PSLE-SCI-REALITY-0353
Wait, What? A Powder Can Have D50 = 20 µm Even When Almost No Particle Is Exactly 20 µm
A laboratory report says D50 = 20 µm. A student imagines a jar containing two groups: half the particles are exactly 20 micrometres wide and the other half are some other size.
That picture is wrong. D50 is a position on a cumulative particle-size distribution. NIST describes d10, d50 and d90 as sizes corresponding to 10%, 50% and 90% of the cumulative distribution; d50 is the median diameter. The value marks a boundary in a distribution. It does not require a pile of particles sitting exactly on that boundary.
There is a second trap. A particle-size distribution can be expressed on different bases, such as number, volume or mass, depending on the measurement method and report. So even the phrase “half the particles are smaller” may be too loose unless the basis is known.
Quick Answer
- D50 is the median size on a stated cumulative particle-size distribution.
- It does not mean all particles are 20 µm, nor that half are exactly 20 µm.
- Check the distribution basis—number, volume, mass or another defined basis—before translating the 50% into words.
- Check the measurement method because irregular particles do not have one uniquely obvious “diameter”.
- D10, D50 and D90 are useful summaries, but a few percentile values cannot reveal the entire distribution shape.
The Exact Learner Job This Page Owns
This page owns one real-world evidence-transfer job: evaluating a D50 particle-size statement by identifying the cumulative-distribution basis and refusing to turn a median boundary into one literal size shared by the sample.
It does not own sediment transport, aerosol health effects, powder engineering or microscopy. It applies existing PSLE Science owners for measurement, distributions, evidence limits and careful comparison.
- How to Read Units, Scales and Measurement Resolution Before Using PSLE Science Data
- How to Check That Two PSLE Science Numbers Measure the Same Scientific Quantity Before Comparing Them
- How to Identify What Evidence a PSLE Science Question Actually Gives You
Original Reality Lab Case: Three Powders With the Same D50
This is an original constructed case. No commercial specification or assessment-book question has been copied.
| Fictional powder | D10 | D50 | D90 |
|---|---|---|---|
| A | 16 µm | 20 µm | 25 µm |
| B | 3 µm | 20 µm | 90 µm |
| C | 8 µm | 20 µm | 31 µm |
All three have the same D50. Are they the same powder-size distribution? Clearly not. Powder B spreads across a much wider size range than Powder A. The median alone cannot tell the whole story.
Observed, Measured, Summarised and Inferred
| Layer | What it means |
|---|---|
| Physical sample | A collection of particles with shapes and sizes |
| Measurement | An instrument responds to properties related to particle size according to its method |
| Distribution | Measured sizes are organised on a stated number, volume, mass or other basis |
| D50 | The median size on that cumulative distribution |
| Inference | A useful description of the sample’s central size, within method and sampling limits |
The Distribution Check: What Does “50%” Refer To?
The number 50 in D50 belongs to a cumulative distribution. Imagine arranging the measured distribution from smaller sizes to larger sizes. D50 is the size where the cumulative distribution reaches 50%.
But 50% must have a basis. In a number-weighted distribution, each particle contributes by count. In a volume-weighted distribution, large particles contribute much more because they occupy more volume. In a mass-weighted distribution, contribution depends on mass. A report that does not say the basis leaves an important interpretation question open.
Why “Half the Particles Are Exactly 20 µm” Fails
A median marks the middle boundary of a distribution. It does not say the data pile up at the median value. Consider five fictional diameters: 8, 12, 20, 27 and 60 µm. The middle observation is 20 µm. Now consider 8, 19, 19.9, 20.1 and 60 µm. A median-like boundary can still sit near 20 even though no particle is exactly 20.
Real particle measurements are more complicated because there may be thousands or millions of particles and because the measurement method may report an equivalent diameter rather than a direct ruler width.
The Shape Check: What Is the Diameter of a Flake?
A perfect sphere has a clear diameter. A long fibre, flat flake or jagged grain does not. Different methods may describe an irregular particle by an equivalent diameter: the diameter of a sphere or circle that would produce a comparable measured response under the method.
This is why “20 µm particle” should not always be imagined as a tiny perfect ball exactly 20 µm across in every direction.
The Method Check: Different Instruments Can See Size Differently
Particle size can be measured using several approaches, such as microscopy, sieving, light scattering and other techniques. Each method has a range, assumptions and way of converting a physical response into a size.
NIST guidance emphasises that particle-size characterisation depends on the measurement technique. Therefore, when two reports give different D50 values, the first scientific question is not “Which laboratory is wrong?” but “Did they sample and measure the same material in comparable ways?”
The Sampling Check: Did the Scoop Represent the Bag?
Even a perfect instrument measures only the material presented to it. If larger grains settle while finer particles remain near the top, a small scoop from one location may not represent the entire bag. Mixing, splitting and sampling method can therefore matter before the instrument even begins.
A D50 result is evidence about the tested sample. Travelling from that sample to a whole batch requires a sampling argument.
Why D10, D50 and D90 Still Do Not Show the Full Curve
Three percentile points can show central size and spread roughly, but NIST warns that quoting d10, d50 and d90 does not reveal the true shape of the entire distribution. Two distributions can share the same three values yet differ between them, or contain more than one cluster of particle sizes.
A summary is not dishonest because it leaves information out. It becomes misleading only when readers forget what has been compressed.
Comparison Check: Is a Lower D50 Always “Finer”?
If two samples were measured using the same valid method, basis and preparation, a lower D50 generally supports the statement that the median of the reported size distribution is smaller. But “finer in every way” is stronger. One sample could have a lower D50 yet also contain a small number of very large particles.
D90, the full curve and the application may matter.
Alternative Explanations for a D50 Shift
Suppose a powder’s D50 changes from 20 µm to 24 µm between two reports. Possible explanations include a real manufacturing change, different sampling location, clumping, moisture, different dispersing procedure, instrument calibration, changed analysis settings or a different distribution basis.
The number changed. The cause is not automatically known.
What Evidence Would Strengthen a Particle-Size Comparison?
- The same material and representative sampling method are used.
- The same measurement principle and preparation procedure are stated.
- The distribution basis is the same.
- D10, D50, D90 or the full distribution curve are reported where spread matters.
- Replicate measurements show the result is stable enough for the intended conclusion.
What Would Weaken the Claim?
- A product advertisement says “20 µm particles” when the only evidence is D50 = 20 µm.
- Two D50 values from different methods are compared without checking the measurement basis.
- A single scoop is assumed to represent a poorly mixed bulk sample.
- D50 alone is used to claim that no very large or very small particles are present.
- The word “diameter” is treated as a direct ruler measurement for irregular particles without checking the method.
Worked Case 1: Number Basis Versus Volume Basis
A fictional sample contains many tiny particles and a few much larger ones. On a number basis, the tiny particles dominate the count. On a volume basis, the large particles can contribute strongly because each occupies much more volume. A D50 from one basis need not equal a D50 from the other. Before comparing, identify what the percentage accumulates.
Worked Case 2: Same D50, Different D90
Powder X has D50 = 20 µm and D90 = 30 µm. Powder Y has D50 = 20 µm and D90 = 100 µm. The identical medians do not support the claim that the powders have the same size distribution. Y has a much larger upper part of the reported distribution.
Worked Case 3: A Flake and a Sphere
A flake and a sphere receive similar equivalent diameters from a particular method. That does not prove their shapes are the same. The metric reports a size representation tied to the measurement response, not a full 3D description.
Worked Case 4: The Bag That Settled During Transport
A bag of mixed grains vibrates during transport. Larger grains concentrate near one region. Laboratory A samples the top; Laboratory B uses a properly mixed composite. Their D50 values differ. The discrepancy could arise before measurement because sample selection changed.
Tempting Reasoning That Fails
- “D50 = 20 means 50% are exactly 20.” D50 marks a cumulative median boundary.
- “Every particle is about 20 µm.” The spread may be narrow or enormous.
- “A diameter is always a direct width.” Irregular particles may be assigned equivalent diameters.
- “Same D50 means same powder.” Distribution shape, chemistry, morphology and many other properties can differ.
Model and Measurement Limits
Particle-size distributions are models built from measurements. Instrument sensitivity, sample preparation, refractive or density assumptions, agglomeration and the chosen mathematical basis can change the reported distribution. That does not make D50 arbitrary. It makes provenance essential.
A good report should make it possible to tell what was sampled, how it was prepared, how size was measured, what distribution basis was used, and which summary values were calculated.
How Far Can the Conclusion Travel?
D50 can support a claim about the median of the stated measured size distribution. It cannot by itself tell you every particle size, the full distribution shape, particle chemistry, shape, safety or performance in a product.
PSLE-Style Transfer Case
A fictional report states: “Sample R: D50 = 15 µm, D90 = 80 µm.” A pupil concludes: “Most particles must be exactly 15 µm, and none is larger than 80 µm.”
Question: Explain two problems with the conclusion.
Reasoned answer: D50 is a median boundary on the stated cumulative distribution and does not mean most particles equal 15 µm. D90 marks the size below which 90% of the stated cumulative distribution lies, so the remaining part can include sizes above 80 µm.
Explained Practice
Practice A: Two reports both say D50 = 25 µm. One is number-weighted and the other volume-weighted. Can the values be treated as identical evidence? Not until the bases and methods are aligned.
Practice B: A powder has D10 = 18, D50 = 20 and D90 = 22 µm. Another has 2, 20 and 200 µm. Which appears more narrowly distributed from these summaries? The first, while remembering that the full curve still contains more information.
Practice C: A product says “particle size 20 µm” but the technical sheet says only D50 = 20 µm. What wording is safer? “Median reported particle size D50 = 20 µm under the stated method.”
Delayed Independent Return: S-I-Z-E
- S — Sample: What material was actually tested?
- I — Instrument: How was size measured?
- Z — Zero in on the basis: Number, volume, mass or another distribution?
- E — Evidence boundary: Median only, or the full distribution?
Parent and Tutor Teaching Guide
Make two bags of paper circles. Give both the same median size but make one bag tightly clustered and the other extremely mixed. Ask the learner whether one median number can describe the difference. Then ask what additional evidence would help. This makes distribution shape visible before introducing laboratory vocabulary.
For transfer, use height, seed mass or rainfall. Ask the learner to distinguish a median boundary from “everyone has the median value”.
Authoritative Sources
- Singapore Examinations and Assessment Board — 2026 PSLE Science Syllabus
- Ministry of Education Singapore — 2023 Primary Science Teaching and Learning Syllabus
- National Institute of Standards and Technology — Particle Size Characterization Recommended Practice Guide
- NIST — Why Particle Size Distribution Matters to Measurement
The Quiet Return
D50 is not the size of the powder.
It is one carefully defined address inside a distribution.
Before turning a median into a picture of every particle, ask what distribution created it.
