What project owners need to know.
Float-zone (FZ) silicon is single-crystal silicon grown without a crucible, by passing a narrow molten zone along a silicon rod, while Czochralski (CZ) silicon is pulled from a melt held in a fused-quartz crucible. Because nothing touches the FZ melt, its oxygen and carbon levels are typically below 5 × 10¹⁵ cm⁻³, against about 5–10 × 10¹⁷ cm⁻³ oxygen in CZ silicon, according to PV-Manufacturing.org. FZ is more expensive and is limited to about 200 mm in diameter, so it serves power devices, detectors and high-resistivity niches, while CZ, at up to 300 mm and lower cost, carries solar and mainstream semiconductor production.
Two ways to grow a silicon crystal
Almost every single-crystal silicon wafer starts in one of two processes. In the Czochralski (CZ) method, polysilicon is melted in a fused-quartz crucible and a rotating seed crystal pulls a cylindrical ingot out of the melt. PV-Manufacturing.org describes CZ ingots of about 200 to 300 mm in diameter, depending on the target wafer size. This is the route behind solar wafers and mainstream semiconductor wafers.
The float-zone (FZ) method has no crucible at all. A radio-frequency coil melts a short zone of a vertically held silicon rod, a seed crystal starts single-crystal growth, and the molten zone is moved slowly along the rod so that it refreezes as a crystal. Surface tension holds the molten zone in place. The technique dates to 1955 at Bell Labs, building on zone refining. PV-Manufacturing.org notes that the feed rod is typically high-purity silicon, often itself grown by CZ, and that a necking step is used, as in CZ, to keep the crystal dislocation-free.
- CZ: melt in a fused-quartz crucible, crystal pulled upward from it
- FZ: no crucible; a molten zone travels along a rod and refreezes as a crystal
- Both start from a seed and use necking to eliminate dislocations
Purity: why the crucible matters
The decisive difference is contact. In CZ growth the hot melt sits against the quartz crucible, which slowly dissolves and adds oxygen to the crystal. PV-Manufacturing.org gives about 5–10 × 10¹⁷ cm⁻³ oxygen for CZ silicon. In FZ growth nothing touches the melt, and most impurities prefer to stay in the moving liquid zone rather than freeze into the crystal, so the zone sweeps them toward the end of the rod. PV-Manufacturing.org gives typical oxygen and carbon levels in FZ silicon below 5 × 10¹⁵ cm⁻³, roughly a hundred times lower.
That matters because oxygen is not inert. PVEducation notes that oxygen in CZ wafers reduces minority carrier lifetime and can become active during high-temperature processing. For a solar cell this is manageable, and CZ remains the industry standard. For devices that need very high resistivity, long carrier lifetimes or tightly controlled behaviour at high voltage, low oxygen is a design requirement rather than a nice-to-have. FZ makers may still add elements on purpose: nitrogen, for example, is added during some FZ growth to control microdefects and improve mechanical strength.
Why FZ stays a niche: diameter and cost
FZ trades scale for purity. Holding a molten zone by surface tension becomes harder as the crystal widens. PV-Manufacturing.org says FZ wafers are usually limited to 150 mm, and an Infineon patent states that FZ ingots are limited to about 200 mm in diameter, while CZ can produce 300 mm and larger. Specialist suppliers do offer 200 mm FZ wafers, which appears to be the practical ceiling today.
FZ is also more expensive. Both PVEducation and the same Infineon patent say so directly, and the patent describes work to make cheaper, larger CZ material usable for devices that have traditionally relied on FZ. That is the long-running tension in power semiconductors: FZ offers the purity, CZ and magnetic CZ (MCZ) offer the diameter and the cost curve.
Neutron transmutation doping (NTD)
High-voltage devices need dopant spread evenly through the wafer, because local variations change how much voltage a device can block. FZ silicon can be doped very uniformly by neutron transmutation doping. About 3% of natural silicon is the isotope ³⁰Si. When a crystal is irradiated with thermal neutrons in a research reactor, some ³⁰Si nuclei capture a neutron and, through a short-lived intermediate, become stable phosphorus-31, an n-type dopant.
The Heinz Maier-Leibnitz research reactor (FRM II) at the Technical University of Munich describes the result as a particularly homogeneous phosphorus distribution, helped by a rotating irradiation basket. It irradiates crystals up to 20 cm in diameter and 50 cm high, to resistivities of 25 to 1,000 Ωcm, and states that it can produce up to 15 tonnes a year, which it puts at about 10% of the global market. NTD needs reactor capacity, which is one reason it remains a specialist supply chain.
- ³⁰Si + neutron → ³¹Si → stable ³¹P (n-type dopant)
- Gives very uniform doping for high-voltage devices
- Depends on research-reactor irradiation capacity
Where each is used
CZ is the volume material. It feeds solar ingots, including the n-type wafers behind TOPCon, HJT and back-contact cells, and the 300 mm wafers used in logic and memory fabs. FZ serves applications where purity or resistivity outweighs cost. Topsil, a Danish FZ maker, lists FZ wafers for IGBTs and diodes in variable-speed drives, NTD FZ wafers for very high-power devices such as HVDC converter stations, and high-resistivity and intrinsic grades for RF, terahertz, detector and quantum uses. FRM II cites thyristors for long-distance power transmission and high-speed trains. PVEducation notes that FZ is mainly used for laboratory or niche high-efficiency solar cells rather than commercial production.
What it means for India
India's semiconductor push is pulling attention toward power electronics for EVs, rail, renewables and the grid, alongside solar. That makes the FZ-versus-CZ question practical for device and module projects: the substrate choice sets purity limits, wafer diameter, supplier concentration and, for NTD material, dependence on reactor capacity abroad. Projects should confirm which substrate their device design actually needs, and whether qualified MCZ material could serve, before locking in a supply chain.
For solar, the lesson runs the other way. CZ will remain the workhorse, and its oxygen story starts with the quartz crucible, which is why crucible purity and hot-zone control are core quality levers for Indian ingot makers. This guide is general awareness, not a substrate recommendation; specifications should be validated with device designers, wafer suppliers and test labs.
How JRST supports this requirement
JRST works on the Czochralski side of silicon: high-purity quartz and crucibles, CZ ingot growth, wafering, equipment sourcing and line consumables for Indian projects. Substrate, purity and dopant specifications for any given project are defined in a project-specific proposal and validated with device and cell partners.
Discuss your requirement
Mohammed Saif Zaveri connects JRST's industrial content to execution conversations.
As Co-Founder and designated partner of JRST Technology LLP, Mohammed Saif Zaveri works across industrial growth, strategic partnerships, client conversations, equipment strategy, and project pathways for silicon, solar, semiconductor, and advanced-manufacturing opportunities.
This knowledge page is part of JRST's public industrial knowledge base, designed to help buyers move from search terms and early research toward a structured technical-commercial discussion.
View Mohammed Saif Zaveri's profileConnect on LinkedInFrequently asked questions
What is the difference between float-zone and Czochralski silicon?
Czochralski silicon is pulled from a melt held in a fused-quartz crucible; float-zone silicon is grown without a crucible by moving a molten zone along a rod. FZ has far lower oxygen and carbon, but is more expensive and limited to about 200 mm diameter, while CZ reaches 300 mm.
Why does float-zone silicon have less oxygen?
Because the melt never touches a crucible. In CZ growth, the hot melt slowly dissolves the fused-quartz crucible, adding oxygen. PV-Manufacturing.org gives typical FZ oxygen and carbon below 5 × 10¹⁵ cm⁻³ versus about 5–10 × 10¹⁷ cm⁻³ oxygen in CZ silicon.
What is FZ silicon used for?
Mainly power semiconductors such as IGBTs, diodes and thyristors, plus high-resistivity uses such as RF, terahertz optics, radiation detectors and some quantum devices. It is also used for laboratory and niche high-efficiency solar cells.
Are solar wafers made from float-zone silicon?
Commercial solar wafers are almost all Czochralski silicon. FZ wafers are more expensive and harder to grow at large diameters, so they are used mainly for laboratory or niche high-efficiency cells.
What is neutron transmutation doping?
A way of doping silicon very uniformly with phosphorus. A crystal is irradiated with thermal neutrons in a reactor; some of the ³⁰Si isotope, about 3% of natural silicon, is converted into stable ³¹P, an n-type dopant.
What is the maximum size of a float-zone wafer?
About 200 mm in diameter. Infineon's patent literature states that FZ ingots are limited to about 200 mm, and PV-Manufacturing.org says FZ wafers are usually 150 mm, while CZ produces 300 mm wafers.
Can CZ silicon replace FZ for power devices?
Partly, and it is an active area of work. Patent literature describes cheaper, larger-diameter CZ and MCZ material being engineered for devices that have traditionally used FZ, but suitability depends on the device design and must be qualified.
Primary sources and further reading
Last reviewed 2026-10-27. Technical scope, policy eligibility, availability, and commercial terms should be independently confirmed for each project.