Purification of kaolin by selective flocculation

---

 

243 

Purification of kaolin by selective flocculation 

A. B. Luz and A. Middea.  

Center for Mineral Technology-CETEM 

Av. Ipê , n

o

 900, Ilha da Cidade Universitária 

21941-590-Rio de Janeiro, RJ, Brasil 

ABSTRACT  

Kaolin clay is heavily used in the paper industry as a coating and a filler, however 

to attend the specifications required by the industry it has to be processed. Kaolin is also 

used as a filler and pigment in various materials, such as paint, plastic, rubber etc. Some 

crude kaolins as mined in Northern Brazil, contain colored impurities such as iron oxide 

and rutile/anatase. These minerals normally are stained by iron and as a result vary from 

yellow to dark brown in color. The kaolin beneficiation process used in that region 

consists mainly of degritting, fractionation by centrifuge, high gradient magnetic 

separation, bleaching with sodium dithionite, filtering and drying. In some kaolin clays 

from that region, titanium oxide mineral occurs in fine fraction, lower than 2 

µm, and so 

making the industrial process more difficult.  

The conventional method of classification and bleaching with sodium dithionite, 

markedly improves kaolin brightness, but has little effect on anatase removal. In fact, the 

fine fractions resulting from centrifugation, in many cases contain more TiO

2

 minerals 

than the original whole clay or the coarse fraction. 

The present work is related to purification of Brasilian kaolin clay using selective 

flocculation for removing its colored impurities such as rutile and anatase. This study 

was conducted in bench scale and it consisted of: blunging, screening, conditioning and 

decantation. Overflow and underflow resulting from settling of the flocculated material 

were then dried and submitted to TiO

2

 and brightness analyses. Sodium polyacrylate and 

sodium silicate were added to kaolin clay suspension as dispersant and ammonium 

hydroxide was also added to adjust pH. After degritting, the minus 44 

µm fraction was 

conditioned with hydroxamate surfactant and then polymers of different anionic charge 

were added. The influence of pH and type of polymer on efficiency of TiO

2

 removal, as 

well as kaolin mass recovery were studied. The results obtained in the present study are 

quite promising. It was possible to diminish TiO

2

 content

 

and increase kaolin brightness 

(ISO) from 82 to 88%. 

Key words: kaolin, selective flocculation, kaolin purification, paper industry, kaolin 

bleaching 

 

 

 

 

 

---

 

244 

INTRODUCTION 

Within the different applications of industrial kaolins, paper industry has been, 

until now, the most important segment and it corresponds to 50% of the kaolin market. 

There are other industrial uses of kaolin such as paint, ceramics, refractory, plastic, resin, 

catalyser, rubber, table ceramic, sanitary ceramic etc. But to attend most of the 

specifications required by the industry, it has to be processed. Kaolin deposits from 

Northern of Brazil contain colored impurities such as iron oxides, rutile and anatase 

(TiO

2

). These minerals are normally stained by iron and vary from yellow to dark brow 

in color. As reported by Monte et al. (1) and Luz (2), the kaolin industrial process used in 

that region consists mainly of dispersion, agitation, degritting, fractionation by 

centrifuge, high intensity magnetic separation, bleaching with sodium dithionite, 

flocculation, filtering and spray drying. According to Luz et al. (3), iron contaminants 

can be removed by leaching, using reducing reagents such as sodium hydrosulphite, but 

there is no chemical leaching economic process capable of discoloring rutile and anatase 

impurities.  In this case, physic-chemical methods have been used such as high intensity 

magnetic separation, flotation or selective flocculation. 

During fractionation of kaolin, because of high centrifuge forces resulted in the 

centrifugation step, mineral particles over 2 

µm in size go to the coarse fraction. 

Therefore, this method is not effective for submicroscopic particles and makes the 

industrial process less efficient. As stated by Maynard et al. (4), the fine fractions 

resulting from the centrifugation step, in many cases contain more TiO

2

 than original 

whole clay or the coarse fraction. 

Some kaolins from Georgia, USA, also present the same type of impurities like 

iron associated with rutile and anatase minerals. To attend the specifications required by 

the paper industry, different processes have been proposed for removing kaolin 

impurities. Sodium hydrosulphite has been used for soluble iron removal during the 

dewatering step. Flotation is another method used for rutile/anatase removal, mostly 

when these minerals are colored by iron. This process consists in increasing pH of 

suspension from acid to alkaline by adding ammonium hydroxide and then adding oleic 

acid as collector. Pine oil (frother) and divalent cationic activators ( Ca

++

 and Pb

++

 ) are 

added and then kaolin is floated as reported by Nott (5), Duke et al.(6); Smith (7); 

Mallary (8), Nott (9), Young and Morris (10). Yoon et al. (11) have studied purification 

of kaolin by flotation and showed that alkyl hydroxamate is more effective for TiO

2 

removal than oleic acid. 

There are basically two ways for removing titanium impurities by selective 

flocculation: i) disperse kaolin in a water suspension, flocculate the TiO

2

 impurities and 

separate them by sedimentation; ii) disperse kaolin in a water suspension, flocculate the 

kaolinite, keep impurities in suspension and separate them by sedimentation. 

Shi (12) reported that many processes have been proposed for TiO

2

 removal by 

selective flocculation. For example, kaolin is dispersed in a water suspension with high 

solid percentage (60%) by adding sodium hexametaphosphate, sodium metasilicate and 

ammonium chloride as dispersants. Then, suspension is diluted to 20% of solids content 

---

 

245 

and anionic flocculant polymer (Nalco 8872) is added to selectively flocculate the TiO

2 

minerals. After sedimentation step, the overflow is removed, coagulated and bleached 

with sodium dithionite. Behl et al.(13), in another method, propose the addition of oleic 

acid and calcium chloride, as a source of calcium ions , to the suspension containing TiO

2

 

impurities. In this case, 113 g/ton of anionic polymer (Sharpfloc 9960) is added to the 

pre-conditioned suspension. After stopping pulp agitation, colored flocks begin to 

sedimentation, the impurities go to the underflow and kaolin goes to overflow. Another 

method has also been proposed by Sheridan and John (14) for TiO

2

 removal and it 

consists in keeping impurities in suspension and then kaolin clay fraction is selectively 

flocculated, with partially hydrolyzed flocculants, such as Polyhall-59. 

According Maynard et al. (4), the TiO

2

 minerals can also be removed from kaolin 

using more dispersant reagents than the amount required. This will promote 

deflocculation of kaolinite particles. In a second step, applying reflocculation, kaolinite 

particles achieve high level of suspensional stability and TiO

2

, particularly anatase can be 

separated by sedimentation. This reflocculation process uses only one dispersant, such as 

sodium haxametaphosphate and it has been used in the beneficiation of fine fractions of 

Georgia kaolin. 

Larroid et al. (15) have studied the purification of kaolins from Northern of Brasil 

using selective flocculation to remove titanium impurities. They showed that is possible 

to promote separation of kaolinite from anatase in alkaline pH with high molecular 

weight weakly anionic polymer. 

Luz (16) developed a purification process, which can be applied to the kaolins 

from Northern of Brasil. This process consists of titanium impurities removal by 

selective flocculation, using anionic polymers of medium and high anionicity. 

The objective of this paper was to increase brightness of the kaolin clay from 

Northern Brasil through titanium impurities removal by selective flocculation.  

METHODOLOGY 

Aqueous suspensions of kaolin with 42% solids content were prepared. Sodium 

polyacrylate (3 kg/ton) and sodium silicate (3 kg/ton) were added to kaolin clay 

suspension as dispersant and ammonium hydroxide (2 kg/ton) was also added to adjust 

pH. The suspension was blunged at 3100 rpm for 15 min period. In a second step, kaolin 

clay suspension was screened at 53 

µm. The fraction over 53 µm was discharged as 

tailings and the fraction minus 53 

µm was conditioned at 1800 rpm for 15 min, with 1 

kg/ton of alkyl hydroxamate (6493) as surfactant. Then, the resulted suspension was split 

into three fractions. To study the influence of pH on selective flocculation, the pH of 

suspensions was adjusted to 9, 9.5 and 10. Initially, the same proportion (150 g/ton) of 

anionic flocculant (Table I) was added to each of the three kaolin samples. Each one of 

these samples was poured into graduated (1000 mL) cylinder

.

 Then, the suspension was 

stirred for 2 min and submitted to sedimentation for 30 min and thereafter the overflow 

was separated of underflow.  

---

 

246 

Table I – Anionic Charge (%) of Nalco Flocculants Used in the Selective 

Flocculation Tests 

0 – 25 % 

25 – 50% 

50 – 75% 

N 7766 

N 9901; N 9286 

N 7875; N 7874 

N 9872 

N 9285; N 9806; N 9825 

N 9878 

 

In the present study, titanium impurities were flocculated and kaolinite was kept  

in suspension. The products obtained in each test were submitted to chemical analysis for 

determination of TiO

2

 and Fe

2

O

3

. Additionally, the brightness of kaolin product was 

determined to evaluate performance of the selective flocculation process. This brightness 

was measured by using Zeiss Fotometer and 453 nm filter. Figure 1 shows the flowsheet 

which has been followed in conducting the tests. 

After those tests, a specific pH was chosen on which the best mass recovery and 

the highest brightness of kaolin product were achieved. Then, the influence of flocculant 

addition was studied.  

Kaolin (500g) 

465 dry base 

Wet Screening (53

µm) 

+ 53

µm 

 (Tailing) 

pH 9;150 g/ton of 

different flocculant 

pH 9.5; 150 g/ton of 

different flocculant 

pH 10; 150 g/ton of 

different flocculant 

Hydroxamate: 1kg/ton 

Cond. (15 min), 1800 rpm 

and pulp spliting 

Cond. 15 min, 42% 

solids, 3100 rpm 

Sodium 

polyacrylate:3kg/ton 

Na

2

SiO

3

: 3kg/ton 

607mL water 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1 - Flowsheet Used to Conduct Selective Flocculation Tests on Kaolin 

from Northern of Brasil 

---

 

247 

 

RESULTS AND DISCUSSIONS 

Chemical Analyses of Kaolin 

The chemical analyses of the + 53 

µm and the – 53 µm fractions are presented in 

Table II. The results showed that Fe

2

O

3

 and TiO

2

 accumulate mostly in the – 53 

µm 

fraction (87.9% and 85.3% recoveries, respectively). 

Table II – Fe

2

O

3

 and TiO

2

 Chemical Analyses of the +53 

µm and the–53 µm Fractions 

of Kaolin from Northern of Brasil 

Open  

Weight (%)  

Fe

2

O

3

 TiO

2

 

(

µm) 

 Grade 

(%) 

Recovery 

(%) 

Grade (%)  Recovery 

(%) 

+ 53 

µm 

12.1 1.45 12.1 

1.50 

14.7 

-53 

µm 

87.9 1.45 87.9 

1.20 

85.3 

Feed 

(calculated)  

100.0 

1.45 

100.0 

1.24 

100.0 

Feed 

(analyzed) 

100.0 

1.50 

100.0 

1.25 

100.0 

Selective Flocculation Tests 

Influence of pH on Selective Flocculation 

Figures 2 and 3 show the influence of pH on mass recovery (yield) and brightness 

of kaolin product of selective flocculation when equal amounts of flocculant were used. 

It can be seen in Figure 2 that the best performance in terms of mass recovery 

from the overflow kaolin product was achieved at pH 9.5 for any flocculant applied in 

the present study. The N-7766 (low anionicity) and N-9901 (medium anionicity) 

flocculants presented the best kaolin mass recovery close to 85% but on the other hand 

the brightness of the kaolin product (Figure 3) was not satisfactory (82.5%). 

Figure 3 shows brightness of the kaolin product from overflow, after 

sedimentation step. It was observed that the N-9825 flocculant produces the highest 

brightness close to 90% at pH 9, at low mass recovery of only 17% (Figure 2). 

---

 

248 

 Nalco Flocculants 

0

20

40

60

80

100

8

8.5

9

9.5

10

10.5

11

pH

Mass Recovery (%)

N-9878__N-7874__N-7875__N-9602__N-9872__N-7766__N-9901__N-9806'>N-9878

N-7874

N-7875

N-9602

N-9872

N-7766

N-9901

N-9806

N-9825

Over 

 

Figure 2 – Influence of pH on Mass Recovery of Kaolin Product from 

Overflow of Selective Flocculation, when Using the Same Addition (150 

g/ton) of Different Anionic Flocculants 

Nalco Flocculants

80

85

90

95

100

8

8.5

9

9.5

10

10.5

11

pH

Brightness (%)

N-9878

N-7874

N-7875

N-9602

N-9872

N-7766

N-9901

N-9806

N-9825

Over 

 

Figure 3 – Influence of pH on Brightness of Kaolin Product from Overflow of 

Selective Flocculation for the Same Addition (150 g/ton) of Different Flocculants 

---

 

249 

Influence of flocculant addition on Selective Flocculation Process 

Figure 4 presents kaolin overflow mass recovery versus addition of 

anionic flocculants (N-9878 and N-9806), after selective flocculation of titanium 

impurities. Note that when using 150 g/ton addition, the mentioned flocculants 

presented, approximately, the same kaolin mass recovery (60%), however, for 

flocculant addition lower 150 g/ton, N-9806 flocculant (medium anionicity) is 

more efficient then high anionic N-9878 flocculant. 

Figure 5 presents brightness of the kaolin in the overflow product, after 

selective flocculation step. Note that N-9878 flocculant showed a better 

brightness (close to 88%) for a flocculant addition between 50 g/ton and 125 

g/ton. On the other hand, N-9806 flocculant presented small oscillation in 

brightness of the kaolin product, around 87%, for addition over 50 g/ton. 

Figures 6 and 7 present the distribution of Fe

2

O

3

 and TiO

2

 in the overflow 

of selective flocculation, as a function of flocculant addition. It can be observed 

that in the case of N-9806 and N-9878 flocculants and addition over 100 g/ton, 

Fe

2

O

3

 and TiO

2

 recovery in the overflow diminish. This indicates that TiO

2

 and 

Fe

2

O

3 

impurities are going, selectively, to the underflow. 

It can also be observed in Figures 6 and 7 that iron and titanium present 

similar distributions in the overflow, because part of iron occurs in the lattice of 

rutile/anatase. 

 Nalco Flocculants 

0

20

40

60

80

100

0

50

100

150

200

250

Flocculant addition (g/t)

Mass Recovery (%)

N-9878

N-9806

Overflow 

pH 9.5

 

Figure 4 – Influence of Flocculant Addition on Mass Recovery of 

Kaolin Content in the Overflow of Selective Flocculaiton, at pH 9.5 

---

 

250 

 

Nalco flocculants 

70

80

90

100

0

50

100

150

200

250

 Flocculant addition  (g/t)

Brightness (%)

N-9878

N-9806

Overflow 

pH 9.5

 

Figure 5 – Influence of Flocculant Addition on Brightness of Kaolin 

Content in the Overflow of Selective Flocculation, at pH 9.5 

 Nalco flocculants 

0

20

40

60

80

100

0

50

100

150

200

250

Flocculant addition (g/t)

 Fe2O3 Recovery (%)

N-9878

N-9806

Overflow

pH 9.5

 

Figure 6 – Influence of Flocculant Addition on  Recovery of Fe

2

O

3

 

Content in the Overflow of Selective Flocculation , at pH 9.5 

---

 

251 

  Nalco Flocculant 

0

20

40

60

80

100

0

50

100

150

200

250

 Floculant addition (g/t)

 TiO2 Recovery (%)

N-9878

N-9806

Overflow

pH 9.5

Figure 7 – Influence of Flocculant Addition on TiO

2

 Recovery in the 

Overflow of Selective Flocculation, at pH 9.5 

Alkyl hydroxamate has been used as collector in floating the colored 

anatase impurities from clay by froth flotation (11). In the present work, the 

selective adsorption of alkyl hydroxamate on anatase/rutile particle surface could 

favor its interaction with anionic polymer, resulting in a co-adsorption 

phenomenon. From this phenomenon, flocculation of anatase/rutile particles 

could occur by hydrophobic forces.  

In the present work, ten anionic polymers with different charge density 

(25 – 75%) were tested with the objective of selecting the most effective polymer 

to yield high mass recovery and high brightness from overflow kaolin product. 

The results showed that medium and high anionic polymers are more effectives.  

Polyacrylamides with medium anionicity (25-30%) have been used to 

dewater tailing containing clay in which occurs flocculation of the total particles 

(17). Our present work is quite different because there is a selective flocculation 

process of the titanium impurities and then its decantation. 

CONCLUSIONS 

•  This study showed the technical feasibility of removing most of titanium 

impurities by selective flocculation, using medium and high anionic 

polyimers.  

•  Close to 80% of TiO

2 

and Fe

2

O

3

 are recovered in the underflow of selective 

flocculation, at pH 9.5. 

---

 

252 

•  The selective flocculation process permited to increase kaolin brightness from 

83% to 88% (ISO) and mass recevery of 60%. 

•  The two flocculants of medium and high anionicity (N-9806 and N-9878) 

presented, aproximately, the same performance in terms of brightness and 

mass recovery. 

•  The use of hydroxamate as surfactant, enhanced the selective flocculantion of 

TiO

2 

impurities. 

•  Titanium and iron impurities presented similar distribution in the overflow of 

the selective flocculation products. This is an indication that significant part 

of iron occurrs in the lattice of rutile/anatase 

REFERENCES 

1  

Monte, M. B. M.; Carvalho, E.A.; Ferreira, O.; Cabo, S. S. “Caulim-CADAM.”  

In: Usinas de Beneficiamento de Minérios do Brasil, Eds. João A. Sampaio, Adão 

B. Luz e Fernando F. Lins,  Rio de Janeiro, CETEM-2001, 11-23. 

2 

Luz, A. B. “Estudo de Reoxidacão de Ferro Contido em Caulins”, PhD Thesis, 

Escola Politécnica da Universidade de S. Paulo, 1998, 16-18. 

3 

Luz, A.B.; Yldirim, I.; Yoon, R-H. “Purification of Brazilian Kaolin Clay by 

Flotation.” Paper presented at XII International Mineral Processing Congress, 

Rome Italy, Vol. C, 2000,  c8b 79-c8b 83. 

4 

Maynard, R. N.; Milman, N.; Lancinelli, J. “A method for removing titanium 

dioxide impurities from kaolin.” Clay and Clay Minerals, Vol. 17, 1969, 59-62. 

5 

Nott, A. J. “Method for Improving Clay Brightness Utilizing Magnetic 

Separation”, British Patent, No. 1,489,158977, 1977. 

6 

Duke, J. B.; Metuchen, N. J. “Improving clay brightness by flotation and fine 

grinding”, U.S Patent, No. 2,920,8321960, 1960  

7 

Smith, S. J. “Kaolin flotation process”, U.S. Patent, No. 3,744,630, 1973. 

8 

Mallary, M. B. “Purification of kaolin clay by froth flotation”, U.S Patent, No. 

3,827,556,1974.  

9 

Nott, A. J. “Brightening of clay by froth”, U.S Patent, No. 4,098,688, 1978. 

10 

Young, R. H.; Morris, H.H. “Method of treating clay to improve its whiteness”, 

U.S Patent, No. 4,492,628, 1985.  

11 

Yoon, R-H; Nagaraj, D. R.; Wang, S. S.; Hildebrand, T. M. “Beneficiation of kaolin clay 

by froth flotation using hydroxamate collectors. Minerals Engineering. Vol. 5, Nos. 3-5, 

1992, 457-467. 

---

 

253 

12 

Shi, J.C.S. “Method of beneficiation kaolin clay utilizing ammonium salts”, US 

Patent , No. 4,604,369, 1986. 

13 

Behl, S.; Will, M. J.; Younh, R. H. “Method for separating mixture of finely 

divided minerals”, U.S Patent, No. 553,5890, 1994.  

14 

Sheridan III; John, J. “Process for purifying clay by selective flocculation”, US 

Patent, No. 3,837,482, 1973. 

15 

Larroyd, F.; Peter, C. O.; Sampaio, C. H. “Purification of north kaolin by 

selective flocculation.”, Minerals Engineering, 15 (2002) 1191-1192, 2002. 

16 

Luz, A. B. “Processo de Purificação de Caulim por Floculação Seletiva”. 

Brasilian Patent, No 008198, 2001. 

17 

Xu, Y; Hamza, H. Thickening and disposal of oil sand tailings, Mining 

Engineering, p33-39, November 2003. 

 

 

 

---

 

Rio de Janeiro, 01 de setembro de 2004. 

 

 

 

 

 

 

Salvador Luiz Matos de Almeida 

Chefe do Serviço de Tratamento de Minérios e Usina Piloto - SETU 

 

 

 

 

 

João Alves Sampaio 

Chefe da Coordenação de Processos Minerais - COPM 

 

 

 

 

 

Adão Benvindo da Luz 

Diretor do CETEM

 

---

Document Outline

• ABSTRACT
  • • N 9878