GB2246362A - Process for treatment of olive oils - Google Patents

Abstract

The process comprises treating olive oil with a metal or metalloid oxide, to reduce the content of chlorophylls and other contaminants. The treatment may be performed in micellar phase in a non-polar solvent.

Description

PROCESS FOR TREATMENT OF OILS The present invention relates to a process for the treatment of edible oils, in particular to a process for the treatment of oils containing partially degraded natural products, and to products obtainable by this treatment.

Olive oil is widely used as a food oil. Olive oil is valued for its natural quality, characteristic flavour and a fatty acid composition which is both relatively healthy and suitable for frying use. However, olive oil is one of the most expensive edible oils due principally to the high costs of olive cultivation and harvesting, it is therefore important to make full use of byproducts from the olive oil production process.

In the production of olive oil, it is commonplace for the oil to be mechanically extracted from the fruit leaving a cake which still contains commercially useful quantities of oil. The expressed oils differ greatly in quality, ranging from high grade 'virgin olive oil' to a lower grade 'LAMPANTE' oil which is suitable, as the name suggests, for use in lamps. For the purposes of the specification the high grade oil will be referred to hereafter as "Type (A)" oil, and the lower grade as "Type (B)" oil. Type (A) oil is fit for consumption without further processing, whereas Type (B) oil can be refined for food use.The refining process removes almost all of the characteristic flavour of olive oil and it is commonplace for around 90% of Refined Type (B) oil to be blended with-around 108.of Type (-A).oil to produce what is commonly-sold as olive oil'., referred to herein as Type (A/B) oil.

It is known, (see Bailey's Industrial Oil and Fat Products, Vol 2, fourth edition, page 217) to extract the oil remaining in the cake with a solvent, such as hexane, and evaporate the solvent from this fraction to yield an olive oil by-product known as 'SANSA' or 'sulphur olive oil' (Bailey cit. ultra page 229). For the purposes of this specification, this SANSA oil will be referred to as "Type (C)" oil.

The cake contains about 10-12% of oil, most of which can be extracted with hexane and subsequently refined for foods use (Bailey cit. ultra, page 245). The quality of the resulting oil is affected by the storage time of the used cake and depending on this time the free fatty acid content of the extracted oil may vary from a minimum of 5% from fresh cake to from 10-40% with an average of around 20% free fatty acids. Due to the extraction process, gums, resins and relatively large amounts of chlorophylls may be present in these Type (C) oils.

Type (C) (SANSA) oil is offered as an edible oil having many of the properties of Type (A/B) oil, at a price intermediate between that of Type (A/B) oil and other vegetable oils. Type (C) oil is also used as a feedstock for soap manufacture.

A disadvantage of Type (C) oil is the presence of components derived from the olive fruit as mentioned above. In particular unstable or partially degraded components reduce the effective frying life of samples of the Type (C) oil due to the rapid development of an unacceptably dark colour and the production of offflavours. In addition, the use of Type (C) oil as a raw material by the food industry is limited by the poor colour of this oil and the frequent presence of offflavours both in the oil per se and in products containing it. While it is difficult to identify all of the components which have to be deactivated by refining it is believed that free fatty acids, gums and waxes, chlorophyll and terpene components and metal traces from the pressing must be removed or bleached.

For the reasons mentioned above the food use of Type (C) oil has been restricted to use of the bleached and rigorously refined oil as an olive oil extender, in much the same way that Type (B) oil is used. Known commercial formulations such as blends for kitchen use ('olio di sansa e di oliva') contain around 90% of bland, refined Type (C) oil and around 10% of a Type (A) ('virgin') oil as a flavouring agent. Such blends are not only believed to contain traces of bleached contaminants, but it is clear that many components of the oil may be altered by high temperature treatment during bleaching.

We have determined that the organoleptic quality of Type (C) oil, as regards both the colour and the taste can be markedly improved by treatment of micellar Type (C) oil in an non-polar solvent with a metal or metalloid oxide to reduce the content of chorophylls and other contaminants by removing these components.

We have also determined that the above mentioned process, and variants thereupon may be applied with advantage in the treatment of other olive oils than Type (C) oil. Consequently, for the purposes of the present specification "olive oil" is used to indicate both all types of olive oil unless the context demands otherwise.

It is known that.theorganolepticqualities of oils in general can be upgraded-hy treatment:ofa.solution of the oil in a non-polar solvent witha -metal or metalloid oxide in a column. as :described in US patent 3955004 and subsequently treating the oil with a bleaching earth after the solvent has been removed.

Surprisingly, we have determined that a colour improvement sufficient to yield an excellent product results from the treatment of olive oil according to the present invention.

According to a further aspect of the invention there is provided an edible oil comprising 50-90% oleic acid residues and less than 0.4 mg/kg chlorophyll.

The oil according to the present invention can be used as a relatively inexpensive frying oil either pure or in combination with other oils. One advantage of Type (C) oil is that like all olive oils it is rich in cis monounsaturated fatty acid residues. These are more stable to oxidation than poly-unsaturated fatty acids and thus give a longer frying life. Moreover, cis monounsaturated fatty acids are believed to be free of some of the risks to health associated with saturated fats.

In certain embodiments of the present invention is essential that the edible glyceride oil should be in solution in a non-polar solvent when being treated with the adsorbent. Suitable non-polar solvents are petroleum fractions, pentane, hexane, cyclohexane, heptane and the like, including mixtures of various nonpolar solvents. Particularly suitable is a petroleum fraction with a boiling point in the range Qf 63"-71"C.

The amount of solvent may vary from 10-95% by weight of the solution, and generally varies from 50-80% by weight of the solution. ..

As stated above, the edible glyceride oil, in- solution in a non-polar solvent,. is contacted with an;. adsorbent.

Although this contacting-could:be carried out in a simple stirring or perc9lati-ng..-rocess, sufficient contact time bering the governing factor, advantageously and hence preferably the contacting is carried out in a column which contains the adsorbent, more preferably as described in British Patent Specification No. 865,807 and French Patent Specification No. 990,704.

The temperature at which the contacting my be carried out may vary widely. For most practical purposes the temperature will lie between 0 and 60"C, and preferably between 10 and 40 C.

The amount of adsorbent to be used is dependent upon a number of factors, e.g. pore size of the adsorbent, type of adsorbent, thickness of the layer in the column, throughput in the column, and so on. In general the ratio of oil to adsorbent varies between 0.3:1 and about 20:1. The adsorbent to be used in the present invention are metal oxides and metalloid oxides, particularly alumina and silica. Particularly silicas in the form of silica gels are especially suitable in the present invention. Mixtures of silicas and aluminas may also be used, as well as mixtures of silicas or aluminas with other adsorbents, provided the amount of silica or alumina is predominant.

The silica or alumina must neither be too fine nor too coarse. In general such silicas or aluminas are used which have an average pore size above 30 A, preferably between 50 and 2000 A. Suitable examples of aluminas and silicas to be used in the present invention are aluminas such as gibbsite or bayerite, and silicas such as silica gels known under the trade names Sorbsil (Joseph Crosfield and Sons, Warrington, UK) and Kieselgel M.(Fa.

Herrman, Cologne, Germany).

Other examples of suitable-aluminas--or-silicas are Aluminiumoxid 504C (Fluka AG, -Buchs, Switzetland), Kieselgel No. 7734 (E.. Merck, Darmstadt, Germany) and Silica Gel Type 62 (Grace GmbH, Bad Homburg v.d.H, Germany).

After contacting with the adsorbent, the edible glyceride oil is subjected to a treatment with bleaching carth. This treatment is carried out under the conditions for a bleaching treatment with a bleaching earth. Such conditions are: temperature generally from 50 to 1200C, pressure generally from 1 to 760 mm Hg, amount of bleaching earth generally from 0.05 to 5%.

Various types of bleaching earths may be used, mostly activated fuller's earth. Suitable examples are commercial products like Tonsil ACCFF, Tonsil 60 C (Fa.

Südchemie, Munich, Germany), furthermore Fulmont C300 (Laporte Industries Ltd., Redhill, U.K.).

It has been found that in the process of the invention less bleaching earth is required than in conventional bleaching processes. Preferably the amount of bleaching earth is from 0.05 to 0.5%. The bleaching earth treatment may be carried out directly after the oil has ben contacted with the adsorbent, or may be carried out after the oil has undergone further refining treatments such as desliming, neutralization and the like. The latter will take place e.g. when a crude oil is contacted with the adsorbent, In this respect it has been found that if silica gel is used as the adsorbent material, a pre-neutralization step is often superfluous, as the silica gel adsorbs practically all free fatty acids present in the crude oil, when only low quantities of free fatty acid are present.

Silica gel has, moreover, the further advantage over e.g. alumina that it hardly adsorbs the tocopherols which are useful as -.antioxidants -and -vitamins in the edible glyceride oil, It is essential that~~the < -trestment with bleaching earth is carried out when the solvent (when present) has been removed from the oil, e.g. by distillation. In carrying out the process of the invention clear edible oils are obtained with improved storage properties, particularly with regard to a significantly reduced tendency to form off-flavours on prolonged standing. In comparison with conventional refining treatments, the fatty acid composition of the glyceride oil remains practically unaltered and particularly the amount of essential fatty acids remains unchanged.

Oils according to certain embodiments of the present invention are sufficiently colourless to be used in the production of cosmetic creams, ungents and medicinal products. With these products it is often of particular importance that the colour of the product is neutral.

We have also determined that an improvement is obtained if the oil is stirred with the metal or metalloid oxide prior to filtration.

While stirring with the metal or metal oxide is less effective than treatment in a packed column, the stirring method requires much less metal or metalloid oxide. This is important as the column method is very expensive and in some circumstances can double the price of the column-refined edible oil as compared to the bleached oil of the prior art.

Preferably, 5-20% metal or metalloid oxide by weight of oil is stirred with the oil. This compares very favorably with the column method in which a given column weight of metal or metalloid oxide can be used to refine about twice its weight of oil..

Good results were obtained with l0*wt-silicå at ambient temperature in a micellar mixture and with 10% silica in solvent-free oil at l0O'C.

In order that the invention may be further understood it will be explained hereafter by way of example.

EXAMPLE 1: FRYING BEHAVIOUR A sample of Refined Type (C) oil (ex San Giorgio) was subjected to repeated frying cycles over a period of 1.5 days, according to the following procedure.

400 g of the Refined Type (C) oil was heated in a stainless steel dish to 1800C over a period of less than 20 min. 1.5 ml of demineralised water was pipetted into the hollow space of a cellulose stopper (Steri-Stopfen No. 9/weich ex. Heinz Herenze Gertigstrasse 14, Hamburg) and a stainless steel spit was introduced into the opening. A second stopper is prepared in the same way and these are both inserted into the hot oil. Frying is simulated for 20 minutes and the stoppers replaced by new and freshly moistened stoppers. After 5 such simulated frying operations the hot oil is allowed to cool to ambient temperature. The cycle is repeated with five more stoppers, and the entire process is repeated for a further half day (a total of 15 cycles).

At the places in the cycle indicated in table 1, samples of the oil are analyzed. The following tests were performed.

a) "Total Polar Components" test. This test identifies the percentage by weight of components which are not eluted on a chromatography column of silica gel with petroleum benzene and diethyl ether as solvents.

The method is specified in HDGF El' nheitsmethode C-Ill 3b/81".

b) "Lovibond Colour Test" :( industty standard method).

c) "Foaming Test": as given by Becker and Rost, Fette, Seifen, Anstrichmittel 66, 123-132, 1964 d) "Oxi-Fritest" (RTM) as supplied by E. Merck, Postfach 4119 Darmstadt 1, article number 10650.

In addition an evaluation of the odour at 1800C was made by an experienced oil-sniffer.

A similar series of experiments was performed with Refined Type (B) oil.

The frying behaviour of these oils (Type (C), and commercially Refined Type (B)) were compared with the frying behaviour of a silica-column Refined Type (C) oil. The results are given in table 1.

Toble 1.

Total Polar Comp.(%) Colour Lov. 2" cell Fooming Oxilritest 180 C Odour at 180 C Yield ml/top ml Samples % Cydes Silica coiumn Yellow+Red+Blue 4.2 . 18 1 Rubbery, shorp note, green 0 20+2.0+0.3 4.8 . 24 2- strongly perfumes, lightly pungent 1A Untreated Type (C) 5 9.4 20+1.0+0.1 5.5 . 20 3 strongly pungent, terpenes# 10 14.2 20+1.1+0.1 8.4 . 23 4 very pungent 15 23.9 31+1.7+0.2 4.5 . 21 1 neutral, hot oil Treated Type (C) 87 0 1.5+0.0 6.1 . 25 1+ lightly pungent, harshy## 1B silica column 5 6.7 4+0.4 7.5 . 22 2 pungent bleaching 0.5% T St FF. 10 16.1 5+0.5 8.6 . 23 3 strongly pungent deod. 230 C/3h 15 22.9 10+0.6 4.5 . 20 1 hot oil, neutral; 0 2.6+0.2 5.1 . 23 1 perfumes, terpenes, harshy 1C Refined 5 13.5 2.8+0.2 6.2 . 25 2- pungent Type (B) 10 21.4 3.0+0.3 8.1 . 26 2+ very pungent, sweet note, stronger 15 28.4 5.0+0.3 than SiO2 refined SANSA Refined Type (B) 0 0.2+0.0 4.3 . 18 1 hot oil, neutral; 1E stirring 20% activated Si02 77 5 8.0 0.7+0.0 5.4 . 19 1 hot oil, neutral at 100 C-bleaching 0.5% 10 14.6 1.2+0.0 6.9 . 23 1+ hot oil, very little sharpe note Tonsil St Ff-deod. 230 C/3h 15 23.1 3.0+0.3 2- hot oil, lightly pungent It can be seen that although the total content of polar materials in the untreated sample Type (C) oil (1A) was initially (5 cycle) quite low (9.4%), this rose markedly during the subsequent frying cycles. In the case of the silica-refined Type (C) oil (1B) the initial (5 cycle) value was lower (6.78) but also rose.

A much clearer difference is seen in the Lovibond colour test (the test was performed in a 2 inch cell). With the untreated Type (C) oil (1A) the final Lovibond colour was 31 yellow, 1.7 red and 0.2 blue whereas the final colour after cycling of the treated Type (C) oil (1B) was 10 yellow, 0.6 red, 0 blue. The lower red and blue scores indicate that the treated Type (C) oil has not darkened as much as the untreated Type (C) oil. The lower blue and red scores are particularly important.

Foaming is in all cases comparable with Refined Type (B) oil, given at (1C) as a comparison.

The "Oxifritest" (RTM) shows that components of the treated Type (C) oil (1B) are oxidised to a lesser extent than those of the untreated Type (C) oil (1A).

The subjective assessment of odour indicates clearly that while the untreated Type (C) oil (1A) is unsuitable for long-term frying, the treated Type (C) oil (1B) is slightly better in quality than the Refined Type (B) oil (1C) used in the comparison, as regards the production of polar components and the development.of odour. It should be noted that the-Refined- Type-(B) oil is not the Type (A/B) blend whidh is commercially available.

The example (1E) in table 1 illustrates the effect of the method of the present invention when applied to Refined Type (B) oil. While the production of polar components in oil treated according to the present invention (1E) is reduced as compared to commercially available Refined Type (B) oil (1C) there is a marked improvement in colour development on frying and a significant improvement in odour development.

EXAMPLE 2: SILICA TREATMENTS: Several silica-refining methods for olive oil were compared. The results are given in table 2. Colour Lov. 2" Unsapon- Chlor Time cell end product Yield ifiable ophyll Oil/Treatment Min. Yellow+Red+Blue (%) % mg/kg 2A/1A Refined Tyoe (C) no treatment 20.0+2.0+0.3 1.36 0.9 2B Refined Type (C) Silica column 4.0+0.4+0.0 87 1.24 0.0 1:2 (M/M) 2C/1B Refined Type (C) Silica column 1.5+0.0+0.0 1:2 (M/M) miscella 2:1 bleaching 0.5% T STFF deod. 3 h - 230 C 2D Refined Type (C) Regenerated 6.0+0.6+0.0 85 1.16 0.0 silica column miscella 2:1 2E Refined Type (C) Miscella 2:1 30 13.0+1.0+0.0 96 1.29 0.2 stirred 10% silica ambient 2F Refined Type (C) Miscella 2: :1 30 13.0+1.3+0.0 96 stirred 10% regenerated silica ambient 2G Refined Type (C)stirred 5% 30 20.0+1.6+0.? 96 silica ambient 2H Refined Type (C) stirred 5% 30 20.0+1.1+0.0 97 1.20 0.4 silica 100 C 3A Type (B) (Tunisia) No treatment 40.0+2.0+0.4 3B Type (B) (Tunisia) stirred 10% 60 21.0+1.2+0.0 90 silica 100 C 3C Type (B) (Spain) No treatment 40.0+5.0+2.2 3D Type (B) (Spain) straight, 40.0+2.3+0.3 but filtered 3E Type (B) (Spain) stirred 10% 60 26.0+1.4+0.0 90 silica 100 C 1C Refined Type (B) 2.6+0.2+0.0 0.2 0.85 1D Refined Type (B) stirred 20% 30 0.9+0.0+0.0 77 silica 100 C 1E Refined Type (B) stirred 20% 30 0.3+0.0+0.0 silica Bleached 1/2% Tonsil deodorise 240 C 3 hrs 0.0 It can be seen that the quality of the untreated Refined Type (C) oil is again shown to be poor by comparative example (2A), which oil is identical to that used in example (1A).

In example (2B) the Refined Type (C) oil was treated as a micellar system in hexane in a method similar to that of US 3955004. The Refined Type (C) oil used here was identical to the Refined Type (C) oil used in example 1B before bleaching and deodorisation.

In example (2C) the Refined Type (C) oil was silica treated by the method of US 3955004 including a bleaching with 0.5% of a commercial bleaching earth and deodorization for three hours at 230"C. A very good quality oil resulted from this trial. This oil was identical to that used for example (1B).

Example (2D) indicates that a regenerated silica column gives reasonable results. The use of regenerated columns is one way of extending the lifetime of the silica and reducing the cost of the process.

Examples (2E) and (2F) relate to the use of the stirring technique as mentioned in the preamble. It can be seen that while the Lovibond scores are not as good as those obtained by the column method they are improved over those of the untreated Type (C) oil.

Comparative examples (2G) and (2H) are comparative examples which illustrate-that the stirring treatment does not have a particularly marked effect on the colour when performed in the:absence.of solvent on Refined Type (C) at lower temperatures,t.ut- that at -elevated temperatures, in this case at-1000C, there is an improvement in the quality of the treated oil.

Comparative example (3A) and (3C) indicate the properties of Type (B) (unrefined LAMPANTE) oil from two sources; Tunisia and Spain. It can be seen that the products have a very high "Lovibond" yellow score.

Examples (3B), (3D) and (3E) show that this is not reduced by filtering (3D) (a conventional treatment step) but is reduced by half on silica treatment: (3B), (3E). Significant improvement is made to the "red" and "blue" scores by filtering.

Comparative example (1C) (repeated in table 1) together with embodiments (1D) and (1E) indicate the effect of the method of the invention on Refined Type (B) (refined LAMPANTE). It can be seen that the colour is improved in examples (1D) and (1E).

Table 2 also shows the chlorophyll levels for a number of the examples. It can be seen that the chlorophyll levels, repeated in table 3 below are reduced to below 0.4 mg/kg by the embodiments of the method aspects of the present invention.

Table 3 Expl. Example type Cholorphyll content 1A/2A comparative 0.9 mg/kg 2B embodiment 0.0 2D embodiment 0.0 " 2E embodiment 0.2 2H comparative 0.4 1C comparative 0-.85 1E embodiment 0.0

Claims (8)

1. CLAIMS 1. Process comprising treatment of olive oil with a metal or metalloid oxide to reduce the content of chorophylls and other contaminants by removal of these components.
2. 2. Process according to claim 1 wherein treatment is performed in micellar phase in a non-polar solvent.
3. 3. Edible oil comprising 50-90% oleic acid residues and less than 0.4 mg/kg chlorophyll or degraded components thereof.
4. 4. Edible oil according to claim 3 comprising less than 0.3 mg/kg chlorophyll or degraded components thereof.
5. 5. Use of metal or metalloid oxide treated olive oil as a component of a fat mixture for deep frying.
6. 6. Process according to claim 1 olive oil is stirred with the metal or metalloid oxide prior to a filtration step.
7. 7. Process according to claim 6 wherein the level of metal or metalloid oxide employed is 5-20% by weight of oil.
8. 8. Process according to claim 1, 6:or 7 wherein the oil is Type (A) or Type (B) or Type- (C) oil as defined herein.