This article is also available online at:

www.elsevier.com/locate/mineng

Minerals Engineering 18 (2005) 969–976

Influence of dry grinding on talc and kaolinite morphology:inhibition of nano-bubble formation and improved dispersion

Marek Zbik a, Roger St. C. Smart a,b,*

a Ian Wark Research Institute, University of South Australia, Mawson Lakes, Adelaide 5095, Australiab ACeSSS (Applied Centre for Structural and Synchrotron Studies), University of South Australia, Mawson Lakes, Adelaide 5095, Australia

Received 1 December 2004; accepted 4 January 2005

Available online 23 February 2005

Abstract

The effects of dry grinding in a porcelain ball mill and a chrome steel ring mill on the structure and morphology of talc and kaol-

inite minerals and mixtures have been studied. It is well known that ground talc is easier to make down as a suspension than

unground talc and that it displays better rheological behaviour in the slurry. Morphological and structural studies (SEM, TEM,

AFM, XRD) of both ground and unground talc and kaolinite samples have revealed new factors accounting for this behaviour dur-

ing make down as a mineral suspension in water. Short-term (<1 min) grinding in the ring mill not only breaks the platelets, which

lowers the aspect ratio, but also disaggregates most of the particles. In both talc and kaolinite, this action destroys voids in aggre-

gates. In talc, it also destroys the splayed ends of the sheets at the particle edges. Both actions reduce trapped nano-bubbles and their

tendency to reduce wetting and promote flotation. Platelets of talc, during grinding, also become more stepped and damaged on

basal surfaces as a result of abrasion. Abrasion on basal surfaces exposes additional edge area, which increases the proportion

of reactive sites and assists dispersion of talc in aqueous solution. Prolonged grinding (60 min) of mixtures (10% talc) produces

rounded aggregates that are composed of nm-sized colloidal particles. These colloids and aggregates are strongly hydrophilic. Struc-

tural observations (XRD and electron diffraction patterns) indicated that crystalline structure destruction occurs during prolonged

grinding in the ring mill making both minerals totally amorphous. This does not occur after mixing (15 min) in the porcelain ball

mill.

� 2005 Elsevier Ltd. All rights reserved.

Keywords: Tailings disposal; Dewatering; Fine particle processing

1. Introduction

Grinding of kaolinite and talc as brightness additives

is an important process in the paper, adhesives, paint

and plastics industries. Because of this, they have beenstudied with great interest by many scientists. The effects

of grinding on clay minerals have been studied, for in-

stance, by Takahashi (1959), Miller and Oulton (1970),

0892-6875/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.mineng.2005.01.005

* Corresponding author. Address: ACeSSS (Applied Centre for

Structural and Synchrotron Studies), University of South Australia,

Mawson Lakes, Adelaide, 5095 Australia. Tel.: +61 8 83023353; fax:

+61 8 83023799.

E-mail address: roger.smart@unisa.edu.au (R. St. C. Smart).

Yariv (1975), Garcia et al. (1991) and Aglietti et al.

(1986) for kaolinite; and for talc by Aglietti (1994) San-

chez-Soto et al. (1997) and Zajac and Malandrini (1997).

Several of these studies came to the conclusion that in-

tense mechano-chemical effects occur during grindingleading to the destruction of kaolinite structure and for-

mation of an amorphous substance. The laminar struc-

ture of kaolinite makes this mineral very sensitive to

amorphization.

Aglietti (1994) revealed that intense mechano-chemi-

cal effects occur as well in the talc surface. Sanchez-Soto

et al. (1997) found that reduction in size of talc particles

by grinding continued up to about 30 min. After 30 min,

970 M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976

the mechanical reduction of the original particles ap-

pears to have a limit and particles start a reaggregation

process in which adhesion forces act. Garcia et al. (1991)

reported similar effects in kaolinite where aggregates ap-

peared to form by tightly bound, extremely small parti-

cles, produced during prolonged grinding. Two factorsseem to be responsible for such structural changes in

the process of dry grinding of kaolinite and talc. One

is the production of a non-crystalline substance attended

by the disordering of the crystalline part, and the other

is a reaggregation process. Despite an extended effort to

investigate mechano-chemical effects, and their conse-

quences for dispersion, caused by grinding, some aspects

of the alteration remain obscure.Zbik and Smart (2002) used micronized and ground

talc samples to prepare dense talc/water suspension.

During these experiments they reported that ground talc

was easier to make down as a suspension than the

unprocessed sample; ground talc also displayed Newto-

nian rheological behavior in contrast to the micronized

sample. During careful morphological studies of both

ground and unground talc samples, Zbik and Smart(2002) found two major reasons for such behaviour dur-

ing make down as a mineral suspension in water.

Firstly, the short grinding time in the ring mill has a

significant effect on the talc platelet diameters, breaking

platelets apart, which lowers the aspect ratio. Secondly,

voids in aggregates and splayed or puckered sheets at

particle edges are destroyed during ring mill grinding

which may reduce trapped nano-bubbles and stop parti-cles floating on the water surface.

SEM observations by Zbik and Horn (2003) of low

solid content cryo-vitrified kaolinite clay suspensions re-

veal that clay platelets build porous three-dimensional

networks with platelets contacting each other mostly

by their edges. To explain this behaviour, which must re-

quire long range edge-to-edge attractive forces, a hydro-

phobic-like interaction has been proposed. Thisinteraction may be induced by the presence of nano-

bubbles existing on the edges of clay crystallites which

may cause clay particles to flocculate. Nanobubble coa-

lescence has recently been presented as a partial explana-

tion (e.g. Meagher and Pashley, 1995; Considine et al.,

1999; Ishida et al., 2000) for strong, long-range attrac-

tive forces that have measured between hydrophobic

surfaces. (Israelachvili and Pashley, 1982; Christenson,1992). The following indirect evidence for such hydro-

phobic behaviour was presented. First, a clay platelet

is shown attached to an oil drop by its edge; second, clay

flocs were attracted by a vertically placed hydrophobic

Teflon strip but not to the hydrophilic mica basal sur-

face; third, a much thicker porous sediment occurred

in CO2-saturated water solution compared to vacuum-

degassed water.The work reported here has revealed new factors

which have been observed when grinding and focuses

on the both the morphological and structural changes

(using SEM, TEM, XRD and AFM) which occur as a

result of prolonged dry grinding.

2. Experimental

North Queensland kaolinite and Commercial Miner-

als Ltd., Talc (CM) were used in this study as single min-

erals and as the kaolinite/talc mixture. The North

Queensland kaolinite was supplied by Comalco Re-

search Centre (Thomastown, Victoria, Australia). It is

the final processed product from their previous North

Queensland operation. Its characteristics, particularlythe layered structure of the particle surfaces, wide steps

and ragged edges revealed in SEM and AFM studies,

have been described by Zbik and Smart (1998). Com-

mercial Minerals Limited (CM) talc T20A used in this

study is a commercial talc described as premium grade

with excellent whiteness, micronized (dry, without addi-

tives) below 20 lm. The micronising process involvescolliding air/talc streams at 120 atm pressure in an aircyclone with classification to extract the �20 lm frac-

tion. It is typically used in adhesives, paints and plastics.

It is produced in Adelaide, predominantly from the Mt.

Fitton (South Australia) deposit.

Samples (50 g) were dry ground during 15, 30, 60 and

120 min in the porcelain ball mill (balls �26 g) and 1 and60 min in the high power Rocklabs ring mill with

chrome steel (AISI, D3) heads. The mineral mixturewas kaolinite with 10 wt.% talc mixed for 5 min in a

mechanical shaker.

The SEM studies were carried out using a Camscan

CS44FE microscope with a field emission gun operating

at 20 kV acceleration voltage. The kaolinite and talc

particles were coated by gold/palladium films to a thick-

ness of 2–3nm using a Dentron Magnetron Sputter

Coater system. A Nanoscope III AFM (Digital Instru-ments) was used with scan heads E (14 · 14 lm2) andscan rate between 5 and 20 Hz in height and deflection

modes. The AFM was calibrated on a gold grid with

5 lm pits separated by 5 lm. The standard pyramidalsilicon nitride tip with a solid angle of 70� and a radiusof curvature at the end of �50 nm was mounted on a

cantilever of nominal spring constant 0.06–0.58 Nm�1.

The standard lateral and vertical deviation of theAFM measurements is ±0.15 nm due, at least in part,

to the uneven nature of the kaolinite and talc particle

surfaces and the relatively low aspect ratio of the

AFM tip.

For the AFM studies, kaolinite and talc particles

were immobilized from a dilute aqueous suspension on

to a freshly cleaved atomically flat mica surface. A small

amount of suspension, i.e. one or two drops was col-lected from a depth of 10 cm below the water surface

and placed on the freshly cleaved mica surface. The par-

M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976 971

ticles were dried at room temperature in a dessicator

overnight. Particles with weak contact to the mica sur-

face were removed in a dry nitrogen gas stream before

SEM examination. The SEM examination showed that

the attached particles were then suitable for AFM study.

3. Results

3.1. Talc: dry grinding

As reported in Zbik and Smart (2002) and confirmed

here, short grinding times in the ring mill had a signifi-

cant effect on the talc platelet diameter, breaking plate-lets and lowering the aspect ratio. SEM micrographs

(Fig. 1) illustrate significant differences in particle size

and morphology between unground and talc ground

short-term (1 min) in the ring mill. For the unground

talc sample, large aggregates of high aspect ratio, curved

platelets, with characteristic splayed and separated edges

are observed in SEM (Zbik and Smart, 2002). Such

platelets are generally larger than 5 lm in diameter(Fig. 1a). In contrast, ground talc contains smaller sep-

arate platelets, which are generally below 2.5 lm in lat-

eral dimension (Fig. 1b). This short grinding (1 min) has

only a minor effect on the particle thickness (i.e. minor

delaminating occurs) whilst reducing the median lateral

dimension of the platelets by approximately half which

significantly reduces the aspect ratio (Table 1). One min-

ute grinding has also a minor effect on the specific sur-face area calculated from the particle geometry, but

significantly increases the edge area of the ground talc

from 13.5% before grinding to 21% after 1 min grinding.

The edge area increase is likely to be beneficial for better

dispersion of talc in aqueous systems.

The results of the aspect ratio calculation, based on

SEM/AFM measurements of �100 talc particles using

Fig. 1. SEM micrographs of (a) Commercial Minerals Talc sample (CM, T2

ground 1 min in ring mill.

Table 1

Geometric characterization of talc samples

Talc Median diameter (nm) Median thickness (nm)

Unground CM 2110 165

Ground CM 1245 165

AFMmethods described in Zbik and Smart (1999), have

been summarized in Table 1. These data have revealed

that the median aspect ratio value of unground CM talc

sample is considerably larger than for the ground

sample.

The second reason for improved make down of theground talc better in aqueous suspension, previously de-

scribed in Zbik and Smart (2002), comes from the differ-

ence in the void availability in individual platelets and

aggregates. Splaying or puckering, frequently observed

on the edges of unground talc platelets (Figs. 1a and

2b), has not been observed after short-term grinding in

the ring mill (Figs. 1b and 2c). Fracturing or compacting

these sheets during grinding contributes to depressingthe apparent hydrophobicity of talc which inhibits dis-

persion of talc particles in water. The likely reason for

this is the loss of suitable voids for nano-bubble

entrapment.

The differences in particle size and void volumes have

a significant impact on the sample structure, which mani-

fests in sample porosity. The bulk gravity measured for

unground talc was 0.3 g cm�3 increased to 0.55 g cm�3

after 1 min in ring mill grinding; resulting in 9%

reduction of interparticle porosity. This ‘‘flattening’’ of

aggregates of talc particles destroys voids suited to

nano-bubbles trapping between individual platelets

and thereby contributes to better mixing of talc with

water after grinding.

Low magnification AFM micrographs of talc (Fig. 2)

show highly aggregated, irregular flocs of talc platelets�15–22 lm in lateral dimension apparently due to floc-

culation in water during sample preparation. AFM

images of the unground CM talc sample (Fig. 2a and

b) display bevelled edges, which are an artefact caused

by the relatively low aspect ratio of the AFM tip (i.e.

sliding down the side of the tip). Details of common arti-

facts in SPM (scanning probe microscope) are discussed

0a) as received, (reproduced form Zbik and Smart (2002) Fig. 2a); (b)

Aspect ratio Surface area (m2 g�1) Roughness (rms nm)

12.8 5.5 1.7

7.5 6 8

Fig. 2. AFM micrographs of CM (T20a) talc: (a and b) before (reproduced from Zbik and Smart (2002) Figs. 4a and 5b); (c and d) after grinding 1

min in ring mill.

972 M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976

in our previous paper (2002). Individual platelets shownin Fig. 2a and b have 1–2 lm lateral dimensions. The

ground talc (Fig. 2c and d) have very ragged, stepped

basal surfaces, with lateral damage and smaller particles

(<100 nm) attached to the surface.

The colloidal platelets lying on the smooth talc basal

surface sometimes overlap each other suggesting that

they do not belong structurally to the underlying platelet

but may have been attached in preparation processes. Itis possible that such colloidal platelets (even below 20

nm in lateral dimension and single unit cell thickness)

were generated during the milling process and were at-

tached to larger particles when drying. At higher magni-

fications (Fig. 2d), lateral shear steps are clearly visible

on basal faces. These steps, studied on this section, are

1.7–3 nm high and (within the instrument error) relate

to 2–4 individual lattice layers of talc. Extremely smallparticles were also observed. These particles in some

parts of the sample form a sort of coating about 1 nm

thick and may be the colloidal products of grinding.

This roughening of the basal surfaces is another rea-

son for improved make down of the ground talc better

in aqueous suspension. This abrasion exposed addi-

tional edge area on basal surfaces, increasing the con-

centration of reactive surface sites and thereforeenhancing talc wettability. This is consistent with expla-

nations which challenge the interpretation of talc as a

hydrophobic mineral, e.g. Michot et al. (1994). Differentvarieties of natural talc have substantially different

water uptakes. Immersion calorimetry has shown that,

after outgassing at 100–400 �C, talc exhibits a strongaffinity for water molecules. Attempts to correlate natu-

ral flotation results to the electrokinetic properties of

talc have not been successful. The poor wettability of

talc and its influence on flotation has still not been

explained.AFM micrograph of unground and ground talc sam-

ples seen on Fig. 2 show visible transformation from

rather smooth blocky talc platelets of unground CM talc

sample (Fig. 2a and b) to highly disordered particles

after milling in the ring mill. Ground talc flakes (Fig.

2c and d) appear highly stepped with increased edge

density on what were previously flat basal surfaces.

The restructuring also suggests that some platelets havedetached from the original particle and moved across

the surface. Such surfaces with high edge density consid-

erably assist make down in the water suspension.

Mean Roughness has been measured for 500 nm

square of the talc basal surface of particle. (Ra)-mean

value of the surface heights relative to the centre plane

and is calculated using:

Ra ¼ ð1=LxLyÞZ Ly

0

Z Lx

0

jf ðx; yÞjdxdy

M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976 973

where f(x,y) is the surface height relative to the centre

plane and Lx and Ly are the dimensions of the surface.

The centre plane is the plane at the mean height of the

sample.

Mean Roughness varies from particle to particle but

those measured for the surface of the unground CM talcparticle shown in Fig. 2a were 1.6–1.9 nm and for the

ground CM talc particle shown in Fig. 2c were 7.2–

8.4 nm.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

0 20 40 60 80 100 120

Time of grinding in min

Spec

ific

Sur

face

Are

a in

m2 /g

kaolinite

kaolinite/talc

ring mill ground

Fig. 3. Specific surface area plot of porcelain mill (upper curves) and

ring mill ground (60 min) sample of kaolinite/talc mixture derived from

Malvern Mastersizer grain size measurement.

4. Kaolinite: Dry grinding

SEM micrograph from unground North Queenslandkaolinite Fig. 4a and b show well-developed euhedral

pseudo-hexagonal platelets of kaolinite crystals up to

1.5 lm on the (001) plane. Surfaces of these large crys-

tals with sharp edges host a number of smaller crystals

with lateral dimension down to 100 nm. These small

particles are euhedral and pseudo-hexagonal slightly

elongated in one direction. Detailed study of North

Queensland kaolinite particles by Zbik and Smart(1998) revealed its quite complex surface morphology

with high density of steps and nm-scale irregularities

on basal surfaces. Also mosaics of smaller crystals

(about 50 nm lateral dimensions) stacked together along

(110) and (010) planes appear to be unique for this

kaolinite. This horizontal stacking has contributed to

the high roughness of this kaolinite sample.

SEM micrographs (Fig. 4c and d) obtained after pro-longed (60 min) grinding in the ring mill, show signifi-

cant changes in morphology. The resulting material is

rounded, or pillow-like, showing little or no defined

structural arrangement, nor regular modular appear-

ance. On these micrographs shapeless larger aggregates

Fig. 4. SEM micrographs. (a and b) Unground kaolinite crystals; (c and d)

grinding in the ring mill.

are constituted of very small rounded particles

<100 nm in diameter and often far smaller than 10–20

nm.

5. Kaolinite/talc mixture: Dry grinding

Other significant changes in morphology observed as

a result of dry grinding were particle aggregation and

amorphization. Malvern Mastersizer results of grain size

distribution show some aggregation of the sample after

15, 13, 60 and 120 min of milling in the porcelain ball

mill. Specific surface area, calculated from grain size dis-

tribution dropped from 0.6 to 0.43 m2/g with increase ofd(50) from 5.44 to 7.29 lm for the kaolinite/talc mix-

ture, ground in the porcelain mill (Fig. 3). More signif-

icant aggregation occurred after high power grinding in

amorphous, pulverized aggregates of nano-sized particles after 60 min

Fig. 5. TEM micrographs of the kaolinite/talc mixture with well-

developed pseudohexagonal symmetry of kaolinite particles shown in

Fig. 5a (unground) is replaced by amorphous, shapeless rounded

aggregates after 60 min grinding in the ring mill in Fig. 5b.

974 M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976

the ring mill. Reduction of the specific surface area was

larger (0.33 m2/g) when the kaolinite/talc mixture was

ground in the ring mill for 60 min, with increase of

d(50) from 5.4 lm (unground) to 18.5 lm (c.f. with

7.3 lm after 120 min grinding in the porcelain ball mill).

Decrease of the specific surface area of ground sam-ples is clearly due to physical aggregation principally

involving the talc. Reduction in surface area is apparent

in talc between 0 and 15 min of grinding in contrast to

kaolinite where the specific surface area initially in-

creases, probably due to stack destruction (Fig. 3). After

15 min grinding, the specific surface area of kaolinite

drops continuously and reaches the same value as the

kaolinite/talc mixture after 120 min grinding in the por-celain ball mill (Fig. 3).

Similar results come from TEM micrographs in

which well-defined pseudohexagonal crystals of un-

ground North Queensland kaolinite in the kaolinite/talc

mixture are found as shown at the micrographs (Fig.

5a). EDX spectra of this sample are typical for kaolinite

Al, Si content with some Fe contribution which is char-

acteristic of this kaolinite. These kaolinite crystals showregular hexagonal arrangement in electron diffraction

patterns proving its crystalline structure. Talc addition

was not recorded on these micrographs because talc

flakes were much larger then those of North Queensland

kaolinite and rarely occurred in the narrow view field.

After prolonged milling, shapeless rounded aggregates

were observed on TEM micrographs (Fig. 5b). The aver-

age diameter of these aggregates was about 100–200 nm.These aggregates no longer display electron diffraction

patterns, suggesting their amorphous state. EDX spectra

of this sample show high uniformity where Mg (from

talc) is detected in all areas additional to the Al, Si

and Fe (from kaolinite) detected before milling. This

may suggest that the resulting amorphous mixture is

now chemically uniform.

Low magnification AFM micrographs prove thesedifferences in morphology between unground (Fig. 6a)

and ground (Fig. 6b) North Queensland kaolinite in

the mixture which transform particles after prolonged

grinding from stacked pseudohexagonal platelets to

shapeless round aggregates. The high magnification

AFM micrograph in Fig. 6c, of unground North

Queensland kaolinite, shows a typical singular kaolinite

platelet, with well-defined pseudohexagonal symmetryabout 200 nm in diameter and 20 nm in thickness, which

gives an aspect ratio of this particle about 10. The disc

shaped particle shown in Fig. 6d belongs to 60 min

ground kaolinite/talc mixture and is 100–150 nm in

diameter and 16 nm in thickness giving an aspect ratio

of about 6. But careful observation of this micrograph

reveal that this disc like platelet is, in fact, an aggregate

and consists of at least 3 smaller particles of �50 nmdiameter giving aspect ratios of each less than 3. AFM

micrographs also revealed much smaller particles of

about 1 nm in diameter shown in Fig. 6e where such

particles looks like needles 5 nm long and 1 nm thick.

Because these needles are arranged always parallel to

the scan direction they are likely artifacts of the AFMtip dragging small particles when scanning producing a

lengthening effect. At lower magnification (Fig. 6f), a

square region has been observed in the area of previous

scanning (Fig. 6e) which appears to be the result of

sweeping away many of the colloidal particles by the

AFM tip in this area. This region is about 1 nm deep

which provides additional information about the dimen-

sions of particles building this thin layer of colloidaldebris on the top of the mica substrate.

The XRD diffraction pattern of the kaolinite/talc

mixture initially mixed in the porcelain ball mill (Fig.

7a) shows crystalline structure of the mixed minerals

with well-defined diffraction peaks from kaolinite, talc

Fig. 6. AFM micrographs of the kaolinite/talc mixture. (a and c) Unground particles; (b and d) ground shapeless rounded aggregates after 60 min

grinding in the ring mill; (e and f) nano-size colloidal film on the mica substrate.

M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976 975

and minor chlorite impurity. The diffraction pattern pre-

sented in Fig. 7b represents the same mineral mixtureafter 60min grinding in the ring mill and shows total

amorphization of feed minerals which confirms our elec-

tron microscope results.

The specific surface analysis performed by BET meth-

od of nitrogen adsorption on the kaolinite/talc sample

ground for 15 min in the porcelain ball mill and after

prolonged grinding (60 min) in the ring mill confirms

substantial aggregation during ring mill grinding. Forcomparison, the starting BET specific surface area of un-

ground North Queensland kaolinite was 26 m2/g, and for

the CM talc sample 5.5 m2/g. After 15 min in the porce-

lain ball mill, the kaolinite/talc mixture had a specific sur-

face area of 23.3 m2/g, which is close to the combinationof specific surface areas of the two minerals implying lit-

tle grinding or aggregation. This value has dropped to

10.7 m2/g after 60 min grinding in the ring mill.

6. Conclusions

The effects of dry grinding in a porcelain ball mill anda chrome steel ring mill on the structure and morphol-

ogy of talc and kaolinite minerals and mixtures have

been studied. It is well known that ground talc is easier

Fig. 7. X-ray diffraction patterns of the kaolinite/talc mixture: (a)

before and (b) after 60 min grinding in the ring mill.

976 M. Zbik, R. St. C. Smart / Minerals Engineering 18 (2005) 969–976

to make down as a suspension than unground talc and

that it displays better rheological behaviour in the

slurry.

Morphological and structural studies (SEM, TEM,

AFM, XRD) of both ground and unground talc and

kaolinite samples have revealed new factors accountingfor this behaviour during make down as a mineral sus-

pension in water. Short-term (<1 min) grinding in the

ring mill not only breaks the platelets, which lowers

the aspect ratio, but also disaggregates most of the par-

ticles. In both talc and kaolinite, this action destroys

voids in aggregates. In talc, it also destroys the splayed

ends of the sheets at the particle edges. Both actions

are likely to reduce trapped nano-bubbles and their ten-dency to reduce wetting and promote flotation. Platelets

of talc, during grinding, also become more stepped and

damaged on basal surfaces as a result of abrasion. Abra-

sion on basal surfaces exposes additional edge area,

which increases the proportion of reactive sites and as-

sists dispersion of talc in aqueous solution. The high

power ring mill has affected kaolinite and talc far more

than the porcelain ball mill.Prolonged grinding (60 min) of mixtures (10% talc)

produces rounded aggregates that are composed of

nano-meter sized colloidal particles. These colloids and

aggregates are strongly hydrophilic. The specific surface

area (measured by BET method) of these aggregates is

smaller than that of the unground mineral mixture.

Structural observations (XRD and electron diffraction

patterns) indicated that crystalline structure destructionoccurs during prolonged grinding in the ring mill mak-

ing both minerals totally amorphous. This does not

occur even after prolonged mixing (15 min) in the porce-

lain ball mill.

Acknowledgements

The authors are grateful to Comalco Research Centre

(Thomastown, Victoria, Australia) for financial support

for this work. Many useful research discussions on this

and other work with Dr. Ray Shaw, Ms. Karen Bur-rows, Dr. Angus Hartshorn (Comalco) and Drs. Peter

Self and Bill Skinner (IWRI) are also gratefully

acknowledged.

References

Aglietti, E.F., 1994. The effect of dry grinding on the structure of talc.

Appl. Clay Sci. 9, 139–147.

Aglietti, E.F., Porto Lopez, J.M., Pereira, E., 1986. Mechano-chemical

effects in kaolinite grinding I and II. Int. J. Miner. Proc. 16, 125–

133.

Christenson, H.K., 1992. In: Schrader, M.E., Loeb, G. (Eds.), Modern

Approaches to Wettability: Theory and Applications. Plenum

Press, New York.

Considine, R.F., Hayes, R.A., Horn, R.G., 1999. Langmuir 15, 1657–

1659.

Garcia, F.G., Abrio, M.T.R., Rodrigues, M.G., 1991. Effects of dry

grinding on two kaolins of different degrees of crystallinity. Clay

Miner. 26, 549–565.

Ishida, N., Inoue, T., Miyahara, M., Higashitani, K., 2000. Langmuir

16, 6377–6380.

Israelachvili, J.N., Pashley, R.M., 1982. Nature 300, 341–342.

Meagher, L., Pashley, R.M., 1995. Langmuir 11, 4019–4024.

Michot, L.J., Villieras, F., Francois, M., Yvon, J., Le Dred, R., Cases,

J.M., 1994. The structural microscopic hydrophilicity of talc.

Langmuir 10, 3765–3773.

Miller, J.G., Oulton, T.D., 1970. Phototropy in kaolinite during

percussive grinding of clays. Clays and Clay Miner. 18, 313–323.

Sanchez-Soto, P.J., Wiewiora, A., Aviles, M.A., Justo, A., Perez-

Maqueda, L.A., Perez-Rodriguez, J.L., Bylina, P., 1997. Talc from

Puebla de Lillo, Spain. II. Effect of dry grinding on particle size and

shape. Appl. Clay Sci. 12, 297–312.

Takahashi, H., 1959. Effects of dry grinding on kaolin minerals I–III.

Bull. Chem. Soc. Japan 32, 235–263.

Yariv, S., 1975. Infrared study of grinding kaolinite with alkaline metal

chlorides. Powder Technol. 12, 131–138.

Zajac, J., Malandrini, H., 1997. Changes in the immersional enthalpy

of talc in n-heptane and water as a function of grinding. Polish J.

Chem. 71 (5), 686–691.

Zbik, M., Horn, R.G., 2003. Hydrophobic attraction may contribute

to aqueous flocculation of clays. In: Adamczyk, Z., Kosmulski, M.

(Eds.), Electrokinetic Phenomena, Krakow 2002. Colloids and

Surfaces A 222, 323–328 (special issue).

Zbik, M., Smart, R.St.C., 1998. Nanomorphology of kaolinites:

comparative SEM and AFM studies. Clays and Clay Miner. 46

(2), 153–160.

Zbik, M., Smart, R.St.C., 1999. Atomic Force microscopy in the

aspect ratio estimation of colloidal kaolinite in clays for our future.

In: Kodama, H., Mermut, A.R., Torrance, J.K. (Eds), Proc. 11th

Int. Clay Conf., Publ. ICC97 Org. Committee, Canada, pp. 361–

336.

Zbik, M., Smart, R.St.C., 2002. Dispersion of kaolinite and talc in

aqueous solution: nano-morphology and nano-bubble entrapment.

Miner. Eng. 15, 277–286.