ORTHOPEDIC IMPLATJTS: MATERIAL AND FAILURE EVALUATION 
by 
W. Phucharoen 
Submitted in Partial Fulfillment of the Requirements 
for the Degree of 
Master of Science in Engineering 
in the 
Materials Science Program 
// &&/983 
Date -- 
ad 23? 1963 
Dean of the Graduate School Date 
Youngstown State University 
August 1983 
ORTHOPEDIC IMPLATJTS: MATERIAL AND FAILURE EVALUATION 
by 
W. Phucharoen 
Submitted in Partial Fulfillment of the Requirements 
for the Degree of 
Master of Science in Engineering 
in the 
Materials Science Program 
// &d /983 
Date -- 
- 
ad 23? 19d~ 
Dean of the Graduate School Date 
Youngstown State University 
August 1983 
ABSTRACT 
ORTHOPEDIC IMPLANT : I-IATERIAL AND FAILURE EVALUATION 
W . PHUCHAROES 
Master of Science in Engineering 
Youngstown State University, 1983 
Two type 316 stainlzss st,zel orthopedic implants, 
Eggers plate and Jewett hip nail-plate, were examined in 
order to find the causes of failure. Unusual crack patterns 
were found on the cylindrical surface of the implants. 
Type 316 stainless steel tensile samples were tested with 
the presence of tensile stresses to determine the low-cycle 
fatigue characteristics and to determine whether or not low- 
cycle tensile stress would produce either of the crack patterns 
- - 
observed in the fractured implants. 
it was found that the Eggers piate seems to fail 
- 
because of bending fatigue and improper design. Low-cycle 
tensile stresses do not produce the crack pattern observed 
on the cylindrical surface of the Jewett nail-plate. 
- - 
However, the most.important result of this work is 
the significance of the Dresence of tensile mean stresses 
that occur during normal walking of a human which can 
significantly shorten the operation life of a type 316 
stainless steel implant. 
- -- 
ACKNOWLEDGEMENTS 
The author wishes to thank his adviser, Dr. R.W. 
Jones, Department of Chemical Engineering and Materials 
Science at Youngstown State University for guidance in 
this study. 
iv. 
TABLE OF CONTENTS 
PA 
ABSTRACT ...................... ii 
ACKNOWLEDGELWNT .................. iii 
TABLE OF CONTENTS ................. iv 
.................. 
LIST OF FIGURES v 
CHAPTER 
I. INTRODUCTION ................ 1 
11. LITERATURE REVIEW ............. 6 
111. EXPERIMENTAL PROCEDURE ........... 11 
IV. RESULTS AND DISCUSSION ........... 21 
V. CONCLUSION ................. 39 
APPENDIX Sample Calculation for Table 3 ..... 40 
REFERENCES ..................... 42 
LIST OF FIGURES 
FIGURE 
PAGE 
1. Eggers Plate and Jewett Hip Nail-Plate . . . . . . 3 
2. A Leg Bone Attached with Eggers Plate and 
Jewett Hip Nail-Plate . . . . . . . . . . . . . 4 
3. Machine Scratch on a Type 316 Stainless Steel 
Standard Tensile Sample . . . . . . . . . . . . -12 
4. Fatigue Curve of Type 316 Stainless Steel and 
Fatigue Curves Obtained from this Experiment . . 16 
5. Fracture Surface of Sample No. 1, at 36X . . . . . 17 
Fracture Surface of Sample No. 2, at 36X . . . . . 17 
Fracture Surface of Sample No. 3, at 36X . . . . . 18 
Fracture Surface of Sample No. 4, at 36X . . . . . 18 
Fracture Surface of Sample No. 5, at 36X . . . . . 19 
Fracture Surface of Sample No. 6, at 36X . . . . . 19 
Fracture Surface of Sample No. 7, at 36X . . . . . 20 - - 
Fracture Surface of Sam$le No. 9, at 36X . . . . . 20 
Fracture Surface of Eggers Plate (SEN), at 600X . - 22 
Fracture Surface of Eggers Plate (SEM), at 1200X . 22 
Fracture Surface of Eggers Place (SEM), at 600X . 23 
Micrograph of Eggers Plate Through Failure Area, 
at 600X . . . . . . . . . . . . . . . . . -.-. . - 23 
Micrograph of Eggers Plate Through Failure Area, 
at 75X. . . . . . . . . . . . . . . . . . . . . 24 
Micrograph of Eggers Plate Through Failure Area, 
at 150X . . . . . . . . . . . . . . . . . . . . 24 
Crack Pattern Found on Cylindrical Surface of 
Jewett Hip Nail-Plate (SEM), at 400X . . . . . . 26 
- - 
Crack Pattern Found on Cylindrical Surface cf 
Jewett Hi? Nail-Plate (SEM), at 450X . . . . . . 26 
LIST OF FIGURES (Cont'd) 
FIGURE PAGE 
21. Crack Pattern Found on Cylindrical Surface of 
...... Jewett Hip Nail-Plate (SEM), at 1200X 27 
22. Crack Pattern Found on Cylindrical Surface of 
...... Jewett Hip Nail-Plate (SEM), at 1200X 27 
23. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 1 (SEM), at 500X. 28 
24. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 1 (SEM), at 800X. 28 
25. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 2 (SEM), at 820X. 29 
26. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 2 (SEM), at 870X 29 
27. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 3 (SEM), at 380X 30 
28. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 3 (SEM), at 840X 30 - - 
29. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 4 (SEM), at 400X 31 
- 
30. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 4 (SEM), at 2000X 31 
31. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 5 (SEM), at 400X 32 
32. Crack Pattern Found on Cylindrical Surface of-- 
.......... SampleNo. 5 (SEM), at1400X 32 
33. Crack Pattern Found on Cylindrical Surface of 
.......... 
Sample No. 6 (SEM), atlOOX. 33 
34. Crack Pattern Found on Cylindrical Surface of 
.......... SampleNo. 6 (SEM), at1900X 33 
35. Crack Pattern Found on Cylindrical Surface of 
........... Sample No. 7 (SEM), at 800X 34 . 
36. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 7 (SEM), at 1600X 34 
LIST OF FIGURES (Cont'd) 
FIGURE PAGE 
37. Crack Pattern Found on Cylindrical Surface of 
........... 
Sample No. 9 (SEM), at 500X 35 
38. Crack Pattern Found on Cylindrical Surface of 
.......... Sample No. 9 (SEM), at lOOOX 35 
39. SEM Fractograph at 4500X of Fracture Surface of 
.... Sample No. 1, Showing Fatigue Striations 37 
40. SEM Fractograph at 4400X of Fracture Surface of 
.... Sample No. 4, Showing Fatigue Striations 37 
LIST OF TABLES 
TABLE PAGE 
1 . Mechanical Properties of Type 316 Stainless 
...................... 
Steel 13 
.... 2 . Test Data Obtained from MTS Testing Machine 14 
....... . 3 Calculated Data Obtained from Table 2 15 
CHAPTER 1 
INTRODUCTION 
An "implant" according to Gray (1962) is a device which 
is temporarily attached to a fractured bone as a structural 
support while the fracture mends (li. The implant is 
removed once the fracture has mended. A "prothesis" is 
a device permanently attached to a bone. 
The first recorded use of an implant was in 1562 when 
gold prothesis was used to close a defect in a cleft 
palate (2) . Ludwigson (3r4) has divi6ed the history and 
development into the three periods: first use to 1825, 
1825 to 1925, and 1925 to the present time. In the first 
- - 
period only pure metals such as gold, silver, and copper 
were used as implants. In the second period, pure metals 
were still used as implants. But surgical techniques were 
greatly inproved which increased the success of the implant. 
In the third period, alloys with improved strengths were 
used as im?lants. 
- - 
Before the latter part of the nineteenth century, 
use of metallic orthopedic implants probably resulted in 
frequent failure of the implant. Advances in mechanical 
design and metallurgical technology, updated sterillization 
- -- 
practices and grzater discrimination in materials' selection, 
have greatly decreased inctdence of failure in these devices. 
Failure (particularly mechanical fracture) is not, however, 
uncommon, as evidenced by the large volume of literature 
concerning analysis of failed orthopedic implants 
(5-3) 
Consequently, further researches are encouraged so that use 
of orthopedic implants can be improved. 
The impetus for this work resulted from the examination 
of two orthopedic implants that failed in service. The 
failed devices were a Jewett hip nail-plate and an Egger 
onlay plate (Figure 1). Figure 2 is a sketch of how these 
two devices are attached to a leg bone. The two implants 
were examined with a scanning electron microscope (SEPl) in 
order to determine the causes of the failures. 
In both cases unusual crack patterns were observed. 
The Eggers plate exhibited numerous cracks and slip lines in 
the fracture surface. In the case of the Jewett hip nail7 
plate no cracks could be found in the fracture surface. How- 
ever, many cracks existed in the cylindrical surface immediately 
adjacent to the fracture surface. Neither of these crack 
patterns has been previously reported in the literature 
concerning.failed implants. The morphology of the two crack 
- - - 
paerns suggested that either fatigue, stress corrosion, 
or a combination of these mechanisms caused the fractures in 
the two implants. The objectives of this research were to 
determine low-cycle fatigue characteristics of type 316 
stainless steel, and to detem-ine whether or not low- - - 
cycle tensile stresses would produce either of the 
(a) Eggers Onlay Plate. 
(b) Jewett Hip Nail-Plate. 
Figure I. Sketch of Eggers Onlay Plate and - 
Jewett Hip Nail-Plate. 
- 
Figure 2. Sketch of How a Jewett Hip Nail-Plate 
and Eggers Onlay Plate are Attached to 
A Leg Bone. 
crack patterns observed in the two fractured orthopedic 
implants. 
CHAPTER I1 
LITERATURE REVIEW 
.A large number of articles have been published 
concerning the failure of orthopedic implants. Some of these 
are Ref. 5-8; and volume 10 of the Metals Handbook 
(9) 
I 
Dumbleton and Miller, devotes one chapter to the failure 
analyses of such devices. Generally, the failure of an 
implant cannot be attributed to only a single cause 
(10) 
One major cause of an implant failure is related directly to 
the human body, which is considered to be one of the most 
hostile environments for metals. In the human body, chemical 
and electro-chemical reactions which may occur are not well 
identified or understood. Moreover, the problem of loading 
geometries involving cyclic loading can be encountered - 
when dealing with particular areas of the musculo-skeleton 
system such as the hip or knee. The combination of the 
hostile or a-ggressive environment and stringent loading 
conditions have brought about two familiar problems: 
corrosion or fatigue and their interaction. Both of these 
phenomena have been identified as major contributors to 
deterioration of implants in service 
(16) 
Imylant failures can be related to five factors.?hey 
are (a) corrosion, (b) patient misuse, (c) materials selection, 
(d) improper installation of the implant and (e) design and 
manufacture. 
The first two factors listed concern directlywith the 
environment. The other three factors are related to the 
implant and its installation into the body. 
(a) . Corrosion 
The corrosion mechanisms ;hi~h:;~eneraI?~ lead to 
degradation or failure of currently used implant devices, 
. 
fretting enhanced crevice corrosion and stress-corrosion, 
are associated primarily with stainless steel usage. Stress- 
corrosion cracking has been recognized as a cause of implant 
failure. However, neither its identification nor its importance 
in most cases has been yet determined because the cyclic loading 
produced by the body movement almost always results in a 
fretted or striated fatigue or corrosion fatigue. 
Pitting 
and intergranular attack, considered as results of improper - - 
chemistry control or mechanical processing, are not yet observed 
in implants. - 
(b) Patient Misuse 
This factor of implant failures is directly related 
tothe role of the patient. Devices may fracture by overloading 
- - 
because of patient misuse. For example, if the patient falls, 
very heavy loads are placed on a fixation device. Ignoring 
instructions about weight limits that a device can bear is 
also hazardous to implants. The patient usually has little 
or no control over the forces that muscles exert on an implant. 
- 
- -- 
- (12 
High loads may result even when a patient is confined to bed 
(c) Material Selection 
The right kind of materials Car implants must be 
carefully selected because only a few alloys have the 
necessary combination of properties; otherwise, the implants 
may be dangerous to the patientl's body (13) 
For surgical implants, a bone plake and bone screw 
must be of similar metals so that galvanic corrosion can be 
avoided. Usually, the corrosion occurs between the plate 
and the underside of the screw heads. The importance of 
material selection is exemplified by Dumbleton and Miller. 
They studied an example of a nail-and-plate device made of 
cast and wrought cobalt-chromium alloys which fractured-as a 
result of dissimilar-metal -co&ct, stress conCentration. and 
crevice corrosion 
(14 1 
Type 316 stainless steel was finally selected for 
- - 
surgical implants since the second world war 
(15) 
. It 
became the desired metal for surgical implants basically 
- 
because of its high pitting corrosion resistance. The 
composition and microstructural variations affect more or 
less the resistance of type 316 stainless steel to pitting 
corrosion. The addition of 2-4 percent of molybd&num in 
composition kas proved to greatly reduce sensitiveness to this 
kind of attack in an environment containing chloride 
(16) 
Warren 
(17) 
revealed that pitting corrosion resistance can be 
decreased with cold work by promoting, along strain lines, - - 
the formation of second phase constituents. 
(d) Improper Installation of Implant 
Faiiure of an implant may result from nonoptimal 
installation. Occasionally, the most desirable device is not 
possibly available for immediate surgical use. For instance, 
the surgeon may be forced to improvise, encountering the 
crucial situation when the operating room runs out of the 
stock for a particular item. 
So~e iailures can be attributed to errors intechnique 
For example, some errors in insertation include use of 
inappropriate size of implant parts (either too large or too 
small a screw, a nail or a p1ate)or use of eccentric or 
misaligned screws. 
Improper installation of.implant can be caused bv over 
applying torque to bone screws. The other aspect of improper 
installation is when the head of the screw is not sheared-off 
immediately. The implant failure may occur later because 
(18 1 
- 
the screw becomes weak 
(e) Design and. i.ialiu?acture 
At the stage of irqlant- fabrication, high qua~itycontrol 
must be carefully exerted in the type of finish om-the 
fixation device. The surface must not be blemished cf 
machining marks that could influence stress raisers which 
eventually activate a fissure and subsequent fractures 
(19 1 
Premature features can be attributed to the design-of 
- ~. 
an implant. For instance, sharp corners of an implant 
can result in stress raisers and lead finally to 
fracture. In addition, the failure can come from screw holes 
which were punched too close to the edge of a bone plate. 
The plate is, thus, not able to withstand bending stresses 
(20 
CHAFTER I11 
EXPERIMENTAL PROCEDURE 
Two implants, a Jewett hip nail-plate and a Eggers 
onlay plate, were sectioned about half an inch from the 
fracture surface using silicon-carbide wheel. A scanning 
electron microscope (SEM) Model S-450 (Hitachi) was used 
to examine fracture surfaces and cylindrical surfaces 
of each implant specimen. For the Jewett hip nail-plate, 
the fracture surface was also examined by metallographic 
examinations. 
Standard 0.505 inch diameter tensile samples of type 
316 stainless steel used for this study were purchased from 
Laboratory Devices Co., Auburn, California. - - 
One tensile sample was used to measure the ncchanical 
properties (see Table 1) with Model M120 HVL, Satec - 
System Inc., Grove City, PA., tension testing machine. 
The remaining samples had a 0.005 inch cylindrical 
grove machined in the center of the gauge length in order 
- - 
to generate a stress concentration (see Figure 3). 
Without the stress concentration, fatigue failure would 
take a long period of time. Each sample was cycled in 
uniaxial tension on an MTS electrohydraulic tesing machine 
with different amplitude load and load range. Sinusoidal - -- 
wave forms were employed for the loading, with the expectation 
Figure 3. Show Machine Scratch on a Type 316 Stainless 
Steel Tensile Sample. 36~. 
-- 
TABLE 1 
MECHANICAL PROPERTIES OF TYPE 316 STAINLESS STEEL 
Offset yield strength ................ 69,750 psi 
.............. Ultimate tensile strength 95,400 psi 
Elongation. ........................ 46.9% 
-- - 
that wave forms would have little effect on tests carried out 
in air at room temperature 
(21) 
. The test data are listed 
in Table 2 and a plot of stress amplitude (Sa , psi) versus 
numbers of cycles to failure (S-N diagram) is shown in 
Figure 4. Also, the accepted fatigue curve for type .316 
stainless steel (22) is plotted on the same graph (see 
Figure 4) . 
The fractured samples were examined on the SEM. Their 
fracture surfaces were photographed (see Figures 5-12) in' 
order to determine the ratio of the ductile fracture arEa 
to the total fracture area (see Table 3). 
14 
TABLE 2 
TEST DATA OBTAINED FROM MTS TESTING MACHINE 
No. of cycles 
to failure 
29,376 
22,456 
6,965 
764 
6 
66 
14 
No failure 
14 
Sample No. 
1 
2 
3 
4 
5 
6 
7 
8 
9 
Load (lbs.) 
14,000 - + 2,000 
12,000 - + 3,000 
9,000 - + 4,000 
8,000 - + 5,000 
14,000 - + 4,000 
16,000 - + 2,000 
12,000 - + 4,000 
9,000 - + 2,000 
9,000 - + 6,000 
Frequency 
(CPS) 
1 
1 
1 
1 
1 
1 
1 
8 
1 
TABLE 3 
CALCULATED DATA OBTAINED FROM TABLE 2 
Sample No. Stress (p'si) Cycles to Amplitude Mean Stress Ratio of 
Failure Stress (psi) (psi) Ductile Area 
. 
to Total 
Fracture Area 
29,376 72,765 
22,456 62,370 
6,965 46,778 
764 41,580 
6 72,765 
66 83,160 
14 62,370 
No failure 46,778 
14 46,776 
I 
Type 316 Stainless Steel 
70,ooo 
(Machine Scratch) 
n 
60,000] m Mean Stress = 41,580 psi 
0 L 
lo0 iol io
2 
b3 104 i o5 lo6 107 
C 
CYCLES TO FAILURE 
H 
vz 
PC 
.. 
cd50,OOO- 
cn 
w 
40,000- 
s 
H 
30,000. 
4 
vl 
H 
vl 20,000- 
10,000- 
Figure 4. Graph Plotted Between Stress Amplitude (Sa,psi) VS. 
Cycles to Failure. 
A Mean Stress = 46,778 psi 
o Mean Stress = 62,370 psi 
Mean Stress = 72,765 psi 
o Mean Stress = 83,160 psi \ \ 
x rve of Type 316 Stainless 
Steel (Notched Specimen) 
Mean Stress = 0 
0 
Figure 5. Fracture Surface of Sample No. 1. 36X, 
- -- 
Figure 6. Fracture Surface of Sample No. 2. 36X. 
Figure 7. Fracture Surface of Sample No. 3. 36X. 
- - 
Figure 8. Fracture Surface of Sarngle No. 4. 36:;. 
Figure 9. Fracture Surface of Sample No. 5. 36X. 
Figure 10. Fracture Surface of Sample No. 6. 36);. 
Figure 11. Fracture Surface of Sample No. 7. 36X. 
Figure 12. Fracture Surface of Sample No. 9. 36;~. 
RESULTS -AND DISCUSSION 
The results are discussed in two sections, micro- 
structural analysis and fatigue analysis. 
(a) Microstructural ~nalysis 
Macroscopic examination of the Eggers plate revealed 
no evidence of corrosion in the vicinity of the fracture. 
Microscopic examination revealed numerous cracks in the 
fracture surface (see Figures 13-15). Figures 16-18 show 
the micrographs of this device through frackured area. There 
was no evidence of stress-corrosion. By comparing Figure 15 
and Figure 16, both pictures having been taken at the same 
- - 
magnification and in the same area, it was found that the 
crack pattern of the fracture surface (Figure 16) was not 
- 
intergranular corrosion pattern since the grain size in 
Figure 15 was larger. Gray and Zirkle (23) showed in their 
examination of type 316 stainless steel Jewett nail that grain 
size of intergranular corrosion pattern and microstructure 
near the fracture are similar. 
The holes in this device were punched too close to 
the edge (see Figure 2). A reduced cross section can make 
a stress riser in the area where high stresses would occsr 
- - 
during walking, and cracks would initiate and propagate. 
Cahoon and Paxton 
(24) 
show that the entire section of the 
Figure 13. SEM Fractograph at 600~ of Fracture Surface 
of Eggers Plate. 
- - 
Figure 14. 5EM Fractograph at 1200X of Fracture Surface 
of Eggers Plate. 
Figure 15. SEM Fractograph at 6OOX of Fractur 
of Eggers Plate. 
Surface 
- - 
Figure 16. Micrograph of Eggers Plate Through Failure 
Area. 6OO;C. 
Figure 17. Micrograph of Eggers Plate Through Failure 
Area. 75X. - - 
Figure 18. Micrograph of Eggers Plate Through Failure 
Area. 150X. 
plate (type 316 stainless steel nail plate) between the 
hole and the edge is severely work hardened and much 
martensite forms. Their picture of plastic deformation 
and martensite in the plate is similar to Figures 16- 
18. The areas between the holes and edge of the plate, 
which have been severely deformed, may fracture with only 
a small amount of bending because both work-hardened 
material and martensite are harder and more brittle than 
the less work-hardened and un-transformed austenite 
which constitutes most of the implant. 
Also, slip bands and associated microcracks pattern 
shown in Figure 14 are similar to what Ligasor 
(25) 
found on a surface of a type 316 stainless steel Jewett 
nail. Results from this examination indicate extensive 
- - 
deformation and accelerated fatigue were chosed by the 
large grain size of the device. 
- 
Therefore, the design of the implant seems to be 
the major cause of failure in the Eggers plate. 
In the case of Jewett hip nail-plate the fracture 
surface was so badly disturbed that the cause of failure- 
could not be determined from either a SEM or metallo- 
graphic examination. A most unusual crack pattern was, 
however, found on the cylindrical surface immediately 
adjacent to the fracture surface (see Figures 19-22). 
- ~ 
Figures 23-28 show the crack patterns found on the 
cylindrical surfaces immediately adjacent to the fracture 
surfaces of the samples tested under low-cycle tension. 
Figure 19. Crack Pattern Found on Cylindrical Surface 
of Jewett Hip Nail-Plate (sEM), at 4OOX. 
Figure 20. Crack Pattern Found on Cylindrical Surface 
of Jewett Hip Nail-Plate (SEM), at 450X. 
Figure 21. 
Crack Pattern Found on Cylindrical Surface 
of Jewett Hip. Nail-Plate (SEM) , at 1200X. 
Figure 22. 
Crack.Pattern Found on Cylindrical Surface 
of Jel~ett Hip Nail-Plate (SEN) , at 1200X. 
Figure 23. Crack pattern Found on Cylindrical Surface 
of Sample No. 1. (sEM) , at 500X. 
- -- 
Figure 24. Crack Pattern ~ou1-12 on Cylindrical Surface 
of Sample No. 1. (SEN), at 800X. 
Figure 25. Crack Pattern Found on Cylindrical Surface 
of Sample No. 2. (SEN) , at 820X. 
Figure 26. Crack Pattern Found on Cylindrical Surface - - 
of Sample No. 2.(SEM), at 870X. 
Figure 27..Crack Pattern Found on Cylindrical Surface 
of Sample No. 3. (SEN' -' 380X. 
Figure 28. Crack Pattern Found on Cylindrical Surface 
- -- 
of Sample No. 3. (SEN'), at 840~ 
Figure 29. Crack Pattern Found on Cylindrical Surface 
of Sample No. 4. (SEM), at 400X. 
- .~- 
Figure 30. Crack Pattern Found on Cylindrical Surface 
of Sample No. $. (Sa) , at 2000X. 
Figure 31. 
Crack Pattern Found on Cylindrical Surface 
of Sample No. 5. (sEM), at 400~. 
Figure 32. 
- -- 
Crack Pattern Found on Cylindrical Surface 
Sample No. 
Figure 33. Crack Pattern Found on Cylindrical Surface 
of Sample No. 6. (SEM), at LOOX. 
- -- 
Fiqre 34. Crack Pattern Found on Cylindrical Surface 
of Sample No. 6. (SEMI, at 1900X. 
Figure 35. Crack Pattern Found on Cylindrical Surface 
of Sample No. 7. (sEM), at 800X. 
- -- 
Figure 36. 
Crack Pattern Found on Cylindrical Surface 
of Sample No. 
Figure '37. 
Crack Pattern Found on Cylindrical Surface 
of Sample No. 9. (SEM), at 500X. 
- ..- 
Figure 38. Crack Pattern Found on Cylindrical Surface 
of Sample No. 9. (SEM), at 1000X. 
None of these crack patterns is similar to what we found 
on Jewett hip nail-plate. This means that a simple 
low-cycle tension stress was not the cause of failure of 
this device. 
(b) Fatigue Analysis 
Fatigue surfaces of samples no. 1-4 are typical 
fatigue patterns (see Figures 5-8). Figures 39 and 
40 show fatigue striations of samples no. 1 and 4 
respectively on the fracture surface. A ductile pattern, 
caused by a high stress in a low number of cycles to 
failure, was found on each fracture surface of samples 
no. 5-7 and 9 (see Figures 9-12). As listed in Table 3, 
the ratio of ductile area to total fracture area 
increases as the mean stress increases. Since the ductile 
area (final fracture area) depends on the stress in,the 
- - 
last cycle, a larger ductile area is produced by a 
larger mean stress. 
- 
The accepted fatigue curve (26) of type 316 stainlc 
stainless steel (notched specimen)v:as plotted in Figure 4 
along with the results of this work in order to compare 
- - 
with the fatigue curves obtained in this experiment. 
Since the largest part of the published fatigue data 
was obtained from fully reversed cycling (mean stress is 
zero) 
(27) 
, a family of curves were plotted with each line 
corresponding to a different mean stress, as shown in 
- -. 
Figure 4, on the S-N diagram. The curves show that 
increased mean tensile stresses severely decrease the 
fatigue life of type 316 stainless steel (samples no. 3, 
Figure 39. SEM Fractograph at 4500X of Fracture Surface 
of Sample No. 1 , Showing Fatigue Striations. 
Figure 40. SEM Fractograph at; h$OOX of Fracture Surface -- 
of Sample No. 9 , Showing Fatigue Striations. 
5 and 7). The fatigue curve of samples no. 3, 8 and 9 
(mean stress of 47 ksi) shows that the endurance limit 
was reduced to about half of the actual endurance limit 
(no mean stress). 
Sander '28) stated that these curves (in Figure 4) 
are not-necessarily parallel to each other and it is 
not easy to predict th-e fatigue strength as influenced 
by mean stresses. Several empirical relations have been 
proposed but they are good only for the simple 
prcblems 
(29) 
A square wave cycle was used to approximate the 
force time profile of a human leg during normal level 
walking 
(30) 
. It was found that the mean stress 
during each square wave cycle is tensile since 
- - 
compressive forces which are a small part of the total 
range do not appear to influence the propagation of 
(31) 
fatigue cracks 
By combining the above fact with that 
type 316 stainless steel is very sensitive to presence 
of tensile mean stresses. It can be pointed out that 
any design and manufacture of type 316 stainless steel 
orthopedic devices must also consider these facts. 
CONCLUSION 
Type 316 stainless sfeel is very sens-itive to tKe 
presence of mean stresses. Tensile mean stresses can 
severely reduce its fatigue life at a given amplitude of 
loading in air at room conditions. 
Low-cycle tensile stresses do not produce the crack 
pattern observed in the two implants. For the Eggers 
plate, the failure seems to be a combination of inproper 
design and bending fatigue. In the case of Jewett nail, 
failure may have been due to overload which resulted from 
- - 
bending stresses. It is fairly certain that low-cycle 
tensile loads did not cause either implant to fail. 
The most important result of this work is the 
significance of the presence of tensile mean stresses that 
occur during normal walking of a human. Such stresses can 
markedly shorten the operation life of a type 316-stainless 
steel implant. 
APPENDIX 
Sample Calculation for Table 3 
Sample No. 1 
Mean Load 
Amplitude Load 
Diameter (initial) 
Mean Stress 
Amplitude Stress 
14,000 lbs. 
2,000 lbs . 
0.495 inch 
Mean Load 
Area 
14,000 lbs 
0.192 sq. in. 
72,765 psi 
2,000 lbs 
0.192 sq. in. 
10,395 psi 
REFERENCES 
1. R.J. Gray, Failure analyses of Surgical Implants 
from the Human Body can Improve Product 
and Performance Reliability, Research 
Sponsored by Union Carbide Corporation 
under Contact with Energy Research and 
Development Administration, 1978, p. 1. 
2. J.K. Wickstrom, Surgical Implants-Their Mechanical 
and Environmental Problems, Journal of 
Materials, Vol. 1, No. 2, June 1966, p. 367. 
3. D.C. Ludwigson, Requirement for Metallic Surgical 
Implants and Prosthetic Devices, Metals Eng. 
Quart. (Amer. Soc. Met.), August 1965, pp. 1-6. 
4. D.C. Ludwigson, Today's Prosthetic Metals, Journal 
- - 
of Metals, Vol. 16, March 1964, pp. 226-231. 
5. R.J. Gray, Bletallographic Examination of Retrieved 
- 
Intramedullary Bone Pins and Bone Screws 
from the Human Body, Report ORNL-TM-4068, 
Oak Ridge National Laboratory, February 1973. 
6. R.J. Gray and L.G. Zirkle Jr., Metallographic -- 
Examination of a Failed Jewett Nail-Plate 
from a Human Femur, Microstructural Science, 
Vol. 4, ed., E.W. Filer, J.M. Hoegfeldt and 
J. McCall, American Elsevier Publishing Co., 
Inc., New York, 1976, pp. 179-189. 
- - 
7. W.E. White and I. Le May, Optical and Electron Frac- 
tographic Studies of Fracture in Orthopedic 
Implants, Microstructure Science, Vol. 3, 
Part B, American Elsevier Publishing Co., 
Inc., New York, 1975, pp. 911-930. 
8. J.R..Cahoon and H.W. Paxton, Metallurgical Analyses 
of Failed Orthopedic Implants, J. Biomed 
Mater Res, Vol. 2, 1968, pp. 1-22. 
9. J.H. Dumbleton and E.H. Miller, Failures of ~etallic 
Orthopedic Implants, Metals Handbook, Vol. 10, 
8th edition, ASM, 1975, pp. 571-580. 
10. Gray, p. 232. 
11. B.W. Lisagor, Corrosion and Fatigue of Surgical 
Implants, ASTM Standardization News, Vol. 3(5), 
May 1975, pp. 20-24 and 43. 
- - 
12. Dumbleton and Miller, p. 578. 
13. Dumbleton and Miller, p. 577. 
- 
14. Dumbleton and Miller, p. 577. 
15. A.C. Fraker, A.W. Ruff and M.P. Yeager, Corrosion of 
Titanium Alloys in Physiological Solutions, 
-- 
Titanium Science and Technology, Year 4, ~ren&n 
Pub. Corp., 1973, p. 2447. 
16. Lisagor, p. 21. 
17. D. Warren, Microstructure and Corrision Resistance of 
Austenitic Stainless Steels, Presented at Liber- 
ty Bell Corrosion Course, NACE and Drexel 
Institute of Technology, September 1968. 
18. Dumbleton and Miller, p. 577. 
19. Gray and Zirkle, p. 186. 
20. Dumbleton and Miller, p. 577. 
21. K.R. Sheeler and L.A. James, Fatigue Behavior of Type 
316 Stainless Steel under Simulated Body con- 
. ditions, J. Biomed Mater Res., Vol. 5, 1971, - 
pp. 267-281. 
22. H. W. Russell and W.A. Welcker Jr., Damage and Overstress 
in the Fatigue of Ferrous Materials, Proc. Am. 
Soc. Test. Mat., Vol. 36, Part 2, 1936, pp. 118- 
138. 
23. Gray and Zirkle, p. 181. 
24. Cahoon and Paxton, p. 16. 
25. Lisagor, p. 23. 
26. Russell and Welcker Jr., p. 118. - - 
27. B. I. Sander, Foundamentals of Cyclic Stress and Strain, 
The University of Wisconsin Press, Madison, - 
Wisconsin, 1972, p. 92. 
28. Sander, p. 92 
29. P.G. Forrest, Fatigue of Metals, Pergamon Press, Oxford, 
-- 
1962. 
30. B. Bresler and J.P. Frankel, Transaction of the ASME, 
Vol. 72, 1950, pp. 27-36. 
31. C.M. Hudson and J.T. Scardina, Engineering Fracture 
Mechanics, Vol. 1, 1969, p. 429-446. - --