Cells are cryopreserved without damage by replacing much of their intracellular water with penetrating cryoprotective agents, cooling them fast enough to avoid ice crystal growth, and storing them below the glass transition temperature of roughly −130 °C, where molecular motion effectively stops. Warming must be equally fast, because recrystallization during thaw can destroy cells that survived cooling.
That is the one-sentence answer. The rest of this explainer is about why each step works, where the margins sit, and what separates a viable protocol from a failed one.
The physics problem: ice is the enemy, not cold
Cold itself is not what kills cells. Ice is. When a cell suspension cools slowly, extracellular water freezes first, pulling pure water out of solution as ice. The remaining extracellular fluid becomes hypertonic, and water flows out of the cell osmotically. If cooling is too slow, cells shrink and dehydrate past tolerable limits. If cooling is too fast, water remains inside the cell and freezes there, and intracellular ice nucleates and expands.
That tension defines the classic "two-factor hypothesis" of cryoinjury: slow cooling causes solution effects, fast cooling causes intracellular ice. The optimum sits in a window that depends on the cell type and the cryoprotectant concentration. For conventional slow freezing of mammalian cells in dimethyl sulfoxide, that window is typically around 1 °C per minute, though the exact rate is empirically tuned per cell type.
Vitrification sidesteps the window entirely. Instead of letting ice form and managing its consequences, vitrification suppresses nucleation and converts the whole sample into an amorphous glass. The word "glass" is literal: the liquid becomes so viscous that molecules stop rearranging into a crystal lattice before they can.
Why cryoprotectants work: colligative and kinetic effects
Cryoprotective agents (CPAs) fall into two functional classes. Penetrating agents enter the cell and act from inside. Dimethyl sulfoxide, ethylene glycol, propylene glycol, and glycerol are the common ones. Non-penetrating agents stay outside and act osmotically. Sucrose and trehalose are typical examples.
Penetrating CPAs depress the freezing point and raise the viscosity of the intracellular solution. The kinetic effect matters more than the colligative one: as CPA concentration rises, the probability that a water molecule joins an ice lattice falls faster than the probability that the solution simply becomes rigid. At sufficient concentration, cooling produces a glass rather than a crystal.
The concentrations required are high, which is the central practical problem. Vitrification of oocytes and embryos typically uses total CPA concentrations in the range of 5 to 8 molar, delivered in a stepwise equilibration rather than a single exposure. That is not a gentle chemical environment. CPAs are cytotoxic at these concentrations, and the exposure time and temperature are tightly controlled to limit damage.
Non-penetrating agents do a different job. By raising extracellular osmolality, they draw water out of the cell before the CPA load is applied and reduce the total water that must be vitrified. Sucrose is the workhorse here.
The quantification: cooling rates and the glass transition
The numbers clarify the difference between the two approaches.
Slow freezing operates at roughly 1 °C per minute, often with a controlled-rate freezer and a seeding step to initiate ice formation outside the cell at a controlled temperature. Samples are then plunged into liquid nitrogen at −196 °C, or stored in the vapor phase at roughly −150 °C to −190 °C.
Vitrification operates at cooling rates three to four orders of magnitude higher, on the order of hundreds to thousands of degrees Celsius per minute or more, depending on the device. A small sample on a thin carrier can cool far faster than a plastic cryovial.
The glass transition temperature of the vitrified solution is the relevant storage threshold. For many CPA mixtures it sits near −120 °C to −130 °C. Below that temperature the sample is glassy and stable in the short term. Above it, even at −80 °C, molecular mobility is sufficient for ice nuclei to grow slowly over time, which is why mechanical freezers at −80 °C are for short-term holding, not long-term banking.
Storage in liquid nitrogen vapor is the standard for long-term banking, and the reason is that no significant ice growth occurs below the glass transition.
The time course of a vitrification cycle
| Stage | Typical temperature or condition | Duration | Primary risk |
|---|---|---|---|
| Pre-equilibration | Room temperature or 4 °C | 5–15 minutes | Osmotic shock if CPA ramped too fast |
| CPA equilibration | Room temperature or 4 °C | 1–10 minutes | Chemical toxicity of high CPA load |
| Loading onto carrier | Room temperature | Seconds to minutes | Evaporation and concentration drift |
| Cooling | Plunge into liquid nitrogen | Sub-second to seconds | Ice nucleation if cooling too slow |
| Storage | −150 °C to −196 °C | Months to decades | Warming above glass transition in transit |
| Warming | 37 °C bath or warmed medium | Seconds | Recrystallization if warming too slow |
| CPA removal | Dilution series, typically with sucrose | 3–10 minutes | Osmotic shock if diluent too abrupt |
The table compresses what is in practice a carefully rehearsed sequence. The point of the sequence is that every stage has a failure mode, and the failure modes are not independent. A cell that survives cooling can still die during warming.
Counterintuitive detail: warming is often the harder problem
Most explanations of cryopreservation treat cooling as the difficult step. In vitrification, warming deserves equal or greater attention. A vitrified sample that is warmed slowly will pass through the temperature range where ice nucleation and crystal growth are kinetically favorable, and small ice nuclei that formed during cooling can grow into damaging crystals. This is called devitrification, and it is the primary reason vitrification protocols insist on rapid warming, not just rapid cooling.
In practical terms, a straw or carrier pulled from liquid nitrogen is plunged directly into a 37 °C bath rather than allowed to sit at room temperature. The seconds matter. A sample left to warm passively on the bench is a sample at elevated risk.
The same logic explains another non-obvious point. Vitrification is not a single physical state but a kinetic achievement. A solution that is glassy at −196 °C can become unstable if the sample is warmed to −80 °C and held there. The glass transition is not a sharp biological safe line; it is a threshold below which the relevant rates become negligibly slow on a human timescale.
Common misconceptions
The first misconception is that cryopreservation freezes cells. Some protocols do freeze, particularly conventional slow freezing for cell lines. Vitrification does not freeze; it avoids freezing entirely. The distinction is not semantic. It determines whether you worry about crystal size or about glass stability.
The second is that cryoprotectants protect the cell from cold. They largely protect the cell from ice by altering the solution's phase behavior. The cold itself is not the primary insult.
The third is that slower cooling is gentler. Slower cooling is gentler on one axis, solution effects, and harsher on another, intracellular ice in slow-freezing protocols. The correct rate is a compromise, not a direction.
The fourth is that long-term storage below −130 °C is risk-free. Storage is stable in the sense that no significant crystal growth occurs, but samples are still vulnerable to warming during transport, inventory handling, or equipment failure. The failure mode is not gradual degradation; it is a thermal excursion that may occur in minutes.
Practical actions and thresholds
If you are working with a cryopreservation protocol, the concrete steps that matter most are the ones at the margins of the protocol.
Verify your storage temperature with a calibrated probe, not with a display. Vapor-phase nitrogen is typically −150 °C to −190 °C, and a partially filled dewar can have warmer zones near the top. Control your warming rate explicitly. Any sample held for more than a few seconds at ambient temperature before dilution is a sample whose protocol has effectively changed. Pre-chill and pre-warm all media and carriers before use; a cold carrier changes the loading and cooling behavior of the sample. Use a seeding step if your protocol calls for it, so ice forms at a controlled site and temperature rather than spontaneously across the sample. Validate by endpoint, not by appearance. Post-thaw viability and functional assays are the only meaningful measure of whether a protocol is working.
A word of caution: a protocol that works for one cell type can fail for another. Oocytes, sperm, hepatocytes, and hematopoietic stem cells have very different membrane permeabilities to CPAs, different surface-to-volume ratios, and different tolerances for osmotic and chemical stress. Do not assume the parameters transfer.
Where the limits are
Vitrification is not a universal solution. The fundamental constraint is the thermal mass of the sample. The larger the volume, the lower the achievable cooling rate, and the higher the CPA concentration required to achieve a glass at that rate. Those requirements scale in opposite directions. This is why most successful vitrification is done on very small samples, on carriers designed to minimize volume.
The second limit is chemical toxicity. The high CPA concentration that enables vitrification is itself a stressor, and the tolerable exposure window narrows with cell type and with the presence of other stresses. That is the central trade-off in protocol design.
The third limit is that the mechanisms described here are physical and fairly stable, but clinical recommendations, storage prices, and approval statuses are not. Those change and should be checked against current authoritative sources rather than generalized from explainers.
There is also a measurement problem. "Viability" after thaw is often assessed by membrane integrity dyes, which measure one form of damage and miss others. A cell with an intact membrane but damaged mitochondria or chromosomal material can look viable by a simple assay. The literature on post-thaw outcomes is full of these discrepancies, and a good protocol is validated against functional endpoints, not only dye exclusion.
The bottom line is that how cells are cryopreserved without damage is a controlled negotiation with kinetics: depress the freezing point with permeating CPAs, suppress ice nucleation, cool into a glass below the transition temperature, and warm fast enough to avoid recrystallization. Done correctly, the cell spends the storage period in a state where nothing moves on a biological timescale. Done with the wrong margin at any stage, the damage can be subtle, delayed, and only visible after thaw.