Author: Chris Geary, Bachelor of Engineering (Hons) in Chemical & Biopharmaceutical Engineering, Department of Process, Energy and Transport Engineering, Cork Institute of Technology [caption id="attachment_13830" align="alignright" width="399"] Fig 1: Microcrystalline cellulose[/caption] FMC BioPolymer in Little Island, Cork, is a diversified chemical company serving agricultural, industrial, environmental and consumer markets. The Cork location produces MCC (microcrystalline cellulose), which is used as a bulking agent, binder and filler. It makes the ideal excipient (MCC powder is pictured right). Line-pressure variance is an ongoing constraint in the production of MCC, however. High viscous suspensions cause inefficient pumping from reslurry tanks to spray dryers and this, in turn, limits production rates. The goal of this project was to identify the cause(s) and bring forward a solution to eliminate the viscous nature of the MCC suspension prior to its transfer to the spray dryer, whilst producing PH product only. This goal was achieved by:

  • Analysing existing data related to line pressure variance;
  • Carrying out a pH investigation on the effect of viscosity of MCC; and
  • Heavy/light pulp weight trials.
ANALYSIS OF VARIABLES The following analysis was completed through examining existing data:
  • Reslurry tank
The temperature rise resulted in high line pressures, i.e. higher viscosity over 40° Celsius. The pH proved to be higher for batches that resulted in low line pressures variance. The higher pH causes MCC suspension to flow more easily. High conductance of suspension results in high line pressures.
  • Quench tank
The use of recycled finished product in batches results in more efficient filtration, leading to stable line pressures. HEAVY AND LIGHT PULP TRIALS These trials were used to investigate the varying ranges in weights of pulp and effect on line pressures. Trial A contained a light and heavy pulp weight batch, with finished product usage (Silo A). Trial B also contained heavy and light batches, but with no finished usage.
Weight difference of total pulp (kg)
Trial A (with Silo A) 178
Trial B (without Silo A) 109
These pulp trials (heavy and light weight pulp), using finished product as quench material, did not cause any viscosity issues prior to drying. Pressure varied from 7 bars to 11.2 bar for the light pulp trial without Silo A usage. Trough levels dropped (these are easier to filter), resulting in thicker cake, less washing and high conductance, whilst high pulp trial resulted in stable line pressures (see the graph of line pressures below for Trial B). [caption id="attachment_13840" align="alignright" width="418"] Fig 2:
Blue line: Filter feed pump
Green line: Reslurry tank line pressure (bar)
Turquoise line: Filter B rate (kg/hr)[/caption] pH INVESTIGATION Reslurry tank samples were taken and the pH was adjusted and viscosity tested using a Brookfield R/S Rheometer with c50-1 spindle for each different pH sample. A constant shear and temperature were maintained for the runs, and a table was produced (see Fig 3). As can be seen, there was a significant drop of viscosity, from pH6 to pH7. There was no significant change in viscosity above pH7, due to the complete ionisation of surface functional groups. CONCLUSION
  • Raising the pH above 7 will not result in easier flow of MCC suspension;
  • A difference of 109kg in pulp weights is significant;
  • Cook is ‘over’ cooked from light cook and results in lowered troughs, lower feed solids and higher conductance;
  • Recycled finished product acts as a filter aid and results in the reduced viscosity of MCC suspension;
  • A temperature of over 50°Celsius will raise the viscosity.
[caption id="attachment_13851" align="alignright" width="427"] Fig 3[/caption] With regard to recommendations, these investigations would suggest that Silo A material should be utilised regularly. When Silo A material is unavailable, it should be ensured that pulp weight is high (approximately 773kg a roll). The temperature may be dropped to 40°Celsius for the pH process, but it appears that further investigation is required. References AM Adel, ZH Abd El-Wahab, AA Ibrahim and MT Al-Shemy, ‘Characterization of Microcrystalline Cellulose Prepared from Lignocellulosic Materials. Part II: Physicochemical Properties,’ Carbohydrate Polymers, Vol. 83, No. 2, 2011, pp. 676-687. doi:10.1016/j.carbpol.2010.08.039