72

A Comparative Study on the Emulsification of

Flotation Reagents*

Amin A. Abdelrahman**

This paper presents the results of a comparative study of three coal flotation systems, one currently being used,
the others being suggested in this work for the treatment of superfine coal matrix. The matrix is a very fine rich
tailings coal, i.e. 54.8 percent ash with 60% of the sample passing through a 46 microns screen. The suggested
systems are simply pre-emulsification of the collectors in the cell of froth flotation, prior to the feed being introduced
into the flotation unit.

INTRODUCTION
The froth flotation process has been researched and employed in many countries throughout the
world, to dress coal. From the reported literature it would appear that coal fines cleaning is almost
exclusively by froth flotation.
The complete area of fine coal cleaning is fraught with difficulties. The raw feed to coal flotation
plant typically contains large quantities of super fine clays. By the mechanism of slimes coating, these
reduce reagent adsorption and result in increased coal losses to the tailings (1). At the same time, clay
slimes are entrained in the produced froth and give rise to high ash levels in the coal concentrate high
reagent levels are also apparent.
One means of surmounting the problem of super fine material, is to remove it from the flotation
feed by a size separation, at say, 30 microns. This should reduce flotation capacity requirements and
improve efficiency (2). However, whilst the efficiency of separation of the 500 microns-I-30 microns
fraction is improved, the 30 microns product from the initial separation stage will still contain
significant proportion of combustible material. This must be recovered. Two possible methods of
achieving this are: The use of oil agglomeration as a complementary process to flotation or by
introducing the reagents to the flotation cells in an emulsion form (i.e. reagent/water emulsion)

The previous work on the emulsification of reagent in coal flotation, although promising in
respect of improved flotation efficiency, (i.e. increased flotation machines capacities and some
reduction in reagent consumption) was disappointing due to the very high operating and chemicals
costs.
Generally, emulsification of oil is carried out by the means of a Jet emulsifier or the use of
surfactants (i.e. chemical emulsification) (5, 6, 7).
In the course of the last years, significant advances have been made in the rheology of coal
preparation slurry and its effects on the treatment operation (3, 8, 9). In the light of these and from the
context above it is apparent that regardless of the efficiency of froth flotation, the final success of this
process is mainly dependent upon adequate dispersion of the reagents of flotation. This specifically
the case where slurries of high densities are to be treated, or a shortage of water is to be faced. As
such, any improvements in the system of emulsification of reagents could be of vital importance.
The present paper describes one simple possible method of producing these significant improvements.
( )
 A Comparative on the Emulsification of Flotation Reagents 73
Material and Experimental Procedure
The coal used was very fine rich tailings and had an overall ash of 54.8 percent (dry basis). A 4
percent of this sample was greater than 212 microns in size, while the superfines material (passing 46
microns) comprised 60 percent.
The reagents which were put forward to perform this work, were those which are used
industrially in coal cleaning (10, 11, 12, 13). They are Pine Oil as a frother and Oleic Acid, Fuel Oil
and Kerosene as collectors.
The froth flotation tests were carried out in a Denver laboratory flotation machine with a capacity
of 2 litres.
Hence the initial objectives of this work were to study the effects of emulsification of flotation
reagents prior to feeding the coal into the cell with the intention of this system reducing the cost of
flotation of very fine coal.
To verify this proposal, it was required that the basic flotation process (i.e. Normal flotation
process) should be established in order to identify the standard values of the flotation variables and for
comparison with the suggested system. Two modifications for the flotation are proposed here in
addition to the basic flotation process were investigated. These methods are emulsified reagent/dry
coal and emulsified reagent/slurried coal. The difference between these systems lies in the form in
which the coal is fed to the flotation cell. In the former, a dry coal is introduced into the cell following
the completion of the emulsification process, while in the latter the feed is slurried coal instead of the
dry coal.
The results of the standard conditions for the froth flotation basic process are summarized in
Table 1 below.
With regard to the emulsification of the collectors; it should be noted that the flotation cell was
initially 3/4 filled with fresh water and then the collector was added, while the impeller was rotating at
1500 rpm and with the air inlet closed. After a half minute the process of emulsification was
completed, this was easily determined through the observation of the water color which did become
milky. Then the coal and the frother were poured off together into the cell. Mixing was conducted
within the cell at 1200 rpm for the conditioning period. This period was found to be shortened to

Table l. Normal Flotation Conditions

Variable Value Variable Value
Solid 10% pH of Pulp 7.2
Concentration w/v 5 Frother Dosage 0.1
Conditioning Time mins lb/ton
Collector Dosage
lb/ton 3. Kerosene
1. Oleic Acid 0.75-30
0.15-2
2. Fuel Oil Impeller Speed of
1-30
Flotation Time 8 mins Conditioning
1200 rpm
Impeller Speed of
900 rpm
Flotation Stage
21

2 minutes instead of 5 minutes which is established for the basic flotation process. The impeller speed
was then reduced to 900 rpm to allow the flotation to proceed. The air inlet was opened.

Vol. 35, No. 2 ('88-0 ( )

 74 Amin A. Abdelrahman
Determination of weight yield and combustible recovery were obtained through filtration, drying and
ash analysis of the products.
By adopting the above mentioned emulsification system, it was surprising to find out that the
time spent in flotation for the standard method, was reduced from 8 minutes to approximately
4 minutes.

Results and Discussion

The results of the tests are given in Tables 2 to 4 and shown by the comparison of the
combustible recoveries (Figs. 1 to 3). In these Tables, columns 4, 5, 6 and 7 show the effects of pre-
emulsification of collectors, while columns 2 and 3 give the comparative results for normal reagent
additions. It was intended to describe the results of each collector individually. However during the
course of the work there was an almost natural progression to update suggestions concerning the
effects of pre-emulsification on the process performance when the results of each reagent did become
available,
Table 2. Results of Oleic Acid
Oleic Acid Efficiency
Emulsified System
Normal System Dry Feed
Slurried Feed
Dosage Combustible
Combustible
1b/ ton
Combustible Ash o/ Ash %
Ash % Recovery % Recovery 0/0
(1) Recovery %
(2) (3) (4) (5) (6) (7)

o. 15 21. o 50. 30 15. o 83. 4 18. o 64.5

o. 30 23. o 52.90 14. 2 85. 8 17.5 67. o
o. 60 24. 5 56. 60 13. 5 88. 4 17. 0 70. o
o. 90 26. o 56.48 13. o 88. 8 16. 8 72. o
1. 25 33. 5 60. 80 13. 5 90. 5 18.o 72. 6
1. 50 32. o 60. oo 13. o 91. 6 19.o 73. 5
2. 00 34. 5 61. 50 12. o 92. 4 20. o 75. o

Table
3. Results of Fuel Oil

Efficiency
Fuel oil Normal System Emulsified System Slurried Feed
Dosage Dry Feed Combustible
Combustible Ash Combustible Ash %
lb/ton Ash % Recovery 0/0 % Recovery % Recovery 0/0
(1) (2) (3) (4) (5) (6) (7)

1.0 16. o 50. o 13. 70. o 17. 0 56.0

1. 25 17.0 53. o 5 74. o 16. 5 58.0
2. 50 18. o 56. 2 12. 89. 5 15. o 60.1
5. 00 18.5 52. 6 8 92.7 14. o 63.0
19.5 53. 6 93. o 13. 5 67.0
10. oo 13.
19.o 55. o 93.2 15. o 69.0
15. 00 8
20. o 58.0 93.2 14.5 71. o
20. oo 12.
60. o 94. o 15.o 73. o
25. oo 22. o 5
30. 00 22. 5 63.3 12. 96. 4 14. 5 76. o
o
11.

( )
 A Comparative on the Emulsification of Flotation Reagents 75
5
11.
2
10.
2
9.
0
22
Study

Table 4. Results of Kerosene

Effciencv
Kerosene Normal System Emulsified System
Dosage Dry Feed Slurried Feed
Combustible Combustible Combusti
lb/ton Ash % Ash % ble Ash %
Recovery % Recovery 0/0 Recovery %
(1) (2) (3) (4) (5) (6) (7)

o. 75 15. o 51. 2 13. 2 89. 4 14. 5 70. o

1. 50 16. o 51. 8 13. o 91. o 14. o 73. 5

2. 50 16. 5 52. o 12. 5 92. o 13. 7 75. o

5. 00 15. o 51. 5 12. o 93. o 13. 5 77. o
7. 50 16. o 52. o 11. 8 93. o 13. o 78. o
10. 00 14. o 52. o 11. o 93. 3 12. 8 79. 2
12. 50 17. o 53. 6 10. o 94. 3 12. 2 80. o
15. 00 18. o 53. 8 9.2 94. 6 12. o 80. 5
20. 00 16. o 52.4 9.0 94. 6 11. 5 81. o

25. 00 16. 5 52. 5 95. 1 11. o

30. oo 15. o 52. o 95. 2 10. 5

which have shown similarity, i.e. all curves follow the normal pattern with regards to reagent dosage,
increased reagent gives increased combustible recovery, hence a basic model describing the various
effects is proposed.
The main differences between the two sets of results (i.e. of normal and emulsified systems) are
that pre-emulsification has improved the effciency of the process. The general improvements are seen
in all of the results. The most prominent feature of these results is in the very dramatic fall in ash
content of the concentrate in all collectors systems without any exception. For example, at 1.25 lb/ton
Oleic acid, the normal addition gives concentrate with ash content of 33.5 0/0, the same dosage, when
the oleic acid was pre-emulsified, produces an ash of concentrate of about 13 0/0, while

Vol. 35, No. 2 ('88-0 ( )

 76 Amin A. Abdelrahman
simultaneously the combustible recovery was increased from 60.8 to 90.5% at the same reagent level.
It is apparent also from Figs. 1 and 3 that fuel oil and kerosene, as expected, exhibited selectivity
IOO

90
B 80
70
% 60

EO 50
40
30
5 10 15 20 25 30
Collector Addition (1b/ton)
Figure 1. Fuel Oil
23

IOO

90

> 80

0
Emulsified Reagent/Dry Coal
A
Emulsified Reagent/ Slurried
Coal o Normal Flotation
System

70
Q)

60
50
40
30
0.5 1 1.5 2
Collector Addition (1b/ton)
Figure 2. Results of Oleic Acid

( )
 A Comparative on the Emulsification of Flotation Reagents 77

o Emulsified Reagent/ Dry Coal

Emulsified Reagent/Slurried
Coal o Normal Flotation System

IOO

90
80
70

60
50
40
30
5 10 15 20 25 30
Collector Addition (lb/ton)
Figure 3. Results of Kerosene

towards the coal in a coal/shale matrix. It was not anticipated however that oleic acid (anionic reagent
would also produce a low ash concentrate, since the principal constituent of shale is kaolinite (14).
Kaolinite has a plate like structure consisting of negatively charged faces and positively charged
edges (15, 16). Anionic reagent such as oleic acid will be attracted to the edges producing a
hydrophobic surface on shale, subsequently the shale will report with the froth floats. The results
shown in Table 2 do not support this. This is probably due to the fact that shale being completely
wetted preferentially by water and in the aqueous phase (i.e. water only) the high shearing rate
enables good dispersion of collectors and hence improved coal collection. The mobility of emulsified
oil is greater and has improved spreading power leading to pick up the hydrophobic super fine coal
particles coupled with improved shale rejection due to the hydrophilic nature that shale exhibits.
In view of the similarities between the three tests series, again emulsified oils are more mobile
and have improved spreading power rendering coal particles more compatible with oils. Surprisingly
24
Study

1 Water Tank
2 . Oil Tank

Vol. 35, No. 2 ('88-0 ( )

 78 Amin A. Abdelrahman
3 . Emulsification Cell
4 . Conditioning Cell

5 . Feed Sump
6 . Frother Tank
7 . Flotation Cells

Concentrate

Plate 1. A Proposed Flowsheet for Pre-emulsification Froth Flotation Process

for both systems of kerosene (i.e. normal and emulsified) the selectivity of coal occurred and a good
separation was obtained. It is considered that this may have resulted from kerosene being characterised
by very low kinematic viscosity, which, of course, will affect the rheology of the flotation pulp. For
example, the viscosity of fuel oil is 0.5 stokes, while of kerosene 0.016 at 20 0 C (16).
These values indicate that the viscosity of fuel oil is 31 times of kerosene at the same
temperature. Finally, some comments with regard to the more highly pronounced selectivity and
improved recovery of the dry feeding system for all series of tests of the three collectors. The results
of slurried feed indicate ash content of the product being higher than those of dry-feeding system. The
corresponding combustible recoveries were low, suggesting that wetting of coal particles prior to the
addition of oils had decreased the spreading power of these oils onto the surface of coal particles;
hence the coal collection belittled clearly. Prime reasons for this, are the higher resistance offered by
the finer coal particles for the replacement of water by collector after being wetted, and the higher
moisture retained by capillary attraction, at the same time, within the coarser coal particles. With the
dry feeding system, one must assume that both coal and shale particles are very thirsty, when these
particles are to be found in an oil/water emulsion, a selective attachment of oil to the coal occurred,
shale is not similarly wetted by oil, but being wetted preferentially by water. This should improve
selectivity and rise combustible recovery.

( )
 A Comparative on the Emulsification of Flotation Reagents 79
Summary
To summarise the results of this work, the successes achieved with pre-emulsification of the collectors are employed to
concentrate the fine coal particles, (being a component of a rich tailings fine coal matrix.) are thus :
With the introduction of collectors into the fresh water within the flotation cell, prior to the addition of coal material, (dry
or slurried.) the process produced a clean coal with an average ash of 13%, giving a combustible recovery of over 90% for dry
feed and of over 70% for the slurried feed.
At economical reagent levels, i.e. 0.15 to 2.5 lb/ton, an increase of between 20% and 34% greater recovery
than that obtained from the normal system, was present. At the same time, a decrease in the ash content of the
concentrate, varied from 5% to 12%.
Emulsification also increased the flotation rate; i.e. the total period to float this coal was reduced by 50%, allowing

25

an increase in the capacity of the flotation machines, equal to twice that of the normal standard flotation process.
An emulsification flotation flowsheet (see plate 1) is therefore suggested. As far as practical implication are concerned, the
validity of the proposed technique, particularly, relating to the method of emulsification is unquestionable. While oil being
emulsified in an aqueous phase by a simple mixer, reagent dispersal will be optimum and energy savings possible.
It is also expected that a reduction in reagent consumption by half, would occur and simultaneously improve the
performance of the process.

Acknowledgement
The author wishes to express his sincere appreciation to his wife for proof reading.

LITERATURE CITED
1. Collins, D. N. and A. D. Read, The Treatment of Slimes, Mineral Sci. Eng., 3, No. 2, p. 19 (1971).
2. Brookes, G. F. , Miles, N. J. and M. G. Ayat, The Possible Use of Small Cyclones in Coal Preparation,
Mining Dept. Magazine of Nottingham Univ., 34, p. 72 (1982).
3. Abdelrahman, A. A. , Spherical Agglomeration of Coal by Diesel Oil, Ph. D. Thesis, University of
Nottingham, Nottingham, U.K. (1983).
4. Plaksin, I. N. and V. I. Klassen, Sci. Technical Progress in Coal Flotation, Fifth International Coal
Preparation Congress, Section A: Paper 4-1.
5. Bensly, C. N. et al., The Effect of Emulsification on the Selective Agglomeration of Fine Coal, Int. Journal
of Min. Processing, 4, p. 173 (1977).
6. Abdelrahman, A. A. and G. F. Brookes, An Economic Assessment of Using Surfactants in Cleaning Coal by
the Oil Agglomeration Method, Energy Progress, 7, No. 1, p. 47 (1987).
7. Melik-Gaykazian, V. I. et al., About Emulsification of Butyric Reagents in Industrial Scale and
Determination of Disperity of the Received Emulsion, Koks i Khimia, No. 3, p. 9 (1964).
8. Brookes, G. F. , The Rheology of Coal Preparation Slurries, Ph. D. Thesis, Univ. of Nottingham,
Nottingham, U.K. (1967).
9. Mersmann, A. and H. Grossman, Dispersion of Immiscible Liquids in Agitated Vessels, Int. Chem. Eng., 22, No. 4, p. 581
(1982).
10. Mining Chemicals Hand Book, Mineral Dressing Notes No. 26, Cyanamid of American, p. 80 (1976).
11. Oliver, C. Ralston, Froth Flotation and Agglomerate Tabling of Nonmetallic Minerals, J. of the Canadian
Institute of Mining and Metallurgy, p. 691 (1937).
12. Cremer, H. YV., Chemical Engineering Practice, London, Butterworths Scientific Publications, p. 209
(1957).
13. Fuerstenau, D. W. , Froth Flotation, New York, the American Institute of *Mining, Metallurgical and
Petroleum Engineers, Inc., 50th Anniversary Volume, p. 518 (1962).
14. van Olphen, H. , An Introduction to Clay Colloid Chemistry, London, John Wiley and Sons, p. 51 (1963).
15. Marshall, C. E. , The Colloid Chemistry of the Silicate Minerals, New York, Academic Press Inc., p. 14 (1949).
16. Chilingarian, G. V. and P. Vorabutr, Drilling and Drilling Fluids, Amsterdam, Elesevier Scientific
Publishing Company, p. 14 (1981).

Vol. 35, No. 2 ('88-0 ( )

80 Amin A. Abdelrahman
26

( )