Originally published July1997
The importance of polymers--and especially thermoplastics--to the medical device industry cannot be overemphasized. According to figures released in 1995 by the American Chemical Society, the production of polymers the previous year rose to a record high of about 75 billion pounds.1,2 Plastics, which now account for approximately 80% of total polymer volume, recorded an annual production of 59 billion pounds. Another recent report states that U.S. demand for medical plastics will increase 3% per year, to more than 3 billion pounds by the year 2000.3 Nearly 80% of the U.S. medical plastic market will be from sales of polyvinyl chloride (PVC), polypropylene (PP), polysulfones (PS), and the various polyethylenes. Projections indicate that PVC will remain the top-volume medical plastic because of the demand for both flexible and rigid products made from the resin and because of the availability of more radiation-sterilizable and environmentally safe grades.
Polymers for medical devices must be safe and mechanically appropriate. Photo: NAMSA
Along with its increased demand for medical-grade plastics, the device industry has intensified its awareness of the need to ensure the safety of materials used. This heightened awareness of the safety of medical devices has taken on an international character largely because of the combination of FDA regulations and International Organization for Standardization (ISO) standards.4 As international standards--including ISO 10993-1 to 10993-13--continue to develop, one commonly hears discussions of the need to meet global biocompatibility requirements. In fact, most manufacturers today prefer to perform their testing only once, in a manner that satisfies regulatory bodies in all the countries in which they intend to sell their products.
The principles applied in testing device materials have altered very little over the past 20 years, but the standards, specific laboratory methods, regulatory attitudes toward testing, and interpretation of results have all changed greatly during this period, as has the availability of new materials. These changes can be expected to continue.
Materials characterization as it relates to biocompatibility is likely to receive considerable attention in the future. Because characterization is so essential, the next revision of ISO 10993-1--being circulated for approval this year--will contain a flowchart similar to the one in FDA's blue book memorandum; in it the first box will instruct the user to characterize the material before proceeding to biological testing.5
The general theme of this article is materials characterization in the design of biocompatibility testing programs for medical materials and devices. Depending on the criticality of the materials, characterization can be accomplished via many methods. For example, characterization may mean determining physical properties, or it may mean confirming major components with infrared analysis or other fingerprinting techniques. It may also include physicochemical tests to determine the weight of nonvolatile residues in an aqueous or alcohol extract prepared under controlled conditions, or tests to identify specific extractables and their relative quantities.
RATIONALE FOR MATERIALS CHARACTERIZATION
When characterizing the biocompatibility of polymers that will contact the human body, two simple but extremely important questions must be asked to guide any testing. The first question, "Is the polymer safe?," should be the initial consideration in a materials-characterization testing program. Nearly every commercial polymer contains additives of various types, including fillers, plasticizers, antioxidants, low-molecular-weight polymers, and even unreacted monomers and solvents. Any of these substances can alter the performance of a given polymer. Adverse effects caused by medical device materials are generally chemical effects, produced by material components, contaminants, or breakdown products that find their way from a device into a patient, resulting in a biological effect. To ensure safety, we need information and testing about potential leachables, which may be toxic or lead to irritation. A materials-characterization testing program must also provide evidence that the polymer properties will not change critically during the life span of the device.
The second question that must be asked is, "Does the material have the necessary physical and mechanical properties for its proposed function?" It is important to match the polymer design with an appropriate intended function, and not force a resin into applications for which it should not be used. If a material does not have the correct properties, a design flaw leading ultimately to product failure or unsafe performance could result. Desired properties might include strength, material hardness or flexibility, surface properties, and even color.
Before an investigator performs any chemical or mechanical analysis, it is very important to have precise information on the synthesis of the polymer itself. Important information includes a description of (1) the monomers used in the polymerization, (2) the solvents used in the synthesis, and (3) any special additives that have been incorporated during production of the polymer. Once this information, if available, has been retrieved, chemical and mechanical testing should be considered.
Over the years, a number of tests have evolved for the characterization of medical device materials. These tests are relevant, sensitive, rapid, and inexpensive, yet provide extremely valuable information to establish material safety. As described below, this group of tests comprises infrared analysis, aqueous physicochemical tests, alcohol physicochemical tests, high-performance liquid and gas chromatography, and atomic absorption spectroscopy.
Infrared Analysis. Infrared instruments measure the vibrational spectrum of a sample by passing infrared radiation through the material and recording which wavelengths have been absorbed, and to what extent. Since the amount of energy absorbed is a function of the number of molecules present, the infrared instru-ment provides both qualitative and semiquantitative information.6
Because the infrared spectrum of a chemical compound or polymer is perhaps its most characteristic physical property, infrared analysis finds extensive application in fingerprinting or identifying materials. Small differences in the structure and composition of a molecule result in significant changes in the distribution of absorbance peaks in the fingerprint region of the spectrum. By matching the infrared spectrum of an unknown substance with that of a known material, proof of identity can be established. A library of spectra of the materials most frequently encountered can be accumulated, or reference spectra available commercially from various sources can be purchased.
Physicochemical Tests. Possible extractables from plastics are quite varied and include such chemicals as base polymers, fillers, lubricants, plasticizers, antioxidants, pigments, and slip agents. In pharmaceuticals, for example, any of these substances can affect the purity of a drug solution; reduce its potency; cause turbidity, precipitates, and particles; or potentially increase toxicity. The manufacturing process itself may further modify polymeric materials through the formation of reaction products or degradants. Many of these chemicals are water soluble, while others are soluble in more nonpolar environments. For these reasons, it is prudent to design tests that characterize both aqueous and nonaqueous extractables.
Aqueous Physicochemical Tests. The United States Pharmacopeia (USP) describes a group of tests used for characterizing the plastic components of pharmaceutical containers and medical devices.7 Commonly called "aqueous physicochemical tests," these procedures are designed to determine the presence of water soluble substances, without regard to their identity. The value of these tests lies in ensuring that the materials examined will not release water soluble chemicals into the drug products or tissue fluids that they contact. The tests can also detect surface contaminants that may find their way into a lot of raw material as it undergoes various manufacturing processes.
Procedures using aqueous extracts derived from a sample material can determine values for nonvolatile residue (chemicals extracted from the test sample or formed by degradation of the test sample during extraction), residue on ignition (inorganic residue from the sample), buffering capacity (estimated strength of the buffering effect caused by extractable chemicals that alter the pH of the extract), and heavy metals. The results of these tests are presented as a set of four values showing the results for nonvolatile residue, residue on ignition, buffering capacity, and heavy metals together with the corresponding USP limits (see Table I).
Table I. Results of aqueous extraction physicochemical testing on polymers commonly used in the manufacture of medical devices. Typical device materials contain very few water-soluble extractables. With reference to these tests, USP limits can be used to establish specifications for raw materials.
The aqueous extract tests are intended to serve as the basis for specifications, and the USP has established limits for each of the four areas tested. Products that fit the classifications of USP chapter 661 must be composed of materials that meet the requirements of the tests both initially and throughout their product lifetimes. Aqueous physicochemical tests can also be used to evaluate raw materials from lot to lot or from different vendors, thereby ensuring chemical equivalence of starting materials.
Nonaqueous Physicochemical Tests. The USP also prescribes the use of isopropyl alcohol (IPA) for conducting physicochemical tests of elastomeric closures used for pharmaceutical containers.8 IPA is a more aggressive extraction fluid than water, and is capable of dissolving a range of chemicals that are insoluble in water.
Nonaqueous extract tests are performed using an alcohol extract, which is analyzed for nonvolatile residue and residue on ignition. Because IPA extracts do not lend themselves to analysis of heavy metals or buffering capacity, these tests are omitted. Instead, turbidity and ultraviolet absorption tests are included as simple procedures for detecting the presence of extractables without specifically identifying their chemical makeup. Results of the alcohol physicochemical tests are presented as a set of five values for each of the end points (see Table II). USP limits do not yet exist for these tests, but they are not necessary for establishing specifications for the acceptance of materials.
Table II. Results of alcohol extraction physicochemical testing on polymers commonly used for medical devices. Isopropyl alcohol (IPA) is a more aggressive extraction medium than water. Most materials used for devices show measurable amounts of extractables when tested with IPA. No USP limits exist to establish an acceptable level of extractables for materials tested by this method.
Chromatographic Characterization. The various chromatographic methods--including gas, liquid, paper, and thin-layer processes--have become indispensable aids in the isolation, separation, and identification of chemicals. Gas and liquid chromatography (especially high-performance liquid chromatography [HPLC]) are powerful analytical tools used to characterize additives in polymeric materials employed in a wide range of medical devices.9
In gas chromatography, the sample is vaporized and injected onto the head of a chromatographic column. Elution is brought about by the flow of an inert gaseous mobile phase. In contrast to what occurs in most other types of chromatography, the mobile phase does not interact with the molecules of the analyte; its only function is to transport those molecules through the packing. Gas chromatography can be used to separate and quantitate both volatile and semivolatile chemicals found in polymeric materials.
High-performance liquid chromatography is undeniably one of the fastest growing and most useful of all analytical separation techniques. Its popularity can be attributed to the sensitivity of the method, its ready adaptability to accurate quantitative determinations, and its suitability for separating nonvolatile or thermally fragile species. Flexible and widely applicable, liquid chromatography can be used for the analysis of more than 80% of all known organic compounds.9
Both gas and liquid chromatography provide an excellent way to characterize the additives, residual monomers, and degradation products from polymers used in medical devices. In addition, the aqueous and nonaqueous extracts obtained from the physicochemical tests discussed above can be subjected to chromatographic analysis to fingerprint the extractables, providing a chromatographic tracing of the chemicals present in the extract. Though not required, it is often desirable to identify all components of these extracts in order to use the information as the basis for establishing specifications for materials used in medical devices. Once established, these chromatographic specifications can be used to evaluate device products throughout their lifetimes. In practice, a copy of this "extractable fingerprint" is used as a reference, and future lots are compared with it to confirm that there have been no changes within the ability of the method to detect them (see Figure 1). Though the extract-preparation technique may vary for different samples, nearly any polymeric material can be subjected to extractable fingerprinting by chromatography.
Figure 1. Chromatographic analysis of PVC additives.
Trace-Metal Analysis. The USP provides guidance for analyzing extracts of plastics and elastomers for the presence of any heavy metal in excess of one part per million (ppm) when compared with a lead standard. This procedure is part of the USP physicochemical test monograph and is a key test in the qualification of a polymer for medical device use; however, the test is nonspecific.7,8
Polymeric materials can contain many chemical substances that are deliberately added in order to obtain desired physical, chemical, or mechanical properties important to a finished device. Additives used as stabilizers might include calcium and zinc. Pigments that contain metals such as lead, tin, arsenic, nickel, cadmium, barium, aluminum, titanium, and iron are frequently employed. Barium oxide is often added to plastic catheter tubing as a radiopaque agent. Still another example is the use of aluminum oxide as a polishing compound during the manufacture of intraocular lenses. This brief list merely illustrates how varied and common are the inorganic chemicals frequently found in polymers.
Except for the above-mentioned nonspecific USP physicochemical test--which is a 1-ppm-limit test for total heavy metals--tests for trace metals are typically specific for one known metal. Atomic absorption spectroscopy has worked well for
this type of test: the metal content can be assayed in an extract of the material or in the material itself through ashing followed by extraction. Passing the extract through a flame causes the metal to plasmatize and give off energy at a characteristic wavelength that can be detected with a lamp detector appropriate for the metal being analyzed. If other elements are also being analyzed, a different lamp must be used for each. With the atomic absorption method, detection is largely limited to ppm levels.
Mechanical testing methods are employed to evaluate materials under a variety of loading conditions (stresses and strains), and they are useful in characterizing materials upon selection and receipt, in analyzing failures and other problems, and in validating manufacturing processes. Mechanical properties can be determined for all types of materials found in medical devices, including metals, ceramics, polymers, and composites. Numerous procedures are used to evaluate mechanical properties; the following discussion will merely highlight some of the more important concepts and available test methods.
Hardness Testing. Hardness is generally defined as resistance to indentation, as scratch resistance, or as rebound resilience. ISO standards report three methods for measuring hardness: tests for Shore hardness, ball indentation, or Rockwell hardness.10,11 It is important to empha-size that hardness values obtained from one method generally cannot be compared with those derived from another, although data can be empirically compared. Generally, hardness is considered as an end-performance property of material employed in a device. The instrument used to obtain hardness measurements is called a durometer.
Stress-Strain Properties and Testing. Mechanical testing that measures stress/strain relationships is used to characterize a material's behavior and performance. Stress-strain behavior greatly depends on temperature, strain rates, and environmental conditions. The strength of a material is expressed by the ultimate values of stress, which can be found by loading a sample until fatigue. Loading can be in tension, compression, flexion, or shear. Both tensile and compression testing are based on the application of forces normal to the plane on which they act. In shear stress testing, however, the forces applied are parallel to the plane.
Tension. The most frequently applied stress-strain measurement is made in tension (stretching of the material specimen), most often using an Instron tensile tester (Instron Corp., Canton, MA). This instrument is essentially a device in which a sample is clamped between grips or jaws that are pulled apart at constant stress rates to determine a variety of parameters such as elongation, elongation at break, breaking strength, and tensile modulus of elasticity.12 The tensile modulus is defined as the ratio of stress to strain, and is derived from the initial slope of the stress-strain curve.
Figure 2. Stress/strain curve, where op is the modulus line, et is the tensile strength, and oe is the elongation.
In the stress-strain curve shown in Figure 2, op represents the modulus line, et is the tensile strength, and oe is the tensile elongation. The modulus of a material is a measure of a specimen's ability to resist deformation (tensile modulus is also referred to as Young's modulus). Some plastic materials are weakened by the process of gamma sterilization, which causes a decrease in the tensile strength of the material. Tensile-strength data collected before and after sterilization can help predict any changes in the strength of a material following gamma processing. Tensile-strength data should also be gathered before and after accelerated-aging studies to help determine or predict the shelf life of a product. Finally, the tensile strength of package seals can be used to evaluate package integrity, to ensure that product sterility will not be compromised.
Compression. In compression testing, the material sample is crushed rather than stretched. Compressive forces are generally significantly higher than the tensile forces generated in a tensile test. The compressive strength of a material is calculated as the pounds per square inch (psi) required to rupture the specimen or deform it to a given percentage of its height, and can be expressed in psi either at rupture or at a given percentage of deformation.12
Knowing the compressive strength of a plastic is of limited design value, since most plastic materials do not fail from compressive loading alone. However, the information may be useful in specifications for distinguishing among different grades of a material, or for assessing the overall strengths of different kinds of plastic materials.
Shear. Shear strength is particularly important for film and sheet products, in which failures from shear loading often occur. Sheer-strength testing is described in ASTM D 732 and has found wide acceptance as an effective materials-characterization procedure for plastics.13 Typically, the specimen is mounted in a punch-type shear fixture and the punch pushed down at a rate of 0.05 in./min until the moving portion of the sample clears the stationary portion. Shear strength is calculated as the force divided by the area sheared. The shear modulus is frequently called the modulus of rigidity.
Flexure. The final stress-strain test that should be considered is that of flexural strength or bending. In a bending test, a sample in the form of a rectangular beam is subjected to both tensile and compression stresses--from maximum tensile stress on one surface to maximum compressive stress on the opposite surface. In other words, the top surface of the specimen is in tension while the bottom surface is in an equal compression; if the tensile and compressive moduli are equivalent, then the stress at the midpoint of thickness is zero. Typically, the specimen is placed on two supports and a load is applied in the center at a specified rate. The loading at failure, in psi, is the flexural strength.12
For most thermoplastics--which do not break in a bending test, even if greatly deflected--the flexural strength cannot be calculated. In this situation, stress at 5% strain is calculated, which is the loading (force) in psi necessary to stretch the outer surface by 5%. A modulus value in flexure can also be obtained with this test. In addition to characterizing polymer bending properties, flexure testing can be used to evaluate the effects of aging and sterilization on a product. For example, radiation can have a dramatic effect on the bending or flexure strength of polypropylene14 (see Figure 3).
Figure 3. The irradiation effect on the bending angle of polypropylene.
Thermal-analysis techniques measure the response of a polymer to controlled heating processes, and they are widely used in the development and characterization of plastics, elastomers, and waxes. For example, characterization of polymer glass-transition temperature, melting point, and extent of crystallinity is important and often used to help produce materials with physical properties tailored to a product's ultimate applications.15
The primary thermal analysis techniques for certifying product quality are differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). Testing can be conducted to ascertain melting point, degree of crystallinity, and glass-transition temperature, or for component quantification. For some materials--such as crystalline polymers and certain organic chemicals--DSC is used to measure melting points and degree of crystallinity. For amorphous polymers, rubbers, and cross-linked thermoset materials, DSC also provides a fast and accurate measure of the glass-transition temperature or the degree of cure. TGA is widely employed to separate and quantitate the components in a mixture. Both DSC and TGA are effective in measuring decomposition of materials.
As mentioned, DSC results offer rapid measures of critical thermal characteristics such as polymer melting point or glass-transition temperature. The particularities of these transitions--in addition to identifying characteristics unique to each polymer--can also provide information about a material's phase structure, thermal history, and purity. The accelerated-life testing of plastics is another area in which thermal analysis has proven beneficial.15 Fairly accurate predictions of the use-life of a number of plastics are now possible when methods such as DSC and TGA are applied to this very important materials-characterization problem.
Successful manufacturing of medical devices requires both the careful selection of materials and consistent monitoring of the processes used to make them. Comprehensive material characterization of polymers should provide sufficient information to allow evaluation of the potential success of a plastic medical device, at least with regard to its chemical and mechanical properties. By using a combination of chemical and mechanical analysis techniques, designers can optimize both materials and manufacturing processes to ensure a safe and effective product.
1. Reisch MS, "Thermoplastic Elastomers Target Rubber and Plastics Market," Chem Eng News, 74(32): 1014, 1996.
2. Thayer A, "Plastic Shipments Outpace Production," Chem Eng News, 74(40):1011, 1996.
3. "Special Report: Medical Plastics," Plastics Eng, LII(10):55, 1996.
4. International Standard ISO 10993-1 to 13, Geneva, International Organization for Standardization.
5. Tripartite Biocompatibility Guidance, General Program Memorandum #87-1, Rockville, MD, FDA, 1987.
6. Sibilia JP, "Molecular Spectroscopy," in A Guide to Materials Characterization and Chemical Analysis, Sibilia JP (ed), New York, VCH Publishers, pp 1340, 1988.
7. "Containers," United States Pharma-copeia XXIII, chapter 661.
8. United States Pharmacopeia XXIII, chapter 381.
9. Lobo H, Bonilla J, and Riley DW, "Plastics Analysis--Improved Characterization of Polymer Behavior and Composition," Plastics Eng, LII(11): 2731, 1996.
10. ASTM D 2238, D 2583, in Annual Book of ASTM Standards, 08.02--Plastics (II), Philadelphia, American Society for Testing and Materials, 1991.
11. International Standard ISO 868, Geneva, International Organization for Standardization, 1985.
12. Palley I, and Signorelli AJ, "Physical Testing," in A Guide to Materials Characterization and Chemical Analysis, Sibilia JP (ed), New York, VCH Publishers, pp 273284, 1988.
13. ASTM D 732, in Annual Book of ASTM Standards, 08.01--Plastics (I), Philadelphia, American Society of Testing and Materials, 1991.
14. Bigg DM, in Polypropylene: Structure, Blends, and Composites, vol 3, Karger-Kocsis J (ed), London, Chapman & Hill, p 263, 1995.
15. DiVito MP, Fielder KJ, Curran GH, et al., "Recent Advances in Routine Thermal Analysis Instrumentation," Am Lab, 24(1):3036, 1992.
David E. Albert, PhD, is manager, chemistry at North American Science Associates (NAmSA, Northwood, OH), where he assists medical device manufacturers and biomaterial suppliers in designing chemical-characterization testing programs for their products. He also prepares protocols for special studies and develops new chemical characterization tests. His primary expertise is in the areas of pharmacology, biochemistry, pathophysiology, and general chemistry. He is currently an adjunct professor in the College of Pharmacy at the University of Toledo.