High-Volume Mineral Admixtures Cement: The Effects of Particle Size Distribution

2024-04-10 10:38HUANGQiminWANGKunLUJipingYUJianpingSHENGZhenhuaYANGLu

HUANG Qimin, WANG Kun, LU Jiping, YU Jianping, SHENG Zhenhua, YANG Lu

(1.State Key Laboratory of Silicate Materials for Architectures, Wuhan University of Technology, Wuhan 430070, China; 2.School of Materials Science and Engineering , Wuhan University of Technology, Wuhan 430070, China; 3.Fujian Jinniu Cement Co., Ltd, Sanming 353300,China)

Abstract: The effects of high-volume slag-fly ash cement with different particle sizes on hydration degree, microstructure and mechanical properties were systematically studied, by means of laser particle size(DLS), X-ray diffraction (XRD), comprehensive thermal analysis (TG-DTA), scanning electron microscopy(SEM) and mechanical properties tests.The results show that suitable particle size distribution of cementitious material has significantly promoting effects on hydration reaction rate and mechanical properties.Compared with slag without further grinding, the slag after ball milling for 4 h has an obvious improvement in reactivity,which also provides a faster hydration rate and higher compressive strength for the cementitious material.When the slag milled for 1 and 4 h is mixed at a mass ratio of 2:1 (i e, slag with D50 of 7.4 µm and average size of 9.9µm, and slag with D50 value of 2.6 µm and average size of 5.3 µm), and a certain amount of fly ash is added in,the most obvious improvement of compressive strength of cement is achieved.

Key words: ultrafine slag; microstructure; compressive strength; particle size

1 Introduction

Mineral admixtures, such as fly ash and slag, have been widely used in cement production and concrete preparation due to their considerable economic benefits and promoting effects on cement performance,known as the indispensable sixth component of highperformance concrete[1-3].It is of great significance for reducing carbon emissions of building materials.

As the two most widely used admixtures in cement and concrete, slag and fly ash, with huge production and good pozzolanic activity, can be used as the supplementary cementitious materials in cementbased materials after certain processing, which can reduce clinker coefficient of cement, and improve the durability as well.Generally, the mixed amount of slag or fly ash in cement is typically maintained at about 20%-30%, which could reduce the consumption of cement clinker and relieve the pressure of carbon emissions in cement production, while also realizing the resource utilization of industrial solid waste, improving economic efficiency and promoting the sustainable development of cement industry[4-7].Previous studies[8-9]have shown that the further increase of the slag or fly ash in cement caused the reduction of strength at early age, with few effects on the improvement of durability.In order to enhance the early strength of cement-based materials with high volume mineral admixtures, some reports indicates that the use of early strength activators and mineral admixtures, such as ultrafine slag and ultrafine fly ash, can achieve a promising result.For the purpose of further decreasing the use-cost of ultrafine mineral admixtures, in this study, the effects of highvolume replacement of cement clinker by mineral admixtures with wide particle size distribution were studied based on the thought of gradation design.The influence mechanism of mineral admixtures with different gradations was investigated through analyzing their performances, such as hydration heat, components and hydration products.

2 Experimental

2.1 Raw materials

Cement: P·I 52.5 cement produced by Fujian Jinniu Cement Co, Ltd.Fly Ash: type II fly ash from Fujian, with density of 2.9 g/cm3; Slag: a commercially available S95 slag with density of 2.6 g/cm3, and the compositions are shown in Table 1; Sand: China ISO standard sand.The particle sizes of mineral admixtures are shown in Table 2, measured by Mastersizer 2000 laser particle size analyzer.

Table 1 Main constituents of cement, fly ash and slag/wt%

Table 2 Particle sizes of cement, fly ash and slag

In order to obtain slag with different particle size distribution, S95 slag was ground by ball mill for 0, 1,2, and 4 h, respectively, which were recorded as Slag A,B, C and D.Fig.1 and Table 3 show the particle sizes and specific characteristic parameters of slag after ballmilling with different time.

Fig.1 Curves of slag particle size distribution

Table 3 Particle sizes of slag after ball-milling with different time

2.2 Preparation of samples

Previous studies[10,11]have demonstrated that slag with particle size smaller than 3 μm has an enhancing effect on the early strength of cementitious materials,slag with sizes of 3-10 μm can provide better later performance.Based on this, the mix proportion used in the experiment is shown in Table 4, in which W0 refers to the cement sample without mineral admixtures,as the reference group.The percentage of mineral admixtures was set as 60 wt%.Groups A, B and Care respectively experimental groups with mineral admixtures of different particle sizes, in which the mix proportions of two slag with different particle sizes in each group were separately 1:1 for Group A, 4:1 for Group B, and 2:1 for Group C.

Table 4 Mix proportion design of cement mortar/g

2.3 Characterization methods

2.3.1 Mechanical strength

Cement mortar was prepared according to“Method of Testing Cements—Determination of Strength” (GB/T17671-1999), by replacing cement with slag and fly ash in proportion, which then was cured in a standard curing room for 3, 7, and 28 d, in purpose of testing the flexural strength and compressive strength.

2.3.2 Isothermal calorimetry test

The isothermal calorimetry test was carried out at the temperature of 20 ℃, by using TAM air hydration calorimeter.

2.3.3 X-ray diffraction analysis (XRD)

The analytical test was carried out with a scanning speed of 5°/min and 2θtest angle ranging from 5° to 70°, using the Empyrean X-ray diffractometer produced by Netherlands’ Panalytical company.

2.3.4 Thermal analysis (TG-DTG)

The hydration products of cement paste were tested by Netzsch STA449F3 thermal analyzer, in the temperature range from room temperature to 1 000 ℃,with a heating rate of 10 ℃/min and nitrogen as the protective atmosphere.

By calculating the mass loss in the ranges of 400-500 ℃ and 600-700 ℃ in the TG curve, the content of CH in the paste was calculated as follows[10-12]:

where,M1refers to the mass loss in the range of 400-500 ℃, whileM2refers to that in the range of 600-700℃.

3 Results and discussion

3.1 Particle size distribution

The particle sizes of ultrafine slag of different groups are shown in Fig.2, and particle size curves of compound cement mixed with different proportions mineral admixtures are shown in Fig.3.Generally,the wide distribution of particle sizes is favorable to improve the packing density of cement paste, while the ultrafine particles are favorable to improve its capability to fill interspace and accelerate the hydration rate of cement, so as to enhance the early performance of cement.As can be seen in Fig.2, after the addition of a certain proportion of Slag D, there is an increase in the proportion of slag with particle size distribution in smaller than 3 µm and 3-10 µm.Among all the groups,group A5 presents the highest content of particle size smaller than 3 µm, reaches to 34.2%, while in Group A6, slag with particle size in 3-10 µm reaches the highest proportion of 45.6%.Also, the addition of ultrafine slag can significantly increase the content of fine particles below 10 µm, which is helpful for filling the tiny interspace in the compound cementitious material, and subsequent improve the strength of matrix.

Fig.2 Particle size distribution of slag in different groups

Fig.3 Particle size distribution of cementitious materials

3.2 Mechanical properties

Fig.4 shows the compressive strength of compound cement mortar at curing ages of 3 and 7 d for the mix proportions.It can be seen that all the compressive strength for 3 d is significantly improved with the decrease of slag particle size.Among these,the compressive strength of A5 group reaches to 16.31 MPa, which is 40.5% higher than that of A1 Group; and the strength of A6 group is 15.78 MPa, 35.7% higher than that of A1 Group.It can be seen that the strength of samples eventually reaches the same level as that of the control group with the decrease of slag particle size.In case of the same amount of slag, increasing the content of slag with particle size smaller than 3 μm can substantially improve the early strength of sample.Meanwhile, its later strength could also attain the level of pure cement with a high-volume mineral admixtures.In addition, with the similar distribution of particle size in group A5 and B6, they are basically at the same level in terms of 3 and 28 d compressive strength, which also confirmed the reasonability of gradation design based on slag grinding.

Fig.4 Compressive strength of cement mortar at different curing ages

Figs.4(b) and 4(c) show that group C6 has the best early compressive strength, which is 52.8% higher than that of A1, and means the particle size of slag has very remarkable impact on the strength of cement mortar.The decrease of slag particle size leads to a certain increase in strength of each curing age, and especially has an obvious effect on the improvement of early strength.Due to the particle packing effects, ultrafine slag fills the interspace between large particles, and improves the overall packing density.Furthermore, slag with smaller particle size has larger specific surface area, increasing the reaction opportunities with cement hydration solution and making the structure more compact, thus enhancing the compressive strength of the matrix.

3.3 Hydration heat

Fig.5 shows the hydration exothermic rate and cumulative hydration heat of compound cement in each group.From Fig.5(a), it can be seen that cement mixed with mineral admixtures has a similar curve of exothermic rate with pure cement, both of which contain two exothermic rate peaks.In the first 40 minutes of hydration, C3A in cement takes the lead in hydration and ettringite (AFt) is rapidly generated in the presence of gypsum, producing a large amount of heat, and achieving the first peak of exothermic rate.As hydration proceeds, the second exothermic rate peak of each cement paste markedly appears at about 14 h.Since mineral admixtures have inferior early hydration activity than cement, their addition noticeably decreases the rate and peak values of hydration heat,which postpones the peaks.The dilution effect of mineral admixtures on cement results in a little shift of the whole hydration process, which also leads to a longer induction period.Specifically, the pure cement group (W0) reaches exothermic rate peak of 3.76 mW/g at about 12 h, while groups of cement mixed with mineral admixtures all reach their peaks after 14 h,with values ranging of 1.93-2.43 mW/g.Meanwhile,the exothermic rate peak rises with the decrease of slag particle size.Group A5 and A6 achieve the highest heat peak and higher early hydration heat, showing better mechanical properties at early age.From the figure of cumulative hydration heat (Fig.5(a)), it can be seen that pure cement has a much higher activity in hydration at early age than slag and fly ash.Cumulative heat of W0 for 3 d is 262.4 J/g, higher than that of any other experimental groups.In the first 24 h, the exothermic curves of cement paste with different particle sizes are almost the same.After 24 h of hydration, groups with higher fineness slag tend to increase the cumulative heat more quickly, which is due to the fact that the slag particles are involved in hydration after 24 h and slag with smaller particle size has higher hydration activity.

Fig.5 Hydration exothermic curves of prepared cement

From 5(b) and 5(c), it can be seen that the decrease of slag particle size increases the peak value and shortens the induction period.Group B6 reaches the peak of 2.57 mW/g at 13.5 h, which is an hour earlier than Group B1 (2.09 mW/g).Group C6 reaches the peak of 2.53 mW/g at 13.9 h, that is 0.3 hour earlier than Group C1 (2.13 mW/g), which indicates the finer slag particles can provide more nucleation sites,resulting in generating more cementitious products and faster the hydration rate.

3.4 XRD

According to the tests of mechanical properties and hydration heat, five samples with promising properties, respectively from groups A5, A6, B6, C5,and C6, were preferably selected for hydration products analysis.The XRD patterns of the above five samples at curing ages of 3, 7, and 28 d are shown in Fig.6.There are similar components contained in hydration products of the five samples with different proportions for 3, 7, and 28 d, which are mainly composed of CH,C3S, AFt and AFm.From Fig.6(a), the diffraction peaks of C3S and AFt have few changes at early age for 3 days hydration.As the main hydration product in the process of cement hydration, CH has a falling peak strength with the decrease of slag particle size, showing that the slag with smaller particle size has more possibility for secondary hydration, corresponding to more consumption of CH.According to Fig.6(b),during 7 days hydration, the C3S peaks of the five groups are reduced markedly in compared with those at curing age of 3 days, while the strength of CH diffraction peaks is obviously enhanced, indicating that continuous hydration reactions occur in all five groups with the increase of curing age.From Fig.6(c), it is found that C3S peaks of all experimental groups are obviously weakened after 28 days hydration.Moreover,the decrease of slag particle size also causes a more significant reduction of C3S peak strength, indicating that the finer slag could better promote the hydration process of cement.

3.5 TG-DTG

Fig.7 shows the TG-DTG curves of hydration products at different curing ages, in which there are two endothermic peaks.The obvious one at 400-500 ℃ is formed by the dehydrating decomposition of CH, and the weak one at 600-700 ℃ is due to the endothermic decomposition of CaCO3[13,14].According to the mass loss respectively in 400-500 ℃ and 600-700 ℃, the CH content at different curing ages can be calculated by Formula (1) and the results are shown in Table 5.As can be seen from Table 5, at the curing ages from 3 to 28 d, there is a certain increase in CH content of different groups, where Group C5 has the best hydration performance at early age due to the highest CH content of 9.13%, proving the better early strength from another aspect.From Fig.7(a), it can be seen that the decrease of slag particle size gradually strengthens the endothermic peak between 400 and 500 ℃ at 3 days curing age, and the corresponding mass loss is also increased accordingly, indicating the growing content of CH.As seen in Fig.7(b), since the CH consumed by pozzolanic reaction is less than the amount of newly generated CH from the cement, the total amount of CH in the system is slightly increased[15-17].

4 Conclusions

a) The decrease of slag particle size in cement with mineral admixtures can markedly increase the compressive strength of samples at different ages,especially in 3 d compressive strength.Under the condition of equal amounts of mineral admixtures, ballmilling slag could improve 3 d compressive strength by 40%, with the same level later strength for 28 d as that of pure cement.

b) The decrease of slag particle size in cement with mineral admixtures can promote the exothermic rate and increase the peak value, indicating that slag with smaller particle size has higher hydration activity.The heat of pure cement paste is obviously higher than that of cement paste mixed with mineral admixtures,and these hybrid pastes all have similar exothermic curves.

c) The decrease of slag particle size in cement with mineral admixtures leads to a gradual decrease in the strength of CH diffraction peak and a significant reduction in the strength of C3S diffraction peak in hydration products, which indicates that the finer slag can better promote the hydration process of cement.

Conflict of interest

All authors declare that there are no competing interests.