Chemistry Reference
In-Depth Information
deodorization), these values can be kept at a constant low level. In cases where the lev-
els of these components are too high, reprocessing may be required if a satisfactory enzyme
working life is to be achieved. Additives such as silica have been suggested as absorbents of
polar compounds. Lee and Sleeter
13
proposed the use of a number of different pretreatment
materials to be used in a column including activated carbon, spent enzyme kieselguhr and
silica. They reported that for a soya bean oil-based blend, the flow rate required for full
conversion was 10% of the starting level after 40 days for the standard blend. For the reactor
in which the oil had received silica pretreatment, the flow was 18% of the starting level,
indicating almost double the residual enzyme activity at this time. Based on the type of ma-
terials being used, it would seem that it was primarily oxidation compounds that were being
removed. Ibrahim
et al
.
14
also focused on oxidation compounds and proposed the use of
spent enzyme catalyst as a means of removing enzyme-inactivating species. Using molecular
sieves, activated carbon and deactivated but unused enzyme, they increased productivity (kg
oil converted/kg enzyme) by a factor of 3.1, 7.4 and 4.1 fold, respectively. From this they
concluded that spent enzyme might be an effective purifying material but this assumes that
the sites of absorption for enzyme-poisoning compounds have not all been occupied during
the normal use of the enzyme product.
Elimination of acids can be achieved by not using phosphoric acid in degumming and
ensuring that a bleaching earth, which has been fully washed to remove residual acid, is
employed. Where this is not possible or where the oils are produced at another location,
alkali treatment of the oil is a possibility. Sodium carbonate and potassium hydroxide have
both been used to remove residual acidity and cause an increase in enzyme productivity.
13
15.5.3 Practical operation of EIE
In factory operations there are two parameters that also need to be controlled. The fat melting
profile or SFC needs to match that of the desired specification, and the quality (measured
as crystallization characteristics - relative content of different crystalline forms; colour -
predominately red and yellow; and odour) of the produced fat needs to be at least as good as
that produced by the alternative technology.
The SFC is a measure of the amount of fat solid at a particular temperature and it is this
which controls the melting properties. The SFC curve shown in Fig. 15.1 does not totally
match the one for CIE. In practice, batch reactions with differing proportions of the two
components of the blend fats would be conducted in order to find the exact mix that gave
the desired results. This is illustrated in Fig. 15.7, which compares the SFC curve for EIE
of three blends of palm stearine and sunflower oil with a product obtained from CIE of
a 30:70 blend of the same fats. The curve for the same proportion blend made by EIE is
slightly below that of the chemical process. By adjusting the blend used for EIE towards a
32:68 proportion, an exact match can be achieved. Similarly to replace partly hydrogenated
fats, a similar approach can be followed but a wider range of fat blend compositions may
be required in order to fully match the properties of the originally hydrogenated fat. For
example, partially hydrogenated soya bean oil can be substituted by the interesterification
of fully hydrogenated soya bean oil with liquid soya bean oil. This produces the desired
physical properties that does not lead to the formation of trans fats nor increase overall the
level of saturated fats.
The composition that provides the desired SFC values is then used for continuous in-
teresterification in the series reactor system. The batch interesterification is carried out to
