PDF Technology Review: 20171201-infinity-electrostatics-laser-diode-melting-additive-mfg
Reference Source: Zavala-Arredondo, M., Boone, N., Willmott, J., Childs, D. T., Ivanov, P., Groom, K. M., & Mumtaz, K. (2017). Laser diode area melting for high speed additive manufacturing of metallic components. Materials & Design, 117, 305-315. http://www.sciencedirect.com/science/article/pii/S0264127516316215
1. Additive manufacturing (AM) is considered a possible alternative to conventional manufacturing with the ability to create geometrically efficient structures and low material wastage.
2. Two processes under AM for metal manufacturing are selective laser melting (SLM) and electron-beam melting (EBM). In both, pre-deposited powder is successively added and then melted through either a laser or electron beam building up the structure of the object.
3. Generally SLM, or direct metal laser sintering (DMLS), is capable of melting a variety of metal powders (steel, nickel, titanium, aluminum alloys) to near full density with laser powers ranging from 100 – 400 W. This wattage determines how fast the laser can scan and is the limit in build speed.
4. EBM have a higher power of up to 3.5 kW and thus a faster build rate (5 – 20 cm3/hr (0.31 – 1.22 in3/hr) for SLM and 80 cm3/hr (4.89 in3/hr) for EBM). However the setups are much more expensive than SLM systems and the surface finish is generally poorer due to the temperamental nature of the electron beam.
5. SLM’s power consumption is considered rather inefficient because they are based off of fiber lasers which have an about 20 percentage wall plug efficiency, of which only 20 percentage consumed by the laser is converted into optical energy, and then strict wavelength outputs mean that most of the energy is reflected instead of absorbed by most metals.
6. The absorption of optical energy by a metal changes depending on the wavelength of the material and is different for each metal.
7. Diode lasers (DL) have a much higher wall plug efficiency of between 50 and 80 percentage, allow for tuning of emission wavelength, have seen performance improvements for more uniform melt and heating zones, as well as consistency in surface applications. This places them as a good option for an AM SLM-like process.
8. The authors of this paper worked to combine several DL onto a bar and have them focus down through curved lenses to create a stripe. Each laser had an elliptical spot and thus the ends of each spot could be covered by the other creating a relatively even stripe and could each be selectively turned off if the object required so. Due to the more compact nature of the DL, the authors suggest that these DL could be stacked to create an even longer stripe of laser to pass over the powder bed.
9. The DL bar they created consisted of 19 individual lasers that are focused to create the elliptical nature to create the stripe. Each laser emitted a power of 2.63 W for a combined power output of 50 W.
10. They first tested this system which they called diode area melting (DAM) on a lower melting temperature metal powder, BiZn2.7. They were able to easily melt the metal with a scan speed of 1 mm/s (0.04 in/s) and the stripe had a width of 4.5 mm. (0.18 in)
11. Temperatures were able to reach in excess of 500 degrees Celsius (932 degrees Fahrenheit) and the lasers could be selectively turned off to create different shapes in a single pass. The structures created also had a density of 99.27 percentage of the true density of the material.
12. They also tested the DAM system on stainless steel 17-4 powder to test it’s capabilities. Iron has a much higher absorption at the wavelength of their DAM and thus had a much quicker temperature rise.
13. At the same scan speed, the stainless steel reached temperatures in excess of 1350 degrees Celsius (2462 degrees Fahrenheit) than for the bismuth zinc alloy. An exact temperature could not be determined due to the range and resolution of their temperature reading camera (the temperature rose at a 100 times the rate as with the bismuth zinc alloy).
14. Using a scan speed of 0.5 mm/s (0.02 in/s) and the full 50 W power, they were able to create a 4.5 mm x 4.5 mm x 6 mm cube (0.18 in x 0.18 in x 0.24 in) structure using manual additional of powder.
15. The structure had bulges at the top indicating that the slow scan speed might have been too slow and that the melt pool would draw from the feed-stock creating peaks and troughs. Also, periodically throughout the structure their would be voids that would indicate that the layers had not completely melted together.
16. Taking cross sections of this structure, at some points the part density was up to 99.72 percentage which is considered on par with conventional SLM processes. At other points the density was lower at 79.18 percentage which was attributed to gas occlusion.
17. The authors note that it is quite promising that the method is able to heat a material to temperatures in excess of 1400 degrees Celsius (2552 degrees Fahrenheit) with such low powered lasers.
18. Most of the issues around surface roughness and areas of porosity could be written off as the process being in early stages of optimization as well as the manual addition of powders causing variation as well as higher than normal minimum deposition of powders (around three times more than in conventional SLM processes).
19. This process of an area based laser would be considered inaccessible by conventional fiber laser systems.
20. Further work must be conducted to see if the laser in this process is vaporizing material which could explain the inclusion of voids, as well as test other materials since the wavelength of the diodes could be tuned to allow for even lower power lasers. One material of interest would be aluminum due to it’s high reflective nature and conduction ability making it a hard material to be used with conventional SLM processes.