The effect of thermal drying on the mechanical strength of South African coal

Abstract

The dewatering of coal, and particularly fine coal continues to challenge coal prepatration engineers to find a cost-effective solution. With known world oil reserves being depleted daily, it is envisaged that the future price of coal may justify the use of thermal drying to achieve lower coal product moisture levels. The effect of exposure of the coal to elevated toemperatures on the mechanical strength of the coal was investigated. It was found that temperature does not play a major role in determining the volume breakage of a particle and that other variables such as orientation during impact has a much greater influence. A double breakage mechanism was reported during the grindability tests. Surface, as well as volume breakage, occurs for the first 4 minutes while only surface breakage takes place thereafter. Due to this double breakage action, it was found that exposure to temperature does play a role in the amount of breakage and breakage rates during grindability tests. It was also concluded that a particle will break to an optimum size due to impact, after which only its surface will grind away as it is subjected to breakage forces. © The Southern African Institute of Mining and Metallurgy, 2008.

Full text

T
r
a
n
s
a
c
t
i
o
n
P
a
p
e
r
Introduction
Reducing the total moisture content of a final
coal product remains one of the more
frustrating areas in coal processing. An
excessive amount of moisture leads to
financial penalties1and can also cause
handling difficulties. Conventional mechanical
methods used to dewater coal do not seem to
be able to do a complete job, while on the other
hand, the value of coal does not justify the use
of thermal drying methods. The presence of
water in product coal continues to be an
operational and economic problem, and while
depleting oil reserves will eventually increase
the value of coal, it is envisaged that thermal
drying of coal will start to play a greater role in
the future.
Apart from the removal of moisture from
coal, thermal drying also has an effect on the
coal’s mechanical strength (from here referred
to as strength)2. The strength of coal is an
important property since it has a direct effect
on the particle size of the coal. The particle size
of the final coal product can then in turn affect
the productivity and efficiency of downstream
processes such as gasification. Studies have
shown that the possibility of fragmentation of
coal after thermal drying would increase with
an increase in the applied temperature during
the drying stage3.
Background
Moisture exists in coal in three different states,
as defined by Rong4. They are surface
moisture, capillary moisture, and chemically
bound moisture. Surface moisture refers to the
inter-particle moisture and can be removed by
mechanical methods such as filtration.
However, studies on coal beneficiation plants
showed that it is not uncommon for fine coal
to have a final moisture percentage between
20 per cent to 30 per cent after filtration. The
capillary moisture, commonly known as intra-
particle moisture, can be removed only by
using thermal methods. Chemically bound
moisture is part of the structure of the ash
fraction of the coal and cannot be removed,
except by pyrolysis. Therefore, to produce a
final coal product containing a single figure
moisture percentage, thermal drying is
inevitable. It is therefore important to
understand what effect thermal drying has on
the strength of a coal product.
In general, size degradation of coarse coal
occurs at conveyor transfer points, in hopper
bins, screening operations, during stockpiling
and reclaiming, and even during more gentle
operations such as conveying. The degree of
size degradation depends on the cumulative
energy imparted to the coal and on the
inherent strength of the particle. Although the
The effect of thermal drying on the
mechanical strength of South African coals
by M. Le Roux*
Synopsis
The dewatering of coal, and particularly fine coal, continues to
challenge coal prepatration engineers to find a cost-effective
solution. With known world oil reserves being depleted daily, it is
envisaged that the future price of coal may justify the use of
thermal drying to achieve lower coal product moisture levels. The
effect of exposure of the coal to elevated toemperatures on the
mechanical strength of the coal was investigated. It was found that
temperature does not play a major role in determining the volume
breakage of a particle and that other variables such as orientation
during impact has a much greater influence.
A double breakage mechanism was reported during the
grindability tests. Surface, as well as volume breakage, occurs for
the first 4 minutes while only surface breakage takes place
thereafter. Due to this double breakage action, it was found that
exposure to temperature does play a role in the amount of breakage
and breakage rates during grindability tests. It was also concluded
that a particle will break to an optimum size due to impact, after
which only its surface will grind away as it is subjected to breakage
forces.
*School of Chemical and Minerals Engineering,
North-West University, South Africa.
© The Southern African Institute of Mining and
Metallurgy, 2008. SA ISSN 0038–223X/3.00 +
0.00. Paper received July 2008; revised paper
received October 2008..
783
The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 108 REFEREED PAPER DECEMBER 2008
▲
193_leRoux:Template Journal 12/11/08 4:30 PM Page 783

The effect of thermal drying on the mechanical strength of South African coals
material strength of coal depends on its rank and
composition5, it is also depends ons its size6. Therefore, it
can be concluded that the energy requirement for size
reduction increases with decreasing particle size.
Two mechanisms of fracture occur when force is applied
to coarse material. They are volume breakage (including
cleavage and shatter) and surface breakage (abrasion)7. The
two mechanisms of fracture are distinguished by the size
distribution of the degraded product. Volume breakage
usually gives a product that has a more progressive size
distribution, while surface breakage will yield some fine
material, and one particle that is closely related to the size of
the original particle.
Experimental design
Two types of tests were used to determine the strength of the
coal. The first is the drop shatter test, based on the ASTM
D44-75 standard, and the second is a normal grindability test.
The drop shatter test is used to determine the relative
size-stability of coal, and is normally an indication of the
ability of coal to withstand breakage during handling and
preparation. The test involves dropping coal fragments from a
fixed height onto a solid surface. A size analysis is then
performed on the product after the drop to determine the
amount of material that has broken to finer size ranges.
Mathematical models of crushing and grinding processes
have been used extensively for circuit design in the minerals
process industry. An important concept of these models is
that size degradation can be considered a first order rate
process7. Therefore, for the drop shatter test:
M1 –M0= –KvM0[1]
where M0is the initial mass of the size fraction, M1is the
mass of the unbroken material after the first drop, and Kvis
the breakage index.
This can be repeated several times on the same sample,
which will yield for Ndrops that:
MN= M0(1 – Kv)N[2]
The numerical value of Kvis determined from the slope of
a straight line plot of ln[M
N
⁄M
0
] versus N. A large Kvvalue
means a greater extent of breakage and a lower resistance to
shatter, which implies a lower strength7.
The grinding test is used to measure the resistance of a
coal particle to abrasion during transport and handling. The
coal is subjected to standard autogenous grinding conditions
and the amount of fines that are generated during fixed
intervals is measured. For this process, the rate of surface
breakage can be given by:
[3]
where Mis the mass of coal remaining in the given size
fraction and Ksis the breakage constant. Integrating this
equation will give:
[4]
The value of Kscan be determined from the slope of a straight
line plot of ln[M
N
⁄M
0
] versus time. As with the drop shatter test,
a large Ksimplies a lower strength coal.
Experimental set-up
Three coal samples were obtained from Witbank Collieries
(Witbank Number 4 seam coalfield in South Africa), New
Vaal Collieries (Free State coalfield in South Africa) and
Middelbult Collieries (Highveld coalfield in South Africa,
adjacent to the Witbank coalfield). Table I shows the
proximate analysis of the coal.
A quantity of sample was thermally dried at 25°C (which
serves as the reference point for no drying), 105°C, 130°C,
160°C and 190°C in a nitrogen rich atmosphere for an hour.
Thereafter it was split into size ranges of –19 mm + 13.2 mm
for the drop shatter test and –13.2 mm + 6.6 mm for the
grindability tests. A 200 g sample from each size range was
used for each test.
Experiments were performed using the two different
procedures for each one of the tests as described above. For
the drop shatter tests, the coal sample was loaded into a
hopper situated 8 metres above a solid steel surface. This
steel surface was enclosed in a 2 metre high drum to prevent
the coal from spilling after impact. The coal was dropped onto
the steel surface and then the amount of +13.2 mm particles
was weighed. The +13.2 mm particles were put through the
test again. This was repeated four times.
For the grindability test, a standard 210 mm internal
diameter laboratory steel mill was used. The mill was loaded
with 200 g of the –13.2 mm + 6.6 mm sample without any
grinding media. The mill was operated at 60 rpm, which is
about 65 per cent of its critical speed. Every two minutes the
mill was stopped and the amount of +6.6 mm coal was
determined. Thereafter the whole sample was placed back in
the mill and allowed to run for another two minutes. This
was repeated 5 times.
The data obtained from both sets of experiments were
reworked to produce the Kvand Ksvalues.
Results and discussion
Drop shatter tests
Drop shatter tests were done on the Witbank, New Vaal and
Middelbult coals as described above, and selected results are
shown in Figures 1–5. Figures 1–3 show the results for the
different coal types. Each one of the graphs shows the mass
fraction of coal remaining in the predetermined size rage after
each drop. The legend of the graphs is an indication of the
different drying temperatures in degrees Celsius.
From Figures 1 –3 it is clear that the data are linear, but
the influence of temperature on the strength of the coal was
minimal. The data points on the graphs lie within acceptable
▲
784 DECEMBER 2008 VOLUME 108 REFEREED PAPER The Journal of The Southern African Institute of Mining and Metallurgy
Table I
Proximate analysis of coal tested
Witbank New Vaal Middelbult
% Moisture (SABS 924) 3.83 6.46 4.00
% Ash (ISO 1171) 11.62 35.79 31.34
% Volatile matter (ISO 562) 27.33 19.18 21.63
% Fixed carbon 57.22 38.56 43.03
193_leRoux:Template Journal 12/11/08 4:30 PM Page 784

error ranges, and there is no clear trend visible that would
lead to a conclusion that the coal is more likely to shatter
after it has been subjected to higher drying temperatures.
This may indicate that insufficient time was given for the
temperature effect to reach the core of the particle. Hence,
there were only minor changes on the surface of the particles,
which led to the constant volume breakage behaviour of the
different particles.
Figure 4 shows the Kvvalues obtained for each coal type
in relation to the drying temperature for the coal. From Figure
4 it can again be concluded that thermal drying has a
minimal influence on the volume breakage of a coal particle.
By following the trend of the graph, it does, however, appear
as if there is a tendency for the Kvvalue to increase slightly
towards the 190°C temperature range. Keeping in mind that a
higher Kvvalue means a weaker coal, it can be stated that the
tendency is for the coal to become only slightly weaker at
elevated temperatures below 200°C.
Taking into account the amount of scatter from the graph
in Figure 4, it is safe to say that the type of coal has a very
limited influence on the volume breakage. This statement can
be quantified by taking the average fraction of coal remaining
in the set size range after impact for every drying
temperature per drop. This can be done for every coal type. A
graph of this average fraction against the number of drops
will give an indication of the influence of the coal type on the
volume breakage strength of the sample.
From Figure 5 it can be seen that the influence of the type
of coal on the breakage is minimal. It is, however,
noteworthy that the Witbank and Middelbult coals, which
have a higher carbon fraction, did break slightly more easily
than the New Vaal coal (a high ash coal). Although not tested
during this work, observations suggested that the orientation
of a coal particle when making contact with the steel surface
does play a significant role.
Grindability results
Selected results from the grindability tests are given in
Figures 6–12. From Figures 6–8 it can be seen that there is a
clear relationship between the applied drying temperature
and the tendency for the particles to grind and chip away on
the surface. More breakage occurs for all three coal types as
the applied drying temperature increases. These results
confirm the explanation given above that the influence of
temperature has taken place only on the surface of the coal
particles, and has not had time to reach the core.
Similarly, the rate of breakage increases as the applied
The effect of thermal drying on the mechanical strength of South African coals
T
r
a
n
s
a
c
t
i
o
n
P
a
p
e
r
785
The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 108 REFEREED PAPER DECEMBER 2008
▲
Figure 3—Drop shatter test results for Middelbult
Figure 2—Drop shatter test results for New Vaal
Figure 1—Drop shatter tests results for Witbank
Figure 5—Average fraction breakage values
Figure 4—K
v
values
193_leRoux:Template Journal 12/11/08 4:30 PM Page 785

The effect of thermal drying on the mechanical strength of South African coals
drying temperature increases. It is, however, interesting to
note that there is a vast difference in the final amount of
breakage and rate of breakage among the different types of
coal.
As described earlier, the breakage rate constant, Ks, is
determined from the slope of the graph of ln[M
N
⁄M
0
] vs. time.
This graph is shown in Figure 9.
The graph in Figure 9 shows that the rate of breakage of
the Middelbult coal is much lower than the rate of breakage
of the New Vaal coal (keep in mind that a lower Ksvalue
means a lower tendency to break). In general, the Free State
coalfields are predominantly dull coal, interlaminated with
sandstone and mudstone, whereas the Witbank coal field has
high coal zones with mudstone and siltstone partings. The
Highveld coalfield on the other hand, consists more of very
thin discontinuous layers8, giving a coal sample that looks
like shale. Therefore, for the New Vaal coal, there is a more
homogeneous distribution of interfaces between the macerals
and the minerals close to the surface than is the case with the
Middelbult coal, meaning more potential weak spots. The
number of weak spots is exaggerated when exposed to the
higher temperature and will give rise to greater breakage
rates.
An interesting trend arose during the determination of
the Ksvalues. The graphs of ln[M
N
⁄M
0
] vs. time did not yield a
linear relationship as was expected from the relevant
literature. An example of the Witbank coals is shown in
Figure 10.
Figure 10 shows a graph where there is a clear discon-
tinuity in each of the tests between 2 and 4 minutes. It yields
a curve with two different slopes, indicating two different
breakage rates. Since the grindability was carried out in a
laboratory-scale mill, the path each particle travels leaves
space for a 200 mm freefall, which will include some degree
of volume breakage, especially on an already weakened
surface. It means that for the first few minutes both surface
and volume breakage may occur. Similar findings were
documented in the paper of Sahoo and Roach7.
To determine the significance of the impact on the surface
breakage of the particle, the slope of the curves for the first 4
minutes and the remaining 6 minutes were drawn separately.
From the slope, the Ksvalues were determined for the
different breakage mechanisms. The results are given in
Figures 11 and 12.
Figure 11 (which is the Ksvalues for the first 4 minutes
of grinding) shows a distinct relationship to the graph in
Figure 9, whereas Figure 12 has yielded a near horizontal
line. These results are very significant. It means that most of
the breakage of the particles takes place in the first 4 minutes
of grinding, whereas very little breakage occurs in the
remaining time. It seems as if there is an optimum size to
▲
786 DECEMBER 2008 VOLUME 108 REFEREED PAPER The Journal of The Southern African Institute of Mining and Metallurgy
Figure 8—Middelbult grindability results
Figure 7—New Vaal grindability results
Figure 6—Witbank grindability results
Figure 10—Example of Witbank reworked grindability data
Figure 9—K
s
values
193_leRoux:Template Journal 12/11/08 4:31 PM Page 786

which a particle will tend to reduce during handling
operations, together with an amount of fines generated from
these breakages. Only when the applied forces to the particles
are increased will they tend to break beyond this optimum
size.
Conclusions
From the r esults above, the following can be concluded:
➤Thermal drying for a time less than 1 hour does not
have a significant influence on the mechanical strength
of a particle greater than 13.2 mm, and the tendency is
for volume breakage to occur. It does, however, show
some influence on the strength of the surface of the
particles, which will cause more breakage to occur if no
drying takes place.
➤The orientation of the particle during impact, rather
than the type of coal, has an influence on the volume
breakage.
➤For the first 2 to 4 minutes of a grindability test, both
surface and volume breakage take place. It is during
this time that most of the breakage occurs. For the
remaining time, only surface grindability takes place.
➤Although not proven, it does seem that coal tends to
break easily to a certain size, whereafter it takes a
definite input of energy to break the coal more finely.
Acknowledgements
The author would like to thank the following people for the
contribution they made in doing experimental work:
➤Mr. C.J. Willemse
➤Mr. R.C. Button
References
1. TAO, D., GROPPO, J.G. and PAREKH, B.K. Enhanced ultrafine coal dewatering
using flocculation filtration processes. Minerals Engineering, vol. 13, no.
2, 2000 . pp. 163–171.
2. ANDERSON, C.M., DEWALL, R.A., MUSICH, M.A. and RICHTER, J.J. Reducing the
moisture in low-rank coals: The stability issues. Conference Article.
ASME: New York. 1994. pp. 375–391.
3. JLANHUI, H., MIGHOU, X., DUNXI, Y. and YUN, Y. Fragmentation by coal
particles by thermal stresses during combustion. Journal of Huazhong
University of Science and Technology. vol. 32, no. 5, 2004. pp. 78–80.
4. RONG, R.X. Advances in coal preparation technology, vol. 2: Literature
review on fine coal and tailings dewatering. JKMRC report on AMIRA
Project P239C, University of Queensland. 1993. p. 30.
5. FALCON, R.M.S. and SNYMAN, C.P. An introduction to coal petrography:
Atlas of petrographic constituents in the bituminous coals of South Africa.
Geological Society of South Africa. Johannesburg. 1986. p. 39.
6. ESTERLÉ, J.S., KOLATSCHEK, Y. and O’BRIEN, G. Relationship between in situ
coal stratigraphy and particle size and composition after breakage in
bituminous coals. International Journal of Coal. vol. 49, nos. 2–3, 2002.
pp. 195–214.
7. SAHOO, R. and ROACH, D. Quantification of the lump coal breakage during
handling operation at the Gladstone port. Chemical Engineering and
Processing, vol. 44, no. 7, 2005. pp. 797–804.
8. JEFFREY, L. Characterisation of the coal fields of South Africa. Colloquium
on sustainability of coal. SAIMM, Sep 2004.
◆
The effect of thermal drying on the mechanical strength of South African coals
T
r
a
n
s
a
c
t
i
o
n
P
a
p
e
r
The Journal of The Southern African Institute of Mining and Metallurgy VOLUME 108 REFEREED PAPER DECEMBER 2008 787
▲
Figure 11—K
s
values for the first 4 minutes
Figure 12—K
s
values for the final 6 minutes
193_leRoux:Template Journal 12/11/08 4:31 PM Page 787