5Mechanical Metallurgy and Passivation of Titanium Implant Alloys

The Devil created the interface.

- WOLFGANG PAULI

When the focus is limited to titanium systems only, very limited information is presented on the morphology of the corrosion products, especially those generated in saline environments similar to that of the human body. The typical quantitative values reported are particle size distributions and the titanium, aluminum, and vanadium ion concentrations expressed in terms of µg/g of retrieved synovial/joint tissue. Other information using potentiostatic measurement techniques state that titanium alloys do not corrode under 'normal' implant conditions. All sources indicate that some sort of depassivation is necessary for any corrosion to occur, thus an exposure to extremely low pH or wear conditions is implicated. Alternatively, other factors, such as the role of hydrogen on titanium's metallurgy, may have been considered to have minimal effect (or even to have been ignored). When these other factors are considered, the story may indeed change...

The Passivation Layer

The passivation layer on metallic surfaces is a thin layer of oxide that forms to varying degrees (depending on the magnitude of the free energy of formation of the metallic oxide and the availability of oxygen or other species at the surface). This oxide layer can serve to greatly reduce the transport of corrosive species to the underlying metal's surface. Freshly exposed metallic surfaces will adsorb and react with oxygen present in the atmosphere almost instantaneously. One exception to this phenomenon is gold. Gold's oxide is unstable, as indicated by the negative value for its oxide's heat of formation. Table 5 provides a list of the heats of formation of some metal oxides in decreasing order.

Metals at the top of the table form oxides quite readily while those at the bottom form passive layers much more slowly (or, in the case of gold, not at all). The tenacity of the passive layer is also higher for the metals at the top of the table. However, in the presence of mechanical abrasion, even the most tenacious passive film can be breached. Once this occurs, the surrounding chemistry (i.e., the availability of free oxygen) and the nature of the metal-oxide bond will determine the rate at which the protective layer is repaired. If an insufficient amount of oxygen is available, the layer may remain damaged and corrosion is likely to occur. A more detailed discussion of the behavior of metals in a corrosive environment will be presented in the later section on The Service Environment for the Implant.

Commercially Pure Titanium

Titanium was discovered in 1790 and is the earth's ninth most abundant element. The Kroll process (developed in 1936) allowed titanium to be produced in commercial quantities. The Kroll process involves the chlorination of the raw ore to produce titanium tetrachloride (TiCl₄). This chloride compound is then reduced with solid magnesium metal in an inert atmosphere to produce MgCl₂ and a porous ingot of titanium (known as a sponge). Iron chloride and residual magnesium chloride are then leached out to purify the ingot, which is then densified. The titanium used in the manufacture of modular implant components is of a nominal purity of 99.0% ASTM Grade 4 with alloy additions as specified in Table 6.

Titanium is an allotropic element. At room temperature, the hexagonal close packed (hcp) structure known as the a phase is thermodynamically stable. When heated to temperatures over 883°C (1621°F), it transforms into the body centered cubic (bcc) b phase. In the annealed condition, it exhibits the mechanical properties given in Table 7. Commercially pure (CP) titanium is often used in medical devices for its relatively easy formability (compared to titanium alloys) and high biocompatibilty. In the case of the stems investigated for this study, the porous surfaces are fabricated from a CP titanium wire mesh pad. The titanium wires are chopped, kinked, and then diffusion bonded into the form of a pad approximately 2 mm thick. The pads are then cut to their final shape to fit recessions on the side and edge of the implant stem. The pads are then hot isostatically pressed (HIP'ed) and diffusion bonded onto the stem surface. Diffusion bonding is preferred over sintering because the temperatures required to sinter porous surfaces onto the stem base material result in very extensive grain growth in the stem, with grains often attaining average volumes of 1 cm3. These large grained components are likely to exhibit insufficient strength when subjected to large or sudden loads.

Ti6Al4V/a + b Alloys

In the 1950's, strengthenable titanium alloys were sought in order to answer a need for materials with high specific strength, low creep, high melting temperature, and high corrosion resistance for aircraft components such as jet engines. Although considered in the early 1950s, alloys of titanium were not used for surgical applications until the 1960's. Now, titanium alloys such as Ti-6Al-4V are beginning to rival stainless steel as well as CoCrMo alloys in terms of implantation uses.

As shown by the Ti-Al phase diagram in Figure 2, the addition of 6 wt% of aluminum to titanium serves to increase the temperature at which the a phase is stable. The a phase is desirable for good strength and toughness and resistance to oxygen contamination at elevated forming temperatures to which this particular type of implant is exposed. Aluminum exhibits a strong solution strengthening effect when added to a titanium matrix, as shown in Figure 3.

Figure 3

Figure 4

The addition of vanadium serves to stabilize the b phase. An addition of only 4 at% of vanadium reduces the a-b transus temperature nearly 200°C, as seen in the Ti-V phase diagram in Figure 4. The bcc b phase allows for better formability at lower temperatures, but is more susceptible to atmospheric contamination The a phase is much more resistant to oxygen contamination and exhibits good strength and toughness. When the a phase is in the platelet Widmänstatten form, it provides excellent fatigue crack arresting properties. A stabilized b phase can be retained upon quenching. In the case of the Ti6Al4V alloy, the b phase can even be retained after a furnace cool. The micrograph presented in Figure 5 shows an a+b Ti6Al4V alloy that was forged and then aged at an intermediate temperature with a subsequent furnace cool to room temperature.

Zirconium and hafnium have extensive solid solubilities in both the alpha and beta phases of titanium. They do not strongly promote phase stability but they retard the rates of transformation and are useful as strengthening agents. Oxygen, nitrogen, and carbon are alpha stabilizers that raise the transformation temperature. As discussed in the next section, hydrogen is a beta stabilizer and lowers the transformation temperature. Niobium is beta isomorphous (bcc) and will not form intermetallic compounds with titanium. The Ti-13Zr-13Nb alloy, mentioned above as a low modulus substitute for Ti6Al4V (Section 3, Overview of the Materials Science of Orthopedic Implant Systems) takes advantage of the solid solubility of these two alloying elements (Nb and Zr) to produce a compound of lower modulus, most likely by 'over-saturating' the titanium matrix beyond the point of maximum solid solution strengthening.

Effect of Hydrogen on Microstructure

Hydrogen in Titanium: Sources of Hydrogen

There are many potential sources for hydrogen to be introduced into the implant during its manufacture and service. Residual hydrogen in the ingot is unavoidable. However, improved refining and furnace techniques have reduced the amount of hydrogen left in the metal after melting. Great care must be taken during the forging operations and subsequent handling prior to an oxidation treatment such that no hydrocarbons (gaseous or liquid) such as methane, or oils from machinery (or even bare hands!) are allowed to contact the bare surface of the component. Though once considered to be a benign treatment, sterilization procedures such as 110°C steam for 15 minutes have been shown to introduce harmful species into the surfaces of biomaterials. In a paper by Keller et al., ratios of carbon, oxygen and nitrogen to titanium were found to increase significantly when analyzed with X-ray Photoelectron Spectroscopy (XPS). Although hydrogen contents were not specifically measured, it is likely that these too are elevated after treatment. In another paper, impurities of fluorine and alkali metals were found on the implant surface using XPS and Secondary Ion Mass Spectroscopy (SIMMS). The titanium implant was thought to be protected from contaminants by keeping it in a titanium box, wrapped in sterile cloth during the autoclaving procedure. The source of the contamination was found to be Na2SiF6 which had been used as an additive to the rinsing water in the final step of the cloth's laundry procedure. The final source of hydrogen for the implant that must be considered is the fluid in the joint cavity into which it is placed.

Hydrogen in Titanium: Kinetics and Thermodynamics

Hydrogen diffuses very quickly in titanium ( Do ª 5 x 10^-3 cm²/sec, Q = 40.2kJ/mol) and grain boundaries have been shown to transport the hydrogen much faster than the bulk microstructure: a fine grained specimen of titanium will absorb over six times more hydrogen than a large grained specimen. In hcp a titanium there are two possible positions available for interstitial atoms: tetrahedral (radius = 0.34Å), and octahedral (radius = 0.62Å). Hydrogen's radius (0.41Å) prevents it from occupying the tetrahedral sites. The low solubility of hydrogen in the a phase is thermodynamically explained as follows: the decrease in free energy associated with the introduction of hydrogen in the octahedral sites is offset by an increase in free energy due to a chemical interaction associated with the large amount of freedom of the hydrogen atoms to vibrate within these sites. The titanium-hydrogen phase diagram (Figure 6) shows this decrease in hydrogen solubility in titanium as temperature is decreased.

The higher solubility of hydrogen in the bcc b phase of titanium alloys (ª 1 wt% or 35 at% max.) is explained by the close match between the site radius (now 0.44Å) and hydrogen's atomic radius, eliminating the free energy increase associated with large atomic hydrogen vibrations. Additions of interstitial oxygen or nitrogen do not influence the amount of hydride phase precipitated in titanium at room temperature. Substitutional aluminum atoms (primarily present in the a phase of the Ti6Al4V alloy) inhibit hydride phase formation.

Hydrogen in Titanium: Effects on Wear Resistance

In research conducted by Xiaoxia et al. concerning the wear behavior of Ti-6Al-4V alloy in an acidic medium, hydrogen was found to play a key role in determining the extent and nature of debris generation. In this paper by Xiaoxia et al. the corrosive wear behavior of Ti-6Al-4V was investigated using a pin on disk apparatus in 1N H₂SO₄ at different applied electrochemical potentials. The quantity of wear debris quadrupled when the wear specimen was polarized from -0.8 V to -1.4 V as shown in Figure 7 from this reference. Forced cathodic polarization of the specimen from -1.0 V to -1.4 V correlates to an increase in hydrogen content by a factor of approximately 2.5. Thus, these authors were able to charge the titanium alloy with hydrogen by simply applying a negative potential in an acidic solution.

The hydrogen content of the jagged wear particle polarized to -1.2V (shown in Figure 8) was approximately 2.3 times higher than that of the ductile wear particle (Figure 9) and 3.5 times greater than that of the original (unpolarized) alloy specimen, as inferred from the results of Secondary Ion Mass Spectrometry (SIMS) (Figure 10). It is crucial to note the difference in the nature of the debris generated at the two different conditions. The high-hydrogen content particle is rough, very thin, and exhibits surface cracking. The low-hydrogen content particle is quite ductile in appearance and exhibits shear banding. The inert gas fusion method (the measurement of the amount of hydrogen out-gassed from the known quantity of metal in an inert gas furnace) was also used to determine the concentration of hydrogen at the surface of the disk and within the wear debris. This method found that polarization of the specimen from -0.48V to -1.2V increased the hydrogen content by a factor of 4.

Figure 8

Figure 9

Figure 10

Other researchers (T.I. Wu and J-K Wu), in their investigation of hydrogen as a surface hardening agent for Ti6Al4V, found that electrolytically charging the alloy in an acidic (or basic) solution led to a microstructural transformation of the acicular b phase to the equiaxed a structure at room temperature. This is significant in that it emphasizes how fast hydrogen can travel in the alloy and how important of an effect it can have upon the alloy's microstructure.

For comparative purposes, Figure 11 shows an electron micrograph of the surface of a Ti6Al4V alloy that has been fretted in air at a very high load and high number of cycles. The severely damaged surface exhibits a grooved, jagged, delaminated appearance indicative of extensive wear.

The Service Environment of the Implant

Description of Fluid in Synovial Region

As described by Wickstrom in Hulbert et al., the body's fluid is an 'angry' environment that encourages the exchange of electrolytes and is thus conducive to corrosion. It is oxygenated at different oxygen partial pressures in different areas. Variation of the oxygen partial pressure is enhanced by the presence of various lymphatic cells that secrete enzymes (thus lowering pH) and absorb oxygen. Figure 12 shows the relationship (measured in vivo) between pO₂ and pH in healthy and diseased synovial fluids. Note how the oxygen partial pressure begins to become less variable as pH is decreased and that at normal pH values there is sufficient oxygen in solution to allow passivation for titanium and cobalt alloys.

Oxygen is brought to all tissues from the bloodstream and ions in chloride solutions rapidly circulate past any given area. The pH of the fluid surrounding a recently implanted prosthesis has been measured to be approximately 5.4, with this value approaching the 'equilibrium' physiological value of 7.35 within approximately ten days. However, the pH of the implant site has been shown to fluctuate to lower values (as low as approximately 4) in cases of loosening or infection.

A localized concentration of dissolved hydrogen is indicated by this reduced pH and is likely to be enhanced by large numbers of crevices in the case of porous coated implants. This increased hydrogen concentration can certainly alter the microstructure and, hence, the wear degradation of titanium implants. As we saw above, titanium has been shown to be highly susceptible to hydrogen absorption and hydride phase formation in the presence of even minor quantities of hydrogen.

The placement of the implant into an acidic solution complicates the issue of hydrogen up-take beyond that of gaseous absorption. In order to understand the extent of protection offered by the passive oxide layer discussed above, potential-pH equilibrium diagrams (also referred to as Pourbaix diagrams for their developer, Marcel Pourbaix) can be consulted that indicate regions of phase stability according to environmental conditions. Pourbaix diagrams consider the effect of pH and applied potential upon metallic species. Solid lines on the diagram indicate a threshold of 10^-6M concentration of metal ions. Thus, enclosed regions denote the phase (or ionic state) of the metal under consideration when immersed in water. The parallel dashed lines indicate the stability region of water and indicate the concentration of oxygen or hydrogen in solution. At the top line, gaseous oxygen is liberated, governed by the reaction

below the lower line, hydrogen is evolved:

Figure 13 shows the Pourbaix diagram for iron and shows that iron exhibits two separate regions of corrosion, a region of passivity, and at the bottom, a region of immunity. The 'Passive' region indicates that corrosion (though possible) is less likely than the formation of a protective oxide layer on the metal surface. The 'Immune' region implies that no corrosion will occur (this is the region sought for cathodic protection of iron). Finally, the two 'Corrosion' regions signify the release of iron ions into solution.

Figure 14 is the Pourbaix diagram for titanium. Some major differences should be noted: In this figure there is only one Corrosion region, and three different favored passive oxide phases. Figure 15 is also a Pourbaix diagram for titanium, however, in this case, the effect of dissolved hydrogen on the metal's behavior is considered. The lower regions in this diagram now indicate that titanium hydride (TiH₂) is the stable phase at all pH values for potentials less than about -0.8V. Essentially, hydrogen serves to shrink the Corrosion region (the triangular shaped region in this figure) while elevating the Immune region.

Figure 15

There is a possibility that piezoelectric potentials generated by the apatite crystals in the bone surrounding the implant may contribute to corrosion or hydrogen uptake of the implant. Strain-related potentials have been recorded for living bone tissue. The potentials released during typical loading conditions, e.g., walking or jumping, have not actually been measured, but recorded values for 0.02 in/min displacement rates (very slow by human loading standards) are -10 to -20 mV for compressive surface strains of 0.0005 to 0.001. It is important to note that for a given peak strain, the generated potential increases in a manner roughly proportional to the square root of the strain rate. When these values are normalized to a more reasonable displacement rate of 10 cm/s, peak generated voltages may reach approximately -140 mV. Thus, it may be possible for the cyclically stressed femur to generate potentials on the order of ±100 mV while in contact with the implant's surface. These potential pulses may help to encourage depassivation. Also, it has been suggested that these pulses may interfere with normal bone regeneration and be a factor in the osteolysis (bone deterioration) frequently observed around the implant.

Surprisingly, there does not seem to be a significant correlation between patient factors and failure occurrence. However, an inverse correlation appears to exist between the amount of use and service life-span of the implant. Unfortunately, mechanical factors alone cannot be implicated in the reduction of the service life of the implant, i.e., failure seems to depend on more than mechanical factors. As discussed below, many failures appear to be heavily influenced by the chemical and biological aspects of the implant's service environment.

Description of Implant Loading Conditions

Immediately after the operation, the patient is in a brace and bandaging that prevents any motion of the limb and joint. There is usually a large amount of swelling and bruising from the surgical damage to the bone and surrounding tissues. During the patient's recovery stage, the swelling subsides and the patient is allowed to apply moderate loads (usually in a swimming pool where buoyancy prevents full loading) to the implant under the supervision of a physical therapist. This recovery period is intended to allow the individual microscopic bone fibers to grow into the porous coating of the implant. Smooth implants rely on the process of bone adhesion directly to the metallic surfaces or PMMA cement's mechanical bonding. Depending on the patient and the time elapsed since the surgery, there may be a small amount of implant-bone micromotion, but this amount is difficult to quantify. In the case of a normal, fully functional implant, normal activities and, hence, full loading of the implant, can be resumed (with caution) by the patient once sufficient bone in-growth has occurred. The design value for the primary applied joint reaction force at the head and neck of the stem is usually taken as five times the patient's body weight. This is thought to be a conservative value, however, many apparently 'gentle' motions of the hip joint can produce a surprisingly large stress build-up and impulse load that are often difficult to predict analytically. Dysfunctional or problematic implant loading can occur when the stem is displaced too much and becomes separated from the bonding bone fibers (or bone cement). Canine studies indicate that very little amounts of micromotion between the stem and bone are required to cause loosening of cementless implants to occur. In this research by Bragdon, et al., micromotions of only 40 µm were sufficient to cause fractures and slippage at the bone-stem interface. Micromotions on the order of 120 µm caused a layer of dense fibrous tissue (known as a 'capsule') to form around the implant. This encapsulation can then cause pain and loosening of the implant and will be discussed as one of the modes of failure for implants.

Modes of Failure

There are many modes of failure that depend upon which hip prosthesis is considered and these are summarized in Table 8. Some failure modes can present themselves independently of other problems, whereas modes such as loosening of the stem often occur in conjunction (or synergistically) with a different mechanism (excessive polyethylene wear, for example).

Corrosion Fatigue

The most catastrophic mode of failure is macroscopic fracture of the stem itself. Due to the high potential for localized corrosive attack in the area of stress, alternating stresses can fatigue crack an implant well within the 'safe' fatigue fracture stress ranges determined under non-corrosive conditions. A subset of the observed implant fatigue failure modes is known as corrosion fatigue. Corrosion fatigue is a common failure mechanism for stainless steel implants, but has been observed in other alloy systems as well. It initiates predominantly in areas that have been subjected to design or surface stresses such as notches, machined grooves, or regions of sharp radii of curvature.

Stem Loosening

Failure can also occur when the stem becomes loose. It may even protrude through the outer wall of the femur if it places too much stress on the inner surface of the cortical (dense) bone of the femur. The first few weeks after the implantation of the hip replacement are critical. If the stem is loaded too soon, any bone-stem bonding that may have formed can be lost. Further loading will only serve to cause more extensive loosening and tissue damage. Loosening can also be caused by biological reaction (most likely to polyethylene or metallic particles surrounding the implant) where a 'capsule' of fibrous tissue made up of proteins and inflammatory cells develops around the implant and isolates it from the bone. This capsule can encourage osteolysis or osteonecrosis (bone loss or bone death) which further enlarge the cavity intended to snugly support the implant, leading to even further loosening. The stem can also loosen when the surrounding bone atrophies (commonly referred to as demineralization) from disuse. Often, this demineralization can be caused by a mechanism known to orthopedic researchers as 'Stress Shielding.' Stress shielding can occur when the metallic stem bears more of the patient's weight than that which is ideal for bone health and maintenance. Indeed, stress is essential for minerals to deposit and remain deposited upon the precursor protein fibers that serve as bone's scaffolding structure. Mother Nature has taken advantage of the piezoelectric nature of bone crystals. Normal physiological stresses upon these hydroxylapatite crystals induce voltages that create a delicate, very local pH range that encourages precipitation of more hydroxylapatite. Polyethylene debris released from the articulating surface has also been implicated in the loosening process owing to its tendency to stimulate bone dissolving cells (osteoclasts) into action.

Excessive Sliding Wear

Figure 16 displays the different physical processes that can occur during sliding wear. Figure 16(a) shows welded junctions that can form on clean surfaces that lead to material transfer. Although this has been reported for some implant systems, it appears to be an infrequent wear mechanism for modular implant systems. When a hard counterbody repeatedly wears against a ductile surface, sheet-like wear particles are formed. This mechanism is shown in Figure 16(b). Polyethylene and titanium, when coupled with harder CoCrMo alloys often exhibit this form of debris. Figure 16(c) shows the wear behavior when surface traction in sliding contact leads to cracking of brittle materials such as ceramics. Figure 16(d) is very important in that it shows how the cracking of a brittle surface layer can result in loose debris particles that later act as abrasives. These wear debris generation mechanisms can manifest themselves at the ball/acetabular socket couple or stem surface. After many wear cycles, i.e., leg extensions/flexions, portions of the stem can loose their protective passive oxide film, especially in the presence of abrasive particles, not only those generated from wear, but also those present from the manufacturing process. These released particles will contribute to 'three body' wear as shown in Figure 16. If a metallic (or ceramic) particle becomes lodged into the relatively soft polyethylene liner, extensive depassivating scratches can form on the CoCrMo head. Chromia particles released in this manner may attack the lower portion of the stem, although the results of most research indicates that extensive CoCrMo head surface wear is uncommon and that very little Co, Cr, Mo, or Ni (a minor constituent of this alloy) are detected in wear debris.

Figure 16

Extensive research has also been conducted on the concentration of metallic ions near the implant in an attempt to demonstrate the extent of simple corrosion of the implant. One interesting finding obtained from these metallic ion concentration measurements is that ions are often found in proportions that differ significantly from those expected from alloy compositions. Thus, leaching and/or corrosion of the implant is suspected. interestingly, these constituents are found in above normal ionic concentrations in regions as removed from the implant as the liver when any burnishing or fretting damage of the CoCrMo head is observed. This information implies that the elemental and ionic species of this alloy are much more soluble in physiological tissue and fluid than those of Ti6Al4V. This observation can be explained by the extremely high thermodynamic instability of the titanium ion in the body. Concentrations of aluminum and vanadium ions vary considerably and are found to be elevated in certain regions around the implant as shown in Table 9. This table shows that ion concentration not only varies according to patient (the reason for the large scatter) but also by the type of ion and the location in which it is measured.

	Ion Concentrations (µg/l)
Tissue Type	Ti		Al		V
Synovial Fluid	556±882 (13±22)		654±743 (109±158)		62±95 (5±1)
Capsule	1540±1238 (723±1217)		2053±1064 (951±586)		288±133 (122±123)
Fibrous Membrane	20813±26467 (N/A)		10581±9764 (N/A)		1027±702 (N/A)
Blood	67±62 (17±60)		218±233 (13±4)		23±31 (6±4)
	Co	Cr		Mo		Ni
Synovial Fluid	588±427 (5±3)	385±232 (3±4)		58±53 (21±8)		32±16 (5±2)
Capsule	821±451 (25±17)	3329±2890 (133±63)		447±247 (17±8)		5789±2535 (3996±6237)
Fibrous Membrane	2229±1583 (N/A)	12554±8055 (N/A)		1524±1399 (N/A)		13234±10074 (N/A)
Blood	20±25 (0.1-1.2)	110±150 (2-6)		10±4 (0.5-1.8)		29±29 (2.9-7)

Debris can be generated at many locations on the implant surface either by corrosive or bone-implant micromotion induced wear mechanisms collectively termed fretting. This debris can then migrate through the fluid filled environment to other surfaces and disrupt the protective passive film. Once the passive film is disrupted in a tribochemical environment, the fretting-wear process often becomes synergistic. A corrosive environment inhibits repassivation after the oxide film has been mechanically removed. Fretting of the exposed surface leads to more wear particles, which in turn cause further removal of the protective film, causing even more degradation of the surface. Figure 17 shows how titanium responds to passive film disruption in a 0.1 N NaCl solution. Peaks on this plot of potential vs. time correspond to oxide film growth while drops in potential are caused by mechanical disruption of the film. Note how the film restores itself when the fretting motion is stopped (at 70 minutes).

The following specific origins of metallic debris have been postulated:

Articulating Surface Wear

Wear can occur at the ball and socket surfaces of the device that are exposed to frequent relative motion. Older implant designs that used titanium for the head (ball) of the stem (and that had not been surface hardened) were prone to rapid and extensive wear at the articulative polyethylene socket-titanium head interface. More recent designs have incorporated a head of different, harder material, such as CoCr alloys or Al2O3, to increase the wear resistance of the implant surface. One drawback observed with the use of the harder head materials is accelerated wear of the polyethylene liner of the acetabular (hip) socket. Adverse biological reactions at the surface and in the vicinity of the implant are now believed to be caused by sub-micron polyethylene debris particles.

Ball-Stem Taper Joint Crevice Corrosion/Fretting Wear

As shown in Figure 18, there is a high potential for fretting at the press-fit taper joint between the neck of the stem and the ball. This opportunity for fretting wear is facilitated by the abutment of two different materials of different hardness at the points of contact of the couple. A common taper combination is 5° 59' 57" for the stem cone and 5° 53' 32" for the bore of the head. Tolerances for medical tapers are up to 8 times looser than those for the automotive and machine tool industries. Taper joints exposed to both actual and simulated implant loading conditions have shown signs of both fretting wear and corrosion. Collier and other researchers have also noted this phenomenon for cobalt alloy balls on cobalt alloy stems. When in contact, two metals (e.g., CoCrMo and Ti-6Al-4V) of different galvanic potential will establish an anodic-cathodic reaction that can exacerbate the corrosion process, especially in the presence of a crevice (as shown in Figure 18). Although this theory is strongly advocated by Collier, a more recent study (sponsored by the orthopedic company Zimmer, Inc.) that examined the corrosion behavior of nitrided Ti6Al4V/CoCrMo alloy taper joints in saline solution under a 700 lb head load for one million cycles found that only crevice corrosion was evident. No micromotion induced damage to the taper joint was observed. The authors conclude that the nitriding process prevented surface damage/depassivation.

Stem-Bone Micromotion Fretting Wear

Motion of the stem portion of the implant relative to the surrounding compact or cancellous (spongy) femoral bone could be implicated in removing the passive and protective oxide film of the titanium stem. When high resolution surface analysis techniques (such as atomic force microscopy or X-ray microanalysis) are applied to Ti6Al4V implant materials, needle shaped oxide growths are found on the surface (Figure 19). Apatite, the main hard mineral component of bone, possesses a hardness of approximately 35 on the Rockwell 'C' scale or 5 on the Mohs hardness scale where diamond is 10 and talc is 1. It is possible that this hard component of the bone could selectively remove portions of the oxide and underlying metal. Retrieved implants often exhibit 'shiny' regions on their normally matte surfaces where the implant had been in direct contact with the inside of the femur.

Porous Coat-Stem Surface Fracture/Fretting/Corrosion

The sintered porous mesh or bead surface layers used on some stems possess elastic properties that are much lower (due to their lower strength and smaller cross-sectional areas) than those of the substrate material. Hooke's law for a two component composite material (in this case, the implant, consisting of the stem and outer mesh) states that

These loose regions may contribute to depassivation of the underlying surface when they are cyclically strained during the implant's service. Evidence for this mechanism of particle generation is difficult to establish since the entire process would be occurring beneath the sintered pad or beaded surface and has not yet been substantiated in the current literature. The increased surface area of these porous regions coupled with an increased probability of finding defects (such as hydrides or crevices) at the implant's surface defects tends to strongly implicate the porous mesh in the debris generation process. Indeed, many researchers have found that when retrieved porous coated implants are examined, the porous layer had delaminated from the surface and that metallic debris and associated biological inflammation were prevalent in the tissue next to the coating.

Bone Cement-Stem Surface Fretting Wear

This mechanism of debris formation is only operative in implant systems that are bonded to the femur with polymethylmethacrylate bone cement (not the implants considered by this research). The cement relies upon mechanical bonding in order to adhere to the interior of the femur, but relies primarily upon chemical bonding to secure the stem within the femur. If the stem manages to loosen from the (primarily chemical) bond of the cement, it will rub and wear against the fairly brittle cement sleeve. Burnished areas are often noted on implants recovered from loosened cement capsules that are indicative of macroscopic wear.^†

The American Society for Testing and Materials (ASTM) Committee F-4 on Medical and Surgical Materials and Devices has established a task force to develop methods for the characterization of particulate debris. Once these methods have been established, researchers will be able to communicate their findings more uniformly and accurately. To date, very few sources have reported on the morphological aspects of the particulate debris released from Ti6Al4V modular hip joint prostheses.

Next Chapters (Chapters 6-8) Previous Chapter (Chapter 4) Table of Contents

This page is maintained by Prof. Stacy Gleixner.  
Questions or problems please send email to gleixner@email.sjsu.edu or call (408)924-4051. 
The page was last updated 09/27/00 .

Some files on this site are in .pdf format. To read these you will need the Adobe Acrobat Reader.