Process parameters are changed from time to time to reduce waste and improve efficiency. Analytical instrumentation provides the insight needed into these parameters, and can therefore hold the key to cement manufacturing process optimisation, says Dr Michael Caves.
The cement industry is one of the world's most energy-intensive sectors. Cement is a crucial global commodity, indeed, usage levels are a useful indicator of economic activity. As an energy-intensive process, cement manufacturers are exposed to increasingly volatile energy prices. We will examine three successful strategies deployed by the cement industry, and the analytical instrumentation that has made them successful.
Smarter definition of product quality
The performance of cement is defined by its composition and fineness. Fineness defines cement performance, because it influences how fast the cement hydrates. Fineness is traditionally quantified by Blaine measurement, a surface area technique, but a recognised drawback of this method is that it only provides a single averaged figure for any given sample. As a result, two samples with the same Blaine may, in fact, contain different proportions of fines, and will therefore hydrate differently and have different strength characteristics.
Consider two model samples with different particle size distributions as shown in Figure 1. Each has the same Blaine value. When mixed with water, these two samples will hydrate differently. For example, the relatively high fines population in Sample 2 will hydrate very quickly, while higher levels of coarse particles in Sample 1 will extend the time taken for complete hydration.
This is a crucial limitation when it comes to optimising manufacturing, since it suggests that Blaine is not a fully reliable detector of whether a modification will impact product performance. Laser diffraction analysis, in contrast, provides a full particle size distribution for each cement sample, rather than a single averaged figure. This enables the correlation of discrete size fractions with critical aspects of cement performance, in a way that is simply not possible with Blaine measurement. This is the primary reason why laser diffraction particle sizing has become widely applied across the cement industry.
Correlations between particle size distribution data and cement performance reveal that different size fractions influence the key parameter of developed strength in various ways. For example, research has shown that particles less than 2 microns hydrate so quickly during product use that they can cause a cement to set exothermically and crack. Coarse particles (>50 microns in size), on the other hand, may fail to hydrate at all over the period of mixing, resulting in a micro-concrete. Strong correlations have been observed between early strength and the fraction of the sample which lies in the 3 - 30 micron range Ref. Figure 2, identifying this as a crucial size range for the control of cement performance. None of this more nuanced understanding can be developed from Blaine measurements.
In the finishing circuit it is common, yet highly inefficient, practice to over-grind, deliberately milling more than is necessary because a cement with higher overall fineness typically delivers acceptable strength characteristics, while one that is too coarse will usually be out of specification. Being able to set particle size specifications based on a sound understanding of the effect of each size fraction on product performance boosts operational confidence and supports the elimination of over-grinding. In fact, by switching to particle size analysis, it is possible to produce cement with higher one-day strength, but with a lower Blaine. In practical terms, this means less milling but a better quality product, a result that delivers a major gain in terms of energy consumption.
Automated process control
Alongside better product quality definition, laser diffraction particle sizing technology has been embraced by the cement industry because it provides options for continuous process monitoring. While Blaine is a manual method ill-suited to online implementation, laser diffraction is fast, highly automated and a well-established process analysis tool. Online systems can measure continuously, in real-time, with minimal manual input, and have the proven track record of reliability needed to drive automated control in a 24/7 operation, increasingly via sophisticated multivariate control platforms.
Just like a good operator, the automated system is simultaneously considering information from a number of sources, and manipulating a number of variables in response, to continually optimise the process. In general, automated systems are far more successful and efficient than a manual approach, provided that sufficient effort is put into developing a model and tuning the resulting control loops.
A switch from manual analysis and control to full automation can deliver significant energy savings. This is in addition to benefits such as increased throughput and a reduction in ball mill charge. All of these improvements drive up the efficiency of the overall process.
Change the product composition
Cement additives such as limestone or fly ash are now used routinely, bringing the economic advantage of being waste streams from other industries. However, their use brings new complexities to cement production.
Additional components in the cement increase the challenge of particle size optimisation, since each ingredient has the potential to perform differently and therefore should ideally have a discrete particle size distribution specification. Furthermore, ingredient replacement raises questions associated with operation of the grinding circuit, such as: 'Can all the ingredients be milled together or it is necessary to mill each ingredient separately and then blend them?'
Instrumentation that combines Raman spectroscopy with automated imaging addresses this analytical requirement, enabling the technique of Morphologically-Directed Raman Spectroscopy (MDRS), which can be used to determine correlations between particle size, shape and chemical composition. For cement manufacturers, this makes it possible to investigate, for example, whether the components of a cement blend are represented equally across all size fractions. Aligning this information with product performance supports the development of precise specifications for the use of cement additives, which can then be used to establish and control processes for their inclusion.
Figure 3 illustrates the information that can be gathered using MDRS. Size data and Raman spectra were measured for a number of particles (typically in the region of 1,000). Comparing the Raman spectra with reference spectra enabled the chemical identification of different particle populations within the blend and the generation of a particle size distribution for each population.
This data suggests that the materials in this sample have been milled together, since the observed differences in particle size distribution correlate directly with differences in grindability between the three components. Clinker is a much harder material than limestone or slag, and if processed under comparable conditions would therefore be expected to exit the mill as the coarsest product. Even if the optimal particle size distribution for each component of the cement is identical, these results indicate that replacement materials cannot be milled alongside fresh clinker in the finishing circuits to produce the required optimised product.
Successfully changing a process or ingredient to reduce waste and increase efficiency relies on a detailed understanding of the parameters that influence product performance and of how to effectively control the manufacturing process. Analytical instrumentation provides the insight needed to develop the necessary knowledge and can therefore hold the key to manufacturing process optimisation. The cement industry provides a powerful example of what can be achieved by embracing new analytical strategies and state-of-the-art techniques.
About the author
Dr Michael Caves is Business Development Manager, Malvern-Aimil Instruments. He has spent over 13 years working in various academic and commercial bioscience laboratories, analysing a range of materials in various contexts.
Laser diffraction analysis, in contrast, provides a full particle size distribution for each cement sample, rather than a single averaged figure.