TLDR
Natural silicon is a mix of three stable isotopes: silicon-28 (92.2%), silicon-29 (4.7%), and silicon-30 (3.1%), plus about two dozen radioactive ones that don’t occur in nature. Silicon-28 is the one chipmakers and quantum computing labs pay a premium to isolate, because its nucleus has no spin and doesn’t scramble a qubit’s quantum state. Silicon-29 has the opposite property — a nuclear spin — which makes it useful for NMR spectroscopy. And ratios of silicon-30 to silicon-28 in rocks, rivers, and diatom shells let geochemists reconstruct weathering rates and ancient climates.
Table of Contents
- What “Silicon Isotope” Actually Means
- The Three Stable Isotopes, Side by Side
- Why Silicon-28 Is the Star of Quantum Computing
- Silicon-29 and the Chemistry of NMR
- Silicon Isotopes as a Geological Clock
- The Radioactive Silicon Isotopes
- FAQ
What “Silicon Isotope” Actually Means
Every silicon atom has 14 protons — that’s what makes it silicon instead of aluminum or phosphorus. What varies is the neutron count, and that’s the whole story of isotopes. Add 14 neutrons to those 14 protons and you get silicon-28. Add 15 and it’s silicon-29. Add 16 and it’s silicon-30. Same element, same chemistry, different mass and — for two of the three — different nuclear behavior.
Silicon isn’t rare. It’s the second most abundant element in Earth’s crust after oxygen, locked up mostly in quartz and silicate minerals. But almost all of that silicon skews toward one isotope: silicon-28 makes up just over 92% of every silicon atom you’ll find, whether it’s in a beach, a computer chip, or your own bones (silicon shows up there too, in trace amounts). The other two stable isotopes fill out the remaining 8%.
The Three Stable Isotopes, Side by Side
| Isotope | Neutrons | Atomic mass (u) | Natural abundance | Nuclear spin | Notable use |
|---|---|---|---|---|---|
| Silicon-28 | 14 | 27.9769 | 92.22% | 0 (spinless) | Quantum computing, high-precision semiconductors |
| Silicon-29 | 15 | 28.9765 | 4.69% | 1/2 | NMR spectroscopy |
| Silicon-30 | 16 | 29.9738 | 3.09% | 0 (spinless) | Geochemical tracer, silicon-31 production |
These figures come from NIST’s Atomic Weights and Isotopic Compositions database, the standard reference chemists and physicists cite for this kind of data. Natural silicon’s abundance ratios do shift slightly depending on where you sample it — meteorites, seawater, and granite each carry a faint isotopic fingerprint — which is exactly why the ratios turn out to be scientifically useful rather than just a curiosity.
Why Silicon-28 Is the Star of Quantum Computing
Ordinary computer chips are made from silicon straight off the periodic table — a natural mix of all three isotopes, and nobody cares, because transistors don’t notice nuclear spin. Qubits do.
A silicon-based quantum computer stores information in the spin of an electron sitting in a tiny pocket of silicon. That spin is fragile. Every nearby atomic nucleus with its own spin acts like a tiny magnet, and the collective noise from billions of those magnets is enough to scramble a qubit’s state in microseconds — a problem called decoherence. Silicon-29, with its nuclear spin of 1/2, is the main offender. Silicon-28 and silicon-30 both have zero nuclear spin, so they’re magnetically silent.
The fix is straightforward in principle and brutal in practice: enrich the silicon down to almost pure silicon-28, stripping out the silicon-29 that’s causing the noise. Labs and fabs now routinely target 99.9% to 99.99% silicon-28 purity for quantum devices, and Chinese researchers reported domestic mass production exceeding 99.99% enrichment in 2026. In isotopically pure silicon-28, the theoretical decoherence limit for a qubit stretches out to roughly 10 hours — an enormous jump from what natural silicon allows. That gap is why isotope purification has become its own commercial industry, with enrichment facilities in South Africa and elsewhere charging on the order of tens of thousands of dollars per kilogram, climbing steeply with each additional nine of purity.
The payoff isn’t limited to qubits. Enriched silicon-28 also conducts heat noticeably better than the natural mix — about 150 W/m·K versus 130 W/m·K — because isotopic uniformity reduces the phonon scattering that comes from having atoms of different masses jumbled together in the crystal lattice. That’s a real engineering advantage for densely packed chips, quantum or otherwise, as detailed in research on targeted isotope enrichment for quantum device fabrication.
Silicon-29 and the Chemistry of NMR
Silicon-29’s nuclear spin, the same property that makes it a liability in a qubit, is exactly what makes it valuable at the chemist’s bench. Nuclear magnetic resonance spectroscopy only works on isotopes that have spin, because the technique depends on nuclei behaving like tiny magnets that flip in response to a magnetic field. Silicon-28 and silicon-30 are invisible to NMR. Silicon-29 is the only stable silicon isotope the technique can see.
That makes 29Si NMR the go-to method for mapping the structure of silicate minerals, zeolites, silicone polymers, and glass — anything built around a silicon-oxygen backbone. It has one quirk that trips up newcomers: silicon-29 has a negative gyromagnetic ratio, meaning its magnetic moment points opposite to its spin. In practice, that inverts the nuclear Overhauser effect and can null out signals entirely when protons are nearby, so chemists lean on pulse sequences like INEPT or DEPT, which sidestep the sign problem, to get a usable spectrum. None of this requires isotopically enriched material — silicon-29’s natural 4.7% abundance is already enough signal to work with, just a faint one.
Silicon Isotopes as a Geological Clock
Move from a cleanroom to a riverbed and silicon isotopes tell a completely different kind of story. Instead of nuclear spin, geochemists care about mass — specifically, the ratio of silicon-30 to silicon-28, written as δ30Si.
Here’s the mechanism: when silicate rock weathers, lighter isotopes react and dissolve slightly faster than heavier ones. That means the dissolved silicon carried off in rivers ends up isotopically lighter than the rock it came from, while the leftover clay minerals get isotopically heavier. Measure that split and you get a proxy for how intensely a landscape has weathered — a signal that shows up in river water, ocean sediment, and even ancient rock records stretching back hundreds of millions of years, as reviewed in research on silicon isotope fractionation and its geological applications.
Diatoms — single-celled algae that build shells out of silica — add a second layer to the story. When diatoms pull dissolved silicon out of seawater to build their shells, they preferentially take up the lighter isotopes, leaving the surrounding water isotopically heavier. Because diatom blooms track ocean productivity, and ocean productivity ties into how much carbon dioxide the ocean pulls out of the atmosphere, the δ30Si trapped in fossil diatom shells buried in seafloor sediment becomes a way to read past carbon cycling and past climate, layer by layer, going back through ice ages. It’s one of the few chemical signals that connects rock weathering on land directly to the carbon cycle in the ocean.
The Radioactive Silicon Isotopes
Beyond the three stable isotopes, silicon has roughly two dozen known radioactive isotopes, ranging from silicon-22 to silicon-44. None occur naturally in any measurable quantity — they’re produced in reactors, cyclotrons, or by cosmic ray interactions, and they decay away fast. Two are worth knowing:
Silicon-31 has a half-life of about 157 minutes. It’s typically made by bombarding silicon-30 with neutrons, and it decays into phosphorus-31 by beta emission. Its main use is academic and analytical — silicon-31 production is the basis of neutron activation analysis for measuring trace silicon in a sample, since you can quantify the original silicon-30 by counting the radiation as it decays.
Silicon-32 is far longer-lived, with a half-life measured in centuries rather than minutes, and it’s produced in small quantities for research through spallation reactions. It’s rare enough that the U.S. National Isotope Development Center lists it as a specialty product available to research groups rather than something with everyday industrial use.
Neither radioactive isotope shows up outside a lab or a research reactor. If you’re holding a piece of silicon — sand, glass, a silicon chip — every atom in it is one of the three stable isotopes.
FAQ
Why is silicon-28 used in quantum computers? Because its nucleus has zero spin, it doesn’t generate the magnetic noise that scrambles a qubit’s quantum state. Removing silicon-29 (which does have spin) by enrichment gives qubits dramatically longer coherence times.
How do silicon isotopes trace geological processes? Lighter silicon isotopes dissolve and get incorporated into biology slightly faster than heavier ones. That fractionation leaves a measurable fingerprint — the δ30Si ratio — in rivers, marine sediment, and diatom shells that scientists use to reconstruct past weathering rates and climate conditions.
What’s the difference between silicon-28 and regular silicon? “Regular” silicon is already about 92% silicon-28 — the difference is purity. Isotopically enriched silicon-28 has had the silicon-29 and silicon-30 stripped out, often to 99.9%+ purity, for applications where nuclear spin or atomic mass uniformity matters.
Is silicon-29 radioactive? No. All three of silicon’s naturally occurring isotopes — silicon-28, silicon-29, and silicon-30 — are stable and don’t decay. Only the lab-made isotopes, like silicon-31 and silicon-32, are radioactive.
How expensive is enriched silicon-28? Pricing scales steeply with purity. Material enriched to around 99.9% has been quoted in the tens of thousands of dollars per kilogram, and pushing purity another order of magnitude higher raises the cost considerably further.
