Table of Contents

Search

The 6th Saudi Engineering Conference, KFUPM, Dhahran, December 2002

Author Index

Vol. 2.

473

RECOVERY OF POTASSIUM SALTS FROM KILN DUST OF

LOCAL CEMENT PLANTS

M. A. Daous1
1: Chemical & Materials Engineering Department, King Abdulaziz University
P. O. Box 80035, Jeddah 21589, Saudi Arabia
E-mail: [email protected]

ABSTRACT
Enriched levels of soluble potassium sulfates in alkali rich cement kiln dust (CKD) makes such dust a
promising raw material for the extraction of potash or its soluble salts. This study was carried out to
investigate leaching of potassium salts from CKD produced in a local cement plant using water, and
20 wt.% calcium nitrate solution as leachants. Leach tests at several liquid/solid weight ratios,
different reaction temperatures, and times revealed that 78.7 % of potash could be extracted using
water at a 5:1 liquid/solid weight ratio, upon leaching for 15 minutes at 70 oC. Similar tests carried
out using 20 wt. % calcium nitrate solutions revealed that 96.1 % of potash could be extracted upon
45 minutes of leaching at these same liquid/solid weight ratio and temperature. Furthermore, the large
reduction in the alkali content of CKD upon such treatment makes this material suitable to recycle as
a kiln feed, as well.

Keywords: cement kiln dust, potash extraction, water leaching of CKD, calcium nitrate leaching of
CKD

.

/ . %
%,
.o / :
%.
. /

.

Vol. 2.

474

M. A. Daous

1. INTRODUCTION
Hundred thousands of tons of cement-kiln dust (CKD) are generated annually from cement
plants in the Kingdom of Saudi Arabia. The bulk of this dust, mostly with high alkali contents,
is land filled with a significant financial loss to the local cement industry in terms of the value
of raw materials, processing, and energy consumption during pyroprocessing, dust collection,
and disposal. This fine dust, which emits from cement kilns, is not totally recycled to prevent
the build up of excessive salts in the cement product. The alkali salts in the dust are derived
from the clay raw materials, which include potassium and sodium feldspars. CKD consists
mainly of partially calcined kiln feed, and as such, is generally suitable for recycling to the
kiln. However, continuous recycling of CKD to the kiln can soon offset the equilibrium
between the alkali levels in the kiln feed and the internal cycle of volatilization and
condensation within the kiln, resulting in high alkali clinker. This, of course, is not suitable for
plants intending to make low alkali cements that meet the national Saudi requirement of low
alkali cement, which specifies a limit of 0.6% equivalent of Na2O for contained alkali. The
alkali content must be kept low to prevent concrete from decrepitating, which might result
from a reaction between the alkali and silica in some reactive aggregates in concrete.
Various methods for utilizing CKD in industrial applications, including existing or proposed
methods for alkali removal, are reported in the literature. Bhatty (1995) provides a general
review of these methods. Because of the generally high lime content of CKD and subsequent
ability to harden upon exposure to moisture, CKD has been used as a binder in soil
stabilization suitable for a sub-base in streets and high way construction. It is commonly used
as a mixture with different solid- waste materials such as waste glass, fly ash, and wastewater
sludge with the addition of cement or other admixtures if necessary (Nicholson, 1978; Sayah,
1993; Miller et. al., 1980, and Bhatty, 1983)
A great deal of work has been done on the use of CKD in blended cements. Most published
work report a loss in strength, setting time, and workability due to the presence of high alkalis
in the dust. It is also been reported that addition of fly ash to these blends tends to improve
these properties and reduce the potential problems from alkali-silica reaction (Daous, 2000;
Klemm, 1993; Bhatty, 1984; Daugherty and Funnel, 1983, Wills, 1983).
CKD has been used as a substitute for lime in stabilizing wastewater streams. This is possible
primarily because of the high neutralizing potential of the CKD (high CaCO3 and CaO) and
fine particle size distribution having Blaine finenesses often greater than 800-m2/ kg. Up to
35% addition of CKD has satisfactorily met the specified pathogen control level in sewage
sludge. High PH, an exothermic reaction, and the resulting accelerated drying when CKD was
added were the factors mainly responsible for the pasteurizing effects on municipal
wastewater sludge (Nicholson, and Burnham, 1988; Burnham, 1987).
Other methods reported in the literature suggest the use of CKD directly as a fertilizer
material because of its high lime content (for acid neutralizing capacity) and its high
Top

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

475

potassium concentrations (Risser et. al., 1981; Baker et. al., 1975; Preston, 1993). Other
miscellaneous methods include the use of CKD as a partial replacement for soda in the
production of glass where color and high chemical stability are not essential (Fraiman et. al.,
1991), as a decolorizing agent in the clarification and purification of sugar solutions (Klemm,
1980), in stabilizing soils contaminated with polychlorinated biphenyls (PCBs) (Einhaus,
et. al., 1991), as an absorbent for cleaning up spills in garages and roads (McCants, 1952), and
as an additive to remove undesirable elements from leachates produced in water leaching of
fly ash (Duchesne and Reardon, 1999).
Enriched levels of soluble alkalis (i.e., K2SO4) in CKD make it a promising raw material for
the extraction of K2O or its soluble salts. A number of patents have been issued on the subject,
most of which utilize aqueous dissolution of soluble potassium salts that are later removed as
a saleable commodity and the water recovered for reuse. The potassium-free dust is usually
recycled as kiln feed.
Dean (1972) recovered a substantial amount of alkali salts from CKD by mixing it with H2O
in successive stages at temperatures between 21 and 99 oC. The supersaturated solution thus
produced was cooled and alkali salts crystallized and separated. Singleton and Bruce (1975)
treated the dust with fresh water at elevated temperatures to essentially dissolve alkali
chlorides and sulfates and subsequently remove them by crystallization. The remaining
solution is carbonated to form calcium carbonates for recycling
McCord (1977) recovered the alkalis by leaching the kiln dust at higher temperatures with an
aqueous solution of KCl. The extract is cooled for crystallization of KCl and removed. The
washed CKD is reclaimed for use as kiln feed. Klemm (1980) proposed the separation of
alkalis as K2SO4 from the high alkali-dust (~ 13 % K2O) for agricultural purposes. The K2SO4
was extracted by mixing CKD with hot water followed by filtering, cooling the aqueous brine,
and K2SO4 crystallization. The remaining material can then be used as a raw feed substitute. A
yield of more than 86 % was reported by lowering the PH with carbon dioxide, thus
preventing the formation of the double salt of K, and Ca sulfates known as syngenite.
Wolfe (1992) reported on leaching tests to remove alkalis from fresh CKD and CKD
stockpiled over the years. Tests on the stockpiled CKD showed that more than 80 % alkali
(K2O) was removable by using ordinary tap water with or without additives at ambient
temperature. Almost 100 % more alkali was removed from stockpiled CKD than from fresh
CKD.
Another approach, proposed by Wilson and Anable (1986), is the removal of alkalis by
pyrotreatment of palletized dust. Here the K2O is volatilized in the presence of SO3 by heating
the dust above 1300 oC in a reducing environment generated by coal addition. The resulting
low-alkali residues were tested as Portland cement, which exhibited strength greater than
ASTM standards.

Top

Vol. 2.

476

M. A. Daous

Earlier, Klemm (1980) reported that high-alkali kiln dust could be fused to separate K2SO4 as
liquid. Several operational difficulties were cited, however. First, although K2SO4 in the dust
has a melting point of 920 oC, the dust itself had to be melted beyond 1400 oC to volatilize
K2SO4 and need to be separated from the fused dust. Second, the K2SO4 is reactive and fuses
with ceramics, and third, the K2SO4 fumes tend to react with the furnace heating elements and
eventually inhibit their proper functioning.
The chemical, mineralogical, and physical compositions of CKD varies from one plant to
another. These depend on raw material, type of kiln operation, dust collection method, and
type of fuel used in the plant (Klemm, 1993).
The objective of this study was to investigate leaching cement kiln dust, produced in local
Saudi cement plants, using water or an aqueous solution of calcium nitrate as leachants to
produce potassium or potassium-nitrogen fertilizers. In this communication, the results of a
series of CKD leaching experiments conducted at three liquid/solid weight ratios, four
different leaching temperatures, and several leaching times are presented. The data are used to
examine the effects of these important parameters on the leaching process and to determine
the most suitable conditions for aqueous leaching of CKD and leaching with aqueous solution
of calcium nitrate. This study constitutes an important step towards the utilization and
treatment of large quantities of waste CKD produced annually in the country to make it
suitable to recycle as a kiln feed, and thus alleviate a standing disposal and environmental
problem, and produce value added products in the process.
2. MATERIALS AND METHODS
2.1 Materials
The cement kiln dust used in this study was obtained from the Arabian Cement Company Ltd.
in Rabigh, Saudi Arabia. The sample, which was rich in calcium, and alkalis, represents dry
materials collected from the electrostatic precipitators above the rotary kiln. The chemical
analysis of CKD and the corresponding analysis of Portland cement produced by the same
company are given in table (1). The CKD had a solid density of 3.08 g/cm3, a bulk density of
1.17 g/cm3, and a porosity of 0.62. The calcium nitrate used in this study was BDH Analytical
Reagent-Grade Ca(NO3).4 H2O.
2.2 Experimental
For each of the leachants (distilled water, and 20 wt% calcium nitrate solution) used in this
study, a series of batch leaching experiments were performed by reacting samples of CKD at
3:1, 5:1, and 10:1 liquid: solid weight ratios for different periods of time ranging from 5 to
60 minutes, and at four reaction temperatures of 30, 50, 70, and 80 oC. All experiments were
carried out in 250 ml glass conical flasks placed in a thermostatically controlled horizontal
shaking bath maintained within 0.5 oC of the set temperature.
Top

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

477

The necessary number of replicate samples for each liquid:solid mix was reacted to provide
the required solution samples for each reaction period and reaction temperature employed in
the study.
In each leaching experiment, the required amount of the used leachant was first added to the
conical flask, which was then placed in the shaking temperature-bath set at the desired
temperature, and was allowed to thermally equilibrate. The required weighed amount of CKD
was then gradually added to the leachant while maintaining continuous shaking of the solution
in the flask. The reaction was timed from the instant all the required amount of CKD has been
added. At the end of the specified reaction period for each sample, the slurry was filtered
through a .22-m paper filter, and washed five times with 20 ml portions of distilled water.
The resulting filtrate of each sample was then analyzed for leached potassium ions using Solar
M5 Atomic Absorption Spectrophotometer (Thermo Elements, 2001).
3. RESULTS AND DISCUSSION
The percent of potassium released from CKD with leaching time in water/CKD leachates at
different leaching temperatures is shown in Figs. (1-4). The leaching behavior of this element
shows that a large percentage of the total mass of potassium present in CKD is released upon
addition of water. Beyond the first 15 minutes of leaching, there is generally no change in
potassium concentration with reaction time. This observation indicates that this element is
originally present in CKD as readily soluble salts.
These data also show the release of K in Leachate solution increases with an increase in the
liquid/solid weight ratio, and with leaching temperature. However, beyond the initial
15 minutes of leaching, the increase in K released in these water/CKD leachates diminishes
beyond a 5:1 liquid/solid weight ratio and 70 oC. This is further illustrated in Fig. 9, which
shows the percent of K released versus leaching temperature at different liquid/solid weight
ratios after the initial 15 minutes of leaching.
Although these figures show high percentage of K released for most runs, the percentage
released for the 3:1 runs is significantly lower at all leaching temperatures employed in this
study. This indicates the existence of a solubility control on K concentration at very low
water/CKD ratios.
The results for the leaching experiments of CKD with 20 wt.% calcium nitrate solutions are
shown in similar plots in Figs. (5-8). These data show similar characteristic leaching behavior
of potassium with a calcium nitrate solution to that with water. However, for any given
reaction time, the percent of K released using calcium nitrate solution is always larger than
that obtained with water at similar leaching conditions of liquid/solid weight ratio, and
leaching temperature. On the other hand, the reaction time needed to reach equilibrium
concentration of K in leachates of all experiments carried with calcium nitrate solution is
approximately 45 minutes, compared to the 15 minutes observed in leaching experiments
employing water.
Top

Vol. 2.

478

M. A. Daous

Similar solubility control on potassium concentration is observed in calcium nitrate solution

leaching-experiments run at 3:1 liquid/solid weight ratio as that observed in leaching
experiments employing water.
These data further show that, beyond the initial 45 minutes of leaching, no further increase in
K released in the calcium nitrate solution/CKD leachates is observed beyond that obtained
with experiments carried at 5:1 liquid/solid weight ratio, and 70 oC. This is further illustrated
in Fig. 10, which shows a plot of percent of K released versus leaching temperature at
different liquid/solid weight ratios.
The increase in potassium release from CKD when using calcium nitrate solution over that
when using water as leachants could be plausibly attributed to the higher solubilizing effect of
nitrate ion, in one hand, and to equilibrium considerations, on the other, resulting from the
possible reaction between potassium sulfate present in CKD and calcium nitrate solution to
give potassium nitrate and the much less soluble calcium sulfate salt. The formation of such
less soluble calcium salt is expected to shift the equilibrium in the direction to increase the
solubility of potassium sulfate in CKD.
4. CONCLUSIONS
Samples of cement kiln dust were leached with water, and 20 wt.% calcium nitrate solution at
three different liquid/solid weight ratios, and four different temperatures for varying leaching
periods as a first step towards characterizing `the leaching behavior of potassium salts from
alkali rich CKD with these two leachants. The following conclusions were made:

CKD leach tests conducted with water, and calcium nitrate solution at different
liquid/solid ratios and different temperatures provide evidence that there is no
solubility control for the concentration of potassium in solutions of leaching
experiments carried out at 5:1 or higher liquid/solid ratios and 70 oC or higher. Under
these leaching conditions, potassium equilibrium concentration in solution is reached
at in approximately 15 minutes in leaching with water, and 45 minutes in leaching
with calcium nitrate solution.

Utilizing the leaching conditions cited above, it is possible to extract at least 78.7 % of
the potassium present in CKD with water, and 96.1 % of this element with 20 wt.%
calcium nitrate solution.

It is technically feasible to extract potassium from waste CKD by leaching with water
or calcium nitrate solutions, however, further studies are needed to investigate the
compositions of the potassium salts present in these leachates, and the crystallization
of these salts, and their mineral compositions to produce potash or nitrogen potash
fertilizers.

Top

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

479

REFERENCES
1. Baker, D.E., Welch, S., Stout, W., and Doty, W., 1975, Kiln Dust from Cement Factories for
Vacuum Filter Processing of Sewage Sludge, Compost Science, Emmaus, Pennsylvania,
U.S.A., 16 (4), PP. 28-3.
2. Bhatty, J.I., 1995, Alternative Uses of Cement Kiln Dust, RP 327, Portland Cement
Association, Skokie, Illinois, U.S.A.
3. Bhatty, M.S.Y., 1983, Use of Kiln Dust in Blended Cements, PCA R&D Serial No. 1717,
Portland Cement Association, Skokie, Illinois, U.S.A.
4. Bhatty, M.S.Y., 1984, Use of Cement Kiln Dust in Blended Cements, World Cement
Technology, London, U.K., 15 (4), PP. 126-134.
5. Burnham, J.C., 1987, Progress Report on the Pasteurization of Municipal Sludge Utilizing
Cement Kiln Dust, A national Conference on Alkaline Treatment and Utilization of
Municipal Wastewater Sludge, The National Kiln Management Association, Toledo, Ohio,
U.S.A., Section II, pp. 2-12.
6. Daous, M.A., 2000, Utilization of Cement Kiln Dust & Fly Ash in Cement Blends in Saudi
Arabia, Proceedings, The 16th International Conference on Solid Waste Technology and
Management, Philadelphia, Pa, U.S.A.
7. Daugherty, E.D. and Funnell, J.E., 1983, The incorporation of Low Levels of By-Products in
Portland Cement and the Effects of Cement Quality, Cement, Concrete, and Aggregates,
American Society for Testing and Materials, Philadelphia, Pennsylvania, U.S.A., 5 (1),
pp. 14-20.
8. Dean, T.C., 1972, Recovering Alkali Metal Salts From Cement Kiln Gases by the Steps of
Condensing Leaching and Crystallizing, U.S. Patent # 3,647,395.
9. Duchesne, J., and Reardon, E.J., 1999, Lime Treatment of Fly Ash: Characterization of
Leachate Composition and Solid/Water Reactions, Waste Management, 19, pp. 221-231.
10. Einhaus, R.L., Honarkhah, I., and Erickson, P., 1991, Fate of Polychlorinated Biphenyls
(PCBs) in Soil Following Stabilization with Quicklime, Technology Applications, Inc.
Cincinnati, Ohio, EPA/600/2-91052.
11. Fraiman, L.S., Ioganson, A.K., and Solovushkov, N.E., 1991, Study of Properties of Bypass
Dust, Tsement, Leningrad, U.S.S.R., 9(10), pp. 27-29.
12. Klemm, W.A., 1993, Cement Kiln Dust: A Look at its Uses and Characteristics, Proceedings,
The 29th International Cement Seminar, Rock Product, San Francisco, California, U.S.A.
13. Klemm, W.A., 1980, Kiln Dust Utilization, Martin Marietta Laboratories Report MML
TR 80-12, Baltimore, Maryland, U.S.A.
14. McCants, H.C., 1952, Method for Conversion of Portland Cement Dust to an Adsorbent
Material and Adsorbent Material Made Thereby, U.S. Patent # 2,6054,236.
15. McCord, A.T., 1977, Process for Reclaiming Cement Kiln Dust and Recovering Chemical
Values There From, U.S. Patent # 4,031,184.
16. Miller, C. T., Bensch, D.G., and Colony, D.C., 1980, Use of Cement Kiln Dust and Fly Ash
in Pozzolanic Concrete Base Courses, Emulsion Mix Design, Stabilization, and Compaction,
Transportation Research Record 754, TRB, Washington D.C., U.S.A., PP. 36-41.

Top

Vol. 2.

480

M. A. Daous

17. Nicholson, J.P., 1978, Stabilized Mixture, U.S. Patent # 4,101,332, Reissue # 30,943, 1982.
18. Nicholson, J.P., and Burnham, J.C., 1988, Treating Wastewater Sludge with Alkaline
Materials for Decontamination and Conversion into Fertilizer, European Patent Application #
EP Z83153.
19. Preston, M.L., 1993, Use of Cement Kiln Dust as an Agricultural Lime and Fertilizer,
Emerging Technologies Symposium on Cement and Concrete in the in the Global
Environment, SP114T, Portland Cement Association, Skokie, Illinois, U.S.A.
20. Risser, J.A., Doty, W.T., and Baker, D.E., 1981, Cement Kiln Dust Useful as Lime-Potash
Fertilizer, Science in Agriculture, The Pennsylvania State University, College of Agriculture,
Agriculture Experiment Station, University Park, Pennsylvania, U.S.A., 28 (2), pp.6-7.
21. Sayah, A.I., 1993, Stabilization of Highly Expansive Clay Using Cement Kiln Dust, M.Sc.
Thesis, University of Oklahoma, Norman, Oklahoma, U.S.A.
22. Singleton, F.H., and Bruce, J.W., 1975, Treatment of Cement Kiln Flue Dust, U.S. Patent #
3,925,534.
23. Thermo Elements, 2001, Absorption Spectrometry, Methods Manual, Co., U.K.
24. Wills, Jr., M.H., 1983, Concrete Masonry Units Incorporating Cement Kiln Dust, U.S. Patent
No. 4,407,677.
25. Wilson, R.D., and Anable, W.E., 1986, Removal of Alkalis from Portland Cement Kiln Dust,
RI # 9032, Bureau of Mines, U.S. Department of Interior, Washington D.C.
26. Wolf, J.R., 1992, CKD Impurity Removal Investigation, Emerging Technologies for Kiln
Dust Management, SP113, Portland Cement Association, Skokie, Illinois, U.S.A.

Table 1. Analysis of Portland cement and CKD, weight percent

Constituent

Top

Portland
Cement

CKD

CaO

63.77

42.02

SiO2

20.36

14.42

Al2O3

5.33

4.14

Fe2O3

3.97

2.61

MgO

2.10

1.55

K2O

0.14

7.2

Na2O

0.48

2.60

SO3

2.72

4.21

-1

Cl

0.03

4.12

L.O.I

--

12.00

Insoluble
Residues

--

5.13

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

100
90

% of K
released

80
70
60
3:1

5:1

10:1

20
30
40
leaching time, minutes

50

60

50
0

10

70

Fig.(1) Percent of K released in the water/CKD leachates at

30 C and different liquid:solid weight ratios.

100
90
80
% of K
released 70
60

3:1

5:1

10:1

50
0

10

20
30
40
leaching time, minutes

50

60

Fig. (2) Percent of K released in the water/CKD leachates at 50 C

and different liquid:solid weight ratios.

Top

70

481

Vol. 2.

482

M. A. Daous

100

90

% of K
released

80

70

60

3:1

5:1

10:1

40

50

60

50
0

10

20

30

70

Leaching time, minutes

Fig.(3) Percent of K released in the water/CKD leachates at 70 C
and different liquid:solid weight ratios.

100
90
80
% of K
released
70
60

3:1

5:1

10:1

50

60

50
0

10

20
30
40
Leaching time, minutes

70

Fig.(4) Percent of K released in the water/CKD leachates at 80 C

and different liquid:solid weight ratios.

Top

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

100
90
80
% of K
released
70
60

3:1

5:1

10:1

50
0

10

20
30
40
Leaching time, minutes

50

60

70

Fig.(5): Percentage of K released in calcium nitrate solution/CKD

leachates at 30 C and different liquid:solid weight ratios.

100
90

% of K
released

80
70
60

3:1

5:1

10:1

50
0

10

20

30

40

50

60

70

Leaching Time, minutes

Fig.(6): Percentage of K released in calcium nitrate solution/CKD
leachates at 50 C and different liquid:solid weight ratios.

Top

483

Vol. 2.

484

M. A. Daous

100

90
% of K
released
80

70
3:1

5:1

10:1

60
0

10

20

30

40

50

60

70

leaching time, minutes

Fig.(7): Percentage of K released in calcium nitrate solution/CKD
leachates at 70 C and different liquid : solid weight ratios.

100
90
80
% of
Kreleased

70
60

3:1

5:1

10:1

50
0

10

20
30
40
leaching time, minutes

50

60

Fig.(8): Percentage of K released in calcium nitrate solution/CKD

leachates at 80 C and different liquid : solid weight ratios.

Top

70

Recovery of Potassium Salts from Kiln Dust of Local Cement Plants

Vol. 2.

100
90
80
% of K
released

70
60
3:1

5:1

10:1

70

80

50
20

30

40
50
60
Leaching temperature, C

90

Fig (9):Percent of Potassium released in the water/CKD leachates

after the initial 15 minutes of leaching at different temperatures and
liquid : solid weight ratios.

100

95
% of K
90
released
85
3:1

5:1

10:1

60

70

80

80
20

30

40

50

Leaching Temperature, C
Fig(10):Percentage of Potassium released in the Calcium Nitrate
solution/CKD leachates after the initial 45 minutes of leaching at
different temperatures and liquid : solid weight ratios.

Top

90

485