Page 55 - MSAM-3-1
P. 55

Materials Science in Additive Manufacturing                       Adhesion study for multi-material 3D printing



            fractured surfaces of the opposing side (the material that is   metrics. We evaluated the SRI (R Sa-A  and R ) of the fractured
                                                                                                a
            printed last) of the coupons with unmodified interface and   surfaces of the material that is being printed first, as shown in
            top infill modifications. More peaks were observed on the   Figure 7A and B. The SRI shows that there is a slight increase
            fractured surfaces of these coupons, suggesting possible   in R Sa-A  and R  when comparing the coupons with top infill
                                                                          a
            impregnation of material from the top material into the   modifications to the baseline coupons that have unmodified
            grooves  on  the  top  layer  of  the  bottom  material.  These   interface. This also helps explain why there is a slight increment
            impregnated material increases the surface area between   in the interlaminar adhesion strength of the coupons with top
            the materials at the interface, which are responsible for   infill modification. The highly uneven fractured surface of the
            the adhesion through surface adsorption and mechanical   coupons with interlocking features can also be reflected by the
            interlocking.  Figure  6C,  on the other  hand,  shows  the   high magnitude of the SRI. Likewise, a comparable trend was
            surface morphology of the coupon with interlocking   observed on the SRI of the corresponding opposite surfaces
            features. It is observed that these coupons generally have   of the fractures, as these surfaces would essentially constitute
            much rougher surfaces regardless of the print order   the negative imprint of the fracture patterns observed on the
            compared with the other two cases due to uniform fracture   initial side, as shown in Figure S1.
            lines that form across the interface.                In general, the presence of internal voids within the
              To enhance our quantitative analysis of the fractured   structure can lead to mechanical imperfections, which, in
            surfaces, we employed two specific metrics to evaluate   turn, contribute to stress concentration and facilitate crack
            surface roughness: (i) the ratio of surface area to projected   propagation under tensile load. This phenomenon results
            area (R Sa-A ) and (ii) the average surface roughness (R ), as   in diverse tensile behaviors, as these voids become critical
                                                      a
            shown in Figure 7. The R Sa-A  metric serves to quantify the   factors in determining the material’s mechanical response
            effective surface area available for adsorption phenomena,   during testing. A micro-CT analysis was conducted on the
            providing insights into the potential interfacial adhesion   three types of coupons to understand how different interface
            capacity. On the other hand, R offers a measure of the mean   designs affect the formation of pores and defect within the
                                   a
            surface texture or topography created during fracture. This   printed structures near the interface (Figure  8A). Here,
            parameter is instrumental in assessing the extent to which   we observed that coupons with no modifications have a
            mechanical interlocking contributes to surface adhesion,   porosity level of approximately 1.47% near the interface
            as it reflects the microscale irregularities that can enhance   region, which is slightly higher compared to that of the
                                                               0.64% of the fully printed regions. Similarly, the coupon
            mechanical bonding between the interfaces for the case of   with interlocking features also has a porosity level of
            samples with clean separation. Although the evaluation   approximately 5.6%. Likewise, the coupons with top infill
            may not be meaningful for the coupons with interlocking   modifications exhibit a porosity level of approximately
            features, the values are also reported here for the sake of   2.1% only (Figure 8B), which is slightly higher than that
            comprehensiveness and comparison purposes.
                                                               of  the  coupons with unmodified interface used  in  this
              In  general,  it  is  observed  that  both  R Sa-A  and  R exhibit   work. A  comparison of the ultimate tensile strength of
                                                    a
            almost similar trend for all cases. To ensure conciseness,   the samples to the corresponding surface properties and
            we combine the discussion of both metrics and use surface   the porosity level at the interface is presented in Table 4.
            roughness indicator (SRI) as the general term to describe both   The pronounced influence of porosity on the ultimate


                         A                                   B















            Figure 7. Graphs showing the (A) surface area/area ratio (the error bars represent standard error) and (B) surface roughness (R ) for the interfaces that are
                                                                                            a
            printed first. Abbreviations: cPLA: conductive polylactic acid; TPU: Thermoplastic polyurethane.


            Volume 3 Issue 1 (2024)                         10                      https://doi.org/10.36922/msam.2672
   50   51   52   53   54   55   56   57   58   59   60