How High-Purity Quartz Sand Is Tested: ICP-MS, ppm Impurities and Why 99.9% SiO2 Is Not Enough

Why 'percent SiO2' is not enough

Many quartz listings quote a single number such as 99.9% or 99.95% SiO2. The arithmetic is simple: 99.9% SiO2 leaves 0.1% for everything else, and 0.1% is 1,000 parts per million. Even 99.99% leaves 100 ppm. India's Minerals (Evidence of Mineral Contents) Amendment Rules, 2026 define high-purity quartz as SiO2 above 99.99% with total impurities below 50 ppm, so a 99.9% product is an order of magnitude away from the legal HPQ line.

A percentage is also usually calculated by difference: the lab measures impurities and subtracts them from 100. That makes the impurity list the real data, and the SiO2 figure a summary of it.

Which elements matter for crucibles

For crucible-grade sand, what matters is not only the total but which elements are present. Aluminium sits inside the quartz lattice and is hard to remove. Alkalis such as lithium, sodium and potassium lower the viscosity of fused quartz and encourage devitrification. Iron, titanium and other transition metals can enter the silicon melt. Boron and phosphorus are dopants in silicon and affect resistivity. The 2024 review by Long and colleagues in Minerals sets out these impurity types and how purification targets them.

• Aluminium: lattice-bound, hard to remove
• Li, Na, K: alkalis that weaken high-temperature behaviour
• Fe, Ti and transition metals: melt contamination
• B, P: dopants that shift silicon resistivity

How the elements are measured

The common laboratory route is inductively coupled plasma analysis. The quartz is first dissolved, typically with hydrofluoric acid, and the solution is passed into an argon plasma at roughly 10,000 K that atomises and ionises it. ICP-OES reads the light emitted by each element; ICP-MS separates ions by mass and reaches far lower levels, which is why it is the usual choice for HPQ.

Glow-discharge mass spectrometry (GDMS) analyses solid samples directly and is used for high-purity materials where dissolution itself could add contamination. Laser-ablation ICP-MS can also map impurities within individual grains. Whatever the method, sample preparation, reagent purity and clean-lab conditions matter, because at ppb levels the lab can contaminate the result.

Reading a certificate of analysis

A useful certificate lists each element separately in ppm or ppb, states the method and the lab, gives detection limits so a 'not detected' means something, and identifies the lot and the sampling basis. Without those, two certificates with the same total can describe very different sand.

• Element-by-element results, not just a total or a percentage
• Method (ICP-MS, ICP-OES, GDMS) and lab name
• Detection limit for each element
• Lot number, sampling method and date
• Whether the sample was taken before or after final processing

Sampling and batch consistency

A single good result proves little. Crucible makers care about consistency from lot to lot because the inner layer of every crucible depends on it. Buyers should ask how often lots are tested, whether retained samples are kept, and how out-of-specification lots are handled. Independent re-testing of a few lots at a second lab is a common way to check a new supplier.

How testing connects to the ingot

Sand purity carries through the chain. Impurities in the crucible's inner layer can enter the melt, promote bubbles and uneven devitrification, and so affect oxygen behaviour and dislocation-free yield in the ingot. Testing the sand is the cheapest point at which to catch a problem that would otherwise surface as a lost pull.

Questions to ask any HPQ supplier

These apply whichever country the sand comes from.

• Which elements do you report, and in what units?
• Which method and which lab, and is the lab accredited?
• What are the detection limits?
• How are lots sampled and how often are they tested?
• Is the product lump, sand or beneficiated sand, and at what size range?
• Can we re-test retained samples independently?