Powder Technology 286 (2015) 610–615

Contents lists available at ScienceDirect

Powder Technology

journal homepage: www.elsevier.com/locate/powtec

Utilization of grinding aids in dry horizontal stirred milling

Okay Altun a,⁎, Hakan Benzer a, Alper Toprak a, Udo Enderle b
a
Hacettepe University, Mining Engineering Department, 06800 Beytepe, Ankara, Turkey
b
Netzsch Feinmahltechnik, Sedanstraße 70, 95100 Selb, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Within the context of the study 3 types of grinding aids, which were glycol-based, TEA-based and TIPA-based,
Received 27 June 2015 were tested in dry stirred milling of cement. The objective was to evaluate their influences on mill performance
Received in revised form 30 August 2015 and cement quality then find out the most proper type with its dosage rate. The results showed that use of grind-
Accepted 1 September 2015
ing aid improved the grinding performance of the mill as finer size distribution than that of no grinding aid was
Available online 4 September 2015
obtained at the same milling condition. The tests at the same dosage rate indicated that, TIPA-based grinding aid
Keywords:
had the lowest specific energy utilization with 16.8 kWh/t compared to that of glycol (17.9 kWh/t ) and TEA-
Stirred mill based aids (19.75 kWh/t) respectively. The product size distributions showed that TIPA-based aid had finer prod-
Dry grinding uct with d50 of 10.54 μm compared to other types. Cement quality tests, which were assessed by measuring the
Fine grinding compressive strength, implied that TIPA had the highest improvement in 28-days strength with 18.8% at 700 g/t
Grinding aid where 8.9% and 5.6% increase were achieved for glycol-based and TEA-based grinding aids respectively. As a con-
Cement clusion of the study, TIPA based grinding aid was found to be the most effective one for dry stirred mill application
of cement due to its provided operational benefits and quality improvements.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction grinding area at different operating conditions and design features

[4,5]. Moreover, tests were performed at different types and dosage
Dry energy intensive processes, such as cement manufacturing, rates of grinding aids, which was investigated within the scope of
demand efficient grinding technologies to reduce the energy costs this paper. Grinding aids have been used in cement grinding applica-
where inefficient ball milling technology is still in use. Development of tions for many years and literature reports energy saving operations
stirred media mills improved the energy utilization in fine grinding when they are utilized [6,7]. They are mostly organic compounds and
area when compared to conventional ball milling. There have been commonly consist of glycols, alkanolamines and phenols [8].
studies compared the energy utilizations of both technologies, which The high polarity in their chemical functioning groups of –OH, –NH2,
reported that stirred milling is more energy efficient [1,2]. This can be –COOR, –SO3–, causes the tendency to adsorb on electrostatic surfaces
attributed to having higher media filling ratio, agitating the fine media from fractured covalent bonds of Ca-O, Al-O, and Si-O, and to resist
at higher tip speeds hence having higher power utilization per unit the agglomeration [9]. The selection of the grinding aids depends on
volume [3]. several parameters e.g., the benefits on overall specific energy con-
When cement grinding operation is considered, there is a ball mill sumption, cement strength, surface area development etc. and Toprak
and air separator operated in closed circuit to achieve required fineness. et al. [6], in their studies compared several types of grinding aids and
Within the circuit, high amount of material is circulated back to the mill concluded that each of them has varied influences on the process and
owing to its coarseness. This coarse product can be ground with an effi- product quality.
cient milling technology, such as stirred milling, hence the production Up to date, the grinding aids have been used in stirred milling appli-
rate of the circuit could be increased. Some other applications could be cations as well. Zheng et al. [10] tested various types of them in lime-
the use of the stirred mill on electro filter product and final product stone grinding and concluded that the surface area of the product
streams with the aim of improving the quality (ultimate strength) of could be doubled with the use of proper grinding aid. Choi et al. [11]
the finished product as well as energy efficiency [4]. Pilot scale dry hor- in their stirred mill test works showed that it was possible to reduce
izontal stirred mill was developed for this purpose with the partnership the utilized energy by 37% with the use of grinding aids. Within the con-
of Netzsch-Feinmahltechnik GmbH. The mill was tested in cement text of the study, 3 types of grinding aids (glycol, TEA and TIPA) were
tested and the influences on mill performance and product quality
⁎ Corresponding author. were discussed by considering the differences in the agglomeration en-
E-mail address: [email protected] (O. Altun). ergies and adsorption mechanisms reported in the literature. The study

http://dx.doi.org/10.1016/j.powtec.2015.09.001
0032-5910/© 2015 Elsevier B.V. All rights reserved.
 O. Altun et al. / Powder Technology 286 (2015) 610–615 611

Fig. 2. Discharge end of the mill with the product separator.

Table 2
The milling conditions of the initial test studies.

Media Type steel

Media size (mm) 4

Stirrer design disc
Stirrer speed (m/s) 5.42
Media filling (%) 60
Feed rate (kg/h) 400

Fig. 1. Dry horizontal stirred mill (1-feed hopper, 2-control panel, 3-grinding cham-
ber, 4-product outlet).
in the hopper before being ground. The tests were performed at differ-
ent dosage rates, which were 500–700–1000 g/t, and at the same mill-
is thought to be beneficial for the development of proper grinding aids
ing conditions given in Table 2, so that the variations coming from the
particularly for dry horizontal stirred mill employed in cement grinding.
operational changes were minimized.
This study assumed that the mill feed and product samples had the
2. Materials and methods
same amount of grinding aids that means the given dosage rate was
fully adsorbed by the material. Therefore, no measurements were
2.1. Description of the mill setup
undertaken on how much of the grinding aids were absorbed by the
product.
The photograph of the mill used in this study is depicted in Fig. 1
where the key components are also illustrated such as control panel,
feed unit (feed hopper and rotary valve), grinding chamber and product
2.3. Material characterization
discharge. In addition, Table 1 gives the technical data of the mill.
The rotary valve mounted under the feed hopper adjusts the mill
The feed sample together with the test products were characterized
feed rate, which is measured from the product outlet by cutting the ma-
regarding to their size distribution curves, specific surface area and ce-
terial flow. A torque sensor installed measures the power draw of the
ment properties. In this context, Sympatec laser sizer (Germany) having
mill, which displays on the control panel instantly. In grinding operation
size measurement range between 1.8 μm and 500 μm was utilized in dry
of the mill, air is supplied from the feed inlet in order to improve mate-
mode then the whole distribution from top size down to 1.8 μm was de-
rial transportation towards the discharge end. The ground particles
termined. The size distribution curves were used to determine the slope
leave the mill from the product outlet where a product separator exists
and the mean size parameters. The slope, which was expressed by n pa-
with the aim of retaining the media inside. Product separator is a cage
rameter in RRBS equation, is an important parameter in cement
having openings smaller than the bead size and attached to the dis-
strength development [14] and within the study the influences of
charge end of the mill directly as illustrated in Fig. 2.
stirred milling were discussed. Specific surface area measurements
The temperature is important parameter that should be controlled in
were undertaken via Blaine method [15]. Moreover, cement strength
cement manufacturing. Therefore, the measurements were undertaken
at different ages (7 days and 28 days) and water demands of cement
via infrared or non-contact thermometer from the outside of the mill
mortar were determined at the cement plant by applying the standard
discharge end. Such information was used to identify if the product of
procedures [16].
the stirred mill was in acceptable temperature limits [12].

2.2. Grinding aids

Although various types of grinding aids are utilized by the industry

[6,13] within the scope of the study, 3 types of them were tested for
dry stirred milling which were; TEA, TIPA and Glycol-based ones. The
grinding aids tested were in liquid form and mixed with feed material

Table 1
Technical specifications of the mill.

Effective diameter (cm) 26.4

Effective length (cm) 75.0

Motor power (kW) 18
Maximum feed rate (kg/h) 500
Stirrer tip speed (m/s) 1.08 - 9.76
Maximum air flow rate (L/h) 1000
Fig. 3. Feed material characteristics.
612 O. Altun et al. / Powder Technology 286 (2015) 610–615

Table 3
Chemical assay of the cement processed in the mill.

Element %

CaO 63.27
SiO2 19.68
Al2O3 4.94
Fe2O3 3.38
MgO 2.03
SO3 3.07
K2O 0.77
Na2O 0.52
Loss of ignition 2.32

Table 4 Fig. 5. The size distributions obtained from glycol-based grinding aid tests.
Change in process data with chemical dosage.

Chemical Specific energy Material amount inside Mean product size, As can be followed from Table 4, the product of the mill gets finer in
dosage (g/t) (kWh/t) the mill (kg) d50 (μm) case of using grinding aids. This can be attributed to the fluidization of
0 21.15 17.1 14.64 bulk material inside the mill that lead the particles already in desired
500 20.18 12.2 14.07 fineness to be transported through the discharge. It can also be conclud-
700 19.55 11.2 13.71 ed that the change in product mean size is dosage dependent. Similar
1000 19.31 11.3 13.65
conclusion was drawn by Hasegawa et al. [18] who noted an increase
in specific surface area of the product with the amount of additives
used. Another observation is the decrease in mill load. There is a consid-
The sample used in grinding tests was collected from the finished erable difference between grinding aid and no aid conditions, which
product stream of a cement grinding circuit when producing CEM I provided a reduction in specific energy consumption by 7.5% from 0 to
42.5R type cement. Fig. 3 illustrates the feed characteristics in terms of 700 g/t. The results also proved that further increase in dosage rate
the cumulative particle size distribution curve by mass, temperature (1000 g/t) is not beneficial in terms of performance improvement. The
(°C), mean size (d50,3), slope, cement strength and water% parameters. improvement in mill transportation can also be expressed by calculating
Additionally, Table 3 gives the chemical assay of the cement processed the retention time of the material. Since the tests were performed at the
in the mill. same feed rate (400 kg/h), retention time of the material reduced from
154 seconds (0 g/t) to 101 seconds (700 g/t), showing the mill condi-
tions were improved. Nair and Paramasivam [19] also observed increase
3. Results and discussions in the flowability of the material in their batch ball milling studies when
grinding aid was utilized.
3.1. Measuring the effects of grinding aids on material transportation

As reported in the literature, introducing grinding aids into the mill- 3.2. Comparison of the grinding aid performances
ing equipment makes the transportation easier since the material
becomes fluidized owing to neutralizing the surface charges of the par- 3.2.1. The tests with glycol-based grinding aid
ticles [6,17]. This study aimed to reveal the influences of grinding aids Fig. 4 illustrates the trends of specific energy, mean size, Blaine num-
on the mill inside and discharge by conducting systematic test studies bers while Fig. 5 depicts the size distribution curves obtained from the
when TIPA-based chemical was used. In this context, 4 tests at changing tests with glycol-based grinding aid. As can be understood from Fig. 4,
dosage rates were carried out initially then the product samples were the condition without grinding aid (no aid) is the least efficient one
collected when the steady state conditions were established. In addi- among the tests since higher specific energy is consumed and relatively
tion, the mill was crashed stopped for each of the condition and cham- coarser distribution is obtained. The tests also showed that dosage rates
ber was removed to weigh the material amount inside. The test results of 500 g/t and 700 g/t had similar performance data i.e., energy con-
are summarized in Table 4. sumption, product mean size and specific surface area. Both conditions
provided reduction in specific energy consumption by 14.5% with re-
spect to no aid condition, which had 20.96 kWh/t. Further increase in
dosage rate (1000 g/t) contributed additional energy saving with 3.3%
(700 g/t-17.9 kWh/t, 1000 g/t-17.3 kWh/t). The size distributions illus-
trated in Fig. 5 imply that, no aid condition has coarser and wider distri-
bution compared to other tests. Additionally it is seen that, changing the
dosage rates has no considerable influence on the mean size and slope

Table 5
Cement properties of glycol-based grinding aid test products.

7 days strength 28 days strength Water demand %

(MPa) (MPa)

Feed 40.8 49.6 28.2

No aid 41.1 51.2 27.2
500 g/t 41.4 52.1 29.2
700 g/t 41.8 52.3 29.2
1000 g/t 43.1 54.0 29.4
Fig. 4. Grinding results of glycol-based grinding aid tests.
 O. Altun et al. / Powder Technology 286 (2015) 610–615 613

Fig. 6. Grinding results of TEA-based grinding aid tests. Fig. 8. Grinding results of TIPA-based grinding aid tests.

illustrating the process performance implies that neither the product

fineness nor the specific energy consumption parameters dependent
on the variation in grinding aid dosage since each test was performed
at 19.7 kWh/t. However they are all more efficient than no aid condition
where the specific energy utilization was 20.96 kWh/t. The particle size
distributions illustrated in Fig. 7 show that the slope and the mean sizes
of grinding aid tests are the same and different from no aid condition
where the distribution was coarser and wider. Finally, the temperature
measured from outside of the discharge end section was 93 °C, which
was similar to that of glycol-based grinding aid test.
The cement properties given in Table 6 indicate that the variation in
dosage rate has no influence on both the strength and water demand
properties. However, all the results are higher than no aid condition.
The maximum 28-days strength was obtained at 1000 g/t, which
corresponded to 5.6% increase compared to that of feed material.

3.2.3. The tests with triisopropanolamine (TIPA)-based grinding aid

Fig. 7. The size distributions obtained from TEA-based grinding aid tests. Figs. 8 and 9 show the variations in process trends of TIPA aid with
the dosage rate. It is understood from the results that no aid condition
of the product size distributions. Throughout the test studies, the tem- has higher specific energy utilization and it decreases substantially up
perature measured from outside of the discharge end section was to 1000 g/t. That is, from 500 g/t to 700 g/t, energy consumption is re-
94 °C that was well below 120 °C where dehydration of gypsum occurs. duced by 4.5% (from 17.6 to 16.8 kWh/t) however further increase
From cement properties point of view, the variations in compressive (1000 g/t) has no contribution. The particle size distributions imply
strength and water demand were investigated. The results given in that, the mean sizes together with the slope of the curves are the
Table 5 imply that, grinding operation has an influence on cement prop- same for all of the test products and parallel to each other except the
erties and additionally, it is dependent on the grinding aid and its ad- no aid case where the curve is wider and coarser. Moreover, the temper-
justed dosage rate. It also indicates that the highest increase in ature was 92 °C that was similar to the previous measurements.
28 days strength was observed at the dosage rate of 1000 g/t where it Regarding to the cement properties, the strength and water demand
was improved by 8.87% with respect to that of feed (non-ground). measurements were undertaken. As can be seen from Table 7, the ce-
When the water demands are compared, it is found that the grinding ment strength at all ages increased with respect to the feed material
aids have higher per cent values than feed and no aid test. Having less as a result of grinding process. The results indicated that the highest in-
n value (slope) in size distribution curve (Fig. 5) compared to other crease in 28 days was observed at 700 g/t with 18.8% where the dosage
products is thought to be effective on obtaining decreased water
demand [14].

3.2.2. The tests with triethanolamine (TEA)-based grinding aid

While testing the TEA-based grinding aid the same methodology as
the glycol-based one was applied. Initially, the influences on the process
performance and product size distribution curves were evaluated. Fig. 6

Table 6
Cement properties of TEA-based grinding aid test products.

7 days strength 28 days strength Water demand %

(MPa) (MPa)

Feed 40.8 49.6 28.2

No aid 41.1 51.2 27.2
500 g/t 43.5 52.2 28.4
700 g/t 43.6 52.1 28.4
1000 g/t 44.8 52.4 28.4
Fig. 9. The size distributions obtained from TIPA-based grinding aid tests.
614 O. Altun et al. / Powder Technology 286 (2015) 610–615

Table 7
Cement properties of TIPA-based grinding aid test products.

7 days strength 28 days strength Water demand %

(MPa) (MPa)

Feed 40.8 49.6 28.2

No aid 41.1 51.2 27.2
500 g/t 44.2 55.4 28.2
700 g/t 46.3 58.9 28.2
1000 g/t 46.6 58.5 28.6

rates of 500 g/t and 1000 g/t increased the strength by 11.69% and
17.94% respectively. The improved results of the grinding aids com-
pared to no aid condition proved that the use of TIPA-based aid is ben-
eficial on quality improvement of cement.
Fig. 11. The change in 28-days strength development of cement with grinding aids.
3.2.4. Comparison in terms of process performance
As presented in the previous chapters, the specific energy utilization
of the mill, the specific surface area and the mean size parameters were
considered while evaluating the influences of the grinding aids. Within TEA increased the grindability of clinker by 26% and 14% respectively. In
the chapter the test results are illustrated in Fig. 10 comparatively. other words, using TIPA based aid improves the energy efficiency of a
Fig. 10 summarizes that, TIPA-based grinding aid has more pro- mill.
nounced effect on mill performance among the grinding aids tested. In
terms of product mean size, its operation has the finer product with
d50 of 10.54 μm (Fig. 10a, at 1000 g/t). Regarding to the specific energy 3.2.5. Comparison in terms of product quality
consumption, its operation utilized 6.4% and 15% less energy than Fig. 11 illustrates the influences of grinding aid types on ultimate
glycol-based and TEA-based grinding aids at 700 g/t respectively strength development (28-days) of cement. The strength results were
(Fig. 10b). The results presented in Fig. 10 prove that using TIPA based compared with reference or feed sample. It is seen from the figure
aid resulted in obtaining less specific energy consumption with finer that, the tests with grinding aids have higher strength values than no
product size compared to glycol based and TEA based aids. The differ- aid condition. This can be attributed to difference in product size distri-
ence in energy utilizations between the grinding aids can be attributed butions since no aid sample is coarser than the test products.
to change in dispersion of the powder inside the mill that is a function of It is also understood from the figure that, TIPA-based grinding aid
the fluidity of the cement powder. As the particles are dispersed, the en- has the highest development with 17.94% at 1000 g/t and 18.75% at
ergy is transferred more efficiently thus the performance is improved. 700 g/t among the grinding aids tested. On the other hand, it was
Sverak et al. [13] in their laboratory scale test studies with Portland ce- observed that TEA-based chemical has the least influence on strength
ment clinker concluded that the mill had different grinding results de- development. Within the stirred mill application, TIPA-based aid was
pending on the type of grinding aid used. Jolicoeur et al. [20] in their found to be more beneficial compared to TEA and glycol-based ones.
studies tested 5 different grinding aids and found out the fluidity The influences of different types of grinding aids on strength develop-
index with their sieving technique. They concluded that, the calculated ment have been reported in the literature [6,22,23]. Toprak et al. [3] in
fluidity index of glycol type aid was two times higher than TEA that sup- their studies on cement grinding circuits indicated that the 28-days
ports the conclusions drawn within the study. In the literature, several strength of cement could be improved by 2% to 4% with the use of grind-
studies regarding to TEA, TIPA comparison can be found that supports ing aids. In particular, TEA-based grinding aid was found to be the most
the outcomes of the dry stirred milling study. Mishra et al. [21] studied beneficial one when the economy of the given circuit was considered.
on adsorption mechanisms of grinding aids on clinker surface. In their Gartner and Myers [23] studied the effects of TEA and TIPA on hydration
studies, agglomeration energies of C3S phase, on which different types of cement. They concluded that TIPA grinding aid had more pronounced
of grinding aids were adsorbed, were calculated. They concluded that effect on early strength development owing to increased hydration of
C3S surface without aid had the highest agglomeration energy and C4AF. The mechanism was proposed to be that TIPA remained in solu-
then TEA and TIPA came respectively. That is, TIPA aid disperses the par- tion and formed iron complexes that affected the strength. Katsioti
ticles more effectively hence improved grinding performance is obtain- et al. [22] in their studies compared the effects of TEA and TIPA on
ed. Finally, Katsioti et al. [22] compared the effects of TIPA and TEA type strength development and showed that TIPA improved the strength
grinding aids on grindability of clinker. The study showed that, TIPA and significantly.

Fig. 10. The variation of mill performance with grinding aids.

 O. Altun et al. / Powder Technology 286 (2015) 610–615 615

4. Conclusions [3] M. Gao, M. Young, P. Allum, IsaMill fine grinding technology and its industrial appli-
cations at Mt. Isa Mines, Proceedings of the 34th Annual Meeting of the Canadian
Mineral Processors, Mingwei Gao, Ottawa, Canada 2002, pp. 171–188.
Within the scope of the study different types of grinding aids, glycol- [4] O. Altun, Investigation of dry horizontal stirred milling applications for cement
based, TEA-based and TIPA based, were tested then the influences on grinding circuits(Ph. D Thesis) Hacettepe University, Ankara, Turkey, 2013.
[5] O. Altun, H. Benzer, U. Enderle, The effects of chamber diameter and stirrer design
dry stirred mill performance and cement properties were discussed. on dry horizontal stirred mill performance, Miner. Eng. 69 (2014) 24–28.
The test studies implied that the use of grinding aids lead to having [6] N.A. Toprak, O. Altun, N. Aydogan, H. Benzer, The influences and selection of grind-
finer product size distribution and decreased specific energy consump- ing chemicals in cement grinding circuits, Constr. Build. Mater. 68 (15) (2014)
199–205.
tion when compared to no aid condition. Among the grinding aids test- [7] F.C. Lai, M.R. Karim, M. Jamil, M.F.M. Zain, Production yield, fineness and strength of
ed, TIPA-based grinding aid was the most effective one at all the dosage cement as influenced by strength enhancing additives, Aust. J. Basic Appl. Sci. 7 (4)
rates. The highest difference was observed at 700 g/t, where the mill uti- (2013) 253–259.
[8] C. Engelsen, Quality improvers in cement making-State of the art, Building and in-
lized 16.8 kWh/t of specific energy, which was 6.4% and 15% less than
frastructure coin project report 22008.
glycol-based and TEA-based grinding aids used at the same dosage [9] A.A. Jeknavorian, E.F. Bary, F. Serafin, Determination of grinding aids in Portland
rate. In addition to influences on the specific energy consumption of cement by pyrolysis gas chromatography-mass spectrometry, Cem. Concr. Res. 28
the mill, TIPA-based grinding aid produced finer product with d50 of (9) (1998) 1335–1345.
[10] J. Zheng, C.C. Harris, P. Somasundaran, The effect of additives on stirred media mill-
10.58 μm. The variations in the performances of grinding aids can be at- ing of limestone, Powder Technol. 91 (3) (1997) 173–179.
tributed to the change in the fluidity of the powder since different grind- [11] H. Choi, W. Lee, D.U. Kim, S. Kumar, S.S. Kim, H.S. Chung, J.H. Kim, Y.C. Ahn, Effect of
ing aids have different influences as reported in the literature. Finally, the grinding aids on the grinding energy consumed during grinding of calcite in a
stirred ball mill, Miner. Eng. 23 (1) (2010) 54–57.
the cement strength measurements were undertaken for the test prod- [12] W.H. Duda, Cement-data-book, International process engineering in the cement in-
ucts. It was understood from the results that TIPA-based grinding aid dustry, 1985 (444 pages).
had the highest development in 28-day strength among the aids tested. [13] T.S. Sverak, C.G.J. Baker, O. Kozdas, Efficiency of grinding stabilizers in cement clin-
ker processing, Miner. Eng. (43-44) (2013) 52–57.
At 1000 g/t, the strength increased by 17.94% and at 700 g/t it increased [14] K. Kuhlmann, H.G. Ellerbrock, S. Sprung, Particle size distribution and properties of
by 18.75% with respect to reference or feed sample. The assessments cement Part I: Strength of Portland cement, ZKG Int. 38 (4) (1985) 169–178.
concluded that TIPA-based grinding aid was the most suitable type for [15] T. Allen, Powder sampling and particle size determination first ed, Elsevier Science,
2003 (682 pp.).
the dry horizontal stirred mill operation as higher mill performance [16] Turkish standard, S EN, 196-1: Methods of testing cement20022002.
and product quality are obtained. [17] R.R. Klimpel, W. Manfroy, Chemical grinding aids for increasing throughput in the
wet grinding of ores, Ind. Eng. Chem. Process. Des. Dev. 17 (4) (1978) 518–523.
[18] M. Hasegawa, M. Kimata, M. Shimane, T. Shoji, The effect of liquid additives on dry
Acknowledgements
ultrafine grinding of quartz, Powder Technol. 114 (2001) 145–151.
[19] P.B.R. Nair, R. Paramasivam, Effect of grinding aids on the time-flow characteristics
Authors appreciate contributions of Assoc. Prof. Abdullah Obut of the ground product from a batch ball mill, Powder Technol. 101 (1999) 31–42.
(Hacettepe University) for his assistance in writing, SET Italcementi [20] C. Jolicoeur, S. Morasse, J. Sharman, A. Tagnit-Hamou, F. Slim, M. Page, Polyol-type
compounds as clinker grinding aids: influence on powder fluidity and on cement
Ankara Cement Plant for the assistance in laboratory studies and hydration, 12th International Congress on the Chemistry of Cement 2007.
Hacettepe University Research Foundation Unit (Project No: 013 T06 [21] R.K. Mishra, R.J. Flatt, H. Heinz, Force field for tricalcium silicate and insight into
604 006) for the financial support. nanoscale properties: cleavage, initial hydration, and adsorption of organic mole-
cules, J. Phys. Chem. 117 (20) (2013) 10417–10432.
[22] M. Katsioti, P.E. Tsakiridis, P. Giannatos, Z. Tsibouki, J. Marinos, Characterization of
References various cement grinding aids and their impact on grindability and cement perfor-
mance, Constr. Build. Mater. 23 (5) (2009) 1954–1959.
[1] J.L. Sepulveda, A detailed study on stirred ball mill grinding(Ph.D. Thesis) Depart- [23] E. Gartner, D. Myers, Influence of tertiary alkanolamines on portland-cement hydra-
ment of metallurgy and metallurgical engineering. , The University of Utah, The tion, J. Am. Ceram. Soc. 76 (6) (1993) 1521–1530.
USA, 1981.
[2] F. Shi, R. Morrison, A. Cervelin, F. Burns, F. Musa, Comparison of energy efficiency
between ball mills and stirred mills in coarse grinding, Miner. Eng. 22 (7–8)
(2009) 673–680.