Originally published March 1997
TECHNICAL PAPER SERIES
Polyvinyl chloride (PVC)based tubing and film are used in numerous medical products. The material is easy to formulate and process, is relatively inexpensive, and performs well in most applications. However, some concerns exist about PVC disposal and the chlorinated by-products of incineration, as evidenced by the list in Table I.1 In addition, there have been reports of the inadequacy of PVC for some drug-delivery applications because of PVC-drug interaction.2,3
Table I. Volatile chlorinated compounds (parts per billion) identified in a PVC incineration study.1
Typical requirements for tubing used in medical devices--such as the intravenous (IV) set shown in Figure 1--include clarity, flexibility, kink resistance, toughness, scratch resistance, ease of bonding with common solvents or adhesives, and suitability for gamma, EtO, or E-beam sterilization. While other thermoplastic polymers have been used to replace flexible PVC in medical tubing, none to date has been able to match the advantages provided by PVC. For example, polyurethane and silicone have been tried, but both are relatively expensive. Nevertheless, there are several commercial offerings of non-PVC medical tubing available on the market. Tubing made by one device company, for instance, employs a polybutadiene-based material that is translucent and kinks easily. Another tubing made by a different company features a three-layer construction, with the outer layer consisting of plasticized PVC and the inner layer of a polyolefin material. As yet, however, no substitute has been widely accepted as a chlorine-free replacement for PVC-based materials.
Figure 1. An example of an intravenous set with the tubing bonded to a variety of IV components.
In the present study, two chlorine-free thermoplastics were examined in an attempt to mimic the key properties of flexible PVC. The research involved blending two classes of polyolefinic resins in order to develop medical tubing that could improve tubing/drug compatibility and offer a product that was both flexible and chlorine-free. The first resin used was a soft, flexible metallocene polyolefin thermoplastic elastomer with excellent clarity. The second resin was a tough, ionic polymer that has good compatibility with the first resin. Results showed performance advantages with the blend tubing when compared with tubing made from the individual resins or from PVC.
Materials. The polymer blends that performed the best in these tubing evaluations were made from a nonionic, metallocene-catalyzed thermoplastic ethylene/butene copolymer, designated as E-1, and either of two ionic copolymers, designated as I-1 and I-2, respectively. Advantages of the metallocene ethylene/butene copolymer compared with a Ziegler polyolefin have been set forth in recent studies by Exxon Corp., which report a reduction in haze and a lower level of extractables with the metallocene polyolefin.4 (The differences are illustrated in Figure 2; extractables identified in the study were mainly low-molecular-weight oligomers.) Ionomer I-1 is a copolymer of ethylene and methacrylic acid. Ionomer I-2 is a terpolymer made of ethylene, butyl acrylate, and methacrylic acid. Both the copolymer and the terpolymer were neutralized with zinc base. Following extrusion, the various tubing was aged at ambient conditions for more than 1 month prior to testing. PVC with di-ethylhexyl phthalate plasticizer (Unichem, manufactured by Colorite Plastics Co., Ridgefield, NJ) was used for comparison.
Figure 2. Comparison of haze and hexane extractables of 1.2-mil ethylene/butene films made with metallocene and Ziegler-Natta catalysts. Extractables comparison shown in weight percent.4
Extrusion. Tubing was extruded in two sizes commonly employed in IV sets, using a 2-in. extruder (Model 200 from Welex Co., Blue Bell, PA). The first size featured an inside diameter (ID) of 0.24 cm and an outside diameter (OD) of 0.35 cm. The second size had an ID of 0.27 cm and an OD of 0.41 cm.
Gamma Irradiation. Tubing samples were gamma irradiated at 25, 50, 75, and 100 kGy, and aged for 1 month prior to testing.
Kink Measurement. All tubing was bent until it kinked. The distance between the two parallel sections of bent tubing was reported as the kink number. The kink point of the tubing was marked and set straight for 30 seconds. The rekink property at the first kink point of the tubing was then measured and reported as the rekink number. The testing device was a "Kink-O-Meter," fabricated in-house and depicted in the schematic in Figure 3.
Figure 3. A schematic of a device used to test the kink property of tubing.
Mechanical Properties. The ultimate tensile strength, elongation, and Young's modulus were tested using an Instron Model-2122 tester. Initial length was set at 2 in., with a head speed of 20 in./min.
Thermal Analysis. The heat stability of the tubing was measured using a thermal mechanical analyzer (TMA) equipped with a 50-mN penetration probe and operated over a temperature sweep from 0° to 100°C at a rate of 10°C/min.
TEM Analysis. Blend uniformity was analyzed using a transmission electron microscope (TEM). The tubing was cross sectioned, and samples from the outer surface, inner surface, and midpoint of the cross section of each blend were examined.
Tables II and III show the kink and mechanical properties of two sizes of blend tubing, each consisting of one of the ionomers and the metallocene polyolefin. Table IV tabulates the same properties for tubing made from the three individual resins.
Table II. Evaluations of extruded tubing with blend of E-1 and I-1 at 75:25 weight-percent ratio.
Table III. Evaluations of extruded tubing with blend of E-1 and I-2 at 70:30 weight-percent ratio.
Table IV. Evaluations of extruded tubing with individual resins.
The blend tubing has much better kink and rekink properties when compared with E-1 tubing or with a stiff tubing made with ionomer I-1. The tubing made from the E-1 resin alone has a poor rekink property, a tacky surface that easily attracts dirt, a strong cold memory, and an unacceptable pullout force when adhesive-bonded to standard components used in IV-set assemblies. Tubing fashioned from the I-1 material is too stiff to be suitable for applications such as IV sets, whereas tubing made from the E-1 and I-1 blend at a 75:25 weight-percent ratio has comparable stiffness to PVC tubing. The blend modulus is approximately 22 MPa, which matches well with the Young's modulus of PVC tubing.
Figure 4. Thermal mechanical analysis shows an increase in temperature stability of the blend tubing compared with the single-polymer tubing made from ionomer resins.
Kink properties of the blend tubing are comparable with some soft versions of ionomer resins, such as I-2, and with PVC. However, the I-2 ionomer has relatively low heat stability, with a softening temperature of approximately 42°C as determined by the TMA. The thermogram in Figure 4 shows that a blend of E-1 and I-2 at a 70:30 weight-percent ratio exhibited a softening temperature of approximately 65°C--almost the same as the softening point of E-1. The result suggests that mixing E-1 and I-2 improves the heat stability of the resulting blend tubing compared with tubing made from ionomer alone. In general, the overall properties of the E-1/I-2 blend, including cost-effectiveness, can be said to be significantly enhanced compared with samples extruded from individual resins such as I-2. A comparison of the rekink property for several types of tubing is shown in Figure 5.
Figure 5. Diameter of curvature at point where tubing kinks in a tubing rekink test. (A limit established by one manufacturer is 2.9-cm diameter.)
Bonding of the tubing to acrylic IV components was done with an ultraviolet lightcurable adhesive. With this adhesive, the blend tubing develops approximately 20% greater bond strength compared with the ethylene/butene copolymer single-resin tubing. Bond strengths for the blend and for the single polymer tubing are shown in Figure 6.
Figure 6. Chlorine-free tubing bond strengths (lb tensile) to an IV-set component. The column at left shows a 20% increase in bond strength with the blend tubing compared with the ethylene/butene copolymers.
Mechanical properties of gamma-irradiated tubing made from the E-1/I-2 blend and from individual resins are presented in Table V. There is little color change for the blend tubing after exposure to various levels of irradiation. The results also show that the mechanical properties of the E-1/I-2 blend--such as Young's modulus, ultimate tensile strength, and elongation--changed only slightly with a higher dose of gamma irradiation. However, in a separate study, the melt index decreased markedly from 2.72 dg/min to 0.05 dg/min for a poly(ethylene butene) copolymer tubing exposed to 50-kGy irradiation.5 Only 24.6% soluble polyolefin was extracted in decahydro naphthalene boiling solvent (at 194.6°C) after the tubing was exposed to a 100-kGy cobalt source. This indicates that there is some grafting and cross-linking occurring in the tubing when it is exposed to higher doses of gamma irradiation. The PVC tubing became brownish after exposure to a 25-kGy or higher dose of cobalt irradiation, suggesting that dehydrohalogenation had taken place in the PVC chains after gamma irradiation.
Table V. Effect of gamma irradiation on various tubing properties.
A cross section of a blend tubing (E-1 and I-2 at a 70:30 weight-percent ratio) was examined with TEM. The micrographs shown in Figure 7 indicate that the blend tubing is more uniform in its middle portion than in its outer or inner layers. This morphology, however, decreases the clarity of the tubing only slightly. Increased extrusion temperatures did not appear to noticeably affect blend tubing clarity.
Figure 7. The morphology of blend tubing as identified by TEM shows the middle section to be more uniformly blended than either the inner or the outer surfaces.
A nitroglycerin adsorption study was conducted using standard IV sets of PVC tubing and IV sets of blend tubing made from chlorine-free polymers. The IV sets were primed with 200-µg/ml concentrations of nitroglycerin in a 0.9% sodium chloride solution. Immediately after priming, the IV sets were loaded into an IV pump and programmed to run at 10 ml/hr. Samples of solution were taken from the outflow of each set at intervals of 0, 0.5, 1, 2, 4, and 6 hours, and tested for nitroglycerin concentration. A comparison with the original solution concentration showed that significantly less adsorption took place with the blend tubing. These data are presented in Figure 8.
Figure 8. A comparison of nitroglycerin adsorption in PVC tubing verses the blend tubing. Nitroglycerin solution was pumped through the tubing at 10ml/hr and samples of the outflow were tested at various times to determine the percent of nitroglycerin adsorbed.
The study demonstrates that tubing made from blends of two types of thermoplastics satisfies various performance requirements for medical applications that tubing made from individual resins fails to do. Specifically, the rekink property of the blend is better than that of the E-1 ethylene/butene tubing, and the temperature stability of the blend is higher as compared with that of the I-2 ionic copolymer. When compared with flexible PVC tubing, the blend tubing offers distinct advantages for clinical applications, such as lower adsorption. Given these performance characteristics and clinical features, the blend provides a viable alternative material for production of a chlorine-free medical tubing.
The authors would like to acknowledge the contribution of S. Larson for drug adsorption studies, J. Tucker for Instron measurement of irradiated samples, and M. Buckett for the TEM study. Support from the medical specialties and infusion therapy departments at 3M is also appreciated.
1. Switzer W, Labscale Incineration Analytical Study of Medical Polymers, Phase 1, Report # 01-5099, San Antonio, TX, Southwest Research Institute, September 1992.
2. Kaul A, et al., "A Review of Possible Toxicity of DEHP," Drug Intell Clin Pharm, 16(9):689692, 1982.
3. Larson S, internal report on evaluation of drug adsorption with PVC and blend tubing, 3M Co., 1995.
4. Speed C, Trudell B, Mehta A, et al., Structure/Property Relationships in EXXPOL Polymers, SPE International Conference No. 7, Houston, February, 1991.
5. Domine J, private communication.
John H. Ko is a product development specialist at 3M Co. (St. Paul, MN), where he is involved in various product development programs for applications that include tubing, film, lenses, adhesives, and specialty coatings. He holds a PhD in physical polymer chemistry from the University of Michigan. Also at 3M, Les Odegaard is a manufacturing engineer specialist with more than 20 years' experience in product and process development for the medical, orthopedic, and infusion therapy businesses.