SK159594A3 - Process for the removal of phosphorus - Google Patents

Abstract

A process for the removal of phosphorous, in the form of orthophosphate, from an aqueous stream comprising the introduction into the stream of an effective amount of calcium ions, preferably gypsum derived, adjusting the pH to between about 5 and 8, introducing the stream into a reaction vessel wherein it is brought into contact with a gently stirred bed of fine particulate magnesia in an amount sufficient to fix the so-formed calcium phosphate, the pH of the stream in contact with the magnesia being from about 8 to 12 and removing a stream from the vessel which is depleted in phosphorous.

Description

Process for removing phosphorus

Technical field

The invention relates to a process for removing phosphorus from watercourses, in particular to a process in which phosphorus is present in the form of orthophosphate and is removed by adsorption of calcium salts to hydrated magnesium oxide.

BACKGROUND OF THE INVENTION

The presence of phosphorus in watercourses entering the environment is considered undesirable in view of the effect this element has on eutrophication. Although significant amounts of phosphorus from surface waters can reach the watercourses that wash out phosphorus from the soil in the basin (especially in agricultural areas fertilized with phosphorus-containing chemicals), there is a significant proportion of phosphorus entering the environment from municipal and industrial waste water. For example, in Hawkesbury, in the Nepean basin near Sydney, Australia, an estimated 1/3 to 1/2 of the total dissolved phosphorus originates from the sewer.

The problems of excessive amounts of phosphorus in watercourses have been known for some time, and in the prior art a number of methods are known to refine and solve the problem. The aim of these procedures is to remove most of the phosphorus's aqueous waste so that its quality complies with the relevant environmental protection regulations.

One general approach to the problem of phosphorus removal is to treat the phosphorus-containing waste water by adding metal salts, thereby forming an insoluble phosphorus-containing metal compound, which is then removed from the water by sedimentation or other methods. Some of these processes use aluminum or iron salts as well as lime. In this process, insoluble aluminum, iron or calcium phosphates are formed.

While these procedures have been successful to some extent because they have allowed the standards to be met, the requirements for limiting the amount of phosphorus in the outgoing wastewater are being tightened to protect the environment, so it would not be possible to meet these stricter standards without significantly increasing the coagulating agent doses. and / or a very significant technological improvement of existing wastewater treatment plants. Data on the amount of iron salts that would be required to reduce the phosphorus concentration to 0.15 mg / L can be found in D M Philp: J. Wat. Field. Cont. Fed. 57 (8), 1984, p. 841-846, which discloses the removal of phosphorus from a large volume of wastewater near Canberry in Australia.

Improvements in the use of metal salts entail an increase in costs and an increase in the cost of the process, but an even more serious disadvantage is that the use of these phosphorus precipitation chemicals produces very gelled sludge products that are difficult to process and difficult to dewater. The use of chemical precipitation of phosphorus usually results in an increase in sludge volume of 30 to 40% and the associated increase in transport and landfill costs. The sludge also contains an increased amount of aluminum or iron, which may make it difficult to use as additives to fertilizers for agriculture. Furthermore, in some cases, the use of these chemicals may complicate the efficacy of the various biological methods used to remove carbon and nitrogen, as the chemicals used to precipitate phosphorus may affect the pH in the bioreactors.

Another disadvantage is that a considerable amount of phosphorus is lost in the form of sludge. For example, a city with 500 000 inhabitants is estimated to lose around 300 tonnes (elemental phosphorus) annually as a result of the use of chemical precipitation in sewage sludge.

Another general approach to solving the problem of phosphorus removal is the use of adsorbents. In this process, the phosphorus precipitates on a suitable medium on which it then remains adsorbed. When the capture medium is sufficiently saturated with phosphorus, it is removed and used either by obtaining phosphorus from it and recovered, or used directly as a phosphorus source, for example in mixed fertilizers. Since this process results in a small volume of phosphorus-saturated product in recyclable form, this method appears to be more advantageous than others. In addition, the total sludge from wastewater treatment is reduced.

Absorption methods have been reported in the prior art as being suitable for removing phosphorus from watercourses. See Pac. Chem. Eng. Cong. 4, 259-264 (1983); Hirasawa et al. disclose a phosphorus removal process in which phosphorus is removed in the form of crystalline hydroxyapatite that is formed on a medium, such as phosphorus rock or bone coal. According to the authors, these media cause crystallization. In this process, the phosphorus-containing waste water is first de-carbonated and then a suitable amount of calcium hydroxide is added. Thereafter, in one method, the purified water is passed through a sand filter and then an active medium in a fixed bed. Purified water flows out of this fixed bed. According to a second method, the purified water passes through a fluid bed of the active medium, then through a sand filter and finally the active medium in the fixed bed. Purified water flows out of this fixed bed.

The authors report that with a secondary effluent of 100 m ³ / day containing 2 to 3 mg / l of phosphorus, the amount of phosphorus was reduced to 0.26 to 0.35 mg / l.

J. Water Pollut. Cont. Fed. 60 (7), 1239-1244 (1988)

Kaneko et al. discloses a process for phosphorus removal using magnesia granulated activated clinker having the following composition: 95% MgO, 3% CaO and 1% SiO ₂ - This process essentially involves adding sodium hydroxide and calcium sulfate to the pre-treated waste water to bring the pH 8 to 9a a calcium concentration of from 50 to 60 mg / l. The purified solution is passed from top to bottom through a sand filter to remove suspended solids and then through a column containing magnesium granulated activated clinker to remove phosphorus. At a flow rate of 3.0 m ³ / m ³ MgO bed volume per hour, the phosphorus concentration decreased from 2-5 mg / L to a level of less than 0.5 mg / L. When the phosphorus concentration increased above this level after about 8 months of operation, the magnesium oxide was regenerated with 2% acetic acid. The authors state that it was not necessary to remove carbon dioxide in the inlet.

Based on the above work, other related Japanese research papers appeared. For example, JP 12 16795-A discloses the removal of phosphorus from wastewater in a fluidized bed system consisting of various phosphorus absorbing materials including light-burned magnesia. The process as described in the application involves a significant recycling of the phosphorus-free waste stream, which aids in fluidization of the reaction bed, and further requires the addition of an anionic polymer to the reaction mixture.

In Wat. Res. 25 (8), 959-965 (1991); Roques et al. disclose another process for the removal of phosphorus from wastewater using half-burned dolomite as a phosphorous fixing medium. They report that this medium consisted of an equimolar mixture of MgO and CaCO ₃ and had a particle diameter of 0.5 to 2 mm with a mean value of 1.5 mm. The examples show that 0.5 mm diameter grains had a phosphorus absorption capacity of 10 mg PO ₄ per 1 g medium. The acid regeneration of the medium was unsuccessful due to the disintegration of its particles. Since the absorption capacity of phosphorus was found to depend on the specific surface area of the medium, the powdered material with a specific surface area of 1200 m ² / m ³ was tested and compared with a standard material with a specific surface area of 3900 m / m 3. The use of this material doubled the absorption of phosphorus, giving an approximate absorption capacity of 20 mg PO ₄ per 1 g medium. The authors state that the finely powdered medium (mean diameter below 100 microns) is to be fed to a slightly axially stirred tank for incorporation directly into the bacterial flakes. This helps to retain solids during their passage through the sedimentation tank. The amount of sludge produced does not significantly increase, and the pH of the outgoing water allows it to be discharged directly into the watercourses.

Obviously, these authors have found out what role this process plays by helping the final deposition of the flakes formed during biological purification while also removing phosphorus.

SUMMARY OF THE INVENTION

The present invention, knowing the usefulness of the previously known phosphorus removal methods, is based on the surprising finding that the use of magnesia in a relatively fine particle grain in a moderately stirred bed provides a highly efficient removal of phosphorus from contaminated water.

The invention relates to a process for removing phosphorus in the form of orthophosphate from contaminated water or de-carbonated water, which comprises adding an effective amount of calcium ions, adjusting the pH to between about 5.0 and 8.0, introducing purified water into the reaction vessel, contacting the fine particulate magnesium oxide in an amount sufficient to trap the magnesium phosphate formed while stirring the magnesium oxide and removing the purified phosphorus-free water from the reaction vessel.

Using this method, the orthophosphate concentrations can usually be reliably reduced from several hundred mg / l (as phosphorus) in contaminated water to 0.3 mg / l (as phosphorus) in the effluent or below, in a continuous design. In addition, the final solid product leaving the reactor is significantly enriched in phosphorus and - depending on the particular process operation - may contain 5 to 18% or more elemental phosphorus. This represents an additional high quality source of this element, usable for its recycling to various productions.

Phosphorus-containing waters may come from a variety of sources, including effluents from the primary, secondary and tertiary stages of wastewater treatment, industrial waste water or process streams. These waters may also contain heavy metals that would be harmful to the environment if they were not removed before the final runoff. However, in addition to phosphorus removal, the essential property of magnesium is to cause the absorption of heavy metals in the form of hydroxides. For waste waters containing significant amounts of organic or insoluble particulate matter, in some cases aerobic or anaerobic treatment is required, possibly followed by sand filtration to remove particulate matter.

Although various combinations of calcium salts can be used to supply calcium to the water, two sources of gypsum - CaSO ₄ .2H ₂ O - and calcium chloride are readily available. The amount of calcium to be supplied to the wastewater depends on the desired concentration at the process outlet, usually in order to reduce the phosphorus concentration from the input value (Pin) in mg / l to less than 0.2 mg / l at the outlet, is added to the feed with at least [2 x Pin + 80] mg / L or more calcium while maintaining the pH in the reactor above 9.7.

Ί

In a preferred embodiment, the feed stream is enriched with calcium to the desired concentration by separating a portion of the stream (typically 10-20% of the total) into a small stirred tank to which solid gypsum is added in a suitable ratio. This separate stream saturated with calcium sulfate is then combined with the main stream. When a more soluble calcium salt than CaCl ₂ is used as the calcium source, it is not necessary to separate the addition stream and the concentrated salt solution can be simply added to the inlet in the desired ratio.

In some cases, it is desirable that the phosphorus-containing waste water be removed prior to contact with the carbon dioxide magnesium. Both bicarbonate and dissolved carbon dioxide can adversely affect the deposition of phosphorus by a competitive reaction with calcium, which leads to the deposition of CaCO ₃ - calcium carbonate. One way to remove carbon dioxide is to adjust the pH of the wastewater to 4 to 4.5 with dilute mineral acid followed by air stripping. Sulfuric acid is excellent for this purpose. After this process, it may be desirable to adjust the pH of the water to higher values by the slight addition of sodium hydroxide, ammonia, lime or other suitable base.

The removal of carbon dioxide may be carried out before or after the addition of calcium. However, it is preferable to remove carbon dioxide before saturating the feed with calcium.

When the treated water is treated in this way, it is fed to the reaction vessel, preferably to the bottom thereof. It has been found that a cylindrical shape with a conical bottom is the preferred shape of the container, since such an arrangement promotes an even distribution of the mixed inflow. However, any shaping of the container to allow distribution of the mixed inflow can be used. The amount of fine magnesia that has previously been introduced into the vessel is mixed gently with an ascending current, while some horizontal movement of the magnesium bed is achieved by slowly rotating the agitator. The reason for this mixing is to prevent the formation of a short connection channel between the inlet and the outlet, which would result in a shortening of the contact time. As the liquid passes through magnesium, phosphorus deposits on the particles in the form of an amorphous precipitate of calcium phosphate.

Magnesium, as used herein, suitable for the reaction bed should be in the form of fine particles in order to maximize the activity of this component. A particle size of about 200 microns, but particularly about 100 microns or less is preferred. However, if the particles are too small, a portion of the bed is lost by leakage to the effluent. For example, it has been found that commercially available Magnesium Enviromag 75 - available from ICI - with a mean particle diameter of about 10 microns is usable and passes through a 75 micron sieve with such a size distribution. 40 m ² / g

MgO 95 CaO 1.5 SiO ₂ 0.5 Al2O3 F θ2θ3

It should be noted that the particles that 95% by weight of this material has a specific surface area of up to 97% to 2.8% to 0.8%

0.1%

A 0.1% listing of this material as suitable does not limit the invention in any way and does not mean that only himself can be used from the sources of magnesium oxide. Magnesium oxides or materials containing them in significant amounts from other sources in hydrated or non-hydrated form and with different particle distributions and mixtures of magnesium oxide containing chemicals can also be used in the method of the invention to remove phosphorus from waste water or other process streams.

When the process according to the invention is carried out with a high flow rate, some of the solid material can be removed from the bed. If the solids concentration at the outlet of the device is unacceptable, these dispersed solids may be filtered off in the usual manner, optionally in conjunction with flocculation and precipitation, using anionic polymers as fine particle flocculation promoters.

With regard to phosphorus deposition, it has been found that the volume of the fluidized bed begins to grow continuous with increasing fluid flow through the reactor, i. depending on the surface loading and also on the phosphorus concentration in the feed stream. However, after some time, the fluidized bed volume stabilizes and the bed's clotting capacity increases significantly. This bed stabilization appears to be dependent upon the degree of hydration of the magnesium oxide and its conversion to magnesium hydroxide. Increasing the bed volume, especially if contaminated water with high phosphorus content is intensively pumped into the reactor, requires some caution when starting the process until the bed stabilizes. However, it has been found that the process can be carried out at surface fill values of 0.5 to 2.0 meters / h at a concentration of up to 90 mg / l of phosphorus at the inlet.

Polluted waters with a higher phosphorus content can also be treated successfully, but in such a case a smaller volume per unit time must be pumped into the inlet, depending on the total reactor volume.

After passing through the reaction zone and the settling process, the phosphorus-free treated water is discharged from the reactor, and is preferably removed from the top of the reactor. Preferably, the mean residence time is 0.5 to 2 hours. At this point, it is important to note that this residence time is governed by the settling properties of the reaction mixture rather than by any inherent slower reaction rate. The thermodynamic chemical equilibrium is reached relatively quickly and the concentration of the orthophosphate solution at the reactor outlet is given by:

The total phosphorus concentration at the outlet = 0.46 x (1 + IC ¹² 'P ^{37 H)} / C and ^{1 »5} (expressed in mg / 1), wherein the symbol Ca" in this respect is the concentration of calcium to the effluent measured in mg / 1st

Overview of the drawings

Fig. 1 is a flow diagram showing one embodiment of the method of the invention.

Fig. 2 is a flow diagram illustrating a second embodiment of the invention.

Fig. 3 is a flow diagram illustrating a third embodiment of the invention.

Fig. 4 is a flow diagram illustrating a fourth embodiment of the invention.

Fig. 5 shows schematically an embodiment of the method used in Examples 1 and 2.

Fig. 6 schematically shows a reactor of larger dimensions and an associated settling tank as used in Examples 6, 7 and 8.

Fig. 7 is a graph representing the method of Example 1.

Fig. 8 is a graph representing the method of Example 2.

Fig. 9 is a graph representing the method of Example 4.

Fig. 10 is a graph representing an evaluation of the efficiency of the method of Examples 6 and 7.

efficiency evaluation

DETAILED DESCRIPTION OF THE INVENTION

Although the particular dimensions of the reactor of FIG. 1, which has been used in the vast majority of the examples, will be given below, other reactor dimensions and shapes are also suitable. The first simple variant is a vertical cylindrical reaction vessel equipped with a suitable stirrer, which again operates with an ascending stream.

In this form, the function of the base vessel is dual, since the vessel forms both a reaction zone for various important ongoing reactions and at the same time serves as a settler. Accordingly, it must be operated with an upward flow at a rate less than the settling rate of most of the magnesium / calcium phosphate containing particles.

In FIG. 1 depicts one embodiment of the method of the invention. In this embodiment, the phosphorus-containing stream 10 enters the vessel 13. The acid 14 is added to the phosphorus-containing liquid 12 in an amount sufficient to adjust the pH to 4.5. Air 11 is then bubbled through liquid 12 to completely remove carbon dioxide.

When the carbon dioxide is removed from the liquid 12, the liquid is pumped to the lower, conical end 23 of the reaction vessel 22 by the pump 21. Part of the carbonated liquid 12 is discharged via pump 18 to another vessel 16. In this vessel 16 is added with stirring gypsum 15 so as to obtain the necessary calcium concentration in the liquid 17. This is then passed through line 20 back to line 19 to effectively obtain a calcium-enriched stream that is fed to the vessel 22.

Magnesium 24 is added to the vessel 22 to form a bed 26 in which, while stirring with the stirrer 25 in conjunction with the upward flow caused by the pump 21, maximizes contact between the magnesium oxide bed and the stream of phosphorus-containing liquid. It should be noted, however, that the mixing intensity is limited to the minimum necessary to prevent a short inlet-to-outlet connection channel.

After the phase of contact between the liquid stream and the magnesia, the liquid is withdrawn at a depth 27 via line 28 into the stirred vessel 32. In this vessel, an anionic polymer 30 is added to flocculate solids in the liquid stream 31 containing them and mixed with stirrer 29. The flocculant 33 is collected over time and is removed 35 together with the phosphorus-saturated magnesium 36 from the reaction vessel 22 for subsequent dewatering to a solid residue 34 to form a liquid stream 37. The solid residue 34 is discharged as a phosphorus-enriched product 39 suitable as a fertilizer. Stream 37 is returned to line 28.

The phosphorus-free stream 40 after flocculation in the vessel 32 passes through a vessel 41 where carbon dioxide 43 is blown into it until the pH of liquid 42 in the vessel is about 7-8. After the pH adjustment, the phosphorus-free stream is discharged as effluent 44.

The embodiment shown in FIG. 2 has the features already mentioned in the embodiment of FIG. 1. However, the main difference in this embodiment is that of FIG. 1, the gypsum bed 45 is added directly to the bottom of the reaction vessel below the magnesium bed 24, instead of supplying the gypsum into a separate carbon dioxide-free stream.

In carrying out the process, the incoming liquid flows first through the gypsum layer, where it is enriched with calcium ions, and then leads to the upper region where the high pH and large surface area of the magnesium oxide particles allows deposition of amorphous calcium phosphate. This mixed fluid bed arrangement typically requires the addition of calcium to the stream of purified liquid prior to entering the reactor. In carrying out this variant of the process according to the invention, it has been found that the heavier calcium sulfate layer remains substantially separated from the lower-density upper region containing magnesium oxide with the captured phosphorus compounds. As the gypsum is quite water soluble, it may be necessary to occasionally add a lower layer of fresh gypsum.

The main disadvantage of this variant of the embodiment of the method according to the invention is that the outgoing stream carries, due to the water solubility of the gypsum, quite a large amount of gypsum. This can be particularly harmful to waters with high salinity. Second, there is contamination of the solid phosphorus-containing product being withdrawn from the reactor with some unused gypsum.

Another viable assembly is one in which the inlet (supplemented with calcium) is pumped into a continuously stirred reactor, which may be of relatively small dimensions, containing a suspension of magnesium. The entire reaction mixture is then passed through a conventional settling and drying apparatus (for example, a conical tank or a sloping flat settler), where after clearing of the solids a clear, phosphorus-free solution is removed from the system. The concentrated phosphorus-containing sludge is then pumped either all or part of it back to the reaction vessel. As the magnesium oxide-containing reagent is consumed, fresh MgO should be added to the reaction mixture in a small amount.

When proceeding in this manner, the process consists of a separate reaction and settling step, and in some cases, the process may offer advantages over the one-step process described previously. In this embodiment, it may be advantageous to use a rod stirrer to stir the space in the reaction vessel.

In FIG. 3 schematically shows an embodiment with the features of FIG. 1 and 2 using the same reference numerals. Specifically, the contaminated calcium-enriched water from which the carbon dioxide has been removed is pumped by means of a pump 21 to a reaction vessel 100 containing a magnesium bed 101. This is continuously stirred by a stirrer 102. After some time, the saturated magnesia reaction mixture is allowed to flow A conventional settling or thickening device, which in this case is a conical tank 104. After the solids 106 have settled, the clear, phosphorus-free liquid is discharged by a stream 107 into the flake removal tank 32, as described above. The solid 106 is collected as a thickened sludge and returned to the tank 100 or, when the magnesium is fully saturated, is led to a dewatering stage as described above.

In another embodiment, the tendency of the bed to increase its volume due to phosphorous saturation is preferably eliminated. In this method, the feed waste water - preliminary, if desired, calcium enriched and / or de-carbonated - is fed at the end of the bottom, which is preferably conical in shape, to ensure uniform distribution of the liquid, cylindrical reactor filled with magnesium oxide. As the process proceeds, the reaction mixture is increasingly enriched for trapped phosphorus and, as a result, the rate of the ascending stream increases throughout the bed volume. The bed volume is increasing and occupying an ever larger proportion of the total reactor volume, and the demarcation line between the reacting mass and the clear supernatant, now largely free of orthophosphate, gradually increases in the cylinder.

In this embodiment, the bed is allowed to rise to a certain height of the reactor, and the phosphorus-enriched sludge is withdrawn at a suitably treated outlet, either continuously or in appropriate batches at specified time intervals. Since the bed volume grows much slower than the incoming feed stream, sludge typically needs to be collected in a volume of only 3-5% of the total flow rate and the bulk of the flow rate exits the reactor through the outlet located at the top of the reaction vessel. It is of course necessary to replace the lost bed volume with fresh magnesium oxide, either in a continuous or semi-continuous manner.

With the appropriate choice of the MgO-based feed and the interdependent feed of the contaminated water and the orthophosphate concentration, it is possible to design the reactor in such a way that the phosphorus concentration in the solids of the sludge taken off has any desired value.

In this embodiment, there is no need to recycle the sludge, with the sole exception that the supernatant resulting from the necessary dewatering of the sludge is returned to the baseline feed stream.

This embodiment is schematically shown in FIG. 4, using the same reference numerals as in the embodiments of FIG. 1 and 2. Specifically, in this case, as the volume of the carbonated bed grows, the outflow 38 flows from the container 22 through the outlet 201 at its upper end. Fresh magnesium 202 is added to the vessel 22 to maintain the bed volume again.

By slightly increasing the inflow of liquid, the properties of the reacting mass, particularly important for settling, may change and the fine particles will cross the area where they are to be maintained. This can be solved by means of a final filtration step optionally supplemented by polymer supplementary flocculation.

In another case, it has been found that it is possible to omit the recycling of a small amount of the formed flakes back into the main reaction vessel because polymer residues reduce the activity of magnesium oxide.

In all cases, the reactor process is carried out at a high pH, wherein the magnesium oxide is contacted with a phosphorus-containing stream at a pH of about 8 to 12. Preferably, the pH is from about 9 to about 11. The pH choice is determined by the solubility of magnesium oxide. If desired, the pH of the incoming polluted water at the inlet can be lowered by dispensing a dilute mineral acid - for example sulfuric acid - or fi16 treatment with carbon dioxide - i. rekarbonáciou.

It will be apparent from the above description that the present process is fundamentally different from the prior art processes, and specifically the present invention is based on the use of finely ground magnesia and a reaction vessel in which the magnesia is present as agitated rather than fixed. the bed.

Both the process described by Kaneko et al. And the process described by Hirasaw et al. Utilize fixed bed granular media, Roques et al. it uses a partially fired dolomite material that is chemically different from the magnesia of the present invention.

It should further be noted that Roques et al. report the addition of partially fired dolomite as the final stage of biological waste treatment. In addition to dissolved phosphorus, a considerable amount of organic material is present at this stage. This organic material adsorbs the medium they use, thus enveloping the active surface and consequently reducing the ability to capture large amounts of phosphorus. This fact was recognized by the authors when referring to semi-fired dolomite as an additive to deposited biological flakes.

Essentially, their process is one of the methods of improving the settling properties of the waste to be treated, in which part of the phosphorus can also be removed.

In order to better understand the nature and scope of the present invention, laboratory-scale experiments using various reactor configurations are described below. These experiments were carried out on a laboratory scale design on a small apparatus capable of handling a flow rate of 50 l / h.

Example 1

This laboratory experiment is characterized by the flow diagram of FIG. 1 and using the device shown in FIG. 5. This is the first experiment that generally served to define and determine the operating conditions that were then used in subsequent experiments.

The apparatus used in carrying out this experiment is further described with reference to FIG. 5th

Inlet 320, devoid of carbon dioxide and containing phosphorus and enriched with calcium ions, was pumped through the bottom 323 conical

A 2.5 liter reactor using a peristaltic pump (not shown). The reactor was previously filled with magnesium oxide to form bed 324 to a height of 326. To prevent the bed 324 from entering the feed line, it was equipped with a glass wool plug 321 at the bottom 323 of reactor 322.

After some time the level of total liquid in the reactor rose to a height of 327. At this point, the outflow was started by gravity 328 through a plexiglass tube 329 containing a glass wool insert 339.

After filtration, the phosphorus-free stream 331 was led to a 3 liter collection vessel 332. The filtered and phosphorus-free effluent was then discharged from the collection vessel 332 via an outlet 340.

The first experiment using this device is described in detail below:

The effluent from the tertiary stage of the Quakers Hill Sewage Treatment Plant (near Sydney, Australia) was enriched by the addition of 18 to 1 mg / l phosphorus as disodium hydrogen phosphate so that the phosphorus concentration at the reactor inlet was between 10 and 40 mg / l. The calcium concentration was also supplemented by the addition of CaCl ₂ with a usual calcium ion concentration between 70 and 140 mg / L.

Bicarbonate and dissolved carbon dioxide, which could interfere with phosphorus capture by competing reaction with calcium ions to form calcium carbonate (CaCO ₃ ) in the reactor, were present in the effluent at a concentration of about 120 mg / l HCO ₃ and were stripped off . The pH was then lowered from about 7 to about 4 to 4.5 using dilute sulfuric acid followed by about half an hour of bubbling air through the aquarium aerator. The pH was then adjusted back to 6.5 to 7.5 with sodium hydroxide solution. However, a basic material other than lime or ammonia solution could be used. Note that the subsequent treatment of the MgO bed would neutralize any residual acid. At the same time, however, this would lead to a partial re-dissolution of the already captured phosphorus and could cause problems in reactor interpretation and process efficiency.

The liquid 320 was then pumped to the bottom 323 of the conical reactor 322 using a peristaltic pump at a flow rate of between 500 and 1,000 mg / h. The reactor used had a conical bottom with a volume

1.5 liter cylindrical 1 liter volume, giving a total reactor volume of 2.5 liters. At the start of the experiment, the reactor was initially filled with 100 g of finely divided magnesium oxide 324 commercially available as Enviromag-75 (ICI) and into the bottom of the conical section. a fine glass wool insert 339 was inserted to prevent flow back into the inlet pipe.

The reactor was also equipped with a stirrer 325 of a shape that allowed slight agitation of the reactor contents. This mixing has been found necessary to prevent the formation of a channel in the bed that would cause a connection in the system for a short time19 and significantly reduce the efficiency of the process. The stirring speed was about 40 rpm.

During operation, the volume of the reactive bed visibly increased by about four times as a result of the upward fluid flow and gentle agitation. Over time, as the phosphorus content of the bed increased, the total bed volume also gradually increased.

Upon reaching 327, the effluent from reactor 322 was allowed to exit through outlet 328 into a downward gravity filter formed simply from a 30 mm diameter Plexiglas tube 329 filled with a layer of glass wool 339, into a three liter collection tank 332.

This filter simply collected particles discharged from the reactor and prevented them from reaching the final outflow 380. It was found that this device was only necessary for the first 200 hours of operation, since after that time the settling properties of the particles in the reaction bed improved as expected due to the fact that they were then coated with calcium phosphate and the reactor was no longer disturbed by the influence of small or non-solid materials, since they were washed away from the reactor vessel.

The experiment was carried out for 1,500 hours at an inlet P between 2 mg / l (untreated effluent leaving the sewage treatment plant) and 45 mg / liter. The system was operated at ambient temperature, i. 20 to 25 ’C.

The effluent samples were taken uniformly from the collection tank and analyzed for the orthophosphate content by the molybdenum blue method described in the 4500-P standard published in r. 1992 American Water Works Association in Standard Methods for Examination of Water and Wastewater.

The total orthophosphate effluent (which is contained both in the fine particles that have been washed out of the reactor and is also dissolved as true ionic phosphorus) from the device according to the invention was also evaluated simultaneously with the concentration of the right solution components after filtration with 0.45 micron filter.

The course of the reactor efficiency is evaluated using the graphs in FIG. 7 and data on the increasing proportion of washed away fine particles can be obtained by comparing graphs of orthophosphate concentrations in filtered and unfiltered effluents. It should be noted that for greater clarity and differentiation of the graphs of phosphorus concentrations in filtered and unfiltered effluents, the values for unfiltered effluents are plotted as double. The amount of phosphorus removed was calculated based on the phosphorus content of the filtered effluent. The gentle mixing was maintained continuously and an adequate amount of calcium was still available for phosphorus capture. The system - as it worked in this laboratory study - deprived the contaminated water of 95% of the orthophosphate contained in the input, even if the contaminated water contained more than 40 mg / l of phosphorus.

At the end of the experiment, the solids contained in the reactor were removed, dried to constant weight at 105 ° C and analyzed. The dry weight was 160 g at 27% after ₄ (i.e. 9% p), 14% Ca and 50% (unused) magnesium hydroxide.

Example 2

Contaminated water obtained from the effluent from the tertiary stage of the Penrith Sewage Treatment Plant - near Sydney, Australia - containing about 4 mg / l of phosphorus (as orthophosphate) was added an additional orthophosphate in such a quantity that the total phosphorus concentration (calculated as element) 20 mg / l. Furthermore, calcium chloride or calcium sulfate (gypsum) was added so that the waste contained about 120 mg / l HCO ₃ , with no carbon dioxide being removed at the inlet of the MgO reactor in this experiment, since no negative effect of this component on the process was found.

The treated feed was pumped into the reactor again filled with 100 g MgO and the experiment was carried out in the same manner as in Example 1 and in the same apparatus for 1034 hours.

Surprisingly, it was found that the tributary had no adverse effect on phosphorus plundering which was so orthophosphate dissolved in that the presence of carbonate on the efficiency of the process was high that more than 99% of the purified liquid was trapped by the MgO bed. Titrations of the incoming and outgoing liquids showed that only about 20% of the bicarbonate present in the feed stream was captured as calcium carbonate in the reactor. However, the effluent was supersaturated with CaCO ₃ and certain calcite deposits were formed in the recovery tank.

In this respect, however, it has been found that, while maintaining moderate mixing, some calcium carbonate is deposited at the bottom of the reactor, particularly on a glass wool plug, which reduces flow and contributes to the formation of a channel in the main body of the MgO bed.

The course of the experiment is shown in FIG. 8th

At the conclusion of this test, it was found that the dry mass of the solids in the reactor was 182 g and contained 26.7% PO ₄ (ie 8.9% P), 15.2% Ca and 49% magnesium hydroxide.

Examples 1 and 2 have shown the applicability of a method for removing orthophosphate from a tertiary effluent from a sewage treatment plant. Further analysis of the inlet and outlet of these ex22 media revealed that some of the fluoride and some of the silica contained in the waste water were also removed. Thus, in Example 2, the fluoride concentration was reduced from 1.0 mg / L in feed to 0.7 mg / L in effluent and silicon was reduced from 3.5 to 2.5 mg / L.

Example 3

This experiment was performed on the same apparatus as the previous tests, but only secondary treated wastewater from the Round Corner Sewage Treatment Plant near Sydney, Australia was used. It is a small treatment plant in which only a two-stage treatment of biologically activated sludge is used and, accordingly, the feed to the MgO reactor contained a higher amount of organic material and suspended solids compared to previous tests. This feed also contained about 8 mg / l of total phosphorus, of which 7 mg / l was in the form of orthophosphate, as no specific phosphorus removal method is used in this treatment plant.

For this test, carbon dioxide was not removed from the inlet and the inlet pre-reaction only consisted of adding 80 mg / l calcium, for which calcium chloride was used.

In this test, the reactor was charged with 50 g of MgO and then pumped with contaminated pre-treated water at a flow rate of between 500 and 1000 ml / hour. The test was performed for over 142 hours and the dissolved orthophosphate was removed by more than 95%. In this test, however, a fairly large portion of the fine particulate material was washed out of the reactor and more than 70 mg / l of suspended solids were detected in the reactor effluent. More than 20% of the phosphorus trapped was fed from the reactor just in conjunction with this fine material. However, when the effluent was treated with an anionic polymer additive to flocculate and settle this material, in this case 2 to 3 mg / L of Alfloc 627 (manufactured by Nalco in Australia) was used at the usual effluent pH

10.2 to 10.5, the phosphorus orthophosphate concentrations (without filtration) decreased to 0.2 to 0.3 mg / L.

It was also not entirely clear whether all solids were washed out as a result of the properties of the secondary effluent mixture, or whether too much agitation was not involved in these losses during the experiment.

In any case, this test has shown that the orthophosphate can be effectively captured in the form of calcium phosphate from the effluent from the secondary stage of the wastewater treatment plant and also that partial leaching of the contaminant can be kept under control by using anionic polyflocculants.

The solids remaining in the reactor after this experiment had a mass of 67 g and contained 2.2% P (6.6% PO ₄ ). As in the previous two examples, no biological growth occurred in the bed, even in this case of the origin of the waste water.

Example 4

This experiment was carried out using a bed of MgO and gypsum to confirm the feasibility of the process of the invention by omitting the calcium pretreatment step by including this step in the main reactor. The process steps are shown in the flow diagram of FIG. Second

The experiment was carried out using the same apparatus as in Examples 1 to 3, but an artificial feed solution containing only orthophosphate at a concentration of 35 to 40 mg / l and a pH of 6.5 to 7 was prepared.

The feed contained no calcium except for a small amount present in the water from which it was prepared, which was about 5 to 10 mg / L. Again, pumping into the reactor at a flow rate of 500 to 1000 ml / hour was used. The reactor was initially charged with the experiment 100 g of gypsum (time0 ₄ .2H ₂ O), which was covered with 100 g of magnesium oxide. As the stream of purified liquid passes through the gypsum bed, it is enriched with Ca ions, which then allows the storage of calcium phosphate when the solution encounters the surrounding environment of the high pH magnesium oxide bed.

During this experiment, it was found that gypsum having a higher density (SG - 2.30) always remains below the layer containing magnesium oxide and calcium phosphate, whose SG is significantly lower, apparently relative to the calcium phosphate trapped.

Due to the relatively high solubility of calcium sulfate (about 2 mg / l in saturated solution at 25 ° C), it was found necessary to replenish the gypsum bed from time to time and therefore, in addition to the gypsum present at the start of the reaction, .

This experiment was carried out for 514 hours, after which 17.1 g of elemental phosphorus were collected in the bed. The dry weight of solids remaining in the reactor after the reaction was 260 g, of which 19.5% PO ₄ (6.5% P), 20%

Ca, 47% Mg (OH) ₂ and about 21% undissolved calcium sulfate.

The course of the reaction is shown in FIG. 9th

Example 5

This experiment was performed with a high influx of orthophosphate, i.e., 20 to 60 mg / L, supplemented optionally with calcium, i. 120 to 230 mg / l, to determine the concentration level to which the solids remaining in the reactor can be enriched for elemental phosphorus when the concentration of phosphorus to be removed in the feed is very high.

The reactor was the same as in the above-described examples and again at a flow rate of 500-1000 ml / hour. The assay was performed for 1,218 hours during which 39.4 g of elemental phosphorus (present in solution as orthophosphate) was passed through the reactor and 37.7 g of elemental P was collected in the bed. This represents 95.6% phosphorus removal.

The solids remaining in the reactor after the test had a dry weight of 205 g and contained 45% PO ₄ (18% P), 29.7% Ca, and only about 16% magnesium hydroxide. This product is a raw material with a high phosphorus content.

In the following three examples, the apparatus used was much larger to verify the efficacy of the method of the invention on a larger scale and also to determine bed properties that are related to its growth as the phosphorus saturation of the bed increases.

A reactor 422 with a conical shaped bottom 423 was used as before, but this reactor had a much larger volume of about 30 L. The reactor dimensions are shown in FIG. 6. Feed 420 was pumped to the conical section at a flow rate of 25 to 28 L / h, corresponding to a reactor surface of 1 m / h. The magnesium oxide that had previously been introduced into the reactor created a bed 424 to a level level at a depth of 426. This was maintained in the reactor by a flexible check valve 429. The inlet 420 was controlled by a manifold 421. In these experiments the reactor effluent was drained 428 to a 5 liter settler 434 to which anionic polymer 431 (Alfloc 627) was added at a rate of 2-3 mg per liter of effluent. This treatment successfully flocculated a small amount of particulate material 433 washed out of the reactor effluent. A stirrer 430 was used to stir this tank. The flocculated and phosphorus-free content 432 was discharged via line 436.

Example 6

This example uses the flow diagram of FIG. 1, using an artificial inflow having a phosphorus content of 15 mg / l and supplemented with Ca to 110 mg / l. The reactor was charged with 50 grams of magnesium oxide and the feed was pumped into the system at a flow rate of 25 L / h for 30 hours.

In this experiment, the concentration of dissolved phosphorus in the effluent filtered through a 0.5 micron filter never exceeded 0.2 mg / l, while the phosphorus concentration in the unfiltered flocculated effluent was usually 0.2 to 0.5 mg / l. The residence time in the settler was very short - about 12 minutes - and by extending the settling time it is likely to achieve a better outflow quality.

During this experiment, it was observed that the bed volume in the reactor increased until about 7 grams of phosphorus was stored and then remained approximately constant.

At the end of the test, the solids from the reactor after drying had a weight of 86 grams and contained 11.7% P (35% PO ₄ ) and 25.6% Ca.

The dry flocculated material was removed from the settler and had a total weight of 3.3 g.

Example 7

The experiment was conducted in a manner same as Example 6 except that the influent containing inter mg / 1 of phosphorus (45 mg / 1 P0 ₄₎ even 120 mg / 1 of sodium.

As in Example 2, it was found in this experiment that the carbonate in the feed had no negative effect on phosphorus removal, reaching a dissolved phosphorus concentration never exceeding 0.26 mg / L. However, the presence of carbonate had a significant effect on the reaction bed volume, which resulted in significantly smaller bed volumes compared to the example where the carbonate in the feed was absent. This effect is illustrated in FIG. 10 where the volumes of both Examples 6 and 7 are plotted depending on the course of the two experiments.

The effect of carbonate on reducing reactor bed volumes has already been noted in other experiments not mentioned here and could be of some importance in considering aspects relevant to the design of a particular process and equipment and its design.

The solids that were removed from the reactor after the process had a dry weight of 107 g, of which 9.5% P and 28.1% Ca. The increased Ca / P ratio observed in this experiment is due to the relatively small amount of CaCO ₃ deposited together with the calcium phosphate. In this experiment, 7.1 grams of dry flocculated material was removed from the settler.

Those skilled in the art can modify the procedures set forth in the examples without departing from or departing from the spirit of the invention as described in its broadest form. The specific embodiments described are, therefore, to be construed in all respects as illustrative and not limiting.

Claims (20)

Process for removing phosphorus in the form of an orthophosphate from an aqueous stream, characterized in that an effective amount of calcium ions is added to the stream, optionally carbon dioxide is removed from said stream, the pH is adjusted to a value between 5 and 8, is introduced into the reaction vessel in such a way that it flows upwardly, in the absence of the anionic polymer, contacting the stream with particles of particle size less than about 200 microns in an amount sufficient to fix the calcium phosphate so formed, the pH of the solution contacted with the magnesium oxide is about 8 to 12, while mixing the magnesium oxide just so as to prevent the formation of channels sufficiently and the stream of reduced phosphorus concentration is removed from the vessel. Method according to claim 1, characterized in that calcium is added to the stream by means of a gypsum bed located at the bottom of the reaction vessel. The method of claim 1, wherein the calcium-containing liquid stream is fed to a reaction vessel containing a continuously stirred suspension of magnesia, which, when appropriately saturated with phosphorus, can flow to a settling or thickening device to form a thickened condensate. , a phosphorus-saturated slurry that is returned to the reaction vessel, and a liquid stream with a reduced phosphorus concentration. Method according to one of Claims 1 to 3, characterized in that the calcium is in the form of a soluble salt, preferably gypsum or calcium chloride. Modified version Valid for the national phase Method according to any one of claims 1 to 4, characterized in that calcium ions are introduced into the stream by separating a portion thereof into the tank, adding calcium to the purified liquid contained therein and returning the calcium-enriched stream to the stream that is fed to the reaction vessel. . Method according to one of Claims 1 to 5, characterized in that carbon dioxide is removed from the liquid stream and then calcium ions are introduced therein. The method of claim 6, wherein the removal of carbon dioxide is achieved by adding sufficient acid to the liquid stream contained in the vessel to achieve a pH of 4 to 4.5 and stripping the stream with air. The method of claim 7, wherein an alkaline substance is added to the carbon dioxide stream to adjust the pH to about 5 to 8. 9. The method of claim 8, wherein the magnesium oxide has a particle size of less than about 100 microns. The method of claim 9, wherein the particle size distribution of the magnesium oxide is such that 95% by weight passes through a 75 micron screen. The method of claim 10, wherein the magnesium oxide has a mean particle diameter of about 10 microns. Method according to one of claims 1 to 11, characterized in that the magnesium oxide is in the form of magnesium. Method according to any one of claims 1 to 12, characterized in that the phosphorus-reduced stream is subjected to filtration and flocculation. Modified version Valid for the national phase The method of claim 13, wherein the flocculation is achieved by adding an anionic polymer to a liquid stream having a reduced phosphorus concentration. Method according to any one of claims 1 to 14, characterized in that the magnesium oxide or magnesia, after the phosphorus enrichment, is drained and dewatered, the resulting liquid fraction being returned to the reaction vessel and the solid fraction being discharged for use as fertilizer. Method according to any one of claims 1 to 15, characterized in that the pH of the liquid stream has a value of 5 to 7 before the addition of calcium. The method of any one of claims 1 to 16, wherein the pH of the contents of the reaction vessel is about 9 to 11. The method of claim 1, wherein the reaction vessel comprises a rod stirrer. Method according to any one of claims 1 to 18, characterized in that the phosphorus-containing aqueous stream is a primary, secondary or tertiary effluent from a water treatment plant. The method of any one of claims 1 to 19, wherein the amount of calcium ions added to the liquid stream to reduce the phosphorus content of the reduced phosphorus concentration stream to less than 0.2 mg / L is sufficient if greater than or equal to [2 x Pin + 80] mg / L, where Pin represents the phosphorus concentration in mg / L in the liquid stream.