Secondary Drying: The Removal of Unfrozen Water - Freeze-drying of Pharmaceuticals and Biopharmaceuticals - page 118

Figure 3 The effect of secondary drying temperature on the aggregation of human serum
albumin (HAS). The solution, containing 8% protein and 8% ethanol, was
freeze dried at a primary drying temperature below 301C; secondary drying
was performed at the temperatures indicated. Reproduced from Pikal (unpub-
lished data).
potentially also aggregation and/or chemical degradation, will inevitably
occur. This is graphically illustrated in Figure 3, which shows how
product quality can be significantly and usually adversely affected by
temperature abuse during secondary drying.
The histogram in Figure 3 shows that, even before the start of freeze-
drying, 10% of the product was already present in its dimerised form,
possibly due to some ''rough'' handling during the purification process.
There was no indication of any more advanced aggregation at this stage.
The product would have contained at least 20% of water, which would
give it an estimated T g of ca. -251C. Even with secondary drying
performed at 101C, considered by many to be a low temperature,
HSA had become prone to aggregation, to a degree that would render
it quite unsuitable for use in parenteral products. As expected, aggre-
gation rapidly increased with rising temperature.
Surprisingly, even modern literature 13 carries statements such as:
''During secondary drying, the product temperature has to be raised
to the maximum tolerable temperature of the dried product. This raise
(sic) can be done as quickly as can be technically achieved in the plant.
Raising it more slowly does not make the product safer. When there is
no more ice in the product, the final temperature can be applied.'' This
advice is dangerously misleading as can readily be concluded from an
inspection of Figure 3.