- Fractionation is the process, by which mixtures of Fatty Acid are separated into distilled pure individual components or combination products. Separation is achieved by contacting a rising vapour stream with a downward liquid stream. The vapour continuously enriches with the light components, while heavier components migrate to the downward liquid. Vapour-Liquid contact is established, using structured packing or equilibrium contacting trays.
- Columns comprise a rectification section, above the feed point, and a stripping section below it. The height of packing or number of trays, in rectification is selected based on the purity requirement of the top product while the height of stripping section determines the purity of the bottom product.
- Tray columns provide total flexibility in sizing the rectification and stripping sections, as the feed can be introduced at any point on the column. By contrast for a packed column, feed can only be introduced between the beds. For a 2 bed column there can be only one feed point and one product draw off. A third bed is usually introduced to provide a second feed and draw off alternative. Designs of packed columns therefore need to be very precise. Packings however are much cheaper than trays and also have much lower pressure drop. HETP is usually lower for packings than trays, requiring shorter columns. (HETP is defined as Height of Equilibrium Theoretical Plate or packing height equivalent to a theoretical tray. NETP is the number of transfer stages per packed bed.)
- Packings provide a 1.5 to 2 times capacity advantage over trays. Structured packings typically used for Oleochemicals provide 250 m2 packing area per m3 of packing volume.
- Column capacity decides the column diameter. A column is designed to ensure that it operates within the operating range of the packing, defined by F-factor and Liquid Load. (F-factor is square root of vapour density multiplied by velocity while liquid load is the volumetric liquid flow per unit cross-section.) Effectively the factors increase and decrease with the gas and liquid velocities. These must be maintained within the operating range of the packing.
- Higher column pressure compresses the vapour stream, reducing vapour velocity, bringing the F-factor within packing operating range. Higher pressure is needed to operate at higher capacities. Typically a 1.45 Φ column processing 12 TPH feed at 30 torr would be able to operate at 15 TPH at 80 torr. This example is for fractionation of C8:10 from C8:18 PKO-CFA. Pick the highest column pressure which maintains a reasonable bottom temperature.
- The role of feed temperature is multifold. It directly changes the reflux required for a particular separation by changing ‘minimum reflux’. A feed temperature below the feed tray bubble point, will derate the stripping section, by diverting its function from mass to heat transfer. Vapour and liquid flows in the stripping section also increase. Specifying a low feed temperature, is a common error, as standard fractionation programs Hysis / ProII / CHEMCAD indicate a lower reflux at lower feed temperatures.
- A pump around condenser can be considered for larger operations. Pump-around provide the option of recycling reflux prior to heat extraction from the external steam generators, making it possible to maintain reflux at column top bubble point and eliminating unnecessary boil-up. By eliminating this additional boil-up, column capacity increases, and energy consumption is reduced.
- If there is a choice to fractionate fatty alcohols instead of fatty acids, it should be opted for. Fatty Alcohols boil on an average about 20°C below Fatty Acids. This provides very significant advantages, as higher column pressures can be used. Both through-put and yields are benefitted. Fatty Alcohols are also non-corrosive.
Simulator: Design Basis & Scope
The unique advantage of this software is that it combines fractionation calculations with packing pressure drop and NEPT calculations in the same program. This facilitates the fractionator design process, which is necessarily iterative. Accurate designs and operation predictions are authenticated with extensive plant data for fractionation of C6 to C22 Fatty Acids, as well as Fatty Alcohols.
- Component wise material balance / Slippages.
- NETP – Rectification / Stripping,
- Operating parameters – Reflux, ΔP, F-factor, Liquid Load,
- Column Temperature and Pressure profiles.
- Re-boiler & Condenser heat duties
A Fouling Factor has been incorporated, to evaluate older columns operating continuously without cleaning.
1. The fractionation process operates within 2 defining limits.
- Minimum NETP at infinite or total Reflux.
- Minimum Reflux at Infinite NETP
2. NETP (min) is a function of component relative volatilities of the feed and top product, calculated using the Fenske Equation.
3. Minimum Reflux is a function of feed component relative volatilities and feed temperature, calculated using the Underwood Co-relation.
4. Reflux Ratio for a separation purity and NETP is derived using the Erbar Maddox or the Gilliland Co-relation. Both co-relations are a function of minimum NETP and minimum Reflux. The simulator uses the Erbar Maddox co-relation as it better duplicates operating conditions.
5. The Design Process is necessarily Iterative. Start with the feed and product relative volatilities, calculate NETP (min), R (min) and reflux for particular NETP, feed and column temperature. Use the Reflux to calculate Vapour / Liquid flows, packing ΔP, F-Factor, Liquid Load and NETP. Generate a new Temperature profile, altering R (min), NETP (min), relative volatilities, and start another calculation cycle.
6. Equations for ΔP / m and HETP as a function of F-factor and Liquid Load have been developed for 250 m2/m3 packing. These have been authenticated using extensive plant data. The published pilot plant data of Professor Zarko Olujic is also extensively used.
7. To assess columns operating without servicing for extended periods a fouling factor is built into the program. The factor reduces the effective contact area and void space in the packing, increasing the packing HETP and ΔΡ.
8. Dissatisfaction with simulations by packing manufacturers using PRO-II, Hysis prompted development of this program.