Polishing 3D printed medical implants

Share this on social media:

Laser polishing could serve as a resourceful alternative to traditional methods for finishing additively manufactured parts

Metal powder-based additive manufacturing has continually grown in popularity as a production process in recent years due to the flexibility and design freedom it offers. Such benefits have led to the process being used to produce medical implants. It enables implants suited to a particular patients’ needs to be designed and manufactured rapidly – vital if a short turnaround time is needed, should the patient require the implant for surgery.

This flexibility comes with the disadvantage of having a rough, low-quality surface finish, leading to a number of postprocesses having to be carried out before the implant is ready for use.

Dental implants are often finished using electrochemical polishing. While this technique has the advantage of being highthroughput, especially considering the intricate design of dental implants, it does require hazardous chemicals – and therefore their storage and disposal. It is also a non-selective technique, meaning the whole dental implant must be polished. This is not desirable, as while the abutment of the implant must be smooth to prevent bacterial growth on the gum, the crownretaining part must be kept rough to enable bonding of the crown. This requires additional post-processing to reintroduce the roughness. Electrochemical polishing is also not suited to thin, smaller parts, as it can damage them.

For cranial implants – large, slightly rounded structures – due to their freeform nature and thinness, mechanical polishing is currently used to finish them. This is a semimanual process that takes a skilled worker many hours and numerous process steps to complete. It’s also not suited to intricate parts.

In collaboration with Renishaw, researchers at Heriot-Watt University in Edinburgh, Scotland are developing laser polishing techniques (see figure 1) that negate the downsides of such polishing methods. They are targeting parts produced using selective laser melting built from either titanium alloy (Ti-6Al-4V) or cobalt chrome (Co-Cr). In recent work, the partners have developed optimised laser polishing parameters using these two materials to provide a high-quality surface finish at reasonable process rates.

‘Laser polishing works by using a high-power laser beam to focus onto a metal surface and create a localised melt pool,’ explained Mark McDonald, a PhD student at HeriotWatt’s Institute of Photonics and Quantum Sciences. ‘The surface tension of the melt pool pulls the surface together, and as the laser scans across the surface, the metal cools down again and resolidifies as a nice, smooth surface.’

As with other laser processes, one of the advantages of laser polishing is that it can be automated. In addition, with the polishing occurring via the melting and cooling of a metal surface, no waste products are produced that later require disposal. The process can also be used selectively to polish intricate shapes of different sizes, as only the parts touched by the beam are polished. Laser polishing also has minimal impact on the surface it is treating, introducing no structural defects.

Weaker focus, stronger results

According to the researchers, the optimisation of laser polishing parameters for various materials has generally been carried out on planar surfaces, in contrast to the actual parts produced using additive manufacturing that are typically not flat and hence have significant height variation across the surface. Using a typical focusing lens and a scan head setup (figure 2b) the spot diameter (and hence the power density) of the incident beam on the surface will be strongly dependent on this height, unless active focus variation compensation is introduced.

Figure 2. (a) the weakly focused telescope setup and (b) the scan head setup. (Image: McDonald et al)

‘To combat this we designed a telescope system that creates a “weakly focussed” beam with low divergence, which provides a more consistent energy density,’ said McDonald.

This system (figure 2a), which used a 100W CW fibre laser, proved much more advantageous than a scan head and focussing lens system, enabling polishing without required additional focus control. To compare the performance of this setup with the focusing lens plus scan head setup, a cobalt chrome cranial implant plate was split into four quadrants, as shown in figure 3. The scanning speed of the laser beam was 16.6mms-1 in all cases, calculated to maintain an energy density of 1.5kJcm−2. A hatch spacing of 280μm was used to maintain a 30 per cent line overlap.

Figure 3. The two setups of figure 2 were tested on a cranial implant to create different surface finishes in each quadrant. (Image: McDonald et al)

For each laser polished region, the surface height map was measured in two regions: bottom, corresponding to the section of the cranial implant where the scan head setup provides a 400μm beam diameter, and top, where the beam diameter from the scan head setup incident on the sample is smaller. Comparing the bottom and top regions for each quadrant shows that when the scan head is used, the top regions are less smooth compared to those polished with the telescope system.

Dental implant polishing

To polish cylindrical parts such as the dental implants, the researchers needed to rotate and move them under a stationary laser spot. To do this, a rotary arrangement was developed that uses a DC motor combined with a linear stage. The rotation was controlled to maintain the required energy density, while the line overlap was controlled using the linear stage travelling at a constant velocity, set to have moved one hatch spacing in the time for one full rotation of the part. This maintained a continuous spiral along the entire length of the component.

For this experiment, the researchers trialled replacing the fibre laser source with a fibre delivered diode array laser with a maximum power of 150W to achieve a more cost-effective solution. The wavelength of this diode array was 915nm, compared with the 1,064nm of the fibre laser.

Despite resulting in a lower beam quality, the almost top-hat shaped beam profile of the multimode fibre delivery aided the laser polishing process. This is because it provided a more uniform temperature gradient in the melt pool, compared with the Gaussian shaped beam of the fibre laser.

The researchers found that the diode array laser setup delivered comparable results to the fibre laser setup.

The size and thickness of dental implants can vary depending on the tooth that is being replaced. This can affect the heat accumulation and thus the difference in laser parameters required to polish each implant. Two variants (smaller and larger) of cobalt chrome implants were therefore studied by the researchers, as shown in figure 4, using the rotational setup for processing cylindrical parts. They found that the smaller implant required lower energy density (2kJcm−2) and less processing time (60s) to polish compared with the larger implant (4kJcm−2 ,120s). A line overlap of 50 per cent was used in both cases. The as-built surface roughness, Sa, was reduced from 5.6 to 0.45μm after laser polishing for the smaller implant and from 4.9 to 0.63μm for the larger implement.

Figure 4. 3D surface maps of small (top row) and large (bottom) as-built and laser polished dental samples. The upper part of each sample is the original as-built surface, and only the lower area was laser polished (Image: McDonald et al)

Out of reach

Additive manufacturing can be used to produce components with complex internal structures such as cooling channels. These internal surfaces, however, pose a challenge for laser polishing, as there is often no direct line of sight.

The researchers therefore experimented with polishing the internal surface of a 30mm diameter hollow cylinder by mounting a rightangled prism within the structure on a lens tube attached to a rotation stage. This enabled the laser light to be aligned to the centre of the lens tube, with the rotating prism providing the spot 360° of rotation around the internal surface.

An improvement of surface roughness from 6.3μm to 0.5μm was achieved, however during processing, some back spatter was observed, which could damage the prism and highlights possible limitations of the process in how close the prism can be to the processed surface.

McDonald explained that implementing a smaller prism and lens tube would enable smaller cylinders to be processed; also, it is planned to further reduce this diameter by using an optical fibre with an angled end-face to replace the prism.

More information on this topic can be read in Journal of Laser Applications 32, 042019 (2020); ‘Practical implementation of laser polishing on additively manufactured metallic components’: https://doi.org/10.2351/7.0000222

Featured Product: Polished mirrors from LBP Optics 

LBP Optics offers an inhouse precision lapping and polishing service for flaw-free, ultra smooth surfaces on many materials including stainless steel, titanium, graphite, brass, ceramics, mild steel, bronze, arcap alloy, electroless nickel and tungsten. We can produce parts with just 1 micron variation in edge thickness. The typical flatness achieved is 0.5 micron or better on lapped parts up to 220mm diameter.

Our chemically polished mirrors have the ultra smooth surfaces needed for very high quality UV grade mirrors. The surface quality is high enough, and the roughness so low, that they can be used in the near IR, visible and the UV.

We also regularly polish metal parts that are not being used as traditional mirrors, but still need optical precision. For example applications such as alignment and positioning systems, test equipment and scientific instruments.








The new method recorded 6GB of data in a one-inch silica glass sample, with each square measuring just 8.8 x 8.8mm.

10 January 2022

Experimenting with circular, square and triangular oscillation patterns for laser surface treatment.

24 February 2021

Figure 5: successful validation of the process and the developed optics at a 4m depth under water

24 February 2021

The thin-disk multipass amplifier module developed by Trumpf. (Image: LAMpAS)

06 December 2021