The process of micro-encapsulation by emulsifying a solution in a stirred tank has been accepted as
the most suitable method for the production of microcapsules for carbonless copy paper and is
currently used by Mondi Paper in Merebank. The focus of this project was to obtain a more
uniform size distribution of the microcapsules so that oversize capsules would not smudge when
they are coated on paper. There was also concern that the formation of very small ink/oil droplets
was consuming wall material unnecessarily and was not contributing to the formation of an image
on paper. The reduction of these tiny droplets would result in a saving of the cost of the wall
material.
Mondi currently produces microcapsules with an average diameter between 4 and 5 microns. The
amount of capsules produced above 10 microns, the oversize, is less than 1 per cent (v/v) and the
amount of capsules produced below 2 microns, the undersize, is between 25 and 30 per cent. Mondi
wishes to reduce the amount of undersize capsules, thereby producing a narrower size distribution.
This could result in large savings, as discussed above. It could also lead to the production of a six-sheet
set of carbonless papers instead of the four-sheet set, which is currently produced. The
production of microcapsules by emulsification was investigated in a 2.5-1iter laboratory tank, using
an impeller measuring 45 mm in diameter.
A range of agitation speeds was investigated and it was seen that at the lowest speed that formed
emulsions, 6600 rpm, 15.03 per cent of undersize capsules was produced and an average capsule
diameter of 7.57 microns, after 40 minutes of agitation. At the highest impeller speed, 8000 rpm,
the average capsule diameter was reduced to 1.93 microns and 67.02 per cent of undersize capsules
were classified as "undersize". No oversize capsules were observed. These capsule specifications
were not favourable. Further experimentation showed that at 7500 rpm, an average capsule
diameter of 5.12 microns and an undersize of 24.20 per cent were observed. The proportion of
oversize capsules was 1.63 per cent. Since these results were similar to the results obtained from
the plant, 7500 rpm was used accepted as the "standard" speed for the experiments. A reduction in
the impeIler speed from 7500 rpm to 7200 rpm after the first 20 minutes of emulsification was one·
way on reducing the proportion of undersize particles further. The proportion of undersize particles was reduced from 20.20 per cent to 19.71 per cent at standard conditions. The average capsule
diameter and the oversize were not affected significantly.
The effect of the emulsification temperature on the particle size distribution was investigated with
temperatures ranging from 22 to 40°C, in increments of 2 QC. A temperature of 30 °c was used as a
standard temperature as this temperature was being used at the plant. A decrease in the proportion
of undersize capsules to 17.12 per cent was seen at temperatures below 30°C and an average of
23.87 per cent was noticed above 30 QC. Although the proportion of undersize capsules decreased,
the average capsule diameter increased beyond the specified range to an average of 7.77 microns at
temperatures below 30°C. At temperatures above 30 °c the average size was reduced to 5.59
microns. Hence the selection 000 °c as the optimum temperature was confirmed.
Experimentation with the emulsification time showed that there were times when a unimodal size
distribution was produced. However, these were at times just after the polymerisation had begun,
and the reaction was not complete at this stage. A bimodal distribution was always noticed after 40
minutes of emulsification, i.e. after the completion of the reaction.
The effects of the baffle widths on the microcapsules were also investigated. Baffle sizes of 5, 10
and 15 mm were used. It was shown that with an increase in baffle width, there was a decrease in
the amount of undersize capsules produced. However, the average capsule diameter became too
large. A baffle width of 5 mm was shown to produce desirable capsule sizes, although the
undersize did not improve, or worsen. Too much of air was trapped in the emulsion when no
baffles were used in the tank.
Alternatives to the current surfactant, called "Lupasol" were tested so that Mondi could produce the
capsules independently instead of relying on the original raw material supplier. This investigation
was done based on limited informa.tion on Lupasol. Results from these experiments were
inconclusive since more data on Lupasol was required.
Samples of the microcapsule emulsion were sent to different companies, in South Africa and
abroad, to determine whether the particle size analyser used at Mondi was giving correct results.
The results obtained from the companies in South Africa differed by a small amount from that
measured at Mondi. However, results obtained from companies abroad varied considerably and it is
recommended that Mondi change their particle analyser settings. The power absorbed by the emulsion, in the laboratory-scale equipment was also found. This was
determined by monitoring torque. The power was found to be 141.97 Wand the power number was
calculated as 0.357. It was noted that the power per unit volume in the laboratory equipment was
significantly higher than the plant data (47 kW/m3 vs. 12 kW/m3).
The design of the impeller was not changed but the effect of baffle spacing was investigated. / Thesis (M.Sc.Eng.)-University of Natal, Durban, 2002.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/4700 |
| Date | January 2002 |
| Creators | Chetty, Ezekiel. |
| Contributors | Loveday, Brian Kelsey., Marsh, Jeremy J. |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0025 seconds