Comparing sieve analysis and laser diffraction
Identifying analytical tools that can set size specifications for metal powders is vital. Sieving has long been the principal technique, but what role could laser diffraction play? Dr Sarennah Longworth-Cook and Parminder Singha*, Malvern Instruments, investigate.
Manufacturing industries using powder metallurgy (PM) techniques – including additive manufacturing (AM) and metal injection moulding – rely on a supply chain of closely specified, high-performance metal powders to produce precision components at a competitive cost.
The physical characteristics of these powders – bulk properties and those of the individual metal particles – define their performance and are closely controlled during manufacture. Packing density and flowability are key bulk properties and are directly, though not exclusively, influenced by particle size.
Accordingly, it is vital to identify analytical tools that can reliably set size specifications for metal powders, to validate quality and manage their use. Sieving has long been a favoured technique for routine particle sizing. However, it also has many limitations that are addressed by laser diffraction. Therefore, it is necessary to explore the importance of metal powder particle size and compare the value of sieving and laser diffraction within this context.
The use of PM methods, especially AM, has intensified pressure on the metal powder supply chain, with manufacturers working to increasingly demanding particle size tolerances. The foremost manufacturing processes for metal powders are attrition milling and gas atomisation, with additional steps such as scalping (the removal of oversized particles), sieving and/or air classification applied to focus the particle size distribution to meet a specification.
A typical particle size distribution for an as-produced gas atomised metal powder usually ranges from around 10-1,000μm (microns), centred on a particle size of 2-300μm. However, much finer powders are optimal for PM processes. For example, metal injection moulding (MIM) powders are in the sub-38μm range, while those for selective laser melting (SLM), a popular AM process, are ideally around 15-45μm. These fine powders ensure the packing densities required to produce high-quality components within well-controlled dimensions. On the other hand, achieving good powder flowability, which is a prerequisite for acceptable processing efficiency, can be challenging with fine powders, as they tend to flow poorly. This is less of a problem with more spherical powder particles, which are preferred for many PM applications. When it comes to assessing sieving and laser diffraction for metal powders, it is important to consider their applicability to powders containing relatively regular particles, ranging in size from several hundred to just tens of microns.
Sieving – an age-old technique
Sieve analysis is one of the earliest techniques developed for particle size measurement and is very simple. A sample is loaded into the top of a sieve stack – a series of sieves with progressively smaller apertures. Vibration of the stack causes the sample to become separated and distributed across the various sieves. The retained sample on each sieve is then weighed to determine a size distribution for the material.
Density (measurements are based on mass/weight) and particle shape will affect the results, not just size. Furthermore, sieve analysis classifies particles according to the sieve aperture dimension they pass through, which tends to correspond to the particles’ second largest, rather than primary, dimension. The result reported is the diameter of a sphere that passes an aperture of a given size. The impact this has on the recorded size distribution is most pronounced for irregular and/or elongated particles.
From the perspective of practical implementation, sieving is inexpensive in terms of equipment outlay and sample sizes can be large, making it easier to gather statistically significant data. Furthermore, separating the sample sieve analysis enables the rejection, removal and/or separate investigation of discrete classes, as well as providing a good representation of in-process behaviour where sieving is used as a separation step.
However, there are limitations. First, sieving is slow and manually intensive – replicating measurements can be tedious and in practice is rarely, if ever, done. In addition, it offers only low precision and resolution, typically separating a sample into just five to eight fractions. Finally, and importantly for metal powder applications, sieving becomes problematic in the sub-100μm range due to cohesive forces between particles, which can lead to agglomeration.
Particle size changes during the process, caused by agglomerate formation or break-up, and poor measurement reproducibility and/or sieve blocking, are more likely with finer powders. The resolution of very fine materials (~one μm or less) is not feasible.
The case for laser diffraction
Over recent decades, laser diffraction has, therefore, become the dominant particle sizing method for industrial applications. Laser diffraction analysers detect the angular dependence of the intensity of light scattered by a sample as it passes through a collimated laser beam. Large particles scatter with high intensity at relatively narrow angles to the incident beam, while smaller particles produce a lower intensity signal at wider angles. The particle size distribution associated with a detected scattering pattern can be calculated using an appropriate theory of light behaviour, typically Mie theory.
Laser diffraction measurements, just like sieve analyses, are influenced by particle shape and size, since Mie theory is based on an assumption of spherical particles. For non-spherical particles, the result is the diameter of a collection of spheres with the same volume as the particles. For elongated particles, laser diffraction reports a larger size than that reported by sieving, which finds the smallest dimension for an elongated particle.
In addition, results rely on the scattering efficiency of the particles – their refractive index. Understanding the differences between the measurement principles underpinning laser diffraction and sieving clarifies why the particle size data measured by either technique will be different. Sieving is a mechanical method, laser diffraction is an optical one, and they both determine a different parameter that is indicative of particle size.
Case study data
The data (in graphs opposite) show particle size distributions for two metal powders, sample A and B, measured using laser diffraction and sieve analysis, according to the respective ASTM standards, ASTM B822–17 and ASTM B814–46.
While sieve analysis gives comparable data to laser diffraction for larger particle sizes, there are significant variations in the size distributions reported below approximately 80μm. It would appear that a significant proportion of particles with a size < 63μm are being retained on the 63μm sieve. This is implied by the small amount of sample making it through to the 53 and 45μm sieves, and the discrepancy with laser diffraction data, which clearly shows a significant fraction of particles in this size range. This may be due to poor powder dispersion in the sieve, where fines are bound together, bound to larger particles or adhere to the sieve itself.
Furthermore, as the lower size range for sieve analysis based on the ASTM standard is 38μm, it is only possible to record the fraction of particles < 38μm. Laser diffraction, however, is capable of measuring size distributions in the sub-micron range, and data for sample B clearly shows there is a significant fraction of particles with a size distribution extending to below 10μm.
A valuable technique
The capital investment for laser diffraction is greater than for sieving but the practical benefits of the former are significant, including:
- Fast measurement times – results are typically reported in less than a minute.
- A broad measurement range, from 0.01-3,500 microns, which comfortably spans the range of interest for metal powders.
- Fully automated analysis, to the point of online implementation, where required.
- High reproducibility/repeatability that can be demonstrated through duplicate measurements.
- The capability to detect and quantify fine and coarse material in a single measurement.
- High resolution – modern systems typical report around 100 size classes, thereby delivering the sensitivity to detect even subtle changes in particle size distribution for effective quality control (QC).
Samples are prepared for laser diffraction to ensure a consistent state of agglomeration, typically complete dispersion, as it is the primary particle size that tends to be of most interest. Rapid dry powder dispersion can be achieved for samples containing very fine or even sub-micron particles and is often suitable for metal powders, which tend to be relatively effective. Liquid-based dispersion is the alternative and is useful for very fine particles.
A winning formula for metal powders
It is clear from the assessments that sieving and laser diffraction have very different strengths and limitations. Those working at the forefront of metal powder supply harness the best attributes of both, rather than choosing between them.
Sieve analysis is typically applied to coarser metal powders, as part of material acceptance (QA), and also to assess how a sieving/screening step will perform in a process. Laser diffraction is valued for its high throughput, especially in QC, and for robust assessment of the level of fines. Fines can have an impact on packing and flow behaviour, and are closely controlled in market-leading metal powders.
Together, the techniques enable metal powder manufacturers to understand and closely control product properties and ensure a robust supply chain for the optimal application of PM processes.
*Dr Parminder Singh, is Product Technical Specialist – Laser Diffraction & Analytical Imaging, Malvern Panalytical India.
Dr Sarennah Longworth-Cook, is a Product Technical Specialist – Laser Diffraction, at Malvern Panalytical UK.