
the temperature in the frozen core during the primary
drying phase is below the melting point but above the
glass transition temperature (T
g
0
), the structural
damage may be significant, and the product may
collapse when the frozen ice crystal is sublimated. A
low temperature (under T
g
0
) maintains a high appar-
ent viscosity; shrinkage and, consequently, collapse
are avoided. The product-collapse temperature is
closely related to the glass transition temperature
but sometimes appears higher than T
g
, especially for
cellular material (biological tissue). For foods with a
high sugar content, the collapse temperature is low
(40 to 60
C) and can be raised by the addition of
high-molecular-weight additives (polysaccharides).
0011 The glass transition temperature decreases strongly
with the water content, due to the plasticizing effect
of water (Figure 2). In the dried layer, a higher
temperature, allowing a higher drying rate, can be
applied without product damage. But over a critical
value of 60
C, another kind of damage occurs in the
biological product, i.e., protein degradation. This
results in a change in the product structure, especially
in protein-based foods (meat, fish), and a reduced
tenderness and hydration capacity.
0012 During the freeze-drying process, the passage of the
product from the frozen state to the dried state is a
critical step in which there is a wide variation in water
content as well as the temperature. The probability of
structural changes (i.e., collapse) is important. As a
general rule, to preserve the original structure, the
product temperature should remain under the glass
transition temperature (T
g
), which depends strongly
on the water content.
0013 On a practical point of view, the control of tem-
perature is based on the temperature and energy sup-
plied from the heating source. Among the traditional
techniques, conductive heating is more preferable for
use with thermally sensitive products than a radiation
heat source. With microwave heating, it is more diffi-
cult to control a homogeneous supply of energy inside
the frozen core of the product, and this often leads to
a reduction in final product quality.
Pressure
0014 In the atmosphere of the freeze-drier under vacuum,
the total pressure is generally assumed to be close
to the vapor pressure. Following the equilibrium
rule, the vapor pressure is directly related to the tem-
perature at the sublimation front inside the product
where the vapor is created. Thus, a control of the
total pressure is an indirect way to monitor the prod-
uct temperature at the sublimation front. At first, the
total pressure has to be sufficiently low to avoid the
melting of ice. After a rapid abnormal increase in
pressure in the freeze-drier, the food product foams
and shrinks due to an increase in front temperature
above the melting point. Furthermore, during pri-
mary drying, if the pressure is sufficiently low to
avoid the melting point but higher than the glass
transition, shrinkage and collapse mechanisms can
be observed. In order to maintain the quality of
the product, the total pressure should be sufficiently
low in comparison with the sublimation front
temperature.
0015In conclusion, the main modification of the struc-
ture of the product is a consequence of the freezing
step, assuming that the drying steps are well con-
trolled (no recrystallization and no collapse). During
the subsequent drying steps, low values of pressure
and temperature (levels depending on the nature of
the dry matter of the product) tend to preserve the
shape and texture of the frozen product. Knowing that
an increase in temperature is a good way to reduce the
freeze-drying time, an optimum temperature has to be
defined for industrial applications.
Flavor Losses
0016Freeze-drying is often considered the best drying pro-
cess as far as flavor compound retention is concerned,
due in a first approximation to the low temperature
level.
0017Molecules, responsible for food flavor, usually
have a high relative volatility to water and a low
molecular weight (less than 300 g mol
1
). Owing to
the low concentration of aroma compounds in the
food products (p.p.b. to p.p.m.), food flavor is usually
evaluated using two kinds of tools: analytical meas-
urements (gas chromatography and mass spectrom-
etry analysis) and/or sensory panel. From these
properties, it is difficult to imagine removing water
as a vapor phase, without any other volatiles like
flavor compounds. Also, flavor losses, even in low
proportions, are rapidly considered to be significant.
As a matter of fact, the loss of flavor compounds can
be limited during freeze-drying with regard to water
elimination. The freeze-drying operating conditions
that improve the drying rates give an increased reten-
tion of volatiles. Rapid primary drying is expected
to result in higher levels of volatiles. This can be
explained from the principle of selective diffusion.
The diffusion coefficient (mobility) of the volatile
decreases more rapidly with the water content inside
the product than the diffusion coefficient of water
itself (Figure 3). The loss of volatile decreases and
stops as soon as a critical moisture is attained. One
way to decrease the first drying duration is to
increase, in a preliminary step, the dry matter (addi-
tives, osmotic dehydration). Based on the techniques
of encapsulation, additives can also decrease the
FREEZE-DRYING/Structural and Flavor (Flavour) Changes 2703