After cryoprotective perfusion, a patient is not simply placed into liquid nitrogen.
The patient follows a computer-controlled temperature programme lasting several days. Sensors inside the patient show whether different anatomical regions are cooling together or separating into dangerous thermal gradients.
The engineering problem changes during the descent. Early cooling should limit ice formation. Near and below the glass transition, precision and thermal uniformity become more important than speed.
Why controlled cooling is necessary
Cryoprotective agents reduce the cooling rate needed to avoid ice. They allow water-rich tissue to enter a highly viscous, glass-like state instead of crystallising extensively.
The relevant glass-transition temperature, written Tg, depends on the solution, its concentration and the measurement method. It is a range, not one universal biological threshold.
For solutions relevant to human cryopreservation, this region is commonly near -120°C to -135°C.
Above that region, cooling too slowly can permit ice nucleation and growth. Below it, the material behaves increasingly like a brittle solid.
If the surface cools much faster than the centre, different regions contract by different amounts. The resulting thermomechanical stress can produce cracks.
This creates a constrained path: cool rapidly enough before vitrification, then control gradients and cooling rate as the patient becomes glass-like.
See the difference between freezing and vitrification for the underlying phase behaviour.
The first-generation cooling box
Tomorrow.bio’s first-generation system, called the V1 cooling box, established the basic architecture for controlled whole-body cooldown.
The patient is placed inside a dedicated insulated chamber. A computer controls the chamber’s cryogenic environment while thermocouples measure temperatures at multiple locations.
The controller follows a programmed curve rather than applying maximum cooling continuously. Every sensor channel is recorded as part of the patient’s thermal history.
This is already fundamentally different from direct immersion. The machine can slow, hold or change the chamber conditions in response to what the patient is actually doing.
The second-generation dewar-based system
Tomorrow.bio’s second-generation design moves the controlled cooldown into a purpose-built dewar.
The patient remains inside the dewar while a specialized lid integrates the cooldown hardware, computer controls, valves, sensor connections and thermocouple feedthroughs.
Thermocouples are positioned at several locations inside the patient. Additional sensors monitor the thermal environment inside the dewar.
The system therefore sees more than the requested setpoint. It sees how quickly different measured regions of the patient are responding.
The computer adjusts the dewar’s temperature gradient and rate of descent to keep those readings within the protocol’s permitted envelope.
A large body cannot be perfectly isothermal during cooling. The practical objective is to minimize the spread and prevent any measured region from running dangerously ahead or behind.
The dewar geometry also reduces handling between controlled cooldown and cryogenic storage. The same enclosed form can provide a stable, symmetric environment around the patient.
Setpoint temperature is not patient temperature
A controller can make the environment cold quickly. It cannot make the centre of a human body instantly match that temperature.
Heat must travel from deeper tissue toward the surface. Geometry, body composition, cryoprotectant distribution and sensor position all affect the observed lag.
This is why one chamber sensor is insufficient. A nominal -130°C environment does not prove that every measured part of the patient has reached -130°C.
The control system must consider both absolute temperature and the difference between channels. The slowest point, fastest point and overall spread each carry information.
Thermocouple calibration, placement and continuity also matter. Implausible readings or sensor failures should be flagged rather than silently averaged into the curve.
Tomorrow.bio’s three-stage temperature programme
The current protocol has three distinct phases. The values are operating targets for this equipment and cryoprotective approach, not universal constants for every vitrification system.
1. Descend rapidly toward -130°C
The first phase moves the patient from dry-ice temperature toward approximately -130°C as quickly as the measured temperature spread safely permits.
Speed matters here because underprotected regions can still form ice before the tissue becomes glass-like.
The computer does not follow elapsed time alone. It monitors the thermocouples and adjusts the cooling environment around the patient.
2. Allow a controlled rebound toward -120°C
After reaching approximately -130°C, active cooling is reduced and the patient is allowed to warm naturally toward approximately -120°C.
This is an annealing phase. Near Tg, the cryoprotected material retains enough molecular mobility for some accumulated mechanical stress to relax.
The rebound also gives warmer and colder regions time to move closer together before the material is taken deeper into its rigid glassy state.
Annealing does not reverse pre-existing biological damage. Its purpose is narrower: reduce thermal gradients and the stresses created by unequal contraction.
3. Descend to -196°C at approximately 1°C per hour
From about -120°C, the target descent is approximately 1°C per hour until the patient reaches liquid-nitrogen temperature.
That final 76-degree interval takes roughly three days even before holds and control adjustments are counted.
Maintaining 1°C per hour in a large, thermally heterogeneous body is a control problem, not merely a timer setting.
The controller must continually adjust the dewar environment so the internal thermocouples descend together without creating excessive gradients.
When one region lags, forcing the environment colder may overcool another region. Precision means balancing the entire sensor field, not chasing one reading.
Why the annealing phase is scientifically plausible
Large-scale vitrification research identifies two competing failure zones: insufficient cooling above Tg and excessive mechanical stress around and below Tg.
A 2024 human-organ-scale vitrification study used multiple internal probes and concluded that annealing reduces gradients before the material enters the glassy phase.
The same study found that slow cooling below the annealing region helps avoid rebuilding large gradients and stress.
Separate thermomechanical research found that annealing near Tg substantially changes residual stress and weakens its dependence on the preceding cooling rate.
These studies support the underlying control strategy. They do not independently validate every exact setpoint in Tomorrow.bio’s whole-body protocol.
Whole human bodies are larger and more heterogeneous than laboratory solutions or organs. Direct temperature records and post-cooldown imaging remain necessary.
What the computer records
The useful output is not a message saying “cooldown complete”. It is the complete time series from every temperature channel.
The record should show rates, holds, rebounds, gradients between sensors, controller actions, alarms and any deviation from the planned curve.
Those data become part of the cryopreservation quality assessment.
After cooldown, standardized CT at liquid-nitrogen temperature can look for cryoprotectant distribution, ice, gross deformation and fractures that temperature data alone cannot reveal.
Why finish at liquid-nitrogen temperature
At normal atmospheric pressure, nitrogen boils near 77.34 kelvin, approximately -195.8°C, according to the NIST Chemistry WebBook.
Liquid-nitrogen storage is passive. As long as sufficient liquid remains in the vessel, its boiling equilibrium maintains a stable cryogenic environment without mechanical refrigeration.
The controlled descent ends only when the patient is thermally ready for that stable state. Storage is not used as a substitute for cooldown.
The design and trade-offs of the next stage are covered in modern cryogenic dewars and intermediate-temperature storage.
TL;DR: Tomorrow.bio uses computer-controlled cooling and internal temperature sensors. The process cools rapidly near glass formation, relieves thermal stress, then descends slowly toward -196°C.
Take the cryopreservation guide with you
Get a practical guide to the procedure, its limits and the decisions involved.
Loading the interactive tool...
Further reading