Investigation on the Role of Steel Slag Powder in Blended Cement Based on Quartz Powder as Reference

To clarify the role of steel slag powder in blended cement, steel slag powders with different amounts and particle sizes were mixed into blended cement and the inert quartz powder was selected as the reference. -e influences of steel slag powder with different amounts and particle sizes on the hydration and hardening properties of blended cement were studied from hydration heat, nonevaporable water content, porosity, hydration products, and strength. -e results show that the influence coefficient of nonevaporable water content (ՓWn) of blended cement paste is in an exponential relationship with the amount of steel slag powder. Moreover, at a dosage of 30%,ՓWn of blended cement gradually decreases with the increase of steel slag particle size. Both the early and late compressive strengths of blended cement are in a binomial relationship with the amount of steel slag powder. -e influence coefficient of steel slag powder on the compressive strength of blended cement is negative at the age of 3 days, whereas it is positive at the age of 28 days. -e chemical filling effect of 30% steel slag powder with different particle sizes in the blended cement paste is very small, only 1.13%–5.06%.-e hydration products of blended cement containing steel slag are mainly amorphous C-S-H gels and platy Ca(OH)2, and their density are consistent with the law of their porosity.


Introduction
It is well known that cement-based material is a nonhomogeneous and multiphase porous material [1][2][3]. e pore structure (including porosity and pore size distribution) and hydration reaction characteristics (including hydration degree, hydration product type, and quantity) are two key factors that affect the overall performance of cement-based materials [4][5][6][7]. Except for cement clinker, modern cementbased materials usually contain supplementary cementitious materials, such as slag and fly ash [8][9][10]. According to a lot of scholars' research, the influences of supplementary cementitious materials components on the properties of cement-based composites include two aspects: chemical action (i.e., the hydration reaction role) and the physical action (i.e., the physical filling role) [11][12][13]. e chemical action is not related to only the intrinsic activity of the supplementary cementitious component itself but also its interaction with cement clinker, while the physical action is mainly related to the filling role of the supplementary cementitious component on the system and effect on the pore structure [14][15][16][17][18].
ere is a great distinction between the different supplementary cementitious components due to chemical composition, activity, and hydration reaction mechanism, so the hydration and hardening performance of cement-based composites containing different supplementary cementitious materials also greatly differ [19][20][21]. erefore, it is necessary to reveal the chemical and physical effects of different supplementary cementitious components on cement-based composites, especially blended cement.
Steel slag is the main industrial waste produced in smelting steel. In China, the basic oxygen converter steel slag is the major kind of steel slag because the main smelting method is dominated by the converter steelmaking. e basic oxygen converter steel slag (hereafter called steel slag in this paper) is rich in silicate minerals, such as dicalcium silicate (C 2 S) and tricalcium silicate (C 3 S), similar to the mineral components of cement clinker, and it possesses a certain hydration activity and cementitious materials after grinding [22][23][24]. erefore, it has great potential as supplementary cementitious materials (or called mineral admixture). Meanwhile, the application of steel slag in cement and concrete is also an important way of steel slag resource utilization [25][26][27][28]. However, the hydration activity and hydration rate of the steel slag powder are much lower than those of the cement. In the blended cement, the cement clinker firstly starts hydration and hardening reaction, resulting in the formation of the system framework by the hydration products [25,29,30]. e effect of steel slag as a supplementary cementitious component on the hydration and hardening of blended cement includes two aspects, i.e., the filling role and the hydration effect. e former leads to the change of pore structure, whereas the latter is reflected in the change of hydration degree of composite cementitious materials. However, most of the current studies in literature focus on the mechanical properties and application effects of steel slag powder in cement and concrete, less paying attention to directly reveal the physical effect (namely, filling role) and chemical effect (namely, hydration role) of steel slag powder with different particle sizes or additional amounts in blend cement. In order to better use the steel slag powder as supplementary cementitious material, it is necessary to reveal the role of the steel slag powder with different amounts or particle sizes in blended cement so as to guide the blended cement containing the steel slag powder to obtain good comprehensive performance.
In this article, steel slag powders with different particle sizes were prepared and mixed into blended cement. e influences of steel slag powder with different amounts or particle sizes on the hydration and hardening properties of blended cement were studied from hydration heat, nonevaporable water content, porosity, hydration products, and strength. Meanwhile, the inert quartz powder, whose particle size distribution is similar to that of the steel slag powder, was also selected as the reference so as to initially reveal the role or contribution of steel slag powder in the hydration and hardening of blended cement. e results of this study also guide how to select or determine the best particle size or addition amounts of steel slag powder in blended cement, which is also meaningful or valuable for the efficient application of steel slag powder building materials.

Materials.
Steel slag used is the converter steel slag after heat-stewing process treatment, which was provided from Laiwu steel company of Shandong province in China. e quartz used is the natural quartz sand with the content of silicon dioxide not less than 95%. e chemical composition and mineral composition, which were determined by X-ray fluorescence (XRF) and X-ray diffraction (XRD) analysis, respectively, are shown in Table 1 and Figure 1.
Steel slag powder and quartz powder with different particle sizes were prepared using a laboratory ball mill under different grinding times. en, four types of steel slag powders with median diameter in the range of 10∼15 μm (denoted by R1), 15∼20 μm (denoted by R2), 20∼30 μm (denoted by R3), and 60∼70 μm (denoted by R4) were selected, respectively. Meanwhile, four kinds of quartz powders, whose particle sizes are similar to those of the steel slag powders, were selected as the reference. e median diameters and particle size distribution of powders selected are shown in Table 2 and Figure 2, respectively. e cement used was prepared using a laboratory ball mill according to the ratio of 95% clinker with 5% gypsum. e sieving residue of 80 μm is 4.4%, with a specific surface area of 402 m 2 /kg and a median diameter of 20.91 μm. e chemical composition of cement is also shown in Table 1.

Mix Design.
e mixture proportions shown in Table 3 were designed for the research of the influence of different percentages of steel slag on the hydration and hardening of blended cement. Meanwhile, the influences of steel slag powder with different particle sizes on the properties of the blended cement were studied using the mixture proportions, as shown in Table 4.

Testing Methods
(1) e hydration heat of blended cement paste containing steel slag powder was measured using a TAM Air isothermal calorimeter from TA instruments. e testing was performed at 25°C, within 96 h, and the water-to-binder ratio of paste is 0.5.
(2) e blended cement pastes with the water-to-binder ratio of 0.35 were prepared for the determination of nonevaporable water content, morphology, and pore structure of hardened blended cement paste. When the paste specimens were cured at a temperature of about 20 ± 1°C and ＞95% relative humidity to testing ages, the middle portions of specimens were obtained, broken, and soaked in absolute ethanol to stop hydration. en, the sample is dried at 105°C for at least 24 h so that it reaches a constant weight. Next, the test methods of nonevaporated water content and porosity are as follows: (a) Before testing the nonevaporable water content, the dried specimens need to be ground into a fine powder (less than 80 μm). en, the nonevaporable water content was determined as the difference in mass between the specimen heated at 105°C and 1000°C normalized by the mass after heating at 105°C and correcting for the loss on ignition of the unhydrated specimen [31]. e calculation formula is as follows: where W n is the nonevaporable water content of specimen, (%); W 1 is the mass of specimen at 105°C (i.e., dried mass of specimen), (g); W 2 is the mass of specimen at 1000°C, (g); L ssc is a coefficient taking into account the loss on 2 Advances in Civil Engineering ignition of cement and steel slag used in blend cement. e latter is calculated as follows: where p ss , and p c are the weight percentages of steel slag and cement in blended cement, respectively, and r ss and r c are the loss on ignition of steel slag and cement, respectively. (b) e morphologies of hydration products of blended cement pastes containing steel slag powder were observed using a FEI Quanta 200F scanning electron microscope (SEM, 15 KV) under high vacuum condition. (c) e porosities and pore size distribution of hardened cement pastes containing steel slag powder and quartz powder were determined using an Autopore 9500 Mercury intrusion porosimetry (MIP) from Micromeritics Instruments Corporation, whose pore analysis range was 3 to 4500000 nm. (d) Paste bars (4 × 4 × 16 cm) were prepared for the determination of the strength of blended cement pastes containing steel slag powder and quartz powders. e compressive strengths of paste specimens after curing for 3 days, 7 days, and 28 days at 20°C were measured according to the Chinese National Standard GB/T17671-1999.

Hydration Heat of Blended Cement
Paste. e hydration heat curves of blended cements containing different amounts of steel slag powder are shown in Figure 3. As can be seen from Figure 3, the incorporation of steel slag powder prolongs the induction period of cement hydration, delays the second exothermic peak, and decreases the peak value. e second exothermic peak becomes broader. e higher the content of steel slag powder, the longer the induction period of blended cement, the later the appearance of the second exothermic peak, the lower the peak, and the broader the peak, showing that the addition of steel slag powder    slows down the hydration reaction rate of blended cement. is is mainly due to the low activity of steel slag powder, which slows down the overall hydration process of cement after replacing cement. In addition, the relatively high content of P 2 O 5 or phosphate in the steel slag powder used will also delay the hydration process of cement. e heat release of the reference cement and the cement mixed with 20% steel slag powder is mainly concentrated before 30 h, while the heat release of the high content (40% and 50%) of steel slag cement is mainly after 20 h. e second exothermic peaks of steel slag cement mixed with 20%∼50% are about 2/ 3, 2/5, and 1/4 of the reference cement, respectively. For the

Nonevaporable Water Content of Blended Cement Paste.
e nonevaporable water content of blended cements containing different amounts of steel slag powder at 28 days is shown in Figure 4. It can be seen from Figure 4 that the nonevaporable water content of the blended cement gradually decreases with the increase of steel slag powder content. e addition of a small amount of steel slag powder (such as 20% content) has little effect on the nonevaporable water content of the blended cement. When the steel slag powder content is greater than 40%, the nonevaporable water content of the blended cement is significantly reduced. Moreover, there is a binomial relationship between the nonevaporable water content of blended cement and the amount of steel slag powder.
In order to further reflect the contribution of steel slag powder to the nonevaporable water content of blended cement, the quartz powder (inert material) with similar particle size distribution was selected as a reference. Moreover, it is assumed that the nonevaporable water content of blended cement is generated by the hydration superposition of clinker and steel slag. Furthermore, the influence coefficient of nonevaporable water content, Փ Wn , is used to represent the contribution of steel slag powder to the nonevaporable water content of the blended cement. e calculation formula is as follows: where Wn (CSS) is the nonevaporable water content of the blended cement paste containing steel slag powder; and Wn (CQ) is the nonevaporable water content of the blended cement paste containing quartz powder. As shown in Figure 5, the influence coefficient of nonevaporable water content (Փ Wn ) gradually increases with the increase of the amount of steel slag powder, and the two   Advances in Civil Engineering 5 meet an exponential relationship. When the amount of steel slag powder is less than 40%, Փ Wn of blended cement is less than 10%, which also indicates that the contribution of steel slag powder to the nonevaporable water content of blended cement is much less than that of cement clinker. Similarly, the quartz powder, an inert material with a similar particle size distribution, is selected as a reference, and it is assumed that the strength of the blended cement is only produced by the hydration of the clinker and steel slag. erefore, the strength influence coefficient, Փ S , is used to represent the contribution of steel slag powder to the strength of blended cement. e calculation formula is as follows (4), and the result is shown in Figure 7: where S CSS is the compressive strength of the blended cement paste containing steel slag powder and Wn (CQ) is the compressive strength of the blended cement paste containing quartz powder. e results in Figures 7(a) and 7(b) show that the compressive strength of the blended cement paste containing steel slag powder at the age of 3 days is lower than that of the blended cement containing quartz powder, but the result is the opposite at the age of 28 days. e reason is that the quartz powder is an inert material and does not participate in the hydration reaction. It plays a "dilution" effect on the blended cement system in the early stage, thereby increasing the actual water-to-cement ratio of the clinker minerals involved in hydration reaction and promoting its early hydration reaction degree. By contrast, the  (2) 7 d fitting curve y = 0.003x 2     steel slag powder has a low hydration degree at the age of 3 days due to the low early activity, so its contribution to the early strength of blended cement is very small. However, at the same time, it needs to occupy the water in the system to dissolve the mineral ions so that the water involved in the hydration reaction of clinker mineral is relatively reduced, delaying its hydration process, resulting in its strength contribution is also reduced.
However, with the prolongation of the age, the hydration degree of steel slag powder in blended cement gradually increases. By the age of 28 d, the strength of blended cement produced by hydration of clinker and steel slag will exceed that of blended cement containing quartz powder. It can be seen intuitively from Figure 7(c) that the strength influence coefficient of the blended cement paste, Փ S , is negative at the age of 3 d. e greater the amount of steel slag powder, the greater the negative influence on the compressive strength. However, Փ S is positive at the age of 28 days and exceeds 25%.

Morphology of Blended Cement Paste.
e morphologies of blended cement containing different amounts of steel slag powder (R2) at 3 days and 28 days are shown in Figure 8.
It can be seen from Figure 8(a) that the morphologies of blended cement have changed obviously with the increase of the amount of steel slag powder. Blended cement with a small additional amount (such as 20%) of steel slag powder has more hydration products, such as amorphous C-S-H gels and Ca(OH) 2 , and compact structure with fewer pores. Moreover, the product system mainly consists of cement hydration products, in which a very small number of hydration products of steel slag are wrapped. On the contrary, when the addition amount of steel slag powder is high (such as 80%), the hydration products of steel slag are the main components in the blended cement, but the hydration products in the system are initially formed with scattered distribution and loose structure with more pores and poor compactness. Figure 8(b) shows that the hydration products of blended cement increased significantly at 28 days, and the amorphous C-S-H gels accumulated on one another and were basically connected as a whole. Moreover, with the increase of the addition amount of steel slag powder, the change of the morphology and microstructure of blended cement at 28 days is relatively smaller than that at 3 days. is is due to the fact that the hydration of steel slag at 28 days is obviously sufficient compared with that at 3 days, and the hydration products generated continuously fill the pore structure and improve the compactness in the system during 28 days. ereout, the effect of steel slag powder content on the morphology and microstructure of the hydration products of blended cement is relatively weakened at 28 days, which is consistent with the above-mentioned influence of steel slag powder content on the strength of blended cement.

Pore Structure of Blended Cement Paste.
e pore volume distribution and aperture parameter value of blended cement paste containing steel slag powder at 28 days are shown in Figures 9 and 10, respectively. It can be seen from Figure 9 that the pore size of blended cement paste is mainly distributed in the range of 3-10 nm and 30-100 nm. e higher the content of steel slag powder in blended cement, the higher the cumulative pore volume of the paste. As shown in Table 5, the total pore volume, median pore size, and porosity of the blended cement paste gradually increase with the increase of steel slag powder content. e total pore volume increased from 0.1178 mL/g to 0.1995 mL/g, the median pore size increased from 21.1 nm to 34.4 nm, and the porosity increased from 21.62% to 34.80%. In particular, the pores of blended cement paste with a large amount of steel slag increased significantly.

Nonevaporable Water Content of Blended Cement
Paste. e nonevaporable water contents of blended cement pastes containing 30% steel slag powder or quartz powder with different particle sizes at 28 days are shown in Figure 10.
e results show that the nonevaporable water contents of blended cement paste are gradually decreased with the increase of particles' size of steel slag powder. However, the nonevaporable water contents of blended cement paste containing quartz powder show an opposite law. e reason is that the quartz powder does not participate in the hydration reaction and plays a "dilution" effect on the blended cement system. e larger the particle size of quartz powder, the stronger the "dilution" effect in a blended cement system. us, the hydration reaction of the clinker minerals will be promoted, resulting in a relatively high nonevaporable water content.
As shown in Figure 11, Փ Wn gradually decreases with the increase of particles' size of steel slag powder. When the particle size of steel slag powder is R1, Փ Wn of blended cement is 18.32%, and when the particle size of steel slag powder increases to R2, the influence coefficient decreases to 9.5%. It indicates that 30% steel slag powder has little contribution to the nonevaporable water content of blended cement, and its contribution is also greatly affected by particle size.

Porosity of Blended Cement Paste.
e porosities of blended cement pastes containing 30% steel slag powder or quartz powder with different particle sizes at 28 days are shown in Figure 12. e results show that the porosities of blended cement paste gradually increase with the increase of particle size of steel slag powder or quartz powder. e reason is that the quartz powder is an inert material and only plays a filling role in blended cement. erefore, quartz powder with a particle size distribution similar to steel slag powder can be used to approximate the physical filling effect of steel slag powder on blended cement.
It can be seen from Figure 12 that the porosity of blended cement paste decreases with the decrease of quartz particle size; that is, the physical filling effect of steel slag powder on blended cement is enhanced with the decrease of particle size. Comparing the porosities of the blended cement paste with steel slag powder and the quartz powder, it can be seen that the smaller the particle size, the greater the gap between the two. is is because the steel slag powder with a small particle size has relatively high hydration activity, which promotes the production of more hydration products and further fills the pores of the paste, thereby reducing the porosity of paste. e filling effect caused by the hydration reaction of steel slag powder can be regarded as the chemical filling role (Փ F ) of steel slag powder on blended cement. e calculation formula is as follows (5), and the result is shown in Figure 13.
where P CSS is the porosity of the blended cement paste containing steel slag powder; S CQ is the porosity of the blended cement paste containing quartz powder. As shown in Figure 13, the chemical filling role gradually weakens with the increase of the particle size of steel slag powder. Generally, the chemical filling role of steel slag powder on the blended cement is relatively small. e chemical filling role of 30% steel slag powder with different particle sizes in blended cement paste is only in the range of 1.13%-5.06%.

Strength of Blended Cement Paste.
e strengths of blended cement pastes containing 30% steel slag powder or Advances in Civil Engineering Nonevaporable water content (%) Figure 10: Nonevaporable water content of blended cement containing steel slag powder with different particle sizes. quartz powder with different particle sizes at 28 days are shown in Figure 14. e results show that the particle size of steel slag powder has different effects on the 3 d and 28 d strength of the blended cement paste. With the increase of particle size of the steel slag powder, the 3 d strength of the blended cement paste first increases and then decreases, while the 28 d strength gradually decreases. e reason is similar to the explanation in Section 3.3. As shown in Figure 15, the results show that the influence coefficient of steel slag powder on the compressive strength of the blended cement paste is negative at 3 days, and the smaller the particle size of steel slag powder, the greater the negative impact on compressive strength. At the age of 28 days, the influence coefficient of strength is positive and in the range of 25%-30%. is is also just close to the amount of steel slag powder in the blended cement in this study.

Morphology of Blended Cement Paste.
e morphologies of blended cement containing steel slag powder with different particle size at 3 days and 28 days are shown in Figure 16.
It can be seen from Figure 16(a) that the type and morphology of the hydration products of the four kinds of blended cement basically did not change much, but the compactness of the paste structure gradually increased slightly with the increase of the particle size of steel slag. Figure 16(b) shows an opposite rule; that is, with the increase of the particle size of steel slag, the hydration products of blended cement gradually decrease at 28 days, the structure of paste becomes loose, and the compactness also gradually decreases. is is consistent with the abovementioned influence of particle size of steel slag powder on the strength of blended cement. e reason is that the early Advances in Civil Engineering activity of steel slag is very low, and the steel slag with a large particle size basically would not participate in the hydration reaction at 3 days, which actually generates "dilution effect" to the water-cement ratio of the cement component in blended cement system. e "dilution effect," in other words, is to increase the water-cement ratio of the cement component in the blended cement system, so the hydration of the cement component in blended cement is accelerated. e smaller the steel slag particle size, the weaker the dilution effect, so the microstructure of blended cement containing steel slag powder with large particle size is relatively good at 3 days. However, with the extension of age, the hydration reaction of steel slag with a small particle size would gradually improve, and its hydration products can further fill the structure of blended cement to make it more compact in later ages. While the hydration effect of steel slag with a large particle size is lower than that of steel slag with small particle size, the blended cement containing steel slag with smaller particle size has more hydration products and a relatively compact microstructure. It also explains the effect rule of particle size of steel slag on the macroscopic properties of blended cement from the view of morphology and microstructure of hydration products.    Advances in Civil Engineering 13

Conclusions
(1) e influence coefficient of nonevaporable water content (Փ Wn ) of blended cement paste is in an exponential relationship with the amount of steel slag powder. Moreover, at a dosage of 30%, Փ Wn gradually decreases with the increase of the steel slag powder particle size. (2) Both the early and late compressive strengths of blended cement paste are in a binomial relationship with the amount of steel slag powder. With the increase of particle size of the steel slag powder, the 3 d strength of the blended cement paste first increases and then decreases, while the 28 d strength gradually decreases. e influence coefficient of steel slag powder on the compressive strength of blended cement paste is negative at the age of 3 days. e greater the amount of steel slag powder or the smaller the particle size, the greater the negative influence on the compressive strength; and at the age of 28 days, the strength influence coefficient is positive.
(3) e pore size of the blended cement paste containing steel slag powder is mainly distributed in the range of 3-10 nm and 30-100 nm, and both the median pore size and porosity increase gradually with the increase of steel slag powder content. e chemical filling role of steel slag powder on the blended cement paste gradually weakens with the increase of its particle size. e chemical filling role of 30% steel slag powder with different particle sizes in the blended cement paste is only in the range of 1.13%-5.06%. e hydration products of blended cement containing steel slag powder are mainly amorphous C-S-H gel and platy Ca(OH) 2 , and its density is consistent with the law of its porosity.

Data Availability
e data used to support the findings of this study are included within the article.

Conflicts of Interest
e authors declare that they have no conflicts of interest.