THE CALCINING ZONE OF THE KILN AND OTHER CONSIDERATIONS

By W. Greer, Lone Star Industries

This section, we will cover the effects of the composition of the kiln charge on calcining and its later effect on
burning. We will cover the alkali cycles and some of the problems that can be expected in this area. We will
consider the information that can be obtained by measurement of kiln torque; the crucial subject of reaction
rates; and we will consider the diagnosis of conditions in the calcining zone as well as the effects of fuel.

Let's consider the basic events that occur in the calcining zone from the standpoint of heat requirements. The
calcining of the carbonate-bearing materials brought about by heat energy is called an endothermic chemical
reaction. Heat energy is not recoverable in the calcining process as the heat in the clinker is partially recovered
in the cooler or it is lost to the system.

The theoretical work to be done in the calcining zone is the decomposition of the calcium carbonate in the raw
feed. The products of this reaction are calcium (lime) and carbon dioxide.
The reaction is depicted in Figure 1.

CaCO3 + Heat = CaO + CO2

calcium 1648°F calcium carbon
carbonate (898°C) oxide dioxide
(limestone) (lime)
1.735 kg 1.0 kg + 0.785.kg

By doing the mathematics properly, it can be determined that 1.785 kg of calcium carbonate in the raw
mix-will yield 1.0 kg of lime and 0.785 kg of CO2.

As was stated, heat is required for this reaction to take place.

The temperature required is 1648°F (898°C). This is not a precise temperature and is shown with different
values in the literature. The reaction involves a phase change from a solid to a solid and a gas.

To help understand what happens in calcination, let's consider the evaporation and boiling of water. We all
know that water boils at 212°F at one atmosphere pressure (sea level). There is slight evaporation from the free
surface of the water .at room temperature. The rate of evaporation continues to increase as the temperature of
the water increases, but the evaporation is confined to the surface. Once the temperature of the water reaches
212°F, the vapor pressure of the water exceeds ore atmosphere, and steam bubbles can form within the liquid.
Boiling begins and continues until all the water is gone. During boiling, there is no increase in temperature of
the water.
An increase in the amount of heat supplied to the pan of boiling water will increase the evaporation rate but
not the temperature. There is a total, fixed amount of heat required to vaporize one pound of water. This is
970.2 BTUs at 212°F and one atmosphere of pressure.
We also know that the evaporation rate of water into air is dependent on the amount of water vapor in the air.
When the air is saturated with water (the partial pressure is equal to the vapor pressure), and the temperature is
below 212°F, then evaporation stops. On a humid day, evaporation of water is much slower than on a dry day.

The calcination of limestone is similar and is dependent on temperatures, pressures, time, and composition of
the immediate atmosphere. With this in mind, we can look at a hypothetical piece of limestone that is a perfect,
homogeneous sphere of one in. radius. It is supported by a scale in a laboratory oven, and the means are
available by which temperature can be measured (see Figure 2).

When the surface temperature reaches about 1200°F in an air atmosphere, incipient calcination appears on the
surface. If the temperature of the sample remained at 1200°F, the rate of calcination would be very slow and
would ultimately cease. This situation differs slightly from our attempted water analogy in that we have a solid
whose composition at the surface cannot be changed by molecular movement, and the source of the gas is a
chemical reaction. Nevertheless, if the temperature is increased to 1500°F, the calcination rate would be
measurable but very slow. It would take a very long time to complete the calcination at 1500°F. If we again
permit the temperature of the sample to rise until the center reaches 1648°F it will remain at this temperature
regardless of the ambient temperature until calcination is complete.

At the completion of calcination, the temperature of our hypothetical piece of lime will increase until the
equilibrium temperature of the furnace is reached. If for some reason the atmosphere in the furnace is 100%
carbon dioxide instead of air, incipient calcination would not begin until the surface temperature was 1600°F.
In an atmosphere similar to that of a kiln, 30% carbon dioxide, the evolution of gas would begin in the area of
1500°F.

Consider now the situation during calcining in a rotary kiln. During the period of active calcination, the
atmosphere within the active bed will approach 100% carbon dioxide. The surface of the bed will be exposed
to an atmosphere of about 30% carbon dioxide and higher kiln gas temperatures. As raw mix particles roll
back into the bed mass, they will be cooled.

I'm sure you can imagine from the description that the calcining zone in the rotary kiln is a dynamic area.

Free calcium oxide absorbs carbon dioxide quite readily. It is easy to see that some particles of free calcium
oxide from the surface of the bed can be recarbonated in the cooler, high carbon dioxide atmosphere of the
mass. It is well documented that this recarbonation occurs until the entire mass exceeds 1648°F.

We should again consider our original limestone sphere of one in: radius. In the oven, with a constant
temperature exceeding 1648°F, the surface will be calcined first with the calcination proceeding toward the
center at a constant, uniform rate. When the calcining surface is one half in from the surface or one half of the
way toward the center, it can be calculated that 87.5% of the required work has been done. Since calcining
proceeds at a uniform rate, we also know that when the calcining surface is one half the way to the center then
one, half the time required to complete calcination has passed. Therefore, if it took one hour to reach the half
waypoint and do 87.5% of the work, it will take an additional hour to complete calcination. This process is
illustrated in Figures 2 and 3.

If by increasing the temperature in the furnace, we can double the calcination rate so that the half waypoint is
reached in one half hour then the work-time relationship is shown in Figure 4. In a traditional kiln it is
impossible to control the calcining rate by varying temperatures in the calcining zone without disturbing other
areas of the kiln. Conversely, if other areas are disturbed, such as a sudden increase in heat in the burning
zone, then the calcining rate will change in this case increase with sometimes undesirable consequences.

We have seen that calcination is a time-temperature reaction. It takes a combination of time and temperature to
perform the desired work. One can achieve the percent completion of calcination as was achieved by
increasing the temperature if one simply reduces the diameter of the theoretical particle. In kiln practice this is
done by grinding the raw mix finer. In Figure 4, curve "C", we see the percent calcination of a particle of one
half the volume of the original one in. radius sphere. To achieve percent calcination to the degree achieved by
the increase in temperature (curve B) the sphere must be reduced is diameter to one half in. with the original
calcining rate of one half in. per hour. In our example the important parameter chane is that the time for
completion of the decarbonization reaction has been reduced by one hour.
By reducing the radius of the sphere from one in. to one-half in. eight of the smaller spheres will be needed to
equal the volume of the larger piece. If time is important to our process, which it is in most industrial
applications, then the size reduction may be advantageous. It seems to be common practice to make raw mix as
coarse as possible without real concern about or examination of what follows in the cement making process.
We will look at this question again a bit later on, but it is now clear that there is an absolute maximum particle
size that can be decarbonated in the transit time available in our kiln system given a fixed temperature profile.
If the particle sizes are larger, then temperatures must be increased to adequately decarbonate the raw mix in
the time allotted. As kiln operators you must determine optimum performance.

The size reduction we are discussing here also prompts consideration-of another factor. As the size of the
particles is reduced, the surface area available for heat transfer or chemical reaction is increased. One of the
ways to increase the speed of a chemical reaction is to increase the surface of the reactants. Similarly, the total
amount of heat transferred by conduction is increased as the surface area is increased.
The fineness of a raw mix is in the range of 4000-5000 square centimeters per gram. At the lower fineness
level of 4000 cm2/ gram, the surface area of one gram of cement is that of a square 24.9 in. on a side. At the
higher level of 5000 cm2/gram, the square increases in dimension to 27.8 in. This is a tremendous surface area
available for heat transfer.

To utilize this surface area and to reduce the time required for transfer of a fixed amount of heat, the
suspension preheater kiln system was developed. In this system the raw material enters the rotary kiln at about
1500°F and is about 40% calcined. The gases leaving the kiln are in the range of 2000°F.
The heat transfer, which occurs in the preheater riders, is facilitated not only by the large, exposed surface
area, which we previously mentioned, but also by the temperature differential between the kiln exhaust gases
and the kiln feed. The temperature differential is the driving force in all heat transfer situations. The
suspension-preheater allows a large temperature differential (driving force) to occur in an area of our system
where the, potential for heat transfer (surface area) is very great.

There are a couple of other points that influence heat transfer in a suspension preheater that we should consider
briefly. In heat transfer involving fluids (kiln exhaust gas), a high velocity at the heat transfer surface increases
the rate of heat transfer. Gas velocities are quite high in a suspension preheater which make them high at the
surface of each raw mix particle when compared to the velocities experienced in the kiln. Moreover, in our
earlier theoretical piece of limestone, there was incipient calcination, which took place from the surface of the
sphere in a much lower temperature that that required for complete calcination. By increasing the surface area
of the raw mix, the area available for incipient calcination at lower temperatures has been increased.

Travel time through a preheater is a matter of just a few seconds during which 40% of the calcination is
accomplished. Typically, a 165 foot preheater has a retention time of 25 seconds. The 40% calcination
represents 15.65% of the total time required at a fixed ambient temperature. If we assume that the transit time
in the preheater zone of adequate temperature was four seconds, then the balance of the work would require
21.56 seconds. I trust you all understand that this discussion is not accurate for actual practice since the
temperature of the gas in the preheater is not constant, the raw mix particles are not spheres, etc., but the idea
is important. A relatively large amount of the work of calcination can be accomplished in a very short period
of time by utilizing to best advantage the principles of heat transfer and chemistry.
We also know that in practice the balance of calcination will not take place in the rotary kiln in anything like
21.56 seconds. The time required will be much greater. All the factors working to our benefit in the preheater
are now working against us. The material enters the kiln in the range of 1500°F and forms the bed in the
bottom of the kiln. For any individual particle there is clearly a reduction in temperature differential and heat
transfer driving force. Heat transfer to the bed is reduced by-surface area considerations alone. The gas
velocity around interior particles is essentially zero. The atmosphere in the bed is close to 100% carbon
dioxide and the temperature of the bed is certainly such that the recarbonization can occur.

What can be done about improvement in this situation? I think you know the answer the precalciner was
developed. By adding another vessel to the bottom of the preheater system and introducing about 60% of the
total fuel requirements in this vessel, calcination can be increased to about 90% in a very short additional time
interval and with increased efficiency. In this vessel the advantages developed in the preheater are restored and
in some cases accentuated.

In the precalciner system there is an additional consideration not present in the simple preheater system.
Combustion air must be supplied to the precalciner vessel. The fuel in the precalciner does not burn with a
conventional flame. It simply oxidizes rapidly (combusts by some descriptions) with localized liberation of
substantial quantities of heat. In this situation it is doubtful if ambient air would support combustion due to the
chilling effect of the air. For this reason air is supplied from a grate cooler in the range of 1650°F. The
preheated kiln feed material enters the precalciner at about 1400°F. Due to the confined space and intimate
contact of material and combustion products, the heat energy supplied to the precalciner is used mainly to
perform calcination.

The precalciner outlet temperature is controlled in the range of 1700°F. The degree of calcination is held to 90
- 95% complete. The question can be asked, "why not 100% calcination?" There are essentially two theoretical
reasons. If the rate of calcination approached 100%, the temperature of the precalciner exit gas would rise to
an unacceptable level since the endothermic reaction of calcination would be complete and the available heat
would simply raise gas temperature. Secondly, the time required to drive out the remaining carbon dioxide is
too slow for a practical process. Again using our admittedly flawed calculations, it can be determined that it
would take 1.7 times longer to drive out the last 5% carbon dioxide as it did the first 95%. As a practical
matter, it has been found that serious buildup problems occur in the precalciner and the following ducts and
vessels when exit gas temperatures are too high.

The high degree of uniform calcination ensures that the materials entering the kiln are relatively stable in their
chemical and thermodynamic parameters. This results in very high production rates compared to a
conventional kiln and a minimum degree of cycling. As a generalization, a precalciner kiln will produce
double the rate of a conventional rotary kiln-preheater combination of the same kiln diameter. In a Precalciner
configuration the kiln is about 60% of the standard four-stage preheater kiln length. Almost the entire kiln is
used for clinkering with the advantage of a clinker coating protection for the refractory.
The temperature profiles in the preheater vessels prior to the precalciner are similar to those in a conventional
preheater. The exit gases can be used for drying raw materials in. any of the accepted systems.
In standard wet or dry process kilns the calcination is, of course, preformed completely within the kiln. The
calcining zones are somewhat similar. It is probable that some of the nodules formed in the chain system of a
wet process kiln survive well into the calcining zone. In this case the temperatures or time required for
calcination will be higher or longer than in the dry process. We know from our experience with the ACL kiln
in Roanoke that the nodules produces for a semi-dry system persist completely through the burning zone.
Presumably other variations occur in kilns fed with filter cake or other pelletized feed.

Perhaps we should spend a moment here discussing a plant manager's nightmare. Few things can be worse that
a crash stop of a rotary kiln with no ability to turn the kiln. For the serious kiln student this can turn into a
once-in-a-lifetime bonanza. Once the kiln has cooled one can enter the kiln for observations and sampling. No
matter what else is at stake, I would recommend that the necessary time and effort be put into thorough study
of the kiln at such a time. The operating kiln will not be totally represented by samples taken after cooling but
there will be sufficient information and clues for serious study.

There are many mechanical means that are employed to accelerate decarbonization and/or reduce the effects of
flushing. I have had experience with a wet kiln system that was equipped with three refractory dams. It worked
quite well. Kiln efficiency improved and flushing was virtually eliminated. Other devices include refractory
crosses and tumbling ledges. The carbonate content of the raw mix has a definite effect on the calcining zone.
The higher the carbonate content the more work there is to do. At the same temperatures, kiln feed rate, and
kiln speed, the calcining zone must occupy more of the kiln length. To shorten the length of the calcining zone
the kiln must be slowed or the temperature of the calcining zone raised. In both cases, particularly the latter,
the temperature profile throughout the kiln would be altered.
The difference between a high carbonate and low carbonate kiln feed can cause substantial differences in kiln
production if feed rates are not changed to compensate. High carbonate kiln feeds result in fewer solids at the
end of the calcining zone-hence lower production without an increased feed rate. Of course, higher carbonate
mixes are most often harder to burn so the lower production may be desirable: There are also concurrent
changes in gas volumes and a velocities within the kiln system as the raw mix carbonate changes. Any
magnesium carbonate in the raw mix also involves an endothermic calcination reaction as shown in Fig. 5.
The amount of MgO is relatively small in most clinkers and is limited to 6% in cement by ASTM
specifications. The same discussions just presented for calcium carbonate are applicable to magnesium
carbonate.

It can readily be seen that the calcining zone is a very complicated area to understand. We have just begun to
consider the possibilities.
All of the common argillaceous material used in cement making contain chemically combined water. These
clay minerals are dehydrated in the calcining zone. The release of this water becomes appreciable at 930°F.
The argillaceous materials contain primarily silica and alumina with frequent significant quantities of iron. The
study of the changes, which these minerals undergo, requires the attention of many chemists. When treated
singularly these materials exhibit endothermic reactions of dehydration with increasing temperature. In the
range of 1800°F there is an exothermic reaction, which is generally the conversion of amorphous anhydrous
alumino-silicates to crystalline forms. Rarely would this be of interest in a rotary kiln since in the raw mix
there are large quantities of free lime available for reaction with the argillaceous components at lower
temperatures.

It has been demonstrated by test in various kiln systems (wet, preheater and lepol), that the free lime which
results from decarbonization of the limestone immediately begins to react with the clay minerals to form CA,
C12A7, C2F and C2S. Contrary to what one might expect, free CaO does not exist in amounts exceeding 2%
until all the alumina, iron and silica have combined. It has been previously stated that calcination begins
around 1200°F so these reactions will begin as soon as free lime is available. Free lime will reach a maximum
of about 17 - 20% (depending on the composition of tie raw mix) at about 1800°F. At the higher temperatures
of the burning zone the CaO will react with the CA, C12A7 and C2S to form, the commonly known clinker
minerals of C3S, C3A and C4AF. As a result the amount of C2S that was originally present will decrease. It
has also been shown that the major exothermic reactions in clinker formation, which have been described
above, really occur in the calcining zone, not the burning zone. The formation of melt, crystal growth and the
reaction of C2S and CaO to form C3S are the primary activities in the burning zone and are endothermic.
There is one interesting mineral that has been found to be present in all kiln systems but was at one time
thought to be an oddity. This mineral is spurrite. It has the formula 2 C2S CaCO 3. Spurrite is formed at about
1300°F and decomposes to C2S and CaO at temperatures above 1700°F. It is recognized that alkalies, chlorine
and sulphates are mineralizers, which promote the formation of spurrite. Spurrite is clearly related to ring
formation in the calcining zone. Spurrite is also, suspected of a tendency to compact into a form which resists
flow down the kiln.

Since spurrite does not decompose until a higher temperature than the decarbonization of limestone, the
calcining zone is lengthened when spurrite is present. Spurrite may be formed by the recarbonization that can
take place in the bed of material in the rotary kiln.

Alkalies and other volatile materials that enter the kiln system can cause a great deal of trouble in modern
kilns. In the old, wet kilns without dust collection the problems were simply blown out the stack. Now the
causes and-solutions of volatile related problems must be recognized and acted upon for a stable, successful
kiln operation.

These minor components in the raw mix and fuel evaporate at the more elevated kiln temperatures and
condense when cooled. If the evaporation and condensation is repeated continuously, then the concentration of
the volatile component within the calcining zone bed material will increase until equilibrium; the amount of
volatile matter leaving the kiln in the clinker must equal the amount entering the system in feed and fuel.
When some amount of the volatile matter is allowed or caused to be vented from the system, then the amount
of the volatile component in the clinker will be reduces at a new equilibrium condition. When the volatile
component is harmful to the product, such as alkalies, or harmful to the process, such as chlorides, the system
must be designed and operated so that the proper amount of volatile matter is withdrawn at the proper point so
that the desired results are achieved. Every combination of raw material, fuel and kiln system must be
considered separately.

The primary components under consideration are potassium, sodium, sulfur and chlorine. There are other
components such as fluorine, arsenic and led that have less practical significance but follow similar patterns.

Three things must be considered when looking at the volatile matter:

1. How much can be tolerated in the clinker?

2. How much can be tolerated in the exhaust gas or discarded dust?
3. Where will the concentrations within the system reach the point that operational problems will result:

It is important to recognize volatile behavior and be able to predict and/or understand what will happen if there
is:

(a) a new process contemplated,

(b) alterations to an existing process, or
(c) changes in raw material or fuel.

Once vaporized within the kiln system, the volatile component travels back up the kiln system until the gas
stream cools below the condensation temperature. The volatile changes from a gas to liquid fog particles.
These fog particles can stick to any solid surface with which they have contact; dust, kiln feed, kiln lining,
chains or whatever. If this volatile material is collected within the system prior to introduction of new feed and
remains within the system, it is part of the internal cycle of volatile returns.

The fog particles may freeze into very fine solid dust particles. These particles of volatile may leave the kiln
system and be collected with the kiln dust. The portion of the volatile matter that is returned to the kiln system
with the dust is part of the external cycle of volatile return.

Each kiln system has a natural outlet for the volatile materials. These outlets can be large, thus reducing the
internal cycle or small, thus increasing the internal cycle. A short, dry kiln with no dust collection has a very
large outlet for volatiles. A four-stage preheater has a very small outlet for volatiles.

Sulfur is particular difficult to evaluate since it is present as gaseous SO 2 which will not condense of its own
accord. It must be reacted with CaCO3 in the raw mix at rather high temperatures (1400°F) or absorbed in the
moisture contained in the raw materials.

When the natural outlet of the kiln system is insufficient to accomplish the desired result in either product
composition or process flow, then an additional outlet must be provided. In the case of a kiln equipped with a
precipitator, the operator might choose to discard any or all of the collected dust. In the case of a precalciner
kiln it might be necessary to bleed off portions of the kiln exhaust gas in which the volatile material is still
gaseous, freeze the vapors and discard the offensive solid.

This outlet is called a "by-pass" and its size is determined by the percentage of kiln exhaust gas that is
withdrawn from the system to achieve the desired reduction in volatile matter. These outlets can vary from 0 -
100% and may be fixed or variable.
There are some practical limits to a by-pass system that are dictated by economics. On a simple preheater the
penalties in fuel consumption and lost production (dust loss in the by-pass) at 30% by-pass may be so great
that such a system is no longer practicable. It is at this point that a precalciner is of real benefit; since only
40% of the gas passes through the kiln; a rather large by-pass can be achieved at a much lower penalty.
From an operational standpoint, once the sticky, fog-like particles have frozen into solid particles of fine dust,
there is no longer a handling problem. With the exception of corrosion of refractory, there is no problem with
gaseous volatiles. The process problems develop from the fog-like particles that tend to agglomerate and
collect dust particles that form rings, restrictions and build-ups. Points of build-up must be removed or
schemes must be developed so that the sticky range can be bypassed while the particles of volatile are still in
suspension.

Of particular concern to cement quality is the alkali content of the finished product. When certain aggregates
are used in making concrete, the cement must be "low alkali" to prevent deleterious reactions in the concrete.
ASTM specifications call for a maximum of 0.60% alkalies expressed as Na 2O. Alkalies can also be involved
in other product and system performance problems such as poor workability of concrete, poor flowability,
warehouse set, depressed late strengths and kiln system buildups.

The ability to produce low-alkali cement depends on two points that were just covered:

1. Alkali content of the raw materials and fuel.

2. Extent of alkali reduction permitted by the kiln system.

Alkalies in raw materials are usually in the form of complex amorphous silicates. These silicates are quite
stable until they are heated to the high temperatures required to produce portland cement clinker in the kiln.
One might then suspect that all alkali activity would be confined to the burning zone. This of course, is not
true because of the internal and external alkali cycles. Keep in mind that any discussion of alkalies applies in
whole or in part to any of the volatiles. Remember that sulfur differs slightly, as some sulfur is contained as
gaseous SO2.
Figure 6 present the melting and boiling points of some pure compounds that may be present in the kiln
system. In a clinkering liquid these temperatures may vary considerably. There are other compounds,
particularly alkali sulfates that may exist as well. When melting first occurs, there also exists slow
vaporization. The vaporization rate increases with increasing temperature. It is very rapid in the burning zone.

K2O and Na2O can form some complex clinker compounds. Some of these are retained in the clinker and some
readily vaporize. Whenever there are sulfur and alkali present in the raw materials or fuel, there are alkali
sulfates in the clinker. In preheater kilns in particular, an effort is made to have a chemical balance between
sulfur as sulfates and alkalies. The alkali sulfates are the least volatile and remain in the clinker. When they
leave the kiln, it becomes impossible for alkali deposits to plug up the preheater at some point. This procedure
works well without a by-pass as long as low alkali cement is not required.

At one time, calcium chloride was added to raw mix to reduce alkalies since the alkali chlorides are much
more volatile than the sulfates. If this addition is to be of real advantage, the kiln must have a natural or
engineered outlet. It is not good procedure to introduce chlorides into a system where they will latter condense
and cause restrictions. Chlorides are also corrosive and accelerate the deterioration of precipitators and duct
work.
When sulfur exceeds the alkali level, the sulfur must exit the system in the clinker, be deposited in the pre-kiln
vessels and ductwork as calcium sulfate, or leave the system as SO 2. In more than one of our plants, we have
found the double sulfate salt of potassium and calcium K2Ca2(SO4) present in the clinker. This salt changes to
syngenite, K2Ca2(SO4)2 H2O when, water vapor is present. The syngenite is implicated in problems relating to
warehouse set. In one of these plants it was frequently impossible to add gypsum-to the clinker and remain
under ASTM specifications because of the sulfate level in the clinker. Of course, setting time problems
resulted because clinker sulfate is not sufficiently soluble to properly retard the setting time. It was necessary
to be more selective in raw materials and to use a lower sulfur coal to eliminate the problem. While it has
nothing to do with the calcining zone, at Lone Star we believe that syngenite can be formed in cement storage
by reaction between K2SO4 and CaSO4 H2O.

In Figure 7 a representation of the internal and external alkali cycles and their relationship to the total kiln
system can be seen. Figures 7A and 7B show a preheater and preheater with by-pass.
Let us now look at the necessary considerations for production of a low alkali clinker. In most kiln systems,
regardless of fuel, most of the alkalies are contained in the raw material.
As previously stated, if there is no outlet to the kiln system, the internal and external cycles increase in alkali
content until the clinker has the same quantity (a higher percent) alkali as the kiln feed. If the raw material
alkali is low, then the clinker alkali will be low; if the feed alkali is high, then the clinker alkali will be high.
Alkali levels can be reduced only by reducing the amount of alkali returned to the system in the external cycle
or by improving the rate of volatilization in the burning zone.

The relative amounts of alkali in the internal and external cycles depend on the kiln system, but the factors
involved in improving volatilization are much the same for all kilns.

These factors are:

1. Total alkali. Naturally, the more alkali present the longer the time required volatilizing it. At higher
levels there is a greater likelihood that it will remain in the clinker.
2. Types of alkali compounds entering the burning zone. The high-boiling sulfates will remain in the
clinker to a greater degree than the more volatile compounds. Generally, there is a greater percentage
reduction of potassium than sodium due to volatility.
3. Temperatures. The higher the temperature the faster the rate of volatilization.
4. Time. If the temperature profile is such that a longer length of the kiln is held at volatilizing
temperatures, clinker has more time to travel this section of the kiln and lose alkalies. In subsequent
sessions you will learn that this situation my not be beneficial to other clinker parameters.
5. Clinker composition. In the turning zone the liquid portion of the clinkering mass has several effects
on the volatilizing rate. The amount of liquid has considerable influence on clinker size (grate
preheater kilns excepted). The bigger the clinker the longer is the time required to volatilize the alkali.
Alkalies volatilize readily from the clinker surface but diffuse slowly from the center to the surface. At
any temperature, the liquid in the clinker seems to have an alkali content level that can be reduced.
The amount of liquid tends to determine the limit of achievable alkali reduction. It has been reported
that at some plants the total alkali has been reduced by as much as 0.5% by reducing the liquid 2 to
3%.
The temperature of burning must be increased to obtain a reasonable level of free lime in the clinker.
All means of attaining this higher temperature tend to lower the potential alkali content of the clinker.
Common means are to raise the lime saturation factor or the silica ratio.
6. Type of kiln. Each type of kiln system has a generally expected reduction in alkalis depending on the
treatment of the internal and external alkali cycles. Figure 8 shows expected alkali reduction expressed
as percent of alkalies in the kiln feed.

Kiln No. 1
They discussed a kiln system consisting of a single stage pre heater and a long, dry kiln. The kiln feed was rich
in all kinds of volatiles. This is shown in Figure 9. This figure also shows the profile of volatile materials in
the calcining and transition zones of the kiln. The total kiln length is 655 ft.

As, the charge moved down the kiln there was a steady and nearly parallel increase in K 2O and Cl content
before 320 ft indicating probable condensation of KCl. The SO3 and Na2O levels remained fairly constant,
which indicated that condensation of alkali sulfates was not appreciable. Beyond 320 ft the rate of increase of
K2O and SO3 suggests condensation of alkali sulfates. The increase of Na2O was small due to its low volatility.
The Cl peaked at 240 ft where the kiln charge temperature is estimated at 1650°F. At this point KCl had begun
to volatilize. The K2O peaked a bit later because of a faster rate of condensation of K2SO4 than volatilization of
KCl. The condensation of sulfate leveled off beyond 220 ft as indicated by the slope of the SO 3 and Na2O
curves and only after that point, the K2O decreased rapidly by the volatilization of KCl. The material
temperature at 140 ft is estimated at 2200°F. The liquid content between 200 to 240 ft amounts to more than
11% of the total material.
The major phase profile of the same kiln is shown in Figure 10. Maximum KCl concentration was about 240
ft. Free lime did not appear in any appreciable quantity before the KCl started to evaporate and reduce the
amount of liquid present. Before this point calcite was gradually converted to spurrite. There were indications
that spurrite increased in direct proportion to the amount of liquid present. The maximum concentration of
spurrite, nearly 60% of the kiln material, existed at the location of maximum liquid concentration. The total
amount of spurrite is very high but remember that alkalies, chlorides and sulfates promote formation of this
mineral and that there is a high level of these materials in this mix.

Subsequent to this point the spurrite decomposed into C2S. The high spurrite concentration and high liquid
content in the kiln charge caused severe operating problems in this kiln. The passage of the lords was
increasingly hampered to the 220 ft point. Here the spurrite diminished rapidly and the liquid decreased
slightly. Suddenly the iron and alumina portion of the melt was formed at 140 ft. The load became a fast
moving lava flow just before the clinkering zone: Needless to say, the kiln was difficult to control and
operation was unsteady and cyclic. Coating was impossible, and brick life was about six weeks.

The addition of gypsum did not prove effective in reducing the recirculation of volatiles. Contrary to
expectation the potassium chloride appeared more stable that K 2SO4 and CaCl2. Reducing the silica ratio to
improve the burnability of the clinker did not help either since KCl volatilized below clinkering temperatures,
because of the long kiln and the resulting relatively cool gas in the preheater risers. A by-pass was only
partially effective. Most of the volatiles have already condensed in the kiln as part of the internal cycle. The
first effective remedy was to reduce the external cycle by wasting increasing amounts of kiln dust (volatiles)
until stable operation was achieved.

Later help was found in a new source of limestone, which reduced the level of Cl to one tenth of the previous
level. Eventually, this proved to be the best solution.

At the fall 1980 G.T.C. meeting, one of the speakers tried to promote the idea that time spent on subjects such
as raw materials investigations in the early stages of a project would prevent situations Such as just described.
This is not always done and frequently a high price is paid.

Kiln No. 2
In the second kiln, a long dry kiln with chains, we want to consider a system with low contents of sulfur and
chlorine in the kiln feed but high sulfur levels in- the fuel. The raw mix volatiles are shown in Figure 11 along
with the volatile profile. The recirculation of chlorine was not of great importance K 2O and SO3 predominate.
Significant amounts of SO3 are combined with CaO as evidenced by the higher SO3 concentration compared to
K2O between 300 and 135 ft.

The calcium sulfate was in the form of langbeinite K2Ca2(SO4)3 sulphospurrite, 2C2S CaSO4 and other similar
minerals. It is interesting to note that the SO3 level at 165 ft continues to increase despite temperatures of
2200°F. The clinker information is not available but it is reasonable to expect that the clinker SO 3 level is quite
high.

The broken lines on this and the next chart indicate compositions in a ring in the chain area. It is thought that
the chains facilitated the condensation of alkali and promoted the formation of spurrite. The spurrite in turn
was responsible for the ring formation and resulting back spill. Removing some of the chains to reduce the
localized alkali and spurrite concentration solved the problem.

Figure 12 shows the phase profile of the kiln and the ring. The ring composition is clearly different from the
kiln feed at that point and contains a large amount of spurrite. Please note on this figure and in Figure 10 that
the free lime curve, the calcium carbonate curve and the C2S curve all behave as was described earlier in this
presentation.
The examples that we have just discussed illustrate the point that kiln chemistry is not so simple and that an
understanding of volatile circulation as well as other facets of kiln chemistry are important. It should be
understood that, for the flow of kiln material to be interrupted, an obstruction has to be present in the kiln. The
obstruction may be caused by the formation of liquids in the kiln load or by the formation of such compounds
as spurrite and resulting rings. The obstruction can be temporary and may not cause a large upset. It can form
again; however, in such a manner that upsets or cycles will result. It can also be possible that a very hard ring
is formed with undesirable consequences resulting.

There is one form of volatile material that does not have a cycle. This is combustible material in the raw mix.
Many raw materials in use today contain carbon, kerogens, and other combustible material. There are certainly
materials on the horizon that will have combustibles. Spent oil shales have residual fuel value. It may even be
possible to use some oil shales directly in a precalciner.

Whenever a new preheater kiln is considered, the raw material should be checked for combustibles. Most often
the carbonaceous material will burn in the preheater at some undesirable spot and contribute to unsatisfactory
temperature profiles and blockage or damage to the preheater.

In non-preheater kilns some combustible material can be tolerated, but attention must be paid to temperature
profiles to prevent high back end temperatures, chain damage, or ring formation.

The precalciner permits use of highly carbonaceous raw materials since this material can be introduced
directly into the calcining vessel.
It is often difficult to measure or observe what is going on in a rotary kiln. We must often make some indirect
measurements and attempt to assess what is really going on inside the kiln. One convenient measurement is
kiln torque. Torque is normally measured in foot-pounds and is defined as that moment of force, which
produces or tends to produce rotation. I have never seen a torque meter on a kiln but amps on the kiln drive
meter are related to torque and can be used to estimate changes in torque. We don't really care what the true
value of torque might be; we do care whether torque (amps) is increasing or decreasing, and we need some
judgment about the magnitude of change.

At any level of kiln speed a certain torque is required. If the amount of kiln feed is changed, the torque will
change. Many mechanical changes in the kiln system such as rubbing seals or tires running on thrust rolls will
change torque. Forces ranging from changes in kiln speed to the weather will change torque. In any use of
torque (amps) as an indirect measurement, care must be exercised to assure that there is no interference with
the observation or conclusions. Clearly an element of experience and knowledge of the system in question is
required for interpretation of events.
Assume that our kiln is running smoothly at set point for feed, speed, firing rate and draft. If there were no
mechanical influences, a recording ammeter would be drawing a straight line. A sudden change such as a ring
dropping out would increase kiln amps sharply; any gradual change would be recorded as a drift in the ampere
readings. Ring build-ups in themselves do little to change amp reading since they are more or less symmetrical
to the center of rotation and would act as flywheels. A ring does dam up material and this increased inventory
within the kiln will be recorded as an increase in amps.
If the build-up is in the raw material area, more torque will be required to lift more material. If there is a
spurrite ring in the calcining zone, the restriction may not be so noticeable due to the airslide effect of the
evolving carbon dioxide.
The effect in the burning zone is quite pronounced and can be a useful measurement. The torque requirement
has a direct relationship to burning zone temperature. As the temperature becomes hotter the clinker material
becomes sticky and rides higher up the side of the kiln. This requires more energy (amps) to lift the bed. If the
kiln increases further in temperature so coating is burned off, the amps will continue to increase since the
amount of material being lifted is increased. Conversely, if the kiln gets cold the material, no longer sticky, lies
in the bottom of the kiln and the amps go down.
Events in the calcining zone follow the same pattern. If the load is calcined too far back in the kiln, it will not
be transported to the burning zone by the airslide effect. It must be lifted, which takes more power. In most
cases when this occurs the burning zone is also too hot. A reduction in fuel rate will correct the kiln problems
and return the drive amps to normal.

I have been responsible for a kiln that used kiln amps to automatically "trim" the fuel rate. Within fairly
narrow limits the system worked well, often the amps indicated subtle changes in the kiln operation before the
operator could have been aware of them.

Later in the course you will receive instructions in the use of the microscope. Although I do not have the skills
to properly use a microscope, I am convinced that the information derived from microscopy can be-one of the
best tools a kiln operator can have. We can actually see what has been done on the calcining zone of the kiln
and if things have not gone properly, we can get clues for identification and solution.

Generally speaking, a good clinker will have a small amount of free lime. This means that we are not wasting
fuel by overburning and sometimes, but certainly not always, that we have had the raw materials combine well
into clinker minerals. With strong reservations about raw mix fineness and proportioning, it can be said that a
small amount of free lime indicates that the kiln is operating below temperatures that damage refractory or
cause the load to go liquid.

Consideration of free lime includes the crystal size, distribution in the clinker, source of the free lime and
temperature of burning.

The size of the free lime crystal is most often determined by the size of the calcium carbonate particles in the
raw mix. Oyster shells can give very large crystals while chalk almost never does. The time-temperature
relationship in the calcining and transition zone can be, such that the CaO particles are permitted to grow to
such a size that they cannot react with C2S in the burning-zone and will appear in the clinker. Free lime can
also result from the decomposition of C3S in overburning. This free lime is sometimes called hard burned free
lime. It is very dense and is not reactive during the standard free lime test nor does it slake during the concrete
mixing period. Hand burned free lime can also result from overburning a coarse mix in a misguided effort to
burn out all the free lime.

Unless the free lime is finely divided and evenly distributed to facilitate slaking, a deleterious expansion
results from the hydration of the free lime in the cement paste matrix. The internal expansive forces literally
destroy the concrete. The autoclave bar test reveals any tendency of the cement toward expansive properties. I
have seen surprises where the "bar blew" when the free lime test indicated no problem.

I am convinced that in each raw mix there is an optimum free lime. The value depends on several factors.
Some of these have to be the reactivity of the raw materials, the raw mix fineness, the raw mix composition
(L.S.F., S.R., etc.) and the kiln system. What we are learning from microscopy and other modern techniques
shows that to arbitrarily establish a level of free lime in a clinker is folly of the first order. There are too many
other important parameters of Portland cement clinker to use free lime alone as the measure of efficiency or
quality.
An old standard test that is now seldom used is the liter weight test. It can be quite useful when properly
applied. The test is based on the premise that the harder a clinker is burned, the denser it will be. This is
reflected in the weight of a liter of clinker.

The test is quite simple and can be performed by a burner or his/her helper as often as necessary. Care must be
used to run the test the same way each time in accordance with standard procedures. The clinker sample can be
taken from the cooler, or the kiln.

The clinker is sized between any convenient screens such as a 3/8 and a No. 4. The material on the No. 4
screen is placed in a hopper with a gate for a fixed fall into a liter container, which is the frustum of a cone.
The top is struck off level and the weight of the contents is determined in grams.
The aim for a particular kiln may be 1300 grams per liter where the free lime is 0.6%. All other things being
equal as the temperature increases in the kiln the liter-weights will increase and the free lime becomes less. It
has been observed that it is impossible to get the high liter-weights (1600 g/1) in the large diameter new kilns
that are possible in the older kilns.

As was previously noted concerning an optimum free lime for each raw mix, each raw mix has its own free
lime vs liter weight relationship. Comparisons should always be related to the same raw mix composition in
the same kiln system.

Some typical examples are shown in Figure 13.

Figure 14 shows the liter weight apparatus.

Curve "A" shows a clinker that seems to have a "natural" level of free lime that is quite low over a wide range
of liter weights. Curve "B" is perhaps more typical clinker a gradual increase in free lime with decreasing liter
weight. Curve, "C" is one that I hope you don't have. The free lime is within acceptable limits over a range of
hard burned liter weights but once the liter weight reaches a more or less "normal" level the free lime quickly
becomes excessive. This is a bad clinker to burn, the kiln will no doubt cycle, the brick life will be terrible and
cement quality will be unacceptable. The operator of such a kiln should look closely at the raw materials and
head for the nearest microscope.

In general liter weight samples should be taken during a period of more or less steady kiln operation. For
example, if a kiln has just cycled and a quantity of underburned material has pushed through the burning zone,
the sample taken will likely be a mixture of high and low liter weight material. The resulting test answer does
not provide much useful information.

The color of clinker is primarily related to the iron-bearing phases: the more iron the darker the clinker; the
less iron the lighter the clinker. Great effort is made to eliminate or minimize iron contents of white clinker. In
a cement where all the alumina is chemically combined with calcium and iron to give a zero C3A clinker, the
clinker color is very dark due to the levels of iron employed in the chemistry.

For normal Portland cement the clinker should be dark velvety gray. Good clinker has the appearance of
relatively small, rounded balls. We used to think that clinker had to sparkle in the sun to be good. This is still
true on many kilns, but I have operated kilns, which were making perfectly satisfactory clinker that did not
have a sparkle in a carload. I understand that sparkle is not necessarily a characteristic of clinker from larger,
modern kilns.
The kiln atmosphere, whether oxidizing or reducing, has a great bearing on clinker color. There can be
localized reducing conditions due to flame position or impingement. Reduction of the ferrite phases changes
clinker from gray to various shades of brown. The resulting cement will be brown or buff. The original buff
cements were made in poorly operated kiln systems.
Some clinker has a black outer shell and a brownish interior. Most often this clinker has experienced a strong
localized reducing condition followed by subsequent oxidation, which has not penetrated beyond the surface
before the clinker was cooled below reaction temperatures. Overburning of a clinker can produce a clinker of
similar appearance with regard to color. At high clinkering temperatures there is some dissociation of ferric
oxide to ferrous even in an oxidizing atmosphere. Reoxidation occurs during cooling but is not complete. It
has been my observation that clinker of this nature has a more yellowish, than brown interior, is usually more
dense and the interior looks like cooled melt rather than crystalline even to the naked eye. It is reported in the
literature that there are measurable FeO contents in both these clinkers.

The rate of cooling also affects color. Very rapidly quenched clinker tends to have a
brownish-reddish-yellowish cast. This is ascribed to a reduction in crystalline iron compounds and a
replacement with glass. Dark brown colors can be produced with slow cooling in a reducing atmosphere.
The literature states that increasing MgO levels tend to darken ferrite crystals. I am aware of one instance
where the cement color was lighted by increasing the level of MgO.

Some other constituents also give color to cement clinker. High quantities of manganese oxide import a brown
color to the clinker. A bit of basic refractory that contains chrome can be found at the center of a clinker that
has a green color.

Of course we all know that underburned clinker is quite different in color from good clinker. This color will
vary from a very dull gray to light colored patches to a chalky white or yellow appearance with lots of dusting.
These clinker nodules are very weak. This cause is obvious. The clinker never got hot enough.

Normally a hotter kiln will produce larger clinker. The converse is also true. Again one must be careful of
generalizations and restrict such comments to a specific kiln system. In a wet kiln system a major change in
the chain system produced a much smaller clinker at the same clinkering temperatures.

The color of the clinker inside the kiln is also important to kiln operation. This action of the bed and how high
the clinker rides up the walls along with color are indications of burning conditions. It is hard to ascribe colors
to certain conditions in the kiln since all of us have our favorite color viewing glass.

No matter what color the filter I suspect that we could all agree that a dazzling, highly incandescent white
indicates that the clinker is pretty hot and the kiln might require a lot of attention. Only through experience can
an operator become proficient in examining the clinker color both within and without the kiln to determine
how well that operator is doing.

Earlier we discussed the fact that the carbon dioxide level in the atmosphere had a definite bearing on the rate
of calcination; the higher the carbon dioxide concentration the lower the calcination rate. The following facts
are somewhat academic since our industry is changing to coal as rapidly as possible. Nevertheless, the higher
the carbon content of the fuel, the higher the percent carbon dioxide in the combustion products. Therefore, in
delivering the same amount of heat, with complete combustion and no excess air the products of combustion
for. The three common fuels would be:

Coal 18.3% CO2

Fuel Oil 15.8% CO2
Natural Gas 12.2% CO2

Based on this information we can expect the calcination rate to be greatest on natural gas and the least on coal.
This effect would be most pronounced in a preheater or precalciner. In a traditional rotary kiln the effect would
be minimized. The atmosphere in the bed is nearly 100% carbon dioxide so the effect would be limited to the
surface alone.

Throughout recent history a tremendous amount of basic and applied research has been carried out to clarify
the process of cement manufacturing.

I hope that you now have an enhanced realization that many different and intermediate reactions take place at
different reaction rates in the calcining zone of a cement kiln system. The processes are complex and the
factors that influence these processes are varied. Certainly there are no two identical kiln systems. It is
important for kiln operators and supervisors to understand the sequence of events in the kiln system and under-
stand the interrelationships of the variables.

The zones of a kiln system are not as clear as the oversimplified Bogue model would have us believe. There is
considerable overlap of the zones, not only as a natural consequence of the process, but due to cycles in the
kiln system, which have been caused by internal and external factors. Reactions, which should theoretically
occur later in the process, may take place before earlier ones are complete.
We have briefly looked at some of the factors that affect reaction rates in the calcining zone. These include
temperature, time, particle size, surface area, composition of the charge, the kiln atmosphere, the mechanism
of the kiln system and the fuel.

The volatile and dust cycles which occur in a kiln system invariably affect operation of the kiln system, the
environment, or the product. The effects are seldom helpful to the operator.

Overall temperature profiles in a kiln system are constantly changing. A single particle undergoes numerous
temperature changes as it comes to the top of the load then is buried in the cooler bed of material. Chemical
reactions start, stop and start again. Some are reversed.