Journal of Materials Research and Technology Journal of Materials Research and Technology
Original Article
Recovery of kaolinite from tailings of Zonouz kaolin-washing plant by flotation-flocculation method
Kianoush Barani, , Masoud Kalantari
Department of Mining Engineering, Lorestan University, Khorramabad, Iran
Received 23 October 2016, Accepted 10 May 2017

The traditional processing of kaolin is achieved by dispersion of the mined ore and classification by multistage hydrocyclone plants. The inefficiencies inherent to cyclones produce a middling product that is commonly disposed back into the quarry. In this research, recovery of kaolinite from tailings of Zonouz kaolin washing plant, which is located in Iran was investigated by flotation and flotation- flocculation. Flotation experiments show that the flotation of kaolinite from the tailings is better in an acidic than in an alkaline medium containing cationic collectors. Flotation under acidic condition causes problems such as equipment corrosion at industrial scale. As a result, the cationic flotation of kaolinite is enhanced by addition of polyacrylamide as a flocculant. The results showed flocculation by polyacrylamide improved flotation of kaolinite within a range of pH. With 300g/t dodecylamine, 500g/t aluminum chloride, 50g/t pine oil (frother), 15g/t polyacrylamide, at pH=7 and without de-slimming a product has 37.19% Al2O3, 54.19% SiO2 and 34.43% mass recovery was archived.

Kaolinite, Flotation, Flocculation, Cetylpyridinium chloride, Dodecylamine, Aluminum chloride, Polyacrylamide
This article is only available in PDF

In general, kaolin is upgraded by physical methods such as water-washing or froth flotation. Kaolin particles are predominantly below 25μm and are flat shaped particles. The industrial separation is commonly achieved by the dispersion of the mined ore and classification using multistage hydrocyclones. The kaolin reports to the overflow which is then reprocessed in other smaller cyclones. Due to inefficiencies associated with hydrocycloning, most plants are not able to produce clean kaolinite products and after the 4th or 5th hydrocyclone stage, the underflow cannot be separated further. The underflow is disposed as waste or sold cheaply to the cement industry and brick-manufacturing or used as back fill in the quarry [1–5].

Zonouz mine, which is located in eastern Azerbaijan Province, has the only kaolin-washing plant in Iran. The plant has a considerable amount of tailings containing about 15% Al2O3. The tailings mineralogical constituents are quartz, kaolin and iron oxide. The tailing has fine particles (100% finer than 100μm and about 30% finer than 20μm); the fine particles can be easily dispersed by wind and settle down by rain and snow. Without treatment or recycling, the tailing would cause environmental pollution. It is important to re-evaluate the tailings because of economic and environmental causes. The evaluation will be economically feasible since it will not require additional mining cost. Kaolinite can be retrieved from the tailings by using flotation. However, flotation of kaolinite from the tailing is difficult because kaolinite and quartz, the two main minerals in the tailings, have the same flotation behavior.

Point of zero charge of quartz is at pH=1.8 and zeta-potential of kaolinite is negative in a wide range of pH values, then both minerals can be floated by amine collectors in acidic and alkaline medium [6–8].

Many studies have been carried out on kaolinite flotation (pure samples) by cationic collectors such as ether diamine, hydroxyl-alkyl-diethanolamine, Cetylpyridinium chloride, cetyltrimethylammonium bromide, and dodecylamine. The results of these studies showed that kaolinite flotation by cationic collectors is better in acidic an environment than in a alkaline environment [7–13]. Yuehua et al. [11] tried to increase kaolinite recovery in alkaline medium by using a macromolecule as a flocculant. Without polyacrylamide, the kaolinite recovery decreased from 70 to 15%, with increasing the pH from 2 to 10. But, with polyacrylamide as a flocculant, kaolinite recovery increased under both acidic and alkaline environments [11].

Comparison of flotation of kaolinite and quartz show that two collectors (dodecylamine and EDA Flotigam) and two quaternary ammonium salts (Cetylpyridinium chloride and cetyltrimethylammonium bromide) are among the most effective collectors for floating both kaolinite and quartz. But, the time required for kaolinite flotation is 5–10 times that for the quartz flotation, depending on the type of the collector [10].

The purpose of the present study is to investigate the recovery of kaolinite from tailings of kaolin washing plants using flotation and flotation-flocculation.


Samples used in this study were taken from the tailing dam of Zonouz kaolin washing plant in East Azerbaijan Province, Iran. The particle size distribution of the representative sample was carried out by laser diffraction (Fig. 1). It is observed that the percent passing 100μm and 50μm is respectively 100% and 50% and more than 30% of the particles were smaller than 20μm (slime particles).

Fig. 1.

Particle size distribution of the tailing sample by laser diffraction.

According to the mineralogical analyses by X-ray diffraction, it was found that the sample contains quartz, kaolinite and calcite as the main minerals, and muscovite, biotite, montmorillonite and iron-rich epidote as the minor minerals. Chemical composition of the sample is given in Table 1.

Table 1.

Chemical composition of the sample.

Composition  Composition 
SiO2  74.67  CaO  1.74 
Al2O3  15.35  SrO  >0.01 
K20.25  P2O5  >0.01 
SO3  0.11  MgO  >0.01 
Fe2O3  0.33  loss on ignition  7.54 
2.2Flotation experiments

Flotation experiments were carried out in a 1000cm3 conventional laboratory cell (Denver model), at 25% solids ratio and agitation rate was 850rpm. The suspension was added to the flotation cell and conditioned for 5min. The pH was adjusted to a desired value by adding NaOH or H2SO4. Cetylpyridinium chloride (CPCI) or dodecylamine (DDA) solution was then added as the collector and the suspension was conditioned for 3min. Aluminum chloride was added as the depressant of quartz and the suspension was conditioned for 3min. Finally pin oil (150g/ton) was added as the frother then after 3min air was delivered to the cell for a total of 8min and the concentrate which was floated and remained in the suspension was collected, filtered and dried in a well-ventilated oven at 120°C. In the flotation- flocculation experiments, before adding the collector, polyacrylamide (PAM) was added as the flocculant for kaolinite particles and the suspension was conditioned for additional 3min. All reagents used in this study were of analytical grade and supplied by Merck company.

The experiments were conducted in 8 series. In each series, the effect of only one parameter was examined and, the other parameters were kept constant

3Results and discussion3.1Effect of pH with CPCI as collector (series 1)

The first series of the experimental investigation was conducted to evaluate the effect of pH on the flotation results when CPCI was used as a collector. The results are illustrated in Fig. 2. It can be seen, with increasing pH from 3 to 9, the Al2O3 recovery decreased and SiO2 recovery increased. Al2O3 and SiO2 assay were respectively assumed as the kaolinite and quartz indicator recovery.

Fig. 2.

Flotation recovery of Al2O3 and SiO2 as a function of pH with CPCI as collector (CPCI=400g/t, without depressant and de-slimming).

Fig. 3 shows the structure of kaolinite crystal. At low pH (pH<4) the layer to layer interaction between the alumina face surface (001¯), and the silica face surface (001) is significant, as well as the interaction between the alumina face surface and the edge surface (010). In this case, the cluster structures, and even more complex network structures, are formed by kaolinite particle aggregation. A simple drawing in Fig. 4 illustrates the adsorption of cationic collector by the kaolinite cluster at low pH (pH<4), forming a hydrophobic structure.

Fig. 3.

Structure of kaolinite [9].

Fig. 4.

Kaolinite cluster adsorption of cationic collector at low pH, dark surfaces represent the silica face surfaces of kaolinite particles, and gray surfaces represent the alumina face surfaces of kaolinite particles [13].

In alkaline medium, the interaction between a cationic collector and the plane (001), which contains oxygen, was stronger than that of the plane (001¯), which contains Al. The plane (001) adsorbed hydrophobic collector; but, the plane (001¯) did not adsorb collector due to having Al, and remained hydrophilic. As a result, collectors of the plane (001) were accumulated and formed structures whose outer surface was hydrophilic (Fig. 5). Therefore, as indicated in Fig. 2, kaolinite recovery decreased with increasing pH. A pH of 3 was chosen as the optimum condition and for further works.

Fig. 5.

Schematic description of the kaolinite–amines interactions [9].

3.2Effect of concentration of CPCI collector (series 2)

The maximum recovery obtained for Al2O3 in the first series was around 45% which was significantly low. For increasing the recovery of Al2O3, the effect of concentration of CPCI on the flotation results was investigated. Also, the slime particles below 38μm were removed and aluminum chloride was used as the depressant.

The results in Fig. 6 indicate that the Al2O3 recovery increased with the increase concentration of CPCI up until a value of 600g/t. The Al2O3 recovery at a 600g/t of CPCI exceeded 85%. By more increasing in the concentration of CPCI to 700g/t, Al2O3 recovery was decreased to 81%, probably because the critical micelle concentration phenomenon or bubble armouring were occurring. Although, more research and evidence are needed to prove it. Also, it is observed that the SiO2 recovery was constant at about 20%. It can be attributed to depress of quartz particles by aluminum chloride. The maximum separation efficiency (81%) was achieved at 600g/t of CPCI.

Fig. 6.

Flotation recovery of Al2O3 and SiO2 as a function of CPCI concentration (pH=3, aluminum chloride depressant=300g/t, de-slimming −38μm).

3.3Effect of aluminum chloride concentration in flotation by CPCI collector (series 3)

The results of series 2 showed that the Al2O3 recovery and separation efficiency were increased by using aluminum chloride as a quartz depressant. In this series the effect and the role of aluminum chloride was investigated. Fig. 7 indicates that with increasing the concentration of aluminum chloride from 300 to 600g/t, significant changes were not observed in Al2O3 and SiO2 recoveries. However, the best result was related to aluminum chloride value of 500g/t which produced a concentrate with recovery of 86.28% Al2O3 and 20.19% SiO2 and assay of 38.42% Al2O3 and 52.96% SiO2 with separation efficiency of 83.51%.

Fig. 7.

Flotation recovery of Al2O3 and SiO2 as a function of aluminum chloride concentration (pH=3, CPCI=600g/t, de-slimming −38μm).

Ions of aluminum chloride (Al3+) play two positive roles during the flotation. From one side, they act as the depressant of quartz. In fact, these ions sit on the surface of quartz, which has negative charges and competes with the collector ions and prevent quartz flotation. However, the excessive increased level of Al+3 increases the kaolin zeta potential. On the other side, these ions increase the ionic strength of the pulp. In a high ionic strength environment, although the negative charge on the basal of planes (001¯) and (001) is constant, the zeta potential caused by the surface charge on both basal planes and edges is significantly screened and thus aggregation of kaolinite particles can occur (Fig. 8). The screened surface charge at high ionic strength also weakens the self-aggregation between the negatively charged (001¯) planes and the positively charged edge planes in acidic pulp. Thus the pH dependence of kaolinite flotation becomes less pronounced.

Fig. 8.

Aggregation of kaolinite particles in a high ionic strength environment [9].

3.4Effect of pH on the flotation with DDA collector (series 4)

Previous studies showed, dodecylamine (DDA) is an effective collector for floating both kaolinite and quartz [10]. In this series, flotation of kaolinite by DDA in the presence of aluminum chloride as quartz depressant was evaluated. Fig. 9 shows that during kaolinite flotation with DDA collector, with increasing pH, both recoveries of Al2O3 and SiO2 were increased. The maximum recovery of Al2O3 and SiO2 was obtained at pH=9. It seems, there are differences between the DDA and CPCI results. As it can be seen in Fig. 2, by using CPCI, with increasing pH, Al2O3 recovery was decreased. Effect of pH on selectivity separation is clearly shown in Fig. 9, so that at pH=3, the difference in Al2O3 and SiO2 recovery was around 28% and good selectivity was achieved; with increasing pH, this difference was reduced and recoveries of Al2O3 and SiO2 were almost equal at pH=9. The best results were achieved at pH=3 where a concentrate with recovery of 66.25% Al2O3 and 37.74%SiO2, respectively, and assays of 20.84% Al2O3 and 69.93%SiO2 with separation efficiency of 35.61% was obtained.

Fig. 9.

Flotation recovery of Al2O3 and SiO2 as a function of pH with DDA as a collector (DDA=600g/t, aluminum, chloride=300g/t, de-slimming −38μm).

3.5Effect of concentration of DDA collector (series 5)

The results of series 4 showed satisfactory separation of kaolinite and quartz by the DDA collector in the presence of aluminum chloride can be approached in acidic pHs, but high dosage of the collector is required. In this series the effect of DDA concentration was studied.

Fig. 10 indicates that the maximum recovery of Al2O3 at a concentration of 300g/t was equal to 80%; but, increasing DDA concentration decreased assay of Al2O3. Also, SiO2 recovery increased from 20% to 38% by increasing in DDA concentration. The best flotation conditions in term of assay and recovery of Al2O3 and separation efficiency was resulted in the collector concentration of 300g/t. In this condition, a concentrate with a assay of 36% Al2O3 and 52% SiO2 and recovery of 80% Al2O3 and 21% SiO2 with separation efficiency of 75.48% was obtained.

Fig. 10.

Flotation recovery of Al2O3 and SiO2 as a function of DDA concentration (pH=3, aluminum, chloride=300g/t, de-slimming −20μm).

3.6Effect of pH on the flotation with DDA collector and PAM flocculant (series 6)

The results from series 1–5 showed that kaolinite flotation by DDA and CPCI collectors is better in an acidic environment than in an alkaline environment. This is consistent with the observations reported by several researchers [7,9–12]. Flotation under acidic condition causes problems such as equipment erosion at industrial scale. However, the results of study by Yuehua et al. [11] showed the recovery of kaolinite in alkaline environments can be improved in the presence of polyacrylamide (a macromolecule) as a flocculant [11], although the effect of the macromolecule on quartz flotation and separation of kaolinite from quartz was not studied.

In this series, the effect of pH on the flotation with the DDA collector in the presence of polyacrylamide (PAM) was studied. Fig. 11 indicates that with increasing pH from 3 to 7, the recovery of Al2O3 slightly was increased from 78 to 79%; but, with increasing pH from 7 to 9 decreased the Al2O3 recovery. SiO2 recovery also increased with increasing the pH; the lowest SiO2 recovery (20%) was related to pH=3.

Fig. 11.

Flotation recovery of Al2O3 and SiO2 as a function of pH With DDA as collector and PAM as flocculant (DDA=300g/t, PAM=25g/t, aluminum, chloride=300g/t, de-slimming −20μm).

Generally, it was observed that with the PAM flocculant, Al2O3 recovery was at an acceptable level within a wide pH range, while SiO2 recovery was low. PAM contains O and COO groups which are expected to bond with the surface of hydroxyl groups. As shown in Fig. 3, in kaolinite minerals, the plane (001¯) has OH group, therefore PAM establishes hydrogen bond with these planes; thus, kaolinite particles can be accumulated by this flocculant. On the other hand, as mentioned earlier, cationic collectors are well absorbed by the planes (001); therefore, cationic flotation can be improved by using PAM. Fig. 12 shows a schematic diagram of kaolinite flotation and flocculation by using PAM. The best results were related to pH=7 where a concentrate with recovery of 78.75% Al2O3 and 25.51% SiO2 and assay of 32.49% Al2O3 and 57.87% SiO2 with separation efficiency 68.18% was obtained.

Fig. 12.

Schematic model of flocculant-enhanced flotation of kaolinite [11].

3.7Effect of PAM on the flotation with DDA as collector (series 7)

Presence of slime increases consumption of chemical reagents and decrease collector's selectivity. In order to increase the recovery in the previous series of 1 to 6, the slime particles below 38μm or 20μm were removed from the initial feed by screening or settling, where this portion constituted a significant amount of the total initial feed.

In this part without de-slimming, the effect of PAM on the flotation with the DDA as a collector was studied. Fig. 13 shows that increasing PAM concentration decreased Al2O3 recovery and increased SiO2 recovery. The best result was obtained at a concentration of 15g/t where Al2O3 recovery and assay in the concentrate were respectively 80.41% and 34.64% and SiO2 recovery and assay in the concentrate were respectively 25.58% and 56.66% with separation efficiency of 72.84%. Comparison of the results of this series with previous series showed that significantly small concentration of PAM is required to improve the flotation.

Fig. 13.

Flotation recovery of Al2O3 and SiO2 as a function of PAM with DDA as collector (DDA=300g/t, pH=7, aluminum, chloride=300g/t, without de-slimming).

3.8Effect of aluminum chloride concentration in flotation by DDA collector and PAM flocculant (series 8)

The purpose of this series, was to find the best condition for separation of kaolinite at pH=7 without any de-sliming.

Fig. 14 indicates that an increase in the concentration of depressant from 300 to 500g/t increased the Al2O3 recovery and decreased the SiO2 recovery; however, the condition did not change considerably with further increase in the concentration from 500 to 600g/t and even Al2O3 recovery deceased from 83% to 82%. Therefore, the best condition for assay and recovery in the concentrate was obtained at a concentration of 500g/ton. At this dosage, recovery and assay of Al2O3 in the concentrate were respectively 83.02% and 37.19% and SiO2 recovery and assay in the concentrate were respectively 22.28% and 54.38% with the separation efficiency of 78.48%.

Fig. 14.

Flotation recovery of Al2O3 and SiO2 as a function of aluminum chloride concentration with (DDA=300g/t, pH=7, PAM=15g/t, without de-slimming).


Results showed that for both CPCI and DDA collectors, the best condition was obtained at acidic pHs. Similar results were achieved by other researchers [7,11]. Comparing the results of the two collectors showed that CPCI collector has better selectivity. The results reported by Köster et al. also confirmed better selectivity of CPCI for kaolinite compared with quartz and feldespat at acidic pHs [7]. Depressant of aluminum chloride had considerable effect on the flotation performance, which is an effective depressant for the quartz and enhanced flotation of kaolinite. The result also agree well with reported studies [7,9]. Köster et al. showed aluminum chloride is a good depressant for quartz and feldspar in presence of kaolinite [7]. Ma et al. demonstrated that flotation of kaolinite enhanced by adding NaCl solutions because the ionic strength of solution was increased [9]. Flotation under acidic condition of pH=3 causes problems such as equipment erosion at industrial scale. For this reason, PAM flocculant was used. Results with PAM showed that this flocculant increased the efficiency of kaolinite flotation at neutral and alkaline pHs. In the last series of the experiments, a significant result was achieved and by using only 15g/t PAM at neutral pH condition, a concentrate with Al2O3 recovery and assay of respectively 83.02% and 37.19% and SiO2 recovery and assay of respectively 22.28% and 54.38% was obtained. While, the assay for Al2O3 and SiO2 of the current concentrate in Zonouz plant are respectively 26% and 62%. This means that using the flocculation and flotation processes at neutral pH, almost 35% of the tail weight can be turned into a concentrate with acceptable assays. On the other hand, Al2O3 and SiO2 assays of the remaining tail were respectively 15% and 74%, where the tail has significantly high SiO2 assay that appropriate applications can be predicted for it as silica concentrate.

Conflicts of interest

The authors declare no conflicts of interest.


The authors of this article would like to thank Zonouz kaolin-washing plant and Lorestan University for funding this research work.

H.H. Murray
Applied clay mineralogy: occurrences, processing and applications of kaolins, bentonites, palygorskitesepiolite, and common clays
Elsevier, (2006)
M. Valášková,G.S. Martynková
Clay minerals in nature: their characterization, modification and application
InTech, (2012)
H.H. Murray
Major kaolin processing developments
Int J Miner Process, 7 (1980), pp. 263-274
R.J. Pruett
Feature articles – Georgia kaolin: development of a leading industrial mineral
Mining Eng, 52 (2000), pp. 21-27
J.E. Kogel
The Georgia kaolins: geology and utilization
SME, (2002)
M.C. Fuerstenau,J.D. Miller,M.C. Kuhn
Chemistry of flotation
American Institute of Mining, Metallurgical and Petroleum Engineers. Inc., (1985)pp. 132
R. Köster,B. Schreck,V. Rybinskil,B. Dobias
New reagent systems for the flotation of kaolinite
Miner Eng, 5 (1992), pp. 445-456
H. Jiang,Z. Sun,L. Xu,Y. Hu,K. Huang,S. Zhu
A comparison study of the flotation and adsorption behaviors of diaspore and kaolinite with quaternary ammonium collectors
Miner Eng, 65 (2014), pp. 124-129
X. Ma,W.J. Bruckard,R. Holmes
Effect of collector, pH and ionic strength on the cationic flotation of kaolinite
Int J Miner Process, 93 (2009), pp. 54-58
O.M.S. Rodrigues,A.E.C. Peres,A.H. Martins,C.A. Pereira
Kaolinite and hematite flotation separation using etheramine and ammonium quaternary salts
Miner Eng, 40 (2013), pp. 12-15
H. Yuehuaa,S. Wei,L. Haipu,Z. Xu
Role of macromolecules in kaolinite flotation
Miner Eng, 17 (2004), pp. 1017-1022
C. Liu,Y. Hu,X. Cao
Substituent effects in kaolinite flotation using dodecyl tertiary amines
J. Liu,X. Wang,C.-L. Lin,J.D. Miller
Significance of particle aggregation in the reverse flotation of kaolinite from bauxite ore
Miner Eng, 78 (2015), pp. 58-65
Copyright © 2017. Brazilian Metallurgical, Materials and Mining Association