The manufacture of metal powders and powder metallurgy (PM) components became very important about 70 years ago, and a number of industrial manufacturing processes have emerged, showing increasing competence and specialization of production. Recently 3D printing, or additive manufacture technology, has led to the latest PM processing development, Selective Laser Melting.
Selective Laser Melting (SLM)
Selective Laser Melting is a 3D printing process whereby a CAD pattern is used to direct a laser beam with the help of a rotating mirror, on to the topmost layer of metal powder on the manufacturing platform. Metal particles are melted by the beam exactly on top of the previous layer but in precise accordance with the CAD pattern, thus creating a 3D object layer by layer.
The un-melted residual metal particles are scraped away for reuse in the next layer/cycle, until they show signs of excessive wear, stick together or are oxidized. This type of synthesis avoids the high levels of wastage found in processes where leftover powder is not recycled, in which up to 90% of raw materials can be unexploited.
The figure on the right below shows how complex metal components can be produced by SLM. The fact that each piece is produced as one, without joints or welding, makes it strong and durable, and allows easy customization. The same platform can be used for any number of CAD programs to produce custom parts one by one on demand, avoiding the use and cost of custom tooling for the sake of manufacturing a single part.
SLM typically demands tight specifications for the metal powder used compared to other part manufacturing processes. This is due to the atomized nature of the powder used and the nature of the parts. Complex components with extremely thin surfaces are produced using powder particles with a smaller mean size and narrow size distribution width.
On the other hand, a bimodal width might be ideal to ensure that the laser melter bed is loaded with loose-packed powder at the highest density, so that the component is solid and strong and voids are minimal in the final product.
Particle shape is another important parameter. The laser bed is most evenly and closely packed when the particles are very round and smooth, with superior flow properties. This allows a proper recreation of the laser melter bed after each new layer is deposited. Such particles help to repeatedly achieve strong and solid structure builds using the fusion process.
Particle size and shape must be detected separately as they are independent parameters. Particles within any size parameter can be rough, round or oval. To ensure quality control, therefore, the metal particles must be measured for both size and shape, ensuring these fit predefined specifications so that the powder is of high quality.
The presence of a single contaminant in the metal powder feed to SLM is a key concern, since this could result in the formation of a point defect in the final part. Image analysis can help detect the contaminant if it is non-spherical, rough on the outside, or translucent, as well as allow for its quantitative measurement as a percentage of the volume or number of particles.
The recycled metal powder must be confirmed to be suitable in terms of the wear and oxidation and contamination it may suffer during each recycle. This crucial quality test is performed by repeating the morphological measurements with each recycle. Powder which contains particles that go beyond the specifications must be melted and atomized before entering the cycle again.
Quality Control for Metal Powders Used in SLM
Metal powder particle size has been analyzed successfully using laser diffraction (LD) technology for many years. Another emerging technology is the Dynamic Image Analysis (DIA) for study of particle size as well as shape. A combination of these two is presented in the Microtrac Sync.
Laser diffraction has now become the go-to choice for particle size detection in quality control of powder metallurgy and metal powder industries. It depends upon the scattering of a coherent laser beam directed through a stream of metal powder by the powder particles, with the degree of scattering and final light intensity depending upon the particle size. An array of detectors is arranged at angles around the sample flow to measure the scattering by picking up the scattered light distribution, which is then calculated into a size distribution using an iterative algorithm.
Dynamic Image Analysis
Dynamic Image Analysis (DIA) depends upon the imaging of flowing particles in a sample cell backlit by a high-speed strobe light, by a high-resolution digital camera. The images are compiled into a video file and exported to a computer which analyzes them using the pixel size and number. Pixel size calibration makes it simple to report on the size and shape of the particles in the sample cell. The saved video file is ready for re-measuring in case another set of Standard Operating Conditions (SOP) is required for a different application.
The blue particle image outline seen in the above figure shows how the values of different parameters can reveal the presence of contaminating or out-of-specification metal particles. It indicates an elongated shape, as shown by the W/L Aspect Ratio. The Solidity and Transparency parameter results show that the particle is rough and translucent, respectively.
The Shape parameters are measured from 0 to 1, with one indicating complete sphericity. Solidity shows perfect smoothness and absence of indentations when its value is 1. A Transparency value of 1 shows the particle is completely transparent. Two or three parameters are always sufficient to report on the shapes of key particles for the majority of processes, and of these, an alteration in any one may reveal the presence of contamination.
For instance, a particle of opaque metal powder may not meet the specifications for W/L Aspect Ratio or for Solidity, indicating that its sphericity and rough surface are not within acceptable limits. The volume percentage or count percentage of powder or contaminant particles that are out of specification can also be identified and measured by this technique.
The Width/Length Aspect Ratio is measured on a scale of 0 to 1, with elongated shapes having lower values and indicating that the particle is not spherical. The Solidity and Transparency values are scaled from 0 to 1 as well, from rough to perfectly smooth surfaces and opacity to perfect transparency, respectively. The latter shows if glass, plastic or other substances have come off the packaging or mixed in with the feed material during handling. An important point is that contaminants in pure metal powder are usually significantly low with regard to the W/L and Solidity values than a metal particle beyond the specification limits would be.
As Figure 9 shows, the W/L Aspect Ratio and Solidity values of the images displayed represent fused powder particles, the fusion having occurred as the result of post-atomization cooling. Such groups of particles would harm the manufacturing process by reducing the density of packing, leading to excessive porosity which can reduce both the green and the sintered strength of parts produced by conventional methods, and produce point defects in the structure of SLM-manufactured parts.
The image analysis tool helps to achieve quality control by providing data to decide the appropriate specification limits of shape for a given metal powder, depending on how complex the part is, the process followed, the yield rate that is targeted, and other relevant aspects.
The View Particles Search feature, or alternatively the Filter feature, enables rapid identification and quantification of elongated particles.
Figure 11 shows how the View Particles works to separate out images in groups using a user-defined search criterion. The present example deals with the search for the image group with a low W/L aspect ratio shown in the scatter diagram in Figure 10. The Search Particles button on the upper left side is selected (shown in callout 1). The Particle query window now opens (callout 2), where W/L is selected from the parameter list and limits are set to below 0.900 (using callout 3).
After beginning the search, one of the results is shown (callout 4). A report displayed in the Search Particles window shows that 21.66 % by volume of the sample comprise particles which fit the search criterion, or in number percentage terms, 5965 out of 51,576 particles or 11.6%. The particles in this group comprised individual particles which had fused, and the number percentage may show that this batch of powder did not conform to shape specifications.
This program can be used to isolate and measure six separate components in a sample using a separate search for each, the whole set being linked to one SOP. The components may include different material types which are either combined on purpose to achieve an optimal set of properties, or have been isolated because they failed to meet shape or size specifications, or simply contaminating particles.
Each component can be measured and viewed by itself, as fractions of a single measurement, provided the different sizes and shapes are selected for individual component search. A search that has been automated and assigned to the SOP can have further selected filters used to report how much of each fraction is present.
The SOP is then applied to get an X-Y graph and table as shown in Figure 13. Callout 1 is the blue curve which is a cumulative % finer curve of the area equivalent diameter (Da) reported for the whole sample. The red curve shown by callout 2 is an equivalent curve for the subset of particles that conform to the search criterion, as seen in the result of the View Particles Search which displays 21.66%.
The real W/L and Solidity values are displayed in Table 3 and are also seen as differential distributions corresponding to individual size fractions.
The View Particles display shows only those images of particles that fit the search settings, as seen in Figure 14. On the left images are less elongated, becoming more and more elongated in the middle and right-hand images. These correspond to the upper, middle and lower ranges of the W/L Aspect Ratio. Using this search analysis it is possible to set the specification limits for the W/L parameter for the given material, depending upon the customer-defined metal powder specification for PM manufacture.
The history of search results for any selected parameter can also be reported using the graph and the table, as seen in Figure 15, showing six records where the Da differential percentage by volume is reported for samples of metal powder.
Figure 16 shows trend charts, which are produced by the DIA software to ensure that a selected parameter can be easily followed between two successive samples, thus picking up important alterations in size and shape in the recycled material feed in SLM.
Once any selected parameter is seen to deviate significantly from the specifications, the stream must be remade from the beginning, and thus the process of determining whether the recycled material is fit to use or not is made simpler by noting the control limits for the selected specifications on the charts.
Figure 17 shows a DIA, the only one that is presently commercially available to yield real-time measurements of metal powder in flowing streams during an industrial or research process. It can be fitted on the recycled metal powder stream during SLM to ensure the identification of beyond-specification metal powder particles, or the presence of contaminant particles.
The need to measure metal powder particle morphology is important in several ways:
- Quality control for both users and supplier specifications
- Identification of the presence and amount of contaminating or beyond-specification particles in industrial or research processes
- Monitoring the quality of recycled metal powder feeds in SLM
Laser diffraction is uniquely efficient in size identification and quantification and has become the standard for these processes in the industry
DIA is the technology of choice for morphological study of the metal powder particles.
- The Microtrac Sync meets the need by combining both LD and DIA in the same measurement of the same sample, and completing the process within minutes.
- It can be set up by selecting any of 30 separate parameters with regard to size and shape, to find and quantitatively measure the important parameters for each application.
- In SLM, this technology allows the recycled material stream to be monitored during the process itself so as to immediately detect the presence of unacceptable deviations of the material from the specifications.
This information has been sourced, reviewed and adapted from materials provided by Microtrac, Inc.
For more information on this source, please visit Microtrac, Inc.