Back to: Projects

Phosphate Recovery

International Conference on Phosphorus Recovery from Sewage and Animal Waste, Warwick University, UK. May 5th & 7th , 1998. Organised by Centre Européen díEtudes des Polyphosphates, Bruxelles, Belgium

ELIMINATION OF PHOSPHORUS FROM WASTE WATER

BY CRYSTALLISATION

Dietfried Donnert, Manfred Salecker*

* Karlsruhe Research Centre, Institute of Technical Chemistry, Section Water- and Geotechnology,

P.O. Box 3640, D-76021 Karlsruhe, Germany

Abstract

By the application of the precipitation process using Fe and Al salts or lime, which is the main process used in Germany for phosphorus removal, some problems occur, such as an increase of the anion concentration of the water and problems to recover the phosphorus from the precipitation sludge. Therefore a process for phosphorus removal from waste waters was developed by a direct precipitation of Calcium phosphate induced by Calcite as seeding material. The adjusting of the pH was done with lime, but some experiments were done with MgO as well.

Basic investigations led to possible applications on both, municipal waste waters with low phosphorus concentrations and industrial waste waters with considerably higher phosphorus contents up to 400 mg/L P. Continuous bench scale experiments using a stirred reactor to remove phosphorus from the waste water of a motorcar factory have been completed. Based on the good results, a full scale device for the treatment of 160 m3 per hour was put into operation in July 1996 and has successfully run since this time. Furthermore, aspects for the recycling of the phosphorus are given.

  1. Introduction
  2. Phosphorus removal in Germany is usually effected by simultaneous and post-precipitation using Fe and Al salts or lime, respectively [1, 2]. But, these precipitation processes have some disadvantages, especially because the anion concentration of the water, i.e. chloride or sulphate, is increased and a recovery of the phosphorus is nearly impossible. In some places enhanced phosphorus removal, which avoids the main disadvantages mentioned above, is applied. Whenever this process is successfully running as in Berlin [3] it is a very suitable process. But, it is not running properly on all places. Therefore, research work on phosphorus removal has been carried out with the aim of avoiding the disadvantages of the precipitation process by developing a process for direct precipitation (crystallisation) of Calcium phosphate from waste waters induced by seeding crystals [4].

  3. Physical-Chemical Fundamentals of the Calcite Crystallisation Process
  4. The general idea of the Calcite crystallisation process [4] is as follows:

    Domestic and industrial waste waters are usually supersaturated with respect to the very low solubility of most of the Calcium phosphate compounds. If the water is in equilibrium with hydroxyl apatite, the phosphorus concentration in the water should be » 0.002 mg/L P. However, despite much higher concentrations and this low solubility no precipitation occurs in real waste waters. Therefore, it was the basic idea of several developments in different countries, especially in Japan and in the Netherlands, to initiate this precipitation by means of seeding crystals [5, 6].

  5. Previous Investigations
  6. The first idea was the addition of Calcium-phosphate compounds and an adjustment of the pH ³ 9.0 to obtain the optimum chemical conditions for the precipitation, because in this pH-range the PO43- ion becomes the predominant phosphorus species which favours the precipitation. Unfortunately, it turns out in all cases [5, 6], that the process is strongly retarded by the hydrogen carbonate which is present in every natural water system. As a consequence, the hydrogen carbonate ions have to be removed prior to the phosphate precipitation by acidifying to about pH = 4, blowing out the carbonate and afterwards by raising pH for the phosphate removal. Of course this is a considerable drawback for the crystallisation process. The acidification causes an increase of the sulphate concentration of the water, whereas the application of NaOH is expensive. And, furthermore it was shown (table 1) that a multiple use of the hydroxyl apatite resulted in a considerable deterioration of the phosphorus removal efficiency.

  7. The Calcite Process ( adjusting the pH with lime)
  8. It was, therefore, the aim of the research work in the Karlsruhe Research Centre to avoid both, the pre-acidification and the use of NaOH. This was achieved by selecting Calcite (CaCO3) as seeding material and using lime to adjust the pH [7].

    1. Basic Experiments

Figure 1 shows the efficiency of phosphate elimination as a function of the pH. No technique for carbonate removal was necessary. The experiments show, too, that the phosphorus inflow concentration has no marked influence on the effluent concentration. This was also valid in the case of the starch water with an inflow concentration of about 120 mg/L P, which is cut off in figure 1. Therefore the investigations, which were carried out in the beginning mainly on the treatment of secondary effluents, were focused towards industrial waste waters with higher phosphorus concentrations.

Figure 1:

Influence of pH on the efficiency of phosphorus removal by Ñcrystallisation"

Further experiments were carried out reusing the seeding material several times because within a certain time of reaction, the surface of the material becomes covered with the precipitated Calcium phosphate, therefore, completely different conditions regarding phosphorus precipitation may appear. For this reason, experiments with secondary sewage effluent were carried out at pH ~ 9.0 [ 7,8] , which showed as listed in table 1 that the phosphorus elimination is improved by a multiple (three time) reuse of the Calcite, and that simultaneously the amount of the Calcium carbonate precipitated is lower. From these results it was assumed that during the multiple use of the Calcite caused by the deposition of Calcium phosphate on its surface the precipitation of Calcium carbonate is inhibited.

On the other hand, a multiple use of hydroxyl apatite crystals caused a considerable deterioration of the phosphorus removal efficiency as already mentioned in section 3. Probably the carbonate layer formed on this surface has another mechanism, but no further attempt was made to investigate

Table 1

Results with Multiple Use of Calcite and HAP Seeding Crystals

HAP = Hydroxyl Apatite

Experimental conditions:

secondary effluent, c(P) in = 8,0 mg/L, c(Ca) in = 42 mg/L

10 g/L Calcite or 1 g/L HAP as seed, residence time 30 minutes,

pH = 9,0 adjusted with lime water (0,15 %),

Calcite reused

0

1

2

3

c(P) out [mg / L P]

2,7

2,0

1,6

1,6

P-Elimination [%]

66,0

74,2

79,7

79,7

D Ca/D P

12,4

6,8

3,8

2,4

c(Ca) [mg / L]

102

89

78

75

D Ca Calcite1)

1,76

0,92

0,37

0,09

HAP reused

0

3

   

c(P) out [mg / L P]

1,2

6,2

   

  1. calculated by assuming a stoichiometric ratio Ca/P of 2.0 for the "crystallised"

Calcium phosphate

The results described above were further verified with throughput experiments [8], because the Calcite was only reused three times, which probably does not represent the stationary state. These first experiments for the treatment of secondary effluent gave promising results, with a concentration of 100 g/L Calcite (added in the beginning and never renewed) and a residence time of 30 minutes it was demonstrated in an experiment which lasted 40 days that the phosphorus concentration could be decreased from 8.2 mg/L P to 0.45 mg/L P depending on the pH-value (9.2 to 9.6) in the reactor. But at such high pH, a final neutralisation step would usually be necessary before the water may be released. This requires additional facilities and cost which, if the objective is simply to remove, rather than recover phosphate, can be avoided by the enhanced biological phosphorus removal in the case of domestic waste water [3].

    1. Long Term Experiments for Technical Applications
    2. Therefore, it seemed more interesting to test the process for the treatment of industrial waste waters with high phosphate concentrations. First, secondary effluent was adjusted to about 110 mg/L P by addition of sodium hydrogen phosphate solution. Already a pH of 7.3 yielded a phosphorus reduction of about 55 % to about 60 mg/L P. Therefore, continuous experiments were carried out with two industrial waste waters with a phosphorus content of ³ 100 mg/L P in order to extend the possible applications of this method. The results are shown in figure 1 [9].

      1. Experiments with the Waste Water of a Starch Factory
      2. Afterwards, further tests were carried out with waste water from a starch factory, which had been anaerobically pre-treated (³ 100 mg/L P, COD » 5.500 mg/L O2). The experiments were performed in a scale of about 300 L/h and lasted three months. The aim was a partial phosphorus reduction to about 30 mg/L P, the average inflow concentration of the domestic sewage into the corresponding treatment plant which services the factory effluent. This aim could be easily fulfilled by adjusting the pH of the water to about 7.8 with lime, also, the sludge content of the water was very high (~ 650 g dry solids /m„). Furthermore, a reduction of about 15 % of the COD-content of the water was achieved. The results were are summarised in table 2 [10].

        Table 2

        Results of Continuous Experiments in a Starch Factory.

        Inflow

        158.9 ± 14.7 mg/L P

        outflow

        27.7 ± 2.7 mg/L P

        effluent pH

        7.8 ± 0.2

      3. Experiments with the Mixed Waste Water from the Mercedes Motorcar Factory in Gaggenau, Germany

      Other long term experiments were performed with a mixed waste water from the Mercedes motorcar factory in Gaggenau, Germany. Here, the goal was an effluent P-concentration of £ 2 mg/L P, the German effluent standard for direct release. To achieve this, the pH value of the water had to be increased to » 9.2. As a consequence, a neutralisation step was added which consisted of a precipitation by means of aluminium sulphate to pH = 7.5 followed by an Electroflotation unit, which was already applied in the old waste water treatment plant of the factory.

      Figure 2:

      Results of continuously running experiments for phosphorus removal from mixed waste water of the Mercedes motorcar factory Gaggenau, Germany. 100 g/L Calcite, detention time in the crystallisation unit: 20 minutes, throughput: 1m3/h.

      The first experiments in the laboratory scale with very promising: the results are described in [ 9] . Additionally, continuous running experiments lasting for about 12 months were carried out in a stirred reactor with a throughput from 0.1 m3/h up to 1.1 m3/h corresponding to a retention time of 100 or 30 minutes, respectively [11]. 100 g Calcite per litre reactor volume were introduced at the beginning and lime was used to adjust pH. Again, the Calcite was never replaced or supplemented.

      The mean values over a 12 month period are shown in table 3 . The phosphorus was removed very efficiently under the conditions described above. Usually the goal of £ 2 mg/l P was readily achieved in the outflow of the crystallisation unit, although the inflow concentrations rose to about 120 mg/l P in several cases and the water contained up to 300 g/m„ dry solids. On working days the pH of the inflow was adjusted to about 9.1, whereas at weekends it turned out to be necessary to raise the pH-value of the water to about 9.8. The reason for this was probably that machines in the factory were cleaned during the weekends with detergents containing organic crystallisation inhibitors. The neutralisation step effected a further reduction of the phosphorus content of the water to £ 1 mg/l P. Additionally, a further considerable COD-reduction of about 40 % was achieved.

      The excellent results are further demonstrated in figure 2. It is obvious that the phosphorus outflow concentration of the crystallisation unit was almost invariably £ 2 mg/l P, although the feed concentration varied throughout the whole experimental period and the pH-adjustment was not very stable. In one case the pH of the water was as low as pH » 8.0, but the phosphorus concentration in the outflow was raised only slightly to about 3 mg/l P.

    3. Development of a Full Scale Device in the Mercedes Motor-Car Factory in Gaggenau, Germany

The promising results described in the previous chapter lead to the development of a full scale plant for the treatment of 160 m„/h water [12]. The reasons were mainly, that most of the old equipment could be further used, only some amendments had to be installed and that it seemed possible to meet the effluent standard of £ 2 mg/L P just by pH-control regardless of the sludge content and the phosphorus concentration of the inflow.

The crystallisation step is carried out in the old precipitation device which an improved stirring unit. The pH of the water is first adjusted with lime to 9.2 to 11 and fed into a stirred reactor with a volume of 18 m„ and with a residence time of 14 minutes. This reactor was filled at the beginning of the operation with a single charge of 1.6 tons of Calcite (grain size 0.35 - 0.7 mm) which has not been replaced to date (23 months). The water - solid - separation is performed in a parallel platter with a volume of 41 m„ followed by the neutralisation step with aluminium sulphate and an Electroflotation for the final removal of the solids in the water.

The operation of the full scale device started in June 1996. The results after one year of operation are listed in the following table 3 . In the meantime the composition of the waste water had changed: the difference between weekdays and weekend being no longer significant. In addition, the phosphorus concentration of the waste water was much lower which can also be followed by the same table 3 . For this reason the phosphorus concentration in the sludge was also lower than in the earlier experiments. At the outflow of the crystalliser the phosphorus effluent standard of £ 2 mg/L P was met throughout the whole period and was further reduced in the final discharge after passing through the Electroflotation unit.

The X-Ray investigations in the sludge again gave no hint on the formation of crystalline compounds. Only after heating to 600oC was it possible to identify hydroxyl apatite - Ca5(PO4)3(OH) and Magnesium Ammonium Phosphate - NH4MgHP2O7 together with small amounts of Whitlockite Ca18Mg2H2(PO4)14. This confirms the findings of earlier investigations [9-12], the phosphorus is mainly eliminated by amorphous precipitation.

Table 3

Waste Water Treatment Plant Mercedes Gaggenau, Germany

Results of the Preliminary Experiments and of the Full Scale Operation

from June 20, 1996 to June 17, 1997 (340 days)

 

Concentrations in the Water

Sludge Analysis

After Drying at 100oC

 

mg/L Dry Solids

pH

[mg/L o-P]

[% P ]

[%C]

organic

[% C]

inorganic

Studies in the Laboratory Scale with 0,1 m3/h

Inflow

---

7,2 ± 0,1

40 ± 18

5,2

16,0± 4,0

-----

Outflow Crystallisation

---

9,6 ± 0,6

1,4 ± 0,9

10,0± 4,0

20,0± 5,0

----

Pilot Scale Studies with 1 m3/h

Inflow

320± 400

7,3 ± 0,4

25 ± 17

5,0± 1,8

15,0± 4,0

0,8

Outflow Crystallisation

500± 450

9,4 ± 0,5

1,3 ± 1,0

8,5± 1,0

15,0± 3,0

2,0

Outflow Electroflotation

£ 0,5

7,6 ± 0,4

0,8 ± 0,8

£ 0,5 mg/L dry solids, no measurements

Full Scale Operation July 1996 ñ July 1997 (80 m3/h)

Inflow

640± 610

7,8 ± 0,6

11,6 ± 5,0

5,0± 1,3

24,1

0,3

Outflow Crystallisation

1,0± 3,0

9,3 ± 0,5

1,1 ± 0,8

6,0± 0,9

21,8± 3,8

3,0± 0,7

Outflow Electroflotation

£ 0,5

7,4 ± 0,4

0,3 ± 0,5

£ 0,5 mg/L dry solids, no measurements

  1. Experiments with Adjusting of the pH with MgO
  2. In the experiments described above Calcite was used as the seeding material and Ca(OH)2 for the adjustment of the pH of the water. In Australia MgO, sometimes in a mixture with gypsum, serves as seeding material and pH-adjusting reagent in a fluidised bed system [13]. For this, pilot plant studies were carried out with municipal waste water [14] in the sewage treatment plant of Warriewood, NSW, Australia.

    Based on this knowledge a joint research project was started in 1993 with the general goal of a further development of the existing German and Australian processes for phosphorus removal. Both systems employ seeding materials (i.e. Calcite and MgO), and the object of the project was to develop a "hybrid" process containing the best elements of both technologies [11, 13].

    1. MgO-System (Australian System)
    2. At the start of each experiment, the seeding crystal feed was set to 1 g MgO / L reactor volume in a fluidised bed system. Then the throughput of the water was started. The pH of the water rose to about 10.8 and the phosphorus concentration dropped to < 1 mg/L P (figure 3). After some time the phosphorus concentration rose to 2 mg/L P and the pH of the water dropped to about 9.8 corresponding to the exhaustion of the MgO. The efficiency of the process was determined by calculating this consumption of MgO as a function of the throughput.

    3. Hybrid System (Karlsruhe Research Centre Development)
    4. First powdered Calcite (50 g CaCO3/L reactor volume) was added. Then the MgO was applied as described above.

    5. Composition of the Waste Waters
    6. In order to evaluate the Ñnew" Australian System the experiments were carried out with ÑAustralian" and ÑGerman" domestic waste water (table 4 ).

      Table 4

      Composition of the German and the

      (Artificial) Australian Domestic Waste Water

       

      German Water

      Effluent STP Warriewood

      (NSW, Australia)

      Mercedes Water

       

      [mg/L]

      [mg/L]

      [mg/L]

      Phosphorus

      9-15

      5

      5-100

      Ca

      100-120

      26

      100-110

      Mg

      11-17

      16

      8-15

      Na

      22-29

      177

      n.d.

      K

      9-11

      18

       

      Zn

      0,2

      0,1

       

      CO3

      120

      132

       

      Cl

      110

      260

       

      SO4

      100

      46

       

      NH4

      1-3

      30

      10-15

      The ÑGerman waste water" was the secondary effluent of the sewage treatment plant of the Research Centre. The artificial Warriewood water (STP near Sydney, Australia) composed as listed in table 2 was used as an example of an Australian waste water. There is one significant difference between these two waters: the Australian water contains much less Calcium than the German one. This has to be taken into consideration, because the Ca-content is very important in the overall efficiency of the crystallisation process. There are also differences regarding the ammonia and sulphate content.

      In order to make an easy comparison possible two "efficiency factors F and Fí were calculated from the results. F gives the ratio of the water throughput up to an effluent standard vs. the MgO used, which is the plot given in figure 3. This means, the higher the value the more favourable is the result.

      Fí gives the MgO consumption per m3 of treated water calculated from F.

      1. Results with Artificial Australian Waste Water ÑWarriewood"

      First of all the Australian method (MgO only) was applied to confirm the results reported in [13, 14], later hybrid experiments (CaCO3 + MgO) were carried out.

      The results of two series with four experiments each where MgO and the hybrid system was applied, are plotted in the following figure 3. Assuming an effluent limit of 2 mg/L P the efficiency F and based on the MgO consumption as described above amounted to

      MgO-System F = 7.2 ± 0.5 L / g MgO or Fí = 0.14 kg MgO / m3

      Hybrid system F = 12.0 ± 0.2 L / g MgO or Fí = 0.075kg MgO / m3

      Thus a remarkable improvement in the utilisation of the MgO occurred in the hybrid system in comparison with the MgO system. Probably the phosphorus crystallisation was improved by the Calcite, which served as additional nuclei. There was no significant change in either the Ca-concentration or the Mg-concentration in the water: the slope was quite the same with and without Calcite, first a slight decrease, afterwards almost constant.

      Figure 3:

      Australian domestic waste water - Results with MgO-suspension and with the hybrid system (MgO + CaCO3)

    7. Experiments with German Domestic Waste Water
    8. Additionally experiments with the German domestic waste water of the Research Centre Karlsruhe (table 5 ) were carried out. The Australian system (with and without addition of Ca2+-ions) was applied and finally again the hybrid system was investigated using Calcite and MgO.

      Assuming an effluent limit of 2 mg/L P the efficiency F and based on the MgO consumption like described at the beginning of the section amounted to

      MgO-System F = 7.5 ± 0.5 L / g MgO or Fí = 0.16 kg MgO / m3

      Hybrid system F = 12.0 ± 0.2 L / g MgO or Fí = 0.08 kg MgO / m3

      There results are remarkably similar to those obtained with Australian waste water. Once again, a significant improvement in efficiency occurred with the German water with the addition of Calcite when compared to the application of MgO alone. The improvement attained was similar to that obtained by a continuous addition of a CaCl2-solution up to 200 mg/L Ca [8] in earlier experiments.

    9. Discussion of the Results

The results show the great advantage using Calcite as a seeding material. This means, that the influence is the same whether lime or MgO is used. Regarding the practical application it has to be mentioned that the MgO process is carried out at a very high pH of ³ 11,0 which means that the neutralisation cost are much higher compared with the Calcite process which is carried out at pH » 9,2.

  • Application of the Calcite Process for the Recycling of the Phosphorus
  • The crystallisation process has proved himself as a very effective tool for the removal of phosphorus particularly at higher concentration ranges of about 100 mg/L P. It is very easy to handle, because the effluent phosphorus concentration can be set by adjusting the pH with lime. Thus, on-line measurements can be simplified or even omitted.

    In the special case of the Mercedes factory no special attention was given to the recycling aspect. The main reason was, that the sludge contained a great amount of organic carbon which derived from the oil components mainly.

    1. Results of Recycling Experiments in the Laboratory Scale
    2. Currently laboratory experiments are being carried out with special regard to the "crystallisation" product for the recycling of the phosphorus. The details of the device and the preliminary results up to now are listed in the following table 5 .

      Table 5

      Results of Recovery Experiments with the Calcite Process

      Device : Stirred reactor (2 litres), sludge return 1,2 L/H

      Calcite : 50 g /L reactor volume; 0,25 ñ 0,4 mm particle size

      pH adjusting : Lime suspension (2 %) or lime water (0,15 %)

      Throughput : 2 L/h

      Residence time : 60 minutes

      Influent : Secondary effluent FZK STP, supplemented to 80 mg/L P

      With Calcite (0,2 ñ 0,4 mm)

      without Calcite

      Ca(OH)2- Consumption for adjusting a pH - value

      pH = 9,54 ± 0,26

      pH = 9,0 ± 0,3

      pH = 9,86 ± 0,50

      Using

      Lime suspension (2 %)

      Lime water (0,15 %)

      Lime suspension (2 %)

      0,21 ± 0,04 g Ca(OH)2/ L

      0,07 ± 0,03 g Ca(OH)2 / L

      0,29 g ± 0,08 Ca(OH)2/ L

      Carry Over

      57,9 ± 8,8 mg/L

      14,3 % P

      41,7 ± 5,1 mg/L

      17,2 % P

      63,4 ± 18,2mg/L

      13,9 % P

      Calcite

      1,6 % P

      2,1 %*)

      -----------

      Settled Sludge

      0,43 g/L

      13,5 % P

      *)12,4% P

      0,31 g/L

      13,1 % P

      *) The corresponding experiments are still running, the value given is intermediate or in progress

      According to the results listed above the Calcite causes a much more stable treatment, which can be recognised on the lower lime consumption as well as on the lower standard deviation. Additionally, the lime water seems much more favourable regarding the chemical consumption.

    3. X-Ray Analysis of the Products
    4. In the diagrams only after heating to 900oC crystalline Ca-P-compounds (different forms of apatite) were detected, which means, that the original products are amorphous as already stated earlier in the Mercedes experiments (section 4.3).

    5. Discussion of the Results for Phosphorus Recycling

    The Calcite process proved itself a more stable one, required less lime suspension for the pH adjusting than the "normal" lime precipitation and, at a pH of 9.5, yielded a sludge with 14 % P. Regarding recycling the application of lime water rather than lime suspension seems to be more favourable, because a sludge with about 18 % P was obtained (pH about 9,2) in the first experiments compared to 15 % P with lime. However, in this case the phosphorus was collected in the carry over instead on the Calcite material.

  • Outlook, Planning of Further Experiments for Phosphorus Recycling
  • The Calcite process seems a promising method to enable a good recycling of the phosphorus. Further experiments are necessary especially with a Calcite with a larger grain size with shorter retention times. Successful laboratory tests need to be followed-up by larger scale field trials where the German - Australian hybrid process described in section 5.2 with MgO should also be employed.

    For the following 1 - 1,5 years the following experiments seem promising and valid:

    1. Comparison column (fluidised bed) - stirring reactor in the laboratory scale for about 3 months

    with a feed of an artificial waste water of the Phostrip process containing about 70 mg/L P

    whilst varying:

      • the pH-value
      • the grain size of the Calcite seed
      • the residence time

    2. Long term experiments in the laboratory scale for about 3 months with the same water as feed

    with the goals:

      • optimisation of the automatic control
      • verification of the previous results

    3. Design and construction of a pilot plant and running of pilot plant experiments in field situations.

     

    1. References

     

    [1] Jedele, K., Bunkofer, A. et al.: Auslegung von Anlagen zur weitergehenden Phosphatelimination. Korrespondenz Abwasser 38 H.2 (1991), 170-183

    [2] Zacher, B.: Weitergehende Phosphorelimination durch Flockungsfiltration im Anschluß an die chemische Fällung. Stuttgarter Berichte zur Siedlungswasserwirtschaft 90 (1986), 67-95

    [3] Sarfert, F., Boll, R., Kayser, R., Peter, A.: Biologische Phosphorentfernung in den Klärwerken Berlin - Ruhleben und Berlin - Marienfelde. Gwf-Wasser/Abwasser 130 (1989) H.3, 121-130

    [4] Donnert, D. et. al.: Elimination von Phosphat aus Wasser durch impfkristallinduzierte Abscheidung von Calciumphosphat. GWF Special I, S51-S56 (1995)

    [5] Eggers, E. et al.: Full-scale Experiences with Phosphate Crystallisation in a Crystalactor. Wat. Sci. Techn. 23 (1986), 819-824

    [6] Sato, K. et al.: Final Report on the crystallisation process. ISSN 0836-5878 (March 1983) Public Works Research Institute, Ministry of Construction, Tsukuba Science City, Japan

    [7] Rieger, J.A., Donnert, D., Eberle, S.H.: Möglichkeiten der Phosphatelimination aus Abwasser durch Abscheidung von Calciumphosphat. Vom Wasser 71 (1988), 27-40

    [8] Donnert, D. Research work on Phosphorus Removal from Waste Water

    Report Research Centre Karlsruhe KfK 4459 (October 1988), 4-33 (ISSN 0303-4003)

    [9] Donnert, D., Salecker, M., Müller, A., Eberle, S.H.: Abscheidung von Calciumphosphat aus Abwasser mit Hilfe von Calcit. Z. Entsorgungspraxis 12 (Dezember 1991), 793-799

    [10] Donnert, D., Salecker, M., Eberle, S.H.: Phosphatelimination aus Wasser durch impfkristallinduzierte Abscheidung von Calciumphosphat. Wasser Boden 5 (1993), 312-316.

    [11] Donnert, D., Salecker, M.: Phosphorus Removal and Recovery from Waste Waters. Proceedings of the 3rd Australian-German Workshop, Dec. 13-16, 1994, 37-38

    ISBN 0 642 244444 8

    [12] Donnert, D., Gensicke, R., Merkel, K., Salecker, M., Eberle, S.H.: Elimination von Phosphat aus industriellem Mischabwasser durch impfkristallinduzierte Abscheidung von Calciumphosphat. Z. Abwasserforschung (in press).

    [13] Brett, S., Guy, J., Morse, J.K., Lester, J.N.: Phosphorus Removal and Recovery Technologies, Selper Publications, ISBN 0 948411 10 0, 1997, 54-56

    [14] Angel, R. AWT Sydney ñ International Conference on Phosphorus Recovery from Sewage and Animal Waste, Warwick University, UK. May 5th & 7th , 1998. Organised by Centre Européen díEtudes des Polyphosphates, Bruxelles, Belgium

    Contents

    1 Introduction *

    2 Physical-Chemical Fundamentals of the Calcite Crystallisation Process *

    3 Previous Investigations *

    4 The Calcite Process ( adjusting the pH with lime) *

    4.1 Basic Experiments *

    4.2 Long Term Experiments for Technical Applications *

    4.2.1 Experiments with the Waste Water of a Starch Factory *

    4.2.2 Experiments with the Mixed Waste Water from the Mercedes Motorcar Factory in Gaggenau, Germany *

    4.3 Development of a Full Scale Device in the Mercedes Motor-Car Factory in Gaggenau, Germany *

    5 Experiments with Adjusting of the pH with MgO *

    5.1 MgO-System (Australian System) *

    5.2 Hybrid System (Karlsruhe Research Centre Development) *

    5.3 Composition of the Waste Waters *

    5.4 Results of the Experiments *

    5.4.1 Results with Artificial Australian Waste Water ÑWarriewood" *

    5.4.2 Experiments with German Domestic Waste Water *

    5.4.3 Discussion of the Results *

    6 Application of the Calcite Process for the Recycling of the Phosphorus *

    6.1 Results of Recycling Experiments in the Laboratory Scale *

    6.2 X-Ray Analysis of the Products *

    6.3 Discussion of the Results for Phosphorus Recycling *

    7 Outlook, Planning of Further Experiments for Phosphorus Recycling *

    8 References *