Cement quality is typically assessed by its compressive strength development in mortar and concrete. The basis for this property is well-burned clinker with consistent chemical composition and free lime.

There are only two reasons for the clinker free lime to change in a situation with stable kiln operation and fuel ash: variation in the chemical composition of the kiln feed or variations in its fineness. Variations in fineness depend on possible changes in raw materials or in operation of the raw mill. Variation in chemical composition is related to raw mix control and the homogenization process.

To ensure a constant quality of the product and maintain a stable and continuous operation of the kiln, attention must be paid to storage and homogenization of raw materials and kiln feed. This article discusses the role of raw mix control and the homogenization process, assuming the raw materials and fuel ash do not vary and the raw mill operation is in full control. Suggestions are made on methods to improve homogenization.

Why homogenization?

Homogeneity and burnability

Before discussing the effects of homogeneity, let's take a general look at the transformation of raw materials into clinker. The process of clinker formation is described in Figure 1. The transformation concludes with the primary clinker phases:

Alite: Impure tricalcium silicate, generally termed C₃S

Belite: Impure dicalcium silicate, normally termed C₂S

Aluminate: Tricalcium aluminate, C₃A

Ferrite: Nominally tetracalcium aluminoferrite, C₄AF

The amount of the clinker minerals formed is determined by 1) the time and temperature treatment of the mix, and 2) the overall chemical composition of the kiln feed. If the pyroprocessing time is too short or the temperature too low, combination of the raw material components may be less complete and some free unreacted lime will be present. Insufficient control of the raw mixture and its blending will cause large variations in the chemical composition of the kiln feed.

If the kiln is operated at a constant material residence time and temperature, such variations also will cause variations in clinker composition, including free lime. This is important because the free lime is usually used as the process parameter to indicate how well the clinker is burned. When unintended variation in kiln feed composition causes large variation in free lime, operators may make incorrect changes to kiln operation, assuming changes are needed when they are not.

Variations in the kiln feed chemical composition affect its burnability and eventually the fuel consumption. This relationship has been described elsewhere (Hills, Johansen, and Miller 2002). In summary, the free CaO remaining after burning kiln feed for a fixed time and temperature can be determined by the following (Theisen 1992):

Variations in composition are directly reflected in CaO₃₀¹⁴⁰⁰. Higher values will result in higher clinker free lime but still give a well burned clinker. If an increase in clinker free lime is perceived as “underburning” and countered by attempts to burn harder, the result will be increased fuel consumption. Introducing the ratio between the CaO₃₀¹⁴⁰⁰ and the setpoint for clinker free lime as the parameter for the degree of burning in the kiln provides an estimate of the relative changes in fuel consumption as a function of variation in the degree of burning. Figure 2 illustrates this relationship.

To discuss the homogeneity aspect of clinkering, let's look at two examples. These scenarios will assume that we begin in the same plant, under the same conditions, with the same inhomogeneous kiln feed. The examples demonstrate two methods of responding to the feed inhomogeneity.

Effect on Fuel Consumption, Kiln Operation, Clinker, Cement Performance

Example I — Response: Burn Harder
Discussion

Insufficient mix control or blending will result in larger variations than anticipated in chemical analysis. The result is variations in clinker free lime. The operator may be obliged to increase the burning zone temperature to achieve the desired free lime level — by keeping the kiln on the hot side, the maximum clinker free lime is brought to the average value and the spikes are eliminated.

As a result of this harder burning, fuel consumption increases. The fuel penalty for burning to an average of 0.8% free lime because of large variability instead of an average of 1% can easily be on the order of 4% (Hills, Johansen, and Miller 2002).

When the kiln is operated on the hot side, alkalis and sulfate become more volatile. This, in turn, might increase the possibility for build-ups in the cooler parts of the kiln system. In severe cases, controlling the kiln may become difficult because of surges of the material through the kiln.

Hard burning tends to cause low clinker porosity, large crystals of alite, and often contributes to generation of dust instead of good, nodular clinker (Miller 1980; Johansen 1990). It also slows down the cooling process, both because the maximum temperature is higher, and because the low-porosity clinker is more difficult to cool.

These effects all can result in cement with reduced strength potential and increased water demand. Reduced clinker porosity can make the clinker harder to grind, increasing finish mill power consumption or reducing mill production. Clinker temperatures exiting the cooler may increase, further increasing fuel consumption and presenting handling problems. The high-temperature conditions may lead to color variations, reductions in clinker alkali and sulfate level, and increases in water demand attributable to increased levels of aluminate.

Variations in clinker alkali and sulfate will affect concrete setting time, and result in strength variations. Periods with decreased clinker alkali content will result in a decrease in early strength and increase in later-age strength; the opposite can occur during periods when the clinker alkali content increases. With such variability, fresh concrete often develops admixture incompatibility and changes in its rheological behavior.

Example II — Response:
Improving Homogenization
Discussion

Sufficient homogenization through mix control or blending reduces the erratic spikes in the clinker chemistry. The free lime target is much easier to obtain, and the kiln is more stable. Fuel consumption will remain similar or decrease slightly. The clinker and resulting cement performance will be much more uniform.

Homogenization — How?

Improving homogenization

Be in full control of raw mix and have full silos. This will save fuel and give consistent quality and thereby same time and money.

Homogenization of raw materials is carried out in pre-blending stores and homogenization of raw meal in silos. With proper layout, all storages for raw materials or raw meal can be operated as blending storages with varying efficiency. Raw materials typically are stored in piles representing seven days production, and raw meal is stored in silos that hold three to dour days production. Strategically located on-line analyzers can provide information on the chemical composition of material fed to the raw mill, making it possible to apply a feed-forward control to the raw mix composition and thereby, in principle, reduce the required blending ratios for kiln feed silos.

The efficiency of a blending system can be characterized by the ratio, H, between variations in composition of the incoming and the outgoing material. An example is the ratio between the standard deviation of CaO in the raw meal before and after the kiln feed silos. Typical values of H for kiln feed blending systems operating at peak efficiency may be as high as 10:1, as reported by system manufacturers. There are various layouts of raw material storages for pre-blending, and silo systems for blending of raw meal.

Raw material storages

There are a number of ways to stack limestone and clay. Most common are the chevron or windrow methods, in which the materials are stacked in many layers in the longitudinal direction of the raw material pile. The material is reclaimed from the entire cross section of the pile and the number of reclaimed layers in the cross section is the important parameter for the blending efficiency.

The composition of the individual layers in the reclaimed cross section is determined by the variation of the limestone or clay in the quarry. By reclaiming a pile cross-section with N layers, the composition represents an average of the N layers. The ratio between the standard deviation of the composition of the material going to the raw material pile and that of the reclaimed material can be shown to be -N. This factor is the efficiency of the blending process. In principle, the greater the number of stacked layers, the better the homogenization will be. An alternative is the continuous stacking in a circular pile; here the number of reclaimed layers is smaller.

Using an on-line analyzer on the material from the crusher to the storage makes it possible to keep records of the reclaimed material composition to control the raw mixture composition (Blumbach and Petersen 2002).

Raw meal storages

Raw meal is homogenized in silos. These may work continuously or batch-wise, in multi-stage systems, and in combination with air agitation. In modern plants, the trend is large, continuously operating silos. In most systems not using air for blending, the material filled into the silo forms layers representing the variation in composition, and the reclaiming process is designed so that a large number of these layers are represented in the material leaving the silo. The simplest example is the funnel flow system, in which an inverted cone is formed over the funnel formed by the central discharge. The material slides down the cone cutting a number of layers. In this way materials from the different layers are mixed.

It follows that the silos, working as continuous blending systems, have to be full in order to have a blending effect. If the silos are only partially full, the variations in the raw mix composition from the raw mill will go directly to the kiln. In modern raw mill control systems, the variation in raw meal composition is controlled by bringing the sum of variations from the set point of a parameter, whether it is C₃S or LSF, to zero within a given time, ts. The t_s has to be lower than the retention time, otherwise the blending system will not be able to blend. If t_s is too large, the variable raw meal composition will pass right through the blending system. In other words, t_s should be as short as possible in order to obtain optimum blending efficiency.

Conclusion

The homogeneity of feed chemical composition and fineness has an important relationship to fuel consumption, kiln operation, and cement performance. Reacting to inhomogeneity by burning harder results in increased fuel consumption, possible kiln buildups due to increased alkali and sulfates, clinker with low porosity, large alite, poor nodulization, and variation in alkali sulfate content. And also may result in cement with increased water demand, decreased early strength, and abnormalities in setting behavior. On the other hand, reacting to variations in feed by improving its homogeneity will avoid these difficulties and produce more uniform clinker, and therefore cement with more uniform performance characteristics.

Suggestions to improve homogenization include:

• When stacking raw materials, stack more reclaimed layers to promote blending efficiency. The higher the number of layers, the better the homogenization.

• In large, continuously operating raw meal silos, keep the silos full to maintain effective blending.

References

Blumbach, J. and Petersen, S.T., “Investing and Renovating”, World Cement, September 2002, pp. 123-128

Hills, L.M, “Clinker Formation and the Value of Microscopy”, Proceedings of the Twenty Second International Conference on Cement Microscopy, Montreal, 2000, pp. 1-11.

Hills, L.M., Johansen, V.C., and Miller, F.M., “Solving Raw Materials Challenges”, Conference Record of IEEE-IAS/PCA 2002 Cement Industry Technical Conference, May 2002, pp. 139-151.

Johansen, V., “Cement Production and Cement Quality,” Materials Science of Concrete, Vol. 1, Edited by Jan Skalny, American Ceramic Society, 1990.

Miller, F. M., “Dusty Clinker and Grindability Problems: Their Relationship to Clinker Formation”, Rock Products, April 1980, pp. 152-157.

Theisen, K., “The Influence of Raw Mix Burnability on the Resulting Cement Clinker”, Proceedings of the Fourteenth International Conference on Cement Microscopy, 1992, pp. 74-88.

Vagn C. Johansen, senior principal process scientist and manager; Linda M. Hills, senior materials scientist; F. MacGregor Miller, senior principal process scientist; and Richard W. Stevenson, senior chemist and supervisor, all work for Construction Technology Laboratories, Inc., Skokie, Ill., (+1) 847-965-7500, www.ctlgroup.com. The authors appreciate the contributions of Kenneth Hooker. This article was adapted from a presentation at International Cement 2002, Chicago, Ill.

To 700°C

Water is lost from clay minerals. Dehydrated clay recrystallizes. Some reactive silica may displace CO₂ from CaCO₃.

700-900°C

As calcination continues, free lime increases. Calcination maintains feed temperature at around 850°C. Aluminate and ferrite form.

900-1,150°C

Reactive silica combines with CaO to begin stages of C₂S formation.

1,150-1,200°C

When calcination is complete, temperature increases rapidly. Small belite crystals form from combination of silicates and CaO.

1,200-1,350°C

Above 1,250°C, liquid phase is formed. Belite and free CaO form alite in the liquid.

1,350-1,400°C

Belite crystals decrease in number, increase in size. Alite crystals increase in size and number.

Cooling

Upon cooling, the C₃A and C₄AF crystallize from the liquid phase. Lamellar structure appears in belite crystals.

(Hills 2000; Hills, Johansen, and Miller 2002)

CaO₃₀¹⁴⁰⁰ = [0.343(LSF-93) + 2.74(SR-2.3)] + [0.83Q₄₅ + 0.10C₁₂₅ +0.39R₄₅]

Where:

CaO₃₀¹⁴⁰⁰ = is the free lime after burning for 30 minutes at 1,400°C

LSF = %CaO/(2.8%SiO₂ + 1.18%Al₂O₃ + 0.65%Fe₂O₃)

SR = %SiO₂/(% Al₂O₃ + % Fe₂O₃)

Q45 = % quartz grains coarser than 45µm

C₁₂₅ = % calcite grains coarser than 125µm

R45 = % other acid-insoluble minerals, (e.g. feldspar) coarser than 45µm

	BEFORE	AFTER — burning harder
Clinker/Kiln Operation Possible Effects:	• large variations in free lime • poor belite distribution	• decrease in free lime • low porosity, difficult grindability • large alite • possible poor nodulization • variation in alkali sulfate content • kiln on the hot side • increase in alkalis and sulfate in kiln internal cycle, possible surges, potential for buildups • low porosity makes it hard to cool • lower clinker reactivity • color differences, brown clinker center
Cement Performance Possible Effects:	• possible erratic expansion results due to free lime	• increased water demand • decreased early strength and increased • admixture incompatibility later strength during periods where alkalis • abnormalities in setting behavior are decreasing • pack set due to static charge (large alites)

	BEFORE	AFTER — burning harder
Clinker Potential Effects:	• poor distribution of free lime and belite	• good distribution of free lime • good distribution of belite • better clinker uniformity • kiln is easier to control
Cement Performance Potential Effects:	• possible erratic expansion results due to free lime	• less variability, more uniformity • smaller alite crystals, enhanced reactivity, possibly allowing lower cement fineness.