ISIJ International, Vol. 50 (2010), No. 5, pp.

695–701

Moisture Capacity: Definition, Measurement, and Application in

Determining the Optimal Water Content in Granulating

Xuewei LV,1) Chenguang BAI,2) Guibao QIU,2) Shengfu ZHANG2) and Meilong HU2)

1) Ph.D Student, School of Materials Science and Engineering, Chongqing University, Chongqing 400044 China.
E-mail: [email protected] 2) School of Materials Science and Engineering, Chongqing University, Chongqing 400044
China. E-mail: [email protected]
(Received on December 21, 2009; accepted on March 15, 2010 )

The optimal granulation moisture, which means the least water content added into the mixture for obtain-
ing suitable size distribution of the granules, is of vital importance for sintering. The moisture capacity,
which means the maximum water content held in the iron ores of unit mass, is suggested, and defined. An
apparatus for the moisture capacity measurment was developed. Five samples were selected for the mea-
surments. In order to study the relationship between the moisture capacity and some characters of iron
ores, the surface area and pore volume was measured by the method of liquid nitrogen absorption. It was
found that the moisture capacity increaseed with increasing the external surface area of the iron ore, and
decreased with increasing the pore volume in the iron ores. The iron ores were granluted with various
water content level in the labtotary rotaing drum, and then loaded in the sintering pot for measuring the per-
meability of the burden. The size distribution of the granules were also measured for asisting confirming the
optimal water content in granlation. The application of moisture capacity indicated that the iron ore of high
moisture capacity needed more water added in the granulation process to get high permeability.
KEY WORDS: granulation; moisture capacity; iron ore; surface area; permeability.

along the liquid surface and the negative capillary pressure

1. Introduction
existing in the liquid bridge. Generally speaking, the
The sintering process consist, in essence, of mixing strength of the granules increases with increasing the water
moisture, iron ore fines and other fines materials with solid content during granulation. However, more water means
fuel, normally coke breeze, and loading the mix on to a per- more energy consumption. The calculation for this balance
meable grate. The upper surface is raised to a high tempera- relation is shown in Fig. 3, in which the strength of granule
ture by oil or gas burners and air is drawn downwards
through the grate. After a short ignition period, heating of
the bed top is discontinued and a narrow combustion zone
move downwards through the bed, each layer in turn being
heated to about 1 300°C. In the combustion zone, bonding
takes place between the grains and strong agglomerate is
formed. Customarily, the first stage of the whole procedure,
namely, spraying water on to iron ore fines and the subse- Fig. 1. Three-phase air–liquid–solid and two phase liquid–solid
quent tumbling, is called granulating. regimes of a particulate ensemble.
In the course of granulation, the agglomerates undergo a
continuous gradual compaction, at least in the initial period
of growth. As a result, the void spaces become increasingly
filled with liquid. Newitt and Jones1) and Rumpf2) have for-
mulated the wet pellet strength in term of three-phase
air–liquid–solid and two-phase liquid–solid regimes of the
particulate ensemble, which is shown as Fig. 1. As for the
granulation of iron ore sintering, the regime of pendular is
the suitable statues for water content.
The iron ore and other particles held together by a liquid
bridge. As shown in Fig. 2, two forces contribute to the ten-
sile strength of the bond in an additive fashion; the pull due
to surface tension at solid–liquid–gas contract line directed Fig. 2. Pendular bonded between two spheres.

695 © 2010 ISIJ

 ISIJ International, Vol. 50 (2010), No. 5

Fig. 4. Schematic of instrument for measurement of moisture ca-

pacity.

the porosity of the particles has a great effect on the mois-

ture capacity.

Fig. 3. Relationship between the water content and granule 2.2. Equipment
strength or energy consumption.
The measurement of moisture capacity of the iron ore
fines were carried out in the self-made instrument which is
is calculated based on the model3) shown in Fig. 2 and only schematically shown in Fig. 4. The iron ore fines are loaded
considering the energy consumption for water evaporation in the sample tube, which is with a diameter of 56 mm and
by heating from the room temperature. The optimal water a length of 300 mm, and hanged to an electric balance.
content is of vital importance for the sintering plant saving There is a sieve covered at the end of the tube. A piece of
energy consumption and remaining the required quality of filter paper, which is used to avoid the leaking of the very
iron ore sinter simultaneously. Many studies4–9) were car- small iron ore fine into the U-shape container through the
ried out on the optimal water content in the granulating, sieve, is put at the bottom of the sample tube and immerged
most of them focused on the granulating experiments, and into the water for a moment before the measurement. The
some studied the contact behavior between the water and sample tube is taken out for loading the samples after the
ores.6–8,10–12) Recently, Litster10–12) studied a lot on the inter- filer paper is saturated with water, and then hanged to the
action mechanism between the water and iron particle and electric balance. The water in the U-shape container is ab-
the growth kinetic of the granules. However, little study in- sorbed into the sample, resulting in an increase of reading
troduced a simple and useful method to measure or predict on the electric balance.
the optimal moisture content during the granulating. In
present study, a moisture capacity of iron ore fines was de- 2.3. Procedure
fined and measured with a novel and simple method, the re- The iron ore fines need to dry at 120°C for 4 h in a dry-
lationship between the moisture capacity with some other ing chamber to remove the original water before measure-
characters of iron ore fines were discussed. ments. The size distribution of the iron ore fines dried are
measured through a series of screens which are classified as
0.2 mm, 0.7 mm, 1 mm, 3 mm, 5 mm, 8 mm, and 10 mm.
2. Method for Moisture Capacity
The iron ore fines of 200 g are naturally loaded in the sam-
2.1. Definition of Moisture Capacity ple tube. The pressing is not allowed during the loading.
Moisture capacity, in present study, is defined as the The sample tube needs to be immerged into the water for
maximum water content which can be hold among the iron absorbing. Two forces affect the accuracy of the measure-
ore particles of unit mass.13) The equation used to calculate ment. One force is the buoyancy, which is expressed by:
the moisture capacity can be expressed as:
fr · p · g · D2 · h/4 ............................(2)
mc(Mw/Mi)100%..........................(1)
where h is the immerging depth, D is the diameter of the
where mc, Mw, and Mi, individually, refer to the moisture tube, and r is the density of the water.
capacity, mass of water absorbed by the bulk, and the mass The other is the interfacial tension between the water and
of iron ores. The water which touches the particles of iron tube, which are shown in Fig. 5, expressed by:
ore can be absorbed into the bulk of iron ore and moves
Fp · D · cos q · s ............................(3)
among the particles under the capillary pressure. Therefore,
the bulk density or the packing conditions will influence the where s is the surface tension of the water, and q is the
total water content. The water hold among the particles can contact angle between the water and the tube.
also be absorbed into the pores of the particles. Therefore, The immerged depth of the tube can be calculated by Eq.

© 2010 ISIJ 696

 ISIJ International, Vol. 50 (2010), No. 5

(4) when the two force balance.

3.2. Granulation and Permeation Test
4 ⋅ cos θ ⋅ σ 4 ⋅ cos 20° ⋅ 72.75 The granulation experiments of iron ore will be carried
h   0.49 mm ....(4)
ρ⋅g⋅D 1000 ⋅ 9.8 ⋅ 0.056 out in the laboratory rotaing mixer at a defined rotating rate
for 3 min with various water content added. The rotaing
In present study, h was calculated as 0.49 mm. Put down the mixer is with a diameter of 600 mm and a length of
tube 0.49 mm under the water, making the balance between 1 200 mm. The particle distribution of the granules was
the buoyancy of sample tube and the interfacial tension. measured using a series of sieves, including 0.2, 0.7, 1, 3, 5,
Therefore, the reading is just the mass of water absorbed 8, 10 mm. The granules were loaded into the laboratory sin-
into the sample. tering pot, which is with a diamter of 200 mm and a height
of 1 000 mm. The negative pressure of the burden in the
sintering pot under a defined gas flow rate was measured to
3. Experimental
represent the permeability of the burden.
3.1. Samples The moisture capacity was defined for the maximum of
Five kinds of iron ores will be used in this study. The the water content hold in the iron ore particles. However,
chemical composition and the size distribution of these five the flux, fuel, and the sinter return are all used in the granu-
ores are listed in Tables 1 and 2 individually. In the indus- lation. In order to give a reasonable explanation of the gran-
trial practice, several iron ores are used together for the ulation in practice, the definition of moisture capacity is ex-
metallurgical and economic reasons. In present study, the panded to the fuel and coal. As for the flux, the moisture
fives mixtures, which are made of these five iron ores, re- capacity of the limestone, dolomite can be measured easily.
turn fines, coke, and flux with various ratios, were used to However, the moisture capacity of lime is very difficult to
find the relationship between moisture capacity of the mix- get an accurate value. The chemical reaction between the
ture and the optimal water content. The mass fractions of water and lime happens during the measurements, which
the mixture studied are listed in Table 3. is exothermic, resulting in the evaporation of the water.
Therefore the water will be consumed in reaction, evapora-
tion, and physical absorption, great error creating during
the measurements. In order to diminish the effect of lime,
the water will be first added into the lime for slaking. The
experience indicates that the suitable water added is 60% of
lime in mass. Several water content levels were selected for
granulation and the negative pressure of the granules in the
sintering pot were measured to compare the permeability.
The water content with the lowest negative pressure of the
burden is the optimal water content for granulation.

3.3. BET Porosometry

In order to study the dependency of the moisture capac-
Fig. 5. Schematic diagram of interfacial tension. ity on the surface area, particle size, and the pore volume of
Table 1. Composition of the five iron ores selected (mass%).

Table 2. Size distribution of iron ores (mass%).

Table 3. Mass fraction of raw materials in the five mixtures (mass%).

697 © 2010 ISIJ

 ISIJ International, Vol. 50 (2010), No. 5

the iron ore were also measured. The particle features like indicated that ore I, IV, and V have great pore volume, and
surface area were measured by ASAP2020 produced by ore II and III have small pore volume. Table 5 shows that
Micromeritics®. ASAP 2020 is the equipment designed by ore I should have the greatest pore volume, but the distribu-
Micromeritics, which can obtain the high-quality data of tion curve of ore I in Fig. 8 is always under the curves of
surface area and porosity of the materials. The principle of ore IV and V. It seems wrong, but actually right. The reason
the ASAP 2020 is the adsorption of nitrogen at low temper- is that the upper width of ore I is much bigger by 16 nm
ature, the mass of nitrogen absorbed can be measured, thus, than ore V, which is not very obvious in Fig. 7 with a loga-
the surface area can be calculated with this data based on rithmic X-axes.
the theoretical absorption model. The pore size distribution The dependency of the moisture capacity on the external
can be calculated based on the theory of capillary conden- surface area and the pore volume was studied by using the
sation, which means the “process by which multilayer ad- measurements. The correlation got was as follows:
sorption from the vapor into a porous medium proceeds to
Mc13.70.374SE2 896Vpore ..............(5)
the point at which pore spaces become filled with con-
densed liquid from the vapor”.14) All the iron ores were
measured using the nitrogen at the temperature of 77 K.

4. Results and Discussions

The moisture capacities and the saturation time (Ts) of
the five iron ores, fuel, and the flux are listed in Table 4. It
shows that coal has a greatest moisture capacity. Return
sinter has a least value for its less surface area and relative
big particle. For the ores, ore V has a greater moisture ca-
pacity, and ore I has a less moisture capacity.
The surface area and the pore volume (Vpore) of the iron
ores are listed in Table 5. The Isotherm linear plots of five
ores are given in Fig. 6. All the plots are similar, belong
pattern III based on the theory of adsorption.15) The materi-
als of pattern III can absorb more nitrogen at high pressure.
Multiple layers of adsorption happens to this materials,
therefore BET model in Table 5 should be more accurate
than Langmuir model. Ores II and III have small surface
area, from 2 to 5 m2/g, ore I, IV and V have large surface
area, from 9 to 30 m2/g. The microspore area and external
area (SE) are also shown in Table 5. The microspore area
only refers to the pore inside the iron ore particles, and the
Fig. 6. Isotherm linear plot of ores obtained by ASAP 2020.
external area refers to the outside of the particles. It shows
that ore I has a greatest microspore surface area, about
5.35 m2/g, and ore V is the second ore with greater mi-
crospore surface area. Ore II has little microspore surface
area, only 0.07 m2/g. Ore III and IV have very similar value
of microspore surface area, about 1.19 and 1.81 m2/g. For
the external surface area, the order of the ore from great to
little is ore V, I, IV, III and II. Figure 7 shows the pore vol-
ume distribution with the pore width for the five ores. It is

Table 4. Moisture capacity and saturation time of iron ores.

Fig. 7. Pore volume verse pore width obtained by ASAP 2020.

Table 5. Surface area and pore feature of the iron ore.

© 2010 ISIJ 698

 ISIJ International, Vol. 50 (2010), No. 5

Fig. 8. Measurements verse calculation for moisture capacity. Fig. 10. Negative pressure of burden as a function of water con-
tent.
Table 6. Moisture capacity and saturation time for the mix-
ture. The Relationship between the negative pressure of the
burden and the water content added for the five iron ores
are plotted in Fig. 10. From Fig. 10, it is indicated that the
permeability of the burden in the sintering pot increases
with increasing the water content added during the granula-
tion first, a maximum permeability, meaning the minimum
negative pressure during the suction measurement, will
reach at suitable water content, then the permeability de-
creases with increasing the water content. The relationship
between the permeability of the iron ore burden in the sin-
tering pot and the water content added in the mixture,
which is similar to a parabola curve, got in present study
agrees with the previous reports.16,17) The optimal water
content was decided according to these measurements, and
the results are listed in Table 7. The rules for this decision
are described as follows: the water content with the mini-
mum negative pressure from the fitting curves is the opti-
mal value if the square of correlation coefficient between
Fig. 9. Water content absorbed as a function of time.
the measurements and the fitting curve is above 0.8, or else
it is below 0.8, the water content with the minimum nega-
where SE and Vpore are the external surface area and pore tive pressure in the measurements is selected to instead the
volume. The values calculated and the measurements for optimal value.
the five samples are plotted in Fig. 8, from which it is indi- The particle distribution of the granules with various
cated the calculation agree well with the measurements. water contents are plotted in Fig. 11. The mass fractions of
Equation (5) indicate that the moisture capacity increase the particles with defined size were represented with his-
with increasing the external surface area of the iron ore, tograms, and the negative pressure of the burden in the sin-
and decreases with increasing the pore volume in the iron tering pot as a function of water content added were also
ores. Generally speaking, the iron ore of small size has the plotted together for easily comparing the suitable size dis-
great surface area. Equation (5) also means the iron ore par- tributions. Figure 11 indicate that the iron ore granule
ticle of small size has a big moisture capacity. This conclu- grows up with increasing the water content added into the
sion agree well with the previous study.13) mixtures. The mass fraction of the particles with small size
The moisture capacity of the five mixtures of iron ores, decreases with increasing the water content, and that of par-
fuels, return fines, and the flux are listed in Table 6. The ki- ticles with big size increase with increasing the water con-
netic of water absorbed into the mixtures are plotted in Fig. tent. Summarizing the five figures, it was indicated that the
9. Mixture E has high moisture capacity and mixture C has mass fraction of the particle with 0.2, 0.2–0.7, 0.7–1 mm
low moisture capacity. The moisture capacities of mixture decreases with increasing the water content added into the
A and B are very close to each other, and the water ab- mixtures, while that of particles with 1–3, 3–5, 5–8, 8–10
sorbed curves are also very similar. The time for water sat- mm increases first and then decreases with the increase of
uration in these five mixtures are different. The saturation the water content, and only the particle with 10 mm in-
time for mixture E is the shortest, only 30 s, while that for creases always with increasing the water content. The opti-
mixtures A and B are longer than mixture E, about 50–90 s; mal size distribution of the particles can be summarized as:
and that for mixtures C and D are longest, about more than about 50% particles with 10 mm, about 30% particles
200 s. It means that mixtures A, B, and E have enough time with 5–8 mm, about 10% particles with 8–10 mm, and 10%
to absorb the water for saturation, while mixture C and D other particles.
can be very close to saturation. According to the measurements of moisture capacity and

699 © 2010 ISIJ

 ISIJ International, Vol. 50 (2010), No. 5

Table 7. Optimal water content for the five iron ores.

Fig. 11. Particle distribution of the granules verse water content for iron ores A–E.

were measured. The moisture capacity of the iron ore was

also applied on optimizing the granulation by finding the
relationship between the moisture capacity of the mixture
and the optimal water content. It was concluded as follows:
(1) The dependency of the moisture capacity on the
external surface area and the pore volume was was ex-
pressed as Mc13.70.374SE2 896Vpore. The corre-
lation agree swell with the measurements and indicate that
the moisture capacity increase with increasing the external
Fig. 12. Relationship between moisture capacity and best water
surface area of the iron ore, and decreases with increasing
content. the pore volume in the iron ores.
(2) The moisture capacity and the optimal water con-
the optimal water contents, the relationship between these tent have high positive correlation. The iron ore which has
two parameters is shown in Fig. 12, from which it can be high moisture capacity need more water added in the gran-
concluded that the mixture with higher moisture capacity ulation process in order to get high permeability.
need more water for getting a best permeability. The mois- (3) The optimal size distribution of the granules after
ture capacity and the best water content have high positive granulation is summarized as: about 50% particles with
correlation. The equation got in present study is: 10 mm, about 30% particles with 5–8 mm, about 10%
particles with 8–10 mm, and 10% other particles.
y6.940.12Mc ...........................(6)
(4) The correlation got by the measurements: y6.92
where y is the optimal water content. The correlation coeffi- 0.12Mc can be used to predict the optimal water content
cient of this equation is 0.907. Using this equation, the opti- in the granulation.
mal water content added into the mixture can be predicted.
Acknowledgments
The authors are especially grateful to Chongqing Univer-
5. Conclusions
sity Graduates’ Science and Innovation Fund (Grant No.
In present study, moisture capacity of the iron ore parti- 200911B1A0120328) and Key Project of Chinese National
cles was suggested, defined, and measured in a self-de- Programs for Fundamental Research and Development
signed apparatus. The surface area and the pore features (Grant No. 2007CB613503).

© 2010 ISIJ 700

 ISIJ International, Vol. 50 (2010), No. 5

REFERENCES 10) J. D. Litster and A. G. Waters: Powder Technol., 55 (1988), 141.

11) J. Litster, B. Ennis and L. Lian: The Science and Engineering of
1) D. M. Newitt and J. M. Conway-Jones: Trans. Inst. Chem. Eng., 36 Granulation Processes 15th ed., Particle Technology Series, ed. by B.
(1958), 422. Scarlett, Kluwer Academic Publishers, Dordrecht, (2004), 247.
2) H. Rumpf: Agglomeration, Interscience, Pennsylvania, (1962), 379. 12) D. H. Lister and F. P. Glasser: Trans. Br. Ceram. Soc., (1967), 293.
3) P. C. Kapur: Adv. Chem. Eng., 10 (1978). 13) X. Lv, C. Bai, C. Zhou and H. Xie: Proc. of the 5th Int. Cong. on the
4) W. Pietsch and H. Rumpf: Ingenieur Technik, 39 (1967), No. 15, Science and Technology of Ironmaking, Shanghai, (2009), 235.
885. 14) L. L. Schramm: The Language of Colloid and Interface Science, ed.
5) H. Schubert: Powder Technol., 11 (1975), No. 2, 121. by ACS Professional Reference Book, ACS, Washington, DC.,
6) S. M. Iveson, S. Holt and S. Biggs: Colloids Surf. A: Physicochemi- (1993), 129.
cal and Engineering Aspects, 166 (2000), 203. 15) S. Kondo, T. Ishikawa and I. Abe: Adsorption Science (In Japanese),
7) S. M. Iveson, S. Holt and S. Biggs: Int. J. Miner. Process., 74 (2004), Maruzen, Tokyo, (1991), 68.
281. 16) J. Khosa and J. Manuel: ISIJ Int., 47 (2007), No. 7, 965.
8) S. M. Iveson, K. F. Rutherford and S. Biggs: Trans. Inst. Min. Met- 17) M. Yoshinaga, S. Sato and T. Kawaguchi: Australia Japan Extractive
all. C, 110 (2001), C113. Metallurgy Symp., Sydney, (1980), 145.
9) P. C. Kapur: Adv. Chem. Eng., 10 (1978), 55.

701 © 2010 ISIJ