Study of the conversion of the dihydrate phase into β-hemihydrate from two varieties of the mineral gypsum

ABSTRACT

This study investigated the optimal calcination conditions for obtaining β-hemihydrates from the “cocadinha” and “rapadura” varieties of gypsum, which are used in plaster production in the Araripe gypsum Pole. The samples were characterized using techniques such as scanning electron microscopy (SEM), thermogravimetric analysis (TGA), derivative thermogravimetry (DTG), and X-ray diffraction (XRD). The analyses indicated a similarity in the morphology of the samples. During calcination in a static furnace, mass losses ranging from 12.8% to 19.8% were observed. The study identified thermal events and crystalline phases as the calcination time and temperature varied. Complete conversion of the dihydrate phase into β-hemihydrate was achieved at 180 °C for 2 h, while partial conversion occurred at 160 °C for 2 h. The Rietveld refinement was successful, with χ² values close to 1 and R indicators below 10% for all analyzed samples. Based on the experimental results, the study identified optimal calcination conditions for producing hemihydrate plaster from the cocadinha and rapadura gypsum varieties, achieving consistent hardness and compressive strength in accordance with NBR 13.207 standards. Both varieties demonstrated setting times comparable to industrial benchmarks after 2 hours or more of calcination at 160 °C and 180 °C. These findings suggest the potential for process optimization, reducing energy consumption while maintaining product quality.

Keywords:
Araripe gypsum Pole; gypsum calcination; plaster manufacturing

1. INTRODUCTION

Gypsum is a hydrated calcium sulfate (CaSO₄·2H₂O) formed from calcium (Ca) and sulfur (S) that originate from the erosion of rocks. These elements are transported by water, which promotes the precipitation of the mineral. Precipitation occurs through the evaporation of groundwater, a process favored in arid environments and shallow saline lakes [¹].

Gypsum is composed of the following chemical elements: calcium (Ca), sulfur (S), oxygen (O), and hydrogen (H), with the mass proportions of oxides and the content of crystallization water being as follows: calcium oxide (CaO) 32.6%; sulfur trioxide (SO₃) 46.5%; and crystallization water (H₂O) 20.9% [²].

Natural gypsum is formed by crystals in the monoclinic crystal system, consisting of SO₄²⁻ and Ca²⁺ sites connected by crystallization water [³]. The phases that constitute the water-sulfate system are: gypsum (CaSO₄⋅2H₂O), bassanite (CaSO₄⋅1/2H₂O), and anhydrites (CaSO₄) α, β, and γ [⁴].

In the Araripe Gypsum Pole, different varieties of gypsum are found, known as: alabaster, “estrelinha,” “cocadinha,” “rapadura,” “johnson,” and “boró” [⁵]. Their applications include the production of α-plaster, β-plaster, Portland cement, agriculture, among others. The manufacturing process of β-plaster primarily involves the calcination of two types of gypsum: “rapadura” and “cocadinha”.

The Araripe Gypsum Pole is responsible for 97% of Brazil’s gypsum production, due to the ease of extraction, logistics, and the quality of the ore [⁶]. In the calcination process to obtain plaster, the dihydrate loses its crystallization water at an initial temperature of approximately 90 °C, and up to about 180 °C, it transforms into calcium sulfate hemihydrate (CaSO₄⋅1/2H₂O) [⁷].

In the Araripe region, the calcination process for obtaining plaster is usually carried out in rotary kilns, and in these calcination processes, significant amounts of CO₂ are emitted, between 5% and 8% of total emissions produced, contributing to increased pollution. This underscores the need to maximize the efficiency of calcination processes for obtaining plaster [⁸, ⁹].

According to previous studies [⁹], there is also a lack of effective control over key parameters in calcination, where different varieties of gypsum are calcined simultaneously, without effective control of temperature and calcination time, potentially leading to an inefficient process marked by excessive energy consumption.

Therefore, the objective of this research was to evaluate the efficiency of the calcination process by varying the temperature and calcination time parameters, with the purpose of determining the ideal conditions for the complete conversion of the dihydrate phase into β-hemihydrate, analyzing separately the “cocadinha” and “rapadura” gypsum varieties.

2. MATERIALS AND METHODS

2.1. Sample collection and powder preparation

The research began with a visit to the Araripe Gypsum Pole to collect samples of two varieties of gypsum known in the region as “cocadinha” (Figure 1a) and “rapadura” (Figure 1b).

Figure 1
Gypsum varieties “cocadinha” (a) and “rapadura” (b).

The fragments of the two gypsum varieties, still within the Araripe Gypsum Pole, were separately crushed using a jaw crusher connected to a 30 hp motor. Then, 12 kg of each of the previously crushed gypsum samples were ground for 4 h in a Los Angeles abrasion machine, brand Pavitest, model I-302, with a rotation speed of 30 rpm and 12 steel balls with approximately 48 mm in diameter, totaling 5.11 kg. Particle size analysis was performed through manual sieving in accordance with NBR 12.127 [¹⁰]. The aim was to determine the passing percentage through a 0.29 mm (mesh 50) sieve. As a characterization requirement, at least 90% of the total sample mass must pass through the sieve openings.

2.2. Morphology analysis

For the morphology analysis, powdered samples of the two varieties of gypsum underwent a metallization process with gold (Au) using a Quorum device, model Q150R ES, operating at a current of 15 mA for 5 min. Subsequently, micrographs of the samples were obtained using a TESCAN scanning electron microscope, model VEGA 3XMU.

2.3. Gypsum calcination

To determine the calcination temperature, both the literature [¹¹, ¹²] and the thermal analyses conducted in this study were considered, leading to the selection of two temperature ranges: 160 °C and 180 °C. The choice of these temperatures for calcination was based on thermogravimetric analysis (TGA), which showed that within this range, a significant loss of crystallization water occurs. To assess the efficiency of calcination, in terms of dihydrate conversion to hemihydrate, times of 1 h, 2 h, and 4 h were established for each of the gypsum varieties separately.

Sample dehydration was performed in a digital sterilization and drying oven (SOLAB, model SL-100) under ambient pressure. To monitor the operating temperature and residence time, an analog thermometer was placed in the upper central opening, allowing for the measurement of the actual temperature in the location. The powder samples were calcined separately, with 700 g of each sample (cocadinha and rapadura) per calcination step. The samples were evenly distributed in aluminum industrial trays, with a height of 10 to 12 mm. The tests were conducted in triplicate.

2.4. X-ray diffraction (XRD)

X-ray diffraction patterns of the powdered samples were obtained using a Rigaku MiniFlex 600 diffractometer with CuKα radiation and a wavelength of λ = 1.54060 Å, operating at a tube voltage of 40 kV and a current of 15 mA. Measurements were taken from 10° to 90° (2θ) at a speed of 10°/min with an angular step of 0.02°. Analyses were conducted at the Multiuser Laboratory for Material Characterization and Development (LaMDeM) at UNIVASF, Juazeiro-BA campus, Brazil.

The X’Pert HighScore Plus software version 3.0 (2009), along with the “Crystallography Open Database” (COD), was used for identifying crystalline phases and converting experimental data for analysis in Profex version 5.2.0 (2023). This software was employed to performing Rietveld refinement (RR) and the VESTA software was used to plot the unit cell.

2.5. Thermal analysis

Thermal analyses were carried out at the Interdisciplinary Laboratory of Advanced Materials (LIMAV) at the Federal University of Piauí (UFPI) using a thermobalance model Q-600 SDT (TA Instruments). The analyses were conducted in an inert argon atmosphere with a flow rate of 100 mL/min, using alumina sample holders and a heating rate of 20 °C/min, ranging from room temperature up to 900 °C. The Universal Analysis (2000) and Origin® version 8 software were used to interpret the results obtained from the curves.

2.6. Sample nomenclature

The nomenclature of the samples was defined according to Table 1.

Table 1
Nomenclature of gypsum samples and symbols.

CONDITION OF THE SAMPLE	VARIETIES
	COCADINHA	RAPADURA
In natura	C-NAT	R-NAT
Calcined at 160 °C for 1 h	C160/1	R160/1
Calcined at 160 °C for 2 h	C160/2	R160/2
Calcined at 160 °C for 4 h	C160/4	R160/4
Calcined at 180 °C for 1 h	C180/1	R180/1
Calcined at 180 °C for 2 h	C180/2	R180/2
Calcined at 180 °C for 4 h	C180/4	R180/4

2.7. Rehydration of the obtained hemihydrates

The preparation of the plaster paste for the tests in this work was based on NBR 12.128 [¹³], with a water/plaster ratio established at 0.7 to meet the normal consistency requirements according to NBR 13.207 [¹⁴]. The tests were conducted in a controlled environment at 25 °C.

2.7.1. Setting time

The setting times of the plasters were analyzed in accordance with NBR 12.128 [¹⁵], using the Vicat needle penetration test.

2.7.2. Mechanical characterization of the plaster

The evaluation of surface hardness and compressive strength of plaster is important to ensure the quality and adequate performance of the material in various applications. The NBR 12.129 [¹⁶] standard provides a standardized methodology for determining the surface hardness and compressive strength of plaster.

The cubic test specimens, with dimensions of 50 mm per side, were molded from the plaster pastes obtained from each of the gypsum varieties, with a water/plaster ratio of 0.7, using a metal mold with smooth and parallel edges. After allowing complete drying, on the fifteenth day, the hardened plaster was evaluated for surface hardness and axial compressive strength.

2.7.2.1. Surface hardness

A device developed at the Mechanical Testing Laboratory of UNIVASF by LÊLA [¹⁷] was utilized, consisting of a Trd 28 load cell adapted to a steel sphere. This setup was coupled to a gear system for manual load application, in compliance with NBR 12.129 [¹⁸]. The equipment features a 1 kN load cell attached to a steel sphere and is designed for manual load application. The test procedure involves positioning a metallic sphere with a diameter of 10.0 mm in contact with the surface of a cubic specimen with dimensions of 50 mm, applying an initial load of 50 N, and increasing it to 500 N over an interval of 2 seconds, maintaining this load for 15 seconds. Three faces of each sample were tested, and the indentation depth for each sample was calculated as the arithmetic mean of the measurements from the three tested faces. The procedure was performed on five cubic specimens, each with dimensions of 50 mm, prepared from the rehydration of hemihydrates obtained under different calcination parameters for each gypsum variety, analyzed separately. The results were subjected to hypothesis testing using the “t-test” to compare means and verify the reliability of the data.

2.7.2.2. Compressive strength

The material characterization in terms of compressive mechanical strength was performed in accordance with NBR 12.129 [¹⁶] at the Mechanical Testing Laboratory of UNIVASF, Juazeiro-BA campus, Brazil, using an Emic DL10000 hydraulic press with a 100 kN load cell, operating at a test speed of 2 mm∙min^–1. For this analysis, five cubic specimens with dimensions of 50 mm were used, prepared under different calcination parameters for each gypsum variety separately.

3. RESULTS

3.1. Scanning electron microscopy (SEM)

The morphologies of the powdered samples of the two gypsum varieties are shown in Figure 2.

Figure 2
Micrographs of the powdered gypsum samples: (a) cocadinha and (b) rapadura.

The cocadinha variety (Figure 2a) exhibits both regular parts and parts formed by lenticular agglomerations; the morphology of the rapadura variety (Figure 2b) shows regular parts as well as platy parts with the presence of agglomerates, as previously reported by [¹⁹].

3.2. Calcination of Gypsum Varieties

3.2.1. Production of β-Hemihydrate (CaSO₄.1/2H₂O)

During the calcination process, partial dehydration of gypsum occurs, forming β-calcium sulfate hemihydrate (CaSO₄.1/2H₂O) [¹²]. Partial dehydration of the two gypsum mineral varieties was observed during calcination at different temperatures and times, as illustrated in Figures 3a and 3b.

Figure 3
Mass loss during the calcination of C-NAT and R-NAT varieties.

It was observed that the increase in mass loss occurs with the increase in temperature and calcination time. From Figure 3, when the temperature is fixed and the calcination time is increased, a higher percentage of dehydration is achieved. According to the literature [¹², ²⁰] the formation of the β-CaSO₄.1/2H₂O phase occurs with an approximate mass loss of 15%. The results obtained ranged from 13–19.8% for the cocadinha variety and 12.8–19.8% for the rapadura variety, which are consistent with the literature. The error bars are overlaid at the top of the graph (Figure 3) because the mass loss variation for each replicate of the same analysis was only 0.02.

3.2.2. Thermal analyses

Thermogravimetric analysis (TGA) allowed the assessment of mass loss at different temperature ranges. According to [²¹], the mass loss of the dihydrate begins at approximately 100 °C, with the release of crystallization water, evidenced by endothermic peaks in the temperature range between 140 and 160 °C in DTG [²²]. Analysis of the TGA curves (Figure 4) revealed the mass loss associated with the dehydration of the natural gypsum samples: cocadinha (C-NAT) and rapadura (R-NAT). The DTG curve (Figure 4) showed that the loss of crystallization water occurs in a single step for all samples, indicating the formation of β-hemihydrate (CaSO₄.1/2H₂O) starting at approximately 100 °C up to approximately 180 °C, and of anhydrite III (CaSO₄.εH₂O) at approximately 210 °C.

Figure 4
TGA and DTG analysis of gypsum varieties C-NAT (a) and R-NAT (b).

As reported by ELERT et al. [¹²], in the temperature range between 210 °C and 700 °C, an average mass loss of 2.5% occurs on the TGA curve, indicating the formation of anhydrite II (CaSO₄). This event is confirmed by the endothermic peaks on the DTG curve.

3.3. X-ray Diffraction (XRD)

The crystal planes and phases present were precisely identified, as illustrated in Figures 5 and 6. The diffractograms obtained under different calcination conditions (Figures 5 and 6) indicate that calcination for 1 h at temperatures of 160 °C and 180 °C was not effective for converting the entire dihydrate phase into hemihydrate when comparing the diffraction peak intensities and positions in 2θ. The diffractograms from these analyses showed peaks at approximately 12° in 2θ, corresponding to the (020) crystal plane of gypsum, as illustrated in Figures 5–6.

Figure 5
X-ray diffraction patterns of cocadinha gypsum before and after calcination.

Figure 6
X-ray diffraction patterns of rapadura gypsum before and after calcination.

Calcinations lasting 2 h at 160 °C partially converted the dihydrate to hemihydrate, evidenced by the disappearance of the plane (020) of greater intensity of gypsum and the formation of the plane (400) of greater intensity corresponding to bassanite at 27.72° in 2θ. Calcinations at 180 °C for 2 h, 160 °C for 4 h, and 180 °C for 4 h achieved full conversion of gypsum to bassanite when comparing the diffraction peaks, their intensities, and positions in 2θ. This result suggests that calcination for 4 h would require more energy to obtain the hemihydrate, which has the same microstructural characteristics as the hemihydrates obtained through 2 h calcination.

In the literature [²³], similar behavior was observed in the diffractograms of the conversion from gypsum phase to bassanite, characterized by the disappearance of peaks below 15° in 2θ and the appearance of new peaks corresponding to bassanite. Through Rietveld refinement, it was possible to compare the experimental X-ray diffraction data with the reference parameters for gypsum (COD-2300258) and bassanite (COD-9012209).

The diffractograms presented in Figures 7 and 8 show the results of the refinement for the natural gypsum sample (dihydrate) and its conversion to bassanite (β-hemihydrate), respectively. The results obtained from the refinement of all samples are detailed in Tables 2–3. According to [²⁴], the weighted profile factor (R_WP) and the statistically expected value (R_EXP) should be below 10%; the quality of the refined model, indicated by χ² and G_oF, should approach 1 to confirm the success of the refinement.

Figure 7
Refined diffractogram of the cocadinha gypsum sample.

Figure 8
Refined diffractogram of the β-hemihydrate obtained.

Table 2
Rietveld refinement indicators for natural and calcined cocadinha gypsum varieties.

INDICATORS	SAMPLES
	C-NAT	C160/1	C160/2	C160/4	C180/1	C180/2	C180/4
RWP(%)	7.32	9.11	8.64	6.51	7.50	7.31	6.71
REXP(%)	5.12	5.44	5.60	5.74	5.51	5.64	5.74
χ2	2.04	2.80	2.38	1.28	1.85	1.68	1.37
GoF	1.42	1.67	1.54	1.13	1.36	1.30	1.17

Table 3
Rietveld refinement indicators for natural and calcined rapadura gypsum samples.

INDICATORS	SAMPLES
	R-NAT	R160/1	R160/2	R160/4	R180/1	R180/2	R180/4
RWP(%)	7.37	7.99	7.85	6.74	7.44	7.52	8.35
REXP(%)	5.00	5.48	5.40	5.59	5.50	5.62	5.82
χ2	2.17	2.12	2.11	1.45	1.82	1.79	2.05
GoF	1.47	1.45	1.45	1.20	1.35	1.33	1.43

Through Rietveld refinement (RR), it was possible to confirm the presence of the major phase of gypsum (CaSO₄⋅2H₂O) in the natural samples. As calcination progressed, gypsum was converted into bassanite (CaSO_4.0.5H₂O), which was also quantified in terms of percentage (Tables 4–5).

Table 4
Phase quantification of cocadinha variety samples by X-ray diffraction (XRD).

PHASES	SAMPLES
	C-NAT	C160/1	C160/2	C160/4	C180/1	C180/2	C180/4
CaSO4⋅2H2O (%)	100.00	7.70	0.50	–	2.00	–	–
CaSO4⋅1/2H2O (%)	–*	92.30	99.50	100.00	98.00	100.00	100.00
Desv.	0.000	0.009	0.000	0.000	0.002	0.000	0.000

• *= unidentified

Table 5
Phase quantification of rapadura variety samples by X-ray diffraction (XRD).

PHASES	SAMPLES
	R-NAT	R160/1	R160/2	R160/4	R180/1	R180/2	R180/4
CaSO4⋅2H2O (%)	100.00	2.53	1.05	–	2.46	–	–
CaSO4⋅1/2H2O (%)	–*	97.47	98.95	100.00	97.64	100.00	100.00
Desv.	0.000	0.01	0.01	0.00	0.01	0.00	0.00

• *= unidentified

The “χ²” values obtained from the refinement ranged from 1.28 to 2.80, and the G_oF values ranged from 1.13 to 1.67. The R_WP and R_EXP values were below 10%, indicating satisfactory refinement quality when comparing the experimental results with the reference standard.

Tables 4 and 5 present the respective proportions of dihydrate and hemihydrate in natural samples and under different calcination conditions.

Phase analyses confirmed that calcination was 100% effective at 180 °C for 2 h. The phase quantifications of natural samples also corroborate the analyses performed in this study, indicating high purity in the dihydrate. The data obtained through RR allowed for the determination of unit cells for both dihydrate and β-hemihydrate phases of calcium sulfate by importing crystallographic information files (CIF) into the VESTA software, as illustrated in Figures 9–10.

Figure 9
Unit cell of the dihydrate phase (CaSO₄⋅2H₂O).

Figure 10
Unit cell of the β-hemihydrate phase (CaSO₄⋅1/2H₂O).

The Figure 9 shows the unit cell of the dihydrate phase (gypsum), which is monoclinic, belongs to the space group C 2/c, and has the following lattice parameters: a = 6.34976 Å, b = 15.19110 Å, c = 6.53280 Å, α = 90.00°, β = 127.38°, and γ = 90.00°. The unit cell comprises 100 atoms forming 12 polyhedra connected by hydrogen bonds. The structure consists of six octahedral sites with a central calcium (Ca) atom and 8 adjacent oxygen (O) atoms, and six tetrahedral sites formed by 4 oxygen atoms around the sulfur (S) atom. The hydrogens (H) constitute the water present in the structure and are bonded to the sulfate group oxygen within the octahedral site in the unit cell. These results corroborate those obtained in references [²⁵, ²⁶].

Figure 10 shows the unit cell of the hemihydrate phase (bassanite), which is monoclinic, belongs to the space group I2, and has the following lattice parameters: a = 12.0354 Å, b = 6.9318 Å, c = 12.6866 Å, α = 90°, β = 90.25°, and γ = 90°. Its unit cell is composed of 210 atoms aggregated into 35 polyhedra primarily connected by (Ca-S) bonds due to the reduction in water molecules during calcination. These results corroborate those obtained in references [²⁵, ²⁶].

Based on these analyses, data were obtained that allow for the optimization of the β-hemihydrate production process, identifying the ideal temperature and time conditions to achieve the desired product with greater energy efficiency.

3.4. Rehydration of the obtained hemihydrates

The results of the setting time, hardness, and compressive strength tests, derived from the hemihydrates of the cocadinha and rapadura gypsum varieties, obtained through calcination under varying time and temperature conditions, are presented in Tables 6 and 7. The consistencies of the plaster pastes for all samples were classified as normal according to NBR 13.207 [¹⁴], with a standard consistency value of 30.00 ± 2.00 mm.

Table 6
Setting times, hardness, and compressive strength for C-NAT derivatives.

SAMPLES	SETTING TIMES		HARDNESS	COMPRESSIVE STRENGTH
	INITIAL TIME (MIN:S)	FINAL TIME (MIN:S)	MPa	MPa
C160/1	07:17 ± 00:14	11:33 ± 00:21	19.77 ± 0.38	6.12 ± 0.11
C160/2	16:08 ± 00:14	24:08 ± 00:14	19.20 ± 0.15	5.63 ± 0.12
C160/4	19:25 ± 00:21	25:52 ± 00:06	19.62 ± 0.21	6.33 ± 0.12
C180/1	07:43 ± 00:06	12:24 ± 00:14	19.24 ± 0.21	6.28 ± 0.07
C180/2	19:34± 00:14	26:07± 00:21	19.21 ± 0.12	6.31 ± 0.12
C180/4	19:47 ± 00:06	26:52 ± 00:14	19.54 ± 0.15	6.34 ± 0.08

Table 7
Setting times, hardness, and compressive strength for R-NAT derivatives.

SAMPLES	SETTING TIMES		HARDNESS	COMPRESSIVE STRENGTH
	INITIAL TIME (MIN:S)	FINAL TIME (MIN:S)	MPa	MPa
R160/1	07:33 ± 00:12	11:53 ± 00:21	19.83 ± 0.14	6.33 ± 0.06
R160/2	15:07 ± 00:14	26:17 ± 00:14	19.32 ± 0.15	5.74 ± 0.12
R160/4	16:25 ± 00:21	26:20 ± 00:06	19.62 ± 0.21	6.36 ± 0.12
R180/1	07:43 ± 00:06	12:24 ± 00:14	19.40 ± 0.21	7.40 ± 0.38
R180/2	16:08 ± 00:14	24:08 ± 00:06	19.38 ± 0.12	6.35 ± 0.12
R180/4	16:28 ± 00:06	26:52 ± 00:14	19.41 ± 0.15	6.48 ± 0.08

As the calcination time and temperature increased, the initial setting times for the samples derived from the cocadinha variety (C-NAT) ranged from 07 minutes and 17 seconds to 19 minutes and 47 seconds, while the final setting times varied from 11 minutes and 33 seconds to 26 minutes and 52 seconds.For the samples derived from the rapadura variety (R-NAT), the initial setting times ranged from 07 minutes and 33 seconds to 16 minutes and 28 seconds, and the final setting times ranged from 11 minutes and 53 seconds to 26 minutes and 52 seconds.

FERREIRA et al. [²⁷], in their study evaluating the setting time using the Vicat method for five plaster samples from different industries in the Araripe Gypsum Pole, observed initial setting times between 10 minutes and 18 minutes and final setting times between 13 minutes and 25 minutes for normal consistency. Therefore, it is noted that for calcination periods of 2 hours or more, at 160 °C and 180 °C, the results were similar to those obtained by the industry.

According to CARDOSO et al. [²⁸], the setting time is influenced by the mixing procedure. JOHN and CINCOTTO [²⁹] report that the setting time is influenced by the raw material used, corroborating the results of this study, which used two gypsum varieties separately under the same calcination conditions, yet presented different setting times.

From the hardness test, it was also possible to assess the applicability of the plaster for civil construction, according to NBR 13.207 [¹⁴], which establishes a hardness greater than or equal to 20 N/mm². Observing Tables 6 and 7, it can be seen that both samples showed similar hardness values, confirmed by the “t-test” for the mean with 95% confidence, and considered statistically equivalent, even when compared to the 20 MPa hardness standard set by NBR 13.207 [¹⁴]. In the compressive strength tests, all samples fell within the acceptable standards for plaster used in civil construction, with a threshold of 5 MPa. Therefore, based on the analyses performed, it was possible to obtain data that allows for the optimization of the hemihydrate production process, identifying the temperature and calcination time conditions to achieve the desired product and avoid energy waste.

4. CONCLUSIONS

Scanning Electron Microscopy (SEM) analyses revealed that the morphology of cocadinha gypsum consists of regular parts and lenticular agglomerations, while rapadura gypsum exhibits regular and platy parts with superficial clusters.

During the calcination process of the C-NAT and R-NAT gypsum varieties, a progressive mass loss was observed, increasing with both elevated temperatures and longer calcination times, indicating partial dehydration of the mineral.

Thermogravimetric Analysis (TGA) revealed that the mass loss during the calcination of gypsum samples is associated with dehydration, with the release of crystallization water beginning around 100 °C and continuing up to approximately 200 °C.

X-ray Diffraction (XRD) analysis accurately identified the crystalline phases and crystal planes present in cocadinha and rapadura gypsum samples, both before and after calcination. Prolonged calcination for 2 h at 180 °C and 4 h at 160 °C and 180 °C resulted in the complete conversion of gypsum to bassanite.

Rietveld refinement enabled the quantification of crystalline phases present in cocadinha and rapadura gypsum samples, both before and after calcination. The presence of the major gypsum phase (CaSO₄⋅2H₂O) was confirmed in the natural samples, with complete conversion to hemihydrate (β-CaSO₄⋅0.5H₂O) during 2 h of calcination at 180 °C. Through Rietveld method, it was possible to determine the unit cells of the phases and their lattice parameters.

The setting time, hardness, and compressive strength tests further confirmed the suitability of cocadinha and rapadura gypsum varieties for industrial applications. The setting times for both samples, following calcination at 160 °C and 180 °C, were similar to those observed in the gypsum industry of the Araripe region for calcinations of 2 hours or more.

The results allowed for the identification of the optimal temperature and time conditions to produce β-hemihydrate calcium sulfate from rapadura and cocadinha gypsum varieties, focusing on energy efficiency and process optimization for plaster manufacturing. These findings suggest that both samples can be calcined simultaneously in an industrial setting.

5. ACKNOWLEDGMENTS

The authors wish to acknowledge the laboratories at UNIVASF, Juazeiro-BA, Brazil, including the Multi-User Laboratory for Materials Characterization and Development (LaMDeM), the Laboratory of Materials and Construction Techniques, and the Laboratory of Mechanical Testing, for providing support and infrastructure during this study. Gratitude is also extended to the Interdisciplinary Laboratory of Advanced Materials (LIMAV) at UFPI, Teresina-PI, Brazil, for their valuable contributions.

6. BIBLIOGRAPHY

• ^[1] REISS, A.G., GAVRIELI, I., ROSENBERG, Y.O., et al, “Gypsum precipitation under saline conditions: thermodynamics, kinetics, morphology, and size distribution”, Minerals (Basel), v. 11, n. 2, pp. 1–38, 2021. doi: http://doi.org/10.3390/min11020141.
  » https://doi.org/10.3390/min11020141
• ^[2] HASHEMI, M., SAKI, A., NIKUDEL, M.R., “Effects of chemical composition on the solubility and geomechanical characteristics of gypsum rocks from Gachsaran Formation, Iran.”, Bulletin of Engineering Geology and the Environment, v. 79, n. 7, pp. 3785–3796, 2020. doi: http://doi.org/10.1007/s10064-020-01794-y.
  » https://doi.org/10.1007/s10064-020-01794-y
• ^[3] CARREÑO-MÁRQUEZ, I.J.A., CASTILLO-SANDOVAL, I., PÉREZ-CÁZARES, B.E., et al, “Evolution of the astonishing Naica giant crystals in Chihuahua, Mexico”, Minerals (Basel), v. 11, n. 3, pp. 1–20, 2021. doi: http://doi.org/10.3390/min11030292.
  » https://doi.org/10.3390/min11030292
• ^[4] SCHMID, T., JUNGNICKEL, R., DARIZ, P., “Insights into the CaSO₄-H₂O system: a raman‐spectroscopic study”, Minerals (Basel), v. 10, n. 2, pp. 115, 2020. doi: http://doi.org/10.3390/min10020115.
  » https://doi.org/10.3390/min10020115
• ^[5] FILHO, P.L.C., ARTHUR, R., FERREIRA, S., et al, “Caracterização estrutural e microestrutural do gesso produzido pelo pólo gesseiro de araripina”, Cerâmica Industrial, v. 13, n. 3, pp. 31–38, 2008.
• ^[6] SOUZA, E.C., “Caracterização da Gipsita da Região do Araripe Pernambucano”, M.Sc. Thesis, UNIVASF, Juazeiro, BA, Brazil, 2018.
• ^[7] SANTOS, J.P.O., GIOVANETTI EL-DEIR, S., “Produção de gesso no Araripe pernambucano: impactos ambientais e perspectivas futuras”, Revista AIDIS de Ingeniería y Ciencias Ambientales, v. 12, n. 3, pp. 496, 2019. doi: http://doi.org/10.22201/iingen.0718378xe.2019.12.3.62953.
  » https://doi.org/10.22201/iingen.0718378xe.2019.12.3.62953
• ^[8] BOUZIT, S., LAASRI, S., TAHA, M., et al, “Characterization of natural gypsum materials and their composites for building applications”, Applied Sciences (Basel, Switzerland), v. 9, n. 12, pp. 2443, 2019. doi: http://doi.org/10.3390/app9122443.
  » https://doi.org/10.3390/app9122443
• ^[9] JOHN, J.P., “Parametric studies of cement production processes”, Journal of Energy, v. 2020, pp. 1–17, 2020. doi: http://doi.org/10.1155/2020/4289043.
  » https://doi.org/10.1155/2020/4289043
• ^[10] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 12127: Gesso para construção civil - Determinação das propriedades físicas do pó, Rio de Janeiro, ABNT, 2019.
• ^[11] CIPRIANO, P.B., GALDINO, T.S.G., SA, C.S., et al, “Evaluation of the calcination parameters of the gypsum waste on the properties of recycled plaster.”, Matéria (Rio de Janeiro), v. 26, pp. e13026, 2021. doi: http://doi.org/10.1590/s1517-707620210003.13026.
  » https://doi.org/10.1590/s1517-707620210003.13026
• ^[12] ELERT, K., BEL-ANZUÉ, P., BURGOS-RUIZ, M., “Influence of calcination temperature on hy-dration behavior, strength, and weathering resistance of traditional gypsum plaster”, Construction & Building Materials, v. 367, pp. 130361, 2023. doi: http://doi.org/10.1016/j.conbuildmat.2023.130361.
  » https://doi.org/10.1016/j.conbuildmat.2023.130361
• ^[13] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 12128:“Gesso para a construção civil - Determinação das propriedades físicas da pasta de gesso, Rio de Janeiro, ABNt, 2017.
• ^[14] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 13207: Gesso para construção civil - Requisitos, Rio de Janeiro, ABNT, 2017.
• ^[15] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 12128: Gesso para a construção civil - Determinação das propriedades físicas da pasta de gesso, Rio de Janeiro, ABNT, 2017.
• ^[16] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 12129: Gesso para construção civil - Determinação das propriedades mecânicas, Rio de Janeiro, ABNT, 2019.
• ^[17] LÊLA, L.A. “Desenvolvimento de dispositivo para determinar dureza mecânica em gesso segundo NBR 12129”. Trabalho de Conclusão de Curso. Engenharia Mecânica - UNIVASF, Juazeiro, BA, Brazil, 2017.
• ^[18] ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS, ABNT NBR 12129: Gesso para construção civil - Determinação das propriedades mecânicas, Rio de Janeiro, ABNT, 2019.
• ^[19] SOUZA, E.C., OLIVEIRA, R.A.P., FERRAZ, A.V., “Characterization of different samples of the gypsum mineral from the Araripe gypsum Pole: from macro to micro with Rietveld refinement”, Holos, v. 8, n. 39, 2023. doi: http://doi.org/10.15628/holos.2023.14902.
  » https://doi.org/10.15628/holos.2023.14902
• ^[20] SCHMIDT, H., PASCHKE, I., FREYER, D., et al, “Water channel structure of bassanite at high air humidity: crystal structure of CaSO₄·0.625H₂O”, Acta Crystallographica. Section B, Structural Science, v. 67, n. Pt 6, pp. 467–475, 2011. doi: http://doi.org/10.1107/S0108768111041759. PubMed PMID: 22101536.
  » https://doi.org/10.1107/S0108768111041759
• ^[21] ENGBRECHT, D.C., HIRSCHFELD, D.A., “Thermal analysis of calcium sulfate dihydrate sources used to manufacture gypsum wallboard”, Thermochimica Acta, v. 639, pp. 173–185, 2016. doi: http://doi.org/10.1016/j.tca.2016.07.021.
  » https://doi.org/10.1016/j.tca.2016.07.021
• ^[22] HOSSAIN, M.S., AHMED, S., “Synthesis of nano-crystallite gypsum and bassanite from waste Pila globosa shells: Crystallographic characterization”, RSC Advances, v. 12, n. 38, pp. 25096–25105, 2022. doi: http://doi.org/10.1039/D2RA04881G. PubMed PMID: 36199888.
  » https://doi.org/10.1039/D2RA04881G
• ^[23] CORDON, H.C.F., CAGNONI, F.C., FERREIRA, F.F., “Comparison of physical and mechanical properties of civil construction plaster and recycled waste gypsum from São Paulo, Brazil”, Journal of Building Engineering, v. 22, pp. 504–512, 2019. doi: http://doi.org/10.1016/j.jobe.2019.01.010.
  » https://doi.org/10.1016/j.jobe.2019.01.010
• ^[24] YOUNG, R.A., The Rietveld Method, Oxford, Oxford University Press, 1993. doi: http://doi.org/10.1093/oso/9780198555773.001.0001.
  » https://doi.org/10.1093/oso/9780198555773.001.0001
• ^[25] FUKAMI, T., TAHARA, S., NAKASONE, K., et al, “Synthesis, crystal structure, and thermal properties of CaSO₄·2H₂O single crystals”, International Journal of Chemistry, v. 7, n. 2, pp. 12, 2015. doi: http://doi.org/10.5539/ijc.v7n2p12.
  » https://doi.org/10.5539/ijc.v7n2p12
• ^[26] BEZOU, C., NONAT, A., MUTIN, J.C., et al, “Investigation of the crystal structure of γ-CaSO₄, CaSO₄·0.5H₂O, and CaSO₄·0.6H₂O by powder diffraction methods”, Journal of Solid State Chemistry, v. 117, n. 1, pp. 165–176, 1995. doi: http://doi.org/10.1006/jssc.1995.1260.
  » https://doi.org/10.1006/jssc.1995.1260
• ^[27] FERREIRA, F.C., SOUSA, J.G.G., CARNEIRO, A.M.P., “Caracterização mecânica do gesso para revestimento produzido no Polo Gesseiro do Araripe”, Ambiente Construído, v. 19, n. 4, pp. 207–221, 2019. doi: http://doi.org/10.1590/s1678-86212019000400352.
  » https://doi.org/10.1590/s1678-86212019000400352
• ^[28] CARDOSO, F.A., AGOPYAN, A.K., CARBONE, C., et al, “Squeeze flow as a tool for developing optimized gypsum plasters”, Construction & Building Materials, v. 23, n. 3, pp. 1349–1353, 2009. doi: http://doi.org/10.1016/j.conbuildmat.2008.07.017.
  » https://doi.org/10.1016/j.conbuildmat.2008.07.017
• ^[29] JOHN, V.M., CINCOTTO, M.A., “Gesso de Construção Civil”, In: ISAIA G.C., Editor, Materiais de Construção Civil, São Paulo, IBRACON, pp. 727–760, 2007.

Publication Dates

• Publication in this collection
  28 Feb 2025
• Date of issue
  2025

History

• Received
  22 Oct 2024
• Accepted
  21 Jan 2025