GB2210627A - Low temperature laundry detergents - Google Patents

Abstract

The performance of conventional sodium perborate-based domestic detergent powders, particularly at cool water temperatures of about 30 DEG C, is enhanced by incorporation of boronated felspathoids which may be pretreated with small amounts (usually less than 0.5% by weight) of metal salts such as copper sulphate or ferrous sulphate. These prepared reagents are added to a detergent "concentrate" to form about one third of the total bulk, or used in whole or partial replacement of any zeolite already incorporated in the commercial formulation. The reagents function partly by a direct catalytic decomposition of the perborate via a bleaching mode, partly by interaction with residual body perspiration which re-ionises the metal then inducing ion-catalysed decomposition of the perborate. The beneficial effects are apparent on treated garments as improved "freshness", a "fuller" handle on cottons, and marked "antistatic" effect on sythetic fabrics.

Description

IMPROVEMENTS IN OR RELATING TO THE FORMULATION OF DOMESTIC POWDER COMPOSITIONS This invention relates mainly but not solely to the manufacture and formulation of detergents containing peroxygen compounds intended for use in domestic automatic washing machines. These fabric washing compositions can be used for washing at high temperatures as well as low temperatures but are principally intended for use at temperatures less than 40 C.

Inorganic persalts such as sodium perborate and sodium percarbonate are very widely used as bleaching agents in detergent compositions especially in UK and Europe. These persalts provide a very satisfactory bleaching action when the detergent composition is used at high temperatures, which may occasionally exceed 800C, but their action becomes progressively less effective the lower the temperature the washing system is operated. This rather critical temperature dependency of the persalt bleaches represents a drawback of continually increasing importance since it can be expected that the common acceptance of the high temperatures demanded will decrease in response to the need to save energy and time.Consequently where an effective degree of bleaching activity is desired resort may have to be made to chlorine-containing bleaches of various types despite their disadvantages, the diminution in fabric strength, the discolouration of the dyes, and, hence, the reduction in the useful life of the article itself.

In t?e course of the efforts to extend tne range of application and to continue to benefit from the inherent technical advantages of the persalt bleach system it has been shown that certain substances now called "activators" have the power to amplify the bleaching action of the persalts at temperatures below 60 C at which temperature many domestic washing machines are operated. Although these activators can be shown to be extremely effective in quantitative in vitro experiments a generally satisfactory material has not thus far emerged.Sometimes the activator must be used in high proportion in order to achieve satisfactory results; or it may be particularly costly; or it may not be environmentally benign; or it must be used only in conjunction with specific peroxygen bleaching agents; or it provokes instability and hence a low shelf life in the detergent powder product; or it requires a non-routine or inconvenient process of incorporation into the detergent product. A review of the activators described heretofor has been given, for example, by Finley and Blumbergs in European Patent Application 79302525.8.It includes references to carboxylic acid anhydrides, carboxylic esters, (N-substituted) Nacylnitrobenzenesulphonamides, N-benzoylsaccharin, aromatic sulphonates and sulphonyl chlorides, N-sulphonylimides, phosphoric-carboxylic and phosphonic-carboxylic dnhydrides, N-acylazoles, phenyl subphonates, and heterccyclic sulphonate esters. An alternative procedure has been described by Clements and Weigh in USP 4,338,210 by which the persalt is converted in situ to a peracid by hydrolysis or perhydrolysis and then treated, again in situ, with an organic or inorganic material which delivers branide ions in aqueous media.

It will be noted that dilute sodium perborate solutions will decanpose spontaneously at roan temperatures - but the reaction takes a long time to go to completion, a time much longer than is normally available in the usual domestic washing machine programmes. On the other hand the reaction proceeds very rapidly if certain metal ions are introduced: d: the most effective are those of metals of variable valency, as for instance copper, iron, cobalt, manganese, nickel.

Konecny and Meeker, in an interesting 1964 contribution (USP 3,156,654), describe processes for improving the bleaching action of peroxy compounds using heavy metal ions which are supported and restrained by special chelating agents. They give as an example the complex formed from cobalt sulpate and pyridine- 2-carboxylic acid.

It is the purpose of the present invention to provide a means of so improving the efficiency of a persalt-based heavy duty danestic detergent powder that it will respond sufficiently well to the lower temperatures which will be increasingly preferred that no difference will be apparent in the appearance of the articles so washed from that appearance which a normally critical person would expect of the same detergent powder used at its more usual high temperatures. By '"more usual temperatures" we mean between 6O0C and 850C and by "lower temperatures" we mean 350C to 400C.

It is also an object of this invention to achieve this desired effect without: (i) significantly increasing the total cost of the raw materials used to make the detergent canposition, (ii) requiring the adoption of inconvenient or unconventional production processes, (iii) introducing biologically undesirable or chemically active species into the effluent, (iv) undermining the effectiveness of the other ingredients in the detergent ccmposition, (v) impairing the properties, in terms of tensile strength and resistance to abrasion, of the washed fabric, (vi) impairing the fastness of the dyes applied to the fabric when compared with the known fastness properties of those dyes to washing at the temperatures specified.

The present invention provides a method of making such a detergent composition which has improved properties when used at temperatures between 350C and 40"C by incorporating a modified "boronated felspathoid" as hereinafter defined with a conventional perborate-based detergent, the improved composition containing between 10% and 30% (by weight) of the modified boronated felspathoid.

A "boronated felspathoid" may be described as a synthetic alkali metal alumino silicate having a crystal structure similar to that of the mineral "nosean" but containing a boron compound entrapped within that structure.

Such compounds have previously been disclosed by Barrer and Freund (J.Chem.Soc. (Dalton), 1974 pp 1049-1053) though the compounds are therein described as "boron-cor.taining zeolites".

The boronated felspathoid powder product used according to the present invention can be made by any of the conventional processes for hydrothermal crystallisation of zeolites (for example, of sodalite hydrate), fran a source of silica, a source of alumina, and a source of alkali metal preferably sodium on the grounds of initial cost, the simplicity of the reaction, and the extent of the conversion to the true "nosean" -type product, but with a source of boron such as boric acid or sodium borate in the synthesis mixture. The source of silica and the source of alumina can he any kaolinite material in which the proportions of Al 0 to SiO are in the molecular ratio of 1:2. Since any other materials will remain largely unreacted in the process to be described it is advisable for the raw materials chosen to be not less that 98% pure.

Two examples of the production of such a boronated felspathoid are described in E.P.C Application No. 81300325.8 as follows: Example 1.

1712g sodium hydroxide and 751g borax decahydrate were dissolved in 4000g distilled water in a stainless steel beaker. 1085g china clay having a nominal particle size below 1 micron were added and the mixture stirred for 24 hours at 850C. (Fran observation it appeared that the process of hydrothermal crystallisation seemed to be complete after about 6 hours). After filtering, washing and drying overnight at 1300C, 1175g of product were obtained. Its X-ray diffraction powder pattern was substantially identical to that of mineral nosean suggesting-the skeletal structure of the cage around the boron moiety.Chemical analysis showed that the product contained 20.6% Na20, 5.0% B2O3, 31.4% Al2O3, 39.8% Ski021 0.8% Foe203, the balance being water. The particle size was mostly less that 5-10 microns (estimated microscopically).

log of this product was stirred with 100g distilled water for 6 hours at 250C but no more than one fiftieth of the B203 was leached out.

A determination of the physical characteristics of this product using the "Sedigraph" instrument and taking a specific gravity figure of 2.38 produced these results: Nominal particle diameter ........ Percentage finer than 35 microns ........ 100% 20 .. ........ 97% 10 .. ........ 85% 6 .. ........ 65% 5 .. ........ 52% 4 .. ........ 35% 1 .. ........ 3% Samples of the product prepared according to the method described in Example I have been presented for assessments of acute toxicity as follows: (i) the modified Draize Eye Irritation Test in the Rabbit, (ii) Draize Skin Irritation Test in the Rabbit, (iii) Gross Acute Oral Tbxicity Assay in the Rat.

The indicative results were that the product, is, respectively: (i) not a severe ocular irritant, (ii) classified only as a mild primary irritant, (iii) exceeding the level of 500 mg/kg of body weight for the minimum acute oral lethal dose.

Example 2 A boronated felspathoid having the same properties as that produced by EXample 1 was made using sodium aluminate (113g) and "M 75" sodium silicate (225g) as the source of alumina and silica (instead of china clay) with 1 54g sodium hydroxide, 96g borax and 11 05g water. The sodium aluminate was dissolved in half the total water, the other materials in the remaining water. The solutions were combined at 400C and then stirred at 800C for one day. The product was filtered off, well washed until the pH of the washing water was not more than 10, and then the product was oven dried at 1100C.

Studies of the effect of the choice of china clay, as in the method of Example 1, showed that there was very little difference in the apparent chemical composition of the product; however there were some minor variations which could be induced in the particle size distribution of the product. As a consequence of the intrinsic fineness of the material produced (about 85% is less than 10 microns) it will not normally be necessary to refine it further, as for instance by grinding or classifying, to render it suitable for the purposes of the invention. The products of these processes are described generically in this text as 'Nosean "B"'.

It is of interest to note that the product made by the methods described above did not exhibit any ion-exchange capacity, that is to say, neither calcium nor magnesium ions could be removed from an aqueous solution to any useful extent. As is well known other types of zeolites are suggested for use in detergent compositions for this particular purpose, e.g.

Cbrkill, Madison & Burns in B.P. 1,429143 (24th March 1976).

The "modification" proposed is intended to develop the knawn ability of a suitable metal compound to decompose sodium perborate but, by combining it with the boronated felspathoid, to do so at a slow, steady rate whilst taking the reaction to completion within the time scales usually provided in domestic automatic washing machines, and without introducing into the effluent undesirable amounts of contaminating heavy metal salts.Copper sulphate has been found to be one such suitable reagent, the complete reagent being made in the following manner: Example 3 Solutioris of copper sulphate crystals were prepared at concentrations of 50g/litre and 5g/litre. 200g of the dried products derived from the procedures described above (Examples 1 & 2) were stirred in a high speed, high shear mixer; the copper solutions were added dropwise to produce loadings of between 0.01 % and 1% copper sulphate on the powder.

The scheme of evaluation preferred for defining the invention is the most direct one possible: a household front ioading washing machine with a chamber capacity of 20 litres was used, and 5 kilos of "normally"-soiled garments weighed for each cycle; the charge of detergent (with and without the added modified boronated felspathoid) was set at half the weight of the manufacturers' recommendation for the "hot" wash (taking approximately 30 minutes at that "hot" temperature), and the cycle time set at half the hot wash recommendation (approximately 15 minutes). In effect this meant that cotton goods would be washed at the "cool" setting recommended as appropriate for "delicate" fabrics; under these circumstances a majority of the soiled cotton goods will be found to be imperfectly washed.

Changes in procedure of this severity are necessary because of the very high efficiency of modern detergent powders. Temperature measurements were taken at the discharge point.

In many of the "activator" experiments described in the literature "tea-stained" or "coffee-stained" fabrics are used and the change in bleaching power measured by the change in ref lectance. This has the advantage of producing objective quantifiable results but the disadvantage that "tea-stains" or "coffee-stains" can not easily be standardised since they are susceptible to variation in reaction according to source, age of stain, liquor/goods ratio in the wash cycle, the temperature tolerance, and adventitious metal ion. But more important, in actual domestic practice, conventional detergent powders will not remove any tea or coffee stains and either a biological-powder treatment or an old-fashioned "rolling boil" process is necessary.Hence "improvements" in detergent power as determined by the change in reflectance of a tea or coffee stain, though they are helpful in advancing an investigation in the laboratory, may only have a marginal practical relevance. . Where instrumental measurements became necessary in this enquiry a coffee-stained cotton poplin aged four days was used. The sophisticated technique used has been described by H.R. Cooper in the "Shirley Institute Rulletin", Volume 57, Number 1, 1984, his results are given in terms both of a visual rating arrived at by skilled assessors and an instrumental rating. For this present test series three other procedures were used to assist in interpreting the instrumental data: (i) colour differences from soiled unwashed sample, (ii) reflectance values at 460 nm, (iii) percentage of recovery towards complete stain removal based upon reflectance differences from those of the unwashed sample.

H.R. Cooper further remarks that commercial detergent compositions differ in their responses to the presence of the primary staining species: fatty, proteinaceous, and "oxidisable"; in a later contribution the significance of a fourth category, of particulate soils, has been emphasized. The application of the techniques of colour measurement, he writes, enables confident assessments to be made of different detergents: but the results are relative not absolute and depend upon the washing machine, the washing conditions, the fabric construction, and the stains selected. In the present case, where a bleaching action is of a principal interest, the important stains are "oxidisable" and the domestically acceptable results include "freshness" as well as freedom from stain.

However, the method of assessment used continually throughout this enquiry was direct and subjective: "was the general quality of the wash acceptable or not?".

The e "Weber-Fechner Rule" is to the effect that "in subjective assessments of intensity the least increase of stimulus that is just discernible bears a constant ratio to the original stimulus: by integration this leads to a linear relation between the logarithm of the stimulus and the measure of the s==sltion". The quotation is from Tabor wno was writing in 1954 on Mohs's Hardness Scale in Proc. Roy.Soc., 65, 3B p249 (1954). Mohs had himself worked out in 1824 a purely subj ective relationship between one mineral and others in terms of his impression of their hardness; by 1954 Tabor had shown that Mohs's subjective assessment has a rational basis; by 1981 Torran writing in "Wear", 68, p263 (1981) had confirmed the validity of Tabor's work in theoretical and quantitative terms. There are other similar examples, and one may conclude in general that if a subjective assessment is carried out in a sufficiently conscientious and "craftsmanlike" manner and if it has a direct relevance to the matter at issue then the quantitative confirmation can be awaited with some confidence.

For these reasons it was decided that the routine method of assessment would be that normally adopted by a critical housewife: the finished goods were required to be washed and clean to the same standards as would be expected of that detergent powder at the full recommended dose, time and temperature appropriate to soiled cotton; constant references back to the "full charge" system were made during the course of the enquiry to determine that the critical standards were maintained.

The charge weight recommended by most manufacturers of the conventional spray dried powder products is 100g; in the first trials the weight taken was 50g (save for the reference tests for which 100g was taken) and in addition quantities of the boronated felspathoid, modified or not of 10g and, in another series of tests, 20g were introduced into the chamber of the washing machine at the same time as the 50g of detergent powder. In another series a "detergent con concentrate" was used at 35g, for reference, but at 15g for test, with 10g and then 20g of the boronated felspathoid, modified or not.

The commercial detergents tested vary quite widely in composition but, in general, fit the pattern illustrated in the table (Nos 1-4 are "conventional", No.5 is a "concentrate").

Detergent Comnositicn 1 2 3 4 5 Sodium alkyl benzene sulphonate 6% 6% 6% 6% 20% Fatty alcohol ethoxylate 4 4 4 4 20 Sodium carboxy methyl cellulose 4 4 4 4 18 Sodium tripolyphosphate 34 34 34 34 28 Sodium perborate 25 25 - 25 9 Sodium percarbonate - - 20 - Sodium silicate 6 6 6 6 4 Sodium sulphate 20 20 25 18 Enzyme concentrate - - - 2 Optical whitener 1 1 1 1 1 Note that in the table above, the water content, either introduced as hydrated material or as moisture pick-up, is ignored. Proportions are given in percentages.

The e standard detergent camposition used in the washing machine tests in which the various copper sulphate loadings were examined systematically was: ...... commercial concentrate ..... 20g i.e. 1g/litre ...... sodium perborate ..... 20g i.e. 1g/litre ..... (modified) boronated felspathoid ..... 20g i.e.Ig/litre Assuming the commercial concentrate has a composition similar to that shown in the table (No.5) the percentage contributions of the functional ingredients, and the final concentrations of active ingredients in the wash water, will be: ............ synthetic detergents ..... 14....0.14 g/litre ............ antiredeposition agents .................... organic ..... 7 0.07 g/litre .................... sodium tripolyphosphate ..... 10 0.10 g/litre ............ sodium sulphate (or silicate) ..... 3 0.03 g/litre ............ sodium perborate ..... 36 0.36 g/litre ............ modified boronated felspathoid ..... 30 0.30 g/litre ..............................................Total 100 Inspection of the fabric articles treated in the manner described show that, relative to the sodium perborate content, the addition of between 20% and 180%, preferably 70% to 130%, of the modified boronated felspathoid containing 0.1% to 0.2% of the copper salt will be effective in producing, in a typical domestic wash, whiteness, freshness, and cleanliness of acceptable standards at wash temperatures not exceeding 400C.

Exploration of this phenomenon over the full range of copper sulphate loadings undertaken showed that the activity was perceptible at 0.01 % in vitro but would not produce washed results acceptable to a critical housewife in the machine. In the washing machine conditions just described a loading of 1% copper sulphate represents a concentration of metal (although the metal ion appears to be imnobilised on the insoluble boronated felspathoid) in the drainings of 100 ppm. Higher loadings of copper sulphate up to 20% have also been examined but do not appear to offer any advantage in domestic conditions though the reaction still appears to be steady and may be appropriate in industrial conditions where proper attention to the effluent may be made. In general, the higher the loading of copper ion on the boronated felspathoid, the faster the deposition of a given charge of perborate goes to completion.

We have also found that it is possible to reproduce, in vitro, the effects of a "0.1 %" copper loaded bornnated felsoathoid by adding the suitable proportion of the highest loaded material to the unmodified boronated felspathoid. Whichever method is used it will be noted that there is a siqnificant decrease in the concentration of sodium tripolyphosphate (fran one third to one tenth) in the powder ccmnosition.

Sane of the tests carried out, with their instrumental results and subjective impressions, follow.

Detergent composition: A.... Powder concentrate ............................... 1 g/litre... A B.... Sodium perborate tetrahydrate..................... 1 g/litre... B C.... Nosean 'B' ....................................... 1 g/litre... C D.... Nosean 'B' modified with ferrous iron 1 g/litre... D E.... Nosean tB' modified with cupric copper 1 g/litre... E Process conditions Stain; solubilised coffee at 1Qgmjlitre, padded, dried, aged 4 days.

Sample Temperature OC Reagents Wash quality 1 75+ AB acceptable 2 35 AB unacceptable 3 35 ABD accsptable 4 35 ABC acceptable 5 75+ ABC acceptable 6 35 ABE acceptable 0 - - not washed Instrumental Assessments Sample Colour difference Reflectance Percent recovery fran stained at 460 nm of reflectance unwashed fabric 1 14.0 71.1 45.2 2 5.2 51.4 11.3 3 5.8 51.2 11.2 4 6.1 52.9 14.6 5 17.3 79.8 54.9 6 6.6 63.7 15.9 Staining Scale Ratings Sample Visual rating Instrumental rating 0 2 2.0 1 2/3 2.8 2 2/3 2.2 3 2/3 2.2 4 2/3 2.3 5 3/4 3.1 6 2/3 2.5 In a previous application, No 8709824 24 April 1987, the properties of the undified boronated feispathoid in imparting an antistatic effect to synthetic fibre fabrics has been described.This is to say that the common means preferred to render such garments wearable, that is, the application of a long-chain quaternary compound, is no longer necessary. This effect is durable only to the extent that it will withstand two further wash cycles, but of course it can be continuously regenerated if the detergent incorporates the boronated felspathoid product.

We have also observed, using test samples of woven terry towelling, that goods washed, as described, in the presence of the boronated felspathoid, modified or not, acquire a fullness of handle and an improvement in folded-bulk effects which persist for several days after the wash. It is further the case that certain wool weight acrylic fabrics which are at a stage of wear which after washing generally means that the critical housewife would prefer to iron such garments will no longer need such ironing.

No complete explanation for the mechanisms which underlie these phenomena has yet emerged. However, there are knawn properties of the materials involved which must contribute, for instance: (i) for an inorganic compound the boronated felspathoid has an extremely high perceived oil absorption (37ml linseed oil/I 00g felspathoid) only exceeded k certain calcined high surface area materials; this may suggest that the contribution to whiteness and hence to apparent bleaching power may be due to adsorption or absorption of greasy matter; (ii) the boronated felspathoid slowly yields up a proportion of its sodium content, which implies a "microclimate" of hydroxyl ions in its very close proximity, and subsequently, in the proximity of the fibre surface as well; this may suggest the possibility of some more active support of the normal detergent process; (iii) in most of the bleaching experiments described there has been a higher weight of iron (in the ferric state) than of copper, ferrous iron is also a known catalyst for the decomposition of perborate: hence it may be that the contribution of the copper is to induce the transformation of the iron into its more active form; (iv) the effects secondary to the promotion of the perborate bleaching action, that is the improvement in handle, the antistatic effects on synthetic fabrics, and the crease shedding effect on acrylics may be explained by the development of a (relatively) swollen and hydrophilic layer around the fabric surface induced by the very high, very localised, alkalinity. We have considered this effect to be a form of "keto-enol" transformation.

In further experiments the properties of boronated felspathoids modified by other variable valency heavy metal salts have been examined, principally cobalt and manganese. A particularly interesting example which gave excellent wash results in practice was prepared from ferrous sulphate stabilised in the ferrous state by ascorbic acid (Vitamin C). The compositions examined contained 0.5% ascorbic acid (by weight) and between 0.1% and 0.5% Fe i.e. SFprnximate1y 0.5P to 2.5% ferry z sulphate heptahydrate (by weight).

Example 4 2g of ferrous sulphate crystals and 1 g of technical quality ascorbic acid were dissolved in 20 ml water to form a clear, stable, slightly yellow solution. The mixed solution was added dropwise to 200g of boronated felspathoid in a high speed, high shear mixer.

This is an example of "redox poising" in that the composition in service must maintain a higher proportion of the metal in its lower valency state - for it to be effective as a catalyst - but to do so under strongly oxidising conditions.

In the tests described above the modified boronated felspathoid was added at the washing machine, but it is also possible to incorporate it with the other ingredients of the commercial detergent powder at the manufacturer's plant. More importantly, because the modified boronated felspathoid is resistant to high temperatures without decomposing, it is possible to incorporate it in the slurry which is to be spray dried as the routine method of manufacture; it is not necessary to incorporate it at any later, and less convenient, stage as is the case with sodium perborate and sodium percarbonate. It will further be observed that the addition of the (modified) boronated felspathoid will usually impart "free flowing" properties to such compositions obviating any practical difficulties which may arise from their hygroscopic nature: it will be found that an addition of between 21% and 18 of the (modified) boronated felspathoid is necessary to provoke this effect; however, compositions which themselves contain more than about 5-7% water (as, for instance, the product from Example 4) will not be so effective.

We have also observed that the slow, steady decomposition of sodium perborate in vitro persists over many days: when the reaction ceases it can be revived by adding further sodium perborate. Hence, the effluents from washings carried out according to the present invention, having the desirable property that they may continually provoke the release of further oxygen, are very much less likely to increase the Basic Oxygen Demand of those effluents and are therefore relatively benign ecologically; however, some environments and some domestic stains have the property of reionising - so resolubilising - the added copper reagent so that this mechanism may not be available.

Th varying degrees the characteristics and performance of a perborate-based domestic detergent concentrate to be operated at 1 gram/litre and at 350C will be improved by the addition of 1 gram/litre of the boronated felspathoid herein described in these respects: 1 .......... detergency, 2 .......... perborate activity, 3 .......... "antistat" finish on synthetics, 4 .......... "full" finish on cottons, 5 .......... water softening, 6 .......... with product (packet) shelf life unaffected.

Two other materials with structural similarities to the boronated felspathoid have been examined to assess their relative contribution to these properties.

(a) a felspathoid prepared in a manner similar to that described under Example 1 but without the borax ("Sodalite") (b) "Zeolite 4A" - a commercial material as produced by Crosfield Chemie BV with the approximate composition: soda ................ 17.5 - 18.5% aluana .............. 28 - 30% silica ............... 31 - 34% water ................ 18 - 22% In some cases, these materials have been modified with 0.15% copper (cu in the way described for Example 3. The dried products were assessed in domestic wash tests as previously described. The impressions of the results obtained are summaries in the Table.

Property Nosean B Sodalite Zeolite 4A detergency xxx xx xxx perborate activity xxx - xxx antistat xxx - xxx fullness xxx - water softening x - xxx shelf life xxx xxx Notes: (i) The symbols mean: xxx ............. effective xx noticeable x not usefully developed - ............. not tested, or not observed ? ............. uncertain (ii) "perborate activity" relates to "coppered" varients only.

Claims (5)

What we claim is:

1. A detergent composition having improved properties at 350C - 400C consisting of a conventional perborate - containing detergent mixture in which has been incorporated between 10% and 35%, by weight of a modified boronated felspathoid as hereinbefore defined.

2. A -detergent composition as in Claim 1 in which the modified boronated felspathoid contains between 0.5% and 5% of a water soluble salt of iron or copper.

3. A detergent composition as in Claim 2 in which the metal salt is a salt of ferrous iron preferably ferrous sulphate which contain in addition, between 0.5% and 5% of ascorbic acid.

4. A detergent composition as in Claim 2 in which the modified boronated felspathoid is replaced by equivalent weights of a conventional detergent zeolite, modified as hereinbefore described.

5. A detergent composition as in Claim 4 in which the metal salt is a salt of ferrous iron which contains, in addition, between 0.5% and 5%