Qing-Jun Guan,
Wei Sun*,
Yue-Hua Hu,
Zhi-Gang Yin and
Chang-Ping Guan
Central South University, No. 932 South Lushan Road, Changsha 410083, China. E-mail: sunmenghu1@163.com
First published on 25th May 2017
A brand new method to transform flue gas desulfurization gypsum (FGD gypsum) into α-calcium sulfate hemihydrate (α-HH) with short hexagonal prisms mediated by succinic acid disodium salt hexahydrate (C4H4O4Na2·6H2O) and NaCl in glycerol-water solution is studied, in which the appropriate reaction temperature is 90 °C. The addition of NaCl facilitates the dissolution of calcium sulfate dihydrate (DH) and creates much higher supersaturation which is a greater driving force for the phase transformation from DH to α-HH. C4H4O4Na2·6H2O as the crystal modifier effectively suppresses the α-HH crystal growth along the c axis, and the products generally change from needle-like particles to fat and short hexagonal prisms, which is attributed to the preferential adsorption of C4H4O42− on the top facets of the crystals by chelating Ca2+.
Numerous researchers have explored methods for the production of α-HH reusing FGD gypsum. One method is the autoclaving method,1 which involves heating DH in an autoclave at elevated temperatures and pressure under an atmosphere of saturated vapor. Obviously, this is an expensive procedure. Another method is the salt solution process under atmospheric pressure and at mild temperature,2–5 which requires high-concentration salt solutions, but evidently results in serious corrosion to equipment.
Recent scientific research has suggested that the transformation from DH to α-HH can be realized in alcohol water solution under mild conditions.6,7 Compared with the autoclave and salt solution methods, this process has the advantages of mild reaction conditions and no corrosion to equipment, and is thus more favorable for continuous industrial production. Although the transition from DH to α-HH is kinetically unfavorable in alcohol water solution, the addition of a small amount of salt as a phase transformation accelerator can significantly shorten the transition time.8 However, few researchers have explored the preparation of α-HH with a specific morphology and size using FGD gypsum in alcohol water solution. The morphology and size of α-HH usually determine its properties and functionalities. For instance, α-HH particles with a low aspect ratio or spherical shape possess better injectability and mechanic strength, and are preferable for use as bone cements.9,10 Therefore, in order to prepare α-HH products with low aspect ratios, crystal modifiers, among which carboxylic acids are highly effective,5,11,12 tend to be added during the transformation process. However, in alcohol water solutions, the influence of crystal modifiers on the morphology of α-HH crystals and transformation time from DH to α-HH has been rarely reported.
Herein, we introduce a brand new process to prepare α-HH with short hexagonal prisms by reusing FGD gypsum in glycerol-water solution. In the procedure, succinic acid disodium salt hexahydrate (C4H4O4Na2·6H2O) and NaCl are used as the crystal modifier and phase transformation accelerator, respectively, and the influence of the crystal modifier on the morphology of the α-HH crystals and the transformation time in glycerol-water solution is studied.
CaO | SO3 | H2O | SiO2 | Fe2O3 | Al2O3 | MgO | PbO | K2O | Total |
---|---|---|---|---|---|---|---|---|---|
36.87 | 42.63 | 14.95 | 0.73 | 0.31 | 0.31 | 0.11 | 0.06 | 0.04 | 96.01 |
Al Kα photon energy of 1486.6 eV. The C 1s peak at 284.8 eV, which is related to the carbon adsorbed on the surface during the exposure of the samples to ambient atmosphere, was used as the reference for all spectra. The interaction between the crystal modifier and the crystal surfaces was analyzed by Fourier transform infrared spectroscopy (FT-IR, IRAffinity-1, Shimadzu, Japan) at a resolution of 4 cm−1 over the frequency range of 400−4000 cm−1. The concentration of Ca2+ in the filtrate was determined via inductively coupled plasma-atomic emission spectrometry (ICP-AES, PS-6, Baird, USA). The average diameters and average aspect ratios of the α-HH products for each sample were determined from about 100 crystal products using the optical images and the Image-Pro plus 6.0 software.
From Fig. 1a, it can be seen that at 80 °C the diffraction peaks of α-HH appear at 18 hours, which suggests that α-HH crystals were formed. After that, the intensity of the DH peaks generally weakened and the α-HH peaks gradually strengthened due to the gradual phase transformation from FGD gypsum to α-HH. At 36 hours, the FGD gypsum was completely transformed into α-HH. When the reaction temperature was increased to 85 °C, the complete transformation time reduced sharply to 20 hours (Fig. 1b), and with an increase in the temperature to 90 °C, the FGD gypsum was completely converted into α-HH within 7 hours (Fig. 1c). Upon further increasing the temperature to 95 °C, the complete transformation time did not significantly reduce, as shown in Fig. 1d. Therefore, considering efficiency and saving energy, the appropriate reaction temperature should be 90 °C.
Without NaCl (Fig. 2a), the transformation from FGD gypsum to α-HH began at 360 min and was completed within 420 min. When 1% NaCl (by weight of FGD gypsum) (Fig. 2b) was added to the glycerol (65 wt%)-water solutions, the transformation started at 45 min and at 90 min the FGD gypsum was completely converted into α-HH. When the NaCl concentration was increased to 3% (Fig. 2c), the gypsum began to dehydrate at 15 min and the phase transformation was completed within 1 h. Upon further increasing the NaCl concentration to 5% (Fig. 2d) and 7% (Fig. 2e), the time of complete transformation from FGD gypsum to α-HH decreased to 45 min. As the phase transition reaction proceeded, α-HH generated by the FGD gypsum continued to dehydrate and was gradually transformed into anhydrous calcium sulfate (AH).
The promoting effect of NaCl on the transformation rate is attributed to the supersaturation (SHH) of α-HH in the solution. A higher supersaturation provided a greater driving force for α-HH nucleus formation, resulting in a shorter phase transformation time from FGD gypsum to α-HH. SHH is defined as:
(1) |
(2) |
As depicted in Fig. 3, as the NaCl concentration increased, the Ca2+ concentration and SHH/S0HH experienced an upward trend, and when the NaCl concentration increased from 0 to 7%, the concentration of Ca2+ increased from 0.24 to 0.90 mg L−1 and SHH/S0HH from 1.00 to 14.06. It is clear that the addition of NaCl facilitated the dissolution of FGD gypsum and provided more available lattice ions, which created a much higher supersaturation and thus was a greater driving force for the phase transformation.
Fig. 3 Influence of NaCl on Ca2+ concentration and SHH/S0HH in glycerol (65 wt%)-water solution at 90 °C. |
To confirm the interactions between the crystal modifier and the crystal surfaces of the α-HH crystals, FT-IR spectra were explored, as shown in Fig. 6. As shown in spectrum a, the peaks at 3612, 3564 and 1620 cm−1 are assigned to the O–H vibration of the crystal water molecules in the α-HH crystals. The three peaks at 1156, 1115, and 1097 cm−1 are assigned to the asymmetric stretching vibration of υ3 SO42−, and the peak at 1007 cm−1 is assigned to the distorted symmetric stretching vibration of υ1 SO42−. Also, the peaks at 660 and 602 cm−1 are indexed to the υ4 SO42− stretching.16,17
As depicted in spectra b and c, the peaks at 2972 and 2951 cm−1 are attributed to the asymmetric and symmetric stretching vibrations of methylene (–CH2–), and the peaks at 1558 and 1435 cm−1 are derived from the asymmetric and symmetric stretching vibrations of the ester group (–COO−), which all demonstrate the adsorption of C4H4O4Na2·6H2O on the α-HH crystal surfaces.18 Additionally, the peak at 808 cm−1 might be due to the distorted vibration of methylene (–CH2–).
Fig. 7 depicts the C 1s XPS spectra of the crystal products formed in the presence of different C4H4O4Na2·6H2O concentrations. As shown in the C 1s spectra (Fig. 7), the most intense C 1s peak observed in all the samples is at 284.8 eV, which is associated with carbon impurity19 or the C–(C, H) species in the crystal modifier.20,21 Without the crystal modifier (Fig. 7a), the C 1s spectra of the α-HH crystals prepared could be fitted with three components. Besides the major component at 284.8 eV, the component located at 286.5 eV is associated with the C–O in glycerol,19,20,22,23 which implies the adsorption of a trace amount of glycerol on the crystal surfaces.24–26 The high binding energy component of C 1s at 288.9 eV is derived from the carbonate species (CO32−) of the limestone existing in FGD gypsum.27,28 As depicted in Fig. 7b and c, when a certain amount of crystal modifier was added to the crystallization reaction system, another C 1s component appeared at 288.2 eV, which is assigned to the carboxylic or carboxylate group (COO−),29–31 and with an increase in the crystal modifier concentration, the intensity of the peak at 288.2 eV improved gradually, implying the adsorption of C4H4Na2·6H2O on the surfaces of the crystal products.
Fig. 8 shows the Ca 2p XPS spectra of the α-HH crystals prepared in the presence of different C4H4O4Na2·6H2O concentrations. As shown in Fig. 8a, in the absence of the crystal modifier, the Ca 2p spectra of α-HH could be fitted with two components. The component at 347.2 eV is attributed to the calcium of limestone existing in FGD gypsum, and the other is derived from the calcium of α-HH. When the crystal modifier was added to the reaction system, a third peak appeared at 347.7 eV, which is associated with the calcium chelated by the carboxyl group of the crystal modifier, as shown in Fig. 8b and c.32 With an increase in the modifier concentration, the strength of the peak at 347.7 eV enhanced correspondingly. These results demonstrate the chelating reaction between the calcium on the crystal surface of α-HH and the carboxyl group of the crystal modifier.
Table 2 shows the surface atomic distribution of the α-HH crystals formed in the presence of different concentrations of the crystal modifier. The surface carbon concentration of the α-HH products prepared without the crystal modifier might be mainly derived from carbon impurities or glycerol adsorbed on the crystal surfaces. From Table 2, it can be seen that with an increase in C4H4O4Na2·6H2O concentration, the surface oxygen and carbon concentrations of the α-HH products were generally enhanced, which implies the adsorption of C4H4O4Na2·6H2O on the α-HH crystal surfaces. However, the adsorption of the crystal modifier on the α-HH surfaces was much less than the amount added, and a certain amount of carboxylate ions (from the crystal modifier) could reduce water-gypsum ratio and improve the product strength to some extent. Therefore, there was no need to remove the crystal modifier from the produced α-HH crystals.
C4H4Na2·6H2O concentration (by weight of FGD) | Ca/% | S/% | O/% | C/% |
---|---|---|---|---|
0 | 15.30 | 16.80 | 64.92 | 2.98 |
0.06% | 14.31 | 15.48 | 65.91 | 4.30 |
0.12% | 11.56 | 12.72 | 66.32 | 9.40 |
As shown in Fig. 9, the α-HH lattices are composed of repeating, ionically bonded Ca and SO4 atoms in chains of –Ca–SO4–Ca–SO4–, and SO4 is a tetrahedral structure in which each S atom is covalently bonded to four O atoms.14,15 The structure of the chains may account for the fact that α-HH tended to grow in a 1D shape. These chains are hexagonally and symmetrically arranged and form a framework parallel to the c axis, and along the c axis continuous channels exist with a diameter of about 4.5 Å, where one water molecule is attached to every two calcium sulfate molecules.33 This crystal structure causes the distribution of Ca2+ to be denser on the 111 top facets and the distribution of SO42− denser on the 110 and 100 side facets, which makes the 111 facets positively charged and the 110 and 100 facets negatively charged.14,33 Therefore, when C4H4O4Na2·6H2O was added to the reaction system, C4H4O42− preferentially absorbed on the top facets by chelating Ca2+ and inhibited the α-HH crystal growth along the c axis, as shown in Fig. 10.34
Although succinic acid disodium salt hexahydrate (C4H4O4Na2·6H2O) as the crystal modifier could effectively tune the morphology of the α-HH products, it suppressed the nucleation and crystal growth of α-HH and prolonged the time for the complete transformation from FGD gypsum to α-HH. As shown in Fig. 11, with an increase in the crystal modifier concentration from 0 to 0.12%, the complete transformation time was extended from 25 min to 17 h. The formation of α-HH can be divided into three processes: dissolution of DH, α-HH nucleation and α-HH growth, of which nucleation is the rate-determining step.35,36 When the crystal modifier was added to the reaction system, C4H4O42− chelated free Ca2+ in the solution and lower the supersaturation for α-HH, which restrained the nucleation of α-HH, and thus prolonged the transformation time.12
The substitution of sodium ions in the modifier with calcium ions might have a considerable impact on the formation and properties of the α-HH products. Succinate calcium does not fully dissociate, and the succinate ions dissociated from succinate calcium in the solution would much less than that from succinate sodium, which would reduce the transformation time, but increase the aspect ratio of the α-HH products, in comparison with succinate sodium. However, the substitution of sodium ions in the modifier with other ions, such as potassium and magnesium, would not have a significant impact on the preparation of α-HH crystals because all of them are fully dissociated.
This journal is © The Royal Society of Chemistry 2017 |