Journal Information
Vol. 8. Issue 3.
Pages 2481-3388 (May - June 2019)
Download PDF
More article options
Vol. 8. Issue 3.
Pages 2481-3388 (May - June 2019)
Original Article
DOI: 10.1016/j.jmrt.2019.02.013
Open Access
Comparative study of K2SO4 production by wet conversion from phosphogypsum and synthetic gypsum
Yassine Ennaciri
Corresponding author

Corresponding author.
, Hanan El Alaoui-Belghiti, Mohammed Bettach
Laboratory of Physical Chemistry of Materials (LPCM), Faculty of Sciences, Chouaib Doukkali University, El Jadida, Morocco
This item has received

Under a Creative Commons license
Article information
Full Text
Download PDF
Figures (10)
Show moreShow less
Tables (4)
Table 1. Operating conditions for the experiments carried out in this work.
Table 2. Infrared frequencies of the PG, calcite P10 and arcanite F10.
Table 3. Distribution of chemical elements between PG and reaction products determined by X-ray fluorescence.
Table 4. Distribution of some trace elements in PG, calcite P10 and arcanite F10.
Show moreShow less

This paper proposes a very attractive process for K2SO4 production via the wet conversion of phosphogypsum (PG) and K2CO3. A parallel study of the conversion of two synthetic types of gypsum by K2SO4 was performed in order to compare their reactivity with that of PG.

We remark that PG is more reactive and yields the desired results while commercial types of gypsum react slowly and generate unwanted secondary products. The factors affecting the PG conversion processes such as the initial concentration of reagents, the reaction time and the reaction temperature was studied to optimize the reaction conditions. X-ray diffraction (XRD) was used to characterize the PG conversion at different conditions while the other techniques (Fourier transform infrared spectroscopy (FTIR), X-ray fluorescence (XRF), scanning electron microscopy (SEM), flame photometer (FP), inductively coupled plasma mass spectrometry (ICP-MS) and thermogravimetric analysis (TGA)) were applied to prove the quality of the final products.

Based on the obtained results, the reaction is carried out with stoichiometric proportions between PG and K2CO3. The maximal conversion of PG is attained at 80°C; this temperature corresponds to the higher solubility of K2SO4, which is 1.2mol/l.

Wet conversion
Potassium sulfate
Synthetic gypsum
Full Text

Potassium sulfate K2SO4 (arcanite) is a type of simple potash fertilizer used especially for high quality cultivation. It allowed bringing to plants two essential nutrients: potassium (up to 54% K2O) and sulfur (18.4% S) [1]. The presence of high chloride content in standard fertilizers, especially potassium chloride, can increase the salinization of soils and accumulate the cadmium in plants [2,3]. For these reasons the use of K2SO4 as a fertilizer is preferred for several advantages such as a good solubility (120g/l at 25°C), low-to-zero chloride content and low salt index. It also costs half of that of the potassium nitrate [1]. In addition, the agriculture experts often recommend supplemental sulfur to go with nitrogen, phosphorus and potassium (NPK). It is a substance that contributes to increase crop yield by supplying a direct nutritional value and by improving the efficiency with which plants use other nutriments elements particularly the nitrogen and the phosphorus [4–6].

The world capacity production of K2SO4 is 11 million tons per year of which China produces about 56%. Among the natural sources of K2SO4 we quote: Kainite KCl·MgSO4·3H2O, Langbeinite K2SO4·2MgSO4, Leonite K2SO4·MgSO4·4H2O, Schonite K2SO4·MgSO4·6H2O, Polyhalite K2SO4·MgSO4·2CaSO4·2H2O and Glaserite K3Na(SO4)2. For not to exhaust the magnesium contained in this different ores, several industrial syntheses have been developed for the production of K2SO4. Approximately a large part of K2SO4 production is based on the Mannheim process according to the reaction of H2SO4 acid and KCl at high temperature (800°C) [7]. In the other hand, the reaction of H2SO4 acid and KCl at low temperature allows forming KHSO4. This latter was desalted with methanol to obtain K2SO4[8,9].

Other processes have been described by numerous researchers to produce K2SO4 by thermal decomposition of K-feldspar KAlSi3O8 and gypsum [10–12]. Several factors affect on this decomposition such as: the reaction temperature (1000–1400°C), the mass ratio of CaSO4/KAlSi3O8 and the amount of the additive used (CaCO3 or CO2 gas). The final product is a water-soluble solid K2Ca2(SO4)3, which could dissolve in water to produce K2SO4.

It is also possible to produce K2SO4 from gypsum and KCl in an aqueous solution containing ammonia. The presence of ammonia induces the reaction between the gypsum and KCl, moves the reaction in the desired direction (selectivity) and prevents the formation of complex salts such as syngenite K2Ca(SO4)2·H2O or/and the pentasulfate K2SO4·5CaSO4·H2O. The yield of this reaction depends essentially on the nature of gypsum used, KCl purity, ammonia concentration and molar ratios of these reagents [13,14].

Generally, several disadvantages intercept the production of K2SO4 by these previous processes such as: high energy consumption, high chloride content in K2SO4 salt and the difficulty of separating this salt from the other salts, which requires the addition of other solvents.

In Morocco, the phosphoric acid production is very important for the economic and social area of the country. In fact, this production is mainly realized by wet process attacking the phosphate rock by sulfuric acid according the following chemical reaction:


The nature and the characteristics of the obtained PG depend strongly on the phosphate ore composition and quality. Although PG is mainly constituted of the CaSO4·2H2O dihydrate form, it contains also various impurities such as fluorides, phosphates, clay minerals, organic matter, trace elements and radioelements. Hence, the presence of these impurities at higher than natural levels, would be considered as a source of environmental contamination [15].

The objective of this work is to present a simple process permitting to produce K2SO4 via the wet conversion of phosphogypsum (PG) waste generated during the manufacture of phosphoric acid, the Prolabo gypsum (PL) and Riedel the Haën gypsum (RH). The aim to use the two latter commercial types of gypsum is to compare their reactivity with that of PG and to study the factors that affect on this conversion. The basic reaction describing this process is:


The important factors that influence on the PG conversion studied are: the initial concentration of reagents, the reaction time and the reaction temperature. Thus, the strengths of this process are as follows: the conversion is carried out with easier conditions; the optimal temperature of reaction is 80°C corresponding to maximum solubility of K2SO4; the medium is purely aqueous without additives; the availability of PG; the main product K2SO4 is easily recovered by simple filtration while the calcite is interesting in several domains of industry and environment.


The PG sample was collected from the fertilizer plant Maroc Phosphore (dihydrate processes) settled at Jorf Lasfar near El Jadida city. The PG has undergone a simple wash to remove some of the soluble impurities and suspension materials as organic matter. The synthetic types of gypsum used are chemical reagents type prolabo and Riedel de Haën with purity 98 and 100% respectively. The potassium carbonate (99%, Riedel de Haën) was also utilized as reactant.

To compare the reactivity of PG and commercial types of gypsum, the reactional mixtures were prepared from the washed PG, gypsum PL and gypsum RH dissolved in K2CO3 solution at different temperature under constant stirring (500tr/min). After a sufficient reaction time, the obtained precipitate was separated from the solution and dried in the oven at 100°C. The transparent salt is recrystallized from the filtrate and introduced in the oven at 40°C (Fig. 1 and Table 1).

Fig. 1.

Different steps of the PG, PL and RH conversion process.

Table 1.

Operating conditions for the experiments carried out in this work.

RunNo.  T(°C)  Type of gypsum  Gypsum(g)  [Gypsum](mol/l)  K2CO3(g)  [K2CO3](mol/l)  Time(h)  PrecipitateNo.  FiltrateNo. 
01  20  Phosphogypsum  17.2  0.1  13.82  0.1  1.5  P1  F1 
02  20  Phosphogypsum  103.3  0.6  82.92  0.6  1.5  P2  F2 
03  20  Phosphogypsum  120.5  0.7  96.74  0.7  1.5  P3  F3 
04  20  Phosphogypsum  17.2  0.1  27.64  0.2  1.5  P4  F4 
05  20  Phosphogypsum  34.4  0.2  13.82  0.1  1.5  P5  F5 
06  20  Phosphogypsum  103.3  0.6  82.92  0.6  P6  F6 
07  20  Phosphogypsum  103.3  0.6  82.92  0.6  24  P7  F7 
08  40  Phosphogypsum  137.7  0.8  110.56  0.8  1.5  P8  F8 
09  60  Phosphogypsum  172.2  138.2  1.5  P9  F9 
10  80  Phosphogypsum  206.6  1.2  165.84  1.2  1.5  P10  F10 
11  20  Gypsum Prolabo  17.2  0.1  13.82  0.1  1.5  P11  F11 
12  20  Gypsum Prolabo  17.2  0.1  13.82  0.1  P12  F12 
13  20  Gypsum Prolabo  103.3  0.6  82.92  0.6  1.5  P13  F13 
14  20  Gypsum Prolabo  103.3  0.6  82.92  0.6  P14  F14 
15  20  Gypsum Riedel de Haën  17.2  0.1  13.82  0.1  P15  F15 
16  20  Gypsum Riedel de Haën  17.2  0.1  13.82  0.1  1.5  P16  F16 
17  20  Gypsum Riedel de Haën  17.2  0.1  13.82  0.1  P17  F17 
18  20  Gypsum Riedel de Haën  103.3  0.6  82.92  0.6  P18  F18 
19  20  Gypsum Riedel de Haën  103.3  0.6  82.92  0.6  P19  F19 

All experiments are carried at atmospheric pressure for total volume of distilled water equal to 1l under constant stirring (500tr/min).

The different analysis carried out on the PG and the compounds produced in this work are accomplished by the following techniques: X-ray powder diffraction (XRD diffractometer BRUKER D8), scanning electron microscopy X-ray analysis (SEM Environmental FEI Quanta 200). Concentrations of chemical elements were determined by X-ray fluorescence (XRF spectrometer S4 PIONEER BRUKER aXS), flame photometer (FP JENWAY 500-731 Model PFP7) and inductively coupled plasma mass spectrometry (ICP-MS Model HP-4500). Besides that, the sulfate ions in different samples can be measured by the gravimetric method. Infrared spectra were performed by Fourier transform infrared spectroscopy (FTIR 8400s SHIMADZU spectrometer) using potassium bromide (KBr) pellets technique. The thermal behaviors of our samples were examined by thermogravimetric analysis (TGA) (DTG-60 type SHIMADZU).

3Results and discussion3.1Characterization of phosphogypsum

The chemical and mineralogical compositions of the PG show that it is composed mainly of 98% of calcium sulfate dihydrate (JCPDS No. 33-0311). The major impurities are SiO2, P2O5 besides Na2O, K2O, Al2O3, Fe2O3 and MgO with lower contents. PG also contains trace chemical elements such as Ba, Cd, Pb, etc. These latter are absorbed in the PG surface or incorporated in its structure [16].

3.2Solubility of potassium sulfate in water

The potassium sulfate solubility in water as a function of temperature are representing in Fig. 2[17]. It is noted that the solubility (S) of potassium sulfate in water increases gradually when the temperature increases. According to the solubility data of potassium sulfate, the predicted concentrations (C=S/M(K2SO4)) of the starting reagents K2CO3 and PG to obtain the maximum solubility of K2SO4 and to avoid the formation of syngenite are 0.6, 0.8, 1 and 1.2mol/l for room temperature 20, 40, 60 and 80°C respectively. We have limited the maximum temperature of PG conversion at 80°C to prevent the transformation of gypsum into plaster, which becomes less reactive.

Fig. 2.

Solubility of potassium sulfate in water and its predicted concentrations as a function of temperature [16].

3.3Effect of initial concentration

The effect of initial concentration of K2CO3 on the conversion of PG to K2SO4 during 1.5h at room temperature was performed by varying its concentration from 0.1 to 0.7mol/l. The mixture of K2CO3 with PG is stoichiometric. The results obtained by DRX analysis reveal that for stoichiometric K2CO3 concentration of 0.1 and 0.6mol/l, the precipitates P1 and P2 (Fig. 3a) and salt crystallized from the filtrates F1 and F2 (Fig. 3b) present a single phase corresponding to the calcite (JCPDS No. 05-0586) for the precipitates and the arcanite (JCPDS No. 44-1414) for the filtrates [18]. However the increase of the K2CO3 concentration above 0.6mol/l allows appearing the syngenite K2Ca(SO4)2·H2O (JCPDS No. 74-2159) besides the obtained calcite P3 (Fig. 3a). Syngenite is a relatively rare double salt of potassium and calcium sulfate, which is deposited by evaporation in addition to normal inorganic salts such as gypsum, anhydrite and halite [19,20].

Fig. 3.

DRX patterns of precipitates (a) and filtrates (b) obtained from PG conversion for different stoichiometric K2CO3 concentration at room temperature for 1.5h.


According to Abu-Eishah et al. [13], the formation of syngenite beside calcite in this case may be explained by two reactions:

2CaSO4·2H2O (s)+K2CO3 (aq)K2Ca(SO4)2·H2O (s)+CaCO3 (s)+3H2O
CaSO4·2H2O (s)+K2SO4 (aq)K2Ca(SO4)2·H2O (s)+H2O

From the precedent results, reaction (3) does not explain the absence of syngenite (Kps=3.54×10−8 at 25°C) in the precipitates relative to concentrations lower than 0.6mol/l.

When the concentration of K2SO4 is very high (oversaturation), this salt reacts with PG to form syngenite. Indeed, the filtrate F3 of the K2CO3 concentration equal to 0.7mol/l (Fig. 3b) reveals that the salt is composed of K2SO4 and K2CO3 (JCPDS No. 87-0730). In this case, there is a consumption of more PG than K2CO3 taking into account the stoichiometry of this reaction. From this result we conclude that reaction (4) is most probable.

The effect of K2CO3 excess on PG conversion to K2SO4 was also examined. Fig. 3a displays that the precipitate P4 obtained for the mixture 0.1mol/l of PG and 0.2mol/l of K2CO3 corresponds to the calcite while the relative filtrate F4 (Fig. 3b) contains a mixture of K2SO4 and K2CO3 salts due to the excess of K2CO3. However, when we inverse the concentrations of PG and K2CO3, the decrease of K2CO3 concentration below its stoichiometric value shows that the PG conversion is not yet achieved and we obtained a precipitate P5 containing gypsum and calcite. For the filtrate F5, its salt is composed only of K2SO4.

We conclude that the conversion of PG to K2SO4 with excess of K2CO3 or PG concentration is possible but not gainful.

3.4Effect of reaction time

The effect of reaction time on the PG conversion to K2SO4 at room temperature was tested by prolonging it to 24h for stoichiometric K2CO3 concentration equal to 0.6mol/l.

The results showed well no significant differences in yield and purity of the final precipitates P2, P6 and P7 (Fig. 4a) and filtrates F2, F6 and F7 (Fig. 4b) when we have increased the reaction time up to 24h, which are always calcite and arcanite respectively. Goerlich [21] predicted the formation of syngenite for a longer reaction time, which is not the case of our reaction. This difference can be interpreted by the quality of the PG used. However, Fernández-Lozano and Wint [14] indicated that the effect of this variable was on the crystal size. According to these authors, crystals produced at shorter reaction time were small and difficult to filter.

Fig. 4.

DRX patterns of precipitates (a) and filtrates (b) obtained from PG conversion for stoichiometric K2CO3 concentration 0.6mol/l at room temperature for various reaction times.

3.5Effect of temperature

To increase the limit concentration of K2SO4 in the solution at the end of the reaction, we performed tests at higher temperatures 40, 60 and 80°C. The effect of temperature on the PG conversion was studied for stoichiometric concentrations equal to the concentrations limit foreseeable. The obtained results for the concentrations of mixtures equal to the concentrations limit 0.8, 1 and 1.2mol/l at 40, 60 and 80°C respectively, reveal that the increase of temperature permits obtaining a pure calcite from the precipitate P8, P9 and P10 (Fig. 5a) and concentrated salt in the filtrate F8, F9 and F10 (Fig. 5b). The absence of syngenite in the precipitate for these concentrations limit is explained by the increase of K2SO4 solubility as a function of temperature, which becomes maximal around 80°C [22]. We conclude that increasing the temperature of PG conversion reaction gives concentrated solution of K2SO4, which allows its rapid recrystallization after cooling. The purity of these products obtained under the optimum conditions at 20, 40, 60 and 80°C are represented in Table 3. The obtained results show that all precipitates contain calcite with purity of 95%. Also, the purity of the final salts recrystallized from filtrates is 97% of arcanite.

Fig. 5.

DRX patterns of precipitates (a) and filtrates (b) obtained from PG conversion for stoichiometric K2CO3 concentrations equal to the concentration limit.

3.6Effect of the gypsum purity

In this part, two types of gypsum PL and RH are used to make a comparison between their reactivity with that of PG generated during the manufacture of phosphoric acid. The conversion of the gypsum PL and the gypsum RH to K2SO4 is performed with stoichiometric mixture between these synthetics types of gypsum and K2CO3. The reaction is carried out for concentrations 0.1 and 0.6mol/l at room temperature for various reaction times.

3.6.1Case of PL gypsum

The DRX patterns of the precipitates obtained by PL conversion are presented in Fig. 6a. All precipitates P11, P12, P13 and P14 contain two phases: calcite and plaster (effectively the gypsum) (JCPDS No. 33-0310). The increase of the reaction time to 4h indicates that the gypsum did not totally disappear.

Fig. 6.

DRX patterns of precipitates (a) and filtrates (b) obtained from PL conversion for stoichiometric K2CO3 concentration equal to 0.1 and 0.6mol/l at room temperature for various reaction times.


Furthermore, the filtrates F11, F12, F13 and F14 (Fig. 6b) relatives to the same concentrations contain mixture of two phases K2SO4 and K2CO3.

3.6.2Case of RH gypsum

Fig. 7a represents the XRD patterns of the precipitates obtained by RH conversion to K2SO4. For K2CO3 concentration equal to 0.1mol/l during a reaction time of 1h, the formed precipitate P15 is composed of small amount of gypsum beside two varieties of CaCO3 calcite and vaterite (JCDPS No. 24-0030). The increase of the reaction time allows transforming the vaterite into calcite P16 and P17 with reappearance evidently of the gypsum. Generally, the vaterite crystallizes practically only under very particular conditions (neutral medium, pure water, pure reagents). At ambient temperature and in an aqueous medium, the vaterite is very unstable and normally recrystallizes in the form of calcite as a function of time [23].

Fig. 7.

DRX patterns of precipitates (a) and filtrates (b) obtained from RH conversion for stoichiometric K2CO3 concentration equal to 0.1 and 0.6mol/l at room temperature for various reaction times.


For K2CO3 concentration equal to 0.6mol/l during the reaction time of 1.5h, the obtained precipitates P18 (Fig. 7a) indicate that the major phase corresponds to vaterite with a very minor phase corresponding to syngenite. When we increase the reaction time to 4h, we observe the presence of vaterite and syngenite with a small amount of calcite P19 (Fig. 7a). In fact, for very pure gypsum and for this concentration, the solution will be saturated with K2SO4. This latter reacts with gypsum to form the syngenite. The same result is attained with PG for the concentrations up to 0.6mol/l for the reason that PG contains less CaSO4·2H2O than RH gypsum.

The presence of gypsum or syngenite in the precipitates recuperated during RH conversion engenders the existence of K2CO3 and K2SO4 in the filtrates F15, F16, F17, F18 and F19 as shown in Fig. 7b.

From these results, we conclude that the conversion PL gypsum and RH gypsum to K2SO4 is not total, which confirms that these types of gypsum are less reactive than PG. This result can be explained by the difference of the purity, the nature of the impurities and the granulometry, which favor the dissolution of PG.

3.7Physicals and chemicals characteristics of final products

In this part, the precipitate P10 and the filtrate F10 recovered after total PG conversion to K2SO4 at 80°C was chosen for all further tests in order to determine the physicals and the chemicals characteristics and the quality of these final products.

The infrared spectrum of the precipitate (Fig. 8b) shows all characteristic bands corresponding to calcite, except for a band observed at 1083cm−1, which is assigned to Si–O stretching mode [16,24,25]. For the filtrate salt, the infrared spectrum is displayed in Fig. 8c. It illustrates the characteristic bands of pure arcanite [26]. The IR spectra are in good agreement with DRX analysis accomplished on these products.

Fig. 8.

IR spectra of the PG (a), calcite P10 (b) and arcanite F10 (c).


Table 2 lists the vibration band frequencies and assignments of the PG as well as the obtained calcite and arcanite.

Table 2.

Infrared frequencies of the PG, calcite P10 and arcanite F10.

Assignment    Wave number ν (cm−1)
    Phosphogypsum  Calcite P10  Arcanite F10 
H2Oν  3600–3400  3700–3400  3600–3300 
δ  1687 ; 1624  1635  1640 
SO42−ν3  1132  –  1194; 1109 
ν1  1004  –  991 
ν4  669; 603  –  – 
δ  –  –  620 
HSO42−  δ2  –  –  1384 
CO32−ν3  –  1418  – 
ν2  –  873  – 
ν4  –  709  – 
HPO42−  ν  837  –  – 
Si–O  ν  –  1083  – 

The results of chemical analysis by XRF realized on the obtained calcite and arcanite are represented in Table 3.

Table 3.

Distribution of chemical elements between PG and reaction products determined by X-ray fluorescence.

Major elements (%)  Phosphogypsum  CalciteArcanite
    P10  P9  P8  P2  F10  F9  F8  F2 
CaO  33.83  53.59  53.86a  53.83a  53.81a  0.28  –  –  – 
SO3  42.52  0.55  –  –  –  43.49  44.05b  44.04b  44.09b 
H221.80c  –  –  –  –  –  –  –  – 
CO2  –  41.70c  41.68c  41.69c  41.72c  –  –  –  – 
K20.02  1.35  –  –  –  53.84  53.59a  53.57a  53.57a 
Na20.12  0.01  –  –  –  1.75  –  –  – 
P2O5  0.67  0.96  –  –  –  0.04  –  –  – 
SiO2  0.49  0.71  –  –  –  0.08  –  –  – 
Al2O3  0.21  0.19  –  –  –  –  –  –  – 
MgO  0.05  0.18  –  –  –  –  –  –  – 
Fe2O3  0.02  0.03  –  –  –  –  –  –  – 
SrO  0.08  0.11  –  –  –  –  –  –  – 
Y2O3  0.02  0.03  –  –  –  –  –  –  – 
Cl  0.02  0.02  –  –  –  0.29  –  –  – 
TiO2  0.01  0.02  –  –  –  –  –  –  – 
0.20d  0.32d  –  –  –  –  –  –  – 
Purity (%)  98.15  95.29  95.54  95.52  95.53  97.33  97.64  97.61  97.66 

Flame photometer.


Gravimetric methods.


Thermal analysis.


Ionometric method.

Calcium carbonate represents around 95.5% of the precipitate composition. In fact the insoluble impurities incorporated in PG pass totally into calcite. Content of potassium in the precipitate is very high compared to that in PG. This is explained by the presence of K2SO4, which is adsorbed in surface of calcite crystals during the filtration. The silica, which is inert during the PG conversion, is detected in the precipitate. It is important to note that the detection of Al, Mg and Fe in the precipitate suggests the presence of clays. These elements can be combined with phosphorus to form insoluble compounds such as AlPO4, Mg3(PO4)2 and FePO4·2H2O. Some elements like as Mg, Sr and Y can react with the carbonate anions during the PG conversion and form insoluble carbonates (MgCO3, SrCO3, Y2(CO3)3, etc.). Fluorine exists in PG in soluble and insoluble form. The insoluble complexes (Na2SiF6, Na3AlF6, K3AlF6, MgSiF6·6H2O, etc.) remain in the powder. Then, the fluorine in soluble form (HF, H2SiF6) passes into the filtrate.

The analysis of the recrystallized salt proves that it is constituted by 97% of K2SO4. The presence of impurities such as Na and Ca is originally reported to the K2CO3 reagent (0.5% of Na2O in K2CO3 reagent) and to the very fine particles of CaCO3 that pass in the filtrate, respectively. The presence of chloride in the salt may be due to chloride existing in initial reagent K2CO3.

According to the results exposed in Table 4, we can conclude that the totality of trace elements is transferred from PG to the resulting calcite. Generally, these elements are adsorbed in the surface of calcite crystals or incorporated into its structure. Several factors favor their presence in the calcite such as: (i) the similar chemical behavior of calcium and these elements; (ii) the presence of these elements in PG under insoluble forms such as (Ba,Ra)SO4, CdSO4·2H2O, CdHPO4, PbSO4·2H2O, ZnSO4·2H2O and CuSO4·2H2O [27,28] that rest insoluble during the conversion of PG; iii) the similarity of the particle size fraction (less than 20μm) of PG and calcite, which facilitates the presence of these elements in the precipitate.

Table 4.

Distribution of some trace elements in PG, calcite P10 and arcanite F10.

Trace elements (ppm)  Phosphogypsum  Calcite P10  Arcanite F10 
Baa  30.08  49.52  0.00 
Cd  1.34  2.23  0.02 
Hg  0.65  0.57  0.00 
Pb  0.73  3.09  0.00 
Cub  55  68  N.d. 
Znb  43  71  N.d. 
Zrb  19  29  N.d. 

Flame photometer.


X-ray fluorescence.

N.d., not detected.

For radioactivity, the element that can cause toxicity problems by decomposing into radon and lead is radium. The various studies carried out on the radioactivity of PG confirm that radium always precipitates with calcium in the solid phase (PG, calcite and portlandite) [29–31]. To avoid this problem, this phase can be simply stored in an open and ventilated environment to dissipate radon into the atmosphere [32]. According to these results, we can suppose that the majority of the radionuclides initially present in PG sample pass to calcite. However, potassium sulfates rest uncontaminated from these radionuclides.

K2SO4 recovered from the filtrate is non-toxic and can be utilized directly in the agriculture as fertilizer [33].

The TGA curves of PG, calcite and arcanite are represented in Fig. 9. The TGA curve of PG (Fig. 9a) exhibits a weight loss (21.8%) at 200°C, which correspond to gypsum dehydration.

Fig. 9.

TGA curves of PG (a), calcite P10 (b) and arcanite F10 (c).


The second TGA curve of calcite (Fig. 9b) presents two weight losses: the first at 450°C is small (∼3%) and it is attributed to the elimination of adsorbed or crystallized water of some chemicals impurities (CaHPO4·2H2O, CaFPO3·2H2O; MgSiF6·6H2O, etc.). The second above 800°C, which is large (41.7%). This is due to the decarbonation of calcite (CaCO3CaO+CO2). The minimization of decomposition temperature of the calcite is the goal of the industrial process to use it in cycle CaCO3/CaO. According to our results, the obtained calcite (CaCO3) can be converted to lime (CaO) at temperature below 800°C while the temperature decomposition of pure calcite is above 900°C. For this reason, we can conclude that the use of the obtained calcite consumes less energy. Thus this reaction is energetically efficient.

In summary, the liberated CO2 may be recuperated for the production of potassium carbonate or used in the Merseberg process for the production of the ammonium carbonate.

For the salt, TGA curve (Fig. 9c) indicates a slight weight loss (0.8%) observed between the temperature 100 and 800°C, which means that the salt obtained is anhydrous. Anooz et al. [34] have detected a weight loss between 0.5 and 0.8% for K2SO4 prepared by the evaporation method.

The morphology of PG, calcite and arcanite are examined by SEM (Fig. 10). We remark that the crystals form of PG is tabular with a size grading from 5 to 30μm (Fig. 10a), whereas the calcite morphology (Fig. 10b) displays an agglomerate of super imposed fine scales with a low surface area. The average diameter of scales is 2μm and thickness is lower than 1μm [16]. Generally, the energy efficiency of synthetic calcite and its low surface area are two favorable parameters for its use in the environment (CaCO3/CaO cycle) as adsorbent of carbon dioxide [35,36].

Fig. 10.

SEM images of PG (a), calcite P10 (b) and arcanite F10 (c).


For the salt, the photograph (Fig. 10c) indicates the presence of macro-crystals in the form of rods of length between 3 and 5mm. This permits to recuperate K2SO4 easily after recrystallization.


During this work, we established a simple and economical process for the production of K2SO4 from the conversion of PG with K2CO3 in an aqueous medium. Compared to the conversion of synthetic types of gypsum or the use of other processes, this proposed process can be applied industrially on a large scale.

The PG is totally transformed to K2SO4 and calcite while the reaction between PL and RH types of gypsum and K2CO3 is not total even if we increase the time reaction and engender secondary products as syngenite and vaterite.

The reaction process of PG conversion to K2SO4 is optimum at 80°C with exact stoichiometric proportion of PG and K2CO3 for concentration limit of 1.2mol/l under a time reaction of 1.5h. Concentrated solution obtained at 80°C allows rapid recrystallization of K2SO4 after cooling.

Under the optimal conditions of PG conversion to K2SO4, the different chemical and physical analyses accomplished on the final products show that the totally of solid impurities in PG are transferred into calcite with the recovery of pure K2SO4.

The environmental interest of this process is reducing the quantity of PG rejected and obtains a calcite as resource. Also, significant amounts of this obtained calcite could be used by the phosphate industry to remedy of acid mine drainage [30,31]. For a numerical and economical approach, about 1ton of PG can be treated by 0.803ton of K2CO3 (595$ per ton [37]) to obtain 0.581ton of calcite (140$ per ton [38]) and 1.012ton of K2SO4 (700$ per ton [38]). In addition, this process does not require any difficult operation for recuperating these valuables products. This economical approach permits to benefit of 312$ per ton of PG converted, which covers very largely the cost of the transport, the energy and the water used during this process.

The purity of K2SO4 tolerates it to be utilized directly in agriculture. While the thermal and morphological characteristic of the obtained calcite allows using it in the environment such as adsorbent for carbon dioxide.

Conflicts of interest

The authors declare no conflicts of interest.

D. Ciceri, M. de Oliveira, A. Allanore.
Potassium fertilizer via hydrothermal alteration of K-feldspar ore.
Green Chem, 19 (2017), pp. 5187-5202
N. Ishikawa, G. Ishioka, M. Yanaka, K. Takata, M. Murakami.
Effects of ammonium chloride fertilizer and its application stage on cadmium concentrations in wheat (Triticum aestivum L.) grain.
Plant Prod Sci, 18 (2015), pp. 137-145
M. Zahedifar, A.A. Moosavi, M. Shafigh, Z. Zarei, F. Karimian.
Cadmium accumulation and partitioning in Ocimum basilicum as influenced by the application of various potassium fertilizers.
Arch Agron Soil Sci, 62 (2016), pp. 663-673
J. Camberato, S. Casteel.
Sulfur deficiency.
Agronomy Department, Purdue University, (2017),
Q. Ma, R. Bell, C. Scanlan, G. Sarre, R. Brennan.
Drought and soil salinity influence response of cereals to potassium and sulfur fertilization.
Better Crops Plant Food, 100 (2016), pp. 20-22
K. Watanabe, Y. Fukuzawa, S.I. Kawasaki, M. Ueno, Y. Kawamitsu.
Effects of potassium chloride and potassium sulfate on sucrose concentration in sugarcane juice under pot conditions.
Sugar Tech, 18 (2016), pp. 258-265
Société Chimique de France.
Sulfate de potassium.
A. Mientka, B. Grzmil, M. Tomaszewska.
Production of potassium sulfate from potassium hydrosulfate solutions using alcohols.
Chem Pap, 62 (2008), pp. 123-126
M. Tomaszewska, A. Łapin.
The influence of feed temperature and composition on the conversion of KCl into KHSO4 in a membrane reactor combined with direct contact membrane distillation.
Sep Purif Technol, 100 (2012), pp. 59-65
L. Lü, C. Li, G. Zhang, X. Hu, B. Liang.
Decomposition behavior of CaSO4 during potassium extraction from a potash feldspar-CaSO4 binary system by calcination.
Chin J Chem Eng, 26 (2017), pp. 838-844
C. Wang, H. Yue, C. Li, B. Liang, J. Zhu, H. Xie.
Mineralization of CO2 using natural K-feldspar and industrial solid waste to produce soluble potassium.
Ind Eng Chem Res, 53 (2014), pp. 7971-7978
Y. Zhong, J. Gao, P. Chen, Z. Guo.
Recovery of potassium from K-feldspar by thermal decomposition with flue gas desulfurization gypsum and CaCO3: analysis of mechanism and kinetics.
Energy Fuels, 31 (2016), pp. 699-707
S.I. Abu-Eishah, A.A. Bani-Kananeh, M.A. Allawzi.
K2SO4 production via the double decomposition reaction of KCl and phosphogypsum.
Chem Eng J, 76 (2000), pp. 197-207
J.A. Fernández-Lozano, A. Wint.
Production of glaserite and potassium sulphate from gypsum and sylvinite catalysed by ammonia.
H. Tayibi, M. Choura, F.A. López, F.J. Alguacil, A. López-Delgado.
Environmental impact and management of phosphogypsum.
J Environ Manag, 90 (2009), pp. 2377-2386
Y. Ennaciri, M. Bettach.
Procedure to convert phosphogypsum waste into valuable products.
Mater Manuf Process, 33 (2018), pp. 1727-1733
Perry's chemical engineer's handbook.
6th ed., (1985),
Y. Ennaciri, F.E. Mouahid, A. Bendriss, M. Bettach.
Conversion of phosphogypsum to potassium sulfate and calcium carbonate in aqueous solution.
MATEC Web Conf, 5 (2013), pp. 04006
J. Fijał, I. Stańczyk, M.Ż.B.W. Inowrocławskim, W. Solnym.
Polska akademia nauk-oddział w krakowie.
Prace Mineral, 15 (1970), pp. 22-27
H. Borchert, R.O. Muir.
Salt deposits: the origin, metamorphism and deformation of evaporites.
Van Nostrand, (1964),
P. Goerlich.
A new process from Voest-Alpine AG for the manufacture of potassium sulfate from potassium chloride and gypsum. A short communication.
University of Linz, (1984), pp. 9
F. Risacher, B. Fritz.
Estimation des variations en fonction de la température des produits de solubilité des principaux sels des milieux évaporitiques.
Sci Géol Bull Mém, 37 (1984), pp. 229-237
S.T. Silk.
Factors governing polymorph formation in calcium carbonate.
New York University, Graduate School of Arts and Science, (1970),
Y. Ennaciri, M. Bettach, A. Cherrat, A. Zegzouti.
Conversion of phosphogypsum to sodium sulfate and calcium carbonate in aqueous solution.
J Mater Environ Sci, 7 (2016), pp. 1925-1933
B. Kanoje, J. Parikh, K. Kuperkar.
Crystallization study and morphology behaviour of calcium carbonate crystals in aqueous Surfactant-Pluronics® prototype.
A. Periasamy, S. Muruganand, M. Palaniswamy.
Vibrational studies of Na2SO4, K2SO4, NaHSO4 and KHSO4 crystals.
Rasayan J Chem, 2 (2009), pp. 981-989
F. Grandia, J. Merino, J. Bruno.
Assessment of the radium-barium co-precipitation and its potential influence on the solubility of Ra in the near-field (No. SKB-TR-08-07).
Swedish Nuclear Fuel and Waste Management Co., (2008),
D. Chafik, R. Bchitou, A. Bouhaouss.
Modélisation et Optimisation du Procédé de la fabrication de l’acide phosphorique en présence des ions métalliques Cu, Zn, et Cd.
Phosphorus Sulfur Silicon Relat Elem, 189 (2014), pp. 353-360
M. Contreras, R. Pérez-López, M.J. Gázquez, V. Morales-Flórez, A. Santos, L. Esquivias, et al.
Fractionation and fluxes of metals and radionuclides during the recycling process of phosphogypsum wastes applied to mineral CO2 sequestration.
Waste Manag, 45 (2015), pp. 412-419
J. Mulopo, D. Ikhu-Omoregbe.
Phosphogypsum conversion to calcium carbonate and utilization for remediation of acid mine drainage.
J Chem Eng Process Technol, 3 (2012), pp. 2-6
W.C. Burnett, M.K. Schultz, C.D. Hull.
Radionuclide flow during the conversion of phosphogypsum to ammonium sulfate.
J Environ Radioact, 32 (1996), pp. 33-51
M. Azouazi, Y. Ouahidi, S. Fakhi, Y. Andres, J.C. Abbe, M. Benmansour.
Natural radioactivity in phosphates phosphogypsum and natural waters in Morocco.
J Environ Radioact, 54 (2001), pp. 231-242
L. Moughli.
Les engrais minéraux caractérisation et utilisation. Bulletin mensuel de l’information et de liaison du PNTTA N° 72 septembre.
S.B. Anooz, R. Bertram, D. Klimm.
The solid state phase transformation of potassium sulfate.
Solid State Commun, 141 (2007), pp. 497-501
B. Cheng, M. Lei, J. Yu, X. Zhao.
Preparation of monodispersed cubic calcium carbonate particles via precipitation reaction.
Mater Lett, 58 (2004), pp. 1565-1570
Z. Zhao, L. Zhang, H. Dai, Y. Du, X. Meng, R. Zhang, et al.
Surfactant-assisted solvo- or hydrothermal fabrication and characterization of high-surface-area porous calcium carbonate with multiple morphologies.
Microporous Mesoporous Mater, 138 (2011), pp. 191-199
D. Ng.
Regeneration of sodium carbonate using solar energy.
US Geological Survey.
Mineral commodity summaries.
Copyright © 2019. Brazilian Metallurgical, Materials and Mining Association
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.