Mixtures and separation · Investigation
Proving something is pure
Someone hands you a white powder and tells you it is pure. Believing them is not chemistry — so what measurement would make them prove it?
Start here
Three bags of white powder. All three labelled pure. One of them is not.
A supplier has sent three batches of the same white solid, all sold as pure. Somewhere in the delivery, one batch has something else mixed into it. Every bag looks identical, and the label is not evidence.
You can make one measurement on each. Which measurement would settle it?
The melting point. A pure substance melts at one temperature — sharply, within a degree. A mixture melts over a range, and it starts melting lower than the pure substance would. That difference is measurable with a thermometer, and it does not care what the bag says.
Before you plan · judge someone else's method
A student has written a plan. Four steps. Rule on each one.
For each step, decide whether it could settle whether the powder is pure — before you read what it is actually worth.
"Look at the powder under a hand lens and check the crystals all look the same."
Tells you nothing about purity.
Two white powders ground together look like one white powder, and a single substance can crystallise in more than one shape. Uniform appearance is exactly the evidence that has failed in every lesson of this unit.
"Taste a little of it and see if it tastes of anything else."
Never done, and it would not settle it anyway.
Never taste anything in a laboratory. Not a drop, not a crystal, not even something you are sure you recognise — an unknown white powder is exactly the thing that can poison you. In a kitchen, tasting is how you check your food. In here, it is how you get hurt. Even if it were safe, most impurities are tasteless at a few per cent.
"Stir some into water and check that it all dissolves."
Useful, and not enough.
It would catch an impurity that does not dissolve — sand mixed into salt would leave a residue. It catches nothing that dissolves, and most things that get mixed into a powder do dissolve.
"Weigh 10 g of it and check that it weighs 10 g."
Measures the balance, not the powder.
Mass tells you how much you have, never what it is. A mixture weighs whatever a mixture weighs. There is no expected value to compare it against, which is what makes it useless here.
Four steps, none of them useless, and not one of them able to answer the question. The plan is not lazy — it is a plan built out of observations rather than out of a measurement with a known expected value. That is the difference you are about to build.
Your turn · the melting point bench
The pure substance should melt at 53 °C. Go and find out what you have.
A little of each sample in a tube, in a heated block, with a thermometer. You record the temperature when melting starts and when it finishes. Two decisions are yours.
How fast you heat the block
How many times you run each sample
Batch 1
| Run | Melting started | Melting finished | Range |
|---|---|---|---|
| 1 | 52.5 °C | 53.0 °C | 0.5 °C |
| 2 | 52.5 °C | 53.5 °C | 1.0 °C |
| 3 | 53.0 °C | 53.5 °C | 0.5 °C |
Batch 2
| Run | Melting started | Melting finished | Range |
|---|---|---|---|
| 1 | 45.0 °C | 52.0 °C | 7.0 °C |
| 2 | 44.0 °C | 51.0 °C | 7.0 °C |
| 3 | 46.0 °C | 52.5 °C | 6.5 °C |
Batch 3
| Run | Melting started | Melting finished | Range |
|---|---|---|---|
| 1 | 52.0 °C | 53.0 °C | 1.0 °C |
| 2 | 47.5 °C | 52.0 °C | 4.5 °C |
| 3 | 52.5 °C | 53.5 °C | 1.0 °C |
Three runs each, heated slowly. Batch 1 melts within a degree, every time. Batch 2 melts over six or seven degrees and starts around 45 °C, well below the expected 53 °C. Batch 3 agrees with batch 1 twice and disagrees with itself once — which is what an anomalous result looks like when you have bothered to repeat.
One run each, heated slowly. Batch 2 stands out — a range of seven degrees, starting eight degrees low. But with one reading each you cannot tell a real difference from a bad run, and batch 3 looks fine on a single measurement that you have no way of checking.
Three runs each, heated fast. The repeats agree with each other, which is reassuring and misleading: they agree because the same error happened three times. Batch 2 is still visibly worse than the others, but every start has been dragged up and every range has been dragged towards the same middling value — batch 2's seven degrees down to under five, batch 1's half a degree up to nearly two. Slow the block down and run it again.
One run each, heated fast. The thermometer cannot keep up with a fast block, so every start reads high, batch 2's seven-degree range collapses to under five, and a sharp melt smears out to nearly two. The gap that gives an impure batch away has shrunk to about three degrees, on one reading you have no way of checking. This table cannot answer the question, and nothing about looking at it harder will fix that.
Which batch is the impure one?
Batch 1 is the clean one — look at its range.
Batch 1 melts within a degree on every run. That is the signature of a pure substance, and it is your reference for what the other two should look like. Batch 2 is the one that melts over six or seven degrees.
Batch 2. A range of six or seven degrees, starting eight degrees low.
Something else is dissolved in it, interrupting the arrangement, so it starts giving way early and finishes late. Your data can carry that claim: three slow runs, agreeing with each other, on all three batches. Your data points the right way but cannot carry the claim — heated fast, or run once, the ranges are not trustworthy. Run it slowly, three times, before you accuse a supplier of anything.
Batch 3 is pure. Its odd run is an anomaly, not evidence.
Two of batch 3's runs melt within a degree, at the expected temperature. The third disagrees with both — one anomalous run, most likely a loosely packed tube. Set it aside with a reason written down and batch 3 is pure. The impure batch is batch 2.
Key fact
A pure substance melts sharply, at one temperature. A mixture melts over a range, and it starts lower than the pure substance would. Purity is proved by a measurement with a known expected value — never by looking.
Five words
Say your answer out loud before you turn each card over. If you cannot say it, you do not know it yet.
Evaluate · what this method cannot do
Three honest limits
A sharp melting point at the expected temperature says the sample is pure. A sharp melting point at the wrong temperature says it is a pure substance that is not the one you ordered.
Without an expected value the measurement means nothing. You have to know what 53 °C was supposed to be.
It says that something else is present, never what. Identifying the impurity is a different job — and chromatography is where you would start.
Think again
“I measured it once and got 53 °C, which is the right answer. That proves it is pure.”
The reading was right and the sample really might be pure. Commit before you read on.
One reading cannot tell you whether it was a good reading. Heat the block too fast and the thermometer lags behind the sample, so a wide range reads narrow and a low start reads high — a single fast run flatters an impure sample, closing much of the gap that would have given it away. Repeat it and the readings that agree with each other are the ones you can defend.
And note what the batch 3 data did. Two runs agreed, one did not. The anomalous run is not deleted and it is not averaged in — it is reported, explained and set aside, and the reason is written down: the tube was packed loosely, or the block was still warming. That sentence is the difference between data and a result.
Mastery ladder
Not started yet.
Rungs 3 and 4 you mark yourself.
Rung 1 · Recall
How does the melting behaviour of a pure substance differ from that of a mixture?
Rung 2 · The one that catches people
A sample melts sharply, within half a degree — but at 61 °C, when the pure substance should melt at 53 °C. What can you conclude?
Rung 3 · Evaluate
In batch 3, two runs gave 52–53 °C and 52.5–53.5 °C, and one gave 47.5–52 °C. Explain what you should do with the odd run, what you must write down, and what your conclusion about batch 3 should be.
Rung 4 · Design it yourself
A shop suspects a supplier is bulking out bags of pure sugar with something cheaper. You have a melting point apparatus, a balance, filter equipment, chromatography paper and a known pure sample of the sugar. Design an investigation that would show whether the bags have been tampered with, and say what result would prove it.
Key note
A pure substance melts sharply at one fixed temperature; a mixture melts over a range and starts lower. That makes melting point the test for purity — but only against a known expected value, only with repeats, and only if the sample is heated slowly enough for the thermometer to keep up. An anomalous run is reported and explained, never quietly deleted.
Going further
The reason a mixture melts low is worth knowing, because it is used on every road in the country. In a pure solid, every particle sits in the same regular arrangement, and the whole thing gives way at one temperature. Mix something else in and the arrangement is interrupted — the particles no longer fit together as neatly, so less energy is needed to break them apart, and different regions give way at different temperatures. That is a lower melting point and a range, from the same cause.
Salt on an icy road is that effect run deliberately: salt water freezes several degrees below 0 °C, so the ice melts at a temperature at which pure water would stay solid. A pharmacist checking a batch of aspirin uses the same physics in the opposite direction — a melting range wider than a degree means something is in there that should not be, and the batch does not ship. Same measurement, same reasoning, and one of them is on a gritter and the other is in a laboratory.
Before this lesson
Next in this unit
At GCSE this becomes
- Melting point as evidence of purity, and the distinction between a pure substance and a formulation.
Where to next
- Next: Chemical change vs physical change
Chemical reactions
- Previous: Chromatography
Ask Mr Badmus AI
Still not sure what to do with the anomalous run in batch 3?
Lesson content © MrBadmusAI.