Received Date: November 30, 2015; Accepted Date: December 14, 2015; Published Date: January 04, 2016
Citation: Daik R, Lajnef M, Amor SB, Ezzaouia H (2016) Application of Gettering Process on the Improvement of the Structural and Mineralogical Properties of Tunisian Phosphate Rock. J Material Sci Eng 5:222. doi:10.4172/2169-0022.1000222
Copyright: © 2016 Daik R, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Material Sciences & Engineering
The present study deals with the effects of gettering process on the structural and mineralogical composition of Tunisian phosphate rock. The treated samples were characterized to investigate the variation of physical structure and chemical composition as compared to the reference phosphate rock. The quantitative analysis of the impurities concentration before and after gettering treatment using energy-dispersive (EDX) reveals a significant reduction of impurity concentration (more than 75%) such as Al, Si, S, Na, and Mg. Scanning electron microscopy (SEM) shows that gettering process promoted structural alterations of phosphate rock sample due to fusion of impurities. The XRD patterns show that the chief mineral constituent of treated sample is only fluorapatite, while those in the reference ore were calcite, dolomite, quartz and carbonate-fluorapatite. FT-IR characterization show a disappearance of the bands related to calcite at 714 cm-1 as well as B carbonate situated at 1430 cm-1, 1458 cm-1 after gettering treatment. This result is in good correlation with Raman analysis.
Phosphate rock; Gettering; Impurities; Structure
Gafsa which is located in the south of Tunisia is one of the largest phosphate producers in the word (more than 10 million tons per year since the early nineties) . The phosphate rock is used to manufacture phosphate fertilizers and industrial products and, also the only significant global resource of phosphorus used in animal feed supplements, food preservatives, anti-corrosion agents, cosmetics, fungicides, ceramics, water treatment and metallurgy . The rock is composed essentially of the apatite group in association with a wide assortment of accessory minerals mainly fluorides, carbonates (calcite and/or dolomite), clays, quartz, silicates, metal oxides as well as organic matters and trace impurities such as U, REEs (rare earth elements), Cd, As, V, Cr, Zn, Cu, Ni, etc., which can be harmful for several application at certain concentration [3-6].
Gettering process typically consists on a combined a rapid thermal treatment (RTP) followed by a chemical etching after the growth of a porous layer in order to reduce impurities amount and to enhance the phosphate quality. The rapid thermal treatment aims to migrate the impurities to the boundaries surface where they undergone an elimination process by a chemical attack.
The gettering treatment is an effective process to eliminate these impurities which was already applied in our laboratory for the purification of silica . Although, characterization and quantification of the impurities contained in Tunisian phosphate is well established, there are not many reports about their elimination and phosphate purification [5,8-12]. In the present work, we aim not only to improve the structural and mineralogical properties but also to eliminate the majority of these impurities contained in Tunisian phosphate by gettering treatment. The changes in physical structure and chemical composition of the samples after gettering treatment have been investigated by using X-ray powder diffraction (XRD), scanning electron microscopy (SEM/EDX), FT-IR, and Raman spectroscopies. Transmission electron microscope (TEM) micrographs were performed to inspect the morphological properties after treatment.
Phosphate rock samples used in this study was obtained from the phosphate deposits from the Metlaoui basins located in the south of Tunisia. It was crushed, ground, and then sieved, the fraction in the range between 180 μm and 600 μm was used. This fraction was crushed by a jaw breaker, reaching a dimension of to 180 μm. Another manual grinding is performed using an agate mortar in order to increase specific surface area.
The experimental procedure consists in two steps:
First step (formation of porous layer): the porous layer of phosphate rock is formed by CAVP technique (Chemical Attack in the Vapor Phase) when the sample is exposed to an acidic vapor composed of 64% HNO3, 20% CH3COOH and 16% HF. The vapor phase etching is performed under heating at 45°C for 60 minutes. The objective of growing of porous layer on the grain surface is to increase the specific surface layer, thus the impurities can be removed easily.
Second step (gettering process): the sample of porous phosphate is introduced in the rapid thermal furnace (RTP) at a fixed temperature 900°C for 45 minutes under a flow of oxygen. In order to remove the impurities from the samples, the thermally treated porous phosphate undergoes four iterative etchings: 1 g of the former sample is etched with 20 ml of diluted solution of CP4 (3 ml HCl + 1 ml HNO3 were dissolved in 996 ml of deionizer water). The mixed solution undergoes a stirring for 3 minutes. The treated phosphate in the final phase is separated from the obtained solution by a filtration system of 0.54 mesh diameter. The solid remaining was washed, dried during 1 hour at 100°C then weighed with a precision. For simplification, we have noted (RP) the reference phosphate sample and (TP) phosphate after gettering treatment (treated phosphate).
X-Ray diffraction were performed using X'PERT Pro Philips analytical diffractometer operating at wavelength Kα copper (λ =1.5418 nm) and the obtained results were analyzed using the software X'PertisHigh Score Plus.
The IR spectra were recorded using a Nicolet 560 spectrometer; samples pelletized using a pressein potassium bromide (KBr) to 2 mg of product 300 mg of KBr. Registration is realized in the range between 4000 cm-1 and 400 cm-1.
Raman shift were recorded with micro-Raman spectroscopy (Jobin Yvon Horibra LABRAMHR) in 400 cm-1-1100 cm-1 range. The excitation source was 632.8 nm line of He-Ne laser. The microstructure of samples was characterized by transmission (Technai G2) electron microscopy. The chemical composition was determined by energy disersive X-ray EDX analysis. For the TEM sample preparation, we employed the ultrasound vibration method . The samples were immersed in ethanol solution and ultrasound vibration was applied to separate precipitates from the phosphate. After that, the precipitates were carefully extracted in the solution and picked up using TEM copper meshes with carbon film coatings.
A powder X-ray diffraction (XRD) analysis was used to determine the crystalline phases of the Tunisian natural phosphates rocks before and after getting process. The XRD patterns of the treated phosphate rocks as well as the raw material are illustrated in Figure 1.
The main minerals in reference phosphate rocks (RP) are carbonatefluorapatite (2?: 25.99°; 28.17°; 29.44°; 31.97°; 33.24°; 34.15°; 40.31°; 42.48°; 44.48°; 47.01°; 49.55°; 50.81°; 53.09°) (JCPDS 00-021-0141), quartz SiO2 (2θ: 26.54°; 51.9°; 56.25°) ( JCPDS 01-080-2146), carbonates which are in the form of dolomite CaMg (CO3)2 (2θ : 30.7°, 41.06°, 50.09°) ( JCP¨DS 01-073-2409) and calcite CaCO3 (2θ : 29.44°, 39.42°, 43.18° and 48.46°) (JCPDS 01-072-1652). Calcite and quartz were the main gangue minerals in the Tunisian phosphate rock. Concerning the treated sample (TP), as expected from Figure 1, the major crystalline phase is hexagonal fluorapatite (FAp) (Ca10(PO4)6F2), space group P63/m (JCPDS 01-079-1459). The highest intensity near 33° confirms the fluorapatite behavior of the treated sample [2,7,9,10,14,15].
Calcite and quartz diffraction lines are disappeared as a result of gettering process, also carbonate-fluorapatite has changed to fluorapatite because carbonates are decomposed by rapid thermal treatment. Therefore, it was proved in others works that rapid thermal treatment at 900°C leads to a phosphate with relatively higher P2O5 and CaO contents and a disappearance of organic matter [16,17]. In this work CaO formed after rapid thermal treatment was eliminated by chemical attack in vapor phase (ACPV).
XRD pattern of treated phosphate compared with reference sample (Figure 1) shows a good resolution of the peaks and a decrease of the width at half maximium which proves an amelioration of crystalinity after gettering process.
X-ray diffraction analysis indicates that certain level of impurities were removed during gettring process of phosphate rock. However, the improvement of the phosphate quality of treated sample depends on the mass percentages of the remaining impurities, notably the quartz, calcite and dolomite. In previous work, it was demonstrated also that the reduction of phosphate impurities was associated with some structural changes in the apatite .
Figure 2 shows the FT-IR spectra of Tunisian phosphate before and after gettering treatment in the region of 4000 cm-1-400 cm-1. From this figure, we can observe that the gettering process had a remarkable an important effect on the vibrational bands intensity and its positions; also we can note the appearance and disappearance of some pics.
The FTIR spectrum indicates that the reference phosphate rock spectrum shows that the characteristic absorption bands corresponds to the carbonate fluorapatite [7,17,18]. The symmetric ?1 (stretching) mode assigned to PO4 3- is represented by a single band at 966 cm-1. The ?2 (bending) mode of phosphate groupment is located at 474 cm-1. The strong absorption band at 1044 cm-1 ascribed to asymmetric ?3 mode. The asymmetric ?4 mode is splited in three bands: 568, 578 and 604 cm-1. The two bands at 1430 cm-1 and 1458 cm-1 were assigned to ?2 vibration of CO3 2- group located in the B site of apatite (carbonate substituting phosphate) . The spectrum of main component of the phosphate rock reference sample is in a good agreement with published IR spectra of apatite [19-21].
FT-IR spectra of the TP sample illustrated in Figure 2 reveals that the vibrational bands of treated phosphate were clearly observed compared with reference phosphate rock. The absorption peaks located at 1098 cm-1 and 1062 cm-1 originated from asymmetrical stretching ?3 of PO3-4 and the peaks localized at 568 cm-1 and 606 cm-1 were attributed to bending modes ?4 of PO3 -4. While the symmetric stretching modes ?1 and ?2 of PO3-4 were also observed at around 964 cm-1 and 520 cm-1 respectively .
Moreover, after gettering process, the band positions and their intensities are slightly affected and we observe a change in the number of phosphate bands, the treated phosphate indicates that the bands at 520 cm-1 correspending to ?2 strongly shifted from 474 cm-1 to 520 cm-1. Concerning the shift, it can be due to the variation repulsion potential of the contracted or dilated crystal lattice which is confirmed by XRD analysis [7,22]. The positions of ?4 and ?1 modes didn’t change but an important increase of intensity was marked. The ?3 asymmetric mode was degenerated in tow distiguitched peacks at 1042 cm-1 and 1062 cm-1. The appearance of the two distinct peaks is due to the presence of different P-O distances in the crystal.
Besides, a considerable reduction in the absorption of carbonate bending is shown clearly after gettering treatment. In fact, we remark a disappearance of the bands related to calcite at 714 cm-1 as well as B carbonate situated at 1430 cm-1, 1458 cm-1. This implies that carbonate and calcite substitutions induce vacancies at the OH sites, and we assume that thermal treatment is responsible of the total decomposition of carbonate bands and intensities decreases . Thus, the results indicate that mixture acids can be used to reduce calcium carbonate in low-grade calcareous phosphate rock as it improves the degree of beneficiation .
Raman scattering is a sensitive tool for studying the phosphate material because it gives direct structural evidence qualitatively related to the different components in the material. Figure 3 shows the Raman spectra for RP and TP. From this figure, we can’t observe any vibrational mode for reference sample (RP). This is due to the fact that Raman bands are completely overlapped by the fluorescence background originated from organic matter, metal compouned and rare earth existing in natural phosphate rock (RP) [25-29]. As a result of this overlap, we can’t differentiate between the different vibrational modes.
For the TP, Raman spectra shows obviously the different vibrational modes of phosphates groupement after Gettering process. The strongest Raman active ?1 of PO3-4 mode appearance in the spectrium of the TP sample at 961 cm-1 [30,31].
To better clarify the vibrationnal modes existing in TP, deconvolution of the Raman spectrum were shown in Figures 4 and 5. The Raman spectrum of phosphate in the 125-300 cm-1 spectral range is illustrated in Figure 4. Raman bands are observed at 139, 169, 214, 234, 265 and 283 cm-1. These bands are assigned to lattice vibrations as it was reported by many authers [11,18,32].
The Raman spectrum of treated phosphate over the 400-620 cm-1 spectral range is reported in Figure 6. This range is assigned to the vibration of ?2 and ?4 PO43- bending modes. It was reported by S. Elgharbi and H. Lefires when they work about Tunisian phosphate rock that the Raman bands at 582, 591 and 607 cm-1 are assigned to ?4 PO43- and the bands at 431 and 435 cm-1 are due to the ?2 PO43- [12,13]. Then, the work reported by Karampasa about calcium phosphate confirmed very well the results above .
The Raman spectrum over the 850-1200 cm-1 range is reported in Figure 6. Similar intensity bands are found at 1056, 1100, 1116 cm-1 which are assigned to ?3 PO43- antisymmetric stretching vibration, the three bandes are attributed to a pure fluorapatite . Low intensity Raman band at 1009 cm-1 is attributed to ?1 PO43- symmetric stretching mode [28,33,34].
From Raman analysis, we can conclude that the gettereing process in necessary to eliminate the impurities and organic matters which are the main causes of overlopped in reference phosphate rock, and consequently improves the structural properties.
The change in the physical structure of the treated sample in comparison with the raw ore was investigated by TEM. This scanning procedure consisted of looking for structure alterations, agglutination, porosity, morphology, compaction, and distribution, with qualitative and semi quantitative identification of elements .
It is shown in Figure 7 that the phosphate rock consists of two different particule phases with estimated sizes of 60 μm. Moreover, these phases are defined with tow portions which can be due to the accumulation of the impurities which escape the dispersion of the particles. The portion in light grey is formed by phosphorous rich components whereas the portion in dark grey are formed by calciumrich components, which can be defined as CaCO3, based on chemical analysis. No phosphorous was found in the carbonate parts. Carbonatefluorapatite existing in the ore has only been observed in the parts with phosphorous-rich components. The surfaces of the parts with phosphate exhibit a compact structure with only little porosity.
The TEM micrographs of the treated sample by gettering process (porous phosphate treated 45 min at 900°C and eatching in mixture acid ) is given in Figure 8 shows that the TP sample is formed by many crystals with baton forms. It seems that the RP sample was subdivided to many particles with different sizes. It was determined that the shrinkage and the cracks at the surrounding parts with phosphate occurring due to thermal and etching treatments . The holes on the surfaces of the parts with phosphate prove that carbonate–fluorapatite was calcined and that the carbonate–fluorapatite changed to fluorapatite. This is due to the disappearance of the impurities which occupied interstitial sites, grain boundaries. Only the preponderant elements appear in the imagery which is confirmed by quantitative analysis. These results are in good agreement with the XRD analysis.
To get more insight of the composition of the RP and TP samples, Energy-dispersive X-ray (EDX) was used in many places of the sample area. The results were summarized in Table 1. From Figures 7 and 8, we noticed that the major elements before treatment are P, Ca, F in addition to the impurities such as Al, Si, S, Na, Mg…Whereas, after treatment only the P, Ca, F are presented with small traces S, Na and Si.
Table 1: Atomic percent of reference phosphate (RP) and treated phosphate (TP).
Table 1 shows the quantitative chemical composition. The analysis shows a homogeneous phase composed by P, Ca and F as being major elements consists mainly of fluorapatite. The chemical composition of phosphate rock shows that after treatement process, it changes to a rather poor in magnesium, in silica and metal such as Al, Fe.
Moreover, the electron microprobe analysis of samples allows us to evaluate the ratio Ca/P. Compared to the reference phosphate are 2.4, treated phosphate (TP) is become 1.7 which is closer to the theoretical Ca/P molar ratio of pure FAP: 1.67. However, this proves the presence of carbonate-FAP and calcite in reference sample and the presence of calcium oxide in excess . This difference in composition may take place by incorporation of ion present in the site of PO42- group. In our case these elements are F-, Na+, CO32- known to be incorporated into the network of the apatite.
A marked change on the properties of Tunisian phosphate rock was observed following the gettering process. The experimental results in this study suggest a significant improvement in the structure as well as the composition of the treated phosphate rock. Therefore, we consider that gettering process is not only a promising way to eliminate the impurities but also it enhances the use of phosphate rock in many fields.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals