Magnetization of Weakly Magnetic Mineral by Iron Ions in Slurry

X. Q.Wu,Y. L.Ai

School of Minerals processing & Bio-engineering, Central south university, Changsha, China

 

 

ABSTRACT

Magnetization in slurry (MIS) of weakly magnetic minerals by iron ions was investigated. The magnetic seeds (strongly magnetic particles) were synthesized by ferruginous chlorides directly in slurry instead of previous preparations of magnetic seeds by finely grinding mineral magnetite or specially synthesizing them separately in an aqueous solution. In this way, the growth of the seeds and their adsorption on the minerals simultaneously take place in the same slurry. That after MIS the recovery of weakly magnetic mineral limonite in a magnetic separator increased means the formation and adsorption of magnetic seeds on the surface of iron mineral limonite. The magnetization effect was evaluated by limonite recovery at a magnetic strength of 3.500 Gs in a magnetic separator. Mineral limonite was properly magnetized in the following slurry condition: molar ratio of Fe³⁺ to Fe²⁺ 1:3, slurry temperature of 45 °C and pH 10.5. The selectivity of the magnetization between minerals limonite and quartz was improved by adding surfactant sodium oleate.

Keywords: Magnetization; Limonite; Iron salts; Sodium oleate; Magnetic separation

 

 

 

INTRODUCTION

Magnetic separation is one of major physical separations, which takes advantage of differences of magnetism between minerals. However, in some circumstances the magnetism of minerals or the magnetic differences between minerals are too small to magnetically recover or separate. There are two types of magnetization:reduction-roasting magnetization and surface- magnetization. The former is a common way to enhance mineral’s volume-magnetism by roasting-reduction, and is feasible technically, but it is energy-intensive and resultantly costly, and also brings about environmental problems. The latter, surface-magnetization, as a way to enhance surfacial magnetism, therefore has attracted some attentions in recent years, and it mainly includes: alkali-leaching magnetization, electro chemical magnetization; hydrophobic magnetization and magnetic seeding magnetization. Relatively speaking, the last one, magnetic seeding magnetization, is dominant among these surfacial magnetization methods (Chao lan Jiang, 1994).

There are quite a few publications about recovering minerals by high-gradient magnetic separator (HGMS) after magnetic seeding magnetization, such as recovering phosphate and heavy metals particles from waste-water (Y. Terashima et al., 1984; Ahamda M. H. Shaikh et al., 1992; D. Feng et al., 2000; Nuray Karapinar, 2003), and separation of weakly magnetic minerals and removal of magnetic impurity from non-magnetic minerals (J. Y. Hwang et al., 1982; P. Parsonage, 1988; Q. Y. Song et al., 1992; Qingxia Liu et al., 1994). The magnetic seeds all of these researches above used were mineral magnetite grains. By adjusting pH and adding surfactant, magnetic seeds, which have been ground into fine particulates, adsorb on surface of target minerals, and then the minerals would be separated by HGMS. In mineral processing field, the method of magnetic seeding magnetization was also used to recover fine iron mineral particles and tiny magnesite (S. Prakash et al., 1998; Georgios N. Anastassakis., 1999, 2002). As a special method to separate minerals, magnetic seeding magnetization is restricted by some factors as the magnetic seeds (crude mineral magnetite) must be very fine and have very high purity.

 

As far as the fineness and high purity of magnetic seeds are concerned, synthetically magnetic particles (Fe₃O₄) are a quite good alternative to the crude mineral magnetite. Hwang used synthetical magnetite which was coated with surfactant to separate rutile from quartz and calcite from quartz rather well (J. Y. Hwang, 1989, 1990, 2002). Xing (Wei-zhong Xing, 1994) reported that limonite ‘s recovery reached 95% with synthetical magnetic seeding at a dosage of 5 kg/t under a magnetic field intensity of 0.3T. Normally, the synthetical magnetic seeds are prepared by mixing Fe²⁺ and Fe³⁺ at certain proportion in the presence of ammonia in an aqueous solution. In this way, the resultant synthetical magnetic particles are very fine in size and have a strong magnetism. Like the magnetization with finely ground crude mineral magnetite, so far the magnetization of synthetical magnetic seeds has not successfully applied in industry, either.

 

Because of the low specific-susceptibility, main way to process weakly magnetic minerals is high-intensity magnetic separator (as high as 1 .5T). But under such high-intensity magnetic field, the separator would lose selectivity and lard gangue minerals with valuable minerals. As reported, the surfacial-magnetism of weakly magnetic minerals can be improved by chemical measures so that they can be recovered by relatively low magnetic field with a result of improved concentrate grade. The main purpose of this paper is to introduce a promising method to magnetize weakly magnetic minerals directly in slurry. Synthetical magnetic particles adsorb on limonite which has a low specific-susceptibility = 2 x i0* 3 x i0 m3/kg) and magnetize it. Magnetized limonite can be recovered in a quite low magnetic field. The essential of this technology is that magnetic particles are synthesized in slurry by adding metal ion salt and adsorb on limonite surface selectively, and effectively enhance the magnetism. Compared with the previous surfacial-magnetization techniques, this method doesn’t need particular steps for preparing magnetic seeds before magnetization process.

 

 

EXPERIMENTAL MATERIALS AND METHODS

The weakly magnetic mineral used in this experiment is limonite (Fe₂O₃.nH₂O), having a Fe content of 55.0%. Quartz is used as a gangue mineral, containing SiO₂ 99.0%. Both of the samples are 95% — 75um in size.

 

FeCl₃.6H₂O, FeCl₂4H₂O and oleic acid used in this experiment are all analytical reagents. Aqueous ammonia contains ammonia of 25% by volume.

 

Mineral was dispersed in water and slurry containing 6.5% solids was formed. Mixed salt solution containing Fe²⁺ and Fe³⁺ was added into the slurry, then adding ammonia with stirring; thereafter the slurry passed through a matrix of HGMS to recover magnetized minerals in order to assess the magnetization. The HGMS used has a matrix of dentate-board configuration, and tooth pitch between two boards is about 3mm, 2.7mm in same board. The maxmum magnetic-flux density of the HGMS is about 16.000Gs, but only 3.500Gs was used in this experiment.

 

 

RESULTS AND DISCUSSION

According to its molecular formula, magnetic particle (FeO.Fe₂O₃) contains two valences of iron ions, Fe²⁺ and Fe³⁺. Basically, it can be synthesized in an aqueous solution, and there is a chemical reaction below:

Fe²⁺ +2Fe³⁺ +8OH^-—Fe₃O₄ +4H₂O (1)

Hence, two types of iron salts (ferrous chloride and ferric chloride) are dissolved in water in certain proportion. After dding alkali solution, magnetic particles Fe₃O₄ should be obtained as precipitators (Ben-lan Lin et al., 2005).

As to this research, the chemical reaction will happen in slurry rather than in an aqueous solution, so it will be more complicated. Several factors, such as dosages of reactants, slurry temperature and pH value, addition of surfactants, which would affect the synthesis and adsorption of magnetic particles, were investigated.

Ratio of Fe³⁺ to Fe²⁺

Two types of iron ions (Fe²⁺ and Fe³⁺ )are basic components of magnetic seeds. It can be seen from the chemical equation (1) that, theoretical value of Fe3⁺/Fe²⁺ molar ratio in a magnetic seed molecule is 2 tol. In sluny, however, reaction and magnetization might be very different from in a pure aqueous system. Under a set of determined conditions (pH10.5, temperature 15 ^oC, etc.), the effect of molar ratio of Fe³⁺ to Fe²⁺ on limonite magnetization are shown in Table 1. When the molar ratio of Fe³⁺ to Fe²⁺ is at the theoretical value (2 to 1) the recovery of limonite is only about 60% while the maximal recovery of limonite reaches 68.0% at the molar ratio 1 to 3.

 

Table 1 Effect of molar ratio of iron ions on limonite recovery

 

The disparity between the theoretical and tested values for the molar ratio of Fe³⁺ to Fe²⁺ might attribute to easy oxidization of Fe²⁺ into Fe³⁺ (Ping-mm Zhang, 2002):

4Fe²⁺ +O₂ +2H₂O —4Fe³⁺ +4OH^- (2)

So, in this reaction system Fe²⁺ would be theoretically superfluous to counteract its oxidation.

 

Dosage of Iron ions

After determining the molar ratio of Fe³⁺ /Fe²⁺, dosage of Iron ions was tested at the ratioi 1:3 and denoted only by Fe³⁺ concentration in slurry. Results are shown in Fig. 1.

 

 

Fig. 1 Effect of iron ion dosage on limonite recovery

Limonite recovery increased with the addition of iron ions. The recovery arrived at 77.2% at Fe dosage of 2.5 x 10^-3 mol/L, and then became slow-moving upward. The maximal recovery 91.6% was obtained at Fe³⁺ concentration of 7.5 x 10^-3mo1/L. This is due to formation of more magnetic particles with the increase of iron ions. So, the Fe³⁺ dosage was set at 2.5 x 10^-3 mol/L in following tests.

 

The effect of ammonia

It is favorable to obtain Fe₃O₄ at pH values above 9.2 (Li-xian Song et al., 2006). Ammonia acts as a precipitator in this reaction, and plays a key role in the process of producing magnetite. In our preliminary test it was found that ammonia was much better than sodium hydroxide solution, so ammonia was chosen as an alkali reactant. Effects of ammonia on limonite recovery and slurry pH are shown in Fig. 2. The recovery of limonite did not necessarily correspond linearly with the slurry pH, especially at high pH range, and at pH10.5 when ammonia dosage was at 0.2mol/L it reached the highest value of 77.0%, and then higher slurry pH deteriorated the magnetization. Ammonia has a buffer function in slurry. This function makes pH value increase slowly with increase of ammonia in high pH range.

 

 

 

 

 

Fig. 2 Effect of ammonia on limonite recovery and slurry pH

Temperature

Temperature is also an influential factor on limonite magnetization in slurry system, as shown in Fig. 3. Limonite recovery increased from 68% to 76% with temperature rising from 15°C to 45°C, and had not dropped obviously until 75°C. It is reported that in an aqueous solution temperature has effects on reaction equation (1) in terms of the average diameter and number of magnetic particles. The average diameter increases with the temperature rising, and the size in 30°C is half of that in 45°C while the number of magnetic particles decreases with the increasing of average diameter size (Runhua Qin et al., 2003; Xin Zhang et al., 2006); on the other hand at higher temperature Fe2 is oxidized more readily. Therefore, probably the main effect of slurry temperature was on the number and size of magnetic particles formed as well.

 

 

 

 

Fig. 3 Effect of slurry temperature on limonite recovery

Selective magnetization

How to make magnetic particles adsorbing selectively on limonite rather than on gangue, namely, selective magnetization, is the most important thing in surfacial magnetization. Our preparatory test indicated that surfactant sodium oleate had a positive role to promote the selectivity between minerals limonite and quartz. The effect of sodium oleate on recovery of limonite and quart is shown in Fig. 4. It can be seen that in the case of absence of the surfactant there is little difference in recoveries of limonite and quartz; after adding sodium oleate, the recoveries of both of the minerals declined but for quartz much more sharply than for limonite, and the biggest recovery gap between limonite and quartz was 52% when sodium oleate dosage is at 3.3x10^-4mo1/L; in addition, with the increase of sodium oleate addition there is very special pattern of recovery variation curve for limonite: after initial drop, there is a rise, and then it drops again.

 

 

 

 

Fig. 4 Effect of oleate sodium addition on recovery of limonite and quartz

Fundamentally, the adsorption of magnetic seeds on the minerals without addition of the surfactant is categorized as physical adherence. As synthetically magnetic particles are very fine in size and have a big surface energy, they adsorb easily on the surface of larger mineral particles without obvious selectivity. Photos of limonite and quartz before and after magnetization were taken by High-power Scan Photomicrography, as shown in Fig. 5. It is observed that the particulates of limonite and quartz in slurry are dispersant before magnetization, and after magnetization magnetic seeds adsorb on their surface and they aggregate.

The introduction of the surfactant into slurry magnetization system could result in competitive adsorption among magnetic seeds, minerals and surfactant ions. As well known, oleic acid is a common flotation collector for iron oxide ores, and oleate ion is prone to adsorb on mineral limonite and magnetic seeds while without a cationic ions as an activator, oleate ions don’t adsorb on gangue quartz. It is these adsorption differences of oleic ion that result in the selective magnetization with the aid of sodium oleate. As far as the special recovery curve of mineral limonite with increase of sodium oleate is concerned, it can be explained in following way: at initial small amount of sodium oleate addition the competitive adsorption of magnetic seeds between limonite and oleate ions consumes some seeds; then when oleate ion concentration exceeds Half Micelle Concentration (HMC 1.8 x 10^-4mo1/L), long hydrocarbon chains of oleate ion adsorbed respectively on limonite and magnetic particles would have an association action, consequently strengthening the magnetization of limonite. So, limonite got the maximal recovery at sodium oleate dosage of 3.3x10^-4mol/L. when sodium oleate dosage was at 5 x 1 0^-4mo1/L, exceeding Critical Micelle Concentration (CMC 4.1x10^-4mol/L), oleate ion in slurry would be in a form of micelle and did not adsorb on minerals (Dian-zuo Wang, 1985). It can be seen from Fig. 4 that limonite recovery declined at high dosage of 5x10^-4mol/L.

 

CONCLUSIONS

There are quite a few ways to magnetize weakly magnetic mineral surface, but magnetization occurring in slurry can cut down needless steps. According to the test results above, several conclusions can be drawn:

1.       Magnetization of weakly magnetic mineral particles with iron salts and ammonia was realized directly in slurry, and magnetized mineral particulates tend to aggregate.

2. According to the recovery of limonite, the optimal conditions for limonite magnetization were: the molar ratio of Fe³⁺ to Fe²⁺ 1:3, temperature 45°C, pH 10.5. The recovery increased with increasing Fe³⁺ dosage, reaching 91.6% at Fe³⁺ dosage of 7.5x10^-3mol/L under the magnetic flux density of 3.500 Gs.

3. with the aid of the surfactant sodium oleate, synthetical magnetic particles was able to selectively magnetize limonite from quartz in slurry. The biggest recovery gap between limonite and quartz happened at the sodium oleate dosage of 3.3x10^-4mo1/L, i.e. between HMC to CMC. The principle of this selective magnetization may lie in that oleate ions adsorb competitively between magnetic seeds and minerals limonite or quartz, and that long hydrocarbon chain of oleate ion adsorbed on limonite and magnetic seeds could have a mutual association.

REFERENCES

Ahamda M.H.Shaikh, S G Dixit, 1992. Removal of phosphate from waters by precipitation and high gradient magnetic separation, Wat Res, 845-852.

 

 

Ben-Ian Lin, Xiao-dong Shen, Shen Cui, 2005 Preparation of Fe304 nanoparticle by liquid—phase coprecipitation, Inorganic Chemicals Industry, 37: 24-25.

Chao-lan Jiang, 1994. Theory and techniques of Magnetic Separation, pp67-68. (Metallurgy Industry Press, Beijing).

 

 

D Feng, C Aldrich, H Tan, 2000. Removal of heavy metal ions by carrier magnetic separation of adsorptive particulates, Hydrometallurgy, 56: 359-3 68.

Dian-zuo Wang, 1985. Solution Chemisty of Floatation, pp5 1- 53. (Hunan Sci and Tech Press. Changsha, China).

 

 

Georgios N Anastassakis, 1999. A study on the separation of magnesite fines by magnetic carrier methods, Colloids Surfaces A: Physicochem.Eng.Aspects, 149: 585-593.

 

 

Georgios N Anastassakis, 2002. Separation of Fine Mineral Particles by Selective Magnetic Coating, Journal of Colloid and Interface Science 256: 114-120.

 

 

Li-xian Song, Zhong-yuan Lu, De-chun Liu, Shao-bo Cui, Xiang-qi Xiao, 2006. Preparation of magnetic nanometer Fe3 04 with the method of complex compound-hydrolyzation deposition, Chemical Industry and Engineering Progress, 25(1): 54-57.

 

 

J Y Hwang, G Kullerud, M Takayasu, F J Friedlaender, P C Wankat, 1982. Selective seeding for magnetic separation, IEEE Trans on Magn, MAG-18(6): 1689-169 1.

J Y Hwang, 1989. Reagents for magnetizing nonmagnetic materials, US Patent. 4834898.

J Y Hwang, 1990. Method of magnetizing nonmagnetic materials, US Patent. 4906382.

J Y Hwang, 2002. Adsorption of surfactant dispersed nanometer

magnetite, Journal of Minerals & Materials Characterization & Engineering, 1(2): 13 1-140.

 

 

Nuray Karapinar, 2003. Magnetic separation of ferrihydrite from wastewater by magnetic seeding and high-gradient magnetic separation, mt. J. Miner. Process, 71: 45-54.

P Parsonage, 1988. Principles of mineral separation by selective magnetic coating, mt. J. Miner. Process, 24: 269-293.

Ping-mm Zhang, 2002. Advanced Chemistry for Engineering, pp 100-102. (Hunan Education Press, Changsha, China).

Q Y Song, F Xu, 1992, Magnetic seeding flocculation of weakly magnetic iron minerals, mt. J. Miner. Process, 34: 219-229.

 

 

Qingxia Liu, F J Friedlaender, 1994. Selective collection of Non-magnetic rutile and quartz by means of a magnetic reagent by HGMS, IEEE Trans on Magn, 30(6): 4668-4670.

 

 

Run-hua Qin, Wei Jiang, Hong-ying Liu, Feng-sheng Li, 2003. Preparation and Characterization of Nanometer Magnetite. Materials Review, 17: 66-68.

 

 

S Prakash, B Das, J K Mohanty, R Venugopal, I 999.The recovery of fme iron minerals from quartz and corundum mixtures using selective magnetic coating, mt. J. Miner. Process, 57: 87-103.

Wei-zhong Xing, 1994. Theory and Practice of Magnetic Seeding Separation, 21. (Metallurgy Industry Press, Beijing).

 

 

Xin Zhang, Xin-gang Li, Bin Jiang, 2006. Preparation and Characterization of Nanometer Magnetite. Chemical Industry and Engineering, 23(1): 54-5 8.

 

 

Y Terashima, H Ozaki, M Sekine, 1984. Removal of dissolved heavy metals by chemical coagulation, magnetic seeding and high gradient magnetic filtration, Wat Res, 537- 545.

Fig. 5 Topography of limonite and quartz before and after magnetization