Glass transition and collapse temperature: the thermal numbers behind lyophilized peptide stability
A lyophilized peptide vial presents as a dry white cake, and the intuition most people carry to it is that a solid is a solid — inert, static, waiting. That intuition is wrong in a way that matters for storage. The cake is not a crystal with atoms locked into a lattice. It is an amorphous glass: a supercooled liquid whose molecular motion has been arrested but not eliminated. Whether that glass stays a stable storage matrix or slowly relaxes toward degradation is governed by two temperatures that never appear on the label — the collapse temperature, which controls what happens during freeze-drying, and the glass transition temperature, which controls what happens for the months or years afterward. Understanding these two numbers explains why lyophilized peptides are stored the way they are, and why a residual-moisture figure and a storage temperature are not independent variables.
Why the cake is a glass, not a crystal
When a peptide solution is frozen and the ice is removed by sublimation, the non-ice components — peptide, buffer salts, and any added sugar or bulking agent — are left behind in the interstitial spaces between former ice crystals. As water leaves, this residual phase becomes progressively more concentrated and more viscous. For most peptide formulations, especially those stabilized with disaccharides, this concentrated phase does not crystallize. It vitrifies: it becomes a rigid amorphous solid, structurally a liquid whose viscosity has climbed by more than ten orders of magnitude to the point where flow is effectively frozen on any observable timescale.
This distinction is the whole point. A crystalline solid stores molecules in a repeating lattice with essentially no translational mobility, but crystallization excludes the peptide from the protective matrix and often destabilizes it. An amorphous glass, by contrast, immobilizes the peptide within a rigid sugar network — the vitrification mechanism discussed in the work on lyoprotectants and bulking agents. The stability of that arrangement depends entirely on the matrix remaining a glass. The moment the matrix softens into a rubbery, mobile state, the molecular motions that drive aggregation, deamidation, and other degradation pathways accelerate by orders of magnitude.
The glass transition temperature and what it governs
The glass transition temperature, written Tg, is the temperature at which an amorphous solid changes between its rigid, glassy state and a softer, rubbery state. It is not a melting point. Nothing liquefies at Tg in the ordinary sense; instead, the material’s molecular mobility, heat capacity, and thermal expansion change slope over a narrow range. Below Tg, large-scale molecular motion is arrested and the matrix behaves as a rigid glass. Above Tg, cooperative motion resumes, viscosity drops steeply, and the once-immobilized peptide gains the translational and rotational freedom that chemical and physical degradation require.
The practical storage rule follows directly: a lyophilized peptide should be held at a temperature comfortably below the Tg of its dried matrix, with margin to spare. This is the physical reason behind the storage recommendations covered in the general storage guidance — the goal is to keep the product in the glassy state where molecular mobility is minimized.
The Tg of the dried solid depends heavily on composition. Pure amorphous sucrose has a dry glass transition in the neighborhood of 65 to 75 degrees Celsius; trehalose, the disaccharide often preferred for its robustness, sits considerably higher, characteristically around 115 to 120 degrees Celsius. These are the values for the essentially dry material. They matter because they set the ceiling — but that ceiling is not fixed, because water moves it.
Residual moisture is a plasticizer, and it lowers Tg
Water is the most effective plasticizer an amorphous pharmaceutical matrix encounters. A plasticizer is a small, mobile molecule that inserts itself between the larger matrix molecules, increasing free volume and lubricating motion. Because water has an extremely low glass transition temperature of its own — on the order of minus 135 degrees Celsius as an amorphous solid — even small amounts sharply depress the Tg of the mixture. The composition dependence is described reasonably well by mixing relationships such as the Gordon-Taylor equation, in which the blended Tg falls steeply as water fraction rises.
The magnitude is not subtle. For a sucrose-based matrix, each additional percent of residual moisture can lower Tg by roughly ten degrees Celsius in the low-moisture regime, though the exact slope is formulation-specific. A cake that measures 1 percent residual moisture and a cake that measures 4 percent are not two grades of the same product; they can have glass transition temperatures separated by tens of degrees, which changes the safe storage temperature entirely. This is why the residual-moisture measurement reported by Karl Fischer titration is not a cosmetic quality attribute. It is a direct input to the temperature at which the glass will soften, and therefore to how much thermal margin the product actually has in a given storage environment.
The interaction runs the other way as well. A matrix stored too close to its Tg has enough mobility to allow slow water redistribution and, in some formulations, gradual crystallization of a component that had been trapped in the amorphous state. Crystallization can expel the peptide from the protective glass and release previously immobilized water into the remaining amorphous phase, lowering its Tg further — a self-reinforcing loss of stability. The margin below Tg is therefore protecting against more than one failure mode at once.
Collapse temperature: the other number, set during drying
The glass transition governs storage, but a related temperature governs whether a usable cake forms in the first place. During primary drying, the frozen concentrate has its own glass transition, denoted Tg-prime, which is the glass transition of the maximally freeze-concentrated solution — the state reached when as much water as possible has frozen out as ice and the remaining unfrozen phase is at its most concentrated. Tg-prime is far below the dry Tg because that concentrated phase still holds substantial unfrozen water; for sucrose it is characteristically near minus 32 to minus 34 degrees Celsius, and for trehalose near minus 29 to minus 30 degrees Celsius.
Closely tied to Tg-prime is the collapse temperature, Tc, usually a few degrees above Tg-prime. During sublimation, the growing dried region is a fragile porous scaffold. If the product temperature during primary drying rises above Tc, the amorphous matrix has enough mobility to flow, and the porous structure collapses under its own surface tension. A collapsed cake is not merely cosmetically shrunken; collapse tends to trap residual moisture, slow subsequent drying, raise the final water content, and — through that moisture — lower the storage Tg of the finished product. The collapse temperature therefore sets the upper bound on how aggressively primary drying can be run, and a process held below Tc is part of why a well-made cake reaches a low, uniform residual moisture in the first place. Tg-prime and Tc are commonly characterized by differential scanning calorimetry and freeze-dry microscopy during formulation development, well before any storage question arises.
The two temperatures thus bracket the product’s life. Tc (near Tg-prime, deep below zero) is the constraint the process must respect while the ice is being removed; dry Tg (tens of degrees above zero for a good sugar matrix) is the constraint storage must respect for the years that follow. A poorly controlled drying step that violates the first ends up degrading the second, because collapse and elevated residual moisture both erode the storage glass transition.
Reading it back at the vial
None of these temperatures are printed on a research vial, but their fingerprints are. A cake that is elegant, uniform, and pharmaceutically sharp in appearance is consistent with a process that stayed below collapse; a shrunken, melted-looking, or shifted cake suggests the opposite and, by implication, a matrix that may carry more residual moisture and a lower Tg than intended. A low, well-controlled Karl Fischer moisture figure implies more thermal margin below the glass transition. And the standard practice of storing lyophilized peptides cold, and of not letting them sit through repeated warm temperature excursions, is precisely a strategy of keeping the product far below the temperature at which its glassy matrix would begin to move.
The larger point is that “keep it dry and keep it cold” is not folklore but a two-variable statement about a single physical state. Dryness raises the glass transition; cold keeps the storage temperature under it. The margin between the two — the gap between how warm the vial gets and how warm its amorphous matrix can get before it softens — is the quantity that actually protects the peptide. A lyophilized cake looks static, but it is a glass held some distance below its own softening point, and the entire storage strategy for these compounds is a matter of preserving that distance. See also the broader treatment in lyophilization chemistry.
Research use only. This post is for educational and reference purposes on peptide formulation and storage chemistry. It does not constitute medical, veterinary, or dosing guidance.