinyl Chloride Contamination in Polyolefins. The contamination of olefin products by polyvinyl chloride is inevitable in a shared facility where both types of resins are processed. The two main sources of contamination are airborne PVC powder and PVC compound residues in the barrel and in the screen pack of the processing equipment. Airborne PVC particles with no stabilizers char rapidly at elevated temperature during polyolefin processing. The charred PVC particles can manifest themselves as dark specks embedded in the finished products. Even though these brown or black specks are often considered a merely cosmetic defect, their presence is not acceptable in the medical industry because of the perception of poor product quality. To reduce or eliminate airborne PVC powder particles, total segregation of the PVC and olefin processing areas is required.

The second source of PVC contamination arises from residual material in the barrel and in the screen pack of the processing equipment. When a production changeover from a previous PVC run is enacted without a thorough equipment cleaning, cross-contamination can result. Although PVC compounds are generally quite stable, long residence times at processing temperatures can degrade residual PVC to brown or black specks and gels that can appear in polyolefin products such as film and tubing. This mode of contamination is more serious than the specks caused by airborne particles, because the larger-sized gels embedded in relatively thin polyolefin films can bring about mechanical failure. For example, during a blood-component harvesting process, film containers are subjected to high centrifugal forces. Gels are the weakest points in the entire film, and dimpling and fracturing can occur under such conditions. To totally eliminate these kinds of failures, shared machinery should be disassembled and completely cleaned between changeovers. Ideally, PVC and polyolefins should each be run on their own dedicated machinery on separate production lines.

Inadequate Dispersion of Antioxidants. Poor dispersion and distribution of antioxidant in a resin often leads to unpredictable product shelf life. When an ~10-µm-thick, medium-density polyethylene (MDPE) film was subjected to accelerated heat aging at 120°C, inadequate antioxidant dispersion resulted in a pattern of very uneven film embrittlement, as seen in Figure 1. Color changes in the film from clear to white translucent to yellow were observed. It was also very interesting to note that the degradation grew in all directions with a whitish front. The white areas were very brittle, whereas the yellowish areas tended to be less so. The darkest areas were those of yellowish color, and the lightest areas were the white front, marked by a dashed line in Figure 1. Infrared spectra suggested that the yellow and white areas were severely degraded—with the evidence of massive C=O carbonyl contents—as compared with the ductile, translucent area (see Figure 2). This was confirmed by results examining oxidative induction time (OIT), with negligible OIT detected at the yellow and white areas, indicating total degradation (see Figure 3).

MDPE that has undergone chain-scission degradation will show evidence of embrittlement, accompanied by a white, translucent color. Scanning electron micrographs showed no sign of surface crazes, indicating that the whitish area was purely bulk color shift from light scattering. The yellow color was suspected to be the degraded hindered phenol antioxidants forming quinone structures. Because antioxidant was used as a sacrificial free-radical scavenger, it degraded to form quinone structures that were frequently deep yellow.

Partially degraded samples were further aged at 120°C for an additional 160 hours in order to observe the growth of the degradation, the change in the color, and the initiation of white spots. Figures 4 and 5 show this progression: the white front, which was very brittle, grew to a much larger size, and light yellow color started to form, increasing in size from the earliest white area seen in Figure 4. Several additional white translucent degradation spots were also initiated. The uneven initiation was an indication of uneven antioxidant dispersion and distribution.

Inadequate dispersion and distribution of antioxidants can shorten product shelf life unpredictably. The dispersion of various additives—including colorants, antistatic agents, impact modifiers, processing modifiers, fillers, etc.—in different polymers requires the proper selection of machinery and careful attention to process conditions.

Catalyst Residue and Processing-Machinery Transition-Metals Contamination. It is known that the thermal oxidation of polyethylene is often catalyzed by transition metals, presumably through the promotion of hydroperoxide decomposition. Polypropylene is also very susceptible to thermal oxidation, even at ambient temperatures, so much so that it always comes with an antioxidant additive package. Catalyst residues from polymerization—for example, titanium—and the presence of transition-metal impurities are frequently the accelerators for hydroperoxide thermal decomposition in polypropylene. Transition-metal contamination can come from different sources, including tools, machinery wear, contaminated additive packages, and rust transfer from autoclave equipment. Figures 6 and 7 show the evidence of metal contamination in MDPE film. Many yellow spots of approximately 1 mm or smaller were seen. High-voltage SEM and EDS analyses identified these yellow spots with dark-yellow core centers as iron, copper, and other organic compounds (see Figure 8). Since transition metals are prooxidants, they initiate the MDPE autocatalytic oxidation.

It is interesting to observe that the fractal pattern of the degraded material flows away from the center when aged at 140°C (see Figure 9). From the flow pattern, it is reasonable to assume that the center is the initiation site, since the low-viscosity degradation products aggregated into fingers while the reaction progressed radically outward. The center yellow-spot contaminant appears to be prooxidant and responsible for the heterogeneous initiation of the degradation. This type of contamination is often ignored because the adverse effects do not become evident in the short term. However, depending on the product's shelf-life requirements, the long-term effects of such contamination need to be considered.
 

Embedded Particulate Contamination in Rigid Components. Particles embedded in a rigid plastic can severely reduce the ductility of the material. We have studied the effect of contamination in a polycarbonate tensile specimen, the elongation-to-break of which before and after steam sterilization can be seen in Table I.

Table I. Effect of contamination in polycarbonate tensile specimen.

It is well known that steam sterilization can cause a loss in ductility in polycarbonate materials. This occurs because the amorphous polycarbonate resin undergoes free-volume relaxation at autoclaving temperatures, and elongation is a measure of ductility. Table I shows that contaminants affect the lower-molecular-weight polycarbonate dramatically. The authors found it very surprising that particles as large as 1 mm could be seen on the fracture surface. For example, Figure 10 shows the fracture surface of a tensile specimen containing a particle ~0.4 mm in size; stress concentration had caused the specimen to break in a brittle manner. Contaminants, identified by microscope IR, included a wide range of materials, among them polyethylene, PVC, nylon, aluminum fragments, chromium titanium, silicone, sulfur, and so on. These specks could have been introduced from the room environment or during any of the manufacturing processes, including polymerization, compounding, pigment formulation, material handling, and injection molding.

CONCLUSION

Plastics failure from heterogeneous contamination was examined. The contaminants were from degraded PVC, inadequate dispersion and distribution of antioxidants, transition-metal prooxidants, and particle inclusions in rigid plastic. PVC contamination is normally purely cosmetic, though it can cause problems if large gels are included in thin plastic films. Poor dispersion and distribution of antioxidants can often shorten in an unpredictable manner the shelf life of a finished plastic component. Transition-metal prooxidants catalyze thermal degradation and also photodegradation, and are the most deleterious species for many plastics. An inclusion—whether a rigid or soft particle—acts as a stress concentrator for rigid plastics. Depending on the size and shape of the particle, the plastic specimen or part could reach a critical stress and fail prematurely.

BIBLIOGRAPHY

Barr-Kumarakulasingshe SA, "Modeling the Thermal Oxidative Degradation Kinetics of Polyethylene Film Containing Metal Pro-oxidants," Polymer, 35(5):995, 1994.

Encyclopedia of Polymer Science and Engineering: Degradation, vol 4, Kroschwitz J (ed), New York, Wiley, 4:630—696, 1986.

Kudoh H, "Application of Target Theory for the Radiation Degradation of Mechanical Properties of Polymer Materials," J Mat Sci Letters, 15:666—669, 1996.

Plastic Additives and Modifiers Handbook, Edenbaum J (ed), New York, Van Nostrand Reinhold, 1992.

All of the authors are current or former employees of Baxter Healthcare Corp. (Round Lake, IL). Michael T. K. Ling is an engineering specialist concentrating on medical product design and troubleshooting and on developing applications for polymeric materials. Stanley P. Westphal, now retired, was an engineering specialist in polymer morphology and rheology. Chuan Qin, PhD, is an engineering specialist involved in biomedical process development for devices and drug-delivery systems. A senior engineering specialist, Samuel Ding, PhD, works in biomedical polymer development. Lecon Woo, PhD, is a Baxter distinguished scientist, specializing in biomedical polymer development and polymer rheology and processing.

Copyright ©1998 Medical Plastics and Biomaterials