AU634588B2 - Water purification process - Google Patents

Description

WATER PURIFICATION PROCESS

This invention relates to a process for the filtration/purification of natural ground or surface waters to provide water satisfactory for general domestic use.

Apart from detritus and other gross solids natural waters contain a variety of dissolved and particulate impurities ranging in size from approximately 10 - ° to 10 ² m. These impurities include ions, molecules, macromolecules, colloids, biocolloids, and larger organic and inorganic particles. The colloidal impurities present in natural waters generally carry a negative charge which keeps them in suspension. Removal of these impurities requires destabilisation of this suspension by the addition of counterions or oppositely charged particles in sufficient quantity to neutralise this negative charge. Once neutralised, the impurities coagulate to form agglomerates which are then separated from the aqueous phase. Electrostatic and chemical interactions are involved in the destabilization processes.

In the conventional processes lime, aluminium or iron salts, or polyelectrolytes are used to coagulate and flocculate the colloidal dispersion. The water to be treated is dosed with an optimum quantity of the above chemicals at optimum pH to achieve effective destabilization of the colloidal system. The dose of the coagulant and the pH at which coagulation is performed depend upon the composition of the raw water, and must be adjusted if changes of the feed water occur. The agglomerates of floes which form are settled and/or filtered out in clarifiers and filters. Such prior art processes frequently require many tanks occupying many hundreds of square metres. The filtration of floes formed by conventional methods must be effected at relatively low lineal flow rates such as 1-2 cm/min. When substantially higher flow rates are employed, floe breakthrough occurs. This requirement for low flow filtration is an important factor making it necessary to use very large equipment and tanks for municipal and other water treatment plants. Apart from large and expensive equipment the conventional processes require strict control of the operational conditions, i.e. pH, dosing rates, mixing conditions, sedimentation and filtration rates.

While the conventional techniques, coagulation sedimentation, filtration and adsorption are" well established, reasonably reliable and widely used, they are complex and require skilled operators to control them, making them unsuitable for small communities or individual households to treat their own water supply.

Several earlier patents have described the use of metallic particles, particularly aluminium, for the destabilization and separation of a colloidal material from an aqueous base. Thus Hayward in South African Patent 70/2516 described the use of metallic aluminium in conjunction with sand and anthracite in one filter unit. Hayward describes the use of a filter bed comprising particles consisting essentially of aluminium mixed with sand particles and covered with a layer of anthracite particles. Hayward states that the aluminium particles are not dissolved during the filtration cycle and that the water borne impurities adhere to the aluminium particles. South African patent 70/2516 exemplifies only separation of relatively coarse clay particles from surface water, wastewater, and sea water; removal over a prolonged time period of the colour and turbidity occurring in natural waters and the suitability of the process for providing water of potable quality were not demonstrated. It does not give any indications of the effects of pH and temperature. In practice Hayward' s process would not give water of suitable quality because at neutral pH aluminium particles will become passivated and be rendered ineffective; mechanical filtration only will result. Turner in US Patent 3,784,014 (1974) describes a complex effluent treatment device for separation of colloidal particles from a liquid medium which uses a plurality of filtration columns packed with, in sequence, aluminium powder, aluminium/granular carbon mixture, and finally activated carbon. Turner claims that the flow of liquid through the mixed filter bed develops an electrical potential between the granules which electrophoretically attracts the colloidal matter. Turner specifies that both the aluminium and the carbon particles must be electrically conductive. Turner does not teach that such a device can produce potable quality water.

It is an object of this invention to provide a process for the treatment of water to produce water satisfactory for general domestic use. The water from this process meets physical quality guidelines for potable water but may need sterilisation to meet microbiological quality guidelines for potable water.

Accordingly, the present invention provides in one aspect a method for treating water to produce water satisfactory for general domestic use comprising:

(a) adjusting the pH of the water to 4.5 or less,

(b) removing carbon dioxide from the water,

(c) filtering the water from step (b) above through a filter medium comprising particles of aluminium and/or particles of metal alloy containing aluminium, and

(d) adjusting the pH of the filtered water to obtain water satisfactory for general domestic use.

If the temperature of the water to be treated is less than 13°C, the water should also have its temperature raised to at least 13°C in or before step (a) of the process.

The process of this invention provides water which meets current world guidelines for the physical quality of potable water. In accord with the WHO (1984) and Australian (1987) guidelines for drinking water quality the residual turbidity in the treated water should be less than 5 Nephelometric Turbidity Units (NTU) and preferably less than 1 NTU. Residual colour in the treated water should be less than 15 Platinum-Cobalt Units (Pt-Co), and residual aluminium should be less than 0.2 mg/1. The pH after treatment should be between 6.5 and 8.5.

The present invention provides a method representing an integrated water treatment process. The invention is based upon passing the water through the aluminium-containing filter, but to produce water of the desired quality it is essential that the water be treated prior to filtration and undergo post-filtration treatment, as described in steps (a), (b) and (d) of the first aspect of the invention.

Water produced by the process of the invention is suitable for general domestic use. In particular, the product of the water may be used in domestic applications such as, for example, washing clothes, flushing toilets, washing cars and watering lawns. Although the process of the invention does remove some of the microbiological material from the feed water, the product water may still contain microbiological material and may not be sterile. Thus, if the product water is to be used for drinking, it is recommended that the water be sterilized,e.g. by boiling or chlorine addition.

For .convenience, steps (a) and (b) , (c) and (d) as described in the first aspect of the invention may be described as pretreatment steps, a filtration step and a post-filtration step respectively, and these terms will be used hereinafter in this specification.

In the pretreatment steps, the feedwater, which may come from a stream or a water storage or an underground aquifer, is pretreated by adjusting the pH to 4.5 or less. The pH is preferably adjusted to a pH within the range of 3.5 to 4.5, more preferably 4.0 to 4.2. The pH is adjusted by adding acid to the feedwater. The acid may be any suitable strong acid, with hydrochloric acid being preferred. Acid addition may be by any known manual or automatic means utilising any suitable pH sensing and acid addition control means.

The temperature of the feed water should be maintained above 13^DC, with the process preferably being operated at a temperature ranging from 15°C to 35°C.

The feedwater must also be treated to remove carbon dioxide formed from bicarbonate ions when the pH is lowered. Carbon dioxide removal may be achieved by any known means, with air injection or gas stripping being suitable methods.

Following the pretreatment step, the water is passed through a bed of filter medium to separate the colloidal material present in the water. The water is passed through a filter bed preferably at least 200 mm deep. The filter bed may consist of:

(i) metallic particles of aluminium or of a metal alloy containing aluminium. These metallic particles preferably have an average particle size in the range of from 0.252 to 1.25 mm; or

(ii) a mixture of the aforementioned metallic particles and sand particles, said sand particles preferably having an average size within the range of 0.5 to 1.0 mm, and preferably with equal volumes of metallic particles and sand; or

(iii) a mixture of the aforementioned metallic particles and coal particles, said coal particles preferably having an average particle size within the range of 1 to 3 mm and preferably with equal volumes of coal and metallic particles.

If a mixed filter bed (i.e. metal and sand, or metal and coal) is used, the particles are preferably intimately mixed.

The metallic particles used should be clean and preferably be pure aluminium. If the metallic particles are passivated with an oxide or other coating their activity can be generally restored by washing the particles with dilute (0.5 - 1.0 ) hydrochloric acid. The temperature for satisfactory operation of the process depends upon the filter medium chosen. If aluminium particles are used the process will operate satisfactorily at temperatures above 20°C. If metal and sand particles are used the process will also operate satisfactorily at temperatures above 20°C, while if metal and coal particles are used the process is effective at temperatures above 13°C. In contrast to conventional clarification processes the filters of this process can be operated at lineal flow rates of up to 9 m/h (15cm/min) with the preferred rate between 1.5 and 5 m/h (2.5 - 8.3 cm/min) .

In operation of the process of this invention the pretreated feed water is passed through the filter bed. The waterborne impurities are coagulated/flocculated and/or electrodeposited on the filter medium and thus separated from the aqueous phase. The purified water leaves the filter while the impurities are captured and retained there. At some stage the filter bed will become saturated with the impurities and must be washed by reversing the flow of the water. In this cle-ansing mode the filter may be first "airscoured" then backwashed with a small volume of suitable water. A volume of approximately 5-10% of the throughput is used in this operation.

After filtration, the effluent from the filter is treated to neutralise the product water to the desired final pH and- to further remove aqueous aluminium species. The pH of the effluent from the filter may be adjusted by any known means. In a preferred form, the effluent is treated in a bed of suitable particulate solid alkaline material, such as limestone or marble granules. If limestone is used, the limestone particles preferably have an average diameter of from 2 to 5 mm and are preferably arranged in a bed 200 - 300 mm deep. Optionally, the alkaline material may be mixed with or followed by an inert particulate material such as silica sand.

The present .invention also provides an apparatus suitable for carrying out the process of the present invention. Accordingly, in a second aspect, the present invention includes an apparatus for treating water to produce water of potable quality comprising first pH adjustment means to adjust the pH of the water to 4.5 or less, carbon dioxide removal means to remove dissolved carbon dioxide from the water, filtration means including a filter medium comprising particles of aluminium and/or particles of metal alloy containing aluminium and second pH adjustment means to effect pH adjustment of outlet water from said filtration means to thereby produce water of potable quality.

If the apparatus is to be used to treat waters that have a temperature below 13^DC, the apparatus should also include means to raise the temperature of the water to at least 13°C.

The process of this invention has been operated under laboratory conditions and also in a full scale demonstration plant (throughput up to 2000 1/day) using feedwaters from different locations. It was surprisingly demonstrated that pH, temperature, pretreatment of the feedwater, and post treatment of the filtered water were all important parameters influencing the effectiveness of the process for the production of potable quality water.

This discovery, and the performance of the prior art "electrocoagulation" processes will be demonstrated by the following Comparative Examples and Examples of the use of the process of this invention. The Comparative Examples demonstrate that the prior art processes using filtration through aluminium, aluminium/sand or aluminium/carbon without all elements of the process of this invention cannot produce water meeting the WHO Standard of pH 6.5-8.5, turbidity <5 NTU, colour <15 Pt-Co units and most importantly residual aluminium <0.2 mg/1.

The feedwaters used in these examples and their characteristics are presented in Table 1 ; all water samples are from the southeast region of Melbourne, Victoria, Australia.

TABLE 1 Characteristics of Feedwaters used

No. Source pH Colour Turbidity Conductivity Alkalinity

(Pt-Co) (NTU) μ/cm mg/lCaC0₃ to ^'pH 4.5

1. Mulgrave 6.6-7.3 53-97 9-90 135-143 49.5 (dam)

2. Lysterfield 7.4 65 13-20 113 25.5 (dam)

3. Aura Vale 6.6-7.3 98-109 21-21 150 20.9 (lake)

4. Monash 6.6-7.3 27-55 5-60 50 47.3

COMPARATIVE EXAMPLE 1

This example demonstrates that a deep bed of aluminium particles used alone and without the pre- and post-treatments of the process of this invention does not purify water to the desired standards. Untreated Feedwater No. 2 was passed through a pilot plant filter of capacity 2000 1/day with a bed of pure aluminium particles 600 mm deep for a period of four weeks at 50-100 1/h. Table A presents a summary of the results under steady state operating conditions (days 15 to 25). The results show that, although feedwater pH and temperature were within the operating range for the process of this invention, product water pH, turbidity, and residual aluminium content are unacceptable.

TABLE A Analyses of composite samples of product water from aluminium filtration without pretreatment; range of values over 10 day period.

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3.6-3.9 20-26°C 4.5-4.8 1-5 Pt-Co 11-17NTU 0.4-l.^"0mg/l

COMPARATIVE EXAMPLE 2

The effect of pH control of feed water alone: This example demonstrates that a bed of aluminium particles over garnet sand used with only the pH and temperature control of the process of this invention does not purify water to the desired standards.

In this example a filter of capacity 2000 1/day was used with a bed of aluminium particles 300 mm deep over a bed of garnet sand 300 mm deep. Feedwater no. 4 was used in the following manner: The water was dosed with hydrochloric acid to the desired pH, heated if necessary to bring the temperature within the range 21-30°C and passed directly onto the filter bed; no attempt was made to control the alkalinity i.e. remove carbon dioxide or adjust the pH filter. Without the pretreatment specified in the process of this invention even after 17 days of operation the product water did not meet drinking water standards being high in residual aluminium. Table B presents analysis of the product water (24 h composite samples) during the operational period.

TABLE B Analyses of composite samples of product water from aluminium/garnet sand filtration with pH and temperature control of f edwater only; range of values over 17 day period.

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3.9-4.1 21-30°C 4.7-5.5 8-19 Pt-Co 1-4 NTU 0.5-1.8mg/l

COMPARATIVE EXAMPLE 3

Effect of aluminium/coal filter bed without carbon dioxide removal or post treatment . This example demonstrates that a bed of aluminium particles over coal used with only the pH and temperature control of the process of this invention does not purify water to the desired standards.

Feedwater No. 4 was passed at 50-100 1/h through the pilot filter with a bed consisting of 300 mm coal particles and 300 mm aluminium particles for a period of 20 days. The product water was marginally acceptable for colour and turbidity but contained excessive residual aluminium. The range of values obtained are given in Table C. Table C Analyses of composite samples of the product water after Al/Coal filter with no carbon dioxide removal or pH control of product water . Range of values over 20 days operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3.9-4.5 15-19°C 4.8-5.5 8.5-16.5 Pt-Co 2. 1-5.0 NTU 0.6-2.0mg/l

In a second run the aluminium layer of the filter bed was increased to 500 mm deep , the filter was operated at 50 1/h and the pH was controlled at 3 . 8-3 . 9 . The results in Table D show that the residual aluminium level increased without any improvement in colour or turbidity .

Table D Analyses of composite samples of the product water after Al/Coal filter with no carbon dioxide removal or pH control of product water . Range of values over 18 days operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3. 8-3.9 15-19 °C 4.8-5. 0 11-18 Pt-Co 1. 8-4.3 NTU 1. 6-2.4mg/l COMPARATIVE EXAMPLE 4

Effect of mixed aluminium/coal filter bed without carbon dioxide removal or post treatment . This example demonstrates that a bed of aluminium particles intimately mixed with coal used with only the pH and temperature control of the process of this invention does not purify water to the desired standards.

Feedwater No. 4 was passed at 80-90 1/h through the pilot filter with a bed consisting of 600 mm of intimately mixed coal particles and aluminium particles for a period of 5 days. The product water was acceptable for colour and turbidity but contained excessive residual aluminium. The range of values obtained are given in Table E.

Table E Analyses of composite samples of the product water after mixed Al/Coal filter with no carbon dioxide removal or pH control of product water . Range of values over 5 days operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

4.0-4.5 16-19°C 5.2-5.3 11-13 Pt-Co 2.4-3.2 NTU 0.8-1.4mg/l COMPARATIVE EXAMPLE 5

Effect of mixed aluminium/ coal filter bed without carbon dioxide removal but with post treatment: This example demonstrates that a bed of aluminium particles over coal used with the pH and temperature control and the post treatment of the process of this invention but without carbon dioxide removal does not purify water to the desired standards.

Feedwater No. 4 was passed at 50 1/h through the pilot filter with a bed consisting of 500 mm of aluminium particles over 300 mm coal particles and for a period of 9 days. The water from this filter was post-treated by passing through a bed of 300 ram of limestone over 300 mm of sand. The product water was acceptable for pH, colour and turbidity but contained reduced but still excessive residual aluminium. The range of values obtained are given in Table F.

Table F Analyses of composite samples of the product water after Al/Coal filter with no carbon dioxide removal . Range of values over 5 days operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3.6-3.8 12-14°C 7.1-7.9 7-14 Pt-Co 0.1-2.8 NTU 0.2-0.7mg/l COMPARATIVE EXAMPLE 6

Effect of mixed aluminium/coal filter bed with carbon dioxide removal and post treatment but without temperature control : This example demonstrates that a bed of aluminium particles over coal used with the pH control, carbon dioxide removal and the post treatment of the process of this invention but with operating temperature below the specified range does not purify water to the desired standards.

Feedwater No. 4 was passed at 50 1/h through the pilot filter with a bed consisting of 470 mm of aluminium particles over 300 mm coal particles and for a period of 9 days. The water from this filter was post-treated by passing through a bed of 300 mm of limestone over 300 mm of sand. The product water was acceptable for pH only, having excessive colour, turbidity and residual aluminium. The range of values obtained are given in Table G.

Table G Analyses of composite samples of the product water after Al/Coal filter with carbon dioxide removal and post treatment but without temperature control . Range of values over 3 day operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

3.7-3.8 10-12°C 7.3-7.4 10-18 Pt-Co 8.5-9.2 NTU 1.3-2. lmg/1 One embodiment of the process of this invention and its operation will now be described with reference to the drawings in which:

Figure 1 is a schematic representation of one embodiment of a water purification process according to the invention;

Figures 2A and 2B show the removal of colour and turbidity as a function of pH;

Figure 3 shows the residual pH and residual aluminium remaining in the water after treatment according to the invention;

Figure 4 shows the residual colour and turbidity in the product water after treatment according to the invention; and

Figures 5 and 6 show the quality of product water after treatment according to the invention.

With reference to Figure 1, raw water 1 flows into storage tank 2 where it is mixed with pH adjusted head tank overflow water and heated if necessary to raise its temperature to a suitable value. Air 7 is blown through the water in tank 2 to remove dissolved carbon dioxide. Alternatively, air may be passed through constant head tank 4 to remove carbon dioxide (not shown) . The conditioned water passes through pump 3 into constant head tank 4 where its pH is adjusted using pH control unit 5 to pH 4.5 or less. The pretreated feedwater passes through valve 8 and flowmeter 9 to the filter 10. The capacity of pump 3 is greater than the flow rate through valve 8 and the filter; the overflow from constant head tank 4 passes back into the raw water storage tank 2. Filtered water passes through valve 11 to a holding tank 12 thence to the post-treatment unit where it passes through a bed of CaC0₃ particles 20. The product water is then passed to storage and distribution.

The filter is backwashed when necessary by closing valve 11, opening valve 17, passing air through valve 16 to "air scour" the filter bed for a suitable time then closing valve 16 and opening valves 15 and 18 and pumping product water through filter 10 in the reverse direction to normal flow then through valve 18 to disposal.

The performance of the process described is demonstrated in the following non-limiting examples.

EXAMPLE 1

The importance of the total feedwater pre-treatment and product water post-treatment prescribed in the process of this invention is demonstrated by the following experiments. The filter described in the Comparative Examples was charged with 470 mm of aluminium particles and 300 mm of coal particles. Feedwater No. 4 was passed at 50 1/h under the following conditions:

Experiment 1: No pre-treatment but pH adjustment (see Comparative Example 2)

Experiment 2: pH adjustment, recirculation, airstripping to remove C0₂.

Experiment 3: pH adjustment, recirculation, airstripping to remove C0₂ and temperature control with filtration followed by limestone percolation for final pH adjustment (the process of this invention) . The results of these experiments are summarised in Table 2 .

TABLE 2 Analyses of composite samples of the product water after Aluminium/Coal filter.

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

Pt-Co NTU mg/1

Experiment 1

3.6-3. 7 7-14°C 4.5-5. 0 16-30 3-10 1. 1-3.4

Experiment 2

4. 1-4.2 14-21°C 5. 8-6.0 0-4. 1 0. 6-0. 7 0. 1-0. 2

Experiment 3

3. 7-4.2 13-18°C 7.5-7. 9 2. 2-5.0 0-0.3 0. 1-0. 2

Only in Experiment 3 does the product water meet WHO or Australian Standards. In addition, preliminary bacteriological tests showed 99.5% bacteria removal in the product water.

EXAMPLE 2

The effect of pH on performance of Aluminium or Aluminium/coal filters: To demonstrate the effect of pH on the filtration process pretreated feedwater No. 1 at various pH values was passed through laboratory scale filters containing either aluminium particles alone or a mixture of aluminium and coal particles at a rate of 5-6 m/h. The colour and turbidity removed in steady state operation are shown as a function of pH in figures 2A and 2B. The temperature in these experiments ranged from 22-27°C. The data demonstrates that the process performs markedly better when the pH is <5; almost complete removal of both colour and turbidity is achieved at pH 4.5. These experiments also demonstrate the congruent performance of aluminium and mixed aluminium/coal filters.

EXAMPLE 3

The effect of pH on performance of Aluminium /sand filter: A laboratory filter with a bed of 325 mm of atomised aluminium particles over 325 mm of fine filter sand was operated at a filtration rate of 5-6 m/h at 22-27°C for a total of more than 30 hours using pretreated No. 3 feedwater. The feedwater pH was changed from 4 to 5 in stages; the results shown in Figures 3 and 4 demonstrate the sharp deterioration in performance at pH 5 by increased turbidity and residual aluminium.

EXAMPLE 4

In this example the laboratory scale filter bed consisted of a 300 mm deep layer of 0.7 - 0.95 mm aluminium particles over a 300 mm deep layer of filter sand of particles size 0.5 to 1.0 mm. Pretreated feedwaters 1, 2 and 3 at pH 4.2 were in turn passed for 24 h through the bed at a linear flow rate of 5-6 m/h at temperatures ranging from 22-27°C. The data in Figures 5 and 6 show that after post treatment drinking quality water was consistently produced in each case.

EXAMPLE 5

Use of aluminium alloys as filter media . A variety of aluminium alloys were compared with pure aluminium eg alloy 214, extrusion alloy, and sacrificial anode alloy. The results of these comparative tests showed that each gave equivalent performance.

EXAMPLE 6

Attempt to eliminate post-treatment step: In an attempt to simplify the process and the necessary equipment limestone was incorporated into the electrochemical filter bed. The filter described in the Comparative Examples was charged with 400 mm of aluminium particles, 340 mm of coal, and 200 mm of limestone particles. Completely pretreated (including CO ₂ strip) feedwater No. 4 was passed at 50 1/h through the filter. The residual aluminium in the product water was excessive as shown by the results in Table 3 below.

Table 3 Analyses of composite samples of the product after Al/Coal /limestone filter. Range of values over 2 days operation .

Feedwater Product Water pH Temperature pH Colour Turbidity Residual Al

4.1-4.2 14-19°C 7.4-7.5 2.2-2.9Pt-Co 1.1-1.6 NTU 0.3-0.4mg/l

These results demonstrate that the final pH adjustment must be undertaken following the coagulation/filtration step. Finally it will be clear to the reader that various modifications and variations may be made to the above described embodiments without departing from the spirit and scope of the present invention.

Claims (16)

The claims defining the invention are as follows:

1. A process for treating water to produce water suitable for general domestic use comprising:

(a) adjusting the pH of the water to 4.5 or less;

(b) removing carbon dioxide from the water;

(c) filtering the water from step (b) above through a filter bed comprising particles of aluminium and/or particles of metal alloy containing aluminium; and

(d) adjusting the pH of the filtered water to obtain water satisfactory for general domestic use.

2. A process as claimed in claim 1 wherein the temperature of the water is at least 13°C.

3. A process as claimed in claim 2 wherein the water is at a temperature of from 15°C to 35°C.

4. A process as claimed in claim 1, 2 or 3 wherein the pH of the water is adjusted in step (a) to a pH within the range of 3.5 to 4.5.

5. A process as claimed in claim 4 wherein the pH of the water is adjusted in step (a) to a pH within the range of 4.0 to 4.2.

6. A process as claimed in any one of the preceding claims wherein the particles of aluminium and/or metal alloy containing aluminium have an average particle size from 0.252 to 1.25 mm.

7. A process as claimed in any one of the preceding claims wherein the filter bed further comprises particles of carbonaceous material.

8. A process as claimed in claim 7 wherein the particles of carbonaceous material have an average particle size in the range of 1 to 3 mm.

9. A process as claimed in any one of the preceding claims wherein the filter bed further comprises solid inert particulate material.

10. A process as claimed in claim 9 wherein the solid inert particulate material has an average particle size of from 0.5 to 1.0 mm.

11. A process as claimed in any one of the preceding claims wherein the pH is adjusted in step (d) by passing the water through a bed of solid particulate alkaline material.

12. A process as claimed in claim 11 wherein the solid particulate alkaline material comprises limestone particles and/or marble particles.

13. A process as claimed in any one of the preceding claims wherein the water has a lineal flow rate through the filter bed of up to 15 cm/minute.

14. A process as claimed in claim 13 wherein the water has a lineal flow rate through the filter bed of from 2.5 to 8.3 cm/minute.

15. Apparatus for treating water to produce water satisfactory for general domestic use comprising first pH adjustment means to adjust the pH of the water to 4.5 or less, carbon dioxide removal means to remove dissolved carbon dioxide from the water, filtration means including a filter bed comprising particles of aluminium and/or particles of metal alloy containing aluminium and second pH adjustment means to effect pH adjustment of outlet water from said filtration means.

16. Apparatus as claimed in claim 17 wherein the apparatus further comprises means to raise the temperature of the water to at least 13°C.