Hot ethanol extraction of lipids from plant or animal materials

Abstract

The present invention relates to a process for the production of a fat rich and a fat lean product from a plant or animal starting material, comprising the steps of: i) providing a particulate fat containing starting material, ii) mixing the particulate fat containing starting material with ethanol of at least 90% w/w concentration, iii) heating the mixture, iv) subjecting the heated mixture to a hydrocyclone treatment to provide a fat reduced underflow stream and an overflow stream, v) recovering the fat lean product from the fat reduced underflow stream, vi) recovering the fat rich product from the overflow stream. The fat rich and the fat lean products are suited for use in human food or for animal feed products.

Claims

1. A process for producing two products from plant or animal starting material, one of the two products being relatively lipid rich compared to the other product so that the one product is a lipid rich product and the other product is a lipid lean product, the process comprising: i) providing a particulate lipid containing starting material, ii) mixing the particulate lipid containing starting material with ethanol of at least 90% w/w concentration to produce a mixture, iii) heating the mixture to a temperature of at least 60 C. to produce a heated mixture in which lipid is solubilized in the ethanol, iv) subjecting the heated mixture to a hydrocyclone treatment to extract lipid from the particulate lipid containing starting material using the ethanol as an extractant and provide an underflow stream with a reduced lipid content relative to the heated mixture and an overflow stream comprising a solution of lipid in ethanol, the heated mixture being at a temperature of at least 60 C. throughout the extraction of the lipid from the particulate lipid containing starting material using the ethanol as the extractant, v) the subjecting the heated mixture to the hydrocyclone treatment comprising subjecting the heated mixture to sequential counter current processing in a plurality of hydrocyclones arranged one after another in a downstream direction so that the lipid reduced underflow stream of a first of the hydrocyclones is fed to a second of the hydrocyclones that is located downstream of the first hydrocyclone, the first hydrocyclone being upstream of the second hydrocyclone, and so that the overflow stream of one of the hydrocyclones is fed to an up-stream hydrocyclone; vi) recovering the lipid lean product from the lipid reduced underflow stream of a last hydrocyclone of the plurality of hydrocyclones, and vii) recovering the lipid rich product from the overflow stream of the first hydrocyclone.

2. The process according to claim 1, wherein the overflow stream of the first hydrocyclone stage is subjected to a clarifying separation to provide a clarified stream and a concentrated stream.

3. The process according to claim 2, wherein clarifying separation comprises a series of from 1 to 3 hydrocyclone stages, or the clarifying separation is accomplished using a centrifuge.

4. The process according to claim 1, further comprising: cooling the overflow stream of the first hydrocyclone stage to a temperature of 40 C. or below to form an ethanolic upper phase and a lower phase, recovering the fat rich product from the lower phase, and optionally recycling the ethanolic upper phase.

5. The process according to claim 1, wherein the heated mixture is subjected to hydrocyclone treatment in two or more hydrocyclones arranged in parallel or wherein each stage in a series of hydrocyclones comprises two or more hydrocyclones.

6. The process according to claim 1, wherein the number of hydrocyclone stages in the series is 3 or more.

7. The process according to claim 1, wherein a diameter of an upper part of the hydrocyclone is 25 mm or less.

8. The process according to claim 1, wherein the mean diameter of the particles of the particulate lipid containing starting material is from about 20 m to about 300 m.

9. The process according to claim 1, wherein a standard deviation Dv95 of a mean diameter of the particles of the particulate lipid containing starting material is 50% or less from the mean diameter of the particles.

10. The process according to claim 1, wherein the particulate lipid containing starting material has a water content below 3% w/w.

11. The process according to claim 1, wherein the temperature is at least 65 C.

12. The process according to claim 1, wherein the heated mixture of the particulate lipid containing starting material and ethanol is superheated above a boiling point of the mixture.

13. The process according to claim 1, wherein the ethanol concentration is at least 95% w/w.

14. The process according to claim 4, wherein the ethanolic upper phase is dehydrated before the recycling of the ethanolic upper phase.

15. The process according to claim 1 further comprising reducing a water content of the particulate lipid containing starting material to below 3% w/w in the steps of: a) mixing the particulate lipid containing starting material with 1 to 20 parts lipid to obtain a slurry, b) heating the slurry to evaporate water and to obtain a water reduced slurry, c) subjecting the water reduced slurry to a solid-liquid separation to provide the particulate lipid containing material having a water content below 3% w/w and a lipid fraction.

16. The process according to claim 15, wherein step b) is performed as a multi-step process comprising subjecting the water reduced slurry to sequential heat treatments, where each subsequent heat treatment is performed at a lower temperature than a preceding heat treatment.

17. The process according to claim 15, wherein the lipid is derived from the lipid rich product produced according to steps i)-vi).

18. The process according to claim 15, wherein the lipid rich product is recovered from the lipid fraction.

19. The process according to claim 1, wherein the lipid reduced underflow stream is subjected to centrifugation to produce the lipid lean product.

20. The process according to claim 1, wherein the starting material is slaughterhouse by-products, dehydrated slaughterhouse by-products, fish by-products, oil seeds, or a press cake remains after extraction of oilseeds.

21. The process according to claim 1, wherein the underflow stream and the overflow stream in each hydrocyclone are the only flows resulting from each hydrocyclone.

22. A process for producing two products from plant or animal starting material, one of the two products being relatively lipid rich compared to the other product so that the one product is a lipid rich product and the other product is a lipid lean product, the process comprising: i) providing a particulate lipid containing plant or animal starting material; ii) mixing the particulate lipid containing plant or animal starting material with ethanol of at least 90% w/w concentration to produce a mixture in which lipid is solubilized in the ethanol; iii) subjecting the mixture to a hydrocyclone treatment to extract lipid from the particulate lipid containing starting material using the ethanol as an extractant to provide an underflow stream with a reduced lipid content relative to the mixture and an overflow stream comprising a solution of lipid in ethanol, the mixture being at a temperature of at least the boiling point of the ethanol throughout the extraction of the lipid from the particulate lipid containing starting material using the ethanol as the extractant; iv) the subjecting the heated mixture to the hydrocyclone treatment comprising subjecting the heated mixture to sequential counter current processing in a plurality of hydrocyclones arranged one after another in a downstream direction so that the lipid reduced underflow stream of a first of the hydrocyclones is fed to a second of the hydrocyclones that is located downstream of the first hydrocyclone, the first hydrocyclone being upstream of the second hydrocyclone, and so that the overflow stream of one of the hydrocyclones is fed to an up-stream hydrocyclone v) recovering the lipid lean product with the reduced lipid content from the underflow stream; and vi) recovering the lipid rich product from the overflow stream.

23. A process for producing two products from plant or animal starting material, one of the two products being relatively lipid rich compared to the other product so that the one product is a lipid rich product and the other product is a lipid lean product, the process comprising: i) providing a particulate lipid containing plant or animal starting material; ii) mixing the particulate lipid containing plant or animal starting material with ethanol of at least 90% w/w concentration to produce a mixture in which lipid is solubilized in the ethanol; iii) subjecting the mixture to a hydrocyclone treatment to extract lipid from the particulate lipid containing starting material using the ethanol as an extractant to provide an underflow stream with a reduced lipid content relative to the mixture and an overflow stream comprising a solution of lipid in ethanol, the lipid being extracted from the particulate lipid containing starting material using the ethanol as the extractant while the mixture is at a temperature of at least 60 C.; iv) the subjecting of the mixture to the hydrocyclone treatment comprising subjecting the mixture at the temperature of at least 60 C. to a sequential counter current processing in a plurality of hydrocyclones arranged one after another in a downstream direction so that the lipid reduced underflow stream of a first of the hydrocyclones is fed to a second of the hydrocyclones that is located downstream of the first hydrocyclone, the first hydrocyclone being upstream of the second hydrocyclone, and the overflow stream of one of the hydrocyclones is fed to an up-stream hydrocyclone, and the lipid lean product is recovered from the lipid reduced underflow stream of a last hydrocyclone of the plurality of hydrocyclones, and the lipid rich product is recovered from the overflow stream of the first hydrocyclone, the plurality of hydrocyclones being at least three hydrocyclones; v) recovering the lipid lean product from the underflow stream with the reduced lipid content; and vi) recovering the lipid rich product from the overflow stream.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) The invention will be readily understood from the following detailed description in conjunction with the accompanying figures, in which

(2) FIG. 1 shows the solubility of oil/fat in ethanol.

(3) FIG. 2 shows a schematic process diagram of an embodiment of the invention.

DETAILED DESCRIPTION OF THE INVENTION

(4) In order to more fully explain the invention it is disclosed in more detail below, and definitions of the terms used throughout the document are given.

(5) The present invention relates to a process for the production of a fat rich and a fat lean product from a plant or animal starting material. In the context of the invention the term fat refers to a triglyceride of biological origin. The fat may also be referred to as oil and the two terms may be used interchangeably in the context of the invention.

(6) The starting material is of plant or animal origin. Any plant commonly cultured to provide a plant oil may be used in the invention, and any part of such plant may be subjected to the process. Typical plant materials are derived from rapeseed, canola, soybean, sunflower seed, peanut, cottonseed, palm, palm kernel, coconut, olive, corn, grape seed, hazelnut and other nut, linseed, rice bran, safflower, sesame, etc. The plant material may for example be the press cake remains after extraction, e.g. expeller extraction, of the oilseeds. In general, any oilseed may be processed according to the invention. The animal may be any land animal, such as livestock, poultry etc. or marine animal, such as fish. In particular, the starting material may be derived from slaughterhouse by-product or dehydrated slaughterhouse by-products, or from by-product from industrial processing of fish.

(7) The process of the invention employs ethanol. In the context of the invention ethanol is referred to in terms of concentration where the ethanol concentration is provided as a percentage by mass, i.e. % w/w. The balance will normally be water unless otherwise indicated. The ethanol may be referred to as azeotropic meaning that it contains 95.6% w/w ethanol and 4.4% w/w water. The term absolute ethanol refers to ethanol of at least 99.9% w/w concentration.

(8) The enclosed nature, and also the small volume of the preferred hydrocyclone size, allows that the feed stream fed to the hydrocyclone is heated above its boiling point without evaporating when processed in the hydrocyclone. This effect of retaining a liquid state is referred to as superheated in the context of the invention. This is particularly advantageous since it allows that the mixture of the particulate fat containing starting material and the ethanol is heated to a temperature where the fat and oil of the particulate fat containing starting material are dissolved in the ethanol and thus may be extracted from particulate fat containing starting material.

(9) The following abbreviations may be used to describe the present invention:

(10) TABLE-US-00002 Abbreviation Term SSP (in kg/h or in %) Suspended Solid (protein) L (in kg/h) Liquid (alcohol) T (in kg/h) Mass flow (SSP + L) FLOW (in m.sup.3/h) Volume flow S.G (in kg/m.sup.3) Specific gravity SLS (in kg/h) Soluble Solids (fat) TS (in kg/h) Total Solids (SSP + SLS) TS (in %) Total Solids (SSP + SLS) TOT (in kg/h) Total mass flow (SSP + L + SSL) F.L. Fresh Liquid W.L. Washing Liquid UF Underflow OF Overflow F Feed flow delta P Pressure drop D10 Cyclonette with 10 mm upper diameter

(11) In a specific embodiment of the invention the fat containing starting material is a slaughterhouse by-product. This starting material typically contains about water. In order to provide a shelf stable protein meal the water content needs to be reduced. Besides protein slaughterhouse by-products contain minerals and a varied quantity of fat depending on the specific by-product fraction. Slaughterhouse by-products can therefore appropriately be treated according to the invention. Conventional dehydration processes are very costly both in terms of energy usage, but also as large and expensive equipment is needed to transfer evaporation heat into wet products. Since products containing high fat levels have better heat conductivity, less drying surface is needed in indirect heaters, and the slaughterhouse by-product can be initially dehydrated according to the Carver-Greenfield process as outlined above.

(12) Existing carver Greenfield installations typically use falling film tubular heat exchangers. However, the present inventors have now found that surprisingly it is possible to use conventional plate heat exchangers of the widegap type. Raising film flash plate cassette evaporators of flash or conventional type provide a compact though flexible and cost efficient design for making dehydration units. Obviously, solid particle size needs to be calibrated to at least less the clearance of the plate heat exchangerand ideally less. For example, a cassette with an 8 mm clearance, will face significantly less risk of clogging if particle sizes are not more than 2-3 mm. Furthermore, the solid particles should ideally be spherical rather than fibrous. Fibres can be caught at supporting points between plates and also in the distribution channel. Consequently, the particle size reduction mill needs to be of a certain quality. We have found that high speed colloid pin mill in combination with a slotted safety filter provides stable particle size output. It has further proven advantageous, that proteinaceous material of animal origin is pre-minced in a conventional hole-plate grinder and thoroughly pre-coagulated before the final milling. Coagulated, even thoroughly milled raw materials tends to re-adhere into fibrous structures.

(13) Smaller particle size further significantly shortens dehydration time as both heat penetration increases and also water/steam diffusion towards the surface is reduced. In conventional Carver Greenfield operations there is however a general reluctance to reduce particle size too much, as this will negatively affect the ability of expeller presses to build up high pressure thus extracting maximum amount of fat from final product. In this invention, however, we will not use expeller presses, but centrifuges, e.g. decanter centrifuges, as the fat left within the solids will be recovered in the following extraction phase. Decanters are not sensitive to product particle size, and will readily recover fine solids particles having a higher density than the carrying fat. Thus, following dehydration in the Carver Greenfield process the slurry may be subjected to solid-liquid separation in a decanter centrifuge to provide the particulate fat containing starting material and a liquid fat phase from which the fat rich product may be recovered.

(14) The slaughterhouse by-product can for example be pre-milled in a hole-plate grinder followed be re-milling, e.g. by a pre-emulsifier, to provide a wet fatty product of a fineness that allows stable operation of the plate heat exchangers in the CG dehydration. Pre-milling may be done directly to the slaughterhouse by-product, whereas re-milling is typically done after suspending the pre-milled slaughterhouse by-product in ethanol.

(15) Pending the milling system, the dehydrated product may still contain pieces (e.g. 2-3 mm) of bones and the like. To ensure that the particulate starting material is within the size range optimal for the operation of the hydrocyclones, the particulate material could undergo a re-milling after dispensing in ethanol, e.g. by a colloid wet-milling in combination with a classifying hydrocyclone treatment, where the underflow is recycled for re-milling, and the overflow containing appropriate sized product is subjected to the extraction process of the invention. The particle sizes of the materials may be monitored throughout the process.

(16) The particulate fat containing starting material to be extracted is mixed with ethanol, milled to appropriate particle size in a colloid mill, preheated in an economiser system before final heating and feeding into the hydrocyclone extraction system. Extraction ethanol is fed into the hydrocyclone system counter-current to the product flow. The underflow of the last hydrocyclone station or stage containing defatted material, may be subjected to heat reduction before being fed to a decanter centrifuge for maximum ethanol removal prior to desolventising.

(17) The overflow from the hydrocyclone treatment containing the fat/oil may be clarified using e.g. hydrocyclones in clarifying configuration and/or a high-speed disk stack centrifuge, before being chilled to separate the fat from the ethanol. After chilling down to e.g. 25 C. or less the optionally clarified overflow may be transferred to a column or cone shaped phase separator, or a disk stack centrifuge.

(18) The recovered fat/oil contains minor amounts of ethanol which can be removed by conventional vacuum stripping technology. If required, phospholipids can be removed using conventional degumming technology, comprising addition of water and lye, mixing and separation.

(19) Solid desolventising can be achieved with any known technology, being indirect heated vacuum driers or by superheated ethanol flash drying.

(20) Processing of a slaughterhouse by-product according to the process of the invention provides a fat rich product comprising triglyceride fats from the material, and a fat lean product which may be referred to as a protein meal. Both the fat rich and the fat lean product can be used for human or animal consumption, e.g. as a component of an animal feed. The fat rich product may also be used as a starting material for the production of biodiesel or the like.

EXAMPLES

Example 1Tests of Hydrocyclone Operating Conditions

(21) A slaughterhouse by-product was treated according to the invention. The separation tests were done with a single 10 mm cyclonette fed with a suspension of fatty meal and alcohol. The fatty bone meal was prepared from the material, which had been dehydrated in a carver Greenfield process, pre-milled in a hammermill and was re-milled to a fineness that allowed a stable operation of the cyclonette and a proper recovery of solids.

(22) The re-milling was done with a suspension, at TS concentration 32%, in a mill with double cone (both with ribbons) of which the clearance could be adjusted to a finer or courser milling.

(23) The first separation test was made with a suspension, at TS concentration approximately 27%, that was re-circulated over the mill for 8 minutes at a clearance of 8/18 (indication on mill).

(24) The test rig operated well for approximately half an hour and then plugged. To assure a consistent stable operation of the test rig it was decided to re-mill the starting material to a finer grade.

(25) The additional re-milling was done during 10 minutes at the finest clearance 0/10 (0.1 mm clearance between the cones). The separation tests were done with this suspension.

(26) Additional separation tests were done with a suspension, at a TS concentration of approximately 7%, from which the coarse fraction had been removed (by discharging the underflow from the cyclonette for a while). These tests were done to get an impression of the operation of clarifier stages that would be installed to clarify the overflow of the washing station.

(27) The test runs are summarised in Table 2 and Table 3. Table 2 shows the test of the hydrocyclone operating parameters without dry solids in the feed. The feed temperature was 22 C., and the feed specific weight was 800 kg/m.sup.3. Table 3 shows the test of the hydrocyclone operating parameters with dry solids in the feed. Table 4 shows the results of hydrocyclone tests with dry solids. The feed temperature was 88 C., except in test 7, and the specific weight of the feed was 824 kg/m.sup.3, except in test 12.

(28) TABLE-US-00003 TABLE 2 Test of hydrocyclone operating parameters without dry solids. Feed OF UF Feed P OF P UF P Flow delta P Flow Flow UF/F Test (Bar) (Bar) (Bar) (l/h) (Bar) (l/h) (l/h) Flow 1 3 0.4 0.4 294 2.6 105 189 0.64 2 4.1 0.8 0.8 338 3.3 136 202 0.60 3 5.3 1 0.8 363 4.3 140 223 0.61 4 6 1.1 1 405 4.9 150 255 0.63 5 6.9 1.2 1 425 5.7 159 266 0.63 6 7.6 1.4 1 431 6.2 167 264 0.61

(29) TABLE-US-00004 TABLE 3 Test of hydrocyclone operating parameters with dry solids. Feed OF UF Feed P OF P UF P Flow delta P Flow Flow UF/F Test (Bar) (Bar) (Bar) (l/h) (Bar) (l/h) (l/h) Flow 7* 4.6 2 1.4 340 2.6 112 228 0.67 8 5.1 1.9 1.7 376 3.2 87 289 0.77 9 6.1 2.1 2 431 4 91 340 0.79 10 6.2 2.2 2.6 425 4 99 326 0.77 11 7.3 3.2 3.8 419 4.1 124 295 0.70 12** 6 2 2.3 440 4 87 353 0.80 13** *Feed temperature was 86 C. **Feed TS content was 25% and Feed specific weight was 840 kg/m.sup.3.

(30) TABLE-US-00005 TABLE 4 Results of hydrocyclone tests with dry solids OF solids UF solids TS recovery Test TS % ml TS % ml % 7* 3.5 0.8 26 3.2 80 8 4 0.5 33 3.7 88 9 0.7 32 4 85 10 0.6 32 4 87 11 33 12** 32 13** 34 *Feed temperature was 86 C. **Feed TS content was 25% and Feed specific weight was 840 kg/m.sup.3.

(31) From the initial experiments presented in Table 2 to Table 4 it is evident that The cyclonette operates stably with consistent results at the tested concentrations and pressures The cyclonette has a capacity at a delta P of 4 Bar of: at 0% TS, 380 l/h at 14% TS, 430 l/h at 25% TS, 440 l/h; The maximum dry solids concentration in the underflow was 32-33%; The dry solids concentration in the overflow was 3.5-4%; The recovery of solids in the underflow was 85-88% of the dry solids; Fat recovery was calculated to be 94% with 4 multi-cyclone washing stages; Fat recovery was calculated to be 97% with 6 multi-cyclone washing stages.

Example 2Simulations of Processes

(32) FIG. 2 shows a schematic process diagram of a five stage process of the invention. In the process of FIG. 2 the particulate fat containing starting material (Feed (fatty meal)) is mixed with ethanol (alcohol) in a mixing tank (Mixing and maceration) and led to a colloid mill (Disintegrater) for further comminution. The mixture is fed to the hydrocyclone of the first stage (W1). The underflow of the first hydrocyclone is fed to the hydrocyclone of the second stage (W2), and the underflow and overflow streams of the second stage hydrocyclone are processed as described above in the subsequent stages of hydrocyclones, W3 to W5, respectively. The overflow of the first hydrocyclone W1 is clarified in a two-step hydrocyclone separation process where the overflow stream of the first clarification hydrocyclone C1 is fed to the second clarification hydrocyclone C2. The underflow streams from clarification hydrocyclones C1 and C2 are combined with the feed stream fed to the first stage hydrocyclone W1. The overflow stream of clarification hydrocyclone C2 is processed further by cooling to separate a fat phase from an ethanolic phase in a phase separator (not shown). The lower phase of the separator may optionally be further processed to provide the fat rich product, and the ethanolic upper phase may be recycled, optionally after dehydrating, to be mixed with the particulate fat containing starting material. The underflow stream from hydrocyclone W5 is subjected to a decanter centrifuge (Decanter) to further concentrate the fat lean product (Discharge (protein)) from the W5 underflow. The overflow of the decanter is combined with the feed stream to hydrocyclone W5. The set-up also comprises a line for applying a washing liquid (Wash liquid (alcohol)) to the final stage hydrocyclone and/or to the W5 underflow. This washing liquid may be combined with the clarified stream from the decanter centrifuge. It is to be understood that while FIG. 2 only shows a single hydrocyclone for each stage, any number of hydrocyclones may operate in parallel at a single stage.

(33) The results obtained in Example 1 were employed to simulate multistage processes having 4, 5 or 6 stages using an empiric mathematical model used and verified in the starch washing industry. The multistage set-ups for the simulations were as illustrated in FIG. 2 except that the numbers of stages were as indicated below. Table 5 shows the default parameters that were used in all three simulations.

(34) TABLE-US-00006 TABLE 5 Defaults for process simulations F.L. rate 4 kg/kg TS Protein recovery in hydrocyclone 85 % of SSP in F Fat concentration OF 70 % of fat concentration UF Rest fat in protein 1 % of SSP Specific weight alcohol 800 kg/m3 Specific weight fat 900 kg/m3 Specific weight protein 1050 kg/m3
4 Stage Process
The simulations of the 4-stage process are summarised in the tables below.

(35) TABLE-US-00007 TABLE 6-1 Feed stream to W1 NEW FEED UF C2 UF C1 SSP in kg/h 700* 22 127 L in kg/h 3967 258 1146 T in kg/h 4667 281 1274 FLOW in m.sup.3/h 5.6 0.3 1.5 SSP in % 15* 8* 10* S.G in kg/m.sup.3 834 842 833 SLS in kg/h 300* 140 280 TS in kg/h 1000 162 408 TS in % 20.1 38.6 26.2 TOT in kg/h 4967 421 1554 *indicates a default value for the calculations

(36) TABLE-US-00008 TABLE 6-2 Feed streams streams of W1 to W4 F W1 F W2 F W3 F W4 SSP in kg/h 999 995 969 824 L in kg/h 11064 7813 7015 6382 T in kg/h 12063 8807 7984 7205 FLOW in m.sup.3/h 14.7 10.7 9.7 8.8 SSP in % 8.3 11.3 12.1 11.4 S.G in kg/m.sup.3 822 825 825 823 SLS in kg/h 1015 399 140 45 TS in kg/h 2015 1393 1108 868 TS in % 15.4 15.1 13.6 12.0 TOT in kg/h 13078 9206 8124 7250

(37) TABLE-US-00009 TABLE 6-3 Underflow streams of W1 to W4 UF W1 UF W2 UF W3 UF W4 SSP in kg/h 849 845 824 700 L in kg/h 2548 2121 1750 1488 T in kg/h 3397 2966 2574 2188 FLOW in m.sup.3/h 4.0 3.5 3.0 2.5 SSP in % 25* 28.5 32 32* S.G in kg/m.sup.3 854 860 867 866 SLS in kg/h 304 108 45 13.60 TS in kg/h 1153 954 868 714 TS in % 31.2 31.0 33.2 32.4 TOT in kg/h 3701 3074 2618 2201 *indicates a default value for the calculations

(38) TABLE-US-00010 TABLE 6-4 Overflow streams of W1 to W4 and C1 to C2 OF C2 OF C1 OF W1 OF W2 OF W3 OF W4 SSP in kg/h 0 22 150 149 145 124 L in kg/h 7111 7370 8516 5692 5265 4894 T in kg/h 7111 7392 8666 5841 5411 5018 FLOW in m.sup.3/h 8.9 9.2 10.7 7.2 6.7 6.2 SSP in % 0 0.3 1.7 2.6 2.7 2.5 S.G in kg/m.sup.3 800 805 810 809 807 805 SLS in kg/h 291 431 711 290 95 31 TS in kg/h 291 453 861 440 240 155 TS in % 3.9 5.8 9.2 7.2 4.4 3.1 TOT in kg/h 7402 7823 9377 6132 5505 5049

(39) TABLE-US-00011 TABLE 6-5 Fresh feeds and decanter centrifuge streams F.L. W.L. OF DEC UF DEC SSP in kg/h 0 0 0 700 L in kg/h 4000 4632 632 856 T in kg/h 4000 4632 632 1556 FLOW in m.sup.3/h 5 5.8 0.79 1.7 SSP in % 0 0 0 45* S.G in kg/m.sup.3 800 800 800 896 SLS in kg/h 0 0 0 8 TS in kg/h 0 0 0 708 TS in % 0 0 0 45.3 TOT in kg/h 4000 4632 632 1563 *indicates a default value for the calculations

(40) TABLE-US-00012 TABLE 6-6 Pump settings and number of cyclones per stage Feed pumps Cyclonettes Flow S.G. Delta P Power D10 m.sup.3/h kg/m.sup.3 Bar kW Pieces C1 10.7 805 4* 0.7 28 C2 9.2 810 4* 0.6 24 W1 14.7 822 4* 0.9 34 W2 10.7 825 4* 0.7 25 W3 9.7 825 4* 0.6 23 W4 8.8 823 4* 0.6 20 *indicates a default value for the calculations
5 Stage Process
The simulations of the 5-stage process are summarised in the tables below.

(41) TABLE-US-00013 TABLE 7-1 Feed stream to W1 NEW FEED UF C2 UF C1 SSP in kg/h 700* 22 127 L in kg/h 3967 259 1147 T in kg/h 4667 281 1275 FLOW in m.sup.3/h 5.6 0.3 1.5 SSP in % 15* 8* 10* S.G in kg/m.sup.3 838 844 836 SLS in kg/h 300* 143 294 TS in kg/h 1000 166 422 TS in % 20.1 39.0 26.9 TOT in kg/h 4967 424 1569 *indicates a default value for calculations

(42) TABLE-US-00014 TABLE 7-2 Feed streams of W1 to W5 F W1 F W2 F W3 F W4 F W5 SSP in kg/h 1000 999 995 969 824 L in kg/h 10435 7192 6439 6059 5750 T in kg/h 11435 8191 7434 7028 6574 FLOW in m.sup.3/h 13.8 9.9 8.9 8.4 7.9 SSP in % 8.7 12.2 13.4 13.8 12.5 S.G in kg/m.sup.3 826 831 832 832 829 SLS in kg/h 1071 465 196 93 34 TS in kg/h 2071 1464 1191 1061 857 TS in % 16.6 16.9 15.6 14.9 13.0 TOT in kg/h 12506 8656 7630 7120 6607

(43) TABLE-US-00015 TABLE 7-3 Underflow streams of W1 to W5 UF W1 UF W2 UF W3 UF W4 UF W5 SSP in kg/h 850 849 845 824 700 L in kg/h 2550 2130 1796 1750 1488 T in kg/h 3399 2980 2642 2574 2188 FLOW in m.sup.3/h 3.9 3.4 3.0 2.9 2.5 SSP in % 25* 28.5 32 32 32* S.G in kg/m.sup.3 862 869 877 877 877 SLS in kg/h 338 138 70 34 11.30 TS in kg/h 1188 987 915 857 711 TS in % 31.8 31.7 33.8 32.9 32.3 TOT in kg/h 3738 3117 2711 2607 2199 *indicates a default value for the calculations

(44) TABLE-US-00016 TABLE 7-4 Overflow streams of W1 to W5 and C1 to C2 OF C2 OF C1 OF W1 OF W2 OF W3 OF W4 OF W5 SSP in kg/h 0 22 150 150 149 145 124 L in kg/h 6479 6738 7885 5062 4643 4309 4263 T in kg/h 6479 6760 8035 5212 4792 4454 4386 FLOW in m.sup.3/h 8.1 8.4 9.9 6.4 5.9 5.5 5.4 SSP in % 0 0.3 1.9 2.9 3.1 3.3 2.8 S.G in kg/m.sup.3 800 806 811 811 809 808 807 SLS in kg/h 295 438 733 327 126 59 23 TS in kg/h 295 461 883 477 276 204 146 TS in % 4.4 6.4 10.1 8.6 5.6 4.5 3.3 TOT in kg/h 6774 7199 8768 5539 4918 4513 4409

(45) TABLE-US-00017 TABLE 7-5 Fresh feeds and decanter centrifuge streams F.L. W.L. OF DEC UF DEC SSP in kg/h 0 0 0 700 L in kg/h 4000 4632 632 856 T in kg/h 4000 4632 632 1556 FLOW in m.sup.3/h 5 5.8 0.79 1.7 SSP in % 0 0 0 45* S.G in kg/m.sup.3 800 800 800 912 SLS in kg/h 0 0 0 6 TS in kg/h 0 0 0 706 TS in % 0 0 0 45.2 TOT in kg/h 4000 4632 632 1562 *indicates a default value for the calculations

(46) TABLE-US-00018 TABLE 7-6 Pump settings and number of cyclones per stage Feed pumps Cyclonettes Flow S.G. Delta P Power D10 m.sup.3/h kg/m.sup.3 Bar kW Pieces C1 9.9 806 4* 0.6 26 C2 8.4 811 4* 0.5 22 W1 13.8 826 4* 0.9 32 W2 9.9 831 4* 0.6 23 W3 8.9 832 4* 0.6 21 W4 8.4 832 4* 0.5 20 W5 7.9 829 4* 0.5 18 *indicates a default value for the calculations
6 Stage Process
The simulations of the 6-stage process are summarised in the tables below.

(47) TABLE-US-00019 TABLE 8-1 Feed stream to W1 NEW FEED UF C2 UF C1 SSP in kg/h 700* 22 127 L in kg/h 3967 259 1147 T in kg/h 4667 281 1275 FLOW in m.sup.3/h 5.6 0.3 1.5 SSP in % 15* 8* 10* S.G in kg/m.sup.3 838 844 836 SLS in kg/h 300* 142 286 TS in kg/h 1000 165 413 TS in % 20.1 38.9 26.5 TOT in kg/h 4967 424 1561 *indicates a default value for the calculations

(48) TABLE-US-00020 TABLE 8-2 Feed streams of W1 to W6 F W1 F W2 F W3 F W4 F W5 F W6 SSP in kg/h 1000 1000 999 995 969 824 L in kg/h 11067 7826 7081 6745 6691 6382 T in kg/h 12067 8826 8080 7740 7660 7205 FLOW in m.sup.3/h 14.6 10.7 9.7 9.3 9.2 8.7 SSP in % 8.3 11.3 12.4 12.8 12.6 11.4 S.G in kg/m.sup.3 824 829 829 830 829 826 SLS in kg/h 1034 421 165 77 34 12 TS in kg/h 2034 1421 1165 1072 1003 835 TS in % 15.5 15.4 14.1 13.7 13.0 11.6 TOT in kg/h 13101 9247 8246 7817 7694 7217

(49) TABLE-US-00021 TABLE 8-3 Underflow streams of W1 to W6 UF W1 UF W2 UF W3 UF W4 UF W5 UF W6 SSP in kg/h 850 850 849 845 824 700 L in kg/h 2550 2132 1804 1796 1750 1488 T in kg/h 3400 2982 2654 2642 2574 2188 FLOW in m.sup.3/h 3.9 3.4 3.0 3.0 2.9 2.5 SSP in % 25* 28.5 32 32 32 32* S.G in kg/m.sup.3 862 869 877 877 877 877 SLS in kg/h 310 115 54 26 12 3.50 TS in kg/h 1160 965 903 872 835 704 TS in % 31.3 31.1 33.4 32.7 32.3 32.1 TOT in kg/h 3710 3097 2708 2668 2585 2191 *indicates a default value for the calculations

(50) TABLE-US-00022 TABLE 8-4 Overflow streams of W1 to W6 and C1 to C2 OF C2 OF C1 OF W1 OF W2 OF W3 OF W4 OF W5 OF W6 SSP in 0 22 150 150 150 149 145 124 kg/h L in kg/h 7111 7370 8517 5694 5277 4949 4941 4894 T in kg/h 7111 7392 8667 5844 5426 5098 5086 5018 FLOW in 8.9 9.2 10.7 7.2 6.7 6.3 6.3 6.2 m.sup.3/h SSP in % 0 0.3 1.7 2.6 2.8 2.9 2.9 2.5 S.G in 800 806 810 810 808 807 807 806 kg/m.sup.3 SLS in 296 439 724 306 111 51 23 8 kg/h TS in 296 461 874 456 261 200 168 132 kg/h TS in % 4.0 5.9 9.3 7.4 4.7 3.9 3.3 2.6 TOT in 7407 7831 9392 6151 5538 5149 5109 5026 kg/h

(51) TABLE-US-00023 TABLE 8-5 Fresh feeds and decanter centrifuge streams F.L. W.L. OF DEC UF DEC SSP in kg/h 0 0 0 700 L in kg/h 4000 4632 632 856 T in kg/h 4000 4632 632 1556 FLOW in m.sup.3/h 5 5.8 0.79 1.7 SSP in % 0 0 0 45* S.G in kg/m.sup.3 800 800 800 912 SLS in kg/h 0 0 0 2 TS in kg/h 0 0 0 702 TS in % 0 0 0 45.1 TOT in kg/h 4000 4632 632 1558 *indicates a default value for the calculations

(52) TABLE-US-00024 TABLE 8-6 Pump settings and number of cyclones per stage Feed pumps Cyclonettes Flow S.G. Delta P Power D10 m.sup.3/h kg/m.sup.3 Bar kW Pieces C1 10.7 806 4* 0.7 28 C2 9.2 810 4* 0.6 24 W1 14.6 824 4* 0.9 34 W2 10.7 829 4* 0.7 25 W3 9.7 829 4* 0.6 23 W4 9.3 830 4* 0.6 22 W5 9.2 829 4* 0.6 21 W6 8.7 826 4* 0.6 20 *indicates a default value for the calculations
Summary of Multi-Stage Simulations
The results with respect to recovery of fat relative to the particulate fat containing starting material are indicated in Table 9.

(53) TABLE-US-00025 TABLE 9 Fat recovery in multistage separation Number of stages Fat recovery (%) 4 95.1 5 95.5 6 97.0

(54) It is clear from the simulations that the process affords a high capacity. For example, with a volumetric flow of 5.6 m.sup.3/h each stage of hydrocyclone treatment requires from 18 to 35 cyclonettes depending on the number of the stage in the process. This number of cyclonettes can be easily implemented in a process using standardised equipment readily available.