Process and quality optimization in cement plant.

Process and quality optimization in cement plant.

The relationship between the burning condition of cement clinker and the ultimate cement properties is well established and being so, determination of burning condition is important in controlling the cement quality. Examination of a large number of cement clinkers from different plants and laboratory clinkers demonstrate that the ratio between the diffraction lines, corresponding to "d" 2.78+ (2?=32.2) and "d"2.74+ (2?=32.7) bears a distinct relationship with the clinkering process. The characteristic of these two lines (d values) is that both of them are combined products of alite (C3S) and belite (C2S) and in case of perfect crystallization, they bear a ratio of 1 to 1.1 only, termed as "C" index [Lea, 1970]. From the examination of industrial clinkers from different plants, it is observed that to achieve normal cement properties, the "C" index must not be above 1.5 [Goswami and Panda, 1986; Goswami et.al, 1991; Goswami et.al, 2016]. Thus "C" index reflects the degree of crystallization, and the higher the "C" index, lesser is the degree of crystallization. Accordingly, there exists a distinct relationship between the "C" index, burning condition and degree of crystallization of cement phases. As the burning condition determines the cement quality, the "C" index can be used directly for predicting the cement properties. For example, during the production of the clinker with"C" index ranging from 1.6 to 3.7 (Table 3), the kiln atmosphere was very dusty, and the cement produced showed abnormal properties. Cement from kiln-2 and 3 was of normal quality, while that of the kiln-3 was the best product. This provides a good relationship between "C" index and clinker quality.

Table 3:'C' indices of cement clinkers from different sources

 

No. of samples analyzed

'C' index

Range

Average

Kiln-1

 

 

 

Early stage-disturbed

15

1.6-3.7

2.5

Later stage-smooth running

35

1.2-2.4

1.5

Kiln -2

20

1.1-1.7

1.4

Kiln-3

25

1.0-2.1

1.2

2.6 Evaluation of Clinker (potential vis-a-vis actual cement phases)
Mineralogy of the cement clinker determines cement properties. It is well known that very often cement clinkers having almost identical chemical composition produce cement of varying properties. This generally happens because of changes in mineralogy due to the presence (absence) of some minor constituents, and mineralizers, a trace of which can alter the mineralogy of the clinker, which was not detected by the widely known "Bogue" calculation, based on major oxide constituents. It is generally observed that clinker phases determined by XRD and Optical microscopy were different in comparison to Bogue calculations, indicating that the quality of clinker entirely depends on the phase composition, and could be determined more precisely only by quantitative XRD. Although, XRD and microscopic analysis produce almost similar values, the former is having number of advantages over the later [Goswami et.al, 1992]

During the use of petcoke as a prime fuel in cement plants, along with the main mineral phases of the clinker like C3S, C2S, C3A, C4AF, various other associated minerals are characterized by XRD like anhydrite (CaSO4), aphthitalite (3K2SO4. Na2SO4), arcanite (K2SO4), calcium langbeinite (K2SO4 2CaSO4) and thenardite (Na2SO4). Overall sulphur content of the clinker increased in proportion with the amount of sulphur in the fuel. Reactivity of clinker also depends on the nature of sulphate bearing phase. The higher sulphur present in the clinker causes fine clinkerization, alite and belite grain size & changing the reactivity.

Higher amount of alkalis (Na2O & K2O), coming from the use of alternative raw materials and fuel during the production of Portland clinker, led to the formation of orthorhombic polymorphic form of C3A instead of cubic (detected by XRD), leading to the different reactivity and hydration characteristics. C3Aorthorhombic is more reactive results in shortening in setting time, problem in reactivity and difficulty in controlling rheology. C3Acubic/ C3A orthorhombic influences the water consumption of the cement and higher water consumption by C3A orthorhombic.

Microscopic evaluation of Portland clinker is another way to investigate quality of clinker, and cement properties by analyzing the morphology (mineralogy as well as granulometry) of clinker and interpreting the OM images as given below in Table 4. Images of major grains observed in Optical microscopy are shown in Fig 4. Optical micrographs of some clinker samples are given in Fig 5.

Table 4:Microscopic images and their influence

Observations in OM images

Interpretations

Increase in alite size and content

Relative high LSF, long sintering zone, coarsening of feed

Inhomogeneity of clinker phases

Lower reactivity of ash, short retention time, heterogeneity of raw mix

Highly porous clinker

Sandy raw meal

Cannibalistic or fused alite

Reduced reactivity of clinker

Inclusion of belite in alite grains

Reducing conditions

Large variations in alite size

Inhomogeneous raw mix

Small alite grains

Short flame, fast heating rate, shorter burning zone

Decomposition of alite

Slow cooling, reducing conditions

Large proportions of belite

Low LSF

Belite nest

Ash shortage, excessive quartz grain size

Large number of belite clusters

Decreased grindability

Large belite

Overheating or overburning

Coarsening of belite, aluminate, ferrite

Slowly cooled clinker

Ragged margins of belite

Also due to very slow cooling

Rounded belite without lamellae

Rapid cooled clinker

Secondary belite on alite grains

Slow cooling

Irregular belite

Over burning

Coarsely crystalline liquid phase

Slow cooliong

Therefore, quantitative estimation of clinker phases through microscopy and XRD is very important in controlling and monitoring the quality of raw material as well as clinker. Monitoring and identifying mineral and morphological features can effectively control the process conditions.

2.7 Cement Grinding
It is generally found that cements prepared incorporating same clinker and gypsum ground to the same fineness, but in different mills show remarkable variation in their properties. This ambiguity in the cement properties can be well explained by analyzing the XRD spectra of two cement samples ground in different grinding system. The XRD spectra of cement samples from different mills show that dehydration of gypsum (CaSO4.2H2O) - formation of hemi-hydrate (CaSO4.0.5H2O) and its crystallinity during grinding vary from mill to mill. Dehydration of gypsum during grinding plays a significant role in deciding final cement properties as gypsum was converted to hemihydrate in plant mill. Setting time of the cement significantly affected by the ratio of hemihydrate to gypsum. Complete conversion to hemihydrate leads to false set, whereas lesser amount of gypsum accounts for flash set and low early strength. Dehydration which leads to loss of crystallinity of gypsum is a function of grinding temperature. Even for identical fineness, temperature generated in different mills may vary and accordingly affect the crystallinity of gypsum. As X-ray intensity is an index of degree of crystallization, XRD of the cement can be used to examine the loss of crystallinity or modification of gypsum. Fig 6 illustrates the effect of grinding time on the transformation of gypsum and with increase of grinding time, the intensity of gypsum peak gradually decreased and diminished at 90 minutes. The effect of grinding time on the cement properties indicated that loss of crystallinity of gypsum results in its less effectiveness in controlling the setting trend as well as compressive strength development (Table 5).

Fortunately, it does not take much time to identify the Ca-sulphate phases by XRD. Once the CaSO4 phase identified, grinding process (mill), time and temperature can be set accordingly to get desired quality of the cement produced [Goswami et.al, 1984; Goswami et.al, 1990c; Goswami et.al 1997; Panigrahy et.al, 2003]

Table 5:Effect of grinding time on polymorphism of gypsum

Gypsum

Grinding time in minutes

20

30

40

90

Cement

Blaine's fineness (cm2/gm)

2932

2925

3000

2988

Setting time (minutes)

Initial

175 (38)

270 (59)

325 (71)

456 (100)

Final

263 (51)

334 (65)

400 (78)

516 (100)

Compressive strength (MPa)

3-days

307 (113)

290 (107)

280 (103)

272 (100)

7-days

350 (99)

325 (92)

347 (99)

352 (100)

28-days

505(116)

438 (100)

457 (104)

437 (100)

 (Figures within parenthesis are in percentage of that of cementground for 90 minutes)

Autoclave expansion is another property of cement that partly depends on the grinding process. More often, it is experienced that cement ground in the laboratory mill shows higher autoclave expansion than the cement produced in the plant mill, even when both of them are having equal amount of gypsum and ground to the same specific surface area. In thermal analysis of such a cement, it is observed that Mg(OH)2 grains developed due to the hydration of periclase of different grain sizes even in the same cement sample get dissociated at different temperatures and thus produce more than one endothermic peak in DTA curve. In another study [Goswami et.al, 1984], it was found that periclase (MgO) in hydration of narrower particle size range produces higher autoclave expansion than the equal amount of periclase of wider size range, although the latter may contain some particles coarser than those in the former. Therefore, variation in grain size of Mg(OH)2 can be easily detected by DTA, the same can be used to ascertain the nature of grinding

2.8 Prediction of cement strength
Several attempts have been made to predict cement strength from the examination of the corresponding clinker and thus to control the clinkering process. The reported studies were generally based on microstructures and optical properties of the clinker phases, phase composition and chemistry like alkali contents, fineness of cement, micro hardness of clinker phases and hydrated and autoclaved cement.

3,0 Concluding Remarks
With the advent of the changes in the cement pyro-processing like use of multi-channel burners, low NOx calciner, use of petcoke, alternate fuels and raw materials during clinkerization and use of various types of gypsum in different comminution techniques like Ball Mill, VRM, Roller Press, there is a requirement of online control of mineralogical investigation by XRD, microscopy , DTA/TGA for process and product quality optimization.

Presently almost all the major cement manufacturers use sophisticated analytical instruments like X-ray fluorescence (XRF) for the elemental analysis, very rarely they possess any instrument for the phased analysis. However, now it is widely recognized that in order to make cement process control system more effective with better product quality control, it is not the chemistry alone, but the phase assemblage, morphology and thermal behavior of materials are equally important. Although above analytical techniques, such as XRD, microscopy, DTA/TGA have their own limitations, but each of them is to be supplemented by the other two methods for proper and adequate evaluation of cement raw materials, process and quality optimization.

References
1.Goswami G, Mohanty S K, Panda J D (1984), "Difference in autoclave expansions of plant and laboratory ground samples", Cem.Conc.Res., 14, p.407-12.
2.Goswami G and Panda J D (1985), "Application of microscopy, XRD and DTA in study of cement raw mix burnability", 7th International Congress on Cement Microsvcopy, Texas, USA, p.81-96.
3.Goswami G and Panda J D (1986), "A study of the effect of clinker size on cement properties", 8th International Conference on Cement Microscopy, Orlands, USA, p.184-96.
4.Goswami G (1987), Indian Cement Review, p.25.
5.Goswami G, Panda J D, Chatterjee S (1990a), "A study of homogeneity of cement raw mixes processed in two types of mills", World Cement, 21(II), p.489-92.
6.Goswami G, Mohapatra B N, Chatterjee S (1990b), "Phase formation during sintering of a magnesium and fluorine containing raw mix in a cement rotary kiln", ZKG, 43(5), p.253-56.
7.Goswami G, Mohapatra B N, Panda J D (1990c), "Gypsum dehydration during comminution and its effect on cement properties", J.American.Ceramic Society., 73(3), p.721-23.
8.Goswami G, Mohapatra BN, Panda J D (1991), ôCharacterization of burning condition of cement clinkers by X-ray diffractomeryö, Cem.Concr.Res., 21, p.1176-79
9.Goswami G, Mohapatra B N, Panda J D (1992), "Effect of fluorine bearing limestone on clinker quality", 9th International Congress on the Chemistry of Cement, New Delhi, p.II-365-71.
10.Goswami G, Mohapatra B N, Panigrahy P K, Panda J D (1997), "Application of X-ray diffractometry in comminution of Gypsum", 10th International Congress on the Chemistry of Cement, Sweden. 11.Goswami G (2014), "Characterization of cement and cement raw materials: Application of thermal analysis", Industrial Angles, November 2014, p. 74-80. 12.Goswami G, Mohapatra B N, Panda J D (2016), "Characterization of burning condition of cement clinker by X-ray diffractometry", Industrial Angles, February 2016, p. 72-75.
13.Lea F M (1970),"The Chemistry of Cement and Concrete", Edward Arnold Ltd., London, p.692.
14.Panigrahy P K, Goswami G, Panda J D, Panda R K (2003), "Differential comminution of gypsum in cements ground in different mills", Cem.Conc.Res., 32, p.945-97.
15.Rao P B, Vishwanathan V N, Raina S J, Arora, V K, Chatterjee A K (1977), Cement 2-1, 5.
16.Webb T L and Stem (1957), "The Differential Thermal Investigation of Clay (RC Mackenzie ed), Mineralogical Society, London, p.329.

Related Stories

Indian Cement Review
indiancementreview.com