**3. Results and Discussion**

The maximum Tresca stress from metal-on-metal bearing with different metallic materials during normal walking cycle is presented in Figure 3. The comparison of the highest, average, and lowest Tresca stresses of metal-on-metal bearing materials under normal walking conditions are described in Figure 4. The maximum Tresca stress value changes due to the difference in the resultant force applied during loading in the normal walking condition, where the highest maximum Tresca stress value is in the seventh phase for all metal-on-metal bearings in this study.

**Figure 3.** Maximum Tresca stress for different metal-on-metal bearing materials in full cycle.

**Figure 4.** Comparison of highest, average, and lowest Tresca stress of metal-on-metal bearings materials on under normal walking condition.

Ti6Al4V-on-Ti6Al4V has the lowest Tresca stress value of 38.31 MPa. Tresca stress values were found to increase in CoCrMo-on-CoCrMo and SS 316L-on-SS 316L bearings by about 45.76% and 39.15%, respectively, compared to Ti6Al4V-on-Ti6Al4V. Apart from the magnitude of resultant force applied, Young's modulus also plays a significant role in the Tresca stress results. The greater Young's modulus of material under the same loading conditions will give a higher Tresca stress. Therefore, metal-on-metal bearings with CoCrMo material have the highest Tresca stress value under normal walking conditions compared to other bearing materials and vice versa for metal-on-metal bearings using Ti6Al4V material. The maximum Tresca stress values of various metal-on-metal bearings can be seen in Table 4.

**Table 4.** Maximum Tresca stress for different metal-on-metal bearing materials at peak loading.


Figure 5 illustrates the distribution contour of Tresca stress from Tresca terminology in ABAQUS [15]. The Tresca stress distribution contour is displayed using 5 selected phases from 32 phases in a normal walking cycle, namely the 1st phase that is the initial cycle, the 7th phase that is the phase with the highest gait loading, the 16th phase that is the middle of the cycle, the 30th phase that is the phase with the lowest gait loading, and the 32nd

phase that is the end of the cycle. It can be seen that the distribution contour of the Tresca stress will be wider and the value of the Tresca stress will be higher along with the greater the resultant force applied.

**Figure 5.** Tresca stress distribution on the acetabular cup for different metal-on-metal bearing materials at selected phases.

The relationship between Tresca stress and acetabular cup thickness on metal-on-metal bearings with different metallic materials during the peak phase and selected phase is illustrated in Figures 6 and 7. The highest Tresca stress does not occur in the surface contact but in the bulk area that is very prone to losing its elastic properties. In addition, the higher Tresca stress experienced by a material, the higher the probability of failure based on the Tresca failure theory [16]. In the results obtained, it is found that Ti6Al4V-on-Ti6Al4V has the lowest probability of failure with the lowest Tresca stress compared to the other two metal-on-metal bearings.

**Figure 6.** Tresca stress profile as a function of acetabular cup thickness for different metal-on-metal bearing materials at peak loading.

**Figure 7.** Tresca stress profile as a function of acetabular cup thickness for different metal-on-metal bearing materials at selected phases.

From the Tresca stress results obtained in the current implant design, various development efforts and further studies for metal-on-metal bearings need to be carried out to reduce the probability of failure. Apart from the selection of materials that have been carried out in the current study, several other efforts can be made, namely, examining the geometry parameters [14,17], the application of textured surfaces [18], and the use of coating layers [19]. The technical aspects of the surgeon's total hip replacement surgery with total hip arthroplasty also contributes to the long-term survival of the implant [20].

There are several focuses of attention to be conveyed regarding the limitations in our study. First, the coefficient of friction used is constant to represent the effect of lubrication that occurs, where the coefficient of friction should have a value that varies with time [13]. Second, the finite element model used is simple with a two-dimensional model that can reduce the accrual of the results compared with the results from three-dimensional model that are closer to the actual conditions [3]. Finally, the computational simulation does not take into account the range of motion during a normal walking cycle by only analyzing vertical loads that are irrelevant to actual conditions [11].
