Microscopic observations of samples of malto- structure of the freeze-dried maltodextrin cake was
dextrin which had been freeze-dried from solutions observed by optical sectioning in the light micro-
containing organic volatile revealed that, in many scope and with the scanning electron microscope.
cases, the dry maltodextrin contained entrapped The structure consisted of numerous intersecting
droplets of the liquid volatile. The amount of plates which had the retained volatile within the
volatile retained as measured by gas chromato- plate. Droplets of similar appearance were ob-
graphic analysis compared favorably with that served within the solids of freeze-dried coffee
calculated from microscopic observations. The samples.
he retention of organic volatiles in freeze-dried food Co., Holte, Denmark) were prepared. To 100 ml of the
Figure 1. Scanning electron micrograph of freeze-dried 207, Figure 3. Hexanal droplets in a freeze-dried maltodextrin and hex-
maltodextrin solution (ZOSX) analsystem(400X)
- SI Many 1.535
infinitely soluble; V very soluble; S soluble; SI slightly soluble. c As
Figure 4. Hexanal droplets in a freeze-dried maltodextrin illid Fieure 5. Scannine electron miaoeraoh of freeze-dried malta-
hexanal system; this sample was frozen more rapidly and crush,:d dextrin and hexanarsystem (note holes i n broken platelet surfaces)
finer than that in Figure 3 (400X) (266X)
The hexanal retained after freeze-drying was determint example, if the droplets are below the resolving power of the
by gas chromatography to be about 1.3 g of hexanal per 1( microscope or if the volatile is actually dissolved in the malto-
g of maltodextrin. The hexanal content was also calculate dextrin), the measured refractive index of the mixture will
by using an average droplet size, known hexanal densify, an decrease due to the presence of the volatile with its much lower
by counting the number of droplets on an easily measurab refractive index. With acetone, for example, no droplets
platelet located on the center of the left edge of Figure 4. E were observed; the weight-fraction average refractive index
assuming the platelet thickness t o be three times the averag was calculated to be 1.529, compared to the measured value
droplet diameter, which seems reasonable based on scannir of 1.530. Because all the very soluble volatiles have similar
electron microscopic and optical sectioning observation refractive indices and retentions, the largest maltodextrin
we calculated the hexanal retention to be 1.56 g of hexan; refractive index for all these volatiles was 1.530. In further
per 100 g of maltodextrin, in good agreement with the me: observations with 1-propanol at lOOOX, small droplets were
sured value. barely discernible in the freeze-dried material, indicating that
Other Volaffles in Maltodextrin. The other volatili:S perhaps the apparent absence of droplets noted for I-propanol
investigated are listed in Table I, together with pertinetIt in Table I is due to limitations of the resolving power of the
information and results. Sizable amounts of each volati:le microscope (at 400X). From the above discussion it seems
were retained by the maltodextrin. It should be note:d obvious that as the volatiles are less soluble, there is more
that the droplets observed microscopically have not been droplet formation and these droplets tend to be somewhat
proven to be the particular volatile added, but are assumed larger, meaning that there is a higher percentage of them re-
to be so on the basis of their similarity in appearance with thle solvable in the microscope, and the maltodextrin refractive
hexanal samples. Of interest is the relationship between thle index is higher.
number of observable droplets, the volatile solubility, an d These observations regarding solubility and droplet size
II_^
LLLC
I""..--'
IdlSCW
...^l.^_l^__ll:_--c.--A:..-
I'IIIII.""~*L,III Ic,,aGL1"=
>_1^__
II1"C.X.
.:'c?L,lrllply
-*. arared, can be related to factors affecting the formation of the drop-
the less soluble volatiles show many droplets and none or only lets. The highly insoluble volatile compounds are present
smdl decreases i n the maltodextrin refractive index. As the in the initial solution as small drops, and remain so as the
volatile solubility increases, fewer droplets are seen and the samples freeze. The size of these droplets can be expected
maltodextrin's refractive index falls to about 1.530. This to depend somewhat on mixing conditions used in the solu-
behavior can be explained by the fact that refractive indices tion preparation. Volatiles of intermediate solubilities will
of mixtures are additive and therefore depend on tbe frac- reach their solubility limits at different stages of the freezing
tion of each component present in the mixture. When the process and can develop as droplets to an extent dictated by
freeze-dried material acts as a homogeneous mixture (for the sample temperature and freezing history. It is uncertain
ACKNOWLEDGMENT
We would like to thank the Statens Teknisk-Videnskahelige
Fond (Denmark) for their support, the Unibersitetets Min-
eralogisk-Geologiske Institut (Copenhagen, Denmark) for
Figure 6. Freeze-dried coffee particles containing small droplets use of the scanning electron microscope, and Lahorant Britt
(400X) Vinderslev Jensen for the freeze-drying preparation.
(i.e., aroma concentrates). For these cases the presence of ReY, L., Bastien, M.-C., in “Freeze-Drying of Foods,” Fisher,
minuteliquid droplets the dry matrix would
not appear to be compatible with concepts based solely on
..-..,w.shlnotnn
rAm.nni 1 1 1 1 .,--,Ilr
n r 70m
~
nn 7c-d7 -_
F. R.,. Ed., National Academy of Science-National Research
-__ ..... -.-., .-.., _-.
Schaeffer, H. F., “Microscopy for Chemists,” Dover Press, N.Y.,
sorption in the dry layer, or diffusion of molecular species. 1966.
Thijssen, H. A. C., Rulkens, W. H., in “Symposium of Thermody-
As the volatile solubility increases, however, the droplet di- namic A~~~~~ of Freeze-~rylng;, intern at ion^ ~~~~i~~~~ of
ameter apparently decreases; diffusion regulation may there- Refrigeration,Commission X, Lausanne, Switzerland, 1969.
fore become of increasing importance in such cases. How-
ever, the presence of droplets in freeze-dried coffee solids Received for reuiew November 8, 1971. Accepted January 3, 1972.