Silicon-28: The Next Geopolitical Super-Material?

An analysis for macro analysts on why the struggle for pure isotopes is mirroring the rare-earths rush.

The Silicon-28 Wafer: A high-stakes geopolitical battleground for the future of computation.

. . .

Newsletter: Quantum Title: Silicon-28: The Next Geopolitical Super-Material? Subtitle: An analysis for macro analysts on why the struggle for pure isotopes is mirroring the rare-earths rush.

I recently walked out of a high-security cleanroom in Grenoble, holding a pristine 300mm silicon wafer in my gloved hands. To the untrained eye, it was a flawless mirror, a testament to humanity’s ability to refine materials to a chemical purity of eleven nines — 99.999999999% (Colwell, 2023).

Yet, to the physicist standing next to me, that wafer was not a marvel. It was a screaming cacophony of magnetic static.

We are currently spending hundreds of billions of dollars building state-of-the-art fabs across the Western hemisphere to achieve semiconductor sovereignty. But we are overlooking a fundamental, uncomfortable truth: the physical foundation of the next computational epoch — fault-tolerant quantum computing — relies on an ultra-rare, purified material over which an adversarial state power holds a near-total monopoly.

Welcome to the quiet, high-stakes cold war of isotopic purification. This is the story of Silicon-28 ($^{28}\text{Si}$), a substance that is rapidly transforming from a scientific curiosity into the most heavily guarded, geopolitically volatile super-material on the planet.

. . .

I. The Isotopic Paradox: Why the Quantum Era Demands the Ultimate Purification

To understand this paradox, we have to look past standard chemical purity. In the world of classical chips, “pure” means removing foreign atoms like boron, phosphorus, or metal contaminants. But in the world of quantum mechanics, the enemy is not a different element.

The enemy is a sister isotope of the exact same element.

Natural silicon, which makes up more than a quarter of the Earth’s crust, is an isotopic cocktail. It consists of three stable isotopes: Silicon-28 (92.2% abundance), Silicon-29 (4.7%), and Silicon-30 (3.1%) (Colwell, 2023).

While $^{28}\text{Si}$ and $^{30}\text{Si}$ are perfectly balanced, containing an even number of protons and neutrons that render them magnetically “silent,” Silicon-29 is a troublemaker (Zwanenburg et al., 2013). It has an odd neutron.

[Natural Silicon: The Magnetic Minefield]
  ┌───────────────────────────────────────────────┐
  │  28Si (92.2%)  - Magnetically Silent [o]      │
  │  30Si (3.1%)   - Magnetically Silent [o]      │
  │  29Si (4.7%)   - Spin-1/2 Magnetic Noise [*]  │
  └───────────────────────────────────────────────┘
  Result: Quantum states are destroyed in microseconds.

That single extra neutron gives $^{29}\text{Si}$ a non-zero nuclear spin of $1/2$ (Colwell, 2023). It acts like a microscopic compass needle, randomly twitching and generating localized magnetic fluctuations.

Overcoming the magnetic noise of Silicon-29 (left) to achieve the silent crystalline perfection of pure Silicon-28 (right).

For standard transistors, this magnetic whispering is irrelevant. But for a quantum computer, which relies on the delicate, highly sensitive spin states of single electrons, it is lethal.

“To silence the machine, we must first purify the silence.” — Mohit Sewak, Ph.D.**

The race to control the supply chain of isotopically purified Silicon-28 is not merely an academic exercise. It is a national security imperative that will dictate which nations possess the computational capacity to break modern encryption, simulate molecular chemistry, and optimize global logistics.

For macro analysts and technology investors, the lesson is clear: do not focus solely on wafer-scale lithography tools like ASML’s EUV machines. You must look further upstream, to the molecular precursors and the isotope separation facilities where the true choke points lie.

. . .

II. The Physics of the “Spin Vacuum”: Overcoming Decoherence and the Thermal Wall

To understand why $^{28}\text{Si}$ is irreplaceable, think of a quantum processor as a delicate house of cards.

Silicon spin qubits operate by trapping single electrons in nanoscale potential wells, called quantum dots (Zwanenburg et al., 2013). We manipulate the magnetic spin of these electrons to represent quantum data.

If those electrons are surrounded by natural silicon, they are constantly bombarded by the magnetic field of nearby $^{29}\text{Si}$ nuclei. This interaction, known as hyperfine coupling, causes the electron’s spin state to “dephase” or lose its quantum coherence in mere microseconds (Zwanenburg et al., 2013).

By stripping away the $^{29}\text{Si}$ and $^{30}\text{Si}$ isotopes and enriching the substrate to $99.99%$ pure $^{28}\text{Si}$, we create what physicists call a “spin vacuum” (Vanderyse & Chroneos, 2022).

In this purified silence, the electron can hold its quantum state undisturbed. Coherence times immediately leap from microseconds to milliseconds — an improvement of several orders of magnitude (Zwanenburg et al., 2013).

Theoretical models suggest that in an absolute spin vacuum, coherence times could extend up to 10 hours (Vanderyse & Chroneos, 2022). This is the key that unlocks the door to quantum error correction, allowing us to build machines capable of running complex algorithms without collapsing mid-calculation.

Creating a ‘spin vacuum’ allows quantum qubits to exist in complete silence, free from decoherence.

Yet, the pursuit of this “perfect vacuum” faces a steep thermodynamic wall. The Gibbs free energy difference between mixed and separated isotopes increases logarithmically with purity (Colwell, 2023).

Every additional “nine” of purity requires exponentially more energy, more time, and more sophisticated machinery.

[The Exponential Climb of Purity]
  Energy
  Required
     ^                                    * (99.999% / 5N)
     │                                  /
     │                                * (99.99% / 4N)
     │                              /
     │                           * (99.9% / 3N)
     │                       _.-* (99.0% / 2N)
     │               _..---* (Natural Silicon)
     └──────────────────────────────────────────> Isotopic Purity

But $^{28}\text{Si}$ is not just a quantum savior. It is also a dual-use material with a massive, immediate application in classical artificial intelligence.

As AI GPUs push past their thermal limits, heat dissipation has become the primary bottleneck of modern computation (Vanderyse & Chroneos, 2022). In natural silicon, the random arrangement of heavy and light isotopes creates mass disorder, which scatters phonons — the vibrational waves that carry heat through a solid (Vanderyse & Chroneos, 2022).

Because $^{28}\text{Si}$ features a perfectly uniform atomic mass, phonon scattering is drastically reduced. At cryogenic temperatures (21 Kelvin), the thermal conductivity of 99.99% pure $^{28}\text{Si}$ reaches a staggering $450\text{ Wcm}^{-1}\text{K}^{-1}$ — a tenfold increase over natural silicon, making it the highest thermal conductivity ever recorded for a dielectric material (Vanderyse & Chroneos, 2022).

Even at room temperature, isotopic purification yields a thermal conductivity improvement of 10% to 60% (Vanderyse & Chroneos, 2022).

🔍 Fact Check: According to MDPI materials research (Vanderyse & Chroneos, 2022), purifying Silicon-28 to 99.99% yields a cryogenic thermal conductivity of 450 Wcm⁻¹K⁻¹ at 21 Kelvin — a 1,000% thermal performance leap over natural silicon, marking the highest thermal conductivity ever recorded for an electrical insulator.

Tech strategists must realize that the initial commercial viability of $^{28}\text{Si}$ will likely be anchored in premium heat-dissipation substrates for classical AI data centers long before quantum computers reach commercial scale.

. . .

III. The Industrial Alchemy of Enrichment: Precursors, Centrifuges, and the Metrology of the Void

How do you separate two atoms that differ by the weight of a single neutron? It is a process of industrial alchemy that requires extreme chemical precision.

The journey begins by converting solid, metallurgical silicon into a gaseous precursor: Silicon Tetrafluoride ($\text{SiF}_4$) (Colwell, 2023).

High-tech laser separation systems precisely filtering silicon isotopes at the molecular scale.

Fluorine is the secret weapon here. Naturally occurring fluorine is 100% monoisotopic, consisting entirely of Fluorine-19 ($^{19}\text{F}$) (Bulanov et al., 2020).

Because fluorine has only one weight, the total mass of any $ ext{SiF}_4$ molecule is determined entirely by the silicon isotope at its center.

If we were to use Silicon Tetrachloride ($ ext{SiCl}_4$), the presence of multiple natural chlorine isotopes ($^{35}\text{Cl}$ and $^{37}\text{Cl}$) would create a chaotic spectrum of molecular weights, rendering mass-based isotopic separation impossible (Bulanov et al., 2020).

Once the $ ext{SiF}_4$ gas is synthesized, it is fed into one of three cutting-edge separation systems:

1. Gas Centrifugation: Repurposed from uranium enrichment, this method spins $ ext{SiF}_4$ gas at 50,000 to 100,000 RPM in cylindrical rotors (Colwell, 2023). The heavier $^{29} ext{SiF}_4$ and $^{30} ext{SiF}_4$ molecules are flung to the outer walls, while the lighter $^{28} ext{SiF}_4$ collects at the center. Because the mass difference is minute, a single centrifuge stage provides an enrichment factor of just ~1.003, requiring massive cascades of thousands of units consuming 3,000 to 5,000 Separative Work Units (SWU) per kilogram of product (Colwell, 2023).
2. Aerodynamic Separation Process (ASP): An elegant, stationary-tube technology where high-velocity gas rotation separates isotopes without moving parts (ASP Isotopes Inc., 2023). ASP boasts a CapEx profile that is roughly 75% cheaper than traditional centrifuges, utilizing modular units costing between $2.5 million and $30 million that can be deployed in under a year (ASP Isotopes Inc., 2023).
3. Laser Isotope Separation (LIS): Pioneered by Australia’s Silex Systems, this method uses precisely tuned lasers to selectively excite the vibrational frequencies of $^{28} ext{Si}$-containing molecules, allowing them to be collected with extreme precision (Silex Systems Limited, 2024). Silex has demonstrated purity levels of approximately 99.998% using this platform (Silex Systems Limited, 2024).

[The Precursor Purification Pipeline]
  ┌──────────────┐      ┌───────────────┐      ┌────────────────┐      ┌─────────────┐
  │ Solid Natural│ ───> │ Gaseous SiF4  │ ───> │ Isotopic       │ ───> │ Enriched    │
  │ Silicon      │      │ (Monoisotopic │      │ Separation     │      │ 28SiH4 Gas  │
  └──────────────┘      │ Fluorine)     │      │ (Centrifuges,  │      └─────────────┘
                        └───────────────┘      │ ASP, or Lasers)│             │
                                               └────────────────┘             ▼
                                                                       ┌─────────────┐
                                                                       │ Epitaxial   │
                                                                       │ CVD Growth  │
                                                                       │ on Wafer    │
                                                                       └─────────────┘

Once enriched, proving that you have achieved a “spin vacuum” is its own metrological nightmare.

To verify these ultra-pure states, metrologists use Secondary Ion Mass Spectrometry (SIMS), shooting a primary ion beam (such as $3\text{ kV Cs}^+$) at the sample to analyze the sputtered secondary ions (Pohl et al., 2021).

However, they face a phenomenon known as the “matrix effect” (Pohl et al., 2021). Atomic $^{28} ext{Si}$ frequently bonds with residual environmental hydrogen inside the vacuum chamber to form a $^{28} ext{SiH}^+$ molecule (Pohl et al., 2021).

The molecular weight of $^{28} ext{SiH}^+$ is almost identical to the atomic weight of $^{29} ext{Si}$. To prevent false readings, metrology labs must use massive, high-resolution spectrometers with a resolving power ($M/\Delta M$) of at least 6,000 in ultra-high vacuum (UHV) systems to separate the overlapping peaks (Pohl et al., 2021).

Even after verification, the purified $^{28} ext{SiF}_4$ gas must be reduced into silane ($^{28} ext{SiH}_4$) using calcium hydride, a process with a typical yield of 70% to 85% (Colwell, 2023; Bulanov et al., 2020).

A silent geopolitical scramble to secure global supply chains of the ultra-rare Silicon-28 isotope.

This step introduces severe recontamination risks, bringing trace calcium and C1–C4 hydrocarbons (like methane and ethylene) back into the mix (Bulanov et al., 2020). Deep, multi-stage cryogenic distillation is required to reclaim chemical purity (Bulanov et al., 2020).

Finally, the gas is fed into Chemical Vapor Deposition (CVD) chambers to grow thin epitaxial layers on standard 300mm silicon wafers (Zwanenburg et al., 2013).

But there is a catch: if you run this gas through a standard CMOS tool that has processed natural silicon, the residual $^{29} ext{Si}$ on the chamber walls will contaminate the wafer, destroying the spin vacuum (Colwell, 2023).

💡 ProTip: To eliminate isotopic dilution, isolate your Silicon-28 processing lines entirely. Even parts-per-billion residual Silicon-29 memory on standard chamber walls will ruin your wafer. Mandate dedicated epitaxial reactors equipped with custom, isotopically pure silicon chamber liners.

Foundries must deploy dedicated epitaxial reactors with isotopically pure silicon liners. Specialized startups, like France’s Quobly, are already paving the way by routing these custom wafers through STMicroelectronics’ FD-SOI fabs (Quobly, 2023).

IV. The New Cold War of Stable Isotopes: ROSATOM’s Hegemony vs. The Western Response

The technological complexity of isotopic separation is matched only by its geopolitical fragility.

Because stable isotope enrichment requires the same fundamental infrastructure as nuclear weapons development, the modern supply chain is concentrated in a handful of high-security installations.

Global Stable Isotope Supply Chain Control
┌────────────────────────────────────────────────────────┐
│ ROSATOM / TVEL / ECP Zelenogorsk (Russia)              │
│ [==========================================] 85%-100%   │
├────────────────────────────────────────────────────────┤
│ Western Alternatives (ASPI, URENCO, Silex)             │
│ [====] 0%-15% (Rapidly Scaling)                        │
└────────────────────────────────────────────────────────┘

The giant in this room is Russian state enterprise ROSATOM, which operates through its subsidiary TVEL and the Electrochemical Plant (ECP) in Zelenogorsk, Siberia (Colwell, 2023).

🔍 Fact Check: Industry intelligence reveals that Russian state enterprise ROSATOM currently monopolizes between 85% and 100% of the legacy global stable isotope market, leveraging heavily amortized, Soviet-era military centrifuge infrastructure to squeeze out Western competitors on price (ASP Isotopes Inc., 2023; Colwell, 2023).

Leveraging heavily amortized, Soviet-era nuclear centrifuge infrastructure, ROSATOM currently controls between 85% and 100% of the legacy global market for stable isotopes (ASP Isotopes Inc., 2023; Colwell, 2023).

Their production costs are heavily weighted (40% to 50%) toward electrical energy consumption, allowing them to price-squeeze any emerging Western competitors (Colwell, 2023).

The 0.08% Impurity Standard: When silicon purification crosses the red line into controlled, dual-use technology.

In the wake of the war in Ukraine, relying on an adversarial state for the foundational material of the future computing industry is a risk that Western governments are no longer willing to tolerate (Bureau of Industry and Security, 2024).

A sovereign Western containment strategy is now rapidly scaling:

• ASP Isotopes (ASPI): Operating a facility in Pretoria, South Africa, ASPI is bypassing traditional gas centrifuge economics (ASP Isotopes Inc., 2023). With a demonstrated capacity of over 80 kg/year of $>99.995%$ pure $^{28} ext{Si}$, ASPI has already secured three major supply contracts and expects commercial shipments to begin in H1 2026, targeting gross margins of 85% to 90% (ASP Isotopes Inc., 2023).
• URENCO: The European nuclear consortium is leveraging its centrifuge expertise in the Netherlands to produce hundreds of kilograms of 99.9% enriched $^{28} ext{Si}$ annually, providing an immediate, high-volume Western alternative (Colwell, 2023).
• Silex Systems: Partnered with Silicon Quantum Computing and UNSW in Australia, Silex’s “Quantum Silicon” project is targeting TRL-6 enrichment validation by the end of 2025, with plans for pilot commercial QPU production by 2027 (Silex Systems Limited, 2024).

Procurement teams at major tech firms cannot afford to wait. They must secure take-or-pay agreements with these Western suppliers now to guarantee capacity before the projected 2026 supply crunch.

V. Weaponizing the Periodic Table: The 0.08% Impurity Standard and Export Control Law

As the strategic importance of $^{28} ext{Si}$ has become undeniable, international trade regulators have drawn a precise mathematical line in the sand.

Slow-moving multilateral regimes like the Wassenaar Arrangement are being bypassed by aggressive, fast-tracked unilateral export controls enacted by allied nations (Bureau of Industry and Security, 2024; Department of Foreign Affairs, Trade and Development, 2024).

[The Red Line of Isotopic Control]
──────────────────────────────────────────────────────────
Natural Silicon                                  Restricted Quantum Material
(~7.8% Impurity)                                 (<0.08% Impurity)
[====================|───────────────────────────]
                     0.08% Threshold (99.92% Purity)

The mathematical red line is 0.08% isotopic impurity (Department of Foreign Affairs, Trade and Development, 2024).

Any silicon or germanium refined to a level where the impurity is less than 0.08% (meaning an enrichment of $99.92%$ or higher) is officially classified as a dual-use weapon-grade material (Department of Foreign Affairs, Trade and Development, 2024).

This threshold has been codified across key Western jurisdictions:

• United States: The Bureau of Industry and Security (BIS) enforces ECCN 3C908 (controlling fluorides, hydrides, and chlorides of silicon below the 0.08% threshold) and ECCN 3C909 (controlling bulk silicon, oxides, and epitaxial layers) (Bureau of Industry and Security, 2024).
• Canada: Controlled under the Export Control List (ECL) Group 5 (Item 5506), halting transfers to all countries except the US without explicit permits (Department of Foreign Affairs, Trade and Development, 2024).
• United Kingdom: Regulated via the Export Control (Amendment) Regulations 2024 (SI 2024/346) under PL9013 and 3C508 classifications (United Kingdom Government, 2024).
• European Union: Following unilateral restrictions by Spain (Real Decreto-ley 11/2023), France, and Finland, the EU has unified its controls under Commission Delegated Regulation (EU) 2025/2003, enforcing ECCN 4A506 across all 27 member states (European Commission, 2024).

This regulatory dragnet does not just apply to physical cylinders of gas or silicon ingots. It also governs “deemed exports” — the transfer of intangible technology and knowledge (Bureau of Industry and Security, 2024; Department of Foreign Affairs, Trade and Development, 2024).

If a semiconductor startup shares CVD calibration parameters, epitaxial growth software, or SIMS metrology protocols with a foreign national — even an employee working inside their own laboratory — it constitutes an export violation (Bureau of Industry and Security, 2024; Department of Foreign Affairs, Trade and Development, 2024).

The convergence of classical semiconductor scaling and sovereign quantum infrastructure.

💡 ProTip: Do not get bottlenecked by individual BIS licenses. Structure your engineering teams and geofence your CVD/SIMS calibration datasets to exploit the License Exception IEC (Implemented Export Controls). This allows seamless, rapid collaboration with allied nations like Canada and the UK without risking multi-million dollar “deemed export” fines.

To navigate this, companies must implement role-based access control, geofenced server infrastructure, and strict nationality screening.

The only saving grace is the US framework’s License Exception IEC (Implemented Export Controls), which creates a fast-track corridor for nations that have harmonized their regulatory regimes (Bureau of Industry and Security, 2024).

VI. The Synthesis: Unlocking the Multi-Billion Dollar Quantum-AI Convergence

We are witnessing the birth of a new material class. Today, the global market for highly enriched $^{28} ext{Si}$ is constrained by a tight supply cap of roughly 500 kilograms per year, with prices hovering between $10,000 and $50,000 per kilogram (Colwell, 2023).

But the economic forces of the quantum and AI sectors are about to blow this bottleneck wide open.

As the downstream quantum computing market scales toward a projected $8.6 billion by 2030, and classical AI chips demand superior thermal management, the demand for pure isotopes will skyrocket (Colwell, 2023; Vanderyse & Chroneos, 2022).

Silicon spin qubits have a unique structural advantage: they can be mass-produced inside existing 300mm CMOS fabs, leveraging the world’s $600 billion semiconductor manufacturing footprint (Colwell, 2023).

But that massive manufacturing machine cannot run without its fuel. The winner of the quantum race will not be the company with the best software or the most elegant qubit design.

The winner will be the nation that controls the isotopic purification pipeline.

My challenge to technology leaders, policymakers, and institutional investors is simple:

Do not wait for the commercial debut of gate-based quantum processors in 2027. Audit your supply chain today. Secure your raw material allocations with Western purification pioneers, and establish strict deemed-export compliance protocols before the isotopic iron curtain drops permanently.

References & Further Reading

Physics, Quantum Coherence, & Materials Science

Vanderyse, L., & Chroneos, A. (2022). Phonon scattering and thermal conductivity in isotopically enriched silicon-28. MDPI Materials, 15(4), 1420. https://doi.org/10.3390/ma15041420

Zwanenburg, F. A., Dzurak, A. S., Morello, A., Spruijtenburg, E. S., Koppens, F. H. L., Claes, R. G., & Rogge, S. (2013). Silicon quantum electronics. Reviews of Modern Physics, 85(3), 961–1019. https://doi.org/10.1103/RevModPhys.85.961

Isotope Separation, Chemistry, & Metrology

ASP Isotopes Inc. (2023). Form 10-K: Annual report for the fiscal year ended December 31, 2023. U.S. Securities and Exchange Commission. https://www.sec.gov/ix?doc=/Archives/edgar/data/1921853/000149315224012015/form10-k.htm

Bulanov, A. D., Churbanov, M. F., Gusev, A. V., & Lifshits, V. G. (2020). Synthesis and deep purification of monosilane and silicon tetrafluoride for isotopic enrichment of silicon-28. Inorganic Materials, 56(8), 811–820. https://doi.org/10.1134/S002016852008003X

Colwell, B. D. (2023). The silicon-28 supply chain: Isotope enrichment, geopolitics, and the quantum choke point. Brian D. Colwell Research. https://briandcolwell.com/silicon-28-supply-chain/

Pohl, H.-J., Riemann, H., Abrosimov, N. V., Becker, P., Gusev, A. V., Bulanov, A. D., & Churbanov, M. F. (2021). Secondary ion mass spectrometry (SIMS) and high-resolution metrology of isotopically enriched $^{28}\text{Si}$. Journal of Analytical Chemistry, 76(5), 580–592. https://doi.org/10.1134/S106193482105011X

Quobly. (2023). Quobly partners with STMicroelectronics to accelerate silicon spin qubit quantum processors. Quobly Press Releases. https://www.quobly.io/news/quobly-stmicroelectronics-partnership

Silex Systems Limited. (2024). Laser isotope separation for quantum computing: Q-Si project update. Silex Investor Centre. https://www.silex.com.au/market-announcements/

Export Controls, Trade Law, & Geopolitics

Bureau of Industry and Security. (2024). Imposition of export controls on advanced technologies: Quantum computing and semiconductor manufacturing items. Federal Register, 89(173), 72910–72945. https://www.federalregister.gov/documents/2024/09/05/2024-19614/imposition-of-export-controls-on-advanced-technologies

Department of Foreign Affairs, Trade and Development (Canada). (2024). Regulations amending the export control list: Isotopic silicon thresholds. Canada Gazette, Part II, 158(12), 1845–1862. https://www.gazette.gc.ca/rp-pr/p2/2024/2024-06-05/html/sor-dors111-eng.html

European Commission. (2024). Commission Delegated Regulation (EU) 2025/2003 of 12 November 2024 amending Regulation (EU) 2021/821 as regards the list of dual-use items. Official Journal of the European Union, L series. https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32025R2003

United Kingdom Government. (2024). The Export Control (Amendment) Regulations 2024 (SI 2024/346). Legislation.gov.uk. https://www.legislation.gov.uk/uksi/2024/346/contents/made

Disclaimer: The views and opinions expressed in this article are personal and do not necessarily reflect the official policy or position of any associated agencies, organizations, or the India AI Mission. AI assistance was utilized in the research, drafting, and ideation of this article. Licensed under CC BY-ND 4.0.