JOURNAL OF COLLOID AND INTERFACE SCIENCE 184, 535–541 (1996)
ARTICLE NO. 0649
Zeta Potential Measurements on Three Clays from Turkey
and Effects of Clays on Coal Flotation
SYED ABID HUSSAIN,* ŞAHIg NDE DEMIg RCIg ,† ,1
*Mining Engineering Department, University of Engineering and Technology, Lahore-31, Pakistan; †Department of Chemistry,
Art and Science Faculty, Middle East Technical University, Ankara, Turkey; and ‡Department of Mining Engineering, Engineering Faculty,
Middle East Technical University, Ankara, Turkey
Received February 16, 1996; accepted August 6, 1996
There is a growing trend of characterizing coal and coal wastes
in order to study the effect of clays present in them during coal
washing. Coarse wastes from the Zonguldak Coal Washery, Tur-
key, were characterized and found to contain kaolinite, illite, and
chlorite. These three clays, obtained in almost pure form from
various locations in Turkey, have been subjected to X-ray diffrac-
tion (XRD) analysis to assess their purity and zeta potential mea-
surements in order to evaluate their properties in terms of their
surface charge and point of zero charge (pzc) values. It was found
from XRD data that these clays were almost pure and their electro-
kinetic potential should therefore be representative of their colloi-
dal behavior. All three clay minerals were negatively charged over
the range from pH 2.5 to 11. Chlorite and illite have pzc at pH 3
and pH 2.5, respectively, whereas kaolinite has no pzc. The effect
of these clays in Zonguldak coal, wastes, and black waters on coal
flotation was studied by floating artificial mixtures of Zonguldak
clean coal (4.5% ash) and individual clay. The flotation tests on
coal/individual clay revealed that each clay influences coal flota-
tion differently according to its type and amount. Illite had the
worst effect on coal floated, followed by chlorite and kaolinite.
The loss of yield in coal was found to be 18% for kaolinite, 20%
for chlorite, and 28% for illite, indicating the worst effect of illite
and least for kaolinite during coal flotation. q 1996 Academic Press, Inc.
Key Words: coal wastes; coal washing; zeta potential; X-ray
diffraction; kaolinite; illite; chlorite; flotation.
INTRODUCTION
Clay minerals in coal not only are a source of ash but
also harmfully affect a number of processes during coal
washing, such as flotation, flocculation, and dewatering.
There are very few studies describing the effect of clays on
flotation and its relationship with their zeta potential (1a,
b). Froth flotation of coal is a physicochemical process based
on surface properties of coal which can be measured through
electrokinetic potential (zeta potential). In general, zeta po-
tential measurement of the solid–liquid interface can be used
1
To whom correspondence should be addressed.
535
AID JCIS 4544 / 6g19$$$$41 12-10-96 01:56:50
AND GÜLHAN Ö ZBAYOĞLU‡
to study the nature of a solid surface. The electrokinetic
measurements are therefore the relative values for the semi-
quantitative estimation of surface charge and adsorption. Sta-
bility and coagulation of colloidal dispersion are considered
in terms of electrokinetic data for systems that cannot be
tested by electrochemical techniques (2).
As flotation depends on water chemistry the presence of
clays may change it through an electrical double layer.
Therefore, zeta potential measurements of the clays can be
of value to process engineers for controlling the washery
processes. It was aimed to determine the zeta potentials of
illite, kaolinite, and chlorite obtained from various sources
in Turkey.
Hussain et al. (3, 4) and Hussain (5) characterized clay
minerals in coal, wastes, and black waters from Zonguldak
Coal Washery in Turkey and found that only three clays,
namely kaolinite, illite, and chlorite, are present. Semiquanti-
tative XRD analysis also indicated dominance of kaolinite
in coal and of illite in black waters (5–8). These clays, in
the forms of slimes, caused loss of recovery in coal flotation
through slime coatings on coal and/or bubble and increased
reagent consumption (9–11). Arnold and Aplan (1a) inves-
tigated the effect of clays on artificial coal–clay mixtures
and concluded that different clays influence coal flotation
differently, and therefore slimes should be well characterized
with respect to clays to determine their effect on actual float-
ing systems involving high ash coals.
MATERIALS AND METHODS
The clays used in this study and their particulars are given
in Table 1.
In X-ray diffraction analysis (XRD) a Jeol JSDX-100S
X-ray spectrometer diffractometer was used. All the analysis
were done using CuKa radiation with Cu/Ni electrodes and
goniometer speed of 17 (2u ) per minute. The slit widths
used were 17 for the divergence slide, 0.27 for the receiver
slide, and 17 for the scatter slide. For XRD analysis samples
were prepared as unoriented mounts, no cation saturation
0021-9797/96 $18.00
Copyright q 1996 by Academic Press, Inc.
All rights of reproduction in any form reserved.
coida
536 HUSSAIN, DEMIRCI, AND ÖZBAYOĞLU

TABLE 1
Turkish Clays Studied

Clay
type Location Variety and preparation

Kaolinite Eskişehir Esk 4045, Kaolin

64.6% passing 5 mm
81% kaolinite
XRD trace obtained showed mainly peaks of
kaolinite and quartz (Fig. 1)
Illite Ordu Crushed in jaw and rolls to minus 10 mesh,
then ground to minus 325 mesh Tyler in
porcelain ball mill XRD trace showed
peaks of micaceous clay minerals and
quartz (Fig. 2)
Chlorite Ordu Concentration after magnetic separation,
crushed and ground in porcelain mortar
grinder to minus 325 mesh.
XRD trace showed mainly chlorite peaks FIG. 2. XRD trace of illite (unoriented mount): I Å illite, Q Å quartz,
(Fig. 3) M Å micaceous clay minerals.

was used, and samples were not subjected to heating. The
cavity filling method of sample preparation was used. During
this preparation the finely ground sample is put in a cavity
and compressed with a spatula or a microscope slide is
moved back and forth. This may cause orientation of the
flaky clays such as chlorite and an unoriented mount may
seem to be an oriented mount. The XRD traces of kaolinite,
illite, and chlorite are shown in Figures 1–3.
A Rank Brothers particle electrophoresis apparatus MKII
(Cambridge, England) was used to measure the electropho-
retic mobilities. During the measurement a quartz rectangu-
lar cell and platinum electrodes were used. After the cell
was filled with the prepared suspension a voltage of 100 V
was applied across it. The velocity of the particles in suspen-
sion was measured at both the front and back stationary
levels in the chamber using the two inner surfaces as refer-
ence settings for focusing the microscope. By altering the
direction of the current after each mobility measurement,
polarization effects were minimized. The mobility of parti-
cles was computed after taking the averages of readings at
the front and back levels. Each reported mobility was the
mean mobility of 20 particles (10 on each level and 5 in
each direction), yielding a standard deviation of about 5%
(5). The equation used for converting observed mobilities
U (U Å q /E, where q is the particle velocity and E is
the applied field strength) into the effective electrokinetic
potential (zeta potential) q depends upon the value of the

FIG. 1. XRD trace of kaolinite (unoriented mount); A Å anhydrite, K

Å kaolinite, Q Å quartz, G Å geothite, R Å green rust. FIG. 3. XRD trace of chlorite (unoriented mount).

AID JCIS 4544 / 6g19$$$$42 12-10-96 01:56:50 coida

 EFFECT OF CLAYS ON COAL FLOTATION 537
dimensionless quantity xa, in which a is the radius of the U r j conversion. If a large (approximately spherical) parti-
particle (assumed spherical) and x is the quantity cle is an agglomerate of small spheres then the question of
which radius to use is seen to arise.

F 4pe 2Snz 2
ekT G 1/2
,
If ( x 1 smaller radius) is small ( õ0.1) then again it is
fairly evident that the larger radius governs the effective xa.
In all the above it has been assumed that the particle itself
familiar in electric double layer theory. From the expression is rigid (i.e., of infinite viscosity). In the absence of firm
given, x É 1 1 10 6 cm at 257C in water containing 1 mM evidence it seems best to treat all particles of radius 1 mm
of a 1:1 electrolyte. or less as rigid. Under all the above considerations it is
If xa ú 200 it will usually be sufficiently accurate to use assumed that the Smoluchowski equation given above was
the Smoluchowski formula, which in the (original) unration- used in the calculation of the nominal zeta potentials.
alized form is Each clay sample before measurement was prepared as a
master suspension in deionized distilled water containing
ej particles less than 10 mm after sedimentation of an already
UÅ , ground material in an agate mortar to minus 300 mesh British
4ph standards. The mobility of each particle was measured in
suspension after soaking it for 10 min at each pH between
where e is the permittivity of the suspending liquid. pH 2.5 and 11. The pH adjustment was done by adding a
Typical units in the older system would be U in microme- small amount of HCl and NaOH and by using a Fisher
ters per second under 1 V cm (usually then in the range (Accumet Model 230) pH meter.
from 0 to 6), in which case for water at 257C q in mV In order to study the effects of clays, namely kaolinite,
would be given by illite, and chlorite, a series of artificial mixtures of clean
coal (4.5% ash) and clays (ground to 0325 mesh B.S.) were
z Å 12.83U. prepared and flotation was carried out using kerosene oil as
collector (340 g/t) and methyl isobutyl carbinol as frother
If xa õ 0.1 then the Hückel equation would be relevant: (85 g/t), which were previously optimized (5). The flota-
tion tests were conducted in a Wedag laboratory flotation
ej machine having a fiberglass cell (capacity Å 500 ml) manu-
UÅ (unrat.).
6ph factured locally. The speed of the impeller was kept constant
at 1420 rpm. All the tests were conducted in distilled water
But this will hardly ever be useful in the case of water as (pH 6.5). Agitation and conditioning time were also con-
suspending medium since even water with no added electro- stants as three minutes each in all the tests. For each test 50
lyte (seen as air equilibrated) has x á 10 6 , so that particles grams of sample consisting of coal/clay mixture according
would have to be as small as 10 nm (0.01 mm) radius to to the amount of clays (0–25%) in each case was used,
reduce xa to 0.1. keeping the pulp density constant (10% solids by weight).
In the intermediate range of xa (0.1 to 200) the Henry The coal samples were ground to minus 28 mesh first in a
equation jaw crusher and then in rolls. Three clays found in Zonguldak
coal, namely kaolinite, illite, and chlorite, were used in this
ej study (3–5). These clays were obtained from various
UÅ [1 / f ( xa)] (unrat. form) sources in Turkey and on XRD analysis were found to be
6ph
almost pure (4, 5).
can be seen if the z potential is small (say õ20 mV), where
f ( xa) is a correction factor taking the values given below: RESULTS AND DISCUSSION

xa 0 0.1 1.0 5 10 50 100 ` The XRD traces showed that each clay selected for this
study contains only one clay mineral, along with nonclay
f ( xa) 0 0.001 0.027 0.160 0.239 0.424 0.458 0.5 impurities associated with clay, such as quartz.
In the traces the broad reflections starting from 3–47 (2u )
If the z potential is not small, then in the range xa Å 0.1 and terminating at about 13–147 (2u ) might be due to the
to 200 the best available computations of the U: z relation- amorphous materials and organic matter present in the sam-
ship are those of Wiersema, Loeb, and Overbeek (12). These ples or to the somewhat poor alignment of the goniometer
authors also considered the effect of different electrolyte (13) (Figs. 1, 2, and 3).
charge types. If particles are not spherical then, unless xa As expected from the preparation method, in chlorite the
is everywhere large, there must be some doubt about the XRD trace shows a regular pattern (Fig. 3). The chlorite

AID JCIS 4544 / 6g19$$$$42 12-10-96 01:56:50 coida

538 HUSSAIN, DEMIRCI, AND ÖZBAYOĞLU
FIG. 4. Zeta potentials of Turkish clays as a function of pH.
has a micaceous nature and it is difficult to grind the sample
to obtain a flakeless powder.
The results obtained for zeta potential values measured
against pH are plotted in Fig. 4 (5). All clays in Fig. 4
were found to have negative zeta potential in the pH range
observed (2.5 to 11). The kaolinite was the most negative
clay, ranging in zeta potential value from 024.0 to 049.5
mV. The illite was the least negative clay, giving a range
of 0 to 042.0 mV. The chlorite gave a range of 0 to 036.5
mV. A lower pzc value at pH 2.5 for illite and at pH 3.0
for chlorite was obtained.
The zeta potential measurements of Turkish clays showed
that kaolinite (unlike others) was more negatively charged
(lowest current) than chlorite and illite and therefore no pzc
value could be found, as this permanent negative charge of
kaolinite due to the isomorphic substitution remains indepen-
dent of pH in the acidic range (14–16). In the basic range,
the edge surface at which the octahedral sheet is broken
may be compared with the surface of an alumina particle.
Hydroxyl ions act as potential-determining ions and cause
the edge surfaces to be negatively charged (16). Therefore,
in kaolinite along the pH range studied no point of zero
charge (pzc) was obtained. Kaolinite was always negatively
charged, but it appeared less negative at about pH 5. A
noticeable increase in negative zeta potential at pH ú5 is
thought to be related to the pzc of the clay edge at about
AID JCIS 4544 / 6g19$$$$42 12-10-96 01:56:50
neutral pH (17, 18). Changes in zeta potential values in
the pH range 2.5–11 showed the presence of complex ion-
exchange reactions on heterogeneous 0SiOH and 0AlOH
sites.
Arnold and Aplan (1b) determined zeta potentials of coal
and clay as a function of pH in both tap and distilled water.
On comparison with zeta potentials of kaolinite determined
by Arnold and Aplan (1b) it was found that three types of
kaolinite, namely, A, B, and C, were used by them. Kaolinite
A and C showed completely different behavior than kaolinite
from Turkey, whereas kaolinite B showed no pzc value, like
kaolinite used in this study. The pzc value of illite was 2.5,
which is similar to the value determined for Turkish illite.
Chlorite could not be compared, as no chlorite zeta potential
value was determined by them (1b). It is evident that the
zeta potential of pure clays in distilled water is an intrinsic
property of clays. Clay is always negatively charged; how-
ever, an increase in zeta potential is noted at around pH 6
due to the pzc of the clay edge.
It was reported by Özbayoğlu (19) that a decrease in
zeta potential means better floatability, and it was found by
Arnold and Aplan (1b) that the presence of ions in water (tap
water) can reduce the overall negative charge of particles in
the pH range from 2.5 to 10. This means that less negative
clays are probably less harmful to coal flotation. Therefore,
in this case illite should give less depression of coal in flota-
coida
 EFFECT OF CLAYS ON COAL FLOTATION 539
tion than chlorite and kaolinite. The use of cations such as TABLE 2
Al 3/ and Fe3/ may easily reverse the charge of those clays Effects of Clays on Coal Flotationd
which are less negative (furnish pzc values easily), thus
enabling these clays to be depressed by the use of dispersants Clay added % Yield % Ash %
Name of
and depressants with advantageous use in flotation (1a). clay added 0 86.7 2.5
As coal is usually negatively charged and gives a negative
zeta potential in distilled water (5, 17, 19) due to its aniso- Kaolinite 5 83.3 3.0
tropic nature, and clays are also negatively charged, only 10 79.5 4.0
clays having charge opposite to that of coal should cause 15 76.9 5.0
20 72.1 6.0
greater problems by adsorbing their positively charged por- 25 68.9 7.0
tions on the coal surface, thereby preventing coal flotation. Chlorite 5 79.0 3.5
Hussain (5) measured the zeta potentials of clean coal 10 76.0 4.5
(4.5% ash) with three dispersants, sodium silicate, calgon 15 73.4 6.0
(sodium hexametaphosphate), and marsperse CB (sodium 20 70.2 7.0
25 66.5 8.0
lignin sulfonate), and three depressants, CMC (carboxy- Illite 5 77.5 4.0
methyl cellulose, Tylose C30), dextrin, and Auro Depressant 10 73.2 5.5
633, and found that only dextrin decreased the negative zeta 15 68.2 7.0
potentials of coal, with increased dosage showing better de- 20 64.3 8.0
pressing effects and better adsorption on coal surfaces. All 25 59.1 9.0
others were not found advantageous in floating the coal. a
Reagents; kerosene oil Å 340 g/t, methyl isobutyl carbinol 85 g/t.
Only dextrin, in the light of zeta potential measurement,
depressed fine high ash coal in single stage reverse flotation.
Effects of various molar concentrations of MgCl2 , CaCl2 , Å 4%) to be included in the artificial mixture of clean coal.
Ca(OH)2 , AlCl3 , and Al2 (SO4 )3 on zeta potentials of clean Flotation results showed that yield and ash values as percent-
coal were also studied by Hussain (5). The calcium cations ages were 70.0 and 9.0, respectively. Without clay, clean
were found to be less negative than the magnesium cations coal flotation gave a yield value of about 90% with an ash
due to higher adsorption with increasing cation radius. Alu- value of 2.5%, showing the effect of clays in reducing the
minum chloride and aluminum sulfate reverse the charge of yield and increasing the ash of the product (coal) (5, 22).
coal by adsorbing on negatively charged surfaces by electro- In the literature there are two opposite views about the
static forces even at lower concentrations. These findings effect of clay slimes on coal flotation. Most investigators
were in line with others (19–21). The results obtained by (10, 12) believe that slime coatings are the real cause of
floating clean coal (4.5% ash) and three different clays (ka- loss in mineral flotation recoveries. The slimes are heaviest
olinite, illite, and chlorite) are given in Table 2 and are when the slimes are uncharged or oppositely charged to the
plotted in Fig. 5. material being floated and thus badly affect the flotation
This figure shows the effect of addition of kaolinite, illite, results. Similarly other researches (13, 23, 24) note depres-
and chlorite (0–25%) on flotation of clean coal. It can be sion of coal due to the clay slimes. However, Firth and
seen that illite has the worst effect both on yield and ash Nicol (25) found no effect of kaolin (clay) on coal flotation.
when floated with clean coal. Chlorite was found to have an Another explanation about the bad effects of clays on flota-
intermediate effect on yield and ash, and kaolinite the least. tion through the entrainment or water carryover mechanism
It can be seen that up to 10% kaolinite and chlorite have no is reported by some investigators (26, 27). The controversy
appreciable effect on flotation of coal. On the other hand, on two opposite views regarding the effect of clays on flota-
up to 25% (maximum amount) of kaolinite and chlorite tion by slime coatings on mineral or coal and water carryover
clays reduced the yield to 18 and 20%, respectively. Illite is still unresolved. It may be said that both these mechanisms
addition up to 25% gave a loss of yield to about 28% and play a part in spoiling the results of coal flotation. The type
increase of ash to 6.5%, showing that coal flotation is af- of clay and the amount have major roles in coal depression,
fected more by illite than by kaolinite. Comparatively higher whereas the water carryover mechanism has a minor effect.
depression of coal by illite than by kaolinite on U.S. coals It was noted that water carryover in coal mainly contributed
was indicated by Arnold and Aplan (1a), in agreement with toward ash increase of the floated product due to clay slimes
the findings of this study. and the coal depression was negligible (i.e., yield was not
In order to compare the effects of clays in amounts deter- affected). The depression of coal can therefore be related
mined by XRD analysis of coal samples from Zonguldak mainly to the clay type and its slime coatings on coal parti-
region (K Å 52%, I Å 31%, Ch Å 17%), these were con- cles, which mask them from the reagent adsorption. How-
verted to their proportions of maximum amount 25% (5). ever, both the mechanisms, one (clay type and amount)
This gave the amount of each clay (K Å 13%, I Å 8%, Ch reducing the yield and the other (water carryover) lowering

AID JCIS 4544 / 6g19$$$$43 12-10-96 01:56:50 coida

540 HUSSAIN, DEMIRCI, AND ÖZBAYOĞLU
FIG. 5. Effect of kind and amount of clay on yield and ash of clean coal (4.5% ash) in flotation.
the grade by increasing the ash of the floated product, are
operative in coal flotation. It is fortunate that Zonguldak
washery coal did not contain any montmorillonite type clays,
which are very harmful for U.S. coal flotation (12, 28–30),
and therefore a less gloomy picture has emerged. However,
the effects of illite, kaolinite, and chlorite in Zonguldak coal
cannot be ignored while this coal is floated.
CONCLUSION
The conclusions drawn from the present investigation are
summarized as follows:
Zeta potentials of all Turkish clays (kaolinite, illite, and
chlorite) determined in distilled water over a pH range from
2.5 to 11.0 were all negative, ranging from 013.1 to 049.5
mV.
Kaolinite was the most negative clay, compared to chlorite
and illite, found electronegative in decreasing order.
No point of zero charge of kaolinite could be found be-
cause of its permanent negative charge due to isomorphic
substitution and/or structural defects. Illite surface exhibited
a point of zero charge (pzc) at pH Å 2.5 and chlorite surface
at pH Å 3.0.
Although zeta potential measurements of these three clays
do not give an absolute clue to their behavior toward flotation
or any other process, they may reflect their nature and hence
these clays may be dealt with accordingly by washery engi-
neers.
AID JCIS 4544 / 6g19$$$$43 12-10-96 01:56:50
Flotation tests on artificial mixtures of clean coal and
individual clays revealed that each clay influences coal flota-
tion differently according to its type and the amount used
for flotation. Illite was found to have the worst effect on
coal flotation by reducing the yield and increasing the ash
of the floated coal, followed by chlorite and kaolinite in
decreasing order. Clay coatings on coal and water carryover
of ultrafine clays were both responsible for deteriorating
yield and ash of floated coal.
REFERENCES
1. (a) Arnold, B. J., and Aplan, F. F., Int. J. Miner. Process. 17, 225
(1986); (b) Int. J. Miner. Process. 17, 243 (1986).
2. Leja, J., ‘‘Surface Chemistry of Froth Flotation.’’ Plenum, New York,
1982.
3. Hussain, S. A., Demirci, Ş., and Özbayoğlu, G., in ‘‘The First Interna-
tional Conference on Modern Process Mineralogy and Mineral Pro-
cessing,’’ p. 228, 1992.
4. Hussain, S. A., Demirci, Ş., and Özbayoğlu, G., Appl. Clay Sci. 7, 471
(1993).
5. Hussain, S. A., ‘‘Characterization of Coal and Wastes from Zonguldak
Coal Washery and Effect of Clays on Coal Flotation,’’ Ph.D. Thesis.
Middle East Technical University, Ankara, Turkey, 1992.
6. Hussain, S. A., Özbayoğlu, G., and Demirci, Ş., in ‘‘International Con-
ference on Recent Advances in Materials and Mineral Resources, 3–
5 May, Penang, Malaysia,’’ p. 554. Pevak Foundation Publication,
1994.
7. Hussain, S. A., Demirci, Ş., Özbayoğlu, G., and Atabey, E., Int. J.
Environ. Issues Miner. Energy Ind., 101 (1993).
8. Hussain, S. A., Özbayoğlu, G., and Demirci, Ş., in ‘‘Proceedings of
coida
 EFFECT OF CLAYS ON COAL FLOTATION 541
Second National Symposium on Modern Trends in Contemporary Turkish Bituminous Coal Seams in Zonguldak Basin.’’ Ph.D. Thesis,
Chemistry, April 4–6, 1995, Islamabad, Pakistan,’’ p. 317. M.E.T.U. Ankara, 1977.
9. Sun, S. C., Trans. A.I.M.E. 153, 479 (1943). 20. Wen, W. W., and Sun, S. C., Trans. A.I.M.E. 262, 174 (1977).
10. Fuerstenau, D. W., Gaudin, A. M., and Miaw, H. L., Trans. A.I.M.E. 21. Sarikaya, M., ‘‘Determination of the Flotation Characteristics of Oxi-
211, 792 (1958). dized Coal from Zonguldak Coal Basin.’’ Ph.D. Thesis, M.E.T.U. An-
11. Iwaski, I., Cooke, S. R. B., Harraway, D. H., and Choi, H. S., Trans. kara, 1988.
A.I.M.E. 223, 97 (1962). 22. Hussain, S. A., and Özbayoğlu, G., in ‘‘International Conference on
12. Wiersema, P. H., Loeb, A. L., and Overbeek, J., J. Colloid Interface Recent Advances in Materials and Mineral Resources, 3–5 May, 1994,
Sci. 22, 78 (1966). Penang, Malaysia,’’ p. 562. Pevak Foundation, 1994.
13. Brindley, G. W., and Brown, G., ‘‘Crystal Structures of Clay Minerals 23. Mishra, S. K., in ‘‘Australasian I.M.M., Mill Operators Conf., Mt. Isa,
and Their X-Ray Identification.’’ Mineralogical Soc. Monograph No. 1978, p. 159.
5, London, 1980. 24. Lynch, A. J., Johnson, N. W., Manlapig, E. V., and Thorne, C. G.,
14. Ferris, A. P., and Jepson, W. B., J. Colloid Interface Sci. 51, 245 ‘‘Mineral and Coal Flotation Circuits: Their Simulation and Control.’’
(1975). Elsevier, Amsterdam, 1981.
15. Wierer, K., Eckl, J., Oberndorfer, J., and Dobias, B., in ‘‘XVI Interna- 25. Firth, B. A., and Nicol, S. K., Int. J. Miner. Process. 8, 239 (1981).
tional Mineral Processing Congress’’ (E. Forssberg, Ed.) p. 819. Else- 26. Warren, L. J., in ‘‘Principles of Mineral Flotation’’ (M. M. Jones and
vier, Amsterdam, 1988. J. T. Woodcock, Eds.), The Wark Symp. Series, Vol. 40, p. 185, 1984.
16. Aplan, F. F., and Feurstenau, D. W., in A.I.M.E. 50th Anniversary 27. Subrahmanyan, T. V., and Fossberg, K. S. E., Int. J. Miner. Process.
Volume,’’ p. 170, 1962. 30, 265 (1990).
17. Van Olphen, H., ‘‘An Introduction to Clay Colloid Chemistry.’’ Wiley, 28. O’Gorman, J. V., and Walker, P. L., Jr., Fuel 50, 135 (1971).
New York, 1963. 29. O’Gorman, J. V., and Walker, P. L., Jr., ‘‘Mineral Matter and Trace
18. White, G. N., and Zelanzy, L. W., Clays and Clay Miner. 36, 141 Elements in U.S. Coals.’’ U.S. Dept. of Interior, Office of Coal Re-
(1988). search, Report No. 2, 1972.
19. Özbayoğlu, G., ‘‘Determination of Flotation Characteristics of Several 30. Gluskoter, H. J., J. Sed. Petrol. 37, 205 (1967).

AID JCIS 4544 / 6g19$$$$43 12-10-96 01:56:50 coida