Chemistry Reference
In-Depth Information
heterogeneities. With increasing length of the column (increasing number
m
Max
) the
maxima of the distribution function
W
(
r
, 0.5) and the distribution function
W
(100,
y
)
shift for the observed volume elements to lower values and both distribution
functions become narrower. For the volume element
v
70 moving through the
column, the length of the column has the opposite effect. The practical consequence
is the relative large polydispersity in the large fraction.
Increasing the number of theoretical plates,
m
Max
, allows establishing a more flat
temperature gradient in the column for the same fractionation results.
In summary, the suggested theoretical model can be applied to answer different
questions arising about the efficiency of copolymer fractionation performed in BW
columns. This type of column is mostly used for analytical purposes.
¼
3.4 Continuous Polymer Fractionation
CPF has been especially developed to produce large fractions in a relatively short
time frame and can be applied for preparative purposes. All traditional procedures,
especially the stepwise methods, require a low polymer concentration for good
efficiency, and large amounts of solutions must be handled to obtain sufficient
material. With this fractionation method, the initial copolymer is divided into two
fractions, where these fractions can be used again as feed for the next fractionation
run. For homopolymers, this fractionation method is a useful tool for cutting the
short molecular weight parts or the extremely high molecular weight parts from the
desired product. In the case of copolymers, those with extremely high or low values
for the chemical composition can also be removed from the product. The simulation
method suggested above is now applied for the optimization of the CPF. This
optimization is always done by variation of the parameters describing the column
and keeping all others constant. The copolymer (
68
) used for the simulations and
the parameters of the
G
E
model (
69
) are identical to those used for the simulation of
the BW columns. The standard parameter set for the operating conditions is:
X
FD
11
Z
FD
5
Z
EA
n
FD
n
EA
¼
0
:
¼
0
:
¼
0
:
15
=
¼
0
:
1
(73)
m
Max
¼
6
m
FD
¼
3
T
¼
520 K
T
condenser
¼
500 K
:
For example, the calculated fractionation data for four CPF runs are collected in
Table
1
, where the initial polymer distribution was a Stockmayer distribution (
15
).
For the fractionation, four CPF runs were simulated in which the obtained gel
fractions were directly used as new feed phase.
The data in Table
1
make it clear that no significant fractionation according to
the chemical heterogeneity took place. However, the fractionation effect with
respect to the molecular weight is characterized by a high effectivity. Except for
the first fraction, all other fractions have a nonuniformity lower than 0.06, similar to
the results obtained for the simulation of the BW column. The nonuniformities are
