Process for Reducing Contaminants in Glyceride Oils
FIELD OF THE INVENTION
The present invention pertains to a composition and method for treating edible glyceride oils to remove contaminants, chiefly free-fatty acids (FFA) . The composition and process may be applied either to rejuvenation of used oils or refining of crude edible glyceride oils.
BACKGROUND OF THE INVENTION
Edible glyceride oils, at various stages of their production and use, contain variable amounts of non-glyceride impurities. In refining crude edible oils, these impurities must be removed through the refining process. In used glyceride oils, these impurities build up as the oil is used, and if removed will increase the useable life of the oil. In the refining process, these impurities influence both the way the oil responds in the various processing steps employed to produce a finished product, and the yield of finished oil. In the reuse of used oil, an increase in impurities can degrade the oil, which can adversely affect its taste and shelf-life and may increase its ability to be absorbed by foods. Accordingly, it is desirable to remove these impurities at whatever stage of oil production and use they occur, whenever possible.
Table 1
Table 1 shows some of the impurities contained in crude glyceride oils, which can be removed by the refining process. In the refining process, the oils are treated with caustic soda in the primary steps. The caustic soda forms a flocculant precipitate of soaps which settle out as "foots." The addition of an alkali solution to crude or crude degummed oil results in chemical reactions and physical changes. The alkali combines with free-fatty acids in the oil to form soaps. The phosphatides and gums absorb alkali and are coagulated through hydra ion or degradation. Much of the coloring matter is degraded and absorbed by the gums, or made water soluble by the alkali. The insoluble matter is entrained with the other coagulated material. The soap-oil mixture is then heated to about 160-180°F (75- 82°C) and fed through a centrifuge for separation into
light and heavy density phases. The light phase comprises chiefly refined oils including traces of moisture and soap. The heavy phase is primarily soap, insoluble matter, free caustic, phosphatides, and 5-9% of neutral oil.
The refined oil (light phase) is discharged from the centrifuge, heated to 190°F (88°C) and mixed with soft water that has been heated to 200°F (93°C). The water-oil mixture passes through a high-speed shear mixer to obtain intimate contact between the oil and water phases for maximum soap transfer from the oil to the water. The mixture next passes through a second centrifuge where the phases are separated. The water- washed oil is discharged as the light phase, and the soapy water as the heavy phase. The water-washing process removes about 90% of the soap content of the refined oil. The remainder of the soap is removed by a subsequent bleaching process.
Caustic refining will remove phosphatides, free-fatty acids (FFA) , and some pigments. However, the oil still contains color bodies, odors, metals, high levels of soaps, and various impurities which need to be removed before the finished oil will be of acceptable color and taste to the consumer. A bleaching process where clay and/or silica hydrogel is added to the oil,
and subsequently removed by filtration, may be used to further reduce the remaining impurities.
In conjunction with the refining process, many different adsorbents have been used, including: silica hydrogel (U.S. Patent 4,629,588); silica hydrogel treated with an organic acid (U.S. Patent 4,734,226); high surface area amorphous silica treated with a strong acid (U.S. Patent 4,781,864); partially dried silica gel (U.S. Patent 4,880,574); bleaching absorbent and phosphoric acid (U.S. Patent 3,895,042); silicon dioxide, aluminum oxide or mixtures thereof (U.S. Patent 3,955,004); activated carbon impregnated with MgO (U.S. Patent 4,125,482 and U.S. Patent 4,150,045); bleaching clay and an alkaline earth metal, lanthanide, or transition metal exchange zeolite Y (U.S. Patent 4,443,379); silica gel and silicic acid (U.S. Patent 4,874,629); metal oxide silica absorbent (U.S. Patent 4,956,126); sodium silicate solution combined with phosphoric acid (USSR patents 992,564-A, 1,386,642-A, 1,148,861-A, and 806,750-B) . Similar methods and adsorbents have been used for removing contaminants from used cooking oils as well.
Although many compositions and methods have been tried, the removal of FFA continues to be a problem. For instance, silica hydrogel and similar compositions, while addressing the removal of soap and color bodies, have not been shown to reduce FFA, as shown in Table 2.
Table 2
Rapeseed Oil %FFA Color- ovibond Soap-Metals (pp ) Red Yellow P Ca Mα Fe
Untreated1 1.2
2.0% ABE2 1.1
0.24% SH3/1.3% ABE 1.13 0.35% SH3/1.0% ABE 1.12
-1 Crude degummed rapeseed oil Activated Bleaching Earth
3 Silica Hydrogel
Accordingly, it would be desirable to reduce the levels of FFA in glyceride oils, as well as reducing color bodies, soaps, and trace metals (Ca, Mg, P, Cu, and Fe) .
SUMMARY OF THE INVENTION
The present invention provides a composition and method for treating edible glyceride oils to remove contaminants therefrom. The composition can be added directly to crude oil, degummed oil, or used oil to reduce FFA, color bodies, trace metals, and other impurities. The composition comprises solid hydrous alkali metal silicates, particularly sodium metasilicate pentahydrate and hydrous sodium polysilicate. This material is added to oil containing contaminants, in an
amount approximately equal to the amount of FFA present in the oil after a small amount of water is added to the oil. The oil may then be heated and agitated. The oil is then filtered or centrifuged to remove the solid hydrous sodium silicate, and vacuum dried, if appropriate, to remove residual water.
DETAILED DESCRIPTION OF THE INVENTION
The type and levels of contaminants present in glyceride oils depend on a number of factors, including whether the oil is crude, whether it has been degummed, and if used, what foods were fried in the oil. Some crude oils, like soybean for example, can have only about 0.7% FFA, while other oils like palm oil, have around 5.0% FFA. Accordingly, in removing FFA, the amount of treating agent (solid hydrous sodium silicate) used should depend upon the amount of contaminants in the oil. It is preferred to use a 1:1 ratio of solid hydrous sodium silicate to the FFA content of the oil on a weight basis. Thus for 100 gms. of palm oil with 5.0% FFA, 5 gms. of solid hydrous sodium silicate would be used as a treating agent.
The oil and solid hydrous sodium silicate should react at elevated temperature. The oil may be heated before (or after) the addition of the solid hydrous sodium silicate. The temperature to which the
oil should be heated will depend upon the processing that the oil has previously received. Crude oils tend to discolor when heated to temperatures over 220°F because of color reactions and phospholipids. Once color bodies are removed, higher temperatures may be used without adversely affecting the oil. For instance, refined oil must have the ability to be heated to 350° or higher in order to withstand the temperatures needed to fry foods.
The temperature to which the oil is heated also depends on what treating agent is used. Sodium metasilicate pentahydrate has a melting point around 162°F (although experiments indicate it is stable in oil at temperatures as high as 220°F) . Therefore oils treated with sodium metasilicate pentahydrate should be heated to a lower temperature than oils treated with hydrous sodium polysilicate, which has a much higher melting point.
Table 3 shows the effect of varying the temperature of the oil on FFA removal when using hydrous sodium polysilicate and sodium metasilicate pentahydrate. As may be seen from the table, hydrous sodium polysilicate is not as effective at removing FFA at low temperatures as sodium metasilicate pentahydrate. However, hydrous sodium polysilicate is useful and effective for removing FFA at temperatures well above the melting point of sodium metasilicate pentahydrate. Therefore, sodium metasilicate pentahydrate is recommended for refining where the oil begins cold and must be heated, since sodium metasilicate pentahydrate allows processing at lower temperature, thereby effecting
a cost savings in heating the oil. Hydrous sodium polysilicate is useful in removing contaminants from used oils, since they are usually at an elevated temperature at which sodium metasilicate pentahydrate would melt. The hydrous sodium polysilicate does not require cooling the used oil and possibly reheating the oil for use after cleaning.
Sodium silicates in general are combinations of sodium oxide (Na2θ) and silicon dioxide (Siθ2) . They may or may not have water chemically bound within them. Sodium polysilicate, for instance, has the formula (Na2θ)x(Siθ2)y*zH2θ where the weight ratio y:x is greater than 2.0 but less than 2.40. The material is generally about 17.5% water on a weight basis when it is hydrous. Specifically tested was a sodium polysilicate sold under the name BRITESIL® C20 (BRITESIL is a registered trademark and BRITESIL products are available through the PQ Corporation, P.O. Box 840, Valley Forge, Pennsylvania 19482). BRITESIL® C20 has a Si02:Na20 ratio of 2.00. BRITESIL® C20 sodium polysilicate is amorphous, has a bulk density of 50 lb/ft3 (.80 g/cm3) , and is 17.5% H20 by weight.
The sodium metasilicate evaluated for performance in removal of contaminants from edible glyceride oils had the general formula
(Siθ2)y(Na2θ)x*zH2θ where the y:x ratio was less than or
equal to 1, and z equaled 5. This makes the sodium metasilicate 42% water. The particular sodium metasilicate tested was METSO PENTABEAD® 20 available from the PQ Corporation (METSO PENTABEAD 20) . The molar ratio of Na2θ:Siθ2 was 1:1, making the composition 29.3% Na20, 28.4% Si02, and 41.6% H20. METSO PENTABEAD 20 has an approximate bulk density of 49 lb/ft3 (0.78 g/cm3) .
The traditional refining process for edible glyceride oils generally begins with a preheating step to heat the oil to treatment temperature. Once hot, the oil is treated with H3PO4 and centrifuged. This treatment turns non-hydratable (unreactive with alkali) phospholipids to hydratable, so they can be removed by the refining process. This treatment is referred to as degumming. As previously described, once the oil has been degummed, diluted caustic (NaOH) is added to neutralize or remove FFA from the oil. The FFA react with the sodium hydroxide to form soaps. The sodium hydroxide is a solution in water, and soaps are contained in the aqueous phase. One problem with this, however, is that it requires several steps to remove the residual alkaline materials. Once the caustic is added, the oil is heated to allow the caustic to react. The oil and water mixture is then centrifuged and the oil is separated from the water and soap phase. The oil is then washed with water again to remove residual caustic and soaps in the oil phase. After washing with water, the
oil must once again be centrifuged and separated. Of course, since soap is an emulsifier, the amount of oil lost in this traditional method can be high. Once the oil is separated, it is vacuum dried to remove any residual water. The oil may then be bleached to help neutralize or remove color bodies.
By using the hydrous solid alkali metal silicate of the present invention, the caustic step in the refining process may be eliminated. After the degumming with phosphoric acid, water is added to the oil in an amount based upon the FFA level. Generally water should be added in an amount of between l.7 and 2.1 times the weight of FFA present in the oil. A ratio of water:FFA of 1.9 is preferred. If too much water is used, the soaps which form will be thin, and will therefore absorb more oil, leading to higher oil losses. Conversely, if too little water is added, the soaps formed will be too thick, making separation difficult.
Solid alkaline hydrous sodium silicate is then added to the oil. To enhance the removal of FFA, the oil may be heated and agitated. The oil may then be centrifuged or filtered, depending upon the amount of sodium silicate added to the oil. When greater amounts of sodium silicate are added, filtering becomes increasingly slow. Therefore if large amounts of sodium silicate are to be added, centrifugation is the preferred
method of separation. The oil may then be washed with water a second time, separated, and vacuum dried to remove residual water.
Some oils like olive and almond oils have a low phosphorus content. In those oils, the step of adding phosphoric acid in order to remove phospholipids from the oil can be eliminated since the sodium silicates will remove some phospholipids. However, the remainder of the refining process is the same, and this process is improved by using solid alkaline hydrous sodium silicate instead of a solution of caustic in that the amount of water added to the oil is reduced.
In rejuvenating used cooking oils, a treating agent is generally added directly to the used cooking oil in the fryer or a separate treatment vessel. The oil is then filtered to remove the treating agent and returned to the fryer to be used. Generally this operation is performed while the oil is hot. The hydrous alkali metal silicate described and claimed herein is useful as such a treatment agent for used cooking oils, either alone or in combination with other rejuvenating compounds.
As a used cooking oil rejuvenator, hydrous sodium polysilicate is generally more useful than sodium metasilicate pentahydrate. As previously explained, the sodium metasilicate pentahydrate used in the tests
conducted for this invention has a melting point of about 162°F (72.2°C), although this sodium metasilicate did not melt in oil until the temperature rose to about 240°F. Nevertheless, oil in fryers is generally at a temperature of around 350°F. This temperature is too high for sodium metasilicate pentahydrate to be useful, since it would melt upon contact with the oil. Therefore, hydrous sodium polysilicate with a higher melting point is more useful for used cooking oil rejuvenation. The sodium metasilicate pentahydrate, while effective in removing contaminants, would require the cooling of the oil prior to treatment. This is generally undesirable in that it involves an additional processing step.
EXPERIMENTAL RESULTS
Both METSO PENTABEAD 20 (sodium metasilicate pentahydrate) and BRITESIL C20 (hydrated sodium polysilicate) were tested with various batches of crude edible glyceride oils to determine the conditions under which the oil should be processed to most efficiently remove FFA from the oil while minimizing undesirable soaps. The results of this test are shown in Table 4.
-r
UJ
_ι m
- sτ -
zOIMSM±Dά S9LIZIP6 0ΛV
The oils with which the compounds were tested can be seen at the top of the table labeled Crude 1, Crude 2, Crude 3, Crude 4, Crude 5, Crude 6, Crude Olive, and Crude Walnut. These represent six batches of crude soybean oils, and one sample each of crude olive oil and crude walnut oil. Only soybean sample 5 was degummed (treated to remove phosphates) prior to treatment with the sodium silicates.
As may be seen from Table 4, the METSO PENTABEAD 20 (sodium metasilicate pentahydrate) performed better with higher contact times, higher temperatures, and larger doses. This was expected, as each factor allows greater contact between oil and treatment agent. The exception to this general observation was sample 9 where METSO PENTABEAD 20 appeared to perform slightly worse with an increase in temperature from 200 to 220°F. Only one sample was tried at this higher temperature. The METSO PENTABEAD 20 even performed well where the initial FFA content was in excess of 5% (see samples 75, 77, 87, and 88) .
Sodium metasilicate pentahydrate of differing particle sizes was also tested and found to perform quite well. METSO (Fine), METSO (Medium), METSO (Granular), and METSO (Oversized) are all sodium metasilicate pentahydrate in fine, medium, granular, and oversize particle sizes. As may be seen from samples 50-74 in
table 4, these sizes of sodium metasilicate pentahydrate were effective in removing FFA from the oils.
Table 4 also shows that BRITESIL C20 was effective in removing FFA from the oils. This may be seen from samples 47-53, 76, and 78. It does not appear from these experiments that filtering or centrifuging the oil made any difference in the performance of the treatment agents. In real life, the time required to filter large amounts of oil may increase the contact time between the oil and treatment agent, thereby increasing the performance of the treatment agent.
An experiment was conducted to show the effect of solid hydrous sodium metasilicate and sodium polysilicate as compared with other treatment agents, in reducing FFA from edible glyceride oils. In this experiment, a single edible oil was used, with a starting FFA content of 1.65%. Several compounds were tested according to the method of the present invention, and the results show that the hydrated sodium polysilicate and sodium metasilicate pentahydrate clearly outperformed the other treatment agents tested. The results of this test can be seen from Table 5.
Both sodium metasilicate pentahydrate and hydrous sodium polysilicate were tested in olive oil.
Olive oil is more viscous than soybean oil, and therefore
agitation should be an important factor in FFA reduction. The results of this testing are shown in Table 6.
TABLE 6 Italian Olive Oil
* - 4% FFA added to the olive oil in the form of oleic acid. +A - These samples were given more agitation than the others. 1 - 50/50 mix of BRITESIL C20 and alumina
In all cases, the oil was filtered after treatment. The sodium metasilicate pentahydrate did not perform well without high agitation where the FFA content was low. However, sodium metasilicate pentahydrate did perform well where the FFA content was high, even without high agitation. It is reasoned that this performance is
due to the larger filtration bed developed where more treatment agent (four times the lower level) is used.
Trials were also conducted to determine the effect of treatment temperature on the effectiveness of sodium metasilicate pentahydrate and hydrous sodium polysilicate in removing FFA from glyceride oil. The oil used for the test was soybean oil, in an amount of 500 g. In several cases, oleic acid was added to the soybean oil to increase the FFA content, before treatment with the treatment agent. The results of these tests are shown in Table 7.
TABLE 7
Temperature Effect on Sodium Metasilicate Pentahydrate and Hydrous Sodium Polysilicate Oil Treatment
(500g Soybean Oil)
^-Filtration time was > 15 minutes.
2Filtration time was > 45 minutes.
3Filtration time was > 2 hours.
4Simultaneous treatment with BRITESIL C20 and Alumina
Sequential treatment with BRITESIL C20, then a mixture of 70% Silica hydrogel and 30% Alumina
In these samples, the addition of alumina, or silica gel and alumina, appeared to have little or no effect on the FFA reduction achieved by the hydrous sodium polysilicate. Generally, the hydrous sodium polysilicate performed better at higher temperatures, although at a treatment agent level of 3% the hydrous sodium polysilicate was quite effective at removing FFA even at temperatures as low as 150°F. The increase in temperature from 200 to 220°F did not appear to result in large changes in the performance of sodium metasilicate pentahydrate, in this series of tests.
Sodium metasilicate pentahydrate and hydrous sodium polysilicate were tested to gauge performance as compared to treatment with a solution of 10% sodium hydroxide, traditionally used to treat oil to remove FFA. The sodium hydroxide was added in an amount of .3% of the weight of the oil to be treated. The results of these tests are shown in Table 8.
TABLE 8
Ratio Color Pigments wt trtmt: Red Yellow Chlorophyll FFA Soap
vo
as
As may be seen from Table 8, both the sodium metasilicate pentahydrate and hydrous sodium polysilicate performed comparably to the sodium hydroxide in both crude and degummed oil, whether or not the oil was bleached. Thus the use of these treatment agents does not sacrifice the quality of the oil produced. Furthermore, an advantage of these treatment agents is that, compared to caustic refining, less soap remains in the oil and less neutral oil is lost in the soap stock. For example, a caustic treated oil was found to contain 319 ppm residual soap and yielded a soap stock that weighed 5.0% of the untreated oil weight. The same oil treated by sodium metasilicate pentahydrate gave 228 ppm and 2.1%.
Palm oil refined with METSO PENTABEAD 20 was also compared to oil refined in the traditional manner described previously. The results of this comparison are shown in Table 9. As may be seen, refining with METSO PENTABEAD 20 is comparable to refining in the traditional manner.
TABLE 9
METSO PENTABEAD 20 STUDY PALM OIL
FFA % SOAP COLOR
TREATMENT DOSE (%) Reduc. (ppm) PCI 520nm.
Untreated 2.52 <1 59.37 2.215
**
Refined Oil unkn. 0.04 98.4 <1 27.44 2.024**
METSO 0.02 99.2 <1 45.47 2.304 PENTABEAD 20
Dosage is 1:1 ratio of METSO PENTABEAD 20:FFA on a weight basis.
**
These color measurements reflect further removal of color bodies in this completely refined oil through bleaching and deodorization.
METSO PENTABEAD 20 sodium metasilicate pentahydrate was also tested to determine the amount of oil absorbed by the treatment agent. The results of this test are shown in Table 10.
Table 10
Oil Absorption Of METSO PENTABEAD 20
Average Type Of Oil % Oil Absorbed
Olive Oil 30.36%
Soybean Oil 38.53%
Cottonseed Oil 34.69%
These tests were conducted with used oils. Oil absorbed is a weight percentage based on the weight of treatment agent used. Thus the sodium metasilicate pentahydrate absorbed only 30 to 40% of its weight of oil, indicating low losses of oil in this type of treatment.
It is understood that various other modifications will be apparent to and can readily be made by those skilled in the art without departing from the scope and spirit of this invention. Accordingly, it is not intended that the scope of the claims appended hereto be limited to the description set forth herein, but rather that the claims can be construed as encompassing all of the features of patentable novelty that reside in the present invention, including all features that would be treated as equivalents thereof by those skilled in the art to which this invention pertains.