Classical cryonics has always assumed that an unfixed brain, once vitrified, preserves neural information in a readable form. This assumption has been repeated for decades, but never demonstrated. The physics of materials, the mechanics of fragile solids, and the informational scale of neural memory all show that this assumption is untenable.
The central point is simple: long-term memory (LTM) lives at the nanometer scale (5–50 nm), while biological glass fractures and relaxes at that same nanometer scale. The two things are incompatible. This is not a technological limitation. It is a structural limitation of the state of matter.
LTM is not a “diffuse pattern” that can be statistically reconstructed. It is a localized physical structure, encoded in:
- spine geometry
- PSD organization
- receptor clusters
- local dendritic micro-architecture
- vesicle distribution
- protein nanodomains
Why biological glass fractures
An unfixed brain, when vitrified, contains four distinct forms of free or quasi-free water, all unavoidable:
- Extracellular water left over from perfusion, not fully replaced by CPA.
- Intracellular water not replaced, because membranes are not fully permeable to CPA.
- Free water intrinsic to the CPA mixture - the unbound fraction that contributes to glass fragility.
- Micro-ice nuclei formed near Tg due to micro-heterogeneities in the mixture, remaining latent during cooling but becoming destructive nucleation points during warming.
But it is essential to clarify that cracking does not depend on free water. Even if it were possible to eliminate it completely, an unfixed vitrified brain would still fracture. Both free water and fragile glass are consequences of the absence of fixation.
Unfixed biological tissue is a heterogeneous material and forms a fragile glass because its components - lipids, proteins, membranes, cytoskeleton, extracellular matrix - have different contraction coefficients and no plasticity. During the glass transition these components contract non-uniformly, generating internal stresses that a fragile solid cannot relax except by fracturing. Free water worsens the problem, but eliminating it would not eliminate glass fragility, which is intrinsic to the nature of unfixed material.
The result is a fragile, heterogeneous solid with no plasticity. A fragile glass cannot flow, cannot deform plastically, cannot redistribute stress. It can only release it through cracking. And cracking is not a single event: it is a field of micro-fractures at the nanometer scale.
This scale coincides exactly with that of neural information.
We must make a distinction.
- Cooling-phase cracking can be minimally destructive, because it is a static phenomenon: the material is already rigid, has no molecular mobility, does not flow, does not deform. The fracture creates an interface but does not drag, distort, or mix. In this elastic regime the fracture can even be nearly planar, because the material is immobile and the crack front remains stable. For this reason, in my previous long post, the inferibility loss associated with cooling-phase cracking can be modeled as below 5%, and in better perfusions the residual ice as even less impactful. In other words: cooling-phase cracking breaks, but does not reorganize.
- Warming-phase cracking, however, is catastrophic. To read an unfixed vitrified brain, it must be warmed at least enough to allow slicing or perforation. But warming crosses the glass transition zone, where:
- the material acquires partial mobility
- internal stresses release chaotically
- membranes begin to soften
- proteins begin to move
- molecular clusters begin to flow
For this reason, even if cooling-phase cracking was minimal, warming-phase cracking drives LTM inferibility close to zero.
I have considered six types of physical loopholes, quite different from each other, that have been imagined by the cryonics community as possible ways to recover — or at least read — in the future the information required for long-term memory (and therefore, by extension, individual identity) from a vitrified brain.
1. The most common loophole is: “What if the glass becomes fluid without cracking?”
The idea is intuitive: “maybe there are conditions under which a biological glass becomes more fluid without fracturing, allowing controlled warming.”
It sounds plausible, because CPAs modify Tg and biological glasses are more complex than silicate glasses. But here physics is unforgiving: what would be required is internally contradictory.
- To avoid cracking, the glass must acquire molecular mobility.
- To preserve memory, the glass must remain immobile at the nanometer scale.
In other words, the physical condition required to avoid fracture is the same condition that erases memory. This is not an empirical limit: it is a logical limit. It is not a matter of “we don’t know everything yet.” It is a matter of logical incompatibility between what is required to avoid fracture and what is required to preserve memory.
Therefore, if the glass acquires the mobility needed to avoid cracking, it must cross Tg; and crossing Tg implies dynamic cracking and unavoidable informational loss. No second physically coherent path exists.
2. The next loophole becomes: “What if we use EM or laser heating matrices?”
One might imagine selective heating using electromagnetic waves or lasers, like a “precision microwave.” But this solves nothing:
- A biological glass does not absorb microwaves in any useful way because it has no mobile dipoles, no free ions, no liquid water, and no dielectric friction.
- It cannot convert electromagnetic energy into controlled heat: absorption occurs only in specific bands and in a non-uniform way, regardless of the number or geometry of beams. Even crossing a hundred laser beams does not create uniform heating: it heats only where the glass absorbs, not where you want to heat. This non-uniformity generates unavoidable thermal gradients, and any thermal gradient increases cracking.
3. But the alternative loophole most often proposed is: “I won’t touch it. I’ll read it from the outside, like with X-rays, but with future technology.”
The problem is not resolution, but the mathematics of the ill-posed inverse problem (see e.g. Bednarski & Roith, 2025) applied to opaque materials (see e.g. Vellekoop et al., 2009).
The idea of reading a vitrified brain from the outside — using X-rays, gamma rays, holography, photon scattering, electron scattering, or any particle — seems intuitively plausible: after all, we are used to “seeing inside” the human body with CT, PET, and radiology, and we tend to think that future technologies could simply increase resolution. But this analogy is misleading: these techniques work only because the body is transparent to X-rays and gamma rays, whereas a vitrified brain is an opaque volume, not a transparent or “less transparent” medium. And an opaque volume cannot be treated as a transparent material by using more energetic radiation. This is not a matter of power, sensitivity, or “future technology”: it is a matter of propagation physics and reconstruction mathematics.
In an opaque material, propagation is not coherent: after a few microns, multiple scattering destroys phase and direction, absorption eliminates deep paths, and the outgoing signal is no longer an invertible transformation of the internal structure.
And even if it were transparent — which it is not — it would not be enough. Even in perfectly transparent materials, it is impossible to reconstruct a macroscopic object at 5–50 nm from the outside: diffraction limits resolution, and no external technique can traverse centimeters of material while preserving nanometric information. Nanometric resolution requires physical conditions — controlled access geometries, coherent paths, thin samples — that cannot be satisfied by a beam crossing centimeters of material. The question “what if it were transparent?” is already physically ill-posed.
In the real case, reconstructing an opaque 10 cm3 volume at 5–50 nm from external signals is an unstable inverse problem. Reconstruction fails for deep physical and mathematical reasons:
- multiple scattering
- absorption
- phase loss
- noise
- non-periodic structure
- no informational redundancy
- no coherent propagation
4. A fallback loophole might be: “Then I’ll perforate the glass with probes or nanobots, entering carefully to observe the inside.”
But a fragile glass cannot be perforated without damage. Every hole creates walls that concentrate stress: the probe tip generates a stress field that exceeds the local strength of the glass, producing radial micro-cracking. An unfixed biological glass cannot be:
- Locally heated: It would cross Tg in a destructive way.
- Locally resin-impregnated, as in electron microscopy preparation to make ultrastructural features readable: The technique itself requires molecular mobility, because the resin must diffuse to impregnate the tissue, and diffusion is physically incompatible with the glassy state.
- Locally read to decide to dig only where no neurons or synapses are nearby: All non-destructive reading techniques discussed in point 3 introduce an interaction that is:
- either destructive (particles, ions, atoms, neutrons)
- or insufficient to generate distinguishable scattering beyond ~50 nm depth
- penetration depth
- local resolution
- damage
Variants imagine EM fields capable of slowing particles after entry so they can nest inside the structure. It is true that the dipole idea would work if introduced before vitrification (as shown by Manuchehrabadi et al., 2017), but after vitrification no known particle can do this: not ions, not neutral atoms, not protons, not electrons, and — counterintuitively — not even neutrons can traverse a glassy material, biological or inorganic, without causing damage:
- If the energy is too low, they bounce or stop at the surface.
- If it is high enough to penetrate, they produce atomic displacements, ionization, micro-cracking, or nuclear transmutations.
Consider, for example, an ion beam intended to nest particles inside the glass — perhaps to form dipoles — and to infer the previous state of the solid from the information lost during their passage, compensating for the collateral “ballistic damage”. Even knowing mass, charge, and trajectory, their traversal does not record information: it erases it. Ion tracks are cylinders of disordered material, with broken bonds, vacancies, and displaced atoms: they are not useful imprints for inference, but regions where the original structure has been eliminated. Ballistic damage is a dissipative, physically non-invertible process dominated by microscopic chaos, and it generates informational collapse: information is not moved elsewhere; it is destroyed.
For this reason, no retroactive technology can introduce dipoles or internal agents into an already vitrified brain.
6. The last loophole, a residue of the previous one, is: “Then I’ll treat some internal molecules as heatable dipoles, even resorting to atomic spins.”
After all, one might think, food in a microwave does not necessarily contain metallic dipoles, but only water with residual free ions. In reality, it is these residual free ions and the weak dipoles intrinsically formed by liquid water molecules that allow heating, whereas a biological glass:
- has no mobile dipoles
- has no free ions
- has no liquid water
- has molecules that cannot rotate
- has astronomical viscosity (10^12–10^15 Pa·s)
- has no dielectric friction
- does not absorb microwaves or other EM radiation in any useful way, because it lacks any physical coupling mechanism
- Achieving ionization would require electric fields on the order of 10^8–10^9 V/m, well above the dielectric breakdown threshold of biological glass. Such fields produce micro-arcs, conductive channels, and immediate structural damage.
- Any ions created would be immobilized in a solid with viscosity 10^12–10^15 Pa·s and could not oscillate or rotate.
- Without oscillation, no dipole heating is possible.
- Ionization in a fragile solid produces bond breaking, vacancies, atomic displacements, and micro-cracking — the same “ballistic damage” described in loophole 5.
- The process destroys information; it does not create usable dipoles
Therefore, no physical mechanism exists to treat internal molecules as heatable dipoles.
Conclusion: the loophole does not exist
The final question is: “Is it physically demonstrable that no unknown properties of biological glass could avoid cracking without destroying memory?”
Yes. Because what would be required is internally contradictory:
- To avoid cracking, you need properties that imply molecular mobility.
- To preserve memory, you need properties that imply molecular immobility.
An unfixed, then vitrified brain is a physically unreadable support for long-term memory: not because of lack of technology, but because of the incompatibility between the glassy state and the nanometric informational scale in which identity lives.
If the brain had been fixed, then yes: it would be possible to read it even with a standard electron microscope, because fixation prevents fragile glass formation, eliminates nucleation, prevents cracking (aside from <1% localized microcracking under very rough heating ramps), and allows controlled slicing and, in the future, even liquid states for passive nanobots.
But classical cryonics chose a completely different path: no fixation → free water + fragile glass → internal stresses → cracking. It was a historical choice, not a physical one.
This does not mean that the people who chose classical cryonics in the past are beyond hope. It means only that the glassy state cannot be read as a physical memory substrate. Identity, however, may turn out to be reconstructible through future approaches — statistical, biological, or computational — that do not depend on reading vitrified tissue. The future of identity preservation may lie not in the glass, but in methods we have not yet invented.