What project owners need to know.
Magnetic Czochralski (MCz) is a variant of Czochralski crystal growth in which a magnetic field is applied to the molten silicon to suppress or reshape convection in the melt. Because oxygen dissolves into the melt from the quartz crucible and is carried by that convection either to the free surface, where it evaporates as silicon monoxide, or to the growing crystal, controlling the flow gives the grower a handle on how much oxygen ends up in the ingot. Patents from established wafer makers describe cusp-field growth reaching below about 15 ppma, or no greater than about 10 ppma, with oxygen held within 5% radially. MCz does not remove the oxygen source, and it adds a magnet and its utilities to the puller, so it is a specification decision to make before equipment is ordered, not a retrofit to assume later.
The oxygen comes from the crucible; the melt decides where it goes
Every Czochralski ingot is grown from silicon held in a fused-silica crucible, and that crucible dissolves. A 1993 oxygen-control patent originally assigned to MEMC describes the mechanism plainly: at the melt temperature of about 1,420 °C the crucible surface in contact with the melt dissolves, part of the dissolved silica evaporates from the melt surface as silicon monoxide, and another part is incorporated into the growing crystal. A 2024 modelling study in Heliyon describes the same path and notes that oxygen reaching the melt–crystal interface is largely taken up by the crystal. PVEducation puts the result at ppm-level oxygen, around 10¹⁸ cm⁻³, in Cz ingots.
Between those two exits sits the melt flow. Heating from the side drives strong thermal convection, and crystal and crucible rotation add forced flow on top. Whichever way the melt carries its oxygen-rich boundary layer, toward the free surface or toward the crystal, sets the final concentration. This is why oxygen in Cz silicon is a fluid-dynamics problem as much as a materials one, and why a magnetic field, which acts on the flow, can change it.
- Treat the crucible as the oxygen source and the melt flow as the oxygen valve
- Remember that rotation, heater layout and argon flow also move oxygen
- Expect oxygen to vary along the ingot as melt depth and crucible contact change
- Pair this guide with crucible and hot-zone choices, not in isolation
What the magnetic field actually does
Molten silicon conducts electricity, so a moving melt in a magnetic field experiences a Lorentz force opposing the motion. A superconducting-magnet patent for single-crystal pullers describes the purpose directly: the field is applied to the melt in the crucible to prevent convection of the melted material. Suppressing turbulent convection steadies the temperature at the growth interface and gives the grower more control over how dopant and oxygen are distributed.
The geometry of the field matters. In a transverse, or horizontal, field the lines cross the crucible in one direction. In a cusp field two coils carry current in opposite directions, so the field is roughly symmetrical around the pulling axis. A 2021 GlobalWafers patent describes a radial cusp field as inducing convection near the solid–liquid interface while suppressing it in the rest of the melt, allowing oxygen to be controlled by magnetic flux density alone, irrespective of crystal rotation speed. It gives typical cusp flux densities of about 0 to 0.2 T axially and up to about 0.6 T in the radial position.
- Ask which field geometry a puller supports: transverse, cusp or both
- Ask whether oxygen is controlled by field strength, rotation, or both
- Request the flux-density range at the melt, not only the magnet rating
- Treat field geometry as part of the growth recipe, not a hardware detail
What the patent record says about oxygen levels
The clearest public numbers come from patents. US 5,178,720, filed in 1991 and published in 1993, combines a cusp field with counter-rotating crystal and crucible, starting at roughly 400 to 1,000 gauss and decreasing the field as more of the melt solidifies. It targets oxygen below 15 ppma with a radial gradient under 5% and an axial gradient under 5–10%, and its worked example holds a constant 13 ppma along the crystal. The GlobalWafers patent states that cusp-field growth may yield ingots with oxygen no greater than about 15 ppma, or no greater than about 10 ppma.
Two cautions apply. These are patent claims and examples from semiconductor-oriented wafer makers, not audited production averages for solar lines. And ppma figures depend on the infrared calibration convention used to convert absorption into concentration, so oxygen limits in a contract should name the measurement method and conversion, not just a number. Equipment makers now present solar pullers as compatible with both standard Cz and MCZ growth, some quoting oxygen as low as 6 ppm; those are supplier claims to be verified on the buyer's own material.
- Read patent oxygen figures as achievable targets, not guaranteed outcomes
- Specify oxygen with a named method and calibration convention
- Ask for axial as well as radial oxygen profiles along the ingot
- Verify supplier oxygen claims with trial pulls on your own feedstock
Why oxygen control matters more as cells move on
Oxygen is not simply bad. The same 1993 patent notes that oxygen in solid solution increases the mechanical strength of silicon wafers and that oxygen-related defects can trap heavy-metal contaminants. But it also causes precipitates, dislocation loops and stacking faults. For solar wafers the balance has moved: as our guide on oxygen precipitates explains, thermal donors and precipitates can distort resistivity and cut carrier lifetime after the cell's high-temperature steps, and high-efficiency n-type architectures such as TOPCon and back-contact reward high-lifetime material.
Continuous Czochralski growth attacks the same problem from another side. A SunEdison paper in Photovoltaics International (2013) described CCz as giving much greater resistivity uniformity and allowing crystal oxygen to be selected independent of production batch size. MCz and CCz are complementary rather than competing: one controls the flow, the other the melt volume and history.
- Link oxygen targets to the customer's cell architecture
- Weigh oxygen's strength benefit against its lifetime cost
- Consider MCz and continuous Cz together when planning the hot zone
- Keep lots traceable to field settings, crucible and growth run
What an Indian ingot plant should decide before ordering pullers
With ALMM List-III bringing ingots and wafers into the approved-list framework from June 2028, Indian ingot plants being specified now will supply cell makers judging them on lifetime and yield. Whether to grow MCz is therefore an early procurement decision. A magnet changes the puller's footprint and utilities: the superconducting-magnet patent describes coils cooled to around 4 K, radiation shields at around 40 K and dedicated cryogenic refrigeration, along with weak-field zones where nearby equipment can be placed.
The practical questions are simple to list. Will the plant's customers ask for oxygen limits that standard Cz cannot hold reliably? Can the chosen puller accept a magnet later, and with what floor space, power and cooling? Who maintains the cryogenics? And how will the recipe that sets field, rotation and argon flow be developed and owned? Current ALMM requirements should always be confirmed directly with MNRE.
- Decide MCz-ready or not at puller specification, not after installation
- Plan floor space, power, cooling and stray-field zones for magnets
- Assign ownership of cryogenic maintenance and spares
- Develop and document the field-rotation-argon recipe as plant know-how
How JRST supports this requirement
JRST supports ingot and wafer projects in specifying crystal growth around the quality their cell customers need: linking puller configuration, crucible and hot-zone choices to oxygen behaviour, and planning procurement, utilities and acceptance testing. Figures on this page are drawn from public patents and published research, some of it historical or semiconductor-oriented, and should be validated for the specific project; crystal quality depends on equipment, recipe and operating conditions, and compliance requirements must be confirmed with the relevant authorities.
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 magnetic Czochralski (MCz) silicon?
Monocrystalline silicon grown by the Czochralski method with a magnetic field applied to the melt. The field damps or reshapes convection, which gives the grower more control over oxygen and dopant distribution in the ingot.
How does a magnetic field reduce oxygen in silicon?
Oxygen dissolves from the quartz crucible and is carried by melt convection either to the free surface, where it evaporates as silicon monoxide, or into the crystal. The magnetic field acts on that flow through the Lorentz force, changing how much oxygen reaches the growing crystal.
What is the difference between a cusp and a transverse magnetic field?
A transverse (horizontal) field crosses the crucible in one direction. A cusp field uses two coils with opposite currents, giving a field roughly symmetrical about the pulling axis. A GlobalWafers patent describes cusp fields as allowing oxygen control by flux density alone.
How low can oxygen go with MCz?
Patents describe cusp-field ingots with oxygen below about 15 ppma, or no greater than about 10 ppma, with a 1993 example holding 13 ppma along the crystal. These are patent figures, not production guarantees, and units depend on the measurement calibration.
Does MCz remove the need for good quartz crucibles?
No. The crucible remains the oxygen source. MCz changes how oxygen is transported, so crucible quality, hot-zone design, rotation and argon flow still matter.
What does an MCz puller need that a standard puller does not?
A magnet and its supporting systems. For superconducting magnets, a published patent describes coils cooled to around 4 K, shields at around 40 K and dedicated cryogenic refrigeration, plus planning for stray magnetic fields around the puller.
How can JRST help?
JRST can help ingot projects decide whether and where MCz fits, frame oxygen as a measured specification, and plan puller procurement, utilities and acceptance terms. Outcomes remain project-specific and must be validated.
Primary sources and further reading
Last reviewed 2026-10-05. Technical scope, policy eligibility, availability, and commercial terms should be independently confirmed for each project.