attributed to the semirigid nature of the cells and the less order- ly packing of the cells in the tissue (Haard 1997). Also, whole fruit freeze-cracking was observed during the freezing of both peach and mango with liquid nitrogen, resulting in irreversible macroscopic damage.
The freezing rate, as it is defined by the International Insti- tute of Refrigeration (IIR 1972), is slightly lower in high-pressure-shift freezing (HPSF) than in liquid nitrogen freezing despite the large difference in the temperatures of their refrigerating mediums ( Table 1). This apparent contradic- tion can be explained if we consider that the 2 processes pro-
ceed in very different ways. Cryogenic freezing is subjected to thermal gradients that initiate ice crystallization on the surface of the sample while the center of the fruit is still eliminating sensible heat in the pre-cooling phase. However, as it can be seen in Fig. 1, in HPSF process, the entire volume of the sample reaches the initial freezing point at the same time, just before the release of pressure. At that instant, theoretically, about 35% of the water in the sample is crystallized instantaneously, as Sanz and others (1997) pointed out. So, the time t0 (see Fig. 1)
from the moment that the surface reaches the initial freezing
point to the moment that the center reaches a temperature 10°C
below this initial freezing point is similar for both processes.
Comparison of local freezing rates at the sample center (characteristic times) showed that times for high-pressure-shift freezing method were almost double those for liquid nitrogen freezing. Characteristic time at the sample center in HPSF pro- cess is longer because when crystallization commences in the center of the fruit, the entire sample is at its phase transition temperature and only about 35% of it is frozen. In cryogenic freezing process, the phase transition at the center of the fruit takes place once most of the product is at a temperature consid- erably lower than the freezing point. So, in HPSF much more heat is necessary to extract with a poorer heat extraction capaci- ty than in the cryogenic freezing process (refrigeration mediums at –25°C and –196°C, respectively).
In HPSF, expansion was followed by uniform supercooling in
the sample, inducing pervasive nucleation of a large number of small intra- and extracellular crystals without this significantly