Biomedical Engineering Reference
In-Depth Information
PU passed the examination imposed by the US Pharmacopeia Class VI of 1980 for
being recognized as “medical grade” materials. These were:
1. Acute systemic toxicity
2. Hemolysis
3. Muscle implantation
4. Mutagenicity
5. Cell adhesion
The examination was a full success, hence followed the next 'go or no-go': perfor-
mance under the expected mechanical constraints.
Fatigue.
The interface of hip stems is mainly exposed to shear forces. For in vivo
simulating of shear fatigue, samples were squeezed between a wedge-shaped steel
anvil and a stamp, immersed in conventional simulated body fluid (SBF). The latter
is a solution copying the ionic composition of body liquids, in particular of blood
plasma. It is obvious that the composition should be adapted to the site in the body
where the tested material is expected to function. In Appendix D, three groups of
compositions are listed: SBF, artificial saliva and heart valves and grafts.
The anvil was sinusoidally loaded at a frequency of 3 Hz and applying shear
stress cycled between 150 and 1,500 kPa. These conditions were derived from nor-
malized theoretical and experimental data. Force and displacement of the anvil were
sampled 64 times per cycle and the number of cycles was registered. This went
on to over 3:10
6
cycles, a common number of cycles in fatigue testing. The loads
exceeded three times the expected real life load and the number of load cycles cor-
responded to an estimated survival period of 5 years [372-374]. Figure
11.18
is an
example of a fatigue set up with sample immersed in SBF (Hanks').
a
b
F
Hanks'
Solutions
- 0.2
Stamp
(AISI 304)
- 0.4
7
Shoulder
piece
- 0.6
Stample
- 0.8
Anvil (AISI 304)
-1
-1.2
Cycle: 3330000
Cycle: 3
-1.4
-1.6
-1.8
-2
-5
-4.5
-4
Displacement [mm]
-3.5
-3
Fig. 11.18
Fatigue test of Sarlink (hardness Shore A 80). Conditions: sinusoidal force variation
between 150 to 1,500 N, frequency of 3 Hz. (
a
) Experimental fatigue set up; (
b
) hysteresis curves
as function of number of load cycles, up to 3:10
6
cycles. The shift of the curves on progressing
number of load cycles is proportional to creep. Courtesy of S.Jaecques [375, Fig. 16, 16]
