Quality Control

To turn SLM into a manufacturing technique for implants or prostheses, some important conditions have to be fulfilled. The laser-melted parts have to meet strict requirements regarding mechanical and chemical properties and the process must guarantee high accuracy. Next to these quality requirements the breakthrough of SLM as a medical Rapid Manufacturing technique will depend on reliability, performance and economical aspects like production and cost. K.U.Leuven has fully characterized the SLM process for cobalt-chromium and titanium alloys (Vandenbroucke et al., 2006).

Many mechanical tests have been performed on both materials. Figure 7.11 shows for example stress-strain curves of tensile tests on titanium SLM samples. The graphs prove high repeatability for mechanical properties. The table in Fig. 7.11 summarizes all measured mechanical properties for Ti-6Al-4V SLM samples. The obtained properties are compared with those of bulk material, STA

















Bulk STA

Density (kg/m3)



Hardness (HV)



Young's modulus (GPa)



Shear modulus (GPa)



Tensile yield str. (MPa)



Ult. tensile str. (Mpa)



Elong. at rupture (%)



Bending yield str. ( MPa)



Ult. bending str. (MPa)



Fig. 7.11 (left) Stress-strain graphs of tensile tests for titanium samples; (right) Mechanical properties of titanium SLM samples and of bulk material STA (= Solution Treated Aged) (Source: K.U.Leuven)

(Solution Treated Aged), from literature. The results for SLM are similar to the stated properties of the bulk material, except for elongation at rupture. The ductility is somewhat lower because during the SLM process the melt pool cools down very rapidly when the laser beam has passed, leading to slight embrittle-ment. The full mechanical analysis shows that the SLM material fulfills the required strength and stiffness for the dental framework applications.

Both selected materials have also been tested for their chemical properties. The corrosion behavior is of high interest to value biocompatibility. Titanium as well as cobalt-chromium alloys are considered as corrosion resistant and biocompatible materials. Nevertheless, the very complex chemistry of the oral cavity may reveal surprises concerning corrosion processes (Strietzel et al., 1998). The corrosion characteristics of SLM samples were examined by static immersion tests in a corrosion solution. The solutions were analyzed by ICP-OES (Inductively Coupled Plasma-Optic Emission Spectrometry) to determine the different ion emissions. Figure 7.12 shows for example the corrosion rate of titanium as a function of time. The emission of Ti-ions decreases rapidly within the first few days to approach a final low value, similar to corrosion rates of bulk titanium material. These chemical analyses demonstrate favorable corrosion behavior for the SLM samples.

Dimensional analyses on benchmark parts have been performed to test the SLM process accuracy. For parts smaller than a dental work model (60 x 60 x 30 mm3) the overall point-to-point accuracy is below 60 mm. In dental literature there is a growing consensus that the final gap size between framework and implant should not exceed 40 mm to avoid failures of the prosthetic system (Renouard, 1999). Therefore, the accuracy of the fit between framework and implants has been analyzed. When the framework is screw-retained on all implants very slightly, the mean gap size between framework and implants amounts to 25 mm. This means that SLM guarantees the needed precision by optimal process parameters and an appropriate production strategy.

The developed procedure allows an efficient production of complex frameworks by SLM. The build time per framework declines for increasing number of frameworks produced during the same build, because powder-depositing time is spread

Fig. 7.12 Titanium corrosion rate of SLM samples as a function of time (Source: K.U.Leuven)

time (days)

Fig. 7.13 Build time per framework versus number of produced frameworks (Source: K.U.Leuven)

1 4 8 14 number of parts

over all frameworks (Fig. 7.13). When eight frameworks are produced during one production run, the build takes sixteen hours or two hours per framework, which is half of the build time when only one framework is produced. A lot of time and money can thus be saved by producing multiple unique frameworks in a single production run, leading to mass customization and to a lower price.

0 0

Post a comment