A new dual coagulant for water purification

PII: S0043-1354(98)00417-5

Wat. Res. Vol. 33, No. 9, pp. 2075±2082, 1999 # 1999 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0043-1354/99/$ - see front matter

A NEW DUAL COAGULANT FOR WATER PURIFICATION R. KUUSIK* and L. VIISIMAA Department of Basic and Applied Chemistry, Tallinn Technical University, Ehitajate 5, 19086, Tallinn, Estonia (First received May 1998; accepted August 1998) AbstractÐA possibility of production of the dual Al±Fe coagulant by sulphuric acid treatment of the new raw materialÐEstonian glauconiteÐhas been shown. The composition, properties and use of re®ned coagulants were investigated. The liquid coagulants contained about 5% of active elements in terms of Al2O3, the solid ones about 15±16%. The content of heavy and toxic elements in the puri®ed water did not exceed the maximum permissible level. # 1999 Elsevier Science Ltd. All rights reserved Key wordsÐglauconite, sulphuric acid, dual coagulant, water puri®cation, toxic elements

INTRODUCTION

The coagulants used for puri®cation of water from suspended impurities and colloids di€er from each other by the type and valency of the main active element, the pH, the physico-chemical properties (solid, liquid, unre®ned, re®ned) and the ®eld of application (for puri®cation of drinking water, treatment of industrial wastewater or municipal sewage, precipitation of certain elements). The coagulating ability of the active element is generally increased by its hydrolysis. The precipitation process is a combination of direct precipitation and coagulation of impurities as well as formation of hydroxide precipitates in which the impurities are incorporated. Depending on the surface properties of the latter, the rate of adsorption or adhesion of the impurities and, accordingly, the eciency of coagulants may vary (Handbook of Water Puri®cation, 1981). In water puri®cation practice salts of aluminium or iron (sometimes their mixtures, the so-called dual or mixed coagulants) are usually used. Dual coagulants may be added to water either separately or in the form of mixtures. The coagulation mechanism, when dual coagulants are used, is complicated and insuciently explored. As declared in the literature (Zapolski and Baran, 1987), mixed coagulants may be used in a larger range of pH values, the particles of hydroxides precipitate more uniformly and a higher degree of water clari®cation is achieved. Other authors (Johnson and Amirtharajah, 1983) have found that combining Al and Fe coagulants does not result in more e€ective coagulation. Pure aluminium coagulants are obtained from bauxite or aluminium hydroxide. Other raw *Author to whom all correspondence should be addressed. [Tel.: +372-6202801; fax: +372-6202801 or 6202020].

materials, e.g. nepheline, a concentrate of Kola apatite production wastes, are also used (Pozin, 1970). Glauconite, a widespread mineral (deposits are found in the USA, Russia, Germany, Belgium, etc.) belonging to a hydrous mica group, contains iron and aluminium silicates as major components (McRae, 1972). That is why glauconite can be considered as an industrial raw material for mixed coagulant production. In Estonia glauconite is found in the form of round pellets blended with sandstone. The reserves of glauconitic sandstone in the main phosphorite deposits in Estonia amount to over 4 billion tons (Viiding, 1984). Using the method of statistical coecients proposed by Nikolayeva (1977) we calculated the chemical composition of the typical Estonian glauconite which may be expressed by the following formula: 2‡ K0;73 Na0;04 Ca0;03 Fe3‡ 0;83 Fe0;17 Al0;93 Mg0;40

Si3;68 O10 …OH†2;39 …H2 O†0;29 : Thus, coagulants obtained from glauconite by sulphuric acid treatment contain aluminium and iron sulphates as main active elements (AE) and di€er in their composition from the mixed coagulants presented in literature (Johnson and Amirtharajah, 1983; Zapolski and Baran, 1987). This di€erence may have an e€ect on their hydrolysis and coagulating ability. The sulphuric acid treatment of glauconite in order to produce coagulants for water puri®cation has been studied at Tallinn Technical University (TTU). In our ®rst research project the suitability of glauconite for obtaining coagulant by the socalled den process was tested (Kudryavtseva et al., 1994). The aim of subsequent studies on the same subject was to increase the content of active

2075

2076

R. Kuusik and L. Viisimaa

elements in coagulants by removing the insoluble residue. This residue consists mainly of silicic acid and may be eliminated by two methods: Юltrating the free-¯owing slurry immediately after digestion of glauconite; Ðdissolving the coagulant obtained by den technology in water followed by ®ltration. In our work the former method was used, as it seemed to be preferable both from the technological and economical point of view. The work was carried out in four parts: 1. research into the kinetics of digesting glauconite with the aim of de®ning a proper technology for gaining high degrees of solubility of active elements; 2. experiments of digesting various samples of glauconite in order to determine the need for the enrichment of the raw material; 3. research into the neutralization and ®ltration of the reaction slurries to ®nd out the most suitable reagent for the process; 4. investigation of the composition and properties of the new dual coagulants and waste components as well as distribution of microelements in the process of obtaining coagulants and water puri®cation. Therefore, the aim of the present work was to investigate Estonian glauconite as an object of sulphuric acid treatment in slurries and to characterize the iron and aluminium salts obtained as dual coagulants for water puri®cation. MATERIALS AND METHODS

In the ®rst stage of the experiments four samples of glauconite were used (Table 1). The parameters of the glauconite digesting varied in the following ranges: . concentration of sulphuric acid from 30 to 70% H2SO4; . quantity of sulphuric acid from 80 to 150% as compared to the theoretical requirement; Table 1. Composition of glauconite Samplea Content, (%) SiOb2tot Fe2Oc3tot Al2O3 K2O MgO Particle size (mm) 0.63 0.18±0.63 0.05±0.18 ÿ0.05

Primare Washed Primare Ground ore 1 ore 2 concentrate 3 concentrate 4 56.3 16.3 7.7 6.7 2.7

54.8 17.6 8.3 6.7 2.9

51.5 20.2 8.7 7.9 3.3

51.1 20.2 8.8 7.9 3.2

1 71 27 1

3 57 36 4

1 71 25 3

0 0 52 48

a samples contained also 2.4±2.7% FeO, 1.5±2.4% CaO and 1.6± 2.1% CO2.bsum of SiO2 from glauconite, other silicate minerals and quartz.csum of all compounds of iron expressed in terms of Fe2O3

. temperature in the reactor from 105 to 1358C; . time of mixing from 5 to 180 min. In the following stages of our work the technological parameters of the process were chosen on the basis of the results of the ®rst series of the experiments. Glauconite digestion was carried out at a controlled temperature in a porcelain reactor with a capacity of 275 cm3. The reactor was equipped with a mechanical stirrer, a re¯ux ¯ow condenser and a heating bath. The initial sample of glauconite (20±50 g) was added to the heated sulphuric acid solution, after which the mixing continued within a ®xed time. During the treatment of glauconite in the reactor, samples of slurry were taken for determining the solubility of the active elements. These samples were immediately diluted in weight ratio 1:10 to interrupt the reaction of glauconite digestion. The diluted samples were ®ltrated and the ®ltrates analysed. In the second and third stages of this work a nonionic ¯occulant (N-300, prepared by Kemira, Finland) was dosed into the reactor 2 min before the end of the mixing, in an amount of 0.0025% from the weight of the reagents. The resulting slurry was diluted with water in the weight ratio 1:0.5 or 1:1 and ®ltrated at 908C, using suction ®lter and a Lavsan (polyester fabric) ®lter-cloth. The ®ltration was carried out at a vacuum of 40 kPa, measuring the ®ltration rate in terms of the volume of the ®ltrate in m3 per l m2 of the ®ltering area per hour. The ®lter-cake was washed up to neutral reaction and dried at 1058C. In the third stage of this work four reagents were tested for neutralization of acidic slurries: . lime (85% CaO); . limestone (39.3% CaO); . ash of oil shale (49.1% CaO, 5.7% Fe2O3tot and 4.9% Al2O3); . nepheline concentrate (1.5% CaO, 3% Fe2O3tot, 28.5% Al2O3, 14% Na2O and 7% K2O). The requirement of the ®rst two reagents was calculated on the basis of their CaO content, but in the experiments with other reagents the content of Fe2O3, Al2O3, Na2O and K2O in the raw material was considered as well. Neutralizing additives were dosed into the reactor after the ¯occulant and water had been added. The resulting slurry was mixed for another 5 min and ®ltrated as described above. When nepheline was used, it was added to the ®ltrate after the insoluble rest had been separated. The resulting mixture, having a porridge-like consistency, could be processed to a solid coagulant. The samples of solid coagulants and wastes, washed and dried at 1058C, as well as primary ®ltrates (liquid coagulants) were analysed with regard to the main chemical components by ordinary methods. The content of microelements in the samples of mineral raw materials, coagulants, solid wastes and the water, puri®ed with coagulants from glauconite, was determined by atomic absorption spectrometry (AAS, Pye Unicam 9100X, England). Photographs of samples of solid coagulants and wastes were made in the Centre for Materials Research of TTU by scanning electron microscope (SEM, ``Jeol'' JSM840A, Japan). Coagulant samples were also examined by X-ray powder di€raction techniques (DRON-4, CuKradiation, Russia) and by di€erential thermal analysis (Q-1000, Hungary). The eciency of the coagulants was tested at the laboratory of the Tallinn water treatment plant by conventional methods. The dose of coagulants was constantly 16 mg/l in terms of Al2O3, as recommended on the basis of the earlier experiments (Kudryavtseva et al., 1994). The testing of the insoluble rest as an adsorbent for removing PO3ÿ 4 ion from waste water was carried out by the method used by Roques et al. (1991), with a few improvements.

A dual coagulant for water puri®cation RESULTS AND DISCUSSION

Digestion of glauconite samples and neutralizing the reaction slurries Extending the reaction time from 1.5 to 3 h (at H2SO4 concentration and amount, 40 and 100%, respectively) caused 10±15% increase in the solubility of AE. In the experiments with sulphuric acid used in the theoretically required quantity the increase in its concentration from 40 to 60% caused a growth in the solubility of the active elements. In case of a further increase in the acid concentration a certain reduction in glauconite digesting due to the thickening of the slurry was observed. Obviously, the thickening of the slurry causes the reduction of di€usion rate of H+ and SO2ÿ ions 4 in it and, as result, the decrease in the digesting degree of the glauconite. Increasing the quantity of acid from 100% (i.e. the theoretical requirement) to 150% caused a growth in the solubility of AE by 12±15%. Raising the temperature from 105 to 1358C brought about an increase in the degree of digestion of AE by 25±30% (Kuusik et al., 1996). On the basis of our critical analysis of the data on the kinetics of the present process mentioned above, the following basic technological parameters were established for further experiments: . time of mixing, 90 min; . temperature in the reactor, 1308C; . amount of suphuric acid, 150% as compared to the theoretical requirements. The four di€erent samples of glauconite (Table 1) were treated applying the chosen experimental parameters. The main parameters of this process varied, depending on the sample, in the following range: degree of iron solubility, 78±86%; degree of aluminium solubility, 92±97%; ®ltration rate, 0.2± 1.9 m3/m2 per h. From the accessory elements magnesium dissolved up to 90±100% and potassium 87±90%. The data showed that washing o€ clay particles from the primary glauconite ore brought about a signi®cant increase in the ®ltration rate, from 0.7 to 1.9 m3/m2 per h, and a slight improvement in the solubility of iron. In the experiments of digesting glauconite concentrates the highest solubility degrees of AE were attained, but the ®ltration rate was low (0.2±0.4 m3/m2 per h), especially when ground material was used. Therefore, washed glauconite ore should be regarded as the most suitable raw material for this process since, in addition to a good ®lterability of the insoluble residue (1.9 m3/m2 per h), fair solubility degrees of AE (82% of iron and 92% of aluminium) were achieved. For comparison it may be pointed out that the ®ltration rate of a similar process with nepheline ranged from 0.7 to 2.5 m3/m2 per hour (Lainer, 1982). A few experiments of digesting washed glauconite ore with a smaller amount of sulphuric acid (80±

2077

100%) were carried out. In that case the concentration of sulphuric acid was reduced to avoid thickening of the reaction slurry. The results of these studies showed that decreasing the excess of sulphuric acid causes signi®cant deterioration of the parameters of this process: both the solubility degree of AE and the ®ltration rate fell. That is why the quantity of acid amounting to 120±150% of the theoretical requirement should be regarded as optimum. However, in that case about 30% of the sulphuric acid remains unused in the product ®ltrate. In coagulants the content of free acid should not exceed 0.1±0.3% (PIX coagulant, 1995). For that reason the ®ltrates from digesting glauconite need to be neutralized. Experiments of neutralizing the slurries obtained by digesting washed glauconite ore showed that all four additives may be regarded as equal with respect to their neutralizing capability. The content of free acid in the product ®ltrate after neutralization was up to 0.1%. Therefore, the second main criterion in this processÐthe ®ltration rateÐmust be considered when choosing the most suitable neutralization additive. In all cases the addition of a neutralizing additive caused a signi®cant deterioration of the ®lterability of the solid phase. The ®ltration rate for the initial acidic ®ltrate was 1.9 m3/m2 per h, but for the neutralized ones only 0.1±0.2 m3/m2 per h. The highest value was attained in the experiment in which ashes of oil shale were used. The ®ltration rate could be signi®cantly increased by increasing the water: slurry ratio and by adding a bigger amount of the ¯occulant into the reactor. At the values of the above-mentioned parameters 1:1 and 0.005%, respectively, the ®ltration rate increased by up to 0.9±1.4 m3/m2 per h. Being an industrial waste oil shale ash is also the cheapest among the reagents used, therefore preference should be given to its use. Limestone is another cheap reagent, but because of its intensive foaming it may cause a certain inconvenience. The method of using nepheline as a neutralizing additive may be regarded as a two-stage process for obtaining coagulants from two di€erent raw materials. In the ®rst stage glauconite, a poorly solving material, is treated with excessive sulphuric acid. In the second stage a readily soluble mineral, nepheline, is added to neutralize the free acid. This is how a high degree of utilization in both the acid and the mineral raw materials is achieved and the content of aluminium in the product increases by about 50% (relative) as compared with using glauconite only. As in this process the adding of nepheline causes a signi®cant thickening of the product slurry, it is advisable to separate the insoluble residue after the ®rst stage of the process, to neutralize the ®ltrate, and to produce a solid coagulant after the evaporation of the neutralized mixture.

2078

R. Kuusik and L. Viisimaa

Fig. 1. Technological scheme for obtaining re®ned coagulants from glauconite.

Potential technological solutions On the basis of the results of our investigations three di€erent methods for obtaining re®ned coagulants from glauconite may be suggested (Fig. 1). All of them are based on a prolonged mixing of glauconite with sulphuric acid taken in excess, followed by separation of the insoluble residue. According to method (a) a large quantity of sulphuric acid (150% as compared to the theoretical requirement) is used. The product ®ltrate is neutralized with nepheline and the neutralized mixture evaporated, thus a solid coagulant is obtained. In case of method (b) a smaller quantity of sulphuric acid (120%) is used and the product pulp is neutralized with ashes of oil shale. After that ®ltration follows. The end ®ltrate may be used as a liquid coagulant or be processed to a solid product. Method (c) is characterized by digesting glauconite by using sulphuric acid close to the theoretically required

quantity (80±100%). Liquid or solid coagulants may be obtained without any addition of neutralizing materials, but the product could contain up to 5% of free acid. Because of the high solubility degree of the raw materials and the highest quality of coagulants obtained, method (a) should be regarded as optimum. Characteristics of the coagulants The samples of coagulants (Table 2) represent products of various technological methods di€ering mainly in the quantity of sulphuric acid for dissolving glauconite and in the type of the neutralizing additive. For liquid coagulant the mass ratio Al2O3:Fe2O3=0.56 is somewhat higher than in the glauconite raw due to a better solubility of aluminium in sulphuric acid. Solid coagulants 2 and 3 contain three times more AE (15±16%) than the liquid one. Sample 2 has the same ratio of Al2O3:Fe2O3 as the liquid coagulant 1 because

Table 2. Characteristics of the coagulants Liquid coagulant Parameters 1.Condition of glauconite digestion .quantity of H2SO4 (% of theoretical requirement) .Concentration of acid (%) .temperature (8C) .time (min) 2. Neutralizing additive 3. Content of water-soluble form (%) .Fe2O3tot of which FeO .Al2O3 .K2O .MgO .Na2O .CaO .Free H2SO4 4. Insoluble residue (%) 5. AE content in terms of Al2O3 (%) 6. Mass ratio Al2O3:Fe2O3 7. Content of particles (%) +0.18±0.63 mm ÿ0.05 mm 8. Requirement of materials for 1 g AE (g) .Washed glauconitic sandstone .Sulphuric acid in terms of 100% H2SO4 .Oil shale ash .Nepheline concentrate

Solid coagulants

1

2

3

120 50 130 90 Oil shale ash

120 50 130 90 Oil shale ash

150 60 135 90 Nepheline

4.3 1.2 2.4 1.6 1.1 Ð 0.08 0.7 Ð 5.2 0.56

12.7 3.2 7.1 4.8 3.2 Ð 0.24 Ð 2.8 15.2 0.56

9.4 2.8 10.3 3.9 1.2 2.6 0.26 Ð 11.7 16.3 1.09

Ð Ð

54 3

72.5 8

6 4.9 1.8 Ð

6.3 5 Ð Ð

4.3 4.3 Ð 1.3

A dual coagulant for water puri®cation

2079

(nearly 12%) may be explained as resulting from a more intensive hydrolysis of iron due to the higher content of aluminium (Zapolski and Baran, 1987) and to the presence of undissolved nepheline. Samples 2 and 3 contain no free acid. More than half of the mass of these products consisted of particles coarser than 0.18 mm. The requirements for the materials per 1 g of AE were the highest to obtain sample 1. For coagulants 1 and 2 the consumption of glauconite was about 1.5 times and the consumption of acid 1.3 times lower. The processing of the liquid coagulant 1 into a solid one (sample 2) caused a slight increase in the requirement for glauconite. An X-ray study of sample 2 showed the main components of this coagulant to be potassium alum and sulphates of aluminium, iron and magnesium, containing various amounts of crystallization water. Electron microscope photographs (Fig. 2) give evidence of the existence of such compounds as alunite K3Al6(SO4)5(OH)10
4H2O and yarosite K2Fe6(SO4)4(OH)12 in the coagulants in the form of quadratic and rectangular particles with a variable degree of distinctness (Zapolski, 1981). The results of testing samples of re®ned Al±Fe coagulants obtained from glauconite are shown in Table 3. As reference materials three samples of industrial coagulants were used: aluminium sulphate from Sweden, iron(III) sulphate (42% solution, named PIX, product of Kemivesi, Tallinn) and unre®ned coagulant produced from nepheline at Eesti Fosforiit (Estonia). Liquid coagulant 1 contains somewhat suspended particles of crystal salts. Therefore, this solution had to be heated before dosing, which complicated the process. The data of Table 3 show that in the case of using new solid coagulants 2 and 3, the required quality of water, according to controlled parameters, was gained. The other coagulants, including laboratory samples of the liquid one and the industrial products, proved to be unsuciently e€ective by one or two parameters at certain testing conditions. For example, in case of using liquid co-

Fig. 2. Electron microscope photographs of solid coagulants. a, sample 2; b, sample 3 (see Table 2 for characteristics).

they are produced from the same raw materials. It contains about 13% Fe2O3 and 7% Al2O3. Besides water-soluble salts, the existence of 2.8% insoluble residue was detected. The latter probably represents yarosite-type ferric sulphates, e.g. K[Fe3(OH)6(SO4)2], which precipitate as a result of hydrolysis at the stage of diluting reaction slurry (Zapolski, 1981). Coagulant 3 has a higher Al2O3 content (about 10%) because nepheline is added to neutralize the free acid. Therefore, Al2O3:Fe2O3 mass ratio in this sample is twice as high as that in samples 1 and 2. On the other hand, product 3 di€ers from sample 2 in having a lower content of iron, potassium and magnesium. The presence of a large amount of insoluble residue in sample 3

Table 3. Coagulating ability of the products Puri®ed water after ®ltration Coagulants from glauconite Liquid Parameter Turbidity (mg/l) Colour (8) pH Content (mg/l) . Fetot . Al3+ Alkalinity (mg-eq./l)

Solids

Permissible level

Raw water

1

2

3

Al2(SO4)3

Fe2(SO4)3

Unre®ned coagulant from nepheline

1.5 20 6.5±7.5

16 27 7.5

1.8 6 6.1

1.1 11 6.8

0.8 7 6.6

1.2 6 6.6

1.7 9 6.6

1.4 6 6.6

0.05 0.03 2.3

0.08 0.64 2.2

0.41 0 2.3

0.11 0.57 2.2

0.3 0.2 1.5±3.3

0.07 0.06 3.45

1.21 0.01 1.35

0.12 0.02 2.3

2080

R. Kuusik and L. Viisimaa Table 4. Characterization of solid wastes Samplesa

Chemical composition (%) Insoluble residue SiO2 Fe2O3 Al2O3 K2O MgO CaO SO2ÿ 4 Properties: A B C

1

3

57.8 55.9 3.9 1 1.8 0.7 15 16.9

69 66.2 3.5 0.9 2.6 1.2 3 17.8

40 1.05/0.98 17

58 0.78/0.53 20

a see Table 2 for process characteristics.A, speci®c suurface area, m2/g; B, amount of the waste, g per 1 g of glauconite/per 1 g of product; C, adsorption ability of PO3ÿ 4 , g per 1 g of solid waste

agulant 1 the puri®ed water contained too much iron and had an acidic reaction. The addition of aluminium sulphate or unre®ned coagulant caused an increase in the content of aluminium, the addition of iron sulphate brought about an increase in the content of iron over the maximum permissible level. Consequently, preference should be given to solid re®ned coagulants made from glauconite.

Fig. 3. Electron microscope photographs of solid wastes. a, sample 1; b, sample 3 (see Table 4 for characteristics).

Characteristics of solid wastes

the same for amorphous SiO2 (aerosil), taken for comparison. Since the speci®c area of the waste samples (40 and 58 m2/g) is much lower than that of aerosil (342 m2/g), the relatively high eciency of samples 1 and 3 may be explained by processes of chemosorptionÐformation of phosphates of calcium, aluminium and iron. The chemical composition of solid wastes suggest too that they can also be used in other processes where materials containing silicates and sulphates of calcium are needed, for example, in producing expanding concretes (The Technology of Coagulants, 1974).

The characteristics of two samples of insoluble residue obtained as waste in the process of producing coagulants 1 and 3 (see Table 2) are given in Table 4. These samples have quite a similar particle size but they di€er in their content of SiO2 and CaO. Sample 3 is mainly glaucosil. It contains 66% of SiO2 and only 3% of CaO. In sample 1 the content of CaO is about ®ve times higher due to the use of oil shale ash, but the SiO2 content is 10% lower. The presence of calcium sulphate crystals in sample 1 and di€erent structural modi®cations of this salt may be observed in Fig. 3(a). The testing of samples of solid waste for removing PO3ÿ 4 ions from water demonstrated the possibility of using these materials as adsorbents. The experiments showed that 1 g of solid waste adsorbed 17±20 mg PO3ÿ 4 . This indicator was about

Distribution of microelements in the processes of producing and using coagulants The water used for drinking and for household needs must have an agreeable taste and odour after puri®cation and be harmless to human health.

Table 5. Content of microelements in raw materials, products and wastes Content (ppm) Products Element Cu Zn Mn Cr Ni Pb Cd As Hg

Solid Wastes

Glauconite

Oil shale ash

Nepheline

2

3

Permissible level (PIX)

1

3

308 179 215 88 49 3.5 0.25 0.8 0.01

85 183 310 86 53 16 1.3 4.5 0.17

233 129 375 194 15 6 0.12 0.75 0.001

1.9 100 141 2.6 17 0.05 0.02 0.12 0.001

4.9 70 95 1.3 18 0.09 0.03 0.1 0.001

1 100 800 3 20 1 0.1 0.5 0.025

288 367 203 69 44 7.4 0.5 4.4 0.05

298 101 164 84 45 0.5 0.2 0.8 0.015

A dual coagulant for water puri®cation

2081

Fig. 4. Distribution of microelements in the process of producing coagulants from glauconite.

Therefore, the content of some heavy metals and toxic elements in puri®ed water is severely limited by standards (EEC, 1980; EVS, 1995). Coagulants may also be regarded as potential sources of some toxic elements. That is why, within the framework of the present work, the distribution of microelements in the process of obtaining coagulants from glauconite and in the process of water puri®cation was also studied. Table 5 presents the content of microelements (ME) in mineral raw materials, in the samples of solid waste and in products 2 and 3. Among the raw materials oil shale ash has the highest content of the toxic elements Pb, Cd, As and Hg. Glauconite contains more Cu and Zn, while nepheline contains more Mn and Cr. In the products made from glauconite the level of ME, excluding Cu, is lower than permitted for PIX. On the basis of these data we calculated the conversion of ME into an insoluble form (Fig. 4). In the calculation for waste 1 we took into consideration the input of ME with the main raw material (glauconite) as well as with the neutralizing additive (oil shale ash). For solid waste 3 only glau-

conite was considered as the source of ME, because the insoluble residue was separated before adding the neutralizing agent (nepheline). The data of Fig. 4 (average for wastes 1 and 3) show that in the process of producing coagulants only 46% of the Zn, which has the highest solubility, is converted to an insoluble residue. The degree of transition of other ME to solid waste exceeds 60% and in case of four toxic elements (Cu, As, Pb, Hg) it is as high as 75±100%. The data presented in Table 6 show that the content of ME in the water puri®ed by using coagulants from glauconite is somewhat higher than in the raw water but does not exceed the permissible level for drinking water (EVS, 1995). Therefore, solid coagulants made from glauconite may be used for water puri®cation, as far as public health service requirements are concerned. CONCLUSIONS

A laboratory study was conducted to elaborate methods for obtaining re®ned Al±Fe coagulants

Table 6. Content of microelements in raw and puri®ed water Content (mg/l) Puri®ed water Element Cu Zn Mn Cr Ni Pb Cd As Hg

Raw Water

Coagulant 2

Coagulant 3

Max permissible level (mg/l)

0.04 0.022 0.01 0.03 0.01 0.001 0.0001 0.003 0.001

0.41 0.24 0.02 0.03 0.01 0.005 0.0005 0.002 0.001

0.05 0.02 0.01 0.02 0.01 0.001 0.0001 0.003 0.001

1 5 0.2 0.05 0.02 0.01 0.003 0.01 0.001

2082

R. Kuusik and L. Viisimaa

from Estonian glauconite. The chemical composition of four various samples of glauconite (primary and washed ore, primary and ground concentrate) was established. The kinetics of the digestion of the above samples by sulphuric acid in slurries was investigated. It was found that increasing the concentration and quantity of the acid as well as raising the temperature in the reactor causes a growth in the solubility of the raw material. Washed ore was found to be the most suitable raw material for this process, considering good ®lterability of the insoluble residue. A basic technology for digestion of the glauconite was established. The possibility of achieving a high solubility of the active elements (82% of iron and 92% of aluminium by treating washed ore) was ascertained. The application of four di€erent reagents (lime, limestone, oil shale ashes and Kola nepheline) for neutralizing free H2SO4 in the product slurries was studied. Two various technological methods for obtaining coagulants from glauconite were proposed. The composition of re®ned Al±Fe coagulants and their eciency in the water puri®cation process were presented. Liquid coagulants contain 5.2% AE in terms of Al2O3, solid ones 15.2±16.3%. The type of the neutralizing additive has a signi®cant in¯uence on the composition and properties of coagulants. In case nepheline was used, the Al2O3:Fe2O3 mass ratio increased up to 1:1. The main components of solid dual coagulant are potassium alum and sulphates of aluminium, iron and magnesium, which contain various amounts of crystallization water. The composition of the solid waste formed in the process allows it to be used for the production of various building materials (including expanding concrete). The suitability of the above waste for sorbtion of phosphate-ions from waste water was shown. The distribution of microelements in the process of obtaining coagulants from glauconite and in water puri®cation was studied. The main part of the total amount of the heavy metals and toxic elements in the mineral raw were converted into insoluble compounds which were separated with the solid waste. By the content of heavy metals and toxic elements the puri®ed water (using the new dual coagulant) corresponded to the public health service standards. The testing of the new coagulants con®rmed that these products have good coagulating ability and

are by their technological parameters comparable to the aluminium and ferric sulphates used at the present time. AcknowledgementsÐThe authors are grateful to the Estonian Science Fondation for ®nancial support (Grant No. 585) and to the Center for Materials Research of TTU for SEM and X-Ray measurements. The constructive criticism of the results by Dr E. AasamaÈe is highly appreciated. REFERENCES

Council Directive. (1980) The quality of water. 80/778/ EEC. Estonian Standard for Drinking Water. (1995). EVS 663:1995. Lorch W, editor. (1981) Handbook of Water Puri®cation. McGraw-Hill Book Co. Ltd, London. . Johnson P.N. and Amirtharajah A. (1983) Ferric chloride and alum as single and dual coagulants. Jour. AWWA 75 (5), 232±239. Kudryavtseva Y., Viisimaa L. and Kuusik R. (1994) A coagulant for water puri®cation from Estonian glauconite. Proc. Estonian Acad. Sci. Chem. 43 (2), 68±76. Kuusik R., Viisimaa L. and AasamaÈe E. (1996) Re®ned coagulant for water puri®cation from Estonian glauconite. 1. Digesting glauconite with sulphuric acid in slurries. Proc. Estonian Acad. Sci. Chem. 45 (1/2), 42±55. Lainer J. A. (1982) Complex treatment of aluminium raw materials by acidic methods In (Kompleksnaya pererabotka alyuminyevogo syrya kislotnymi metodami). Nauka, Moskva (in Russian). McRae S.G. (1972) Glauconite. Earth Sci. Rev. 8, 397± 440. Nikolayeva I. V. (1977) Minerals of glauconite group in sedimental formations In (Mineraly gruppy glaukonita v ocadochnykh formatsiakh). Nauka, Novosibirsk (in Russian). PIX coagulant. (1995) Prospectus of Kemivesi. Tallinn (in Estonian) . Pozin M. E. (1970) The technology of mineral salts. (Tehnologya mineralnykh soley) Part 1. Khimiya, Leningrad (in Russian) . Roques H., Nugroho-Jeudy L. and Lebugle A. (1991) Phosphorus removal from wastewater by halfburner dolomite. Wat. Res. 25 (8), 959±965. Tkachev KV, editor. (1974) The Technology of Coagulants. (Technologya koagulyantov) Khimiya, Leningrad (in Russian) . Viiding H. (1984) Estonian minerals and rocks In (Esti mineraalid ja kivimid). Valgus, Tallinn (in Estonian). Zapolski A. (1981) The sulphuric acid treatment of aluminium raw material with high content of silicon In (Sernokislotnaya perepabotka vysokokremnistogo alyuminiyevogo syr'ya). Naukova dumka, Kiev (in Russian). Zapolski A. K. and Baran A. A. (1987) Coagulants and ¯occulants in water puri®cation proceses In (Koagulyaty i ¯okkulyanty v protsessah ochistki vody). Khimiya, Leningrad (in Russian).