How Particle Size and Shape is Defined
Aluminum, -325 mesh
Aluminum, -325 mesh, spherical, 22 micron
Do you really know what those particle sizes really mean? What is really being described? When they say "-325 mesh" and "22 micron", what's the difference? And why does it matter to you?
Well it can definitely help you to know how the particle "size" ratings get assigned to metal powders. Most of the size ratings come directly from the wholesaler or manufacturer. But every so often we buy surplus materials which may not come with any additional information about the manufacturer, the size or shape of the powder. Recently, we received a surplus lot of magnesium powder, including several drums with almost no information available from the seller. Before we can sell it to you, we need to be able to tell you what it is, so you can figure out if it suits your purposes.
The first step in the identification process is a visual inspection. You may be surprised how much you can tell about a sample just by looking at it. By observing the flow characteristics of a powder, and how it feels between your fingers, you can approximate particle size and shape. If you have experience with metal powders, for instance, you can often tell if a sample is granular (rough feeling), or atomized (round particles, feels smooth, pours and flows quickly and smoothly). If you cannot feel any particles between your fingers, you can assume the powder is probably finer than 200 mesh, or even less than 325 mesh (written as "-325 mesh.")
The next step is to verify those assumptions though quantitative and qualitative testing.
To determine if a material is appropriate to be used in a given formula you'll need to know the particle's shape (morphology), size, and distribution (granulometry). Shape is easily determined under a microscope and classified as atomized (spherical or spheroidal), granular, or flake.
Particle size is reported in one of two ways: either by mesh size (large and medium particles, generally larger than 325 mesh) or by microns (very small particles).
Why use two measurements?
US mesh size describes the number of openings per inch in a screen. So if a material is listed as -60 mesh it will all pass though a 60 mesh screen (the minus sign in front of the 60 means that all particles are smaller then 60 mesh). Conversely, if the material is described as +60 mesh, it would mean that all particles would be retained on a 60 mesh screen and are therefore larger than 60 mesh.
But mesh sizes can only go so far. After a point the individual wires that make up the screen are so close together it is no longer practical to measure using screens. In practice, particles smaller than 325 mesh are usually described in microns. A micron is one thousandth of a millimeter, or one millionth of a meter. The unaided human eye can see particles of about 40 microns. Smaller than that, you need magnification.
There is no truly accurate conversion from mesh size to microns, because the wire thicknesses in screens vary all over the place. But approximate conversion tables are commonly used anyway. (In the table below, screen sizes of smaller than 600 mesh are shown, even though they don't exist in practice.)
|More detailed conversion charts|
"Mass fraction analysis" is used to determine large-to-medium size particle distribution in a sample. The powder is sifted through a set of nesting screens, each with progressively smaller openings (higher mesh numbers). By measuring the percent of material that remains on each screen, we can classify a material by its size distribution.
If you were to sift Skylighter's #CH2080 Magnesium-Aluminum (described as 180-325 mesh) through a stack of 180 mesh, 200 mesh, and 325 mesh screens, a mass fraction analysis yields a particle size range that looks like this:
If the 180 mesh size was critical to your firework formula, you can interpret this to mean that 26% would remain on the 180 mesh screen (larger then 180 mesh) and 74% would pass through it (be smaller than 180 mesh).
Mass fraction by sieve analysis is a very helpful method of classifying coarse-to-medium particles, but what about the really small stuff?
When the average particle size is around 50 microns, sieve analysis is no longer practical, and doesn't adequately describe the particle sizes. Several methods are commonly used to measure really fine stuff: gravitational sedimentation, laser light diffraction, optical light microscopes, scanning electron microscopes (SEM) and transmission electron microscopes (TEM). The most accessible method to an amateur is an optical light microscope.
So how is a particle measured with a microscope? Do you need some kind of tiny ruler? As funny as that might sound, that's exactly how it's done. The microscope can be fitted with a gizmo called a reticule micrometer. After it is calibrated, it can be used to measure the size of individual particles in a powder sample right down to 1 micron.
But just because you can measure it doesn't mean it's a simple task.
Sure, measuring spherical material is fairly straightforward. After all, you're really just measuring the diameter of little balls. But what about flake, granular, and spheroidal samples? Digital imaging and software can drastically decrease the time needed to perform measurements and reduce error rates. But it appears that most if not all of the automated equipment measures any particle shape as if it is spherical. Because of this, there is not really a standard method for assigning a particle size.
Selecting the method seems to be based mostly on what you'd like your results to state. Below is an imaginary particle and three circles representing different measurement methodologies.
In the first example the measurement is across the smallest dimension of the particle. This method might be used to describe the particle in terms of its reactivity by describing the particle in the smallest possible size. Method B might be used conversely—to describe the particle's largest dimension. Arguably the most accurate methodology would be using example C, where an average size is calculated.
No matter what method is used, the results would normally be presented to you, the buyer, as an average size (3 micron), a particle range (3 to 15 micron) or a frequency distribution (30% <5 micron, 10% 5-10micron, 60% 10-15 micron), or some variation thereof.
So why does particle size or shape matter?
Many amateur fireworks makers only consider particle shape and size when a formula calls for a specific material. Even fewer consider particle size distributions. The shape and size of a particle has a huge impact on its reactivity. Flake particles have a large surface area that can be in contact with an oxidizer when compared with a spherical particle. Granular particles often have sharp edges that can ignite more easily than the smooth, round edges of an atomized powder.
Selecting powder with a different particle size or shape can create a wide variety of changes in the pyrotechnic effect, from hang time of a spark to delay of strobing. Even controlling the burn time can be accomplished by altering the particle size and shape.
Look what happens when we change particles in a real example.
The glitter formula below calls for -325 mesh spherical aluminum. Skylighter sells 3 aluminums that are -325 mesh spherical. One is further described as 5 micron (CH0100), one is 12 micron (CH0103) and another is 22 micron (CH0105).
|D1 Glitter Formula|
|Aluminum (-325 mesh, spherical)||7%|
Using the 5 micron aluminum did not produce a usable glitter. Instead it produced a bright star with an unattractive, dense, short-lived flitter-like tail. This aluminum was simply too reactive and started burning both in the flame envelope as well as after, creating poorly defined flashes.
The 12 micron aluminum produced a wonderfully dense, but short tail of fairly evenly-spaced flashes. Because the particle size distribution was within a fairly small range (mostly 6-18 microns), the glitter effect appeared fairly closely behind the star.
The 22 micron produced the best effect of all, creating a long tail that maintained good distribution of flashes over its entire length (with a few long delay pops). The 22 micron contains particles over a very wide range with most particles appearing between 5 to 38 micron.
It is clear from the results of the test above that tracking the average particle size and shape may not be enough to reproduce a specific effect, tracking the particle size distribution (if you know it) may also be worth noting in your formula book.
- Material to be measured
- Optical light microscope
- Screen Set (TL2011)