CA1300965C - Decolourising of whey and products derived from whey - Google Patents

Abstract

25.
ABSTRACT
"DECOLOURISING OF WHEY AND PRODUCTS DERIVED FROM
WHEY"
The invention provides a process of decolourising annatto-containing whey or products produced therefrom, by oxidation with hydrogen peroxide, characterised in that the oxidation is carried out in the presence of a catalytic amount of an active peroxidase, in that the amount of hydrogen peroxide used is sufficient to effect the oxidation of the annatto and in that the time for which the peroxidase is exposed to a deactivating concen-tration of hydrogen peroxide is insufficient to permanently deactivate the peroxidase. The process is generally applicable to inter aria whey, demineralised whey, reconstituted whey powders, whey protein concentrates and reconstituted whey portein concentrate powders. It may be applied as a batch, semicontinuous or a continuous in line process.

Description

i30096S
--1--`
DESCRIPTION
DECOLOU~ISING OF WHEY AND PRODUCTS DERIVED FRO~I~EY.
The present invention relates to the decolourising of whey and products derived from whey such as demineralisèd whey and whey protein concentrates.
The most commonly used material for dyeing cheese is annatto, a natural colouring material derived from the fruit bixo orellena. The presence of annatto, whilst desirable in the cheese is undesirable in products produced by further processing such as demineralization or concentration as by ultra-filtration techniques. Thus the use of annatto in cheese produces a coloured whey so that protein concentrates produced from such whey will also be coloured. This colourisation renders such secondary products less valuable than secondary products produced from uncoloured starting materials and not infrequently renders them unusable for some applications.
The decolourisation of coloured whey and secondary whey products ls thus very desirable. Thus whey and secondary products such as whey protein concentrates (WPC) which contain annatto must be decolourised if they are to have a value approaching that of uncoloured whey and secondary products produced therefrom. This problem is 8 longstanding one in the art.
Studies of the destruction of the much less stable compound ~-carotene in peroxidase/hydrogen peroxide/
halide systems ha~e been reported in Lipids Vol. 18 N~.3 (1983) at page 198 in connection with the study of the oxidation of lipids in foods. Since carotene is not normally soluble in aqueous syste~s substantial quantities of the surfactant Tween*80 were used to bring the carotene into aqueous solution. The ~ ade mark C' ~300965 destruction of,~-carotene was used as a sensitive method to determine the activity of the enzyme. In the said aqueous system, in the presence of optimal concentration of hydrogen peroxide the rate of ~-carotene destruction was reported ~s slowly increasing until a critical concentration of the halides was reached, followed by a rapid increase in the rate when the halide concentration was further increased.
This system was further studied by Ekstran~ and Bjorc~ and reported in J. Agric. Food Chem. 1986 34, 412-415. Such studies of artificially produced aqueous systems containing substantial quantities of Tween 80 however, have not provided any basis for the creation of a viable process for decolourising the much more stable annatto in whey, even on a l~boratory scale and still less on a commercial scale.
The only method which has been suggested as a decolourising process for whey is that proposed by McDonough et al. in J. Dairy Science Vol~ 51 No.3 pages 471 and 472 to decolourise annatto-containing Cheddar-cheese whey before drying by oxidation with benzoyl peroxide or hydrogen peroxide. The paper by McDonough et al states that the efficiency of decolourisation was determined by measuring the transmittance of light by clarified whey with A Bausch and Lomb*Model 340 Spectrophotometer at a wavelength of 600 m~ or in modern terms 600 nm~ Uncoloured wney and unbleached coloured whey were used as controls.
As described in more`detail hereinafter with reference to the accompanying drawings, annatto has a maximum absorbance in whey at 460 nm and has virtually no absorbance at a wavelen~th of 600 nm. It is impossible to determine therefore what relationship, if any, the transmission figures for clarified whey *Trade mark ~300965 reported by McDonough et al hava to the decolourisation of annatto by hydrogen peroxide.
~othin~ is said in the paper about the actual colour of the 'decolourised whey powders' mentioned in the psper as having been made by concentration in an evaporator under vacuum and freeze-drying. The paper also reports that catalase was added to the H202 containing lots to inactivate residual H202 indicating that even after 2 hours residual H202 remained in the whey. The whey used by McDonough et al was produced during laboratory or pilot scale Cheddar*cheese making and was pasteurized before use~
In attempting to apply the teaching of McDonough et al to the decolourisation of commercially produced annatto coloured whey it has been found that using 300 and 500 ppm of H~o2, in the concentrations recommended by McDonough et al little decolourisation is achieved with fresh p~steurized whey. It has also been found that even at a starting concentration of 100 ppm H2O2 fresh whey still contained residual H2o2 even after 4 hours, and little decolourisation had taken place.
McDonough et al reported that the decolourising effectiveness of hydrogen peroxide on annatto was reduced as the amount of curd particles ir. the whey increased and gave as the probable reason the inactivating effect of the natural peroxidases in the protein. They indicated that sufficient amounts of hydrogen peroxide must be added to QVercO~e the natural peroxidase before oxidation of annatto can be 3o effective.
It has now been surprisingly found contrary to the teaching of McDonough et al that the peroxidase catalyses the oxidation of annatto by hydrogen peroxide but is deactivated by high concentrations of hydrogen peroxide. Furthermore the deactivation ~ade mark i3~096S

effect has been found to depend both on the concen,ration of hydrogen peroxide and on the time the peroxidase is exposed to the deactivating concentration and that if the concentration of hydrogen peroxide present in the whey falls below the deactivating concentration for that whey before complete deactivation of the peroxidase has occurred decol~ourisation then occurs but at a slower rate.
In its broadest aspect therefore the present invention provides a method of decolourising annatto-containing whey~ or products produced therefrom, by ' oxidation with hydrogen peroxide, characterised in that the oxidation is carried out in the presence of a catalytic amount of an active peroxidase, in that the amount of hydrogen peroxide used is sufficient to effect the oxidation of the annatto and in that the time for which the peroxidase is exposed to a deactivating concentration of hydrogen peroxide is insufficient to permanently deactivate the peroxidase.
The peroxidase may be that inherent in the whey, -namely lactoperoxidase and/or may be added to the syste,m such as by the addition, for example, of a catalytic amount of active lactoperoxidase or an active peroxidase of similar enzymatic action such as horseradish peroxidase.
For any given whey there is an optimum amount of hydrogen peroxide which effects rapid pero*idàs`e'``` ``` '`
catalysed decolourisation of annatto in whèy. Below that opti~mum amount of hydrogen peroxide peroxidase catalysed decnlourisation of the annatto may not occur at all, or the-hydrogen peroxide is exhausted soon after the catalysed oxidation of the annatto ` ' commences. As the initial concentration of hydrogen peroxide in the whey is increased, for any given whey , J 35 there is an initial lag time before peroxidase catalysed decolourisation of the annatto commences.

. ;,~
. ' ' ~:~00~65 The extent of the time lag lncreases with increasing in`itial concen~ration of ~ydrogen peroxide. The rate of peroxidase catalysed decolourisation of the whey after the time lag decreases with increa~ing initial S concentration of hydrogen peroxide in the whey.
Without wishing to be bound by any particular theory, i~ is believed that more than one factor is involved. The annatto is believed to act as a substrate for the peroxidase. If other substrate(s~
are present in the whey they may compete preferentially for the attachment points or sites on the peroxidase and react with a part of the hydrogen peroxide. Furthermore, above a critical initial ~ concentration of hydrogen peroxide in any given case, the peroxidase catalyst is deactivated. This is believed to be due both to inhibition of the ability - of the peroxidase to act as a substrate base for annatto which is a reversible effect and to denaturing of the peroxidase whlch increases with time of exposure of the peroxidase to the hydrogen peroxide.
If the initial concentration of hydrogen peroxide is subsequently reduced, e.g. by reaction with other substrates, or by reaction with other materials in the whey, such as catalase, then if the concentration of hydrogen peroxide in the whey falls at a sufficient rate the activity of the peroxide will in general be restored at least in part. The rate of decolourisation will then be governed by the residual activity of the peroxidase to catalyse the reaction of annat~o with hydrogen peroxide which in turn will depend on the initial concentration of the hydrogen pero~ide and the time for which the peroxidase has been expvsed to a deactivating concentration of hydrogen peroxide. If that time is short the subsequent rate of decolourisation is high, that is to say, the activity of the peroxidase as a catalyst is ' .
.

1300g65 not ma~erially affected by a short exposure to an initial deactivating concentration of hydrogen peroxide. At high initial concentrations of H22 however, it may take some considerable time for the excess hydrogen peroxide to react with materials other than annat~o in the whey and in this time~ at least some of the peroxidase will be permanently deactivated.
If the initial concentration of hydrogen peroxide is so great that reactive materials in the whey are insuffi.cient to reduce the hydrogen peroxide ~
concentratlon sufficiently to allow restoration of the catalytic activi~y of the peroxidase then peroxidase catalysed oxidation of the annatto will not occur and decolourisation will proceed only very slowly or possibly no~ at all.
The concentration of hydrogen peroxide which effectively deactivates peroxidase is not a constant value but has been found to depend on the protein concentration in the whey or whey-derived product.
Thus reconstituted liquid whey protein concentrates, or reconstituted whey protein concentrate powders, having high protein contents compared to the original whey, have been found to tolerate higher initial concentrations of hydrogen peroxide before effective deactivation of the peroxidase takes place.
As already stated, it is believed that the annatto acts as a substrate for the peroxidase so that if any other materials are present which also--act as substrates they may compete pre~erentially with the-3o annatto for he peroxidase and hydrogen peroxide.Where such materials are present the amount of hydrogen peroxide added to-the whey or whey-derived product must be sufficient to accommodate both such substrate(s) and the annatto. An example of such a preferentially competing substrate is the thiocyanate ion which is present in bovine milk. The i30~19SS

concentration of thiocyanate ion is substantially rèduced by, for example, demineralisation or ultrafiltration and prevents less or no real problem in demineralised whey or whey pro.ein concentrates.
Increasing temperature increases the rate of peroxidase catalysed decolourisation of annatto in whey up to the temperature at which deactivation of the peroxidase takes place. In general, the temperature used will depend on the processing conditions. Thus for example, in the processing of whey to produce for example a whey protein concentrate, the point of injection of hydrogen peroxide into the whey being processed will be material in terms of the temperature which can be used. Where the whey can, as a normal practice, be held for fairly long periods, quite low temperatures of 5C to 10C could effectively be used. Where, because of the point of in~ection of the hydrogen peroxide, rapid decolourisation i8 necessary or required, higher temperatures would then be necessary and for the maximum rate of decolourisation a temperature of between 55C to 60C would be preferred. Higher temperatures might be used for very short periods, but at such higher temperatures, e.g.
above 65C the whey could not be held at such a temperature for very long and certainly not long enough to deactivate the catalyst.
The peroxidase must be present in an active state to catalyse the oxidation of annatto by hydrogen peroxide. If ion exchange is carried out on the whey prior to decolourisation, it has been found that a catalytic amount of active peroxidase must be added to the treated whey in order to effect decolourisation.
It is believed that the extremes of pH which occur in - 35 the whey during ion exchange deactivates any naturally occurring peroxidase in the whey. Demineralisation which is effected without such extremes of pH as by .

, ~0~96s dialysis has not been found to affect the activity of the peroxldase catalyst~ A~ain, drying, such as spray drying of a whey protein concentrate has also been found to deactivate naturally occurring peroxidase, and reconstituted whey protein concentrate powders require the addition of a catalytic amount of peroxidase to the reconstituted system.
The pH of the whey or whey-derived product has an effect on the rate of oxidation of the annatto and should in general be maintained in the range of from 4.5 to 7.5, preferably 5.5 to 6.3.
In a more specific embodiment the invention provides a method of decolourising whey or whey-derived products (such as whey protein concentrates) containing annatto, by oxidation of the annatto with hydrogen peroxide, characterised in that the oxidation i8 efected in the presence of an active peroxidase catalyst at a pH of from 4.5 to 7.5, preferably 5.5 to 6.3, and in that the amount of hydrogen peroxide used is sufficient to effect the oxidation of the annatto whether or not other materials which preferentially c~ompete with the annatto for the hydrogen peroxide are ~resent and in that when the initial concentration of hydrogen peroxide used is initially sufficient to deactivate the peroxidase as a catalyst said initial concentration of hydrogen peroxide is insufficient to permanently deactivate the peroxidase before the concentration of hydrogen peroxide in the whey has been reduced by reaction with materials other than annatto to below a deactivating level.
Generally the lactoperoxidase naturally present in whey is sufficient to catalyse the oxidation of the ~nnatto provided that the whey has not previously been sub~ected to condition~ which deactivate the lactoperoxidase, e.g. ion-exchange, high temperature or spray drying. Where deactivation of the naturally occurring lsctoperoxidase has occurred, active ..

.
.. . .

130~)965 g peroxidase which can be lactoperoxidase or another peroxidase ~aving similar enzym~ activity such as horseradish pero~idase must be added to the whey or whey-derived product prior to the addition of hydrogen peroxide. Thus a reconstituted whey protein concentrate powder that has been spray dried will need such an addition of active peroxidase if decolourisation by the method of the presen~ invention is to be effective.
In practice, where for example annatto-containing whey is to be subjected to ultrafiltration the hydrogen peroxide can conveniently be dosed in-line into the whey e.g., before entry into the bulk storage ~ tan~(s) from which the whey i~ later conveyed to an ultrafiltration unit. ~nere the whey is to be demineralised other than by ion exchange the decolourisation according to the invention is preferably effected after such demineralisation. Whey protein concentrates may be decolourised before drying to form a powder or (with addition of active peroxidase) a reconstituted whey protein concentrate powder may also be decolourised by the method of the invention.
The invention will be further illustrated by reference to the accompanying drawings in which:-Figs.la to lf are curves showing change inabsorbance at 460 nm against time and illustrating the effect of hydrogen peroxide concentration on peroxidase catalysed decolourisation of pasteurised whey which has been stored for several days at oQc to 4 C;
Figs. 2a to 2c are curves showing the change in absorbance at 460 nm against time and illustrating the effect of hydrogen peroxide concentration on peroxidase catalysed decolourisation of fresh pasteurised whey;

Figs. 3a to 3d are curves showing ~he change in absorbance a~ 460 nm against ~ime and illustrating the effect of increasing co~centr~tions of thiocyanate ion on peroxidase catalysed decolourisation of whey;
Figs. 4a and 4b are curves showing the change in absorbance of 460 nm against time and illustrating the effect of demineralisation on peroxidase catalysed decolourisation of whey;
Fig.5 shows two curves, curve A illustrating the effect of pH on the rate of peroxidase catalysed decolourisation of whey and curve B the effect of pH
on lag time before onset of peroxidase catalysed decolourisstion of the whey commences;
Figs.6(a) to 6(d) illustrate the absorption characteristics of annatto;
Figs. 7 and 8 illustrate the deactivating effect of higher concentrations of hydrogen peroxide on peroxidase;
Figs. 9 and 10 illustrate the effect of concentration by ultrafiltration on peroxidase catalysed decolourisation of whey; and Fig.ll is a block flow diagram illustrating continuous peroxidase catalysed whey decolourisation by hydrogen peroxide on a laboratory scale.
Unless otherwise stated the wheys used in producing the following figures of the drawings were commercially produced during Cheddar cheese manufacture, the whey pasteurisation conditions being either 71.5C/17 secs or (72.2 to 72.6C)¦18 secs.
3o Typical analysis figures in percent by weight for liquid whey, whey powder and whey powder concentrate (75%) are as follows:-Liquid Whey WPC 75 l~ey Powder Powder Total Solids 6.1 96.0 96.1 Protein (Nx6.38) 0.8 12.1 76.9 Fat 0.04/0.25* 0.8 6.5 Ash 0.52 8.0 2.75 Sodium 0.053 0.85 0.13 Potassium 0.15 2.47 0.49 Calcium 0.062 1.02 0.46 Magnesium 0.008 0.13 0.048 Phosphorus 0.046 0.79 0.24 pH 6.3 6.0 6.18 * separ~ted/unseparated.

Whey protein concentrates ~e.g. WPC 75) were producedon DD~ plate and fra~e UF equipment and then spray dried.
Referring to Figs. la to lf, H22 solution was added to six separate samples of pasteurised white whey containing 1 ppm added norbixin which had been stored for æeveral days at 0C ~o 4C and the change in absorbance at 460 nm compared to a white whey was - m!easured against time. The H22 solution was added so as to provide 10 ppm, 20 ppm, 30 ppm, 50 ppm, 100 ppm and 300 ppm of H202 in the respective samples of the whey. The resulting curves are reproduced in Figs. la to lf. It is seen that a lag time occurs before the onset of peroxidase catalysed decolourisation, the lag time increasing with increasing initial concentration 3 of the H22 in the whey. Furthermore, it is seen that at high time lags the ensuing peroxidase catalysed decolourisation after lag time is slower. It was observed that the concentration of the H202 continuously decreased but more slowly during the lag time. The experiment was repeated with three samples ..... ~

i;~O0965 of fresh pasteurised whey containing lppm added norbixin and using 10 ppm, 30 ppm and 100 pp~ initial concentration of H202. The results are shown in Figs.
2a to 2c.
A co~parison of Fig.la with Fig.2a shows a marked similarity. At 30 ppm H202 however, the lag time has considerably increased from 15 minutes in Fig.lb to 57 minutes in Fig.2b and the ensuing peroxidase catalysed decolourisation rate is much slower. At the end of the lag time in Fig.2b the H202 concentration had dropped ~o 4 ppm. At 100 ppm initial H202 concentration the time lag in Fig.2c is extended to include the whole curve (5 hours) and no peroxidase catalysed decolourisation section of the curve is apparent. Furthermore, the rate of fall in H22 concentration over the whole of this curve was observed to be very slow, the H202 concentration after 2 hours being still 25 ppm, after 3 hours 10 ppm and after 4 hours 5 ppm.
From the results set out in Figs. 1 and 2 it is seen that an initial concentration of H22 above a~out 10 ppm deactivates the peroxidase catalyst but that partial reactivation occurs if the H22 concentration in the whey is reduced below about 10 before all the peroxidase has been permanently deactivated. It is believed that the hi8her rate of fall of H202 concentration in the stored whey is probably due to reaction of the H22 with catalase formed during storage of the whey.
The effect of thiocyanate concentration is illustrated in Figs. 3a to 3d. The control samples were fresh white pasteurised whey with 1 ppm added norbixin. Two control samples were used with initial H22 concentrations of 7 ppm and ~ ppm H22 respectively. The resulting curves produced as for Figs. l and 2 are set out as Fig.3a (i) and (ii). The experiment was repeated using four samples containing 1 ppm added norbixin and l ppm added thiocyanate ion (added as sodium thiocyanate), 7 ppm, 8 ppm 9 ppm and 10 ppm H202 respectively initially added to the four samples. The results are shown in Fig.3b (i), (ii), (iii) and (iv). The experiment was repeated with four further samples containing 1 ppm added norbixin and 2 ppm added thiocyanate ion but using respectively lO
ppm, ll ppm, 12 ppm and 13 ppm H2o2, The results are shown in Fig. 3c (i), (ii), (iii~ and (iv). A final experiment was carried out using three samples of the same whey with l ppm added norbixin and 4 ppm added thiocyanate ion using 13 ppm, 16 ppm and 17 ppm H202 in the respective samples. The resulting change in absorbance at 460 nm with time is shown in Fig. 3d (i), (ii) and (iii).
The curves in Figs. 3a to 3d show that with increasing concentration of thiocyanate in the whey higher initial concentrations of H2o2 are required to effect peroxidase catalysed decolourisation of the whey. It is believed that the thiocyanate in whey competes preferentially for substrate sites on the peroxidase and is oxidised first by the hydrogen peroxide present. Thus enough hydrogen peroxide needs to be added to react with both the thiocyanate and with the norbixin. The control samples contained thiocyanate ions and exhibit the same effect. The time lag before onset of decolourisation is here being caused by the thiocyanate ion independently of lag time due to the initial H202 being at a deactivating level.
Figs. 4a ~i), (ii) and (iii) and Figs. 4b (i), (ii) and (iii) illustrate the effect of demineralisation by dialysis on the decolourisation of 13()Q~6S
fresh whey. The deminerali~ed whey was produced by dialysis o~ fresh pasteurised white whey for 2.5 hours against deionised water 77 ml whe~/1500 ml water. The final volume of the dialysed whey was 85 ml. A
control sample of whey was diluted with deionised water in the proportion 77 ml whey to 8 ml deionised water. 1 ppm norbixin was added to each of three samples of control whey and three samples of the dialysed whey. All samples were then subjected to the peroxidase catalysed decolourisation process of the invention, 1 ppm, 2 ppm and 3 ppm H22 being added respectively to the dialysis samples and 8 ppm, 9 ppm and 10 ppm H202 to the control samples. The change in absorbance at 460 nm was plotted against time in each case and the results are set out in Figs. 4a(i), (ii) and (iii) for dialysed whey and in Figs. 4 (b) (i), (ii) and ~iii) for the control samples. It is immediately apparent that the redllction in mineral content has a marked effect on the amount of H22 required to efect peroxidase catalysed decolourisation of the whey, about 2 ppm to 3 ppm being required for the dialysed samples whereas 9 ppm ~o 10 ppm H22 were required for the conn-ol samples.
The lag time n Figs. 4a (ii) and (iii) is also shorter than for Figs. 4b (ii) and (iii). Taken in conjunction with the results shown in Figs. 3a and 3b it is reasonably apparent that the reduction in thiocyanate ion by dialysis effectively reduces the lag time and the amount of H22 required for decolourisation.
The effect of pH on rate of decolourisation and on the lag time is illustrated in Fig.5. Eight samples of a fresh Red Leicester cheese whey were used and the decolourisation was effected in each sampie with 10 ppm H22 at various pH levels and the rate of colour .

*Trade mark ~300 ~ 5 -lS-loss in absorbance units per hour was plotted againse pH to obtain curve A and lag time in minutes against pH to obtqin curve B. From curve A it is apparent that whilst a range of pH of from about 4.5 to about 7.5 is possible, a rànge of about 5.5 to about 6.3 would be preferable. The lag time has a different optimum range of about 4.5 to about 6.
The absorbance curve for annatto is reproduced in Figs.6 (a) to 6 (d). In Fig.6 (a) the absorbance c~rve of 200 ppm annatto (1 ppm norbixin) in distilled water is reproduced. It is seen that the peak absorbance occurs at 460 nm which is the wavelength chosen in producing the absorption curves in the remainder of the accompanying drawings. It is to be noted that there is very little absorbance at 600 nm.
Fig.6 (b) is a repeat of Fig.6 (a) but with 0.1% v/v Boehringer lactoperoxidase added. In Fig.6 (c) 1.0 ppm H22 has also been added and the curve shows a marked drop in absorbance at 460 nm. To produce the curve 6(d) lactoperoxidase (0.1% v/v) was added to the reference solution and Fig.6 (d) shows that this addition does not alter the curve shown by Fig.6 (c) for which the reference solution was distilled water.
In all four curves 6 (a) to 6 (d) the absorbance at 600 nm remains constant at a low level. The change in visible colour of the aqueous system apparent to the eye in moving from curve 6 (b) to curve 6 (c) is that of from a strong yellow to a very pale yellow. In producing the three curves 6(a) to 6(c) the reference used was distilled water~ In all four curves 6(a) to 6(d) the pH was 6.2, the scan speed was 2nm/sec, chart speed 20 sec/cm, the expansion 40 nm/cm, the absorbance range 0 to 0.5 and path length 10 mm.
The activation of peroxidase by hydrogen peroxide is illustrated in Figs. 7 and 8. To produce the curves reproduced in Figs. 7 and 8 5 ul of purified lactoperoxidase (Boehringer*) were added to each of *Tr ~ mark i300965 five tubes containing S ml of approximately 20 mM
potassium phosphate buffer having a pH cf ~.2. H202 was then added to five of the tubes to give a concentration of H2O2 of lO ppm. Beads of agar-entrapped yeast cells containing high catalase activity were added to one o~ the tubes immediately after addition of the H202 Beads of the same agar-entrapped yeast were added to a second tube 30 minutes after addition of the hydrogen peroxide and to the third and fourth tubes 60 minutes and ]20 minutes respectively ater addition of the hydrogen peroxide.
The purpose of these additions was to react t~le catalase with H202 to eliminate it from the system in each of the four tubes to which H2O2 had been added.
The catalase took about 25 to 30 minutes to eliminate the H2o2 in each case.
3ml samples of liquid (separated from the beads in our of the tubes) were taken from ea~h of the five tubes and 20 ul of Fulwoods*annatto were added to each 3 ml sample followed by H202 to gi~e a concentration of l pp~ H2O2 in each 3 ml sample. Absorption curves at 460 nm were made in respect of each sample and the rate of decolourisation in absorbance units per hour determined from the resulting curves. In Fi~.7 the values for rate of decolourisation are plotted against the time in minutes after addition of H202 when the beads were added. No difference was found in ~he rate of decolourisation as between the 3 ml sample taken from the tube to which l~0 ppm H22 had not been added and the 3 ml sample taken from the tube to which the beads were added immediately after the lO0 ppm H202 In each of the remaining 3 ml samples the rates of decolourisation had dropped considerably and were lower the longer the time interval between addi.tion of the lO0 ppm H202 and .he addition of the beads. The *Trade mark ~a a~

~30096S

absorption curves are r~produced in F~g. 8 aQ curves (a) to (e), curve (a) ~elng the control sampl2 to which neither 10 ppm H2~2 nor beads were added and curves (b), (c), (d) and (e) are the samples to which 100 ppm H202 were added foliowed by the beads at intervals of 0, 30, 60 and 120 min~tes, respectlvely.
The effect of concentration by ultrafiltration on peroxidase catalysed decolourisation of whey is illustrated in Figs. 9 and 10. ~ey protein concentrates WPCl and WPC2 having the following analysis were used:
WPCl WP~2 (Fig.9) (Fig.10) Total Solids 9.5 22.2 Protein (Nx6.38) 3.63 15.2 Fat 0.3 1.37 Ash o 73 0 94 pH 6.15 6.15 .

WPCl had a concentration factor (CF) of 5 and WPC2 a CF of 15. To a sample of WPCl, 19ppm H202 were added. The H2o2 treated material was stored for 2 days in a refrigerator. The absorption at 460 nm of a sample dilu*ed 5-fold was measured and taken as the reference for the zero point. Various concentrations of H202 were added to further samples of WPCl, the treated materials stored in a refrigerator for 2 days and the absorption of 5-fold diluted samples measured as against the reference sample. The results are shown in Fig.9 in graphical form, the absorbance in arbitrary units relative to the reference being plotted against the initial concentration of H202 added in each sample in ppm. Since WPCl had a protein content of 3.63% as opposed to about 0.8% in the original wheyj the amount of annatto (assoclated with the protein) to be decolourised was considerably , ` 1300965 higher in WPCl than in whey. Consequently the amount of H2O2 required also increased. The higher initial concentrations of 50 9 100 and 200 ppm H2O2 showed deactivating effect but it is seen that initial concentrations of as high as 40 ppm did not in this instance cause much, if any, deactivation of ~he peroxid~se. A similar series of measurements was conducted with samples of WPC2 except that since the CF of WPC2 was 15 the di~lution prior to spectro-photometry was 15-fold and the control reference had ~0 ppm H2O2 added. The protein concentration in WPC2 was 15.2%. From Fig.10, toleration of the peroxidase to H2O2 concentration is seen to have increased to about 50 ppm in this series of samples. The residual H2O2 concentration of the sample marked A in Fig.10 was zero after 4 minutes, and in the case of the æamples marked B and C in Fig.10 was zero after 5 minutes and 0.5 ppm after 9 minutes respectively.
The invention will be furthe. illustrated by reference to the following Examples which are purely illustrative.
_ ample l - I A laboratory scale continuous whey treatment is illustrated in Fig.ll. The whey used was fresh pasteurised white whey from commercial Cheddar Cheese (white) with 1 ppm added norbixin. The whey at 40C
was pumped from a supply tank 1 via pump 2 to holding tank 3, entering the bottom of tank 3 and exiting from the top thereof. Hydrogen peroxide held in supply tank 4 was pumped via pump 5 into the whey stream between pump 2 and holding tank 3 to provide 9 ppm H22 in the whey. 9 ppm H22 was just sufficient to effect decolourisation of a small sample of the whey as a ba~ch addition. The flow rate of whey was such as to fill holding tank 3 in 34 minutes and the flow was maintained at this level throughout the .. . . .

-19- ' experiment. Samples of whey were taken at sampling point 6 at regular time intervals from commencement of whey flow and the absorbance at 460 nm measured for each sample together with residual H22 (if any). The results are shown in Table I.
TABLE I.
` Absorption Residual Time (Mins) (460 nm) H22 (ppm) 34 0.14 0-0.5 54 0.105 0-0.5 79 0.095 0 99 0.105 0 131 0.13 0 ~ 145 ` `i 0.15 0 152 0.12 0 ` ``
174 0.12 0 190 0.122 0.1 219 0.125 0.2 260 0.13 0.1 280 0.128 0.1 : (Untreated Whey) (0.36) - From Table I it is seen that reduction of whey colour to very low levels can be achieved on a continuous basis by quite small continuous in-line addition.of H22 ~

Example 2 The procedure of Example 1 was repeated using the same whey except that the H202 was added to provide an initial concentration of 7 ppm H202 in the whey i.e. insufficient to effect decolourisation of a sample of the whey by batchwise additon (cf Example 1). The results are shown in Table II.

~30096~

TABLE II
Absorption ReSidual Time (Mins) (460 nm) _ H202 ~ppm) (Untreated Whey) 0.4 44 0.175 64 0.175 0 74 0.175 0 84 0.175 0 ~xample 3 Example 1 was repeated using the same whey as in Example 2 but increasing the H22 addition to an initial level of 11 ppm. It was found that build-up of H22 with time occurred in the holding tank at this higher level of H2o2 addition which was reflected in the decolourisation achieved when the build-up occurred. The results are set out in Table III.
TABLE III.
Absorption Residual 20 Time (Mins) (460 nm) H202 (ppm) 44 0.18 0 54 0.185 0.5 j 74 0.21 O.S
84 0.21 0.8-1.0 97 0.22 0.5 108 0.28 0.5 118 0.315 0.5-0.8 Example 4 In this Example hydrogen peroxide was injected into Red Leicester whey being processed in a full-scale commercial plant for the production of whey powder by evaporation and drying. The hydrogen peroxide was injected at a rate to provide 11 ppm H2o2 in the whey, the injection point being into the whey .

.

i3009~S

after pasteurisation but before the whey reached the storage silos in which the whey was held for a minimum of 1 hour prior to evaporation and drying.
Pasteurisation was effected at 71.5C/17 secs. The temperature of the whey during injection and in the storage silos was 42C. A decolourisation of 60% was achieved. -Exam~le S
The procedure of Example 4 was carried out on a different Red Leicester whey and it was found thàt 15 ppm H202 injection into this whey achieved about the `
same degree of decolourisation as in Example 4, namely 60%. It is believed the higher concentration of H202 required (lS ppm as opposed to 11 ppm in Example 4) was due to there being a higher concentration of thiocyanate concentration in this whey compared to that in Example 4.
Exam ~
The procedure of Example 4 was repeated in a commercial plant producing whey powder arising from coloured Cheddar cheese except that the H202 was added to provide an initial concentration of 6 ppm H22 in the whey. The degree of colour removal was 50% to 60%.
Example 7 The procedure of Example 6 was followed but on a di~ferent coloured Cheddar cheese whey and it was found necessary to inject 15 ppm H2o2 into this whey to achieve about the same decolourisation as that achieved in Example 6, namely 50% to 60%. Again, it -is believed-that a higher concentration of thiocyanate ion in the whey of this Example made it necessary to use the higher concentration of~15 ppm H202.
~xample 8 The procedure of Example 7 was repeated except that the concentration of H202 was 14 ppm and the point of injectLon of the H2o2 was into the whey 1:~0096S

stream feeding the whey evaporator. The reaction time was 4 minutes, the whey temperature increased from 20C to 71C during the reaction time and the decolourisation was 85%.
Example 9 Hydrogen peroxide was injected into unseparated whey in a commercial plant producing a 75% whey protein concentrate from coloured Cheddar cheese whey. The H22 was added to provide an initial concentration of H22 in the whey of 10 ppm. The colour value of the resulting powders as measured by a Colorguard* 1000 Series Colorimeter made by Pacific Scientific together with that for a WPC75* from white Cheddar whey and the untreated coloured Cheddar whey are as follows:-White ~heddar Whey Concentrate Powder 12.1 Coloured Cheddar Whey Concentrate Powder 28.5 H202-Treated Coloured Cheddar Whey Concentrate Powder 16.2 *Trade mark

Claims (16)

1. A method of decolourising annatto-containing whey or products produced therefrom, by oxidation with hydrogen peroxide, characterised in that the oxidation is carried out in the presence of a catalytic amount of an active peroxidase, in that the amount of hydrogen peroxide used is sufficient to effect the oxidation of the annatto and in that the time for which the peroxidase is exposed to a deactivating concentration of hydrogen peroxide is insufficient to permanently deactivate the peroxidase.

2. A method as claimed in claim 1, characterised in that the oxidation with hydrogen peroxide is effected at a pH of from 4.5 to 7.5.

3. A method as claimed in claim 1, characterised in that the oxidation with hydrogen peroxide is effected at a pH of from 5.5 to 6.3.

4. A method as claimed in claim 1, 2 or 3, characterised in that the temperature at which decolourisation is effected is not greater than 75°C.

5. A method as claimed in claim 1, characterised in that the whey is concentrated by ultrafiltration before addition of hydrogen peroxide.

6. A method as claimed in claim 5 in which the initial concentration of hydrogen peroxide is not greater than 100 parts per million.

7. A method as claimed in claim 6, characterised in that the concentration factor of the concentrate is at least 5 and the initial concentration of hydrogen peroxide is not more than 40 parts per million.

8. A method as claimed in claim 5 or 6, in which the concentration factor of the concentrate is at least 15 and the initial concentration of hydrogen peroxide is not more than 50 parts per million.

9. A method as claimed in claim 1, 2 or 3, in which the whey is decolourised is fresh whey and the initial concentration of hydrogen peroxide is not more than 30 parts per million.

10. A method as claimed in claim 1, 2 or 3, in which the whey which is decolourised has been stored for 2 or more days before addition of hydrogen peroxide and in which the initial concentration of hydrogen peroxide is not more than 100 parts per million.

11. A method as claimed in claim 1, 2 or 3, in which the whey has been pasteurised before addition of hydrogen peroxide.

12. A method as claimed in claim 1, 2 or 3, in which the whey has been separated before addition of hydrogen peroxide.

13. A method as claimed in claim 1, in which the whey is dialysed before addition of hydrogen peroxide.

14. A method as claimed in claim 13, in which the initial concentration of hydrogen peroxide is not greater than 10 parts per million.

15. A method as claimed in claim 1, 2 or 3 characterised in that the active peroxidase catalyst comprises the lactoperoxidase naturally occurring in the whey or whey derived product to be decolourised.

16. A method as claimed in claim 1, 2 or 3 characterised in that the active peroxidase catalyst comprises an active peroxidase added to the whey or whey derived product.