CHAPTER 5

METALLIC IMPLANT
MATERIALS

Metals have been used in various forms as implants. The first metal developed
specifically for human use was the "Sherman Vanadium Steel," which was used
to manufacture bone fracture plates and screws. Most metals such as Fe, Cr, Co,
Ni, Ti, Ta, Mo, and W used for manufacturing implants can be tolerated by the
body in minute amounts. Sometimes those metallic elements, in naturally occur-
ring forms, are essential in cell functions (Fe) or synthesis of a vitamin BJ2 (Co),
but cannot be tolerated in large amounts in the body. The biocompatibility of
the implant metals is of considerable concern because they can corrode in the
hostile body environment. The consequence of corrosion is loss of material, which
will weaken the implant, and probably more important, the corrosion products
escape into the tissue resulting in undesirable effects. In this chapter we study
the composition-structure-property relationship of metals and alloys used for
implant fabrications.

5.1. STAINLESS STEELS

The first stainless steel used for implant materials was the 18-8 (type 302 in
modern classification), which is stronger than the vanadium steel and more
resistant to corrosion. Vanadium steel is no longer used in implants since its
corrosion resistance is inadequate as discussed in Section 5.6. Later, 18-8sMo
stainless steel was introduced, which contains molybdenum to improve the
corrosion resistance in salt water. This alloy became known as type 316 stainless
steel. In the 1950s the carbon content of 316 stainless steel was reduced from
0.08 wt% to 0.03 wt% maximum for better corrosion resistance in chloride
solution; this became known as 316L.

79
80 CHAPTER 5

5.1.1. Types and Compositions of Stainless Steels

Chromium is a major component of corrosion-resistant stainless steel. The

minimum effective concentration of chromium is 11 wt%. The chromium is a
reactive element but it and its alloys can be passivated to give an excellent
corrosion resistance.
The austenitic stainless steels, especially types 316 and 316L, are most widely
used for implants. These are not hardenable by heat treatment but can be hardened
by cold-working. This group of stainless steels is nonmagnetic and possesses
better corrosion resistance than any others. The inclusion of molybdenum
enhances resistance to pitting corrosion in salt water. The ASTM (American
Society for Testing and Materials) recommends type 316L rather than 316 for
implant fabrication. The specifications for 316 and 316L stainless steels are given
in Table 5-1.
The nickel serves to stabilize the austenitic phase at room temperature and,
in addition, to enhance corrosion resistance. The austenitic phase stability can
be influenced by both the Ni and Cr contents as shown in Figure 5-1 for 0.10 wt%
carbon stainless steels.

5.1.2. Properties of Stainless Steel

Table 5-2 gives the mechanical properties of 316 and 316L stainless steels.
As can be noted, a wide range of properties can be obtained depending on the
heat treatment (to obtain softer materials) or cold-working (for greater strength
and hardness). The designer must be careful when selecting materials of this
type. Even the type 316L stainless steels may corrode inside the body under
certain circumstances such as in a highly stressed and oxygen-depleted region.

Table 5-1. Compositions of 316 and 316L Stainless

Steels"

Composition (wt%)

Element Grade 1 Grade 2

Carbon 0.08 max 0.03 max

Manganese 2.00 max 2.00 max
Phosphorus h 0.03 max 0.03 max
Sulfur 0.03 max 0.03 max
Silicon 0.75 max 0.75 max
Chromium 17.00-20.00 17.00-20.00
Nickel 12.00-14.00 12.00-14.00
Molybdenum 2.00-4.00 2.00-4.00

U From Annual Book of ASTM Standards, Part 46, American Society for
Testing and Materials, Philadelphia, 1980, p.578.
h Slight variations are given (0.025 max) for special quality stainless steels
(F318 and F139 of ASTM).
METALLIC IMPLANT MATERIALS 81

Ni-Cr
25

20 \ I

Ni(w/ol!5 1\ Austenite

10
'\ /
Ferritic or ""'- V
intermediate struCiur'i(
5

o
o 5 10 15 20 25
Cr(w/o)

Figure 5-1. The effect of Ni and Cr contents on the austenitic phase of stainless steels containing
0.1 wt% C. From F. H. Keating, Chromium-Nickel Austenitic Steels, Butterworths, London, 1956.

They are, however, suitable to use in temporary devices such as fracture plates,
screws, and hip nails.

5.1.3. Manufacturing of Implants Using Stainless Steel

The austenitic stainless steels work-harden very rapidly as shown in

Figure 5-2 and therefore cannot be cold-worked without intermediate heat treat-
ments. The heat treatments should not induce, however, the formation of
chromium carbide (CCr4) in the grain boundaries; that may cause corrosion. For
the same reason, the austenitic stainless steel implants are not usually welded.
The distortion of components by the heat treatments can occur but this
problem can be solved easily by controlling the uniformity of heating. Another
undesirable effect of the heat treatment is the formation of surface oxide scales,
which have to be removed either chemically (acid) or mechanically (sandblasting).

Table 5-2. Mechanical Properties of Stainless Steel Surgical Implants"

Ultimate tensile Yield strength Elongation 2 Rockwell

strength, min, (0.2% offset), min, in. (50.8 mm), hardness,
Condition psi (MPa) psi (MPa) min, 0/0 max

Grade 1 (type 316)

Annealed 75,000 (515) 30,000 (205) 40 95 HRB
Cold-finished 90,000 (620) 45,000 (310) 35
Cold-worked 125,000 (860) 100,000 (690) 12 300-350
Grade 2 (type 316L)
Annealed 73,000 (505) 28,000 (195) 40 95 HRB
Cold-finished 88,000 (605) 43,000 (295) 35
Cold-worked 125,000 (860) 100,000 (690) 12

a From Annual Book of ASTM Standards, Part 46, American Society for Testing and Materials, Philadelphia,
1980, p. 579.
82 CHAPTER 5

14oo.-------------------------------,
Type. 316L. 317L. 318

1200

1000
STRESS
(MPa)
800
100

600 80
STRAIN
60 ("to)
400 ,
,
40
200
""""=> 20
.................. _-- ______~~~?!'_9Q'ion
O~~--~--~--~~~~--~--~--JO
o 20 40 60 80
"toCOLD REDUCTION

Figure 5-2. Effect of cold-work on the yield and ultimate tensile strength of 18-8 stainless steel.
From Source Book on Industrial Alloy and Engineering Data, American Society for Metals, Metal
Park, Ohio, 1978, p. 223.

After the scales are removed, the surface of the component is polished to a mirror
or mat finish. The surface is then cleaned, degreased, and passivated in nitric
acid (ASTM Standard F86). The component is washed and cleaned again before
packaging and sterilizing.

Example 5-1

Calculate the amount of the volume change when iron is oxidized to FeD (p = 5.95 g/ cm 3 ).
The density of Fe is 7.87 g/cm 3 .

Answer

Since the molecular weight of Fe is 55.85 g/mol

55.87 g/mol 3
= 7.1 cm /mol
-----.::::......----0-
3
7.87 g/cm

The molecular weight of FeD is 71.85 g/mol, hence

71.85 g/mol
= 12.08 cm 3 / mol
--:::......----0-
3
5.95 g/cm

Therefore, a v = (12.08 - 7.1)/7.1 = 0.7 (70% volume increase by oxidation). This increase
in volume by oxidation causes the oxides formed to be very porous. Continued oxidation
can then take place via further diffusion of oxygen to the underlying metal.
METALLIC IMPLANT MATERIALS 83

5.2. Co-BASED ALLOYS

These materials are usually referred to as cobalt-chromium alloys. There

are basically two types; one is the CoCrMo alloy, which is usually used to cast
a product, and the other is the CoNiCrMo alloy, which is usually wrought by
(hot) forging. The castable CoCrMo alloy has been in use for many decades in
dentistry and recently in making artificial joints. The wrought CoNiCrMo alloy
is a newcomer now used for making the stems of prostheses for heavily loaded
joints such as the knee and hip.

5.2.1. Types and Compositions of Co-Based Alloys

ASTM lists four types of Co-based alloys that are recommended for surgical
implant applications: (1) cast CoCrMo alloy (F76), (2) wrought CoCrWNi alloy
(F90), (3) wrought CoNiCrMo alloy (F562), and (4) wrought CoNiCrMoWFe
alloy (F563). The chemical compositions of the first three types are summarized
in Table 5-3. At the present time, only two of the four alloys are used extensively
in implant fabrications, the castable CoCrMo and the wrought CoNiCrMo alloy.
As can be seen in Table 5-3, the compositions of the alloys are quite different.

5.2.2. Properties of Co-Based Alloys

The two basic elements of the Co-based alloys form a solid solution of up
to 65 wt% Co and the remainder is Cr as shown in Figure 5-3. The molybdenum
is added to produce finer grains, which results in higher strengths after casting
or forging.

Table 5-3. Chemical Compositions of Co-Based Alloys"

CoCrMo (F75) CoCrWNi (F90) CoNiCrMo (F562)

Element Min. Max. Min. Max. Min. Max.

Cr 27.0 30.0 19.0 21.0 19.0 21.0

Mo 5.0 7.0 9.0 10.5
Ni 2.5 9.0 11.0 33.0 37.0
Fe 0.75 3.0 1.0
C 0.35 0.05 0.15 0.025
Si 1.00 1.00 0.15
Mn 1.00 2.00 0.15
W 14.0 16.0
p O.ot5
S 0.010
Ti 1.0
Co Balance

a From Annual Book of ASTM Standards, Part 46, American Society for Testing and Materials, Philadelphia, 1980.
84 CHAPTER 5

CoCr
1800

1600
L
/' ~
1400
""'---. :::....... ~
1400'
~
TEMP. a
~~. l~

'"l'"
(OC) 1200
a+y
1000

\
Igeoo
800 ~
I' "" ,,
600
"", I (3+y Yt~ ,
/ : ,,,
400
/~ :
I ,
,
,
I ,
I
I
,,
200 I

o 20 40 60 80 100
Cr (w/o)

Figure 5-3. Phase diagram of Co-Cr. From C. J. Smithells (ed.), Metals Reference Book, Butter-
worths, London, 1976, p. 549.

One of the most promising wrought Co-based alloys is the CoNiCrMo alloy
originally called MP35N (Standard Pressed Steel Co.), which contains approxi-
mately 35 wt% Co and Ni each. The alloy has a high degree of corrosion resistance
to seawater (containing chloride ions) under stress. The cold-working can increase
the strength of the alloy considerably as shown in Figure 5-4. However, there is

ULTIMATE
TENSILE 2
STRENGTH
(GPo)

-~-\

CoNiCrMo

0
0 15 25 35 45 55 65 75
PERCENT REDUCTION

Figure 5-4. Relationship between ultimate tensile strength and the amount of cold-work for
CoNiCrMo alloy. From T. M. Devine and J. Wulff, "Cast vs. Wrought Cobalt-Chromium Surgical
Implant Alloys," J. Biomed. Mater. Res., 9, 151-167, 1975.
METALLIC IMPLANT MATERIALS 85

a considerable difficulty of cold-working especially when making large devices

such as hip joint stems. Only hot-forging can be used to fabricate an implant
with the alloy.
The abrasive wear properties of the wrought CoNiCrMo alloy are similar
to those of the cast CoCrMo alloy (about 0.14 mm/year in joint simulation test);
however, the former is not recommended for the bearing surfaces of a joint
prosthesis because of its poor frictional properties with itself or other materials.
The superior fatigue and ultimate tensile strength of the wrought CoNiCrMo
alloy make it very suitable for the applications that require long service life
without fracture or stress fatigue. Such is the case for the stems of the hip joint
prostheses. This advantage is more appreciated when the implant has to be
replaced with another one since it is quite difficult to remove the failed piece of
implant embedded deep in the femoral medullary canal. Furthermore, the revision
arthroplasty is usually inferior to the original in terms of its function owing to
poorer fixation of the implant.
Table 5-4 shows the mechanical properties required of Co-based alloys. As
is the case with other alloys, the increased strength is accompanied by decreased
ductility. Both the cast and wrought alloys have excellent corrosion resistance.
Experimental determination of the rate of nickel release from the CoNiCrMo
alloy and 316L stainless steel in 37°C Ringer's solution produced an interesting
result. Although the cobalt alloy has more initial release of nickel ions into the
solution, the rate of release was about the same (3 x 10- 10 g/ cm 2 / day) for both
alloys as shown in Figure 5-5. This is rather surprising since the nickel content
of the CoNiCrMo alloy is about three times that of 316L stainless steel.
The modulus of elasticity for the cobalt-based alloys does not change with
the changes in their ultimate tensile strength. The moduli range from 220 to
234 GPa, which are higher than other materials such as stainless steels. This may

Table 5-4. Mechanical Property Requirements of Co-Based Alloysa

Wrought CoNiCrMo (F562)

Cast Wrought
CoCrMo CoCrWNi Solution Cold worked Fully
Property (F76) (F90) annealed and aged annealed

Tensile strength 655 860 795-1000 1790 600

(MPa)
Yield strength 450 310 240-655 1585 276
(0.2% offset) (MPa)
Elongation (%) 8 10 50.0 8.0 50
Reduction 8 65.0 35.0 65
of area (%)
Fatigue strength 310 340
(MPa)b

U From Annual Book of ASTM Standards. Part 46, American Society for Testing and Materials, Philadelphia, 1980.
b Data from M. Semlitsch, "Properties of Wrought CoNiCrMo Alloy Protasul-10, a Highly Corrosion and Fatigue
Resistant Implant Material for Joint Endoprostheses," Eng. Med., 9,201-207, 1980.
86 CHAPTER 5

0.04
CoNiCrMo -
hot forged ( t ! . ) /
passivated (0) L
Ni ION 0.03

RELEASE
(fLg/cm 2 )
0.02 316L ss /

0.0 I

0
0 20 40 60
EXPOSURE TIME (days)
Figure 5-5. Nickel ion release versus time for hot-forged and passivated CoNiCrMo and 316L
stainless steel in 37°C Ringer's solution. From Biophase Implant Material, Technical Information
Publ. No. 3846, Richards Mfg. Co., Memphis, Tenn., 1980, p.7.

have some implications of different load transfer modes to the bone although it
is not established clearly what the effect of the increased modulus is.

5.2.3. Manufacturing of Implants Using Co-Based Alloys

The CoCrMo alloy is particularly susceptible to the work-hardening so that

the normal fabrication procedure used with other metals cannot be employed.
Instead the alloy is cast by a lost wax (or investment casting) method that involves
the following steps (Figure 5-6):

1. A wax pattern of the desired component is made.

2. The pattern is coated with a refractory material, first by a thin coating
with a slurry (suspension of silica in ethyl silicate solution) followed by
complete investing after drying.
3. The wax is melted out in a furnace (lOO-I50°C).
4. The mold is heated to a high temperature burning out any traces of wax
or gas-forming materials.
5. Molten alloy is poured with gravitational or centrifugal force. The mold
temperature is about 800-1000°C and the alloy is at 1350-I400°C.

Controlling the mold temperature will have an effect on the grain size of the
final cast; coarse ones are formed at higher temperatures, which will decrease
the strength. However, high processing temperature will result in larger carbide
precipitates with greater distances between them resulting in a less brittle material.
Again there is a complementary (trade-off) relationship between strength and
toughness.
Figure 5-6. Lost wax casting of femoral joint prosthesis. (Courtesy of Howmedica, Inc., Rutherford,
N.J.) (a) Injection of wax into a brass mold. (b) Wax patterns assembled for a ceramic coating
(note the hollow part of the femoral head). (c) Application of the ceramic coating. (d) A hot
pressure chamber retrieves the wax, leaving behind a ceramic coating. (e) Pouring molten metals
into the preheated ceramic mold.
88 CHAPTER 5

Figure 5-6. (Continued).

Example 5-2

Calculate the number of Co atoms released during a year from the femoral head of a hip
joint prosthesis made of CoCrMo alloy. Assume that the wear rate is 0.14 mm/yr and that
all of the atoms become ionized.
Answer
Assume a nominal diameter of the prosthetic femoral head of 28 mm . The surface area is
A = (4/3)7T (1.4 cm)2 = 8.21 cm 2 • Half of this area is in contact with the socket portion
of the joint. Therefore, the volume of wear material is 0/2) x (1.4 cm)2 x 0.014 cm/yr =
0.055 cm 3/yr. Since the density of Co is 8.83 g/cm 3, and the atomic weight is 58.93, and
the alloy is about 65% cobalt,
atoms 0.65 x 0.055 cm 3/yr x 8.83 g/cm 3 x 6.02 x 10 23 atoms/mol
yr 58.93 atoms/mol
= 3.2 x 10 21 atoms/yr, or 10 14 atoms/sec.
METALLIC IMPLANT MATERIALS 89

5.3. Ti AND Ti-BASED ALLOYS

Attempts to use titanium for implant fabrication date to the late 1930s.
It was found that titanium was tolerated in cat femurs as was stainless steel
and Vitallium® (CoCrMo alloy). Its lightness (4.5 g/cm 3 compared to 7.9 g/cm 3
for 316 stainless steel, 8.3 g/ cm3 for cast CoCrMo, and 9.2 g/ cm 3 for wrought
CoNiCrMo alloys) and good mechanochemical properties are salient features
for implant application.

5.3.1. Compositions of Ti and Ti-Based Alloys

There are four grades of unalloyed titanium for surgical implant applications
as given in Table 5-5. The impurity contents separate them; oxygen, iron, and
nitrogen should be controlled carefully. Oxygen in particular has a great influence
on the ductility and strength.
One titanium alloy (Ti6AI4V) is widely used to manufacture implants and
its chemical requirements are given in Table 5-5. The main alloying elements of
the alloy are aluminum (5.5-6.5 wt%) and vanadium (3.5-4.5 wt%).

5.3.2. Structure and Properties of Ti and Ti Alloys

Titanium is an allotropic material that exists as a hexagonal close-packed

structure (Q:- Ti) up to 882°C and body-centered cubic structure (f3- Ti) above that
temperature. The addition of alloying elements to titanium enables it to have a
wide range of properties:

1. Aluminum tends to stabilize the a phase, that is, increase the transforma-
tion temperature from Q: to f3 phase (Figure 5-7).

Table 5-5. Chemical Compositions of Titanium and Its Alloy (ASTM F67, F136)

Element Grade 1 Grade 2 Grade 3 Grade 4 Ti6AI4V"

Nitrogen 0.03 h 0.03 0.05 0.05 0.05

Carbon 0.10 0.10 0.10 0.10 0.08
Hydrogen O.ot5 O.ot5 O.ot5 O.ot5 0.0125
Iron 0.20 0.30 0.30 0.50 0.25
Oxygen 0.18 0.25 0.35 0.40 0.13
Titanium Balance

" Aluminum 6.00 wt% (5.50-6.50), vanadium 4.00 wt% (3.50-4.50), and other elements 0.1 wt% maximum or 0.4 wt%
total.
h All are maximum allowable weight percent.
90 CHAPTER 5

1200

1000

,,
TEMP. 800 ,,
(Oel ,,
600 ,,,
-}<l(HCP)
400
Ti-AI-4V
200~--,---,---,---.---,
o 2 468 10
AI (w/ol

Figure 5-7. Part of phase diagram of Ti-AI-V at 4wt% V. From C. J. E. Smith and A. N. Hughes,
"The Corrosion Fatigue Behavior of a Titanium--6 wlo Aluminum-4 wlo Vanadium Alloy," Eng.
Med., 7,158-171,1966.

2. Vanadium stabilizes the {3 phase by lowering the temperature of the

transformation from a to {3.

The a alloys have single-phase microstructure (Figure 5-8a), which promotes

good weldability. The stabilizing effect of the high aluminum content of these
groups of alloys results in excellent strength characteristics and oxidation resist-
ance at high temperature (300-600°C). These alloys cannot be heat-treated for
strengthening since they are single-phased.
The addition of a controlled amount of {3-stabilizers causes the higher
strength {3 phase to persist below the transformation temperature, which results
in the two-phase system. As discussed in Section 3-4, the precipitates of {3 phase
will appear by heat treatment in the solid solution temperature and subsequent
quenching, followed by aging at a somewhat lower temperature. The aging cycle
causes the precipitation of some fine a particles from the metastable {3 imparting
a structure that is stronger than the annealed a-{3 structure (Figure 5-8b).
The higher percentage of {3-stabilizing elements (13 wt% V in Ti13VllCr3AI
alloy) results in a microstructure that is substantially {3 which can be strengthened
by the heat treatment (Figure 5-8c).
The mechanical properties of the commercially pure titanium and 6Al4V
alloy are given in Table 5-6. The modulus of elasticity of these materials is about
110 GPa, which is half the value of Co-based alloys. From Table 5-6 one can
see that the higher impurity content leads to higher strength and reduced ductility.
The strength of the material varies from a value much lower than that of 316
stainless steel or the Co-based alloys to a value about equal to that of annealed
316 stainless steel of the cast CoCrMo alloy. However, when compared by the
specific strength (strength per density) the titanium alloy excels over any other
implant material as shown in Figure 5-9. Titanium, nevertheless, has poor shear
strength, making it less desirable for bone screws, plates, and similar applications.
It also tends to gall or seize when in sliding contact with itself or another metal.
METALLIC IMPLANT MATERIALS 91

C
Figure 5-8. Microstructure of Ti alloys (all are 500 x ). (a) Annealed a alloy. (b) Ti6AI4V, a-f3 alloy,
annealed. (c) f3 alloy, annealed. From G. H. Hille, "Titanium for Surgical Implants," J. Mater., 1,
373- 383, 1966.
92 CHAPTER 5

Table 5-6. Mechanical Properties of Ti and Ti Alloy (ASTM F136)

Properties Grade 1 Grade 2 Grade 3 Grade 4 Ti6Al4V

Tensile strength 240 345 450 550 860

(MPa)
Yield strength 170 275 380 485 795
(0.2% offset)
(MPa)
Elongation (%) 24 20 18 15 10
Reduction of area 30 30 30 25 25
(%)

Titanium derives its resistance to corrosion by the formation of a solid oxide

layer. Under in vivo conditions the oxide (Ti0 2 ) is the only stable reaction product.
The oxide layer forms a thin adherent film and passivates the material. Corrosion
resistance mechanisms are discussed further in Section 5.6.

5.3.3. Manufacturing of Implants

Titanium is very reactive at high temperature and burns readily in the presence
of oxygen. It therefore requires an inert atmosphere for high-temperature process-
ing or is processed by vacuum melting. Oxygen diffuses readily in titanium and
the dissolved oxygen embrittles the metal. As a result, any hot working or forging
operation should be carried out below 925°C. Machining at room temperature
is not the solution to all of the problems since the material also tends to gall or

6AI4V Heat treated

6AI4V Annealed I
Unallayed- Grade4 I
Cold-worked I
J Wrought annealed
J As cast
-.J Cold-worked I
CD(/)
i(;(/)
=:=J Annealed
=::::J Cold-worked
c
r- ] Annealed

o 50 100 150 200

YIELD STRENGTH/DENSITY
[MPa /(g/cm3)l
Figure 5-9. Yield strength-to-density ratio of some implant materials. From G. H. Hille, "Titanium
for Surgical Implants," J. Mater., 1, 373-383, 1966.
METALLIC IMPLANT MATERIALS 93

seize the cutting tools. Very sharp tools with slow speeds and large feeds are
used to minimize this effect. Electrochemical machining is an attractive means.

Example 5-3

The allotropic phase change of titanium at 882°C is from hcp structure (a = 2.95 A,
C = 4.683 A) to bcc structure (a = 3.32 A). Calculate the volume change in cm 3/g by
heating above the transformation temperature. The density is 4.54 g/ cm 3 and the atomic
weight is 47.9 g.

Answer

Area of hexagon = ax a ~-2 x -21 x 6 = ~a2

3)3
2
A2
0

3)3 2 03
V hcp = ~ a c = 106.3 A which contains 6 atoms
2

Vbcc = a3 = 36.59 A3 which contains 2 atoms

106.3 A 3 106.3 A 3 x 6.02 X 1023

hcp:--- -------- = 0.2227 cm 3 / g
6 atoms 6 x 47.9 g/mol atoms
36.59 A3 36.59 A3 x 6.02 X 1023 3
bcc:--- -------- = 0.23 cm /g
2 atoms 2 x 47.9 g/mol atoms
The difference is 0.0073 cm 3/ g.

5.4. DENTAL METALS

5.4.1. Dental Amalgam

An amalgam is an alloy in which one of the component metals is mercury.

The rationale for using amalgam as a tooth filling material is that since mercury
is a liquid at room temperature, it can react with other metals such as silver and
tin and form a plastic mass that can be packed into the cavity, and that hardens
(sets) with time. To fill a cavity, the dentist mixes solid alloy, supplied in
particulate form, with mercury in a mechanical triturator. The resulting material
is readily deformable and is then packed into the prepared cavity. The solid alloy
is composed of at least 65% silver, and not more than 29% tin, 6% copper, 2%
zinc, and 3% mercury. The reaction during setting is thought to be

Y + Hg ~ Y + YI + Y2 (5-0

in which the Y phase is Ag 3 Sn, the YI phase is Ag 2Hg 3 , and the Y2 phase is
Sn7Hg, as shown in Figure 5-10.
The phase diagram for the Ag-Sn-Hg system shows that over a wide composi-
tional range, all three phases are present. Dental amalgams typically contain 45%
94 CHAPTER 5

1000
960.5 0

800
Liquid (Ll
TEMP.
(OC) 600

400
y+L

221 0
200

O+-~-r~~~--r-~.-~~
Ag 0 20 40 60 80 100 Sn

COMPOSITION (w/o)

Figure 5-10. Phase diagram of Ag and Sn.

to 55% mercury, 35% to 45% silver, and about 15% tin, when fully set. Strength
of the restoration increases during the setting process, so that the amalgam has
attained one quarter of its final strength after 1 hour, and almost all of its final
strength after 1 day.

5.4.2. Gold

Gold and gold alloys are useful metals in dentistry as a result of their
durability, stability, and corrosion resistance. Gold fillings are introduced by two
methods: casting and malleting. Cast restorations are made by taking a wax
impression of the prepared cavity, making a mold from this impression in a
material such as gypsum silica, which tolerates high temperature, and casting
molten gold in the mold. The patient is given a temporary filling for the intervening
time. Gold alloys are used for cast restorations, since they have mechanical
properties that are superior to those of pure gold. Corrosion resistance is retained
in these alloys provided they contain 75 wt% or more of gold and other noble
metals. Copper, alloyed with gold, significantly increases its strength. Platinum
also improves the strength, but no more than about 4% can be added, or the
melting point of the alloy is elevated excessively. Silver compensates for the color
of copper. A small amount of zinc may be added to lower the melting point and
to scavenge oxides formed during melting. Gold alloys of different composition
are available. Softer alloys containing more than 83% gold are used for inlays,
which are not subjected to much stress. Harder alloys containing less gold are
chosen for crowns and cusps, which are more heavily stressed.
Malleted restorations are built up in the cavity from layers of pure gold foil.
The foils are degassed before use, and the layers are welded together by pressure
at room temperature. In this type of welding, the metal layers are joined by
thermal diffusion of atoms from one layer to the other. Since intimate contact is
required in this procedure, it is particularly important to avoid contamination.
METALLIC IMPLANT MATERIALS 95

The pure gold is relatively soft, so this type of restoration is limited to areas not
subjected to much stress.

5.4.3. Nickel-Titanium Alloys

The nickel-titanium alloys show an unusual property, i.e., after the material
is deformed it can snap back to its previous shape following heating the material.
This phenomenon is called shape memory effect (SME). This effect of Ni-Ti alloy
was first observed by Buehler and Wiley at the U.S. Naval Ordnance Laboratory.
The equiatomic Ni-Ti alloy (Nitinol®) exhibits an exceptional SME near room
temperature: if it is plastically deformed below the transformation temperature,
it reverts back to its original shape as the temperature is raised. The SME can
be generally related to a diffusionless martensitic phase transformation that is
also thermoelastic in nature, the thermoelasticity being attributed to the ordering
in the parent and martensitic phases. The thermoelastic martensitic transformation
exhibits the following general characteristics:
1. Martensite formation can be initiated by cooling the material below Mso
defined as the temperature at which the martensitic transformation begins.
Martensite formation can also be initiated by applying a mechanical stress
at a temperature above Ms.
2. Ms and As (the temperature at which the reverse austenitic transformation
begins upon heating) temperatures can be increased by applying stresses
below the yield point; the increase is proportional to the applied stress.
3. The material is more resilient than most metals.
4. The transformation is reversible.
Some possible applications of shape memory alloys are orthodontic dental
archwires, intracranial aneurysm clips, a vena cava filter, contractile artificial
muscles for an artificial heart, orthopedic implants and other medical devices.
In order to develop such devices, it is necessary to understand fully the
mechanical and thermal behavior associated with the martensitic phase transfor-
mation. A widely known Ni-Ti alloy is 55-Nitinol (55 wt% or 50 at% NO, which
has a single phase and the "mechanical memory" plus other properties, e.g., high
acoustic damping, direct conversion of heat energy into mechanical energy, good
fatigue properties, and low-temperature ductility. Deviation from the 55-Nitinol
(near stoichiometric NiTi) in the Ni-rich direction yields a second group of alloys
that are also completely nonmagnetic but differ from 55-Nitinol in their capability
of being thermally hardened to higher hardness levels. Shape recovery capability
decreases and heat treatability increases rapidly as the Ni content approaches
60 wt%. Both 55- and 60-Nitinols have relatively low moduli of elasticity and
can be tougher and more resilient than stainless steel, Ni-Cr, or Co-Cr based
alloys.
Efficiency of 55-Nitinol shape recovery can be controlled by changing the
final annealing temperatures during preparation of the alloy device. For the most
efficient recovery, the shape is fixed by constraining the specimen in a desired
96 CHAPTER 5

Table 5-7. Chemical Composition of

Ni-Ti Alloy Wire

Element Composition (wt%)

Ni 54.01
Co 0.64
Cr 0.76
Mn 0.64
Fe 0.66
Ti Balance

configuration and heating to between 482 and 510°C. If the annealed wire is
deformed at a temperature below the shape recovery temperature, shape recovery
will occur upon heating, provided the deformation has not exceeded crystallo-
graphic strain limits (-8% strain in tension). The Ni-Ti alloys also exhibit good
biocompatibility and corrosion resistance in vivo.
The mechanical properties of Ni-Ti alloys are especially sensitive to the
stoichiometry of composition (typical composition is given in Table 5-7) and the
individual thermal and mechanical history. Although much is known about the
processing, mechanical behavior, and properties relating to the SME, considerably
less is known about the thermomechanical and physical metallurgy of the alloy.
The differential scanning calorimeter (DSC) is a device that is capable
of measuring the heat capacity of materials as a function of temperature.
Figure 5-11 shows a typical DSC plot for Ni-Ti alloy and identifies several

1
u
:E
II:
W
:I:
l-
e
o
z
w

Af

u
:E
II:
w
:I:
l-
ex
w

1
0 10 20 30 40 50
TEMPERATURE (DC)

Figure 5-11. A typical DSC curve of the specific heat (thermal energy) versus temperature. From
J. H. Lee, J. B. Park, S. F. Andreasen, and R. S. Lakes, "Thermomechanical Studies of NiTi Alloys,"
J. Biomed. Mater. Res. 22, 573-588, 1989.
METALLIC IMPLANT MATERIALS 97

relevant parameters. The Ni-Ti alloys generally display two peaks that are
associated with the martensitic phase transformation temperature in cooling (Ms
and M r) and heating (austenitic phase transformation temperature; As and Ar)
where subscripts sand f indicate the starting and finishing of transformations,
respectively. The area under the specific heat versus temperature curve can be
used to calculate the amount of thermal energy used for the phase transformation.
The typical curves for bending moment versus bend angle are shown in
Figure 5-12. Ni-Ti alloy wire specimens were tested at O°C and room temperature.
As can be seen from the curves, the samples deformed at room temperature
recovered almost completely to their original states, indicating that the transforma-
tion temperature is close to room temperature. From these curves the elastic
stiffnesses of the alloy were calculated using the straight portions of the curves
and are given in Table 5-8. The results also indicate that the elastic modulus is
higher at higher temperature.
Figure 5-13 shows the typical microstructure of a Ni-Ti specimen at room
temperature. The optical microscopic picture of the cross section of the wire
shown in Figure 5-13a illustrates the evenly dispersed nonmetallic inclusions in
the Ni-Ti matrix. The inclusions are presumed to be primarily titanium car-
bonitrides, with a few nickel-titanium oxides. Panels band c show SEM photo-
micrographs of the unbent and bent portions of specimens, respectively, at room
temperature. From these photomicrographs one can see that long pores are aligned

8 9

LEGEND
'"o
I O· c--- ---
~6
ROOM TEMPERATURE-- '"~
o

.r; x
6 •
E
z
t- t-
Z4 Z
lJJ LLJ
~ ~
o o
~ ~

(!) 3(!)
Z
Z
~2
oZ
LLJ lJJ
CD CD

5 10 15 20 25 30
BEND ANGLE (0)

Figure 5-12. The bending moment versus bending angle curves for a NiTi specimen at O°C and
room temperature. From J. H. Lee, J. B. Park, S. F. Andreasen, and R. S. Lakes, "Thermomechanical
Studies of NiTi Alloys," J. Biomed. Mater. Res. 22, 573-588,1989.
98 CHAPTER 5

Table 5-8. Elastic Properties of Ni-Ti A"oy Wire

Elastic stiffness (Eb)a

Test temp. Ibf in/ degree, N m/degree, Young's modulus (E)

(0C) X 10-3 xl0-4 (GPa)

o 1.4 1.58 31
Room temp. 4.3 4.86 30

" Eh = Ilib/!!.w (bending moment/angle of bending).

with the longitudinal direction of the specimen. The long pores appear to be
created by the polishing and chemical etching steps used to prepare the specimen.
However, Figure 5-13c shows transformed martensitic structure near the surface
in contrast to the undeformed structure shown in Figure 5-13b, which does not
show martensitic structure at room temperature.

5.5. OTHER METALS

Several other metals have been used for a variety of specialized implant
applications.
Tantalum has been subjected to animal implant studies and has been shown
very biocompatible. Due to its poor mechanical properties (Table 5-9) and its
high density (16.6 g/ cm3 ), it is restricted to few applications such as wire sutures
for plastic surgeons and neurosurgeons and a radioisotope for bladder tumors.
Platinum and other noble metals in the platinum group are extremely cor-
rosion resistant but have poor mechanical properties. They are mainly used as
alloys for electrodes such as pacemaker tips because of their high resistance to
corrosion and low threshold potentials.

5.6. CORROSION OF METALLIC IMPLANTS

Corrosion is the unwanted chemical reaction of a metal with its environment,

resulting in its continued degradation to oxides, hydroxides, or other compounds.

Table 5-9. Mechanical Properties of Tantalum (ASTM F560)

Properties Fully annealed Cold-worked

Tensile strength (MPa) 205 515

Yield strength 140 345
(0.2%) offset) (MPa)
Elongation (%) 20-30 2
Young's modulus (GPa) 190
Figure 5-13. Microstructure of a NiTi specimen at room temperature. (a) Optical micrograph of
the cross section (100X) showing nonmetallic inclusions (black particulates) and NiTi matrix (white
background). (b) Scanning electron micrograph of unbent specimen at a longitudinal section
(1000X) showing elongated pores in the wire-drawing direction. (c) Scanning electron micrograph
of bent specimen at a longitudinal section (1000X). Note the martensites formed at about 45° to
the wire-drawing direction (top to bottom in this photo).
100 CHAPTER 5

Tissue fluid in the human body contains water, dissolved oxygen, proteins, and
various ions such as chloride and hydroxide. As a result, the human body presents
a very aggressive environment to metals used for implantation. Corrosion
resistance of a metallic implant material is consequently an important aspect of
its biocompatibility.

5.6.1. Electrochemical Aspects

The lowest free energy state of many metals in an oxygenated and hydrated
environment is that of the oxide. Corrosion occurs when metal atoms become
ionized and go into solution, or combine with oxygen or other species in solution
to form a compound that flakes off or dissolves. The body environment is very
aggressive in terms of corrosion since it is not only aqueous but also contains
chloride ions and proteins. A variety of chemical reactions occur when a metal
is exposed to an aqueous environment, as shown in Figure 5-14. The electrolyte,
which contains ions in solution, serves to complete the electric circuit. In the
human body, the required ions are plentiful in the body fluids. Anions are negative
ions, which migrate toward the anode, and cations are positive ions, which migrate
toward the cathode. The electrical component V in Figure 5-14 can be a voltmeter
with which to measure the potential produced; or it can be a battery in which
case the cell is an electroplating cell; or it can be tissue resistance, in which case
the electrochemical cell is an unwanted corrosion cell for a biomaterial in the
body. In the body, an external electrical driving source may be present in the
form of a cardiac pacemaker, or an electrode used to stimulate bone growth. At
the anode, or positive electrode, the metal oxidizes. The following reactions
involving the metal M may occur:

(5-2)

At the cathode, or negative electrode, the following reduction reactions are

important:

Electrolyte

Figure 5-14. Electrochemical cell.

METALLIC IMPLANT MATERIALS 101

(5-3)
(5-4)

2H30+ + 2e- ~ H 2i+2H 20 (5-5)

!02 + H 20 + 2e- ~ 20H- (5-6)

Consider as an example the corrosion of a metal such as iron. Metallic iron

goes into solution in ionized form as follows:

(5-7)

In the presence of oxygen, rust may be formed in the following reactions:

(5-8)

4Fe3+ + 120H- ~ 4Fe(OHh~ (5-9)

If less oxygen is available, Fe304, magnetite, may precipitate rather than ferric
hydroxide.
The tendency of metals to corrode is expressed most simply in the standard
electrochemical series of Nernst potentials, shown in Table 5-10. These potentials
are obtained in electrochemical measurements in which one electrode is a standard
hydrogen electrode formed by bubbling hydrogen through a layer of finely divided
platinum black. The potential of this reference electrode is defined to be zero.
Noble metals are those that have a potential higher than that of a standard
hydrogen electrode; base metals have lower potentials.
If two dissimilar metals are present in the same environment, the one that
is most negative in the galvanic series will become the anode, and bimetallic (or
galvanic) corrosion will occur. Galvanic corrosion can be much more rapid than
the corrosion of a single metal. Consequently, implantation of dissimilar metals
(mixed metals) is to be avoided. Galvanic action can also result in corrosion
within a single metal, if there is inhomogeneity in the metal or in its environment,
as shown in Figure 5-15.

Table 5-10. Standard Electrochemical

Series

Reaction !:;'Eo (volts)

Li_ Li+ -3.045

Na- Na+ -2.714
AI-AI+ -1.66
Ti_Ti3+ -1.63
Fe- Fe2+ -0.44
H 2 -2H+ 0.000
Ag-Ag+ +0.799
Au-Au+ +1.68
102 CHAPTER 5

Matrix Crack
(cathode) (anode)

Grains (cathode)
-- Boundaries (anode)

Figure 5-15. Micro-corrosion cells. (a) Grain boundaries are anodic with respect to the grain
interior. (b) Crevice corrosion due to oxygen-deficient zone in metal's environment.

The potential difference E actually observed depends on the concentration

of the metal ions in solution according to the Nernst equation,

E = Eo+ (RT/nF) In[Mn+] (5-10)

in which R is the gas constant, Eo is the standard electrochemical potential, T

is the absolute temperature, F is Faraday's constant, 96,485 coulombs/mole, and
n is the number of moles of ions.
The order of nobility observed in actual practice may differ from that
predicted thermodynamically. The reasons are that some metals become covered
with a passivating film of reaction products, which protects the metal from further
attack. The dissolution reaction may be strongly irreversible so that a potential
barrier must be overcome. In this case, corrosion may be inhibited even though
it remains energetically favorable. Finally, the corrosion reactions may proceed
slowly: the kinetics are not determined by the thermodynamics.

5.6.2. Pourbaix Diagrams in Corrosion

The Pourbaix diagram is a plot of regions of corrosion, passIvIty, and

immunity as they depend on electrode potential and pH. The Pourbaix diagrams
are derived from the Nernst equation and from the solubility of the degradation
products and the equilibrium constants of the reaction. For the sake of definition,
the corrosion region is set arbitrarily at a concentration of greater than 10-6 g
atom per liter (molar) or more of metal in the solution at equilibrium. This
corresponds to about 0.06 mg/liter for metals such as iron and copper, and
0.03 mg/liter for aluminum. Immunity is defined as equilibrium between metal
and its ions at less than 10-6 molar. In the region of immunity, the corrosion is
energetically impossible. Immunity is also referred to as cathodic protection. In
the passivation domain, the stable solid constituent is an oxide, hydroxide, a
hydride, or a salt of the metal. Passivity is defined as equilibrium between a metal
METALLIC IMPLANT MATERIALS 103

+1 CORROSION

POTENTIAL
(V)
........................... -
.......
/Gastric
Fluid
o r-~~~~~--~'r------+----~--
0--- 7 pH
---
14
--.. --.. -Urine Bile

-I
IMMUNITY

Figure 5-16. Pourbaix diagram for chromium, showing regions associated with various body fluids.
From J. H. Dumbleton and J. Black, An Introduction to Orthopaedic Materials, Charles C. Thomas,
Springfield, III., 1975.

and its reaction products (oxides, hydroxides, etc.) at a concentration of 10- 0

molar or less. This situation is useful if reaction products are adherent. In the
biomaterials setting, passivity mayor may not be adequate: disruption of a passive
layer may cause an increase in corrosion. The equilibrium state may not occur
if reaction products are removed by the tissue fluid. Materials differ in their
propensity to reestablish a passive layer that has been damaged. This layer of
material may protect the underlying metal if it is firmly adherent and nonporous;
in that case further corrosion is prevented. Passivation can also result from a
concentration polarization due to a buildup of ions near the electrodes. This is
not likely to occur in the body since the ions are continually replenished. Cathodic
depolarization reactions can aid in the passivation of a metal by virtue of an
energy barrier that hinders the kinetics. Equations (5-5) and (5-6) are examples.
In the diagrams shown in Figures 5-16 to 5-19, there are two diagonal lines.
The top ("oxygen") line represents the upper limit of the stability of water and
is associated with oxygen-rich solutions or electrolytes near oxidizing materials.

I
POTENTIAL
(V)
o
Figure 5-17. Pourbaix diagram for an immune
-I
metal: gold. Stippled region: immunity;
hatched region: corrosion. From M. Pourbaix, : :: " . .: ', ... :
-.: .. . , - "
Atlas of Electrochemical Equilibria in Aqueous
Solutions, NACE, Houston/CEBELCOR,
o 7 14
Brussels, 1974. pH
104 CHAPTER 5

POTENTIAL
(V) 0

-I Figure 5-18. Pourbaix diagram for a passive

metal: titanium. Stippled region: immunity;
hatched region: corrosion; unmarked region:
-2
passivity. From M. Pourbaix, Atlas of Electro-
0 7 14
chemical Equilibria in Aqueous Solutions,
pH NACE, Houston/CEBELCOR, Brussels, 1974.

In the region above this line, oxygen is evolved according to 2H 2 0 ~

O 2 + 4H+ + 4e-. In the human body, saliva, intracellular fluid, and interstitial
fluid occupy regions near the oxygen line, since they are saturated with oxygen.
The lower ("hydrogen") diagonal line represents the lower limit of the stability
of water. Hydrogen gas is evolved according to Eq. (5-5). Aqueous corrosion
occurs in the region between these diagonal lines on the Pourbaix diagram. In
the human body, urine, bile, the lower gastrointestinal tract, and secretions of
ductless glands occupy a region somewhat above the hydrogen line.
The significance of Pourbaix diagrams is as follows. Different parts of the
body have different pH values and oxygen concentrations. Consequently, a metal
that performs well (is immune or passive) in one part of the body may suffer an
unacceptable amount of corrosion in another part. Moreover, pH can change
dramatically in tissue that has been injured or infected. In particular, normal
tissue fluid has a pH of about 7.4, but in a wound it can be as low as 3.5, and
in an infected wound the pH can increase to 9.0.
Pourbaix diagrams are useful, but do not tell the whole story; there are some
limitations. Diagrams are made considering equilibrium among metal, water, and
reaction products. The presence of other ions, e.g., chloride, may result in very
much different behavior and large molecules in the body may also change the

2
Mg

POTENTIAL
(V) 0

-I
Figure 5-19. Pourbaix diagram for magnesium.
Hatched region: corrosion; unmarked region:
-2
passivity. From M. Pourbaix, Atlas of Electro-
0 7 14 chemical Equilibria in Aqueous Solutions,
pH NACE, Houston/CEBELCOR, Brussels, 1974.
METALLIC IMPLANT MATERIALS 105

situation. Prediction of "passivity" may in some cases be optimistic, since reaction

rates are not considered.

5.6.3. Rates of Corrosion and Polarization Curves

The regions in the Pourbaix diagram specify whether corrosion will take
place, but they do not determine the rate. The rate, expressed as an electric
current density (current per unit area), depends on electrode potential as shown
in the polarization curves in Figure 5-20. From such curves, it is possible to
calculate the number of ions per unit time liberated into the tissue, as well as
the depth of metal removed by corrosion in a given time. An alternative experiment
is one in which the weight loss of a specimen of metal due to corrosion is measured
as a function of time.
The rate of corrosion also depends on the presence of synergistic factors,
such as those of mechanical origin. For example, in corrosion fatigue, repetitive
deformation of a metal in a corrosive environment results in acceleration of both
the corrosion and the fatigue microdamage. Since the body environment involves
both repeated mechanical loading and a chemically aggressive environment,
fatigue testing of implant materials should always be performed under physiologi-
cal environmental conditions: under Ringer's solution at body temperature. In
fretting corrosion, rubbing of one part on another disrupts the passivation layer,
resulting in accelerated corrosion. In pitting, the corrosion rate is accelerated in
a local region. Stainless steel is vulnerable to pitting. Localized corrosion can

Active
,.. Base Metal Substrate
0.2

o
0.2
POTENTIAL
(V)
0.4

0.6

0.8

-Ti Alloy
1.0
Noble
10- 4 10- 3 10- 2 10- 1
CURRENT DENSITY (mA/cm2 )

Figure 5-20. Potential-current density curves for some biomaterials. From E. H. Greener, J. K.
Harcourt, and E. P. Lautenschlager, Materials Science in Dentistry, Williams and Wilkins, Baltimore,
1972.
106 CHAPTER 5

occur if there is inhomogeneity in the metal or in the environment. Grain

boundaries in the metal may be susceptible to the initiation of corrosion, as a
result of their higher energy level. Crevices are also vulnerable to corrosion, since
the chemical environment in the crevice may differ from that in the surrounding
medium. The area of contact between a screw and a bone plate, for example,
can suffer crevice corrosion.

Example 5-4

Consider a restoration made of dental amalgam. Suppose that the corrosion current density
is 100 }.tA/ cm 2 . Is this a reasonable value? This material is an alloy; assume the density
is 11 g/ cm 3 and that the mean atomic weight is 150 amu.
(a) How many univalent ions will be released per year from a restoration surface 3.16 mm
by 3.16 mm?
(b) If the corrosion is uniform, how many millimeters of depth will be lost per year?
Discuss your answer.

Answer

(a) Ion flow,

J= [
4 A ] [
10 -- coul x e- ]
cm 2 sec x 1.6 x 10- 19 coul

[0.1 cm 2 ][3.15 x 107 sec/yr] = 1.97 x 1021 e- /yr

(b) Depth loss,

L = J[l ion/ e- 150 g/mol] [ 2! . / ]

6.02 x 10 IOns mol

[1/(11 g/cm3 )][1/0.1 cm 2 ] = 0.45 cm/yr

(Depth loss is far in excess of what is known to occur in dental amalgams. However, the
value of 100 }.tA/ cm 2 is a conservative estimate of the corrosion current density from
the polarization curve in Figure 5-19. Surface changes that depend upon time may
occur in vivo.)

5.6.4. Corrosion of Available Metals

Choice of a metal for implantation should take into account the corrosion
properties discussed above. Metals that are in current use as biomaterials include
gold, cobalt-chromium alloys, type 316 stainless steel, titanium, nickel-titanium
METALLIC IMPLANT MATERIALS 107

alloys, and silver-mercury amalgam. This section deals with the corrosion aspects
of these metals. Their composition and physical properties are discussed in
Section 5.4.
The noble metals are immune to corrosion and would be ideal materials if
corrosion resistance were the only concern. Gold is widely used in dental restor-
ations and in that setting it offers superior performance and longevity. Gold is
not, however, used in orthopedic applications as a result of its high density,
insufficient strength, and high cost.
Titanium is a base metal in the context of the electrochemical series; however,
it forms a robust passivating layer and, as shown in Figure 5-18, it remains passive
under physiological conditions. Corrosion currents in normal saline are very low:
10- 8 AI cm 2 • Titanium implants remain virtually unchanged in appearance. Ti
offers superior corrosion resistance but is not as stiff or strong as steel.
Cobalt-chromium alloys, like titanium, are passive in the human body. They
are widely in use in orthopedic applications. They do not exhibit pitting.
Stainless steels contain enough chromium to confer corrosion resistance by
passivity. The passive layer is not as robust as in the case of titanium or the
cobalt-chromium alloys. Only the most corrosion resistant of the stainless steels
are suitable for implants. These are the austenitic types 316, 316L, and 317, which
contain molybdenum. Even these types of stainless steel are vulnerable to pitting
and to crevice corrosion around screws.
Dental amalgam is an alloy of mercury, silver, and tin. Although the phases
are passive at neutral pH, the trans passive potential for the 'Y2 phase is easily
exceeded, due to interphase galvanic couples or potentials due to differential
aeration under dental plaque. Amalgam, therefore, often corrodes and is the most
active (corrosion prone) material used in dentistry.

5.6.5. Minimization of Corrosion: Case Studies

Although laboratory investigations are essential in the choice of a metal,

clinical evaluation in follow-up is also essential. Corrosion of an implant in the
clinical setting can result in symptoms such as local pain and swelling in the
region of the implant, with no evidence of infection; cracking or flaking of the
implant as seen on X-ray films, and excretion of excess metal ions. At surgery,
gray or black discoloration of the surrounding tissue may be seen, and flakes of
metal may be found in the tissue. Corrosion also plays a role in the mechanical
failures of orthopedic implants. Most of these failures are due to fatigue, and
the presence of a saline environment certainly exacerbates fatigue. The extent to
which corrosion influences fatigue in the body is not precisely known. Some
specific case histories follow.
Case 1. A total hip replacement broke after 1.5 years of service.
The prosthesis was X-rayed and found to be broken high in the femoral
stem area. The femoral implant was retrieved at surgery and analyzed. It was
found to be made of cast stainless steel. Cementation of the implant was not
optimal but was adequate. The choice of material was considered poor for this
108 CHAPTER 5

demanding application. Cast stainless steel is mechanically inferior to the wrought

steels commonly used in hip nails. Consequently, scratches on the stem during
implantation could initiate corrosion fatigue cracks. The stem was found to have
failed by fatigue followed by overload fracture. Moreover, in this implant,
excessive inclusion content and porosity were found in the area of the failure,
indicating poor manufacturing processes.
Case 2. A patient's arm was X-rayed. A bone plate, shown in Figure 5-21,
had been left in place for 30 years.
The screws had lost their clear outline due to corrosion and the irritating
effect of the corrosion products resulted in osseous proliferation. The plate was
found to be vanadium steel, a metal that was considered suitable in the 1920s
but that has since been abandoned for implants.
Case 3. A radiograph of a mold arthroplasty revealed that the surgeon had
used an ordinary iron nail (Figure 5-22) to reattach the osteotomized greater
trochanter.
The cup was found to be made of CoCrMo. The nail was removed and
found to be grossly corroded. The mold technique is rarely used today, and
surgeons would now not even think of using a carpenter's nail.
Case 4. A patient experienced pain and disability in a repaired shoulder
fracture (Figure 5-23).

Figure 5-21. Corrosion of an obsolete metal. Radiograph of a vanadium steel bone plate in place
30 years. The screws have lost their clear outline as a result of corrosion. From C. O. Bechtol,
A. B. Ferguson, Jr., and P. G. Laing, Metals and Engineering in Bone and Joint Surgery, Williams
and Wilkins, Baltimore, 1959.
METALLIC IMPLANT MATERIALS 109

Figure 5-22. Corrosion due to use of improper metals. (A) Radiograph of a CoCrMo mold hip
arthroplasty. An ordinary iron nail was used to reattach the osteotomized greater trochanter. (B)
The retrieved nail is grossly rusty. From C. O. Bechtol, A. B. Ferguson , Jr., and P. G. Laing, Metals
and Engineering in Bone and Joint Surgery, Williams and Wilkins, Baltimore, 1959.

The screws were removed and examined. One was found to be CoCrMo and
the others of stainless steel. Bimetallic corrosion had resulted. Such cases can be
avoided by better efforts by both manufacturers and surgeons to avoid mixed
metals.
Case 5. A broken drill bit was found in a radiograph of a repaired broken
hip (Figure 5-24).
Removal of the metal revealed severe corrosion. Mixed metals had been
implanted unintentionally.
Experience in the orthopedic setting suggests that corrosion is minimized
by the following:

1. Use appropriate metals.

2. Avoid implantation of different types of metal in the same region. In the
manufacturing process, provide matched parts from the same batch of
the same variant of a given alloy.
110 CHAPTER 5

Figure 5-23. Mixed metals . Radiograph of a repaired shoulder. This patient experienced pain and
disability. Examination of the screws following removal revealed one to be made of CoCrMo and
the others of stainless steel.

3. Design the implant to minimize pits and crevices.

4. In surgery, avoid transfer of metal from tools to the implant or tissue.
Avoid contact between metal tools and the implant, unless special care
is taken.
5. Recognize that a metal that resists corrosion in one body environment
may corrode in another part of the body.

Experience in the dental setting has led to the following suggestions for the
minimization of corrosion.

1. Avoid using mixed metals to restore teeth in apposition, and if possible,

in the same mouth.
2. Use an insulating base when seating a metallic restoration to minimize
electrical conduction below the restoration.
METALLIC IMPLANT MATERIALS 111

Figure 5-24. Unintentional implantation of mixed metals. (A) Broken and corroded screws and a
broken drill bit removed from a patient. (B) Radiograph of fracture fixation device showing screws
and drill. From C. O. Bechtol, A. B. Ferguson, Jr., and P. G. Laing, Metals and Engineering in Bone
and Joint Surgery, Williams and Wilkins, Baltimore, 1959.

3. Avoid conditions that lead to plaque buildup, since regions covered by

plaque will experience a reduced pH. This could result in corrosion.

PROBLEMS

5-1. A I-mm-diameter eutectoid steel is coated with I-mm-thick aluminum. Using the following data,
answer (a)-(d) below.

Young's Yield Expansion

modulus strength Density coefficient
Material (GPa) (MPa) (g/cm 3 ) ere)

Eutectoid steel 205 300 7.84 10.8 x 10- 6

Aluminum 70 100 2.7 22.5 x 10- 6

(a) If the composite bar is loaded in tension, which metal will yield first?
(b) How much load can the composite carry in tension without plastic deformation?
(c) What is the Young's modulus of the composite?
(d) What is the density of the composite?
112 CHAPTER 5

5-2. Cu has the following characteristics:

Atomic no. 29
Atomic mass (amu) 63.54
Melting point (0C) 1084.5
Density (gjcm 3 ) 8.96
Crystal structure fcc
Atomic radius (nm) 0.1278
Ionic radius (nm) 0.096

(a) Calculate the density of Cu.

(b) Calculate the packing efficiency (or factor) of Cu.
(c) Would the grain size increase, decrease, or remain the same if the Cu was cooled more slowly
than the original Cu after remelting?
(d) Would the grain size increase, decrease, or remain the same if the metal of (c) was annealed?

5-3. Silver has the following characteristics:

Atomic no. 47 Crystal structure fcc

Atomic mass (amu) 107.87 Atomic radius (nm) 0.1444
Melting point (0C) 961.9 Ionic radius (nm) 0.126
Density (gj cm 3 ) 10.5

A bioengineer is trying to construct a binary phase diagram of Cu-Ag and the following additional
information was given:

Eutectic temperature 779.4°C

Eutectic composition 28.1 Cu-71.9 Ag
Maximum solubility of Cu in Ag 8.8%
Maximum solubility of Ag in Cu 8.0%

(a) Construct the binary phase diagram by using the information on Cu in Problem 5-2.
(b) Label all the phases in the phase diagram.

5-4. Sterling silver is made of 92.5% Ag and 7.5% Cu.

(a) Construct a temperature versus % fraction chart for liquid phases. Indicate specific temperatures
where the phases start to change.
(b) Calculate the density of the sterling silver by using the mixture rule.
(c) When the sterling silver was equilibrated at 780°C, what phase(s) exist? What are their composi-
tions?
(d) Can the sterling silver be age-hardened? Give a reason.

5-5. In general, pearlite steels are susceptible to corrosion. Explain why this is to be expected and
how it may be prevented.

5-6. Type 316L stainless steel has a maximum carbon content of 0.03%. Welding of finished com-
ponents is acceptable for this steel but not for type 316. Why? Explain how you would expect the
mechanical properties of each metal to differ.

5-7. Calculate the densities of cast and wrought cobalt alloys (CoNiCrMo). Densities for Co, Cr,
Mo, and Ni are 8.8, 7.9,10.22, and 8.91 gjcm3 , respectively. Use the mixture rule.

5-8. The phase diagram of dental amalgam is shown in Figure 5-10.

METALLIC IMPLANT MATERIALS 113

(a) Calculate the theoretical weight percentage of the silver and tin of the I' phase (Ag 3 Sn).
(b) What is the eutectic temperature?
(c) Can you age-harden the amalgam?
5-9. Suppose a titanium alloy (Ti6AI4V) implant experiences a corrosion current density of
0.1 fLA/cm 2 .
(a) Determine the depth of metal removed each year.
(b) Suppose no titanium were excreted and that all of the metal became lodged in a 1-cm thickness
of tissue near the implant. Determine the average concentration of metal in the tissue, in parts
per million by weight, in one year.
5-10. Magnesium is used in airplanes. Is it sensible to use it in an implant? Explain.
5-11. A patient is to be treated for a fracture from a hunting accident. Lead shotgun pellets remain
in the wound. The surgeon decides a bone plate should be used. What is your recommendation?
5-12. Chromium is considered as a candidate material for a staple to be exposed to stomach contents.
Is this suitable? Why?
5-13. An amalgam filling has been in service for 30 years and has corroded. The filling surface has
become black and the patient complains of tooth sensitivity. A gap has formed at the margin of the
restoration, allowing leakage. Should the filling be drilled out and replaced by gold?
5-14. A patient with a stainless steel bone plate complains of pain. The plate has been in service for
30 years and the original fracture has long since healed. Should the plate be removed and replaced
by gold?
5-15. Aluminum and iron are used as electrodes in an electrochemical cell. A voltmeter is connected
across the electrodes. What potential difference is observed?
5-16. The ASTM (American Society for Testing and Materials) issued standard specifications for
thermomechanically processed CoCrMo alloy for surgical implants (F799-82). What is the rationale
of the new standard and what are the differences from the earlier standard (F76)?

SYMBOLS/DEFINITIONS

Greek Letters

I' Silver tin intermetallic compound (Ag 3 Sn).

1'1 Silver mercury intermetallic compound (Ag 2 Hg 3 ).
1'2 Tin mercury intermetallic compound (Sn7Hg).
W Bending angle of the three-point mechanical bend test.

Latin Letters
Temperature for the start of reverse transformation, NiTi (III)-NiTi
(II). NiTi (II) and NiTi (III) designate structural variations
(phases) produced by diffusionless (martensitic) transformation.
Observed electrochemical potential difference.
Standard electrochemical potential difference, for negligible con-
centration.
Faraday's constant, 96,485 coulombs per mole.
Temperature where martensitic transformation ceases.
Martensitic transformation starting point in cooling.
Number of electron charges removed from ion.
Number of moles.
Gas constant.
Absolute temperature in K.
114 CHAPTER 5

Terms

Age harden: Hardening of alloys by precipitating second phase during aging

process.
Amalgam: An alloy obtained by mixing silver tin alloy with mercury.
Anode: Positive electrode in an electrochemical cell.
Cathode: Negative electrode in an electrochemical cell.
Corrosion: Unwanted reaction of metal with environment. In a Pourbaix diagram,
it is the region in which the metal ions are present at a concentra-
tion of more than 10-6 molar.
Crevice corrosion: A form of localized corrosion in which concentration gradients around
preexisting crevices in the material drive corrosion processes.
Galvanic corrosion: Dissolution of metal driven by macroscopic differences in elec-
trochemical potential, usually as a result of dissimilar metals in
proximity.
Galvanic series: Table of electrochemical potentials associated with the ionization of
metal atoms. These are called Nernst potentials.
Immunity: Resistance to corrosion by an energetic barrier. In a Pourbaix diagram,
it is the region in which the metal is in equilibrium with its ions
at a concentration of less than 10-6 molar.
Nernst potential: Standard electrochemical potential measured with respect to a stan-
dard hydrogen electrode.
Noble: Type of metal with a positive standard electrochemical potential.
Noble metals resist corrosion by immunity.
Passivation: Production of corrosion resistance by a surface layer of reaction
products.
Passivity: Resistance to corrosion by a surface layer of reaction products. In a
Pourbaix diagram, it is the region in which the metal is in equili-
brium with its reaction products at a concentration of less than
10-6 molar.
Pitting: A form of localized corrosion in which pits form on the metal surface.
Pourbaix diagram: Plot of electrical potential versus pH for a material in which the regions
of corrosion, passivity, and immunity are identified.
Shape memory Thermoelastic behavior of some alloys that can revert back to their
effect (SME): original shape when the temperature is greater than the phase
transformation temperature of the alloy.

BIBLIOGRAPHY

c. O. Bechtol, A. B. Ferguson, and P. G. Laing, Metals and Engineering in Bone and Joint Surgery,
Balliere, Tindall and Cox, London, 1959.
T. W. Duerig, K. N. Melton, D. Stockel, and C. M. Wayman (eds.), Engineering Aspects of Shape
Memory Alloys, Butterworths-Heinemann, London, 1990.
1. H. Dumbleton and 1. Black, An Introduction to Orthopedic Materials, Chapter 9, Charles C. Thomas,
Springfield, IL. 1975.
E. H. Greener, 1. K. Harcourt, and E. P. Lautenschlager, Materials Science in Dentistry, Williams
and Wilkins, Baltimore, 1972.
B. Harris and A. R. BunselI, Structure and Properties of Engineering Materials, Chapters 7 -9, Longmans,
London, 1977.
S. N. Levine (ed.), Materials in Biomedical Engineering, Annals of the New York Academy of Sciences,
Vol. 146, 1968.
D. C. Mears, Materials and Orthopedic Surgery, Chapter 5, Williams and Wilkins, Baltimore, 1979.
1. Perkins (ed.), Shape Memory Effects in Alloys, Plenum Press, New York, 1975.