Production of potassium phosphate

Abstract

An apparatus (100) for producing fully soluble, pure and defined mono- or dipotassium phosphate, comprising an extraction section (10), a stripping section (20) and a finishing apparatus (90). The extraction section performs liquid-liquid extraction of phosphate between a feed solution (1) comprising phosphoric acid. The stripping section performs liquid-liquid extraction of phosphate between a phosphate-laden solvent and a stripping solution (4). The phosphate-depleted solvent is recycled to the extraction section for further extraction of phosphoric acid. The stripping solution is an aqueous potassium phosphate solution. The finishing apparatus comprises a source of potassium base (60), an addition apparatus (70), a cooling apparatus (50), a precipitate remover (40) and a recycling system (80).

Description

TECHNICAL FIELD

[0001] The present invention relates generally to the production of potassium phosphate from phosphorus-containing solutions, and in particular to the production of potassium phosphate from feed solutions comprising phosphoric acid. BACKGROUND

[0002] All water-soluble phosphate salts, such as soluble fertilizers, are derived from phosphoric acid. Phosphoric acid is produced commercially by either the "wet process" or the thermal process. The wet process digestion of phosphate rock is the most commonly used process. The thermal process is energy intensive and therefore costly. As a result, the amount of acid produced by the thermal process is much less and is mainly used for the production of industrial phosphates.

[0003] Phosphoric acid for fertilizer production is almost exclusively based on the wet process digestion of phosphate rock. The process is mainly based on the dissolution of apatite with sulfuric acid. After the rock is dissolved, the calcium sulfate (gypsum) and phosphoric acid are separated by filtration. To produce commercial grade phosphoric acid, a high acid concentration is required and water is evaporated. Depending on the main conditions such as temperature, phosphorous concentration in the slurry, and content of free sulfate, the calcium sulfate is present in a variety of different crystalline forms. The calcium sulfate precipitates as dihydrate (CaSO4-2H2O) or as hemihydrate (CaSO4-1 / 2H2O). The phosphoric acid produced by this process is characterized by a relatively low purity.

[0004] To obtain potassium phosphate salts, commercial grade phosphoric acid with a concentration of about 54% P2O5 is neutralized with potassium hydroxide (KOH) or potassium carbonate (KHCO3 or K2CO3) to form mono-potassium phosphate (MKP, KH2PO4, also known as potassium dihydrogen phosphate, KDP) or di-potassium phosphate (DKP, also known as potassium phosphate dibasic). Di-potassium phosphate has the formula K2HPO4-(H2O) x (x = 0, 3, 6) and is produced by controlling the molar ratio of KOH to phosphoric acid during the neutralization process. MKP and DKP are commonly used as fertilizers, food additives, and buffering agents.

[0005] The neutralization of commercial grade phosphoric acid with potassium hydroxide or potassium carbonate is an exothermic reaction.

[0006] Potassium phosphate is considered a high value fertilizer. There are many reasons why plants need potassium. Potassium enables enzyme activity in plants. It also promotes respiration, transpiration, and enables plants to take up other important nutrients. Potassium is also a key player in the transport of water and nutrients in the xylem of plants. A healthy supply of potassium is essential for the functioning of the transport system of plants. Healthy plants with sufficient potassium supply also promote root growth and help to stop the development and spread of plant diseases. A common source of potassium for fertilizers is potassium chloride (KC1) because it is recovered in this form from ores and natural brines. However, plant species differ greatly in their sensitivity to high concentrations of chlorides in the soil solution, which can be divided into four main groups in this respect: a) chloride-loving crops, b) chloride-tolerant crops, c) partially chloride-tolerant crops, and d) chloride-sensitive crops.

[0007] Chloride-sensitive crops include many fruit and vegetable varieties, as well as special crops such as hops or tobacco. For these crops, potassium chloride cannot be used as a fertilizer, and the chloride ions must be replaced by sulfate or phosphate anions. Phosphorus is an important plant nutrient. Without phosphorus, photosynthesis cannot occur. Phosphorus, as a major component of ATP (adenosine triphosphate), plays a key role in the complex energy conversions that are essential for all life. It is also a core component of DNA and RNA - and is essential for building proteins and other compounds. It is a component of cell membranes and plays a central role in regulating the pH of plant cells. Plants require phosphorus from seedling stage to maturity - and phosphorus has a measurable impact on crop quality and yield. Phosphorus helps plant growth and plant health through the following mechanisms: a) enabling photosynthesis (energy conversion), b) building nucleic acids, proteins, and enzymes, c) promoting root growth, d) strengthening stems and stalks, e) improving flower formation and seed production, f) promoting crop uniformity, g) helping to mature early, h) increasing disease resistance, i) increasing overall crop quality, and j) promoting nitrogen-fixing ability in legumes.

[0008] Due to the above, potassium phosphates are very valuable fertilizers because they consist of both potassium and phosphorus and do not contain any chlorides.

[0009] For several applications such as fertigation, i.e. the application of water-soluble fertilizers in irrigation water, and foliar fertilization, i.e. the spraying of fertilizers on the leaves, fully soluble potassium phosphates are required to avoid clogging of the fertigation devices by undissolved solids. Wet-process phosphoric acid contains a large amount of impurities such as iron, aluminum, calcium, magnesium, cadmium, etc., which form water-insoluble solids after neutralization with potassium hydroxide or potassium carbonate, and thus fertilizer-grade potassium phosphates are not fully soluble in water. Therefore, P fertilizers for fertigation purposes that are fully soluble must be produced exclusively from purified phosphoric acid, which means additional processing.

[0010] The current technology for phosphoric acid purification is based on extraction of the impure wet-process phosphoric acid into an organic solvent (ketones, trialkyl phosphates, alcohols, etc.) and then back-extraction with water to form a purified phosphoric acid, but at a lower concentration, which is then concentrated by water evaporation. Thereafter, the purified phosphoric acid is neutralized with potassium hydroxide or potassium carbonate according to the procedure described above to form a fully soluble potassium phosphate product.

[0011] There are many disadvantages to the current technology for producing potassium phosphate. Phosphoric acid produced from a gypsum filter in the dihydrate process is not suitable for direct use in the manufacture of potassium phosphate salts. The acid must be further concentrated by water evaporation to the appropriate phosphoric acid concentration (typically about 54% P2O5). Typically, the concentration of the phosphoric acid is carried out in three stages. The acid from the filter (28% P2O5) is evaporated in a single stage vacuum evaporator to 40% P2O5. The acid is then clarified to remove precipitated solids and then the clarified acid is concentrated in two stages to 54% P2O5. The concentration between stages is about 48% P2O5. The 54% P2O5 acid is used to produce potassium phosphate according to the procedure described above.

[0012] The concentration of the acid by evaporation is a very energy intensive process. Depending on the production conditions, the amount of steam required to concentrate the phosphoric acid typically varies between 2.5 and 5 tons of steam per ton of phosphorus. If the phosphoric acid is purified by solvent extraction, the energy requirement is about 7 tons of steam per ton of phosphorus. The energy requirement for concentrating the phosphoric acid is a major production cost. Expensive equipment such as steam distribution systems, evaporators, off-gas scrubbers, condensing systems, cooling water systems, waste liquid treatment systems, and acid storage facilities are necessary for the production of commercial grade phosphoric acid. About 50 tons of cooling water are required to condense one ton of steam. In a barometric condenser, the steam is in direct contact with the water and, therefore, the impurities in the steam contaminate the cooling water, which results in a large amount of contaminated waste water. In addition, the neutralization of the phosphoric acid with potassium hydroxide or potassium carbonate requires additional equipment. The production of industrial quality potassium phosphate requires additional processing steps as described above.

[0013] U.S. Patent 4,132,540 describes a process for removing solvent from the raffinate of an extracted phosphoric acid. The residual solvent is removed from the raffinate by the addition of ammonium or alkali or alkaline earth metal cations in an atomic ratio of between 0.1 : 1 and 0.6: 1 of ammonium or alkali or alkaline earth metal cations to phosphorus.

[0014] U.S. Patent 4,311,681 describes a process for separating impurities (such as silica and organic impurities) from an organic solvent by washing with an aqueous alkali orthophosphate solution sufficient to maintain the pH of the solvent-water mixture at from about 9.5 to about 12.5.

[0015] US Patent 4,678,650 describes a process for producing an aqueous alkali phosphate salt by mixing an aqueous phase containing an alkali compound with an organic phase containing phosphoric acid in a volume ratio greater than 1 : 1 and then separating the resulting aqueous alkali phosphate salt from the organic phase.

[0016] US Patent 4,751,066 describes a process for base stripping of phosphoric acid from a water-immiscible organic solvent to produce a sodium phosphate solution.

[0017] US Patent 4,112,118 describes a process for stripping phosphoric acid from an organic solvent with anhydrous alkali compounds of ammonia, sodium or potassium, or with up to 10 moles of water, or with solid dihydrogen phosphate salt to obtain a liquid phase mixture comprising an organic solvent phase substantially free of phosphoric acid and an aqueous phase comprising dissolved phosphoric acid and dissolved dihydrogen phosphate salt of the alkali. A major disadvantage of this process is that the stripping process requires the organic solvent to carry > 60 wt% of a highly concentrated phosphoric acid solution. Therefore, the process requires means for concentrating the phosphoric acid by water evaporation, such as evaporators, steam distribution systems, etc., and energy. Another major disadvantage of this process is that the product is an aqueous solution which requires further processing. SUMMARY

[0018] It is a general object of the present invention to produce fully soluble, pure and defined mono- or di-potassium phosphate, avoiding the problems of the prior art. Additional objects will be discussed in connection with the different embodiments presented herein below.

[0019] The above objects are achieved by the methods and apparatus according to the appended independent patent claims. Preferred embodiments are defined by the dependent patent claims.

[0020] In general, in a first aspect, a method for producing pure potassium phosphate comprises extracting a phosphate salt from a feed liquid comprising phosphoric acid into a solvent by liquid-liquid extraction. The solvent having at least a portion of the phosphate salt is stripped into a stripping solution by liquid-liquid extraction. The stripping solution is an aqueous potassium phosphate solution. The stripping solution loaded with the stripped phosphate salt is separated from the solvent which is at least partially depleted of the phosphate salt. The solvent which is at least partially depleted of the phosphate salt is recycled for further extraction of the phosphate salt in the extraction step. A potassium base is added in at least a portion of the stripping solution. The heat generated upon addition of the potassium base in at least a portion of the stripping solution is cooled. Crystals are removed from the loaded stripping solution. After the step of removing crystals, the stripping solution is recycled for use as a stripping solution input in the stripping process.

[0021] In a second aspect, an apparatus for producing pure potassium phosphate comprises an extraction section, a stripping section, and an end treatment apparatus. The extraction section is configured to perform liquid-liquid extraction of phosphate salt between a feed solution comprising phosphoric acid and a solvent. The extraction section has a first extraction inlet for providing the feed solution, a second extraction inlet for providing the solvent, a first extraction outlet for delivering at least a partially phosphate salt-depleted feed solution, and a second extraction outlet for delivering a phosphate salt-laden solvent. The stripping section is configured to perform liquid-liquid extraction of phosphate salt between the phosphate salt-laden solvent and a stripping solution. The stripping section has a first stripping inlet connected to the second extraction outlet for providing the phosphate salt-laden solvent, a second stripping inlet for providing an input stripping solution, a first stripping outlet for delivering at least a partially phosphate salt-depleted solvent, and a second stripping outlet for delivering an output stripping solution. The first stripping outlet is connected to the second extraction inlet to recycle the at least partially phosphate salt-depleted solvent for further extraction of phosphate salt. The stripping solution is an aqueous potassium phosphate solution. The end treatment apparatus is connected to the second stripping outlet and comprises a potassium base source, an addition apparatus, a cooling apparatus, a precipitate remover, and a recirculation system. The addition apparatus is connected to the potassium source and configured to add potassium base from the potassium base source to at least a portion of the stripping solution. The cooling apparatus is configured to cool heat generated by a chemical reaction when potassium base from the potassium base source is added to at least a portion of the stripping solution. The precipitate remover is configured to separate crystals from the loaded stripping solution. The recirculation system is connected between an outlet from the precipitate remover and the second stripping inlet of the stripping section. The recirculation system is configured to reuse the stripping solution from the precipitate remover as the input stripping solution.

[0022] An advantage of the present invention is that a well-defined fully soluble, pure and well-defined monopotassium phosphate or dipotassium phosphate can be produced in an efficient and economic way by an industrially applicable process. Additional advantages will be discussed in connection with the different embodiments further presented below. BRIEF DESCRIPTION OF DRAWINGS

[0023] The present invention can best be understood by reference to the following description taken in conjunction with the accompanying drawings and table, of which:

[0024] Figure 1 is a flow chart of steps of an embodiment of a method for producing potassium phosphate;

[0025] Figure 2 is a block diagram of an embodiment of an apparatus for producing monopotassium phosphate;

[0026] Figure 3 is a block diagram of an embodiment of an apparatus for producing dipotassium phosphate;

[0027] Figure 4 is a block diagram of one embodiment of an apparatus for producing both monopotassium phosphate and dipotassium phosphate;

[0028] Figure 5 is a block diagram of another embodiment of an apparatus for producing both monopotassium phosphate and dipotassium phosphate;

[0029] Figure 6 is a block diagram of one embodiment of another apparatus for producing dipotassium phosphate;

[0030] Figure 7 is a block diagram of one embodiment of another apparatus for producing monopotassium phosphate;

[0031] Figure 8 is a block diagram of yet another embodiment of an apparatus for producing both monopotassium phosphate and dipotassium phosphate;

[0032] Figure 9 is a block diagram of one example of an apparatus for producing monopotassium phosphate;

[0033] Figure 10 is a flow diagram of steps of one example of a method for producing potassium phosphate;

[0034] Figure 11 is a block diagram of another example of an apparatus for producing monopotassium phosphate;

[0035] Figure 12 is a block diagram of yet another example of an apparatus for producing monopotassium phosphate;

[0036] Figure 13 is a block diagram of yet another example of an apparatus for producing monopotassium phosphate;

[0037] Figure 14 is a graph showing H3PO4 distribution data using an aqueous phase with about 40% H3PO4 (4.85 M H3PO4), 7% H2SO4 (2.8 M), and 2% F (as H2SiF6) and an organic solvent: 80% vol. TBP and 20% vol. kerosene;

[0038] Figure 15 is a graph showing H2SO4 distribution data using an aqueous phase with about 40% H3PO4 (4.85 M H3PO4), 7% H2SO4 (2.8 M), and 2% F (as H2SiF6) and an organic solvent: 80% vol. TBP and 20% vol. kerosene;

[0039] Figure 16Figure 1 is a graph showing the extraction of H3PO4 using 80% vol. TBP and 20% vol. kerosene at O:A 3:1 in a five stage counter current mixer settler setup;

[0040] Figure 17 Figure 1 is a graph showing the extraction of H3PO4 using 80% vol. TBP and 20% vol. kerosene at O:A 3:1 in a five stage counter current mixer settler setup;

[0041] Figure 18 Figure 1 is a graph showing the extraction of H3PO4 using 80% vol. TBP and 20% vol. kerosene at O:A 3:1 in a five stage counter current mixer settler setup;

[0042] Figure 19 Figure 1 is a graph showing the extraction of H3PO4 using 80% vol. TBP and 20% vol. kerosene at O:A 3:1 in a five stage counter current mixer settler setup; 2- , and wherein the content of H3PO4 in the organic / Na2CO3 system is 1.27 M, and the content of H3PO4 in the organic / K2CO3 system is 1.05 M. DETAILED DESCRIPTION

[0043] Some commonly used terms in this disclosure should be interpreted as follows:

[0044] Solvent - liquid phase, typically organic, which preferentially dissolves the extractable solute material in the aqueous solution.

[0045] Extractant - active component in the solvent which enables extraction, typically organic.

[0046] Solvent extraction (liquid-liquid extraction) - separation of one or more solutes from a mixture by mass transfer between immiscible phases, wherein typically at least one phase is an organic liquid.

[0047] Strip - displacement of ions or acid removed from the process solution from the solvent, so that the solvent can be used again.

[0048] Diluent - liquid, typically organic, in which the extractant is dissolved to form the solvent.

[0049] Raffinate - aqueous phase from which the solute is removed by extraction.

[0050] Strip product - aqueous phase in which the solute removed by solvent extraction is recovered.

[0051] It is a further object of the present technology to produce solid potassium phosphate without the need to evaporate water using associated equipment (such as evaporators, vapor distribution systems, etc.) and energy producing heat. To this end, the present disclosure is based on the extraction of phosphoric acid with a solvent immiscible or at least substantially immiscible with water.

[0052] Several water immiscible solvents are suggested in the literature as suitable for the extraction of phosphoric acid.

[0053] The suggested solvents can be generally divided into the following groups: a) phosphonic acid alkyl esters, such as tributyl phosphate, b) amines, such as tri-n-octylamine, c) alcohols, such as isoamyl alcohol, n-pentanol, cyclohexanol, methylcyclohexanol, tert-amyl alcohol, isobutyl alcohol, n-butyl alcohol, heptanol, d) ketones, such as methyl isobutyl ketone, methyl propyl ketone, diethyl ketone, methyl ethyl ketone, methyl n-butyl ketone, e) amides, such as butyl acetamide, f) aldehydes, such as benzaldehyde, g) esters, such as ethyl acetate, butyl acetate, pentyl acetate, cyclohexanone, h) ethers, such as diethyl ether, di-n-pentyl ether, and glycol ethers, such as diethylene glycol.

[0054] All the above solvents are classified as water immiscible. However, most of the mentioned solvents actually have a relatively high solubility in water, which can result in contamination of both the water stream and the final product with traces of the solvent. High water solubility usually requires recovery of the dissolved solvent from the water stream by distillation, which is both expensive and complex. For example, n-butyl alcohol has a water solubility of about 90 grams per liter at room temperature. In addition to the high solubility, several of the mentioned solvents also have other disadvantages, such as flammability and risk of explosion, for example diisopropyl ether.

[0055] Furthermore, several of the mentioned solvents show very small extraction capacity for phosphoric acid below a certain threshold concentration. This means that the feed phosphoric acid must initially have a high concentration, which usually requires concentration of the acid by water evaporation. In addition, phosphoric acid can only be partially extracted with these solvents. A typical example of a solvent having a threshold concentration for the extraction of phosphoric acid is methyl isobutyl ketone. In general, ethers, esters, and selected ketones have a threshold concentration for the extraction of phosphoric acid and are therefore not suitable for the extraction of phosphate salts from low concentration sources.

[0056] In the phosphate industry, tributyl phosphate is commonly used for the purification of phosphoric acid by liquid-liquid extraction. Tributyl phosphate is not flammable, has low toxicity, and has a very low solubility in water, about 0.4 grams per liter at room temperature. In addition, the solubility increases with increasing temperature. Tributyl phosphate also has a fairly constant distribution coefficient, i.e. the ability to extract phosphoric acid at low phosphoric acid concentrations. Due to the relatively high density of tributyl phosphate, it is usually mixed with a diluent, such as aliphatic kerosene, in order to improve the physical separation of the immiscible phases.

[0057] Tributyl phosphate enables the preferential extraction of phosphoric acid over dissolved salts (such as chlorides or sulphates) and dissolved acids (such as hydrochloric acid or sulphuric acid). The presence of dissolved salts or acids enhances the extraction of phosphoric acid by a salting out mechanism, which can enable almost complete extraction of phosphoric acid.

[0058] However, tributyl phosphate preferentially extracts nitric acid over phosphoric acid, which makes it impossible to selectively extract phosphoric acid from nitric acid. In general, solvents with low solubility in water preferentially extract nitric acid over phosphoric acid. Solvents with increased selectivity for phosphoric acid, such as pentanol, have a high solubility in water and still co-extract large amounts of nitric acid. The main advantage of using tributyl phosphate as a solvent for the extraction of phosphoric acid is that the low solubility in water enables operation without the need to distil the solvent from the water stream, which is expensive and complex.

[0059] As mentioned above, most suitable solvents have a relatively high solubility in water, for example most alcohols. However, alcohols with a relatively long carbon chain also have a relatively low solubility in water. Therefore, an alternative solvent for the extraction of phosphoric acid is a long carbon chain alcohol, for example heptanol, which has a solubility in water of less than 2% and preferably less than 1%. If the solubility is lower than such a level, the amount of solvent following the water stream will become rather low, which can be handled by less expensive and complex equipment.

[0060] The extraction of phosphoric acid with a water-immiscible solvent has previously been used for the production of ammonium phosphate.

[0061] International patent application WO 2008 / 115121 discloses a process and apparatus for phosphorus recovery. Phosphorus ions are extracted from a solution by adsorbing them in a scavenger and releasing them into an eluate during regeneration of the scavenger. The regeneration is performed by means of ammonia. Upon introduction of an excess of ammonia, the phosphate anions precipitate as triammonium phosphate. The ammonia that remains in the solution after the triammonium phosphate has precipitated is used again for the regeneration of the scavenger. Unfortunately, triammonium phosphate is not stable at ambient temperature and atmospheric pressure, which leads to the decomposition of the crystals and the release of ammonia in the process, which needs to be further processed into a stable form of ammonium phosphate. Triammonium phosphate is not suitable for direct use in agriculture.

[0062] GB 636,035 discloses an improvement in the process for the production of diammonium phosphate. Crystals of monoammonium phosphate are introduced into a solution of diammonium phosphate in a reactor and anhydrous ammonia is fed into the reactor. Diammonium phosphate crystals are collected at the bottom of the chamber.

[0063] International patent application WO 2010 / 138045 assigned to the present assignee describes a process comprising adding ammonia to a phosphorus-loaded water-immiscible liquid phase in order to precipitate ammonium phosphate. The precipitated ammonium phosphate is washed with a saturated aqueous ammonium phosphate solution and the washed crystals are dried. The residual scavenger washed out of the crystals is separated by phase separation of the scavenger and the saturated aqueous ammonium phosphate solution and the separated residual scavenger is reused for further adsorption of phosphorus to be reused for further extraction. The scavenger-depleted wash liquid is reused for further washing of the crystals. The drawback is the need for three phase separations.

[0064] US patent 9738522 B2 assigned to the present assignee discloses a process for producing ammonium phosphate.

[0065] It is a further object of the present technology to, instead, achieve the production of solid potassium phosphate by using liquid-liquid extraction in a robust way that enables efficient recycling of process liquids.

[0066] The present invention is therefore based on a process of stripping a phosphoric acid loaded solvent with a potassium phosphate solution. Since stripping with monopotassium phosphate solution is based on extraction equilibrium, complete stripping of phosphoric acid from the solvent can require more than one contact stage. However, as will be shown in the later description, stripping with monopotassium phosphate solution is very efficient.

[0067] Generally, since monopotassium phosphate has a higher density, it was found that stripping with monopotassium phosphate is significantly superior to conventional stripping with water in terms of separation time and completeness of separation.

[0068] One advantage of using monopotassium phosphate for stripping is that, in most embodiments, the stripping involves only two phases, an organic phase and an aqueous phase, without any precipitate being formed. This enables operation with conventional liquid-liquid extraction devices such as pulse columns, mixer-settlers, or any other liquid-liquid extraction device such as stirred columns, non-stirred columns, in-line mixers, centrifugal contactors, etc.

[0069] Based on the surprising results presented above, an advantageous apparatus for producing pure potassium phosphate can be outlined. The apparatus comprises an extraction section, a stripping section, and a final treatment apparatus. The different parts are closely related in terms of the composition of the liquids used. The extraction section performs a liquid-liquid extraction of phosphate between a feed liquid comprising phosphoric acid and a solvent. The solubility of the solvent in water is preferably less than 2%. This enables the production of a potassium phosphate precipitate with a satisfactory level of attached solvent. The stripping section performs a liquid-liquid extraction of phosphate between a phosphate-loaded solvent and a stripping solution. The phosphate-poor solvent is connected back to the extraction section for further extraction of phosphate. The stripping solution is an aqueous potassium phosphate solution. By these apparatuses, a phosphate-loaded stripping solution is provided in a way that is suitable for industrial production. The loaded stripping solution is then treated in different ways in order to obtain a distinct MKP and / or a distinct DKP. Such a final treatment apparatus uses a source of potassium base, such as KOH. An addition apparatus is connected to the source of potassium base and adds potassium base to at least a part of the stream of stripping solution that exits the stripping section. The chemical reaction that takes place when potassium base is added generates heat, and therefore, a cooling apparatus is provided for removing this heat. The apparatus also comprises a precipitate remover that separates crystals of MKP or DKP from the loaded stripping solution. Finally, a recirculation system is connected between the precipitate remover and the stripping section for enabling the use of the stripping solution from the precipitate remover again as the stripping solution input to the stripping section.

[0070] The above disclosed method here achieves the production of clean and distinct mono- or di-potassium phosphate in one and the same plant without the need for concentrating phosphoric acid by water evaporation. This is achieved by a combination of liquid-liquid extraction and chemical precipitation, which, due to the special properties of the stripping solution and / or the solvent, avoids emulsification and the formation of a crude product due to non-settled precipitate. In a preferred embodiment, in addition, operational problems such as the formation of precipitate during the stripping solution, the production of potassium phosphate precipitate with an unsatisfactory level of attached solvent are avoided.

[0071] A process for producing water-soluble ammonium phosphate is disclosed in US patent 9,738,522 B2. As discussed before, for certain crops, potassium phosphate is a valuable specialty fertilizer. In addition, there are other advantages to producing potassium phosphate instead of ammonium phosphate. Ammonia is a toxic gas, so handling ammonia requires special care and it can be difficult to obtain a permit for handling ammonia at certain sites. The handling of potassium base is much easier since no gas phase is involved. At several phosphoric acid plants, ammonia is not handled on site and it is easier to produce potassium phosphate instead of ammonium phosphate in terms of permits, logistics, storage, and transportation.

[0072] Patent US 9,738,522 B2 discloses a process for producing ammonium phosphate by combining liquid-liquid extraction and chemical precipitation of ammonium phosphate. In the disclosed process, solvent loaded with phosphoric acid is back-extracted with an ammonium phosphate back-extraction solution to obtain an ammonium phosphate solution containing back-extracted phosphoric acid. However, when the same process was tested to recover potassium phosphate, emulsification occurred in many cases, rendering the process inoperable. After extensive investigation, the cause of the problem was revealed.

[0073] Tributyl phosphate solvent can extract both fluorosilicic acid and phosphoric acid. When solvent loaded with both fluorosilicic acid and phosphoric acid is back-extracted with a potassium phosphate solution, a precipitate of potassium fluorosilicate forms, causing the crude product / emulsification. This occurs because the solubility of potassium fluorosilicate is very low. In contrast, ammonium fluorosilicate has a high solubility in water, and therefore, a similar problem does not occur in the process of US patent 9,738,522 B2.

[0074] From these findings, it was concluded that the basic process described above can indeed be operated on feed solutions that do not contain fluorosilicates.

[0075] However, for feed solutions that contain fluorosilicates, the solution to the technical problem is to reduce the content of fluorosilicates in the solvent at least prior to back-extraction with potassium phosphate. This can be done by reducing the fluorosilicate content of the aqueous original feed to the extraction step, or by reducing the fluorosilicate content of the solvent itself from the load, as described in the further examples of the specification. After reducing the fluorosilicate content, or if the feed solution is essentially free of fluorosilicates to begin with, the fluorosilicate content in the loaded solvent is preferably lower than 0.3 M, and most preferably lower than 0.001 M.

[0076] Figure 1 A flowchart showing the steps of a method for producing pure potassium phosphate is shown. The procedure starts at step 200. In step 210, phosphate salts are extracted from a feed solution containing phosphoric acid into a solvent. The extraction is preferably carried out by liquid-liquid extraction. The solvent has a solubility in water of preferably less than 2%. In step 212, at least a portion of the solvent with phosphate salt content is back-extracted into a back-extraction solution by liquid-liquid extraction. The back-extraction solution contains potassium phosphate. In step 214, the back-extraction solution loaded with back-extracted phosphate salts is separated from the at least partially phosphate salt-depleted solvent. In step 216, the at least partially phosphate salt-depleted solvent is recycled for further extraction of phosphate salts in the extraction step 210.

[0077] In step 220, a potassium base (such as KOH) is added in at least a portion of the stream of stripping solution. In step 222, the heat generated in the addition of the potassium base in at least a portion of the stream of stripping solution is cooled, in whole or in part. In step 224, crystals are removed from the loaded stripping solution. In this figure, steps 220-224 are shown as subsequent steps. However, since they describe different cyclic processes, their relative dependency on time and process liquids can vary from one embodiment to another. However, all of steps 220-224 are performed in one way or another in all embodiments. In step 226, after the step 224 of removing crystals, the stripping solution is recycled for use as stripping solution input in the stripping step 212. The procedure ends in step 299.

[0078] In applications where the feed liquid comprises fluosilicates, a process for removing fluosilicates is preferably included. In one embodiment, a step 208A of removing fluosilicates from the feed liquid is performed prior to the step 210 of extracting phosphates. This step removes or at least significantly reduces the content of fluosilicates in the feed liquid. Different embodiments of this process will be further discussed below in connection with examples. In another embodiment, a step 208B of removing fluosilicates from the stream of water-immiscible solvent loaded with phosphoric acid is performed. This step thus occurs after the step 210 where phosphates are extracted together with some fluosilicates, but before the step 212 of solvent stripping. This step can preferably be combined with, for example, any washing procedure. Step 208B removes or at least significantly reduces the content of fluosilicates in the water-immiscible solvent loaded with phosphoric acid. Different embodiments of this process will be further discussed below in connection with examples.

[0079] Some embodiments of the present concept will be presented hereinafter in order to illustrate its advantages and various possibilities.

[0080] Figure 2 An embodiment of an apparatus 100 for producing pure potassium phosphate, in this embodiment monopotassium phosphate, is shown. A phosphoric acid containing feed liquid 1 is fed to an extraction section 10 configured to perform a liquid-liquid extraction of phosphates between the feed liquid 1 and a solvent 5. The solvent 5 is water-immiscible and has a solubility in water preferably less than 2%. In this particular embodiment, the water-immiscible solvent 5 is tributyl phosphate in aliphatic kerosene.

[0081] The phosphoric acid feed liquid 1 is typically obtained by digesting a phosphorus-containing material with a mineral acid. The phosphorus-containing material can be phosphate rock or other phosphorus-containing material such as ash from incinerated sewage sludge, ash from incinerated slaughterhouse waste, ash from manure, etc. The mineral acid used for the digestion should preferably be sulfuric acid, hydrochloric acid, nitric acid or phosphoric acid in order to obtain selective extraction of phosphoric acid.

[0082] The concentration of phosphoric acid in the leach solution can be very low, such as less than 7% P2O5or even less than 4% P2O5. For a sludge ash leach solution, a diluted phosphoric acid solution is typical. The advantage of treating a diluted leach solution is that the solution also contains dissolved salts or acids which do not get extracted in preference to the phosphoric acid, but which have a salting-out effect, which enables almost complete extraction of the phosphoric acid at low concentrations. Of course, leach solutions with a higher concentration of phosphoric acid can be treated according to the present disclosure. There is no limitation on the maximum concentration of phosphoric acid.

[0083] In alternative embodiments, the phosphoric acid feed solution 1 can be provided by other means. The specific manner in which the phosphoric acid feed solution 1 is provided does not substantially affect the main concept of the present disclosure and is therefore not discussed further. The only detail which is of interest is the fluosilicate content in the feed solution, which is discussed further below.

[0084] Since the main objective of the concept presented in the present disclosure is to enable the production of potassium phosphate without the need to concentrate the phosphoric acid by evaporation of water, it is obvious and most advantageous that the concentration of the phosphoric acid is the maximum concentration which is actually possible by digestion with acid. Digestion of phosphate rock with sulfuric acid according to the dihydrate process typically results in a concentration of the phosphoric acid of about 28% P2O5. Optionally, the aqueous leach solution is pre-treated to remove ionic compounds, such as iron, fluorine, etc.

[0085] The extraction section 10 has a first extraction inlet 11 for providing the feed solution 1 and a second extraction inlet 13 for providing the solvent 5. The extraction section 10 also has a first extraction outlet 12 for transporting the raffinate or at least partially depleted phosphate feed solution 2 and a second extraction outlet 14 for transporting the phosphate loaded solvent 3.

[0086] As discussed previously, any organic solvent which is capable of removing phosphorus from an aqueous solution can be used. The mechanism of extraction of phosphorus can be solvation of the phosphoric acid, or both ion association and solvation. The composition of the organic solvent should be selected in dependence of the concentration of the phosphoric acid feed, the presence of additional acids or salts, etc. in order to obtain a high loading capacity and an efficient operating extraction process. In the present embodiment, phosphoric acid tributyl ester in aliphatic kerosene is used as the preferred solvent.

[0087] Since lower temperatures are generally advantageous for the extraction of phosphoric acid, the temperature of the water immiscible solvent 5 is preferably below 60°C.

[0088] The liquid-liquid extraction process in the extraction section 10 is preferably a continuous liquid-liquid extraction process, preferably using a liquid-liquid extraction device such as a pulse column. However, any other liquid-liquid extraction device can be used, such as a stirred column, a non-stirred column, a mixer-settler, an inline mixer, a centrifugal contactor, etc.

[0089] The raffinate 2, i.e. the feed solution at least partially depleted of phosphate, is directed to further processing, such as re-use for dissolution, etc.

[0090] Optionally, the water-immiscible solvent 3 loaded with phosphoric acid is washed with an aqueous solution to remove co-extracted impurities.

[0091] The loading of the solvent 3 depends on the concentration of phosphoric acid in the feed solution 1, the concentration of dissolved salts and acids, the ratio of the input water-immiscible solvent 5 compared to the feed solution 1, and the number of contacting stages during extraction in the extraction section 10. A particular advantage of this embodiment is that potassium phosphate can be produced in high yield even if the phosphoric acid loading in the solvent 3 is very low. The phosphoric acid loading in the solvent 3 can be below 2% P2O5, and still a high yield of solid potassium phosphate can be obtained by increasing the concentration of phosphoric acid during the stripping process. Of course, the loading of the solvent 3 can be higher. A typical feed solution obtained from the digestion of phosphate rock with sulfuric acid according to the dihydrate process typically results in a phosphoric acid loading of about 6% P2O5 in the water-immiscible solvent 3.

[0092] Returning to Figure 2 Thereafter, the phosphorus-loaded water-immiscible solvent 3 is provided to a stripping section 20. The stripping section 20 is configured for liquid-liquid extraction of phosphate between the phosphate-loaded solvent 3 and a stripping solution 4. The stripping solution 4 is an aqueous potassium phosphate solution. In this embodiment, essentially all of the potassium phosphate in the input stripping solution 4 is monopotassium phosphate.

[0093] The stripping section 20 has a first stripping inlet 21 connected to the second extraction outlet 14 for providing the phosphate-loaded solvent 3. The stripping section 20 also has a second stripping inlet 24 for providing the input stripping solution 4. The stripping section 20 also has a first stripping outlet 22 for delivering the at least partially phosphate-depleted solvent 5, and a second stripping outlet 23 for delivering the output stripping solution 6. The first stripping outlet 22 is connected to the second extraction inlet 13 to recycle the at least partially phosphate-depleted solvent 5 for further extraction of phosphate in the extraction section 10.

[0094] The stripping section 20 is preferably a continuous liquid-liquid extraction process, preferably using a liquid-liquid extraction device such as a pulse column or a mixer-settler. However, any other liquid-liquid extraction device can be used, such as a stirred column, a non-stirred column, an inline mixer, a centrifugal contactor, etc.

[0095] When producing monopotassium phosphate according to this embodiment, the potassium phosphate solution 4 is preferably a recycled monopotassium phosphate solution, e.g. with a concentration slightly below the solubility limit at the operating temperature.

[0096] With low phosphoric acid loading in the solvent, the phosphoric acid concentration is increased during the stripping process by using a ratio of organic phase to water phase greater than 1. This requires several contact stages to obtain complete stripping of the phosphoric acid. During the stripping process, the phosphoric acid concentration can typically be increased to between 2 times and more than 5 times the original concentration. In the present embodiment, where stripping is performed using a KH2PO4solution, the output stripped solution 6 is typically composed of a mixture of KH2PO4and H3PO4. The output stripped solution 6 is provided to the end treatment device 90.

[0097] The end treatment device 90 of the present embodiment comprises a source of potassium base 60. Preferably, it can be a source of KOH. Furthermore, an addition device 70 is connected to the source of potassium base 60. The addition device 70 is configured to add potassium base from the source of potassium base 60 to at least a part stream 71 of the stripped solution. In this embodiment, a part of the stripped solution 9 leaving the precipitate remover 40 (which will be further explained below) is diverted, and potassium base from the source of potassium base 60 is added to this part stream 71. In the present embodiment, the part stream 71 comprises a monopotassium phosphate solution 72, and by adding potassium base, this solution is converted into a dipotassium phosphate solution 73.

[0098] The addition of potassium base to the part stream 71 of the aqueous potassium phosphate solution generates heat. Therefore, this chemical process is an exothermic process, and in the present embodiment for the production of monopotassium phosphate, the solution 73 resulting after the addition of potassium base comprises a mixture of KH2PO4and K2HPO4, which is further heated. Therefore, a cooling device 50 is provided, which is configured to cool the heat generated when potassium base from the source of potassium base 60 is added to the part stream 71 of the stripped solution. This neutralization of heat is preferably removed by a heat exchanger 52. The cooled solution 51 of the mixture of monopotassium phosphate and dipotassium phosphate leaves the cooling device 50.

[0099] As will be further discussed below, the solution 73 of the mixture of monopotassium phosphate and dipotassium phosphate is preferably cooled more than the cooling given by the addition of potassium base. The addition of potassium base to a saturated KH2PO4solution 72 leads to a significant increase in the solubility limit of the ionic species in the solution as soon as the amount of potassium added reaches a certain level. This enables the cooling of the potassium phosphate solution 51 to a temperature lower than the temperature of the stripped solution 6 without any precipitate being formed. Therefore, heat exchange can be performed on a heat exchanger without fouling.

[0100] Preferably, the amount of potassium base added should be proportional to the phosphate load in the water-immiscible solvent 3. As the concentration of phosphoric acid in the solvent decreases, the conductivity decreases and the pH level increases. Therefore, the addition of potassium base can be controlled by monitoring the pH and / or conductivity of the solvent 3 with suitable sensors. An alternative approach is to monitor the pH and / or conductivity of the stripping solution 6 or in the recycling loop with suitable sensors before or after the addition of potassium base.

[0101] In an alternative embodiment, the potassium base from the potassium base source can be added directly to the stripping solution 6 in order to form monopotassium phosphate from the remaining phosphoric acid. However, the neutralization of phosphoric acid with potassium base is highly exothermic and leads to a large amount of heat generation which has to be removed in the process afterwards. The solubility of potassium phosphate is very temperature dependent. For example, the solubility of monopotassium phosphate is about 23 g per 100 ml water at 20 °C and rises to about 84 g per 100 ml water at 90 °C. This means that if the potassium base is added directly to the stripping solution 6 containing phosphoric acid, no precipitation of potassium phosphate occurs due to the higher solubility at higher temperatures. Cooling of the stripping solution 6 by heat exchange after the addition of potassium base is indeed possible. However, since the solubility of potassium phosphate is temperature dependent, this procedure can lead to crystallization of potassium phosphate on the heat exchanger which reduces the heat exchange efficiency and requires frequent descaling which is undesirable and makes it difficult to operate continuously.

[0102] Therefore, preferably, according to the embodiment of Figure 2 mixing the stripping solution 6 with the recycled and cooled dipotassium phosphate solution 51 in the mixing reactor 30, the monopotassium phosphate precipitates spontaneously instead. The precipitation kinetics can be enhanced in the presence of seed crystals. It was found that if the stripping solution 6 and the recycled dipotassium phosphate solution 51 have the same temperature, the temperature increase during the precipitation of monopotassium phosphate per mole of phosphate precipitated is only a few °C. This temperature increase is easily manageable compared to the significant temperature increase of directly neutralizing the stripping solution with potassium base. As mentioned above, the temperature of the recycled and cooled dipotassium phosphate solution 51 can be controlled to be lower than the temperature of the stripping solution 6 which can compensate for the temperature increase during the precipitation of monopotassium phosphate obtained by combining the stripping solution 6 with the recycled dipotassium phosphate solution 51.

[0103] In other words, the cooling device is configured to keep the temperature of the loaded stripping solution below the saturation temperature of monopotassium phosphate, so that monopotassium phosphate crystals precipitate. The cooling is preferably performed in the stream of potassium phosphate that does not precipitate. In a preferred embodiment, the addition device 70 comprises a cooling device 50. The actual cooling can also be performed in different ways. The cooling device 50 can for example operate on the liquid stream to which the potassium base is to be added, i.e. before the mixing. Alternatively or in combination, the cooling device 50 can also operate on the liquid stream to which the potassium base has been added. Finally, the potassium base to be added can also be cooled. The cooling device 50 is typically a heat exchanger 52. However, any other alternative cooling device 50 can also be used, such as a Peltier element, cooling by heat conduction, etc.

[0104] Thus, according to the present embodiment, the precipitation of monopotassium phosphate occurs in the mixing reactor 30, without the need to cool the mixing reactor 30 itself. In this way, the crystallization of potassium phosphate on the heat exchanger is avoided, which improves the heat exchange efficiency of the process and makes the continuous operation easy and robust, without the need to frequently remove scale from the heat exchanger.

[0105] The slurry 31 leaving the mixing reactor 30 is composed of KH2PO4 crystals in a saturated KH2PO4 solution. The slurry 31 leaving the mixing reactor 30 is fed to the previously mentioned precipitate remover 40. The precipitate remover 40 is configured to separate the crystals 41 from the loaded stripping solution 31 leaving the mixing reactor 30. The separation can be performed by any solid-liquid separation technique, such as filtration, sedimentation, centrifugation, etc.

[0106] In other words, the end treatment device 90 comprises a dipotassium phosphate supply device 35 and a mixing reactor 30. The mixing reactor 30 is connected to the second stripping outlet 23 of the stripping section 20 and to the dipotassium phosphate supply device 35. The mixing reactor 30 is configured to mix a dipotassium phosphate solution in the stripping solution 6 from the output of the stripping section 20. In this embodiment, the cooling device 50 is configured to maintain the temperature of the solution 31 leaving the mixing reactor 30 below the saturation temperature of monopotassium phosphate. This consequently results in the precipitation of monopotassium phosphate crystals from the saturated monopotassium phosphate solution. In this embodiment, the precipitate remover 40 is thus configured to separate the monopotassium phosphate crystals. In this embodiment, the dipotassium phosphate supply device 35 comprises an addition device 70. In this embodiment, the addition device 70 comprises an inlet connected to an outlet from said precipitate remover 40 for supplying a partial flow 71 of the stripping solution 9 leaving the precipitate remover 40. The addition device is configured to add a potassium base from the potassium base source 60 to the partial flow 71 of the stripping solution leaving the precipitate remover 40, thereby forming a solution comprising dipotassium phosphate. The addition device is further configured to return the solution comprising dipotassium phosphate to the dipotassium phosphate mixing reactor 30. The cooling device is configured to maintain the temperature of the loaded stripping solution below the saturation temperature of monopotassium phosphate, thereby precipitating monopotassium phosphate crystals.

[0107] The separated crystals can be dried in a dryer 42 and / or granulated according to known processes, forming a final monopotassium phosphate product. Since the adhering solution is a saturated monopotassium phosphate solution, the water content of the separated crystals is low. The separated crystals can also be mixed with other ingredients, such as nitrogen, forming different fertilizer products.

[0108] It was found that according to the present embodiment, the residual solvent in the precipitated crystals is very low, as it only corresponds to the dissolved solvent in the solution that adheres to the separated crystals. The water solubility of water-immiscible solvents, such as tributyl phosphate, is low, which results in a solvent level in the precipitated crystals after separation of less than 20 ppm.

[0109] Furthermore, according to a further embodiment, trace amounts of dissolved solvent can be removed from the aqueous solution prior to the potassium phosphate precipitation. This can be done by adding an oxidizing agent, such as hydrogen peroxide or the like, in order to oxidize the trace amounts of solvent to carbon dioxide and phosphoric acid. In this way, even residual solvent present in the product can be cleaned up. Furthermore, if other contaminants, such as fluorine or the like, are also co-extracted with the phosphoric acid, they can be removed from the stripping solution prior to the potassium phosphate precipitation by precipitation, extraction or the like.

[0110] Returning to Figure 2In an embodiment of the application, the aqueous potassium phosphate solution 9 leaving the precipitate remover 40 is divided into two parts. One part of the aqueous monopotassium phosphate solution stream 4 is recycled back to the second stripping inlet 24 of the stripping section 20. Hence, the apparatus comprises a recycling system 80 connected between the outlet from the precipitate remover 40 and the second stripping inlet 24 of the stripping section 20, wherein the recycling system 80 is configured to recycle the stripping solution 4 from the precipitate remover 40 again for use as input stripping solution. The second part of the aqueous potassium phosphate solution stream 71 is treated as mentioned above for addition of potassium base.

[0111] The amount of potassium base introduced into the system is used entirely for the generation of potassium phosphate crystals. In order to achieve a balance in the process, the addition apparatus 70 is preferably configured to add an amount of potassium base to the addition apparatus 70 which depends on the amount of stripped phosphate salt in the stripping solution 6 leaving the second stripping outlet 23 of the stripping section 20. The amount of potassium base added also depends on the relative amounts of monopotassium phosphate and / or dipotassium phosphate produced, see further described embodiments below.

[0112] There are many objectives of the present technology. One objective is to provide pure potassium phosphate with very low contamination of adhering water immiscible solvent. Another objective of the present invention is to provide high yield of potassium phosphate by spontaneous precipitation or seed induced precipitation without the need for cooling during precipitation. Yet another objective of the present invention is to achieve improved phase separation during stripping of phosphoric acid from the solvent. Another objective of the present invention is to achieve solvent recycling without the need for further treatment by liquid-liquid extraction or distillation. By Figure 2 All these objectives are achieved by

[0113] Another objective is to avoid operational problems such as emulsification and crude product formation. Specifically, the aim is to form a regenerated solvent that does not precipitate upon contact with the potassium phosphate back-extraction solution. In applications where the feed liquid contains fluorosilicates, a device for removing fluorosilicates is preferably included. In one embodiment, a fluorosilicate remover 8A is provided in the feed liquid stream 1 before it enters the extraction section 10. In other words, the outlet of the fluorosilicate remover 8A is connected to the first extraction inlet 11 of the extraction section 10. The fluorosilicate remover 8A is configured to remove or at least significantly reduce the fluorosilicate content in the feed liquid 1. Different embodiments of this device will be further discussed below with reference to examples. In another embodiment, a fluorosilicate remover 8B is provided in the water-immiscible solvent 3 containing phosphoric acid before it enters the back-extraction section 20. These can preferably be combined, for example, with any of the washing devices mentioned earlier. The fluorosilicate remover 8B is configured to remove or at least significantly reduce the fluorosilicate content in the water-immiscible solvent 3 containing phosphoric acid. Different embodiments of this device will be further discussed below with reference to examples.

[0114] If dipotassium phosphate is the desired final product, the precipitated monopotassium phosphate can be converted into dipotassium phosphate by feeding precipitated monopotassium phosphate and potassium alkali into a reactor containing a dipotassium phosphate solution.

[0115] also, Figure 3 An embodiment of an apparatus for the direct production of dipotassium phosphate is shown. Extraction section 10 and back-extraction section 20 are connected to... Figure 2The same is presented in the middle and the fluosilicate removers 8A and 8B are also the same if needed, however, the end treatment equipment 90 has certain variations. As before, the dipotassium phosphate supply equipment 35 provides to the mixing reactor 30 a solution comprising dipotassium phosphate to precipitate monopotassium phosphate. However, at this time the details of the dipotassium phosphate supply equipment 35 are somewhat different, which will be further discussed below. The precipitated monopotassium phosphate 41, possibly together with some remaining monopotassium phosphate solution, is provided as a slurry to the DKP conversion reactor 32. Potash from the potash source 60 is fed to the DKP conversion reactor 32. In other words, the dipotassium phosphate supply equipment 35 is configured to add potash to a stream derived from the loaded stripping solution, for example, as in this embodiment, via the precipitate remover 40, or as discussed in the embodiments below, directly from the stripping section 20. The addition of potash causes the precipitation of dipotassium phosphate and provides a slurry 33 of dipotassium phosphate crystals in a dipotassium phosphate solution. The slurry 33 is cooled by heat exchange by the cooling equipment 50 to a temperature preferably below 90°C. In the DKP precipitate remover 43, the dipotassium phosphate crystals 44 are separated from the mother liquor 45. A first portion 46 of the separated dipotassium phosphate solution is recycled to the DKP conversion reactor 32. A second portion 51 of the separated dipotassium phosphate solution 45 is recycled to the mixing reactor 30 to form the precipitated intermediate monopotassium phosphate. In this way, the addition of potash and the heat exchange during the conversion of monopotassium phosphate to dipotassium phosphate are utilized to produce the intermediate monopotassium phosphate.

[0116] Figure 2 and Figure 3 Embodiments of the above can also be combined in such a way that only a portion of the MKP crystals separated in the precipitate remover 40 is taken to the DKP conversion reactor 32. The remaining portion is still MKP. In this way, both a defined monopotassium phosphate and a defined dipotassium phosphate can be produced in the same plant.

[0117] Figure 4A further embodiment is shown which produces both defined monopotassium phosphate and defined dipotassium phosphate. Here, a part stream 81 of the strip solution 6, which carries monopotassium phosphate and phosphoric acid, enters the DKP conversion reactor 32. A solution or slurry 83 of potassium phosphate trihydrate is added. At a suitable mixing ratio and at a suitable temperature, a dipotassium phosphate slurry 33 will be formed. The cooling device is preferably configured to keep the temperature of the loaded strip solution below the saturation temperature of dipotassium phosphate, so that dipotassium phosphate crystals precipitate. The slurry 33 contains precipitated dipotassium phosphate crystals as well as a saturated dipotassium phosphate solution. The dipotassium phosphate crystals 44 are removed in the DKP precipitation remover 43, leaving a saturated dipotassium phosphate solution 45. The saturated dipotassium phosphate solution 45 is directed to the second addition device 70, where potassium from a second potassium base source 60 (or from the potassium source used for the production of monopotassium phosphate) is added. A slurry of potassium phosphate trihydrate 82 is formed, which is cooled in the heat exchanger 52 to provide the slurry of potassium phosphate trihydrate 83 to be used in the DKP conversion reactor 32.

[0118] The disadvantage of this embodiment is that the part producing dipotassium phosphate will continuously collect a large amount of dipotassium phosphate solution, which is not recycled back to the stripping section. At the same time, the part producing monopotassium phosphate will lose a corresponding amount, which has to be replaced. Figure 5 A further embodiment is shown which at least partially solves this problem. Here, the potassium phosphate trihydrate supply device connected to the stream of monopotassium phosphate is omitted. Instead, the dipotassium phosphate supply device 35 producing dipotassium phosphate is used to supply dipotassium phosphate 51 to the mixing reactor 30. To this end, a part stream 84 of the potassium phosphate solution 45, in this embodiment a solution of dipotassium phosphate, leaves the precipitation remover, in this embodiment the DKP precipitation remover 43, and is recirculated to be used in the production of monopotassium phosphate.

[0119] In other words, the part producing dipotassium phosphate is configured to add potassium base to the stream containing monopotassium phosphate. This causes dipotassium phosphate crystals to precipitate from the saturated dipotassium phosphate solution. The dipotassium phosphate supply device comprises a reconnection of dipotassium phosphate, which connects a part stream of the saturated dipotassium phosphate solution from the part producing dipotassium phosphate to the production of monopotassium phosphate as supplied dipotassium phosphate.

[0120] In a further embodiment, Figure 4 and Figure 5 Embodiments of

[0121] Additional embodiments of the device for producing dipotassium phosphate according to the invention will be described hereinafter with reference to Figure 6 Most of the parts are identical to those described in connection withFigure 2 The process is similar to the one in Figure 1, and the difference mainly lies in the type of solution used in the different parts of the apparatus. The feed solution 1 containing phosphoric acid is fed to the liquid-liquid extraction section 10. The temperature of the water-immiscible solvent 5A is preferably below 60 °C, as lower temperatures favor the extraction of phosphoric acid.

[0122] The liquid-liquid extraction section 10 is preferably configured to perform a continuous liquid-liquid extraction process, preferably using a liquid-liquid extraction device such as a pulse column. However, any other liquid-liquid extraction device can be used, such as a stirred column, a non-stirred column, a mixer-settler, an inline mixer, a centrifugal contactor, etc. The raffinate 2, which is depleted in phosphate, is directed to further processing, such as being used again for dissolution, etc. Optionally, the water-immiscible solvent 3A, which is loaded with phosphoric acid, is washed with an aqueous solution to remove co-extracted impurities.

[0123] Thereafter, in the stripping section 20, the water-immiscible liquid phase 3A, which is loaded with phosphorus, is mixed with the recycled potassium phosphate dibasic solution 4A as stripping solution. As above, the stripping section 20 is preferably configured for a continuous liquid-liquid extraction process, preferably using a liquid-liquid extraction device such as a mixer-settler. However, any other liquid-liquid extraction device can be used, such as a pulse column, a stirred column, a non-stirred column, an inline mixer, a centrifugal contactor, etc.

[0124] Stripping with a potassium phosphate dibasic solution is based on the conversion of phosphoric acid to dissolved salt, as opposed to stripping with a potassium phosphate monobasic solution, which is found to be based on the extraction equilibrium.

[0125] It is believed that the reaction is as follows:

[0126] KH2PO4 + H3PO4 = 2 KH2PO4

[0127] As stripping with a potassium phosphate dibasic solution is not based on the extraction equilibrium, complete stripping can be obtained in a single contact stage. Therefore, when stripping with a potassium phosphate dibasic solution, the preferred liquid-liquid extraction device is a single mixer-settler unit.

[0128] According to the present embodiment, the stripping of phosphoric acid from the loaded water-immiscible solvent is preferably performed with the recycled potassium phosphate dibasic solution 4A in such a way that the solubility of potassium phosphate monobasic or potassium phosphate dibasic is not exceeded. This enables the operation with only two phases during the stripping in the mixer-settler.

[0129] According to the present embodiment, the phosphoric acid is stripped from the loaded water-immiscible solvent 3A with a solution of dipotassium phosphate 4A, whereby no precipitation of monopotassium phosphate or dipotassium phosphate occurs. It was found that by controlling the phase ratio of solvent 3A to dipotassium phosphate solution 4A during the stripping, the precipitation of monopotassium phosphate can be completely avoided. The phase ratio of solvent 3A to dipotassium phosphate solution 4A at which no precipitation of monopotassium phosphate occurs depends on the loading of phosphoric acid in the water-immiscible solvent 3A and the concentration of the dipotassium phosphate solution 4A. This means that the phase ratio has to be adapted to each operating condition, preferably.

[0130] In the test setup of this embodiment, the potassium phosphate solution, i.e. the incoming stripping solution 4A, is composed of a recycled dipotassium phosphate solution.

[0131] Similar to the stripping with a solution of monopotassium phosphate, the stripping with a solution of dipotassium phosphate can be used to increase the concentration of phosphate in the stripping solution 6 compared to the original concentration in the water-immiscible solvent 3A. This is achieved by having a ratio of organic phase to aqueous phase greater than 1 during the stripping process. Generally, a higher concentration of dipotassium phosphate 4A will allow a higher ratio of organic phase to aqueous phase during the stripping. Higher concentrations of dipotassium phosphate can be obtained by operating at higher temperatures. Preferably, the ratio of organic phase to aqueous phase is not increased to a level at which a precipitate of monopotassium phosphate occurs during the stripping process.

[0132] For a specific operating condition and temperature, the phase ratio is calculated from the solubility of the ion species involved.

[0133] Returning to Figure 6 , the stripping solution 6 is composed of a mixture of KH2PO4 and K2HPO4. Therefore, the ratio of K2HPO4 to KH2PO4 in the obtained stripping solution 6 is important, as a sufficient amount of K2HPO4 is preferably required so that no precipitate of monopotassium phosphate is formed during the stripping process in the stripping section 20. The stripping solution 6 is mixed with a potassium phosphate solution or a slurry 51A comprising K3PO4. The slurry 31A leaving the mixing reactor 30 is composed of K2HPO4 crystals in a saturated K2HPO4 solution. The slurry 31A leaving the mixing reactor 30 is fed to the precipitate remover 40. In this embodiment, the precipitate remover 40 is configured to separate the precipitated dipotassium phosphate crystals 44 from a dipotassium phosphate solution 9A. The separation can be performed by any solid-liquid separation technique, such as filtration, sedimentation, centrifugation, etc.

[0134] The separated crystals can be dried and / or granulated according to known processes, forming a final dipotassium phosphate product. The separated crystals can also be mixed with other ingredients, such as nitrogen, forming different fertilizer products.

[0135] The aqueous dipotassium phosphate solution 9A leaving the precipitate remover 40 is divided into two parts 4A and 72A. One part 4A of the aqueous dipotassium phosphate solution is recycled back to the stripping section 20. The second part 72A of the aqueous dipotassium phosphate solution is treated for addition of potassium base. The addition of potassium base to the aqueous dipotassium phosphate solution 72A generates heat and can form a slurry 73A. The neutralization heat is removed by heat exchanger 52. The solution / slurry 73A consists of a mixture of K2HPO4 and K3PO4. The addition of potassium base should be proportional to the phosphate load in the water-immiscible solvent 3A. The cooled slurry 51A is provided to the mixing reactor 30 as further described above.

[0136] In another embodiment, the input stripping solution is a mixture of KH2PO4 and K2HPO4. The input stripping solution has a K / P ratio and a concentration of phosphate ions that prevent monopotassium phosphate or dipotassium phosphate crystals from precipitating when in contact with the phosphate-loaded solvent in the stripping section.

[0137] If the input stripping solution has a K / P ratio and a concentration of phosphate ions that cause monopotassium phosphate crystals to precipitate when in contact with the phosphate-loaded solvent in the stripping section, the stripping section must be designed as a three-phase stripping section. The precipitate remover is then configured to separate the monopotassium phosphate crystals from the loaded stripping solution leaving the stripping section.

[0138] When the input stripping solution is a mixture of KH2PO4 and K2HPO4, the apparatus preferably comprises a mixture control unit configured to control the composition of the input stripping solution 4.

[0139] Figure 7Another embodiment of the production of monopotassium phosphate is shown. In this embodiment the stripping solution 4A consists of K2HPO4. The loaded water immiscible solvent 3A is stripped with the stripping solution 4A and results in three phases; stripped solvent 5A, loaded stripping solution and formation of precipitated KH2PO4 crystals. The stripped solvent 5A is separated and used again for the extraction of phosphoric acid. The aqueous phase leaving the stripping section, the stripping solution 6B, consists now of KH2PO4 crystals in a saturated KH2PO4 solution. A small problem of this embodiment is that the KH2PO4 crystals thus formed contain a too high level of residual solvent. To remove this residual solvent the stripping solution 6B is mixed with the recycled monopotassium phosphate solution 26 in a washing volume 25. This washing enables the residual solvent to be separated at a higher aqueous to organic phase ratio and the residual solvent 15 can be recycled to the extraction section 10. After separation of the residual solvent 15 the slurry 27 of KH2PO4 crystals in a saturated KH2PO4 solution is fed to a precipitate remover 40. In the precipitate remover 40 the monopotassium phosphate crystals 41 are separated from the mother liquor.

[0140] A first part 28 of the separated monopotassium phosphate solution 9 is provided to an addition device 70 to be subjected to treatment for the addition of potash from a potash source 60. A solution 4A consisting of K2HPO4 is formed. This neutralization reaction also generates heat which is cooled by means of a heat exchanger 52 of the cooling device 50. After cooling in the heat exchanger 52 the dipotassium phosphate solution 4A is used again for stripping.

[0141] A second part 26 of the separated monopotassium phosphate solution 9 is used again for the separation of residual solvent as described above.

[0142] The three phase stripping section is also applicable to a system in which a mixture of KH2PO4 and K2HPO4 is used as input stripping solution and the K / P ratio and the phosphate ion concentration result in the precipitation of monopotassium phosphate crystals when contacted in the stripping section with the phosphate loaded solvent.

[0143] Then small modifications can be made to the equipment. These modifications are indicated by dashed lines in Figure 7 To this end a third part 29 of the separated monopotassium phosphate solution 9 is allowed to bypass the addition device, which means that after cooling in the heat exchanger 52 the dipotassium phosphate solution is mixed with the third part 29 of the separated monopotassium phosphate solution to form a solution 4A consisting of a mixture of KH2PO4 and K2HPO4. Preferably the addition device comprises an addition control device configured to add a stream of stripping solution directly from the outlet of the precipitate remover 40 together with a stream of solution containing dipotassium phosphate from the heat exchanger 52 to obtain the required K / P ratio in the stripping section 20.

[0144] The use of a three-phase stripping section 20 can also be combined with the production of either or both MKP and DKP. Figure 8 An embodiment is shown. The slurry 85 of monopotassium phosphate crystals in a saturated monopotassium solution is withdrawn from the precipitate remover 40 and enters the DKP mixing reactor 32. Potash is added and the neutralization heat is cooled, which is similar to Figure 3 An embodiment similar to that of FIG. 1 is shown. However, in this case, the DKP precipitate remover 43 is connected to the stripping section 20. Figure 8 In the embodiment of FIG. 2, the dipotassium phosphate solution 45 from the DKP precipitate remover 43 is instead returned to be used as the stripping solution. If the amount of dipotassium phosphate crystals is too low to maintain the incoming stripping solution, a stream 86 of monopotassium phosphate can be withdrawn from the precipitate remover 40 and potash can be added to convert this solution to a dipotassium phosphate solution. Optionally, a cooling device can also be connected to this stream.

[0145] If the stripping solution is composed of a mixture of KH2PO4and K2HPO4, a portion 87 of the monopotassium phosphate solution from the precipitate remover 40 can be mixed into the incoming stripping solution without passing through the addition device 70. Here again, an addition control device can be provided and configured to add a stream of stripping solution directly from the outlet of the precipitate remover 40 together with a stream of solution containing dipotassium phosphate to obtain the required K / P ratio in the stripping section 20.

[0146] The applicant has also found that by adding potash and mixing it with the loaded solvent, monopotassium phosphate can be formed directly in the loaded solvent, as presented in further examples herein.

[0147] However, in this case, a relatively large amount of solvent is still attached to the separated potassium phosphate crystals. These amounts are typically large enough that the loss of the expensive solvent mixture is generally economically unacceptable. Therefore, it is also preferred to recover these amounts of solvent.

[0148] Figure 9In the embodiment, the solvent 3 carrying the phosphate is provided to a first inlet 122 of a mixing reactor 30. Potassium hydroxide 118 from a potassium hydroxide source 60 is provided to a second inlet 124 of the mixing reactor 30. Monopotassium phosphate is formed in the mixing reactor 30. An addition control 128 senses the composition in the mixing reactor 30, for example by means of a pH meter or other sensor 130, and controls the amount of potassium entering through the second inlet 124 above the solubility level of monopotassium phosphate. Heat is generated during the mixing, which is absorbed by the heat exchanger 50. A temperature sensor 132 senses the mixing reactor temperature, and a temperature control 134 controls the operation of the heat exchanger 50 to maintain a reasonable temperature. The mixed solution containing the precipitated monopotassium phosphate crystals is provided to the precipitate remover 40, which separates the potassium phosphate crystals from the solution 5 and returns it to the extraction section 10 for further extraction.

[0149] In the embodiment, the solvent 3 carrying the phosphate is provided to a first inlet 122 of a mixing reactor 30. Potassium hydroxide 118 from a potassium hydroxide source 60 is provided to a second inlet 124 of the mixing reactor 30. Monopotassium phosphate is formed in the mixing reactor 30. An addition control 128 senses the composition in the mixing reactor 30, for example by means of a pH meter or other sensor 130, and controls the amount of potassium entering through the second inlet 124 above the solubility level of monopotassium phosphate. Heat is generated during the mixing, which is absorbed by the heat exchanger 50. A temperature sensor 132 senses the mixing reactor temperature, and a temperature control 134 controls the operation of the heat exchanger 50 to maintain a reasonable temperature. The mixed solution containing the precipitated monopotassium phosphate crystals is provided to the precipitate remover 40, which separates the potassium phosphate crystals from the solution 5 and returns it to the extraction section 10 for further extraction. Figure 9 In the embodiment, the solvent 3 carrying the phosphate is provided to a first inlet 122 of a mixing reactor 30. Potassium hydroxide 118 from a potassium hydroxide source 60 is provided to a second inlet 124 of the mixing reactor 30. Monopotassium phosphate is formed in the mixing reactor 30. An addition control 128 senses the composition in the mixing reactor 30, for example by means of a pH meter or other sensor 130, and controls the amount of potassium entering through the second inlet 124 above the solubility level of monopotassium phosphate. Heat is generated during the mixing, which is absorbed by the heat exchanger 50. A temperature sensor 132 senses the mixing reactor temperature, and a temperature control 134 controls the operation of the heat exchanger 50 to maintain a reasonable temperature. The mixed solution containing the precipitated monopotassium phosphate crystals is provided to the precipitate remover 40, which separates the potassium phosphate crystals from the solution 5 and returns it to the extraction section 10 for further extraction.

[0150] According to one embodiment of the present invention, the solvent adhering to the separated potassium phosphate crystals is removed by washing the potassium phosphate crystals with a saturated aqueous potassium phosphate solution. The solvent initially adhering to the crystals forms a separate phase, which is typically lighter than the dense aqueous phase, and as further described above, is immiscible with water. Thus, the two phases spontaneously separate from each other. The separator 160 of the present embodiment is therefore a phase separator 158, which is arranged to separate the solvent from the saturated aqueous solution of potassium phosphate. Surprisingly, the above washing procedure was found to be very effective. It is believed that the washing with the saturated potassium phosphate solution is a dynamic process, in which the potassium phosphate crystals are constantly dissolving and recrystallizing, enabling efficient removal of the adhering solvent. The washing procedure is simple to operate and does not consume energy. The saturated potassium phosphate solution separated from the crystals is continuously recirculated for further washing. The composition of the saturated potassium phosphate solution is made by dissolving the produced potassium phosphate salt in an aqueous solution, such as water, phosphoric acid or other acid / salt solution. As mentioned above, the separated water immiscible solvent is continuously recirculated in order to extract the phosphate salt from the feed solution.

[0151] The washed potassium phosphate crystals are subsequently dried in a dryer 42. Preferably, the drying can be at least partly performed by means of the heat obtained in the heat exchange process of cooling the mixed potash and acid. For this purpose, the dryer 42 is connected to the heat exchanger 50 as indicated by the dashed arrow 137. Thus, the dryer 42 is arranged to dry the washed crystals 41 using at least part of the heat extracted in the heat exchanger 50.

[0152] The produced potassium phosphate is fully water soluble, metal depleted, and can be used for agricultural purposes, such as fertilization or fertigation.

[0153] Another important advantage of the washing process according to the preferred embodiment of the present application is that it enables the production of potassium phosphate to produce MKP or DKP, independent of the initial composition of the precipitated crystals. If MKP is the desired end product, the washing solution used is preferably composed of a saturated aqueous solution of monopotassium phosphate. By adding, for example, phosphoric acid or potash, the pH of the slurry is controlled and adjusted to the optimal value. Thus, a pH control device 156 can be provided to control the pH in the washing volume device 25. This procedure enables the production of MKP independent of the initial composition of the precipitated crystals. In a similar manner, if DKP is the desired end product, the washing solution used is composed of a saturated aqueous solution of dipotassium phosphate. By adding, for example, potash, the pH of the slurry is controlled and adjusted to the optimal value for DKP. This procedure enables the production of DKP independent of the initial composition of the precipitated crystals. In this way, both MKP and DKP can be produced according to the present application. For this purpose, the washer 25 is thus arranged to control the pH of the saturated aqueous solution of monopotassium phosphate and / or dipotassium phosphate.

[0154] Figure 10A flow chart showing the steps of a method according to an embodiment of the application is shown. The method for producing potassium phosphate starts at step 200. Step 210 is performed as described earlier. If necessary, steps 208A and 208B are performed as described earlier. Step 210 provides a phosphorus loaded water immiscible liquid phase. In step 230, a potassium base is added to the water immiscible liquid phase. In one specific embodiment, the adding step 230 comprises monitoring the conductivity of the water immiscible liquid phase, and controlling the amount of potassium base added in dependence of the monitored conductivity. In another specific embodiment, the adding step 230 comprises monitoring the pH of the water immiscible liquid phase, and controlling the amount of potassium base added in dependence of the monitored pH. In step 232, monopotassium phosphate and / or diphosphates are precipitated from the water immiscible liquid phase. In step 234, the temperature of the water immiscible liquid phase is controlled within a predetermined temperature interval during the adding and precipitating steps. As will be discussed in further detail below, the actual control step can be performed before, during and / or after the adding and precipitating steps. The important feature is to ensure that the temperature during the adding and precipitating is kept within a predetermined range. It is less important when the actual heat removal instant occurs. Thus, in time, step 234 can be located before, simultaneously and / or after steps 230 and 232. The temperature control typically comprises extracting heat from the water immiscible liquid phase. In step 236, the precipitated monopotassium phosphate and / or diphosphates are extracted from the water immiscible liquid phase. The water immiscible liquid phase is again used for further extraction in step 238.

[0155] In Figure 10 In the shown embodiment, the method further comprises step 240, in which the extracted precipitated monopotassium phosphate and / or diphosphate crystals are washed. In step 242, the remaining water immiscible liquid phase (i.e. typically the solvent) washed out from the crystals is separated. As indicated by the dashed arrow 244, the separated remaining wash-out agent is preferably used again for further adsorption of phosphorus to obtain a phosphorus loaded water immiscible liquid phase. Similarly, as indicated by the dashed arrow 246, the washed out depleted solvent is used again for further washing of the crystals. In this specific embodiment, the washing is performed with a saturated aqueous solution of potassium phosphate, and the depleted solvent is separated by phase separation of the solvent from the saturated aqueous solution of potassium phosphate. In step 248, the washed crystals are dried. Preferably, the drying utilizes at least a part of the heat extracted from the temperature controlled step.

[0156] There are alternative ways of washing the crystals. The adhering solvent can be removed from the crystals by washing with an organic solvent having a boiling point significantly lower than the boiling point of the extraction solvent and in which the potassium phosphate crystals cannot dissolve. The organic solvent used for washing can be completely miscible with water. Examples of solvents that can be used for washing the potassium phosphate crystals include acetone (boiling point 56.5 °C), methanol (boiling point 64.7 °C), etc. The potassium phosphate crystals that can be recovered by distillation treatment to remove the residual washing solvent. The washing solution obtained can be collected and the organic solvent used for washing can be separated from the solvent by distillation.

[0157] Figure 11 This embodiment is illustrated. Here, the washing volume device 25' is arranged to wash the crystals with an organic washing solvent in which the potassium phosphate cannot dissolve. The separator 160 comprises a distiller 164 that separates the washing solvent 161 in gaseous phase from the residual extraction solvent 17 that is still in liquid phase. Preferably, the heat extracted from the mixing reactor 30 can also be used as at least part of the heat source required for the distillation operation, as indicated by the arrow 138. The gaseous washing solvent 161 is condensed in a condenser 162. Gaseous washing solvent is also produced in the dryer 42, which is preferably also connected back to the condenser 162 for further reuse. However, as a minor drawback, it is found that washing the crystals with an organic solvent requires careful consideration of the flow and amount of washing solvent. The amount of washing solvent required for washing can be relatively large. Furthermore, the process is somewhat more complex than the previously proposed processes, requiring more energy to separate the organic washing solvent from the organic extraction solvent by distillation. At present, the embodiments using a saturated aqueous solution of monopotassium phosphate and / or diphosphate are considered to be preferred.

[0158] As briefly mentioned above, the heat can be actually extracted from the said phosphorus-laden water-immiscible liquid phase in different ways. In Figure 9 and Figure 11 In the embodiment of the heat exchanger 50 integrated in the mixing volume device 30. At present it is believed that this is the preferred method because it gives a well-controlled temperature.

[0159] However, alternatives are also possible. Figure 12An embodiment is shown, wherein the heat exchanger 50 is arranged in contact with the water-immiscible liquid phase 116 leaving the precipitate remover 40. The controller 134 can still be controlled based on the temperature in the mixing reactor 30 as measured by the thermometer 132. Alternatively, or in addition, the controller 134' can operate based on the temperature of the solvent entering the extraction section 10 as obtained by the thermometer 132'. In this way, the temperature of the solvent entering the extraction section 10 is primarily controlled, which in turn keeps the temperature of the phosphorus-laden water-immiscible liquid phase within the mixing reactor 30 within the required temperature interval, especially if there is information about the phosphorus content of the phosphorus-laden water-immiscible liquid phase envisaged to leave the extraction section 10. In other words, by controlling the temperature of the solvent entering the extraction section 10, an indirect control of the temperature in the mixing reactor 30 will also be achieved. This can be a good alternative from a plant point of view, in case the initial phosphorus content is relatively stable or at least predictable. The solvent 5 entering the extraction section 10 can then be optimized in terms of its affinity for phosphorus in terms of temperature.

[0160] Figure 13 Yet another embodiment is shown, wherein the heat exchanger 50 is arranged in contact with the water-immiscible liquid phase 3 leaving the extraction section 10 before entering the mixing reactor 30. Here, the control can be based on either or both of the temperature in the mixing reactor 30 or the temperature of the loaded solvent before entering the mixing reactor 30. The temperature in the solvent before entering the mixing capacity device is then measured by the thermometer 132" and using the controller 134". In this way, the temperature of the loaded solvent is reduced, and the exothermic reaction expected in the mixing reactor 30 will bring the water-immiscible liquid phase to the predetermined temperature interval.

[0161] Different systems have been tested to show and verify the advantages obtained with the method and the plant according to the above-described principles. Some examples are presented in the following.

[0162] Example 1

[0163] This example demonstrates the extraction of H3PO4, H2SO4, and (SiF6) 2- A filter grade phosphoric acid having approximately 28% P205(approximately 40% H3PO4or 4.85 M H3PO4), 7% H2SO4(2.8 M), and 2% F (as H2SiF6) was used as the aqueous phase. An organic phase containing 80% vol. TBP and 20% vol. kerosene was contacted with the above aqueous phase at organic phase: aqueous phase (O:A) ratios ranging from 1 : 10 to 10: 1. Figure 14 Distribution data for H3PO4(the concentration ratio of H3PO4in the aqueous phase to the organic phase for the tested O:A ratios) are shown, and Figure 15Distribution data for H2SO4 is shown. The trend indicates that as the organic phase is loaded with more H3PO4, H2SO4 is excluded. This can be exploited to minimize the co-extraction of H2SO4 by TBP, for example, by performing the extraction of H3PO4 close to the H3PO4 loading capacity of the organic solvent. Between 5% and 15% of the total F in the aqueous phase is extracted by the organic solvent at the different O:A ratios tested.

[0164] Example 2

[0165] This example demonstrates the extraction of H3PO4 at pilot scale (counter-current mixer-settler, 5 / L stage). The extraction of H3PO4 from a filtered grade of H3PO4 was performed using an organic solvent containing 80% vol. TBP and 20% vol. kerosene. Two systems were tested: a five-stage counter-current setup and an eight-stage counter-current setup, using an O:A ratio of 3:1. Figure 16 and 17 The extraction of H3PO4 in both systems is shown. In the five-stage setup, 0.92 M H3PO4 was loaded in the organic solvent, leaving about 2 M H3PO4 in the aqueous phase Figure 16 ). With the increase in the number of stages to eight, the efficiency improved: 1.4 M H3PO4 was extracted in the 8-stage setup, leaving about 1.3 M H3PO4 in the aqueous phase Figure 17 .

[0166] Example 3

[0167] This example highlights the importance of the treatment step (to prevent K2SiF6 precipitation during stripping with a KH2PO4 solution) and shows the stripping isotherms of H3PO4. Stripping of a solvent (80% vol. TBP and 20% vol. kerosene) loaded with 1.25 M H3PO4 and 0.04 M F in the form of (SiF6)2"was performed with a 1.58 M KH2PO4 solution at four different O:A ratios: 1:1, 2:1, 5:1 and 10:1. 2- The importance of the treatment step (to prevent K2SiF6 precipitation during stripping with a KH2PO4 solution) and shows the stripping isotherms of H3PO4. Stripping of a solvent (80% vol. TBP and 20% vol. kerosene) loaded with 1.25 M H3PO4 and 0.04 M F in the form of (SiF6)2"was performed with a 1.58 M KH2PO4 solution at four different O:A ratios: 1:1, 2:1, 5:1 and 10:1. 2- The importance of the treatment step (to prevent K2SiF6 precipitation during stripping with a KH2PO4 solution) and shows the stripping isotherms of H3PO4. Stripping of a solvent (80% vol. TBP and 20% vol. kerosene) loaded with 1.25 M H3PO4 and 0.04 M F in the form of (SiF6)2"was performed with a 1.58 M KH2PO4 solution at four different O:A ratios: 1:1, 2:1, 5:1 and 10:1. 2- The importance of the treatment step (to prevent K2SiF6 precipitation during stripping with a KH2PO4 solution) and shows the stripping isotherms of H3PO4. Stripping of a solvent (80% vol. TBP and 20% vol. kerosene) loaded with 1.25 M H3PO4 and 0.04 M F in the form of (SiF6)2"was performed with a 1.58 M KH2PO4 solution at four different O:A ratios: 1:1, 2:1, 5:1 and 10:1. Figure 18 The stripping yields at each O:A ratio tested are given in Table 1.

[0168] Strip O:A ratio Strip yield of H3PO4 (%) 1∶1 94.4 2∶1 82.9 5∶1 54.3 10∶1 34.4

[0169] Table 1. H3PO4 stripping yields with a 1.58 M KH2PO4 solution at different O:A ratios. Solvent: 80% vol. TBP and 20% vol. kerosene loaded with 1.25 M H3PO4.

[0170] As shown in the McCabe Thiele diagram in Figure 18 Figure 3, at 3: 1 O:A, all of the H3PO4 in the organic phase can be effectively recovered in the KH2PO4 solution in four stages. Of course, other O:A ratios and different number of stages can be used.

[0171] Example 4

[0172] This example presents two options for removing (SiF6) 2- ions from the wet phosphoric acid solution by precipitation with Na2CO3 or K2CO3. Filter grade phosphoric acid with approximately 28% P2O5 (40% H3PO4 or 4.85 M H3PO4), 7% H2SO4 (2.8 M), and 2% F (as H2SiF6) was used. Different stoichiometric amounts of Na2CO3 or K2CO3 were added to 100 mL of the acid. This was done for Na / K to (SiF6) 2- at 1 : 1 Na:(SiF6) 2- or 2: 1 K:(SiF6) 2- . The resulting precipitates were filtered, dried, and weighed. The results are presented in Table 2. By adding Na2CO3 or K2CO3 to the acid and filtering the precipitated Na2SiF6 / K2SiF6, the fluosilicate ions can be efficiently removed.

[0173]

[0174] Table 2. Removal of (SiF6) 2- from wet H3PO4 by precipitation with Na2CO3 or K2CO3.

[0175] Example 5

[0176] This example shows that if either of the two methods described in Example 4 is used to remove (SiF6) 2- during stripping, the K2SiF6 precipitation during stripping can be avoided, and the stripping process is not hindered by this treatment. Using both methods in Example 4, (SiF6) 2- was removed from the wet acid stream before the H3PO4 was extracted with TBP. Na2CO3 was dosed at 3 times the stoichiometric amount Na:(SiF6) 2- . K2CO3 was dosed at 3 times the stoichiometric amount K:(SiF6) 2-twofold addition. The precipitated Na2SiF6 / K2SiF6was filtered and the H3PO4in the resulting solution was extracted with 80% vol. TBP and 20% vol. kerosene. The organic phase was back extracted with 1.58 M KH2PO4solution at different O:A ratios (1.05 M H3PO4loaded for the aqueous phase treated with K2CO3and 1.27 M for the aqueous phase treated with Na2CO3)(Table 3). No precipitation occurred at any O:A ratio. Figure 19 The back extraction isotherms for both systems are shown. As seen from the values in Table 3, the back extraction yield is not affected by the (SiF6) 2- treatment step.

[0177]

[0178] Table 3. H3PO4back extraction yield with 1.58 M KH2PO4solution for the system where (SiF6) 2- was removed by precipitation with Na2CO3or K2CO3prior to the solvent extraction step. Solvent: 80% vol. TBP and 20% vol. kerosene loaded with 1.27 M H3PO4(Na2CO3treated) or 1.05 M H3PO4(K2CO3treated).

[0179] Example 6

[0180] The removal of (SiF6) 2- from 80% vol. TBP and 20% vol. kerosene containing 1.15 M H3PO4and 0.04 M F as (SiF6) 2- was tested using other processes as follows.

[0181] i) Solid K2CO3was mixed with the organic solvent (20 g / L). This caused (SiF6) 2- to precipitate as K2SiF6which was removed. Subsequently the resulting organic phase was effectively back extracted with 1.58 M KH2PO4solution at 3:1 O:A without any precipitation occurring.

[0182] ii) Similar to i above except that solid KH2PO4(30 g / L) was used instead. This also caused K2SiF6to precipitate and, after removal of the precipitate, allowed back extraction with 1.58 M KH2PO4solution at 3:1 O:A without any precipitation occurring.

[0183] iii) Washing the organic solvent with water, for example, contacting the solvent with water at high O:A ratios of 10:1 and 20:1. After removing the wash water, the resulting organic solvent is stripped with 1.58 M KH2PO4solution at 3:1 O:A. Precipitation of K2SiF6occurs. The water wash step removes some H3PO4from the organic phase according to Table 4.

[0184]

[0185] Table 4. Washing of 80% vol. TBP and 20% vol. kerosene loaded with 1.15 M H3PO4with water at 10:1 and 20:1 O:A ratios.

[0186] iv) Similar to iii above, except using a 1.58 M KH2PO4wash solution instead of water. Precipitation of K2SiF6occurs during the wash. The precipitate and wash solution are removed, and the resulting washed organic phase is then stripped with 1.58 M KH2PO4solution at 3:1 O:A. Precipitation of K2SiF6occurs, but is significantly reduced compared to the reference sample that was not treated with (SiF6) 2- The amount of precipitate is significantly reduced compared to the reference sample that was not treated with (SiF6)

[0187] v) Stripping of the loaded solvent with 1.58 M KH2PO4solution at 50°C and 3:1 O:A ratio. This significantly improves the phase separation time, but also notes the precipitation.

[0188] Example 7

[0189] Potassium phosphate precipitation was performed by adding solid KOH to the organic solvent loaded with 1.18 M H3PO4. Two systems were tested: one added the stoichiometric amount of solid KOH to form KH2PO4; the other used twice this amount. In both cases, potassium phosphate precipitated with high yield (Table 5).

[0190]

[0191] Table 5. Potassium phosphate precipitation by adding solid KOH to the organic solvent containing 1.18 M H3PO4.

[0192] Example 8

[0193] In this experiment, a nearly saturated monopotassium phosphate solution (KH2PO4, 1.58 M) was reacted with KOH to produce a dipotassium phosphate solution (K2HPO4, 1.58 M). The pH of the solution increased from 4.23 to 9.69. The reaction is exothermic; the temperature of the solution increased from 23°C to 50°C, and no precipitation occurred.

[0194] Example 9

[0195] Solid KH2PO4crystals were precipitated from the stripping product (KH2PO4containing 4 M H3PO4and 0.5 M H2SO4; initial pH 0). To the stripping product was added a 1.58 M K2HPO4solution (see Example 8) until the pH reached 4.65 (ratio of stripping solution : K2HPO4solution 1 : 3.5). The formation of KH2PO4crystals was slow, but the process could be accelerated using KH2PO4crystal seeds. The solution was filtered the next day. At this point the precipitation yield was 82% (356 g of crystals per L of stripping product, dry content 97%); it was noted that the precipitation continued slowly in the filtrate. 621.2 mg of the precipitate was dissolved in 10 mL of deionized water and the solution was analyzed using ICP-MS to confirm the composition and purity of the solid (Table 6). The crystals were composed of KH2PO4(K:P = 1.04, 99% by mass). The filtrate was sampled and analyzed at the time of filtration. Almost all of the sulfate ions remained in the filtrate (Table 7).

[0196]

[0197] Table 6. Analysis of the solution obtained by dissolving 621.2 mg of precipitated KH2PO4crystals in 10 mL of deionized water.

[0198]

[0199] Table 7. Analysis of K, P and S in the filtrate after precipitation of KH2PO4.

[0200] The above embodiments should be understood to be illustrative examples of the application. Those skilled in the art will understand that various modifications, combinations and changes can be made to the embodiments without departing from the scope of the present application. For example, the production of monopotassium phosphate and dipotassium phosphate in parallel according to the principles described above, the production of dipotassium phosphate by reacting a potassium phosphate trihydrate slurry with a stripping solution composed of monopotassium phosphate and phosphoric acid, and the combination of stripping with dipotassium phosphate solution to produce dipotassium phosphate with monopotassium phosphate as an intermediate. However, the scope of the present application is limited only by the claims set forth below.

Claims

1. A method for producing pure potassium phosphate, comprising the following steps: - Phosphate is extracted from a feed liquid containing phosphoric acid into a solvent by liquid-liquid extraction, wherein the extraction solvent is tributyl phosphate or a long-chain alcohol; - The solvent containing at least a portion of the phosphate is back-extracted into the back-extraction solution by liquid-liquid extraction; The back-extraction solution is an aqueous solution of potassium phosphate; - Remove fluorosilicates from the feed solution before the step of extracting phosphates from the feed solution, or remove fluorosilicates from the solvent before the step of back-extracting the solvent carrying phosphates; The steps for removing fluorosilicates include using Na / K:(SiF6) 2- At least one of Na2CO3 and K2CO3 is added in a ratio of at least 1:1; - Separate the back-extraction solution containing back-extracted phosphate from the solvent that is at least partially deficient in phosphate; - The solvent for at least a portion of the phosphate-poor salt is recycled for further extraction of phosphate in the extraction step; - Add potassium base to at least a portion of the back-extraction solution containing the back-extracted phosphate; - The heat generated when the potassium base is added to at least a portion of the back-extraction solution containing the back-extracted phosphate is cooled. -Removing crystals from the back-extraction solution containing the back-extraction phosphate; - After the crystal removal step, the back-extraction solution after crystal removal is recycled and used as the input back-extraction solution in the back-extraction step.

2. The method according to claim 1, characterized in that, The at least partial flow containing the potassium base is the at least partial flow following the crystal removal step.

3. The method according to claim 1 or 2, characterized in that, The solvent treated in the step of back-extracting the solvent has a fluorosilicate content of less than 0.3 M.

4. The method according to claim 3, characterized in that, The solvent treated in the step of back-extracting the solvent has a fluorosilicate content of less than 0.001 M.

5. The method according to claim 1 or 2, characterized in that, The potassium phosphate in the back-extraction solution is monopotassium phosphate, and is characterized by the following further steps: The temperature of the back-extraction solution containing the back-extraction phosphate is maintained below the saturation temperature of potassium monophosphate. A dipotassium phosphate solution is mixed into at least a portion of the back-extraction solution containing the back-extraction phosphate. To precipitate potassium monophosphate crystals from a saturated potassium monophosphate solution; The step of removing the crystals includes separating the potassium monophosphate crystals.

6. The method according to claim 5, characterized in that, The potassium base is added to at least a portion of the back-extraction solution containing the back-extraction phosphate to form a solution containing dipotassium phosphate. The solution containing dipotassium phosphate is returned to the back-extraction solution containing the back-extracted phosphate.

7. The method according to claim 1 or 2, characterized in that, The step of adding potassium alkali is performed by adding a certain amount of potassium alkali according to the amount of phosphate in the back-extraction solution containing the back-extracted phosphate after the back-extraction.

8. An apparatus (100) for producing pure potassium phosphate, comprising: - Extraction section (10) is configured to perform liquid-liquid extraction of phosphate between a feed liquid (1) containing phosphoric acid and a solvent (5), wherein the extraction solvent is tributyl phosphate or a long-chain alcohol. The extraction section (10) has a first extraction inlet (11) for providing the feed liquid (1), a second extraction inlet (13) for providing the solvent (5), a first extraction outlet (12) for conveying at least a portion of the phosphate-poor feed liquid (2), and a second extraction outlet (14) for conveying the phosphate-loaded solvent (3). - The back-extraction section (20) is configured to perform liquid-liquid extraction of phosphate between the phosphate-loaded solvent (3) and the back-extraction solution (4); The back-extraction section (20) has a first back-extraction inlet (21) connected to the second extraction outlet (14) for providing the phosphate-loaded solvent (3), a second back-extraction inlet (24) for providing the input back-extraction solution (4), a first back-extraction outlet (22) for conveying at least a portion of the phosphate-poor solvent, and a second back-extraction outlet (23) for conveying the back-extracted phosphate-loaded back-extraction solution (6). The first back-extraction outlet (22) is connected to the second extraction inlet (13) to recycle the solvent of at least a portion of the phosphate-poor phosphate for further extraction of phosphate; The back-extraction solution (4) is an aqueous solution of potassium phosphate; - Fluorosilicate remover (8A, 8B); The fluorosilicate remover is configured to use Na / K:(SiF6). 2- At least one of Na2CO3 and K2CO3 is added in a ratio of at least 1:1; The outlet of the fluorosilicate remover (8A) is connected to the first extraction inlet (11), or the fluorosilicate remover (8B) is connected between the extraction section (10) and the back-extraction section (20). - An end-of-line processing device (90) is connected to the second back-extraction outlet (23); The end-of-line treatment device (90) has a potassium alkali source (60); The end-of-line treatment device (90) has an addition device (70) connected to the potassium alkali source (60); The end-of-line processing device (90) has a cooling device (50) configured to cool the heat generated from the chemical reaction when potassium alkali from the potassium alkali source (60) is added to at least a portion of the back-extraction solution carrying the back-extraction phosphate. The end-of-line treatment device (90) has a precipitate remover (40) configured to separate crystals from the back-extraction solution containing the back-extracted phosphate. The addition device (70) is configured to add potassium base from the potassium base source (60) to at least a portion of the back-extraction solution containing back-extraction phosphate; as well as - A recirculation system (80) is connected between the outlet from the precipitate remover (40) and the second back-extraction inlet (24) of the back-extraction section (20), the recirculation system (80) being configured to reuse the back-extraction solution (9) after crystal removal from the precipitate remover (40) as the input back-extraction solution (4).

9. The device according to claim 8, characterized in that, The at least portion of the flow containing the potassium base is from the outlet of the precipitate remover (40).

Images