Method for purifying calcium-sulfate-containing material

Abstract

The invention relates to a method for purifying calcium sulfate, in particular for reducing radioactive components, phosphorus-containing or fluorine-containing impurities and heavy metals originating from what is known as phosphogypsum resulting from phosphoric acid production, based on recrystallization of the anhydrite form into the dihydrate form in a sulfuric acid solution and separation into a coarse fraction and a fine fraction, wherein the fine fraction contains the undesired impurities, in particular the radioactive component. The invention also relates to a system for carrying out such a method.

Description

[0001] Method for purifying calcium sulfate-containing material

[0002] The invention relates to a process for purifying calcium sulfate, in particular for reducing radioactive components, phosphorus or fluorine-containing impurities and heavy metals, starting from so-called phosphogypsum from phosphoric acid production.

[0003] In addition, the invention also relates to a plant for the purification of calcium sulfate.

[0004] Calcium sulfate is a waste product of the phosphoric acid-producing industry and is formed as a dihydrate and / or hemihydrate during the digestion of phosphate ores with sulfuric acid. The calcium sulfate formed during the production of phosphoric acid is also known as phosphogypsum.

[0005] Due to the phase-out of coal-fired power generation, large quantities of gypsum from the flue gas desulfurization (FGD) plants of German coal-fired power stations will no longer be available. The technical FGD gypsum produced in the lime washing process is chemically identical to natural gypsum and has the same structural properties. Like natural gypsum, FGD gypsum is used in the building materials industry for construction products. Phosphorus gypsum could therefore serve as a replacement for the future loss of FGD gypsum.

[0006] Besides impurities such as phosphorus, fluorine, and heavy metals, another reason why only about 15% of the annual global production of calcium sulfate from phosphoric acid production is currently used for limited applications (e.g., in road construction, as fertilizer, or cement additive) is the naturally occurring radioactivity of phosphogypsum. Due to its production process, phosphogypsum is classified as TENORM (Technologically Enhanced Natural Occurring Radioactive Material).

[0007] The radioactive components in phosphogypsum are crucial 226 Ra and its decay products. In magmatic phosphogypsums, these can also be found in increased amounts. 228 Ra and its decay products are present. A compilation of phosphogypsum from different countries and their radioactivity can be found, for example, in the "Safety Reports 230102P10WG".

[0008] Series No. 78, Radiation Protection and Management of NORM Residues in the Phosphate Industry” of the IAEA.

[0009] Following the predominantly used wet chemical manufacturing process with sulfuric acid, approximately 80 to 90% of the radium from the phosphate ore remains in the phosphogypsum due to the poorly soluble nature of radium sulfate, whereas approximately 80% of thorium and uranium are found in the phosphoric acid.

[0010] Therefore, in order to utilize phosphogypsum as a valuable product, the reduction of radioactive components, especially radium, is desirable.

[0011] In the building materials industry, for example, a so-called activity concentration index I is defined. According to the EU Directive Radiation Protection 112 (1999) "Radiological Protection Principles concerning the Natural Radioactivity of Building Materials", this index provides information based on the specific activities of 226 Ra, 232Th (or 228 Ra) and 40 K is an estimate of the effective annual dose to which a person is exposed by a material used as a building material. The following activity concentration index (I) is derived to determine whether a dose criterion is met: C K 3000 Bq kg- 1

[0012] Depending on the quantity, four limit values ​​are defined for the In of the material used in a building. For bulk quantities such as concrete structures, the limits are set at 0.5 and 1, which correspond to an effective annual dose of 0.3 and 1 mSv a, respectively. -1 For surface building materials, such as plasterboard, the corresponding index values ​​are 2 and 6, respectively, which should not be exceeded. Essentially, an index value of I = 1 defines the safe limit below which a trivial dose of 1 mSv a -1 is not exceeded.

[0013] In other areas, such as their use as fertilizers or soil improvers, the boundaries are not clearly defined. However, it is known that the use of phosphate-containing fertilizers leads to uranium contamination of agricultural soils. The uptake by plants is less problematic; rather, the continuous accumulation in groundwater and thus its ingestion through drinking water poses a potential future problem.

[0014] Due to its low solubility, radium sulfate is neither absorbed by plants nor is groundwater contamination expected. However, due to its low solubility and long half-life of 1600 years, 230102P10WG 226 Ra is expected to accumulate on fertilized fields, which could lead to problems for future generations.

[0015] According to the German Radiation Protection Ordinance, the following applies: 226Ra sets a limit for the activity concentration of 1000 Bq / kg. However, depending on the origin of the gypsum phosphate, specific activities of sometimes significantly >1000 Bq / kg are to be expected. Therefore, the use of gypsum phosphate as, for example, a soil improver or fertilizer can be problematic.

[0016] Various approaches to reducing the radioactive content in phosphogypsum are known from the state of the art.

[0017] The co-crystallization of radium through incorporation into foreign crystals has already been described (IAEA Technical Report Series 476; "The environmental behaviour of radium: Revised Edition" (2014); p. 17). Experimental studies have shown that radium can be incorporated into barite (BaSO4) through recrystallization. The tendency of radium to incorporate into a foreign crystal lattice is defined by its so-called partition coefficient. For barite, this value is approximately 1.2 to 1.8, and for gypsum in dihydrate form, it is only about 0.3. For gypsum in anhydrite form, estimates range up to 800. Based on this, one would expect that, at equilibrium, a significant enrichment of radium would occur in the anhydrite phase while the dihydrate phase is also present.

[0018] Patent GB 1394734 describes a process for reducing the radioactivity of untreated waste gypsum by classifying it into fine and coarse fractions using a hydrocyclone. The resulting fine fraction is only slightly enriched with radioactive components. However, this process presents two significant problems. Firstly, the radioactivity levels in the fine and coarse fractions already differ depending on the original phosphate ore source and the process conditions. Secondly, the enrichment effect in the fine fraction is very low and insufficient for further use.

[0019] US patent 4,146,568 discloses a process in which gypsum phosphate is suspended in dilute sulfuric acid with the addition of a barium sulfate solution and subsequently classified by wet sieving or hydrocyclone. However, the use of barium sulfate is undesirable because, depending on the application of the gypsum, there are limit values ​​for barium that would be exceeded by this process. 230102P10WG

[0020] US patent 4,388,292 describes a process in which gypsum phosphate is first converted to anhydrite at a sulfuric acid concentration of approximately 8 to 10 M, and then partially converted to the dihydrate at 0 to 42 °C and approximately 1 M sulfuric acid. The radium-enriched anhydrite fraction is then separated. As also mentioned in the patent, however, process control appears to be challenging, as there is a risk of metastable hemihydrate forming and causing problems for the equipment.

[0021] US patent 4,421,731 relates to a process in which gypsum phosphate is first calcined to the hemihydrate and then partially converted back to the dihydrate in an aqueous solution at 55 to 90°C. However, the process described therein is difficult to control, as it is essential to avoid excessively rapid and uncontrolled setting into a solid gypsum mass, while ensuring a controlled conversion to the dihydrate is challenging.

[0022] Above all, none of the aforementioned methods provides a comprehensive solution for phosphogypsum, which includes both the purification of the gypsum and, in some cases, the utilization of further impurities with regard to, for example, P2O5 and F content, heavy metals such as Cd, As, Pb and others, as well as integration into the overall complex of a phosphoric acid plant.

[0023] US 2003 / 163014 A1 describes a method for reducing the content of radioactive isotopes in a phosphogypsum material containing a radioactive impurity, the method comprising the steps of: mixing the phosphogypsum material with an acidic decontamination solution; separating the mixture into a liquid stream containing a large proportion of the radioactive contamination and a solid stream containing a residual component of the acidic decontamination solution; and removing the residual component of the acidic decontamination solution from the solid stream.

[0024] The object of the invention described below is to provide a process for purifying calcium sulfate, in particular for reducing radioactive components such as radium isotopes, so that the resulting, processed calcium sulfate can be used as a dihydrate (phosphogypsum) for downstream applications, such as gypsum in the gypsum industry, as a soil improver or fertilizer, as a sulfate carrier in the cement industry, as a feedstock in the paper industry, and others, in compliance with the respective prescribed regulations or guidelines for the specific 230102P10WG

[0025] activity can be used.

[0026] Another objective according to this invention described here is also to achieve additional fine purification with regard to unwanted impurities, such as P2O5 and F content, and heavy metals, such as Cd, As or Pb.

[0027] This problem is solved according to the invention by a method according to claim 1. In particular, the problem is solved by a method for reducing radioactive content from calcium sulfate, wherein the method comprises the following steps: a) providing a calcium sulfate comprising one or more of the components selected from dihydrate (CaSO₄-2H₂O), hemihydrate (CaSO₄²⁻FW), or anhydrite (CaSO₄), wherein the anhydrite is present in an amount of at least 50 wt%, based on the dry calcium sulfate, and has a particle size D v (50) of less than or equal to 25 pm, measured by laser diffraction, b) recrystallization of a suspension containing the calcium sulfate obtained in step a) and an aqueous sulfuric acid solution with a concentration of 2 to 18 wt. %, at a temperature of 0° to 35°C to the calcium sulfate dihydrate form, c) classification of the calcium sulfate from step b) and separation of the fine fraction from the coarse fraction, wherein the D v(50)-value of the fine fraction < 50 pm, and wherein the coarse fraction contains at least 70 wt% dihydrate, wherein the classification step c) is divided into two classification steps, wherein in the first classification step a coarse fraction and a partial fine fraction are produced and in the second step a partial coarse fraction and a fine fraction are produced, or wherein the classification step c) is divided into two classification steps, wherein in the first classification step a partial coarse fraction and a fine fraction are produced and in the second step a partial fine fraction and a coarse fraction are produced.

[0028] For the purposes of the present invention, with reference to step c), the coarse fraction is the fraction with the highest D v (50) value and the fine fraction is the fraction with the smallest D v(50)- value, in each case based on the total amount of calcium sulfate from step b). In addition to anhydrite and dihydrate, the individual fractions of the process according to the invention may also contain small amounts of, for example, quartz and / or hemihydrate. In a preferred embodiment, the coarse fraction contains dihydrate in an amount of 70 wt.% to 99 wt.% and anhydrite in an amount of 1 wt.% to 30 wt.%.

[0029] In a preferred embodiment of the process according to the invention, the production of the purified calcium sulfate, which is provided in step a), comprises the following steps:

[0030] I) Conversion of raw phosphate with concentrated sulfuric acid to calcium sulfate in the form of dihydrate (CaSO4-2H2O), hemihydrate (CaSO4-1 H2O) or a mixture thereof,

[0031] II) Separation of calcium sulfate as a solid,

[0032] III) Recrystallization of the dihydrate or hemihydrate to the anhydrite by preparing an aqueous suspension of the calcium sulfate separated in step II) and / or calcium sulfate in the form of phosphogypsum (CaSO4-2H2O) from the stockpile with sulfuric acid and

[0033] IV) Separation of the purified calcium sulfate after step III) in the form of the anhydrite (CaSO4) as a solid with a particle size D v (50) of less than or equal to 25 pm, measured by laser diffraction, of the liquid phase of the suspension obtained, wherein the anhydrite is present in an amount of at least 50 wt.%, based on the dry calcium sulfate.

[0034] In one embodiment, step IV) is followed by a washing step V).

[0035] The raw phosphate used in step I) is preferably apatite, Ca5(PO4)3X with X = F, OH or CI).

[0036] The separation in steps II) and IV) can be carried out, for example, by centrifugation, filtration, settling or evaporation, with mechanical separation being preferred, and separation by filtration being particularly preferred.

[0037] In one possible embodiment of the process, the calcium sulfate obtained from step III) can be separated from the aqueous phase in step IV) in such a way that residual amounts of sulfuric acid remain in the filter cake. This has the advantage that filter area in step IV) can be saved, since no further washing is necessary.

[0038] The particle size D v (50) of the calcium sulfate anhydrite in step a) is 1 to 25 pm, preferably 2 to 20 pm, particularly preferably 5 to 15 pm, wherein at a D v (50) 50 vol% of the particles have a smaller diameter than the specified value. The D v (50)- 230102P10WG

[0039] The value was determined using laser diffraction particle size analysis or laser granulometric measurement on a Malvern Panalytical GmbH Mastersizer 3000 with ethanol as the dispersing medium. The Fraunhofer model was applied as the scattering model.

[0040] The above-mentioned particle size of the calcium sulfate anhydrite enables the most effective possible separation of the coarse fraction (predominantly dihydrate phase) and fine fraction (predominantly anhydrite phase) in the subsequent classification step c).

[0041] For example, the particle size of the resulting anhydrite can be adjusted or set in steps I) to IV) by varying the reaction parameters such as temperature, sulfuric acid concentration, reaction time, and S / L ratio. It is also possible to adjust the particle size by adding additives such as bauxite, iron ore, or organic additives such as CPC and CTAB, or by adding seed crystals (anhydrite).

[0042] For the recrystallization to the dihydrate in step b), sulfuric acid concentrations in the range of 2 wt.% to 18 wt.%, preferably in the range of 3 wt.% to 18 wt.%, and particularly preferably from 6 wt.% to 18 wt.%, and especially from 8 wt.% to 16 wt.%, are used. In a particularly preferred embodiment, the sulfuric acid concentration is 10 wt.% to 12 wt.%.

[0043] The temperature in step b) is in the range between 0°C and 35°C, preferably 10°C to 30°C and particularly preferably in the range of 15°C to 25°C and especially at 15°C to 20°C.

[0044] In a particularly preferred embodiment, step b) is carried out at a temperature of 10°C to 30°C and a sulfuric acid concentration of 6 wt.% to 16 wt.%.

[0045] The reaction times in step b) are between 1h and 240h, preferably between 2h and 120h, particularly preferably between 3h and 24h.

[0046] In a preferred embodiment, the sulfuric acid is added in step b) in such an amount that the weight ratio of solid to liquid (S / L ratio in kg / kg) in the suspension is between 1 / 10 and 1 / 1, preferably 1 / 6 and particularly preferably 1 / 4 and 1 / 1.

[0047] In a particular embodiment of the invention, the sulfuric acid concentration and the S / L ratio can be adjusted via the amount of sulfuric acid remaining in the filter cake at step IV) of the process, such that the filtrate quantity from step IV) is monitored appropriately and thus conclusions can be drawn about the remaining sulfuric acid in the 230102P10WG.

[0048] The filter cake can be closed and the desired sulfuric acid concentration brought to the desired concentration by adding suitable amounts of water.

[0049] In the event that the production of the calcium sulfate for step a) is carried out by the process comprising steps I) to IV), it is advantageous to save the filter area, reduce the amount of wash water and reuse the sulfuric acid.

[0050] The particle size Dv(50), measured by laser diffraction, of the calcium sulfate produced in step b) in the dihydrate form after recrystallization is, for example, 25 pm to 150 pm, preferably 30 pm to 125 pm, particularly preferably 35 pm to 100 pm.

[0051] The mass fraction of anhydrite phase after step b) is less than 70 wt.%, preferably less than 50 wt.%, particularly preferably less than 40 wt.% and particularly less than 30 wt.%.

[0052] Since one mole of anhydrite is converted to one mole of dihydrate, two moles of water from the aqueous phase are required, which can lead to an increase in the acid concentration of the aqueous phase in step b).

[0053] To directly monitor the reaction progress, the water of crystallization content of the solid can be determined, from which the dihydrate and anhydrite content can be directly calculated.

[0054] It has been shown that, largely independent of the reaction time in step b), a small residual amount of anhydrite phase always remains, and therefore reaction control is not strictly necessary. With longer reaction times, the particle size (D) increases. v (50)-value) of the crystals of the dihydrate on average, however, the particle sizes of the anhydrite are not affected or only slightly affected by this.

[0055] In a preferred embodiment, the suspension remains at rest during the recrystallization in step b), preferably for 1 h to 240 h.

[0056] Alternatively, and also preferably, the solution can be kept in moderate motion (e.g., without significant shear forces) during the recrystallization in step b). This allows for a further improvement in the particle size of the dihydrate, favoring larger particle sizes. Suitable methods for keeping the suspension in moderate motion include, for example, blowing in air, stirring with a stirrer that minimizes damage or shear forces, or pumping.

[0057] In a particularly preferred embodiment of the process according to the invention, in step b) the calcium sulfate anhydrite is mixed with sulfuric acid of the concentration ranges mentioned above, and at the desired temperature, in a suitable reaction vessel or tank with moderate agitation for the duration required to achieve the desired degree of recrystallization. The agitation of the reaction mixture can be effected, for example, by blowing in air.

[0058] In a further preferred embodiment, one or more calcium sulfate dihydrate crystals are added as seed crystals in step b). It has been shown that the size and quantity of the seed crystals have a decisive influence on the particle size of the final product (dihydrate) and on the rate of recrystallization from the anhydrite to the dihydrate. The size and quantity of the seed crystals determine a (theoretical) total surface area available for the dissolved calcium sulfate to grow on. Thus, the surface area provided by the seed crystals can significantly optimize the recrystallization rate. The following basic rules apply:

[0059] The larger the total surface area provided by seed crystals, the faster the recrystallization rates are observed.

[0060] The larger the total surface area provided by seed crystals, the smaller the dihydrate crystals obtained in the final product.

[0061] Smaller seed crystals further accelerate the reaction rate.

[0062] The smaller the seed crystals, the smaller the crystals obtained in the final product.

[0063] Preferably, 0.1 wt.% to 70 wt.%, more preferably 2.5 wt.% to 40 wt.%, and particularly preferably 5 wt.% to 30 wt.% dihydrate crystals in solid, moist, or suspended form, such as filtrate and / or wash water with a suitable solids content of calcium sulfate dihydrate, are added to the suspension in step b). The mass fraction refers to the total mass of the purified calcium sulfate used in step a).

[0064] As previously described, the seed crystal size and quantity are responsible for optimizing the reaction rate and particle size of the final product. The dihydrate seed crystal size can be determined either by a suitable classification method, such as hydrocyclone, classifier, wet or dry sieving, or by targeted milling to achieve the desired crystal size for a corresponding quantity of the dihydrate.

[0065] In a particularly preferred embodiment, seed crystals are used which have a particle size D v (63) in the range of 15 to 100 pm, preferably 20 to 80 pm, particularly preferably 25 to 60 pm, wherein at a D v (63) 63 vol.-% of the particles have a diameter smaller than the specified value.

[0066] In a particularly preferred embodiment, 5 to 30 wt.% dihydrate crystals with a particle size D are added to the suspension in step b). v (63) used from 15 to 60 pm.

[0067] In a further embodiment, organic additives such as cetylpyridinium chloride (CPC) and / or cetyltrimethylammonium bromide (CTAB) can be used in step b). These have also been shown to have a positive effect on induction time and crystal growth. CTAB and / or CPC are used at a concentration of < 200 mg / kg, preferably < 100 mg / kg, and particularly preferably < 50 mg / kg, based on the suspension.

[0068] The classification of the calcium sulfate in step c) can be carried out, for example, by one or more of the following methods selected from i) one or more hydrocyclones connected in series, ii) one or more classifiers connected in series, iii) dry sieving, or iv) wet sieving. Method i) one or more hydrocyclones connected in series is preferred, and two hydrocyclones connected in series are particularly preferred. Classification methods i) and iv) are wet classifications, while classification methods ii) and iii) are dry classifications. Depending on the plant capacity, it may be technologically and economically advantageous not to perform each classification step in a single apparatus, but rather to divide the process stream to be classified among apparatuses connected in parallel. Following this principle, it is therefore conceivable to proceed according to the above.preferred method i) to sequentially connect several parallel hydrocyclones for the first classification step, followed by several parallel hydrocyclones for the second classification step.

[0069] In one embodiment of the present invention, a combination of the above-mentioned classification steps can be selected. For example, a coarse and fine fraction can first be generated via wet classification (e.g., using a hydrocyclone), and after the interposition of a separation and drying unit, the coarse and / or fine fraction can be separated again into a further coarse and fine fraction by means of dry classification (e.g., using an air classifier).

[0070] In one embodiment, the first step involves classification into coarse and fine fractions using a hydrocyclone. The fine fraction, now depleted of solids content, can then be separated into liquid and solid phases via a suitable separation process (e.g., filtration). After drying the solid phase using a suitable method, a dry classification can then be carried out, again producing coarse and fine fractions in this second step.

[0071] In a dry classification process that requires a dry material, a further process step according to step b) of the process according to the invention may be necessary. Preferably, the aqueous phase is first separated from the solid phase, optionally with washing of the gypsum. The gypsum is then dried. Suitable drying methods are those in which the gypsum cannot form agglomerates during the drying process, such as fluidized bed drying or spray drying. The dried gypsum can then be separated into fine and coarse fractions using the aforementioned (dry) classification methods.

[0072] For the purposes of the present invention, the following terms are used for the different particle fractions in step c) of the process according to the invention:

[0073] The largest faction is the faction with the highest D v (50) value.

[0074] The sub-group is the group with the second-highest D v (50) value

[0075] The sub-faction is the faction with the third highest Dv(50) value.

[0076] The fine fraction is the fraction with the smallest D v (50) value.

[0077] The highest D rating v (50) value, second highest D v (50) value, third highest D v (50) value and smallest D v (50)-value always refers to the sum of the amount of calcium sulfate obtained in step b) of the process according to the invention.

[0078] The process according to the invention is, in particular, an integrated process. An integrated process within the meaning of the present invention is a process in which all process parameters and process stages of the overall complex are coordinated to achieve a specific product quality starting from a variable reactant property, such as the reactant property of the apatite used in phosphoric acid production or the phosphogypsum from the stockpile. The separated liquid phases can be introduced into the reactor unit of the phosphoric acid production or into the reactor unit for the purification of the phosphogypsum as feedstock.

[0079] Due to radium's tendency to co-crystallize more readily in the anhydrite phase (fine fraction) and less so in the dihydrate phase, an enrichment of radium in the fine fraction occurs after the previously described recrystallization step b) to the dihydrate. 230102P10WG

[0080] In a preferred embodiment, the separation of the fine fraction from the coarse fraction takes place in a separate step d), wherein a partial or complete separation of the solid phases from both the fine and coarse fractions obtained after step c) is carried out. For example, during the wet classification in step c), after recrystallization in step b), at least a partial separation of the solid and liquid phases can first be performed. It can also be advantageous to at least partially wash the solid phase with water. The solid phase is then subsequently suspended with a suitable amount of water for a subsequent wet classification and fed to the classification according to step c).

[0081] The advantage of this embodiment is that only a diluted sulfuric acid-gypsum suspension or water-gypsum suspension is added to the wet classification process, thus simplifying the material selection for the classification process and increasing occupational safety.

[0082] By appropriately selecting the classification method, in step c) the coarse fraction, consisting mainly of dihydrate, and the fine fraction, enriched with the anhydrite phase, can now be separated from each other.

[0083] The anhydrite fine fraction is characterized by smaller crystal sizes on average compared to the dihydrate coarse fraction. The separation of the fine and coarse fractions in step c) must be selected such that the D v (50)-value of the fine fraction < 50 pm, preferably < 30 pm, particularly preferably < 20 pm and particularly preferably < 15 pm.

[0084] This can be achieved, for example, by appropriately selecting the sieve fractions for wet or dry sieving, by adjusting the pump pressure, choosing the geometry of the hydrocyclone, and selecting the apex nozzle and / or vortex finder for hydroclassification, or by adjusting the airflow and type of classifier when, for example, wind classification is chosen as the method.

[0085] In one variant of the inventive method, classification step c) is divided into two classification steps, wherein in the first classification step a coarse fraction and a partial fine fraction are initially produced, and in the second step a partial coarse fraction and a fine fraction are produced. This can be achieved by selecting a classification method, for example by a hydrocyclone, with a high particle size, preferably > 40 pm, in the first classification step and a classification method, for example by a hydrocyclone, with a low particle size, preferably < 25 pm, in the second classification step. 230102P10WG

[0086] According to this embodiment, the coarse fraction represents the product stream, the semi-coarse fraction the seed crystal stream, and the fine fraction the waste stream (AH / DH waste).

[0087] In another embodiment of the inventive process, classification step c) can be divided into two classification steps, wherein in the first classification step a partial coarse fraction and a fine fraction are initially produced, and in the second step a partial fine fraction and a coarse fraction are produced. This can be achieved by selecting a classification process, for example, a hydrocyclone, with a low particle size, preferably < 25 pm, in the first classification step, and a classification process, for example, a hydrocyclone, with a high particle size, preferably > 40 pm, in the second classification step. According to this embodiment, the coarse fraction represents the product stream, the partial fine fraction the seed crystal stream, and the fine fraction the waste stream (AH / DH waste). The particle size is defined as the particle size at which a particle has a 50% chance of being separated into the coarse or fine fraction.

[0088] In both described embodiments, a preferred variant allows the fine or semi-fine fraction, or both fractions, to be concentrated (thickened). This can be achieved using suitable methods such as hydroclassification, settling tanks, centrifugation, or similar processes. The resulting liquid phase can then be fed into the process for washing the gypsum, for example, in a separation unit, for diluting the suspensions in the reactor vessel, to a mixing tank, or to a dilution unit.

[0089] Furthermore, it can be advantageous to dilute the coarse fraction in both described embodiments. The liquid phase required for this can originate from the thickening units after the fine and semi-fine fractions, the liquid phases from the separation units, the reactor vessels, or any combination thereof.

[0090] In a preferred embodiment, the separation of the fine fraction from the coarse fraction takes place in a separate step d), wherein the fine and coarse fractions are separated from the suspension as solid phases.

[0091] In one embodiment, after step c), preferably between steps c) and d), a further recrystallization of the fine fraction and / or coarse fraction to the dihydrate form can take place in a further reactor vessel. This is advantageous for removing any remaining anhydrite from the coarse fraction. 230102P10WG

[0092] In a preferred embodiment, in a further step e) the solid phases of the fine and coarse fractions are separated from the sulfuric acid phase, washed, and dried. Preferably, a filtration unit can be used for the separation and washing.

[0093] The fine fraction enriched with radium (Ra) (anhydrite phase) can be stored as a waste stream or used as an additive for various applications, such as as a sulfate carrier in the cement industry or for example for the production of Ra-containing pharmaceuticals (e.g. Xofigo®).

[0094] In a further embodiment of the process, the anhydrite phase (fine fraction) can be recrystallized to the dihydrate after separation from the dihydrate phase (coarse fraction) in step d). This can be achieved, for example, by adding dihydrate seed crystals again. Optionally, after this process step, the dihydrate can be reclassified from the anhydrite to obtain a further quantity of "good fraction" or processed directly. Depending on the requirements, the dihydrate obtained in this way can also be used in the gypsum industry, in road construction, as an additive for the cement industry, and in other applications.

[0095] The solubility of calcium sulfate in aqueous and sulfuric acid solutions (approx. 2 to 3 g / l) is significantly higher compared to the solubility of radium sulfate (approx. 2 to 3 mg / l). With the S / L ratios underlying the process according to the invention, between 1 / 10 kg / kg and 1 / 1 kg / kg, the higher solubility of calcium sulfate can lead to an undesirable concentration of radium.

[0096] In an alternative embodiment of the process according to the invention, the liquid streams saturated with calcium sulfate can therefore be reused, or fresh aqueous and sulfuric acid solutions saturated with calcium sulfate can be used. For this purpose, for example, a portion of the final and / or initial and / or an intermediate product from steps a) to e) of the process can be used to generate calcium sulfate-saturated liquid streams. This prevents an undesired concentration of radium and thus an increased specific activity in the solid.

[0097] During the conversion to the dihydrate from anhydrite, the sulfuric acid is concentrated in such a way that, after separation and washing of the solid phase from the liquid phase, a desired dilution of the sulfuric acid to the original 230102P10WG is achieved.

[0098] Concentrations can be better controlled. This allows the reaction solution (sulfuric acid) to be almost completely recycled back into the process and thus reused.

[0099] The advantage of this embodiment is that only small amounts of reaction solution are consumed, simple temperature control is possible since hardly any heat of dilution is generated by mixing sulfuric acid and water, and the small quantities withdrawn can be more easily fed into one of the integrated upstream or downstream processes.

[0100] The resulting Ra concentration in the fine fraction after step e) of the process according to the invention depends on factors such as the original Ra concentration of the gypsum, the distribution coefficient of radium in the anhydrite and dihydrate phases, the mass fraction of anhydrite after step b) of the process according to the invention, the efficiency of the classification and consequently the percentage of anhydrite in the fine fraction and the mass fraction of the final fine fraction, based on the original mass.

[0101] The first two factors are largely uncontrollable, as they are determined by the choice and process parameters of the phosphoric acid production process. By using the appropriate particle size of the anhydrite in step a) of the process according to the invention, the appropriate particle size of the dihydrate in step b), maintaining the mass fraction of anhydrite in step c), selecting the required and most suitable classification method in step c), and adjusting the number or combination of classification methods, for example by connecting several classifiers or hydrocyclones in series, the Ra load in the fine fraction can thus be optimized.

[0102] The invention further relates to a plant for reducing radioactive components from calcium sulfate, wherein the plant comprises the following equipment:

[0103] 1) A mixing tank having at least one feed for feed material and at least one feed line for aqueous sulfuric acid solution with a concentration of 2 to 18 wt.%, wherein the feed material comprises calcium sulfate comprising one or more of the components selected from dihydrate (CaSO4-2H2O), hemihydrate (CaSO4-1J H2O) or anhydrite (CaSO4), wherein the anhydrite is present in an amount of at least 50 wt.%, based on the dry calcium sulfate, and has a particle size D v (50) of less than or equal to 25 pm, measured by laser diffraction, exhibits, 230102P10WG

[0104] 2) at least one reactor unit, which is fed from the mixing tank via at least one supply line, wherein at least the reactor unit is designed to hold a suspension comprising calcium sulfate and an aqueous sulfuric acid solution with a concentration of 2 to 18 wt.% at a temperature of 0° to 35°C,

[0105] 3) at least one first and one second classification unit, each selected from one or more of the group consisting of: i) one or more hydrocyclones connected in series, ii) one or more classifiers connected in series, iii) one or more dry sieves or iv) one or more wet sieves, wherein these are designed to separate the calcium sulfate into at least one fine and one coarse fraction and

[0106] 4) at least one separation unit designed to separate the solid and liquid phases of the fine and coarse fractions.

[0107] In a preferred embodiment, the system according to the invention comprises at least one further reactor unit designed to age the fine and / or coarse fractions obtained from at least one of the classification units, thereby forming a dihydrate. The aging process involves a further conversion of the anhydrite fraction to the dihydrate. If two further reaction vessels are used, the fine fraction can be aged in one and the coarse fraction in the other.

[0108] In a preferred embodiment, the system comprises a separation unit 9, particularly preferably the separation unit 9, 7a and 7b.

[0109] In a further preferred embodiment, the system further comprises

[0110] 5) at least one fluidic connection designed for the return of the

[0111] Sulfuric acid solution from the reactor unit into the mixing tank as well as

[0112] 6) at least one fluidic connection designed for the return of the

[0113] Sulfuric acid solution from the separation unit into the mixing tank.

[0114] In a preferred embodiment of the system, for example in wet classification, the mixing tank is fluidically connected to a feed line for the feed material and at least one feed line for the aqueous sulfuric acid solution. The mixing tank is preferably configured to contain a suspension comprising calcium sulfate and an aqueous sulfuric acid solution.

[0115] to form a sulfuric acid solution. In a further preferred embodiment, at least one outlet from the mixing tank is connected to at least one feed line to at least one reactor unit. The reactor unit is preferably configured to receive and / or maintain the suspension comprising calcium sulfate and an aqueous sulfuric acid solution. The reactor unit is thus configured to enable the conversion of the anhydrite contained in the feed material to the dihydrate.

[0116] In another embodiment, the at least one mixing tank and the at least one reactor unit can be located in a common container or designed as a single, interconnected container.

[0117] In a further embodiment, at least one outlet of the at least one reactor unit can be fluidically connected to at least one inlet of at least one pre-separation unit. In a further embodiment, at least one outlet of the at least one pre-separation unit can optionally be connected to at least one inlet of at least one dilution unit. In a further embodiment, the at least one outlet of the at least one reactor unit can be fluidically connected directly to at least one inlet of at least one dilution unit. In a further embodiment, at least one outlet of the at least one dilution unit can be directly connected to at least one inlet of at least one first classification unit.In a further embodiment, at least one outlet of the at least one reactor unit can be directly connected to at least one inlet of the at least one first classification unit, which is designed to form at least one coarse and at least one fine fraction.

[0118] At least one outlet of the at least one first classification unit is connected to at least one inlet of at least one second classification unit. Particularly preferably, the at least one outlet of the at least one first classification unit for the fine fraction is connected to at least one inlet of a second classification unit.

[0119] The first classification unit at least includes at least one outlet for a fine fraction and one outlet for a coarse fraction.

[0120] Preferably, at least one outlet of the at least one first classification unit is connected to at least one inlet of at least one second downstream reactor unit. Particularly preferably, the at least one outlet of the at least one first classification unit for the coarse fraction is connected to at least one inlet of a second downstream reactor unit (230102P10WG).

[0121] Preferably, the at least one second classification unit has at least one outlet for a fine fraction and at least one outlet for a coarse fraction, wherein the at least one outlet of the fine fraction is connected to at least one inlet of the at least one mixing tank and / or at least one inlet of the at least one reactor unit and at least one inlet of at least one downstream reactor unit. Preferably, the outlet of the coarse fraction of the second classification unit is connected to at least one inlet of the at least one mixing tank and / or to at least one inlet of the at least one reactor.

[0122] Preferably, the fine fraction of the first classification unit is the sub-fine fraction, and the coarse fraction of the first classification unit is the coarse fraction. In this case, the fine fraction of the second classification unit is the fine fraction, and the coarse fraction of the second classification unit is the sub-coarse fraction.

[0123] Preferably, at least one outlet of the first downstream reactor unit is fluidically connected to at least one inlet of at least one first separation unit, wherein the at least one first separation unit comprises at least one outlet for a substantially solid-containing waste stream and at least one outlet for at least one substantially liquid-bearing stream. The at least one first downstream reactor unit preferably also has an outlet for a substantially liquid-bearing stream.

[0124] Preferably, the at least one outlet of the at least one first downstream reactor unit, which essentially comprises a liquid-carrying stream, and / or the at least one outlet of the at least one first separation unit, which essentially comprises a liquid-carrying stream, is fluidically connected to at least one inlet of the at least one dilution unit and / or to at least one inlet of the at least one mixing tank.

[0125] Preferably, at least one second downstream reactor unit is connected via at least one outlet to at least one inlet of at least one second separation unit. The at least one second separation unit has at least one outlet for a product stream consisting mainly of solids and at least one outlet for a stream consisting mainly of liquids. The at least one second downstream reactor unit further has at least one outlet for a stream consisting mainly of liquids.

[0126] Preferably, the at least one outlet of the at least one second downstream reactor unit, which essentially comprises a liquid-carrying stream, and / or the at least one outlet of the at least one second separation unit, which essentially comprises a liquid-carrying stream, is fluidically connected to at least one inlet of the at least one dilution unit and / or to at least one inlet of the at least one mixing tank.

[0127] In a further preferred embodiment of the wet classification system configuration, at least one outlet of the at least one first classification unit is connected to at least one inlet of at least one second classification unit. Particularly preferred is the at least one outlet of the at least one first classification unit for the coarse fraction connected to at least one inlet of a second classification unit.

[0128] Preferably, at least one first classification unit has at least one outlet for a fine fraction and one outlet for a coarse fraction.

[0129] Preferably, at least one outlet of the at least one first classification unit is connected to at least one inlet of at least one first downstream reactor unit and at least one inlet of at least one mixing tank and / or at least one inlet of at least one reactor unit. Particularly preferably, at least one second outlet of the at least one first classification unit for the coarse fraction is connected to at least one inlet of a second classification unit.

[0130] Preferably, the classification unit comprises at least one second unit via at least one outlet for a fine fraction and at least one outlet for a coarse fraction, wherein the at least one outlet for the fine fraction is connected to at least one inlet of the at least one mixing tank and / or at least one inlet of the at least one reactor unit. Preferably, the outlet for the coarse fraction of the second classification unit is connected to at least one inlet of at least one second downstream reactor unit.

[0131] Preferably, the fine fraction of the first classification unit is the fine fraction, and the coarse fraction of the first classification unit is the partial coarse fraction. In this case, the fine fraction of the second classification unit is the partial fine fraction, and the coarse fraction of the second classification unit is the coarse fraction.

[0132] The fluidic connections described above apply analogously to the interconnections of at least one first downstream reactor unit with at least one first separation unit, as well as to the interconnections of at least one second downstream reactor unit with at least one second separation unit. The same applies to the connection of at least one first and at least one second downstream reactor unit and separation unit with at least one mixing tank and at least one reactor unit, as well as at least one dilution unit.

[0133] In one embodiment of the dry classification system, the system configuration described above applies analogously. Preferably, however, the at least one dilution unit is replaced by at least one drying unit. In a preferred embodiment of the dry classification system, the at least one pre-separation unit can be fluidically connected to at least one inlet of the at least one drying unit via at least one outlet. In a further embodiment, the at least one drying unit can further comprise at least one outlet which is fluidically connected to at least one inlet of the at least one mixing tank. The at least one drying unit can further comprise at least one outlet which is fluidically connected to at least one inlet of the at least one first classification unit.

[0134] In another embodiment of the dry classification, the above-mentioned at least one first and second downstream reactor units and the at least one first and second separation units, as well as the associated fluidic connections to the at least one inlet of the at least one mixing tank, can be components of the plant.

[0135] The equipment can be used for the methods described above according to the invention. For details of the individual devices or units, reference is made to the preceding information. The preceding information for the method(s) applies accordingly to the equipment. The methods according to the invention, as described above and in the claims, are preferably carried out in the equipment described in this document and in the claims.

[0136] Unless otherwise stated in individual cases, the various embodiments of the invention mentioned in this application can be advantageously combined with one another.

[0137] The invention is explained below in exemplary embodiments with reference to the accompanying drawings. The drawings show: 230102P10WG

[0138] Figures 1a and 1b show a schematic representation of two embodiments of the process according to the invention for the purification of calcium sulfate from phosphoric acid production by means of wet classification.

[0139] Figures 2a and 2b show a schematic representation of two embodiments of the process according to the invention for the purification of calcium sulfate from phosphoric acid production by means of dry classification.

[0140] Figures 3a and 3b show a schematic representation of an embodiment of wet classification with thickening of the fine and semi-fine fraction (Figure 3a) and an embodiment of wet classification with dilution of the semi-coarse fraction (Figure 3b).

[0141] Figures 4a, 4b, 4c show a schematic representation of three embodiments of wet classification with variants of seed crystal extraction.

[0142] Figures 5a, 5b, 5c show a dry classification with variations in seed crystal extraction.

[0143] Figure 6 is a graphical representation of the concentration effects of sulfuric acid.

[0144] Figure 7 shows a graphical representation of the temperature influence from 15°C to 30°C (with two different IK values).

[0145] Figure 8 shows a graphical representation of the influence of the IK mass or the total surface area IK, for the same IK size.

[0146] Figure 9 shows a graphical representation of the influence of the IK size or the total surface area IK, for the same IK mass.

[0147] Figure 10 shows a graphical representation of the influence of the mixing intensity (shaken versus still),

[0148] Figure 11 is a graphical representation of the influence of the S / L ratio,

[0149] Figure 12 shows a first graphical representation of the particle size distribution after recrystallization,

[0150] Figure 13 shows a second graphical representation of the particle size distribution after recrystallization. 230102P10WG

[0151] Figures 1a, 1b, 2a and 2b show schematic flow diagrams for the purification of calcium sulfate from phosphoric acid production, in particular for the reduction of the radioactive content in calcium sulfate according to a possible embodiment of the process according to the invention, wherein figures 1a and 1b each show a schematic flow diagram for a wet classification and figures 2a and 2b each show a schematic flow diagram for a dry classification.

[0152] In a mixing tank 3, the aqueous sulfuric acid solution 1, which has a concentration of 2 to 18 wt%, and the feed material 2, which contains at least 50 wt% calcium sulfate anhydrite (CaSO4), are mixed. The feed material 2 comprises calcium sulfate from phosphoric acid production, which contains one or more of the components selected from the dihydrate (CaSO4-2H2O), hemihydrate (CaSO4- 1The suspension contains calcium sulfate (calcium sulfate) or anhydrite (CaSO4), wherein the anhydrite is present in an amount of at least 50 wt%, based on the dry calcium sulfate, and has a particle size Dv(50) of less than or equal to 25 pm, measured by laser diffraction. A preferred S / L ratio is in the range of 0.1 kg / kg to 0.33 kg / kg. The temperature of the suspension is set and maintained in the mixing tank 3, or at the latest in the reactor unit 4, at a temperature between 0 and 35 °C. For this purpose, heat exchangers operated with a cooling medium, preferably cooling water, can be used, for example. In an alternative embodiment, the mixing tank 3 and the reactor unit 4 can form a single unit and are therefore not designed as separate units.

[0153] As the reaction progresses in reactor unit 4 (from anhydrite to dihydrate), the aqueous sulfuric acid solution 1 experiences an increase in concentration (for example, up to almost 1 wt% increase at a starting concentration of 10 wt% and S / L = 4, with almost complete conversion to dihydrate). After separation of the coarse and fine fractions via wet or dry classification (Figure 5a and Figure 5b), the solid and liquid phases are separated in separation units 7a and 7b, and the solid fraction is washed. In this process, as much of the sulfuric acid solution 1 as possible, originating from the reaction mixture, is recycled back into the process (see the dashed line in Figures 1a and 1b and 2a and 2b, respectively), and only small amounts of wash water are used to dilute the concentrated solution from reactor unit 4 in a dilution unit 8. As shown in Figures 1a, 1b, 2a and 2b, a pre-separation unit 9 can be installed between reactor unit 4 and dilution unit 8.It is also possible that a liquid phase of sulfuric acid and / or wash water from the downstream reaction vessels 6a or 6b or the separation units 7a or 7b or any combination of the aforementioned is supplied to the dilution unit 8 and / or that fresh water is supplied directly.

[0154] In a preferred embodiment, as schematically illustrated in Figures 1a and 2a, the coarse fraction from the second classification unit 5b is recycled to the reactor unit 4 and / or the mixing tank 3. This coarse fraction serves primarily as the seed crystal stream. If necessary, the fine fraction from the second classification unit 5b can be at least partially recycled to the mixing tank 3 or reactor unit 4 for further recrystallization of the remaining anhydrite phase.

[0155] In a further embodiment, as schematically illustrated in Figures 1b and 2b, the partial fine fraction from the second classification unit 5b is recycled to the reactor unit 4 and / or to the mixing tank 3. In this case, the partial fine fraction serves primarily as a seed crystal stream. If necessary, the fine fraction from the first classification unit 5a can be recycled, at least partially, to the mixing tank 3 or to the reactor unit 4 for further recrystallization of the remaining anhydrite phase.

[0156] In an alternative embodiment of the variant shown in Figure 1a, as illustrated in Figure 3a, the fine and semi-fine fractions after the second classification unit 5b or the first classification unit 5a can be concentrated via a thickening unit 11. The thickening unit 11 can consist of one or more hydrocyclones, a separation unit, a drying unit, or a combination thereof. The resulting liquid phases can be fed to the separation units 7a and / or 7b, the dilution unit 8, the mixing tank 3, the reactor unit 4, or a combination thereof for washing. Using a thickening unit 11 for the semi-fine fraction from the first classification unit 5a has the advantage that the solids concentration, in the case of a subsequent classification step in the second classification unit 5b with a hydrocyclone, is within a more favorable range.Thickening the fine fraction from the second classification unit 5b has the advantage that no undesirable reduction in the concentration of sulfuric acid occurs when it is fed into the mixing tank 3 or into the reactor unit 4.

[0157] Figure 3b shows a variant of the embodiment as described in Figure 1b. As described for Figure 3a, the fine and semi-fine fractions are concentrated after the classification step in the classification units 5b and 5a, respectively, via a thickening unit 11. Additionally, the semi-coarse fraction from the first classification unit 5a is first diluted via a dilution unit 8 before being fed into the second classification unit 5b. The 230102P10WG

[0158] Dilution unit 8 for the partial coarse fraction from the first classification unit 5a has the advantage that the solids concentration in the case of a subsequent classification step in the second classification unit 5b with a hydrocyclone is in a more favorable range.

[0159] Any shortfall in aqueous sulfuric acid solution 1 is added to the mixing tank 3, whereby the sulfuric acid solution 1 should be brought to the desired sulfuric acid concentrations in advance if possible, in order to avoid heat of dilution in the mixing tank 3.

[0160] The size of the mixing tank 3 should be kept as small as possible so that the calculated total volume can be divided, for example, between mixing tank 3 and reactor unit 4 in a ratio of 1 / 99 to 10 / 90. The function of mixing tank 3 is to thoroughly mix the fresh feed material 2, fresh and / or refluxed sulfuric acid 1, and / or the reflux fractions from the second classification unit 5b (see Figure 1a) or the reflux fractions from classification units 5a and 5b (see Figure 1b). Mixing tank 3 and / or reactor unit 4 can each be a single container or any number of containers connected in parallel and / or in series.

[0161] Reactor unit 4 should be designed to minimize movement in its lower section. The feed from mixing tank 3 can, for example, be made in the middle section of reactor unit 4, also with minimal turbulence of the reaction solution. A reactor unit 4 with three zones is therefore preferred. In the upper section, for instance, solid particles can settle, allowing largely solid-free aqueous sulfuric acid solution 1 to be fed into mixing tank 3 or (not shown) used for other purposes, such as diluting the thickened suspensions before / after classification units 5a and 5b, preferably hydrocyclones. The fresh product 2 from mixing tank 3 can be fed into the middle section, ideally without turbulence. Continuous settling of the "mature" product can take place in the lower section. Preferably, no or only minimal mixing occurs.

[0162] Figures 1a and 1b show two classification units 5a and 5b connected in series. Preferably, the classification units 5a and 5b are hydrocyclones. Alternatively, special hydrocyclones capable of simultaneously producing three fractions can be used. The two classification units produce the four fractions: coarse fraction, semi-coarse fraction, semi-fine fraction, and fine fraction. (230102P10WG)

[0163] The coarse fraction is the product stream. In this embodiment, the partial coarse fraction serves as a seed crystal stream and can be fed to the mixing tank 3 and / or the reactor unit 4 (see Figure 1a), or alternatively, the partial fine fraction can serve as a seed crystal stream and can be fed to the mixing tank 3 and / or the reactor unit 4 (see Figure 1b). If necessary, the fine fraction can be at least partially returned to the mixing tank 3 or the reactor unit 4 for further recrystallization of the remaining anhydrite phase.

[0164] In a preferred embodiment, a first separation of particle sizes > 500 pm can be carried out before feeding the material onto the first and second classification units 5a and 5b, respectively, preferably a first hydrocyclone 5a and a second hydrocyclone 5b. This pre-separation can be performed, for example, by various separation methods such as wet sieving, flotation, or hydrocyclone. This allows, for example, the removal of quartz components and unreacted apatite particles, thus protecting the lining of the hydrocyclones. Additionally, the final product (dihydrate) can be purified of quartz components and other impurities in this way.

[0165] In a further embodiment, a first separation of particle sizes > 500 pm can already take place with the feed material 2 or before step a) of the process according to the invention. This pre-separation can also be carried out by various separation processes such as wet sieving, flotation, hydrocyclone, air classification or dry sieving.

[0166] As shown in Figures 1a, 1b and 2a, 2b, the material streams (AH / DH waste and DH product) can be aged in further downstream reactor vessels 6a and 6b before the liquid phase is separated in the separation units 7a and 7b, respectively. This ensures that as much dihydrate product as possible is obtained. A further advantage is that the accompanying particle growth promotes the separation of the solid and liquid phases in the separation units 7a and 7b.

[0167] Figures 4a, 4b, and 4c show wet classification with variations of seed crystal removal. In a preferred embodiment, seed crystal removal is carried out via the partial coarse fraction after the second classification unit 5b (see embodiment shown in Figure 1a) or via the partial fine fraction after the second classification unit 5b (see embodiment shown in Figure 1b). In some cases, however, it may be advantageous to carry out seed crystal removal partially or completely via the embodiments shown in Figures 4a, 4b, and 4c. This is the case when the lowest possible 230102P10WG

[0168] A change in the concentration of sulfuric acid 1 in the mixing tank 3 and / or the reactor unit 4 is desired. In the embodiment according to Figure 4a, the sulfuric acid concentration remains virtually unchanged. In the embodiment according to Figure 4b, for example, a thickened suspension and, according to the embodiment according to Figure 4c, solids (wet or dry) from the separation units 7a and / or 7b can be fed into the mixing tank 3 and / or the reactor unit 4.

[0169] Figures 5a, 5b, and 5c show embodiments of the dry classification process with seed crystal extraction. In the dry classification process, a separation unit 9 is additionally used to separate the aqueous phase from the solid phase (calcium sulfate dihydrate), and optionally, the gypsum can also be washed. The gypsum is then dried in a calcium sulfate drying unit 10.

[0170] In another embodiment, a portion of the seed crystals can be added to the first downstream reactor vessel 6a to stimulate the growth of the AH / DH waste particles. This allows for improved filterability of the AH / DH waste stream.

[0171] In a preferred embodiment, the product streams AH / DH waste and DH product can be thickened before being fed to the solid / liquid separation units 7a and 7b. This significantly reduces the loss of the liquid phase. Thickening can be achieved, for example, by a settler and / or an additional hydrocyclone unit.

[0172] In accordance with the present invention, the activity of a radioactive sample serves to characterize the "strength" of the radiation source. The activity (with the unit becquerel; Bq) indicates the number of decays in a radioactive source per time interval. To better compare the activity of different samples, the ratio of the activity to the mass of the sample (Bq / kg) is calculated. This ratio is called the specific activity of the sample.

[0173] If the gypsum (product) purified by the inventive method is used directly for the production of plasterboard, plaster of Paris, or similar products, an index value of 0.5 should be targeted. Since Ra-226 is usually the main source of radioactivity, this results in a target value for Ra-226 of < 150 Bq / kg.

[0174] If the same index value is applied for use as an additive in the cement industry (as a setting retarder), a target value for Ra-226 of < 3000 Bq / kg is calculated based on the permitted maximum ~5% addition of additive (assuming that this is the only source of radioactivity in the finished cement).

[0175] Consequently, with appropriate gypsum and adjustment of the split from coarse to fine fraction, this offers the possibility of using both the finally radioactivity-depleted gypsum fraction and the waste stream for further use as a building material.

[0176] Since the radioactive components are primarily concentrated in the fine fraction, and in the case of high specific activities of the starting material, a split of 70 / 30 to 95 / 5 (coarse fraction to fine fraction) is preferred.

[0177] For lower specific activities or when both fractions are to be reused, a split of 50 / 50 to 95 / 5 is preferred, since a lower ratio of coarse fraction to fine fraction simultaneously reduces the residence time and thus the required equipment size.

[0178] Examples

[0179] Examples of recrystallization (kinetics and final particle size distribution), reaction control and reduction of specific activity using classification methods are listed below.

[0180] Three different phosphogypsums, designated PG A, PG B and PG C, were used as starting materials (see Table 1).

[0181] The phosphogypsums were dried for at least 24 hours at 40 to 50 °C before and after treatment according to the described procedures to remove free water. The chemical composition of the gypsums was determined before and after treatment by X-ray fluorescence analysis (XRF) on an Axios Advanced spectrometer from PANalytical using the SuperQ 5.3B software package. For this purpose, the gypsum was digested using lithium tetraborate. The fluoride content was determined after digestion of the gypsum with sodium peroxide and hydrochloric acid using an ion-selective electrode. In addition, the mineralogical composition of the gypsums was analyzed before and after treatment by powder diffractometry on a Bruker D4 Endeavor diffractometer. The Bruker Topas 4.2 software package was used for evaluation using the Rietveld method. vThe (50) value of the particle size distribution was determined on a Malvern Mastersizer 3000 using ethanol as the dispersing medium. The Fraunhofer model was applied as the scattering model. The specific activities of the gypsum were determined externally at lAF-Radioökologie GmbH, Radeberg 230102P10WG.

[0182] Table 1: Analyses of the untreated phosphogypsums designated PG A, PG B and PG C (The samples were dried at 40 °C before analysis). The three phosphogypsums (PG-A, PG-B, and PG-C) were treated as described below and converted to anhydrite. Typical analytical results of the treated phosphogypsums (PG-LE-A, PG-LE-B, and PG-LE-C) are listed in Table 2. Such treated phosphogypsums were used for the process according to the invention and (partially) recrystallized and subsequently partially classified. Preparation of PG-A

[0183] 150 g of a phosphogypsum designated "PG A" was stirred with 300 ml (S / L = 0.5) of 5 molar sulfuric acid for 90 minutes at 75 °C using a KPG stirrer. After this time, the suspension was rapidly filtered and washed twice with 172.5 ml of water at room temperature. The reaction was repeated several times under identical conditions to obtain suitable quantities for the subsequent steps. The analysis shown (Table 2) was performed on the pooled sample from the multiple reactions.

[0184] Production of PG-B

[0185] 150 g of a phosphogypsum designated as "PG B" was reacted with 300 ml (S / L = 0.5) 5 molar 230102P10WG

[0186] Sulfuric acid was stirred for 45 minutes at 75 °C using a KPG stirrer. After this time, the suspension was rapidly filtered and washed twice with 172.5 ml of water at room temperature.

[0187] Preparation of PG-C: 150 g of a phosphogypsum designated as "PG C" was reacted with 300 ml (S / L = 0.5) 5 molar

[0188] Sulfuric acid was stirred for 55 minutes at 75 °C using a KPG stirrer. After this time, the suspension was rapidly filtered and washed twice with 172.5 ml of water at room temperature. The reaction was repeated several times under identical conditions to obtain suitable quantities for the subsequent steps. The analysis shown (Table 2) was performed on the pooled sample from the multiple reactions.

[0189] For example 22, the treated phosphogypsum PG-LE-A was finely ground in a ball mill (designation PG-LE-A_M).

[0190] Table 2: Analysis results of the treated phosphogypsums PG-LE-A, PG-LE-B and PG-LE-C; For PG-LE-A_M, a quantity of PG-LE-A was ground to the specified fineness using a ball mill. 230102P10WG

[0191] Examples 1 to 22:

[0192] Reaction rate of recrystallization

[0193] The effects of parameter variations of influencing factors such as concentration of

[0194] Sulfuric acid, temperature of the recrystallizations, amount of seed crystal, seed crystal size, moving versus stationary suspension, influence of the S / L ratio and prior milling of the anhydrite are shown below:

[0195] Several aspects of the recrystallization can be seen in the results (see Table 3). For examples 1 to 22, the previously treated phosphogypsum PG A (PG-LE- A) was used and recrystallized.

[0196] Table 3: Examples of the reaction rate of recrystallization under different reaction conditions

[0197]

[0198] 230102P10WG All examples (except example 22) were carried out using 5 g to 10 g of purified phosphogypsum (designation PG-LE-A; see also Table 2 for typical composition, particle sizes and specific activity).

[0199] The amount of sulfuric acid (in g) was added according to the specified S / L ratio (in g / g).

[0200] For example 22, 200 g of phosphogypsum designated PG-LE-A_M (see Table 2 for particle size characteristics) were mixed with 800 g of 14% sulfuric acid and stirred at 25 °C at 500 revolutions per minute; no seed crystals were used; the dihydrate content was determined not by acid-base titration as in examples 1-21, but by XRD measurement; see also the following section vii) under "Discussion of the Results" for further explanation.

[0201] Explanation of the Settings column:

[0202] T: corresponds to the temperature in °C; all experiments, with the exception of examples 1–6, were carried out in an INCU-Line® 150R incubator from VWR. Experiments 1 to 6 were performed in an air-conditioned room, with the temperature recorded by a data logger, using a Reax 20 / 8 overhead shaker from Heidolph Instruments GmbH & Co. KG at 10 revolutions per minute. c(SA): corresponds to the sulfuric acid concentration used. The exact concentrations were determined in each case by acid-base titration with a 1 M sodium hydroxide standard solution from CarlRoth using an Eco-Titrator from Metrohm.

[0203] S / L: corresponds to the S / L ratio; the S / L ratio refers to the mass of phosphogypsum PG-LE-A or PG-LE-A_M used, without taking into account the amount of seed crystal to the mass of sulfuric acid.

[0204] IK %: corresponds to the added mass of seed crystals relative to the mass of phosphogypsum PG-LE-A or PG-LE-A_M used; e.g., IK = 10% and 10 g of phosphogypsum PG-LE-A means an addition of 1 g IK to 10 g of phosphogypsum PG-LE-A

[0205] IK m: corresponds to the size of the seed crystals used; the position parameter x63;3 of the RRSB distribution was defined as the size value; the position parameter was determined using Mastersizer 3000 software of the Malvern 3000 from Malvern Panalytical GmbH. Agitated / unagitated: The samples were either agitated using the aforementioned overhead shaker (agitated) or briefly shaken after preparation of the suspension and then left to stand. The unagitated (unagitated) samples were only briefly shaken again when a sample was taken.

[0206] Explanation of the DH(ti) column:

[0207] Representative samples were taken at the specified times and filtered through a 1.6 pm glass microfiber filter from Whatmann. The filtrate (sulfuric acid) was collected, and its concentration was determined using the acid-base titration described above. The conversion of the anhydrite to the dihydrate results in the continuous removal of water molecules from the filtrate, which manifests as a steadily increasing sulfuric acid concentration. Knowing the initial mass of the gypsum phosphate, the mass and concentration of the original sulfuric acid, and the initial composition of the gypsum phosphate used, this change in sulfuric acid concentration can be directly attributed to the conversion to the dihydrate.The dihydrate concentrations in the table above refer solely to the convertible anhydrite concentration of the phosphogypsum used; dihydrate content of the amount of seed crystal used is not taken into account according to the calculation in Table 3.

[0208] Discussion of the results:

[0209] I) Influence of sulfuric acid concentration:

[0210] A comparison of Examples 10 to 12 and 19 (see Table 3 and Figure 5) shows that lower initial sulfuric acid concentrations favor the reaction rate of the conversion to the dihydrate. A lower limit of the accelerating effect is estimated to be around 5 wt% sulfuric acid. ii) Influence of temperature:

[0211] The influence of temperature was investigated in more detail in the range between 15 and 30 °C (see Figure 6). Examples 7, 11, 15 and 21 (with seed crystals of a size defined as position parameter x63;3 according to the RRSB function; x63;3 = 45.14 pm) and examples 6, 8, 17 and 20 (x63;3 = 11.54 pm) show that a significant acceleration of the reaction rate results at lower temperatures.

[0212] Hi) Influence of the amount of seed crystals (or total surface area of ​​the seed crystals)

[0213] A comparison of Examples 1, 2, and 3 (see Figure 7) shows that increasing the number of seed crystals, and thus increasing the total surface area of ​​seed crystals, leads to a significant acceleration of the reaction rate. 230102P10WG iv) Influence of seed crystal size (or total surface area of ​​seed crystals):

[0214] Comparisons of examples 1 (x63;3 = 45.14 pm) and 4 (x63;3 = 11.54 pm), 11 (x63;3 = 45.14 pm) and 6 (x63;3 = 11.54 pm), as well as 7 (x63;3 = 45.14 pm) and 8 (x63;3 = 11.54 pm) show (see Figure 8) that, with a constant seed crystal mass, the size (and thus the total surface area) of the seed crystals has a drastic influence on the reaction rate. With smaller seed crystals (and thus a higher total surface area), the reaction rate is increased. v) Influence of mixing intensity of the suspension:

[0215] Comparisons of Examples 2 (moving) and 5 (still) as well as 3 (moving) and 11 (still) show that mixing the suspension leads to an improvement in the reaction rate (see Figure 9). vi) Influence of the S / L ratio:

[0216] A comparison of Examples 11 (S / L = 1 / 4) and 13 (S / L = 1 / 3) shows that the S / L ratio has only a negligible influence on the reaction rate (see Figure 10). Only with longer reaction times is there a virtually insignificant improvement in conversion with a lower S / L ratio. This can be attributed to the fact that, with the same initial sulfuric acid concentrations, a higher S / L ratio results in a greater change in sulfuric acid concentration at the same conversion to the dihydrate. Thus, the measured sulfuric acid concentrations after 20.72 h for Example 11 and 20.98 h for Example 13, with nearly the same conversion to dihydrate of 71.5% versus 70.5%, are 14.62% and 14.83%, respectively (with an identical initial concentration of 13.98%). As already shown under i) Influence of sulfuric acid concentration, a higher sulfuric acid concentration causes a decrease in the reaction rate.vH) Influence of the particle size of the purified phosphogypsum (anhydrite) used.

[0217] As can be seen from Example 22, a drastic acceleration of the reaction rate can be achieved by prior grinding of the anhydrite. For this purpose, finely ground gypsum PG-LE-A_M (see Table 2) was prepared, and 200 g of it was mixed with 800 g of 14% sulfuric acid. The reaction was carried out in a double-walled, temperature-controlled reaction vessel at 25 °C with stirring at 500 revolutions per minute using a KPG stirrer. After only 3.45 h, almost complete conversion of 230102P10WG to the dihydrate (see Table 3) was observed.

[0218] Examples 23 to 28:

[0219] Influence on particle growth and final particle size after recrystallization

[0220] For examples 23 to 28, 10 g of purified phosphogypsum PG-LE-A were mixed with 40 g of 14% sulfuric acid and shaken at room temperature (RT) in a Reax 20 / 8 overhead shaker from Heidolph Instruments GmbH & Co. KG at 10 revolutions per minute. Different seed crystal sizes and quantities were used for these examples. After 20.85 to 21.4 h, the reactions were stopped, the suspension filtered, and the filter cake washed and dried. Examples 26 and 28 are duplicate determinations to verify reproducibility. The uncorrected dihydrate concentration—determined by the described acid-base titration—ranged from 88.7 to 92.3 wt% for all examples. Particle sizes of the recrystallized samples were determined using Malvern 3000 (see Table 4).Plotting the obtained distribution density against the size classes (see Figures 11 and 12) reveals two clear facts. First, larger seed crystals, with the same amount of seed crystals, result in a final product with larger dihydrate particles. Second, a larger amount of seed crystals, with the same seed crystal size, results in a final product with smaller dihydrate particles. For the subsequent classification, the largest possible dihydrate particles are advantageous, as this allows for a sharper separation between dihydrate (coarse material) and anhydrite (fine material).

[0221] Table 4: Examples of crystal growth when different amounts and different sized seed crystals are added under otherwise identical reaction conditions.

[0222] Examples of reaction control via measurement of the acid concentration of sulfuric acid in the suspension

[0223] To verify whether the sulfuric acid concentration was suitable for determining the conversion to the dihydrate, several phosphogypsum samples from Examples 1 to 22 were analyzed by XRD. As previously described, samples were taken at the times specified in Table 3, filtered through filter paper, and the sulfuric acid concentrations were determined. The filter cakes were then washed twice with water to remove the sulfuric acid from the phosphogypsum. The samples were subsequently dried at a maximum temperature of 40 °C, and the mineralogy was determined using the Rietveld method. As previously described, the calculated dihydrate and anhydrite concentrations from Table 3 and Table 5 (column DH (Tit.) and column AH (Tit.)) refer to the phosphogypsum sample used, without considering the amount and composition of the seed crystals.To compare the actual composition (determined by XRD) and the amount of anhydrite reacted (determined by titration), the dihydrate and anhydrite concentrations determined by acid titration were corrected for the amount and composition of seed crystals used (see Table 6 for the composition of the seed crystals used) (see Table 5, columns DH (corr.) and AH (corr.)). Table 5, columns AAH and ADH, show the absolute deviations in wt% between the corrected concentrations of anhydrite and dihydrate after acid-base titration and the concentrations determined by XRD. The results show very good agreement between titration and XRD.

[0224] Table 5: Examples of reaction control via measurement of the acid concentration of sulfuric acid in the suspension; given are the mineralogy of the examples determined according to the Rietveld method, the uncorrected anhydrite and dihydrate mass fractions after acid-base titration, the corrected mass fractions for dihydrate and anhydrite, and the absolute deviation of the corrected mass fractions from the mineralogy.

[0225] Table 6: Mineralogy of the seed crystals used

[0226] Example 29:

[0227] Cleaning efficiency with regard to other trace elements

[0228] As a further aspect, the purification efficiency with respect to other trace elements was investigated using an ICP-OES analyzer from SPECTRO Analytical Instruments GmbH (model Spectro Arcos SOP). For this purpose, the purified phosphogypsum obtained were dissolved in aqua regia by microwave digestion (MLS-MWS ETHOS from MLS Microwave Laboratory Systems GmbH), and the solutions were analyzed for the various trace elements. Table 8 shows, as examples, the results for untreated gypsum A (designated PG A), a typical result for treated gypsum PG-LE-A, and gypsum recrystallized according to the inventive process (Example 29). For direct comparability, the values ​​are calculated without loss on ignition. Loss on ignition, determined at 1050 °C in a muffle furnace, was 25.89% for PG A, 4.94% for PG-LE-A, and 21.66% for Example 29.The number following the corresponding element symbol indicates the detected element line in nm. The depletion (expressed as a percentage) was calculated using the following formulas.

[0229] Depletion in % = 1 - (PG-LE-A / PG A) or

[0230] Depletion in % = 1 - (Example 29 / PG A)

[0231] The abbreviation NWG in Table 8 means limit of detection. The value in parentheses indicates the exact limit of detection.

[0232] It is shown that for a large number of trace elements, an additional purification effect is achieved compared to the purification efficiency of the purified phosphogypsum PG-LE-A produced according to the process described in patent EP 4087817 A1. 230102P10WG

[0233] Table 8: Cleaning efficiency with regard to various trace elements: 230102P10WG

[0234] Table 8 shows the analytical results by ICP-OES for the untreated phosphogypsum PG A, the treated phosphogypsum PG-LE-A, and a phosphogypsum recrystallized according to the inventive process (Example 29). For better comparability, the data refer to the loss-on-ignition phosphogypsum in each case.

[0235] Example 30

[0236] Reduction of specific activity after partial recrystallization and hydroclassification

[0237] The treated phosphogypsum PG-LE-B was recrystallized according to the inventive process in 10% sulfuric acid with an S / L ratio of 1 / 4 with the addition of 25 mg / l cetylpyridinium chloride (based on the sulfuric acid) without seed crystals for approximately 5 days at room temperature. The composition of the calcium sulfate thus obtained was determined by the Rietveld method using XRD (Table 9, 30_before HZ). The particle size distribution found using a Malvern 3000 is given as D vValues ​​were also determined. After filtration, washing, and drying of the calcium sulfate, it was suspended in water (solids concentration in the suspension was approximately 4.5%) and passed through a hydrocyclone from FLSmith (Krebs urethane U0.5) using a centrifugal pump (model). The flow rate was determined to be approximately 140 l / h at a differential pressure of 0.55 bar between the pump discharge and the hydrocyclone inlet. Both the underflow (solids concentration of the underflow was approximately 6.9%) and the overflow (see Table 9, 30_OL_1; solids concentration of the overflow was approximately 1.7%) of the hydrocyclone were collected. The underflow was then resuspended in water (solids concentration in the suspension was approximately 1.9%) and again passed through a hydrocyclone from FLSmith (Krebs urethane U0.5) using a centrifugal pump (model company). This time the flow rate was approximately...A flow rate of 195 l / h was determined at a differential pressure of 1.2–1.25 bar between the pump's pressure side and the hydrocyclone's inlet. Both the underflow (see Table 9, 30_UL_2; underflow solids concentration was approximately 3.0%) and the overflow (see Table 9, 31_OL_2; overflow solids concentration was approximately 0.4%) of the hydrocyclone were collected, filtered, and dried. Together with the overflow from the first pass, they were analyzed for particle size distribution, mineralogy, and specific activity (gamma spectroscopy performed in an external laboratory). For the isotopes U-238, Ra-226, Pb-210, and Ra-228, depletions of > 86%, 28%, 16%, and 26%, respectively, were achieved.

[0238] Table 9: 230102P10WG

[0239] Notes: For the sake of clarity, the remaining mineralogical components have been omitted. nb = not determined

[0240] The depletion is relative to the underflow 31_UL_2 compared to untreated original sample PG B

[0241] Example 31

[0242] Reduction of specific activity after partial recrystallization and classification using a classifier

[0243] The treated phosphogypsum PG-LE-C was recrystallized according to the inventive process in 14% sulfuric acid with an S / L ratio of 1 / 10 with 1% unspecified dihydrate seed crystals for approximately 161 h at 25 °C with stirring (KPG stirrer at 200 rpm). For Example 31, four batches were prepared under nearly identical conditions, differing only in the additional additives used (cetylpyridinium chloride and sodium citrate) (see Table 10). The composition of the calcium sulfate thus obtained was determined by XRD according to the Rietveld method (see Table 11). The particle size distribution found using a Malvern 3000 is given as D vValues ​​were also determined (see Table 11). After filtration, washing, and drying of the four recrystallized batches, these were first carefully deagglomerated by hand using a mortar and pestle, combined, and then passed twice as a single sample through a classifier (Alpine zigzag classifier, model MZR, Augsburg, Germany). The first run used the combined sample of the four batches, and the second used the coarse fraction (31_C_1). The auxiliary air was set to 23 Nm for both runs. 3 The rate was set to / h. For the isotopes U-238, Ra-226, Pb-210, and Ra-228, reductions of 88%, 32%, 36%, and 48% were achieved after the first pass, and reductions of >92%, 46%, 52%, and 59% after the second pass. This example shows that two or more classifiers connected in series can optimize the result with regard to the reduction of radioactive isotopes.

[0244] Table 10: Additives used for example 31 (CPC and Na citrate)

[0245] Table 11: Analysis results for example 31 and untreated phosphogypsum PG C: particle size characteristics (DV), mineralogy and specific activities.

[0246] Note on the terminology:

[0247] 31_1 to 31_4 denote the 4 different approaches.

[0248] The index F and C stand for fine fraction and coarse fraction respectively, and the following number indicates the 1st or 2nd pass.

[0249] The depletion_1 refers to the coarse fraction 31_C_1 to untreated original sample PG C and depletion_2 accordingly for 31_C_2 to untreated original sample PG C.

[0250] Example 32

[0251] Reduction of specific activity after partial recrystallization and classification using a classifier

[0252] The treated phosphogypsum PG-LE-A was recrystallized according to the inventive process in 14% sulfuric acid with an S / L ratio of 1 / 4 with 10% seed crystals (x63;3 = 45.14 M, see also Table 6) for 21.6 h at 25 °C without stirring. The composition of the calcium sulfate thus obtained was determined by the Rietveld method using XRD (see Table 12). The particle size distribution found using a Malvern 3000 is given as D v Values ​​were also determined (see Table 12). After filtration, washing, and drying of the mixture, it was first carefully deagglomerated by hand using a mortar and pestle and then passed through a classifier (zigzag classifier from Alpine, Augsburg, Germany, model MZR). The auxiliary air was set to 23 Nm 3 The rate was set to / h. For the isotopes U-238, Ra-226, and Pb-210, depletion rates of 72%, 54%, and 42%, respectively, were achieved. 230102P10WG

[0253] Table 12: Analysis results for example 33 and untreated phosphogypsum PG A: particle size characteristics (DV), mineralogy and specific activities.

[0254] Reference symbol list

[0255] 1 Aqueous sulfuric acid solution

[0256] 2. Goods

[0257] 3 Mixing tank

[0258] 4 reactor units

[0259] 5a first classification unit

[0260] 5b second classification unit

[0261] 6a first downstream reactor unit

[0262] 6b second downstream reactor unit

[0263] 7a first separation unit

[0264] 7b second separation unit

[0265] 8 dilution units

[0266] 9 Pre-separation unit

[0267] 10 drying units

[0268] 11 Thickening unit

Claims

230102P10WG Patent claims 1. A method for purifying calcium sulfate, the method comprising the following steps: a) Providing a calcium sulfate comprising one or more of the components selected from dihydrate (CaSO4-2H2O), hemihydrate (CaSO4- 1 / H2O) or anhydrite (CaSO4), wherein the anhydrite is present in an amount of at least 50 wt.%, based on the dry calcium sulfate, and has a particle size D v (50) of less than or equal to 25 pm, measured by laser diffraction, b) recrystallization of a suspension containing the calcium sulfate obtained in step a) and an aqueous sulfuric acid solution with a concentration of 2 to 18 wt.%, at a temperature of 0° to 35°C to the calcium sulfate dihydrate form, c) classification of the calcium sulfate from step b) and separation of the fine fraction from the coarse fraction, wherein the D v(50)-value of the fine fraction < 50 pm, and wherein the coarse fraction contains at least 70 wt% dihydrate, wherein the classification step c) is divided into two classification steps, wherein in the first classification step a coarse fraction and a partial fine fraction are first produced and in the second step a partial coarse fraction and a fine fraction are produced from the partial fine fraction, or wherein the classification step c) is divided into two classification steps, wherein in the first classification step a partial coarse fraction and a fine fraction are first produced and in the second step a partial fine fraction and a coarse fraction are produced from the partial coarse fraction.

2. The method of claim 1, wherein the production of the purified calcium sulfate from 230102P10WG Step a) includes the following steps: I) Conversion of phosphate rock with concentrated sulfuric acid to calcium sulfate in the form of dihydrate (CaSO4-2H2O), hemihydrate (CaSO^FW) or a mixture thereof, II) Separation of calcium sulfate as a solid, III) Recrystallization of the dihydrate or hemihydrate to the anhydrite by preparing an aqueous suspension of the calcium sulfate separated in step II) and / or of calcium sulfate in the form of phosphogypsum (CaSO4-2H2O) from the stockpile with sulfuric acid and IV) Separation of the purified calcium sulfate after step III) in the form of the anhydrite (CaSO4) as a solid with a particle size D v (50) of less than or equal to 25 pm, measured by laser diffraction, of the liquid phase of the suspension obtained, wherein the anhydrite is present in an amount of at least 50 wt.%, based on the dry calcium sulfate.

3. Method according to claim 1 or 2, wherein in step a) the calcium sulfate anhydrite has a particle size Dv (50) from 2 to 20 pm.

4. Method according to any of the preceding claims, wherein the concentration of sulfuric acid in step b) is in the range of 8 wt.% to 16 wt.%.

5. A method according to any of the preceding claims, wherein the sulfuric acid is added in step b) in such an amount that the weight ratio of solid to liquid (S / L ratio) in the suspension is between 1 / 10 kg / kg and 1 / 1 kg / kg.

6. Method according to any one of the preceding claims, wherein the particle size D v (50) the dihydrate form in step b) after recrystallization is 25 pm to 150 pm, as measured by laser diffraction. 230102P10WG 7. Method according to one of the preceding claims, wherein in step b) one or more calcium sulfate dihydrate crystals are added to the suspension as seed crystals.

8. The method of claim 7, wherein the particle size D v(63) the seed crystals are in the range of 15 to 100 pm.

9. Method according to any of the preceding claims, wherein the classification of the calcium sulfate in step c) is carried out by one or more of the methods selected from i) one or more hydrocyclones connected in series, ii) one or more classifiers connected in series, iii) dry sieving or iv) wet sieving.

10. Method according to one of the preceding claims, wherein in the first classification step a classification method with a high separation particle size, preferably > 40 pm, is selected, and in the second classification step a classification method with a low separation particle size, preferably < 25 pm, is selected, or that in the first classification step a classification method with a low separation particle size, preferably < 25 pm, is selected, and in the second classification step a classification method with a high separation particle size, preferably > 40 pm, is selected.

11. Method according to one of the preceding claims, wherein between step c) and d) a further recrystallization of the fine fraction and / or coarse fraction to the dihydrate form takes place in a further reactor vessel.

12. Method according to any one of the preceding claims, wherein in step d) the separation of the fine fraction from the coarse fraction is carried out as a partial or complete 230102P10WG Separation of the solid phases from both the fine and coarse fractions obtained after step c) is carried out.

13. Plant for the purification of calcium sulfate, the plant comprising the following facilities: 1) a mixing tank (3) which has at least one feed of feed material (2) and at least one feed of aqueous sulfuric acid solution (1) with a concentration of 2 to 18 wt.%, wherein the feed material (1) comprises calcium sulfate comprising one or more of the components selected from dihydrate (CaSO4-2H2O), hemihydrate (CaSO^F ) or anhydrite (CaSO4 ), wherein the anhydrite is present in an amount of at least 50 wt.%, based on the dry calcium sulfate and has a particle size D v (50) of less than or equal to 25 pm, as measured by laser diffraction, 2) at least one reactor unit (4) which is fed via at least one supply line from the mixing tank (3), wherein at least the reactor unit (4) is designed to accommodate a suspension comprising calcium sulfate and an aqueous sulfuric acid solution (1) with a concentration of 2 to 18 wt.% at a temperature of 0° to 35°C, 3) at least one first and one second classification unit (5a, 5b), each selected from one or more of the group consisting of: i) one or more hydrocyclones connected in series, ii) one or more classifiers connected in series, iii) one or more dry sieves, or iv) one or more wet sieves, wherein these are designed to separate the calcium sulfate into at least one fine and one coarse fraction, wherein the first classification unit (5a) has at least one outlet for the fine fraction and one outlet for the coarse fraction of the first classification unit (5a), and wherein at least one outlet of the first classification unit (5a) is connected to at least one inlet of the second classification unit, and 4) at least one separation unit (7a, 7b, 9) designed to separate the solid and liquid phases of the fine and coarse fractions.

14. Plant according to claim 13, wherein this comprises at least one further reactor unit designed to age the fine and / or coarse fractions obtained from at least one of the classification units (5a, 5b), forming dihydrate.

15. System according to claim 13 or 14, wherein this 5) at least one fluidic connection designed for the return of the sulfuric acid solution (1) from the reactor unit (4) to the mixing tank (3) and 6) comprising at least one fluidic connection designed for the return of the sulfuric acid solution (1) from the separation unit (7a, 7b) to the mixing tank (3).

Images