REPRINT OF PAPER PUBLISHED IN INTERNATIONAL CEMENT REVIEW MAY 2015

CEMENT ADDITIVESPOWERFUL GRINDING AIDS

GRINDING AIDS

INTERNATIONAL CEMENT REVIEW MAY 2015

Practical experiences with all major cement producers around the world show that stronger grinding additives

enhance cement particle dispersion. Besides higher production rates, this also results in more favourable particle size distribution (PSD) which physically accelerates strength development and allows for higher clinker replacements. Consequently, stronger grinding additives are particularly suitable for blended cements.

A three per cent increase of the production rate could mean a reduction of specific energy consumption of around 1.2kWh/t and thus save a 1Mta plant E0.07m in energy costs. If the market environment allows, the resulting 30,000tpa increased sales volume could result in an extra profit of E0.3m per year. Alternatively, this grinding potential could be used to target a higher fineness at the same production rate.1 The resulting higher strength could be used to reduce the clinker factor by two per cent, adding E0.4m of profit. Additional potential could be taken from maximising production at times with lower electric energy cost or shifting capacities to more efficient mill systems while reducing or stopping the usage of inefficient mills.

Standard vs solution providerStandard grinding additives are based on existing raw materials from mass production. The cost of purchase is compensated by energy cost savings of approximately E0.2/t. Every E0.01/t less treatment cost leads to E10,000 in annual savings for a 1Mta plant. Hence the target is to find the cheapest possible products leading to the typical 10 per cent production increase.

Stronger technologies are based on customised raw materials that need special production. Their higher cost will be compensated by the additional benefits

they deliver, eg raw material savings, reduced CO2 emissions, increased cement sales or minimised claim costs. In this case, the target is to find a technical solution under economic conditions. Consequently, a stronger grinding aid will be a solution provider.

Three-fold approachTo gather comprehensive knowledge for developing high-performance grinding aids, Sika has followed a three-fold approach: the basis has been the evaluation of the physical and chemical background of forces between cement particles. Then, extensive screening with all types of compounds has been carried out at its central cement laboratory, using its own batch grinding equipment. Finally, molecular modelling, in cooperation with the University of Akron and the Institute for Building Materials of the ETH Zürich (Swiss Federal Institute of Technology) was performed to provide insights into the atomistic interactions. More detailed information has been published by Weibel and Mishra.2

Physical and chemical backgroundCritical to the understanding of the working mechanisms of grinding

additives are the first and second laws of thermodynamics: ‘Everything tends towards a state with the lowest possible energy and the greatest possible disorder.’ These general principles of nature are also valid for all aspects of grinding. Coated mill internals represent the lowest energy level for grinding without chemical additives. The presence of a grinding aid leads to a different minimum energy state which leads to blank mill internals without coating and hence higher mill efficiency.

From a scientific perspective, a cleavage with significant charge separation is only possible in the complete absence of water and even then would be very unlikely. Surface atoms of freshly-cleaved polar surfaces reposition themselves in accordance with minimum energy and optimised charge equilibration very quickly. Consequently, surface charges – as frequently mentioned – cannot be the reason for particle agglomeration. Atoms, molecules and ions always interact with one another. Within homogeneous solids and liquids, the internal forces cancel each other out. The bulk energy is zero. At the surface, there is a resulting force directed inwards.

In liquids this force is called surface tension. It has the same value and dimension as the surface energy. The

IPowerful grinding aids by Jorg M Schrabback andDr Martin Weibel, Sika, Germany

Chemical additives play a key role in achieving the desired cement performance. They support obtaining the required fineness while also improving the energy efficiency of the grinding system. But are improvements possible? Are they technically needed and what is their economic value?

Extensive theoretical analyses combined with vast laboratory trials and molecular modelling improve the understanding of the working mechanisms of grinding aids during the cement milling process

©Copyright Tradeship Publications Ltd 2015

GRINDING AIDS

INTERNATIONAL CEMENT REVIEW MAY 2015

Practical experiences with all major cement producers around the world show that stronger grinding additives

enhance cement particle dispersion. Besides higher production rates, this also results in more favourable particle size distribution (PSD) which physically accelerates strength development and allows for higher clinker replacements. Consequently, stronger grinding additives are particularly suitable for blended cements.

A three per cent increase of the production rate could mean a reduction of specific energy consumption of around 1.2kWh/t and thus save a 1Mta plant E0.07m in energy costs. If the market environment allows, the resulting 30,000tpa increased sales volume could result in an extra profit of E0.3m per year. Alternatively, this grinding potential could be used to target a higher fineness at the same production rate.1 The resulting higher strength could be used to reduce the clinker factor by two per cent, adding E0.4m of profit. Additional potential could be taken from maximising production at times with lower electric energy cost or shifting capacities to more efficient mill systems while reducing or stopping the usage of inefficient mills.

Standard vs solution providerStandard grinding additives are based on existing raw materials from mass production. The cost of purchase is compensated by energy cost savings of approximately E0.2/t. Every E0.01/t less treatment cost leads to E10,000 in annual savings for a 1Mta plant. Hence the target is to find the cheapest possible products leading to the typical 10 per cent production increase.

Stronger technologies are based on customised raw materials that need special production. Their higher cost will be compensated by the additional benefits

they deliver, eg raw material savings, reduced CO2 emissions, increased cement sales or minimised claim costs. In this case, the target is to find a technical solution under economic conditions. Consequently, a stronger grinding aid will be a solution provider.

Three-fold approachTo gather comprehensive knowledge for developing high-performance grinding aids, Sika has followed a three-fold approach: the basis has been the evaluation of the physical and chemical background of forces between cement particles. Then, extensive screening with all types of compounds has been carried out at its central cement laboratory, using its own batch grinding equipment. Finally, molecular modelling, in cooperation with the University of Akron and the Institute for Building Materials of the ETH Zürich (Swiss Federal Institute of Technology) was performed to provide insights into the atomistic interactions. More detailed information has been published by Weibel and Mishra.2

Physical and chemical backgroundCritical to the understanding of the working mechanisms of grinding

additives are the first and second laws of thermodynamics: ‘Everything tends towards a state with the lowest possible energy and the greatest possible disorder.’ These general principles of nature are also valid for all aspects of grinding. Coated mill internals represent the lowest energy level for grinding without chemical additives. The presence of a grinding aid leads to a different minimum energy state which leads to blank mill internals without coating and hence higher mill efficiency.

From a scientific perspective, a cleavage with significant charge separation is only possible in the complete absence of water and even then would be very unlikely. Surface atoms of freshly-cleaved polar surfaces reposition themselves in accordance with minimum energy and optimised charge equilibration very quickly. Consequently, surface charges – as frequently mentioned – cannot be the reason for particle agglomeration. Atoms, molecules and ions always interact with one another. Within homogeneous solids and liquids, the internal forces cancel each other out. The bulk energy is zero. At the surface, there is a resulting force directed inwards.

In liquids this force is called surface tension. It has the same value and dimension as the surface energy. The

IPowerful grinding aids by Jorg M Schrabback andDr Martin Weibel, Sika, Germany

Chemical additives play a key role in achieving the desired cement performance. They support obtaining the required fineness while also improving the energy efficiency of the grinding system. But are improvements possible? Are they technically needed and what is their economic value?

Extensive theoretical analyses combined with vast laboratory trials and molecular modelling improve the understanding of the working mechanisms of grinding aids during the cement milling process

©Copyright Tradeship Publications Ltd 2015

GRINDING AIDS

MAY 2015 INTERNATIONAL CEMENT REVIEW

surface tension reduces the surface area and leads to the lowest-possible energy level. This phenomenon enables water striders to be carried on the water surface and causes free-falling droplets to be spherical.

In powders this surface energy is the attractive force which causes ground particles to agglomerate to reduce the surface area and energy.

The high surface polarity and energy of dry clinker is reduced to a certain level by hydroxylation. The required water is available from gypsum, wet clinker replacements and cooling water injection. This is the ‘blank’ condition with coated mill internals and low grinding efficiency. A further decrease of the surface energy and thus attractive forces between the particles is only possible with organic grinding aids. It results in clean grinding media and improved mill efficiency.

Laboratory grinding trialsTo determine the most powerful grinding additive technology, all types of organic and many inorganic compounds as well as mixtures thereof have been tested. The laboratory grinding results are repeatable and deliver clear trends. Consistent structure performance relations have been found. They can be explained by the mentioned physical and chemical effects.

Some common assumptions on the working mechanisms of grinding aids have been disproved. The findings clearly show that the strength of adsorption of compounds does not correlate with the grinding performance. Also the reduction of the mechanical toughness by surface-active compounds (Rehbinder effect) is not verifiable.

Clear evidence was found that grinding aid molecules, with

their polar and non-polar parts, compensate and reduce the polarity of the clinker surfaces to a considerable extent. In parallel with the reduction of polarity there is also a reduction of surface energy.

Molecular modelling The third step to understanding in detail why some substances lead to better grinding performance than others is molecular modelling. It provides insight into values and mechanisms which are otherwise not accessible. The complex calculations of how grinding aids interact with the clinker surface (Figure 1) are made with ‘molecular dynamics simulation’ with experimentally-validated force field parameters for each compound.

One of the important findings was that the adsorption energy – which frequently is regarded as decisive for the efficiency – could not be correlated with grinding performance.

Agglomeration energyMost important are the calculations of the agglomeration energy. This energy is released when particle surfaces come together to achieve the lowest energy level. Alternatively expressed, it is the energy which is required to separate agglomerated particles.

Figure 2 shows the values for dry C3S – the dominant part of clinker composition – in comparison with hydroxylated C3S surfaces (HC) without and with chemical compounds. Already one monolayer of the various grinding additives significantly reduces the agglomeration energy to approximately 25-50 per cent compared to hydroxylated C3S surfaces. The

agglomeration energy correlates inversely with the grinding performance of the laboratory screenings. Consequently, lower agglomeration energy leads to a better particle separation and hence increases mill output.

Agglomeration energy C3S > HC > HC-glycerine > HC-TEA > HC-TIPA > HC-MDIPA

Grinding performance Clinker < HC < HC-glycerine < HC-TEA < HC-TIPA < HC-MDIPA

Not all clinker phases behave the same. In the case of tricalcium aluminate (C3A, Figure 3), the agglomeration energy without grinding additives is approximately twice that of C3S. This confirms that C3S should be easier to grind, as it is assumed in Engelsen (2008).3 Also the performance rankings of individual chemical compounds are quite different compared to C3S (red arrows).

Customised multi-compound grinding aids (blends) ensure that the agglomeration energy of all clinker

phases is significantly reduced. Consequently, these blends can achieve a stronger grinding performance compared to pure chemical compounds.

Particle separationFigure 4 shows the computed distribution of glycerine molecules on 50 per cent-covered clinker surfaces. As a result, the particles are separated by one complete layer (100 per cent coverage) and can approach each other to a minimum distance of typically about 4Å. The necessary quantity of a chemical

Figure 1: interaction of TIPA molecules with a hydroxylated surface (C3S reacted with water molecules)

Figure 2: agglomeration energy of C3S and the influence of water (HC) and chemical compounds

Figure 3: agglomeration of energy C3A and the influence of water (HC) and chemical compounds

©Copyright Tradeship Publications Ltd 2015

GRINDING AIDS

INTERNATIONAL CEMENT REVIEW MAY 2015

compound to reach this separation is related to the specific surface area of the cement. In case of glycerine, a cement fineness of 3000cm²/g according to Blaine requires a dosage of 0.015 per cent of the active compound, while 5000cm²/g make it necessary to dose 0.025 per cent. This layer reduces the agglomeration energy to a large extent. Higher dosages do not significantly improve grinding performance.

Further improvement can only be achieved by spatial means. Crushed clinker particles are jagged. To separate them most efficiently larger polymers like PCEs (polycarboxylate ethers) are necessary. Figure 5 displays the calculated distance which is double compared with glycerine.

PCEs improve grinding in two ways. On the one hand, they reduce the agglomeration energy like traditional compounds. On the other hand, they cause a steric separation of the clinker particles due to size and configuration.

We can conclude that the combination of the physical and chemical background, empirical testing and molecular modelling provides comprehensive understanding of the working mechanisms of grinding aids. This allows the development of further improved grinding aids and customised cement additives.

Case study: Portland cementTo verify these conclusions, Sika executed and analysed plant trials which compare traditional grinding additives and blends with PCE.4, 5 The grinding system at this 1.2Mta plant (serving a sold-out market) was a traditional ball mill in a closed-circuit system with a third-generation separator. The blank production rate of CEM I 32.5 R (3200cm²/g according Blaine) was rated at 108tph. Dosing 300g/t TEA based grinding aid already achieved a good value of a 13 per cent higher production

rate. The same amount of a TEA/PCE blend performed significantly better and reached a 20 per cent production increase compared to the blank benchmark (Figure 6). This confirms the findings of the molecular modelling.

The technical improvement of a six per cent higher production compared to a traditional TEA leads to two advantages: 2.3kWh/t reduced energy consumption and 70,000t higher volume. The economic value is calculated by E0.14/t less energy cost minus E0.06/t additional treatment cost for the higher value grinding aid and accumulates to an annual saving of E0.1m. Additional sales generate a further contribution of E0.7m. In total, the technical solution of a PCE-powered grinding aid offers an extra E0.8m annual profit to the cement plant.

SummaryExtensive theoretical analyses combined with vast laboratory trials and molecular modelling have made it possible to improve the understanding of the working mechanisms of grinding aids during the cement grinding process. Energy on the clinker surface is larger than inside the particle. As a consequence, powder particles agglomerate to reach a lower energy level. The energy which is released is called agglomeration energy. The strong agglomeration energy of dry clinker is usually reduced by the moisture existing in the grinding atmosphere. The remaining agglomeration energy is still so high that it negatively affects the grinding and separating process (coating, agglomeration).

Traditional grinding aids reduce agglomeration energy to a level where most particles are dispersed and coating is reduced to a large extent. The performance of Sika‘s new grinding aid technology is boosted by specific ‘sterically acting molecules which extend the distance between the particles due to large PCE molecules. Molecular modelling confirms these results.

Stronger grinding aids offer opportunities to improve cement plant profitability, especially when producing blended cements. ________________________________I

References1 SCHRABBACK, JM (2009) ‘Finest strength development’ in: ICR, September, p75-80.2 WEIBEL, M AND MISHRA, RK (2014) ‘Comprehensive understanding of grinding aids’ in: ZKG, June, p28-39.3 ENGELSEN, CH (2008) Quality improvers in cement making – State of the art – COIN Project report no 2. Oslo, Norway: SINTEF, 24p.4 SCHRABBACK, JM (2010) ‘Concepts for ‘green’ cement’ in: ICR, April, p75-78.5 HELLER, T, MÜLLER, T AND HONERT, D (2011) ‘Cement additives based on PCE’ in: ZKG 2, p40-48.

Figure 4: side view of two C3S surfaces, each half covered with glycerine molecules in separated state (left) and smallest approximation (right)

Figure 5: two C3S surfaces separated with a PCE molecule and the resulting distance

Figure 6: increased production and reduced specific energy consumption of a ball mill grinding system for

OPC using PCE upgraded grinding aids

©Copyright Tradeship Publications Ltd 2015