PROOFS Use of cement grinding aids to optimise clinker factor J. J. Assaad*, S. E. Asseily* and J. Harb Holderchem Building Chemicals S.A.L., Lebanon; Notre Dame University, Lebanon A comprehensive research project was undertaken to evaluate the effect of grinding aids (GAs) on the percentage of clinker decrease and reduction of energy needed during the grinding process. Three GAs, based on amines, glycols and acids, were tested at various concentrations. The tested cement properties include water demand, Blaine fineness, sieve residue, setting time and compressive strength. The results showed that the use of higher concentrations of GAs can lead to significant improvements in the performance of grinding mills characterised by higher Blaine and lower sieve residue values. Compared with the control mix, setting times were found not to be significantly affected. However, the mortar compressive strength increased by 15, 8 and 7% when using GA based on amine, glycol and acid, respectively. A decrease in both clinker factor (up to 15 . 8%) and grinding energy (up to 4 . 5%) can be simultaneously achieved through substitution of clinker with pozzolan materials with lower Mohs hardness values and the incorporation of a relatively high concentration of GA. Introduction After the aluminium and steel industries, the manu- facture of Portland cement is considered to be the most energy-intensive process, consuming around 4 GJ of energy per tonne; the process also releases large volumes of carbon dioxide into the atmosphere (Cembureau, 1998). Energy consumption during the grinding of Portland cement clinker and gypsum can be significantly reduced by adding a small quantity of a grinding aid (GA), generally in the range of 0 . 02–0 . 1% of the manufactured cement weight. The chemical basis of commonly used GAs includes ethanolamines such as triethanolamine (TEA), monoethanolamine (MEA) and triisopropanolamine (TIPA), as well as glycols such as ethylene glycol (EG) and propylene glycol (PG). Be- cause of their highly organic polar nature, GAs are preferentially adsorbed on surfaces formed by the frac- ture of electrovalent bonds (i.e. Ca–O and Si–O), thus reducing the surface energy forces that cause attraction and re-agglomeration of the newly produced cement particles (Assaad et al., 2008). In the literature, limited data are available pertaining to the direct impact of GAs on the percentage of clinker reduction and/or energy savings that may result during the grinding of clinker. This is due to the fact that most of the published data deal with the hydration processes and interaction mechanisms of the main ce- ment compounds in the presence of GAs. For example, Ramachandran (1976) studied the hydration and hard- ening characteristics of C 3 A with or without gypsum, C 3 S, C 2 S and Portland cement following the addition of various dosages of TEA varying from 0 . 1–1% of the cement weight. Ramachandran found that TEA acceler- ated the hydration of C 3 A systems and extended the induction period of C 3 S. In Portland cement pastes, TEA decreased the strength at all ages and setting characteristics were found to be drastically altered, especially at higher TEA contents. It is important to note that, compared with TEA, use of TIPA is known to yield a reduction in setting times and significant increases in strength development at early and late ages, regardless of the cement type (Perez et al., 2003; Sandberg and Doncaster, 2004). The effect of glycol-based GAs from the dihydroxy compound class (including EG, PG and polypropylene glycol (PPG)) on the grindability of clinker was studied by Teoreanu and Guslicov (1999) at dosage rates of 0 . 03, 0 . 05 and 0 . 1% of cement weight. For the same grinding duration, they concluded that the cement spe- cific surface area increased by 29% when using PPG and by 20% with the use of EG and PG. For longer grinding durations when cement Blaine fineness ex- ceeds 3500 cm 2 /g, growth of the specific surface area Advances in Cement Research, 2009, 21, No. 1, Month, 1–8 doi: 10.1680/adcr.2008.21.1.1 1 Article number = 900007 www.cement-research.com 1751-7605 (Online) 0951-7197 (Print) # 2009 Thomas Telford Ltd * Holderchem Building Chemicals S.A.L., Lebanon † Department of Civil and Environmental Engineering, Notre Dame University, Lebanon (ACR-D-09-00007) Paper received 17 February 2009; last revised 5 May 2009; accepted 26 June 2009