**2. Resistance evaluation of titanium alloys to cyclic fatigue by dynamic tests with crevice corrosion stimulation**

The role of biomaterials is to aid or totally replace the functions of living tissues. In case of orthopedic implants, the loading response has to match the natural bone. The average load on a hip bone, estimated to be thrice the body weight, may increase to a value of 10 times the body weight during heavy exercise [15]. Therefore, the ideal orthopedic implant should manifest appropriate mechanical properties and be highly biocompatible with existing tissues [40]. In case of a metallic implant, the potential corrosion of the material in the body environment has to be considered.

Corrosion fatigue is defined as the process in which a metal fractures prematurely under conditions of simultaneous corrosion and repeated cyclic loading at lower stress levels or fewer cycles than would be required in the absence of the corrosion environment (ASTM G15-97).

The term cyclic dynamic test (fatigue) with crevice corrosion stimulation covers various phenomena, namely, crevice corrosion, fatigue, and tribocorrosion. The term stress corrosion (static) with crevice stimulation covers two entangled phenomena, namely, crevice corrosion and stress corrosion. Stress corrosion cracking is the result of a joint action between corrosion and a constraint of reaction or of static compression, applied or residual.

The aim of our research was to assess the mechanical properties of two titanium alloys currently used for orthopedic implants, namely, Ti6Al7Nb and Ti6Al4V, as reference.

The tested modular prostheses, type PL-06, consist of a distal and a proximal module, interlocked by a screw (**Figure 1**).

Two sample series were used for testing. The main characteristics are given in **Table 1**.

**Figure 1.** *The modular prostheses used for testing.*


#### **Table 1.**

*Main characteristics of the sample series.*

For each series, four samples were evaluated: three in cyclic dynamic test (fatigue) with crevice stimulation and one constrained with crevice stimulation (static) (**Table 2**).

The organization and coding of the samples subjected to the tests are given in **Table 2**. Samples # 1 and # 2 were used for the cyclic adjustments of the fatigue test machine and verification of electrochemical corrosion programs.

For the fatigue tests, a Walter & Bai AG, Switzerland, LFV 10 KN series machine, adapted for fatigue testing (**Figure 2a**) and research of biomedical implants (hip implant prostheses according to ISO 7206-4 and 6), was used. This compact testing system has a hydraulic power pack integrated in its base. The crosshead features automatic adjustment with hydraulic unlocking and hydraulic moving through two long stroke actuators. The prosthesis, more precisely the distal module, was embedded in a non-conductive composite resin, held by a metal sample holder (**Figure 2b**). The embedding of the tapered shape was up to 1 cm from the boundary between the two modules (distal module/proximal module). The specimens were loaded with an average stress of 1.4 and amplitude of 1.1 mm at a frequency of 10 Hz.

Two types of mechanical tests were conducted, the first one under dynamic loading for 5 million fatigue cycles and the second one under a static force of 981 N, during the equivalent time corresponding to 5 million dynamic fatigue cycles, which correspond to approximately 5 years of walking for a person with a bodyweight of 100 kg. The parameters of sample placement followed the requirements of the ISO 7206-6:2013(E) standard.

The potentiostatic measurement technique (controlled-potential coulometry), adapted according to the ASTM F746-87 standard, consists in performing an excitation at a given potential for a very short period of time and then positioning itself on a fixed potential for a certain time. The composition of a measurement cycle is shown


**205**

considerations:

**Table 3.**

**Figure 2.**

*adapted to the fatigue testing machine.*

*Composition of an electrochemical measurement cycle.*

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

in **Table 3**. Ten cycles correspond to 1 million cycles of mechanical fatigue. The test was carried out in increments of 1 million cycles for a total of 5 million cycles. The choice of the measuring technique is motivated by the following

*The fatigue testing system. (a) The fatigue testing machine and (b) the metal sample holder, specifically* 

**Potential type Level (mV) Time** Stimulated potential 800 60 s Preselected potential 600 36 min Stimulated potential 800 60 s Preselected potential 650 36 min Stimulated potential 800 60 s Preselected potential 700 36 min Stimulated potential 800 60 s Preselected potential 750 36 min Preselected potential −500 20 min

• The materials the components are made of do not pose any significant problems regarding the general corrosion resistance. On the other hand, electrolyte infiltration into the spaces of the distal/proximal module joint may lead to localized crevice corrosion, accompanied by a tribocorrosion process.

• The values of the electrical potential suggest that no decomposition process of

• A cathode potential equalizer at −500 mV SCE was added in order to analyze the depassivation-repassivation capacity of the materials in the tested areas.

• Using this technique, the electrical charge used for the experiment may be easily measured. For analytical estimations the total electrical charge passed in the experiment is easily related to the concentration of electroactive species in the cell.

the electrolyte into hydrogen and oxygen takes place.

**Table 2.**

*Organization and coding of the samples subjected to the tests.*

#### **Figure 2.**

*Engineering Steels and High Entropy-Alloys*

*Main characteristics of the sample series.*

of the ISO 7206-6:2013(E) standard.

**Sample series no. Code Mechanical** 

#1 #2

*Organization and coding of the samples subjected to the tests.*

(static) (**Table 2**).

**Table 1.**

For each series, four samples were evaluated: three in cyclic dynamic test (fatigue) with crevice stimulation and one constrained with crevice stimulation

Surface No specific treatment type 2 ("Ti anodizing") Tolerances Uniforms Adapted to each cone level

machine and verification of electrochemical corrosion programs.

Composition Ti6Al7Nb Ti6Al4V

average stress of 1.4 and amplitude of 1.1 mm at a frequency of 10 Hz.

Two types of mechanical tests were conducted, the first one under dynamic loading for 5 million fatigue cycles and the second one under a static force of 981 N, during the equivalent time corresponding to 5 million dynamic fatigue cycles, which correspond to approximately 5 years of walking for a person with a bodyweight of 100 kg. The parameters of sample placement followed the requirements

The potentiostatic measurement technique (controlled-potential coulometry), adapted according to the ASTM F746-87 standard, consists in performing an excitation at a given potential for a very short period of time and then positioning itself on a fixed potential for a certain time. The composition of a measurement cycle is shown

**solicitation**

Simulation, setting tests

 #3 Fatigue Yes #4 Fatigue Yes #5 Fatigue Yes #6 Constraint Yes #7 Fatigue Yes #8 Fatigue Yes #9 Constraint Yes #10 Fatigue Yes

**Crevice electrochemical test**

Simulation, test, and control of the corrosion program

The organization and coding of the samples subjected to the tests are given in **Table 2**. Samples # 1 and # 2 were used for the cyclic adjustments of the fatigue test

For the fatigue tests, a Walter & Bai AG, Switzerland, LFV 10 KN series machine, adapted for fatigue testing (**Figure 2a**) and research of biomedical implants (hip implant prostheses according to ISO 7206-4 and 6), was used. This compact testing system has a hydraulic power pack integrated in its base. The crosshead features automatic adjustment with hydraulic unlocking and hydraulic moving through two long stroke actuators. The prosthesis, more precisely the distal module, was embedded in a non-conductive composite resin, held by a metal sample holder (**Figure 2b**). The embedding of the tapered shape was up to 1 cm from the boundary between the two modules (distal module/proximal module). The specimens were loaded with an

**First sample series Second sample series**

**204**

**Table 2.**

*The fatigue testing system. (a) The fatigue testing machine and (b) the metal sample holder, specifically adapted to the fatigue testing machine.*


#### **Table 3.**

*Composition of an electrochemical measurement cycle.*

in **Table 3**. Ten cycles correspond to 1 million cycles of mechanical fatigue. The test was carried out in increments of 1 million cycles for a total of 5 million cycles.

The choice of the measuring technique is motivated by the following considerations:


The potentiostat used is a model PAR 273A, EG&G (Princeton Applied Research). The electrochemical cell has been specially designed for these types of measurements. It is a cell with three electrodes: working electrode (green wire), platinum counter electrode (red wire), and the saturated calomel reference electrode on the right (**Figure 3a**). It is fitted on the head of the cyclic fatigue machine (**Figure 3b**). The test environment was a solution of NaCl at a concentration of 9 g/l (ASTM F746-1998) in ultrapure water (electrical resistivity 18 MΩ cm).

After the assembly of the two modules, there will always be a space which will allow the diffusion of the fluids (**Figure 4a**). Thus the presence of fluids in the interstice can generate crevice corrosion. After the fatigue corrosion test, it is possible to notice the corrosion by strong staining (**Figure 4b**).

Behavior to localized corrosion of samples #3, #4, and #5 (series 1) during 5 million mechanical fatigue cycles is shown in **Figure 5a**.

Behavior to localized corrosion of samples #7, #8, and #10 (series 2) during 5 million mechanical fatigue cycles is shown in **Figure 5b**. Sample #6 and #9 were evaluated for corrosion resistance without cyclic dynamic forces but under a load of 100 kg (**Figure 5**). The comparative behavior to localized corrosion for samples #6 and #9 is presented in **Figure 6**.

**Figure 3.** *The electrochemical cell adapted on LFV 10 KN machine.*

#### **Figure 4.**

*Sample #3 before and after testing. (a) (A) Distal module and (B) proximal module. Before testing. (b) Coloration of the two modules after the corrosion test.*

**207**

**Figure 7.**

*Electrical charges accumulated during testing.*

**Figure 5.**

**Figure 6.**

*(series 2-cyclic fatigue dynamic test).*

*Behavior to localized corrosion of samples #6 and #9 (static).*

By analyzing the sum of the accumulated charges during the total duration of the test, we note a worse behavior of the series 1 samples compared to series 2

*Behavior to localized corrosion of tested samples. (a) Behavior to localized corrosion of samples #3, #4, #5, and #6 (series 1-cyclic fatigue dynamic test) and (b) behavior to localized corrosion of samples #7, #8 #9, and #10* 

(**Figure 7**) (samples #6 and #9 were tested under a stress of 100 kg).

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

#### **Figure 5.**

*Engineering Steels and High Entropy-Alloys*

and #9 is presented in **Figure 6**.

*The electrochemical cell adapted on LFV 10 KN machine.*

*Coloration of the two modules after the corrosion test.*

The potentiostat used is a model PAR 273A, EG&G (Princeton Applied Research). The electrochemical cell has been specially designed for these types of measurements. It is a cell with three electrodes: working electrode (green wire), platinum counter electrode (red wire), and the saturated calomel reference electrode on the right (**Figure 3a**). It is fitted on the head of the cyclic fatigue machine (**Figure 3b**). The test environment was a solution of NaCl at a concentration of 9 g/l

(ASTM F746-1998) in ultrapure water (electrical resistivity 18 MΩ cm).

sible to notice the corrosion by strong staining (**Figure 4b**).

million mechanical fatigue cycles is shown in **Figure 5a**.

After the assembly of the two modules, there will always be a space which will allow the diffusion of the fluids (**Figure 4a**). Thus the presence of fluids in the interstice can generate crevice corrosion. After the fatigue corrosion test, it is pos-

Behavior to localized corrosion of samples #3, #4, and #5 (series 1) during 5

Behavior to localized corrosion of samples #7, #8, and #10 (series 2) during 5 million mechanical fatigue cycles is shown in **Figure 5b**. Sample #6 and #9 were evaluated for corrosion resistance without cyclic dynamic forces but under a load of 100 kg (**Figure 5**). The comparative behavior to localized corrosion for samples #6

*Sample #3 before and after testing. (a) (A) Distal module and (B) proximal module. Before testing. (b)* 

**206**

**Figure 4.**

**Figure 3.**

*Behavior to localized corrosion of tested samples. (a) Behavior to localized corrosion of samples #3, #4, #5, and #6 (series 1-cyclic fatigue dynamic test) and (b) behavior to localized corrosion of samples #7, #8 #9, and #10 (series 2-cyclic fatigue dynamic test).*

**Figure 6.** *Behavior to localized corrosion of samples #6 and #9 (static).*

By analyzing the sum of the accumulated charges during the total duration of the test, we note a worse behavior of the series 1 samples compared to series 2 (**Figure 7**) (samples #6 and #9 were tested under a stress of 100 kg).

It should be noted that the curves presented in **Figure 5a** and **b** are very different from one series to another. The harmonic shape of the series 2 curves is probably due to the surface treatment.

Since the last stage of the measurement cycle (**Table 3**) is −500 mV vs. SCE; thus a depassivation cathodic plateau, the repassivation capacity of the layer designed for the first series samples shows an ability of slower repassivation. In the first stage (600 mV), one still finds higher amounts of electrical charges compared to the later stage (750 mV) vs. SCE (**Tables 4** and **5**). The curves of the series 2 samples have no correlation with the testing fatigue cyclic motion.

In case of series 1 samples, after about 3 million cycles of the dynamic tests, crystallized sodium chloride can be found in the area of the tightening screws of the modular prostheses (**Figure 8**). This means that there is an electrolyte pumping effect in the assembly space of the two parts, and therefore crevice corrosion is quite possible to occur. For series 2 samples, this process can be noticed earlier, starting from about 2.5 million cycles. In the case of electrochemical static measurements, no phenomena of electrolyte pumping were observed. The quantities of electrical charges consumed in the corrosion process (**Tables 4** and **5**) are greater in series 1 than in series 2 (**Figure 7**).

When interpreting the microscopic observations, it is essential to take into account that the various phenomena involved, namely, crevasse corrosion, fatigue (cyclic dynamic test), stress (static test), and tribocorrosion (cyclic dynamic test), are impossible to be considered separately. These processes are intimately intertwined by complex mechanisms; observation is limited to their combined effects. This is why the presented phenomenon is referred to as corrosion fatigue.

The distal module of sample #3, observed by scanning electron microscopy (SEM) at a 560× magnification, is presented in **Figure 9**.

In the analyzed area, cracks have developed (**Figure 9a**), which are probably due to the cyclic dynamic process (fatigue), accompanied by the corrosion processes (**Figure 9b**).

In case of samples #4 and #5, similar phenomena may be observed. In case of sample 4, the deposition is present at the level of the crevice as well as the interferential colorations (same as in **Figure 4b**). The EDX analysis of the area reveals only Al, Ti, and Nb.


#### **Table 4.**

*Quantity of electrical charge (mC) consumed for the first million fatigue cycles during testing of sample #4, series 1.*

**209**

**Figure 9.**

*SEM 560×.*

**Table 5.**

*series 2.*

**Figure 8.**

Sample #5 shows gaps in the distal part (**Figure 10b**) at the level of the crevice as well as coloration at the same level on the proximal module (**Figure 10a**). The spectrum EDX analyses 1, 2, and 3 (**Figure 10c**) reveal depassivation (lack of oxygen in spectrum 1) of the gap interior; the gap's margins contain Na, Cl, Si, and K.

*Sample #3. Distal module. (a) Sample #3. Distal module. SEM 560× and (b) sample #3. Distal module.* 

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

**Cycle no. 600 mV 650 mV 700 mV 750 mV Total** 37.3 21.9 17.8 16.5 93.5 12.3 7.4 7.3 8.1 35.1 10.1 5.3 5.1 5.7 26.2 8.9 4.1 3.8 4.0 20.8 8.4 3.5 3.0 3.3 18.2 8.2 3.1 2.7 2.9 16.9 7.8 2.9 2.4 2.6 15.7 7.9 2.8 2.3 2.5 15.5 7.9 2.7 2.2 2.4 15.2 8.1 2.7 2.2 0.9 13.9 Total 116.9 56.4 48.8 48.9 271.0

*Quantity of electrical charge (mC) consumed for the first million fatigue cycles during testing of sample #10,* 

*Sample #3. (a) Salt deposition in sample #3 and (b) crack of the distal module of sample #3.*

**Cycle no. 600 mV 650 mV 700 mV 750 mV Total** 37.3 21.9 17.8 16.5 93.5 12.3 7.4 7.3 8.1 35.1 10.1 5.3 5.1 5.7 26.2 8.9 4.1 3.8 4.0 20.8 8.4 3.5 3.0 3.3 18.2 8.2 3.1 2.7 2.9 16.9 7.8 2.9 2.4 2.6 15.7 7.9 2.8 2.3 2.5 15.5 7.9 2.7 2.2 2.4 15.2 8.1 2.7 2.2 0.9 13.9 Total 116.9 56.4 48.8 48.9 271.0

#### *Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

**Table 5.**

*Engineering Steels and High Entropy-Alloys*

due to the surface treatment.

than in series 2 (**Figure 7**).

(**Figure 9b**).

Al, Ti, and Nb.

It should be noted that the curves presented in **Figure 5a** and **b** are very different from one series to another. The harmonic shape of the series 2 curves is probably

Since the last stage of the measurement cycle (**Table 3**) is −500 mV vs. SCE; thus a depassivation cathodic plateau, the repassivation capacity of the layer designed for the first series samples shows an ability of slower repassivation. In the first stage (600 mV), one still finds higher amounts of electrical charges compared to the later stage (750 mV) vs. SCE (**Tables 4** and **5**). The curves of the series 2

In case of series 1 samples, after about 3 million cycles of the dynamic tests, crystallized sodium chloride can be found in the area of the tightening screws of the modular prostheses (**Figure 8**). This means that there is an electrolyte pumping effect in the assembly space of the two parts, and therefore crevice corrosion is quite possible to occur. For series 2 samples, this process can be noticed earlier, starting from about 2.5 million cycles. In the case of electrochemical static measurements, no phenomena of electrolyte pumping were observed. The quantities of electrical charges consumed in the corrosion process (**Tables 4** and **5**) are greater in series 1

When interpreting the microscopic observations, it is essential to take into account that the various phenomena involved, namely, crevasse corrosion, fatigue (cyclic dynamic test), stress (static test), and tribocorrosion (cyclic dynamic test), are impossible to be considered separately. These processes are intimately intertwined by complex mechanisms; observation is limited to their combined effects.

The distal module of sample #3, observed by scanning electron microscopy

In the analyzed area, cracks have developed (**Figure 9a**), which are probably due to the cyclic dynamic process (fatigue), accompanied by the corrosion processes

In case of samples #4 and #5, similar phenomena may be observed. In case of sample 4, the deposition is present at the level of the crevice as well as the interferential colorations (same as in **Figure 4b**). The EDX analysis of the area reveals only

**Cycle no. 600 mV 650 mV 700 mV 750 mV Total** 108.2 69.7 54.1 50.3 282.3 30.9 26.3 25.9 25.3 108.4 23.0 18.4 18.9 20.3 80.6 19.3 15.4 16.0 16.7 67.4 15.7 13.2 13.3 14.8 57.0 10.8 10.5 10.2 10.5 42.0 10.7 8.3 7.9 8.6 35.5 8.7 6.8 7.3 7.8 30.6 8.1 6.4 6.4 6.8 27.7 7.8 5.7 5.9 6.1 25.5 Total 243.2 180.7 165.9 167.2 757.0

*Quantity of electrical charge (mC) consumed for the first million fatigue cycles during testing of sample #4,* 

This is why the presented phenomenon is referred to as corrosion fatigue.

(SEM) at a 560× magnification, is presented in **Figure 9**.

samples have no correlation with the testing fatigue cyclic motion.

**208**

**Table 4.**

*series 1.*

*Quantity of electrical charge (mC) consumed for the first million fatigue cycles during testing of sample #10, series 2.*

**Figure 8.**

*Sample #3. (a) Salt deposition in sample #3 and (b) crack of the distal module of sample #3.*

*Sample #3. Distal module. (a) Sample #3. Distal module. SEM 560× and (b) sample #3. Distal module. SEM 560×.*

Sample #5 shows gaps in the distal part (**Figure 10b**) at the level of the crevice as well as coloration at the same level on the proximal module (**Figure 10a**). The spectrum EDX analyses 1, 2, and 3 (**Figure 10c**) reveal depassivation (lack of oxygen in spectrum 1) of the gap interior; the gap's margins contain Na, Cl, Si, and K. At the distal/proximal part interface of the distal module, deposits may be observed. The EDX analyses particularly show the presence of elements C, Na, Si, Cl, and K (**Figure 10d**).

Sample #7 reveals the same phenomenon of electrolyte penetration at the interface of the distal/proximal modules. The optical examination establishes the presence of wear and corrosion in this area, revealed by a rough appearance of the proximal module surface, as well as on the distal module but to a lesser extent. Examination of the distal module does not reveal the presence of cracks. The phenomenon is probably due to tribocorrosion and crevice corrosion.

Sample #8 also shows an increase in the roughness of the surface (**Figure 11a** and **b**). **Figure 11c** and **d** show the EDX spectrum analysis in the rough zone of the distal modules and reveal the presence of Na, Cl, and Si elements.

In case of sample #9, having undergone only a static test, the electrolyte does not enter the distal/proximal module interface, and no corrosion phenomenon is highlighted. These remarks are also valid for sample #6, which was subjected to the same static test. The EDX analysis does not reveal any traces of corrosion, the chemical composition being the same in the three zones subjected to evaluation.

**Figure 12a** shows a sectional view of the proximal module of sample #10; as in case of samples #7 and #8, salt deposits and a phenomenon of wear and/ or corrosion in the crevice area are present. On the other hand, the distal module (**Figure 12b**) is much less marked by this phenomenon than the distal module of sample # 8. The EDX analysis in **Figure 12d** is measured on the distal module in the crevice area (**Figure 12c**).

According to the results obtained, the electrochemical quantities are examined, and the optical observations reveal a better corrosion behavior on the part of the series 2 samples (Ti6Al4V—anodized type 2) compared to the series 1 samples (Ti6Al7Nb).

**211**

**Figure 12.**

*module. Crevice area and (d) spectrum EDX analysis.*

**Figure 11.**

*Spectrum EDX analysis.*

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

*Sample #8. (a) Sample #8. Proximal module. Rough area, (b) sample #8. Distal module. Rough area, (c) sample #8. Distal module. SEM 560×. Spectrum EDX analysis and (d) sample #8. Distal module. SEM 560×.* 

*Sample #10. (a) Sample #10. Proximal module, (b) sample #10. Distal module, (c) sample #10. Distal* 

**Figure 10.**

*Sample #5. (a) Sample #5. Proximal module, (b) sample #5. Distal module, (c) sample #5. Distal module. SEM 230× and (d) spectrum EDX analysis.*

*Multicomponent Alloys for Biomedical Applications DOI: http://dx.doi.org/10.5772/intechopen.88250*

#### **Figure 11.**

*Engineering Steels and High Entropy-Alloys*

Cl, and K (**Figure 10d**).

crevice area (**Figure 12c**).

(Ti6Al7Nb).

At the distal/proximal part interface of the distal module, deposits may be

phenomenon is probably due to tribocorrosion and crevice corrosion.

distal modules and reveal the presence of Na, Cl, and Si elements.

observed. The EDX analyses particularly show the presence of elements C, Na, Si,

Sample #7 reveals the same phenomenon of electrolyte penetration at the interface of the distal/proximal modules. The optical examination establishes the presence of wear and corrosion in this area, revealed by a rough appearance of the proximal module surface, as well as on the distal module but to a lesser extent. Examination of the distal module does not reveal the presence of cracks. The

Sample #8 also shows an increase in the roughness of the surface (**Figure 11a** and **b**). **Figure 11c** and **d** show the EDX spectrum analysis in the rough zone of the

In case of sample #9, having undergone only a static test, the electrolyte does not enter the distal/proximal module interface, and no corrosion phenomenon is highlighted. These remarks are also valid for sample #6, which was subjected to the same static test. The EDX analysis does not reveal any traces of corrosion, the chemical composition being the same in the three zones subjected to evaluation. **Figure 12a** shows a sectional view of the proximal module of sample #10; as in case of samples #7 and #8, salt deposits and a phenomenon of wear and/ or corrosion in the crevice area are present. On the other hand, the distal module (**Figure 12b**) is much less marked by this phenomenon than the distal module of sample # 8. The EDX analysis in **Figure 12d** is measured on the distal module in the

According to the results obtained, the electrochemical quantities are examined, and the optical observations reveal a better corrosion behavior on the part of the series 2 samples (Ti6Al4V—anodized type 2) compared to the series 1 samples

*Sample #5. (a) Sample #5. Proximal module, (b) sample #5. Distal module, (c) sample #5. Distal module.* 

**210**

**Figure 10.**

*SEM 230× and (d) spectrum EDX analysis.*

*Sample #8. (a) Sample #8. Proximal module. Rough area, (b) sample #8. Distal module. Rough area, (c) sample #8. Distal module. SEM 560×. Spectrum EDX analysis and (d) sample #8. Distal module. SEM 560×. Spectrum EDX analysis.*

**Figure 12.**

*Sample #10. (a) Sample #10. Proximal module, (b) sample #10. Distal module, (c) sample #10. Distal module. Crevice area and (d) spectrum EDX analysis.*

In cyclic dynamic tests with crevice stimulation, the electrolyte enters the interface between the distal and proximal modules, which is not the case during static tests (#6 and #9).

Samples #3 and #4 of series 1 reveal cracks in the distal module. Samples #3 and #5, also series 1, reveal holes in the crevice proximity. Metallic interferential staining of the distal/proximal module interfaces of the series 1 samples (#3, #4 and #5) is indicative of electrolyte reactions with the substrate and helps highlighting the corrosion process. This coloration does not appear in case of series 2 samples.

Series 2 samples (#7, #8, and #10) do not show cracks or holes as observed in case of series 1 samples. On the other hand, at crevice level, the surface of the proximal module and to a lesser extent the surface of the distal module present an increase of roughness after the cyclic dynamic corrosion test with crevice stimulation. This phenomenon is particularly visible on sample #8.

The observation of the samples only subjected to the static test does not reveal any sign of corrosion.
