**3. Results and discussions**

#### **3.1 Workability of SCC**

Figure 2 illustrates slump vs. different w/cm ratios for both OPC and CSC; the figure indicates that all slump values are greater than 230 mm, the specification for SCC with high flow. Concrete with a lower w/cm ratio than 0.28—implying significantly high binder content—may lead to higher slump and satisfactory flowability. For an identical dosage of SP mixtures, CSC has higher slump than OPC mixtures. Hence, it is clear that the use of CSS might improve the workability of SCC; and this is similar to the test results of the research papers referenced. Hence, the use of CSS in SCC can lead to high flow properties.

### **3.2 Setting time of SCC**

Figure 3 shows the effects of CSS on the penetration resistance of concrete, indicating that as the w/cm ratio of OPC or CSC increases, the penetration resistance decreases and the setting time increases. Further, the setting time of CSC is longer than that of OPC irrespective of the w/cm ratio. This is due to the low PAI values of CSS. Thus, the setting time increases with the amount of CSS. This result is similar to the results in a previous study, which showed that the addition of BFS decreased the setting time of SCC [Mihashi H, Yan X, Arikawa S.,1995.].

In order to obtain high-strength SCC with lower water content (160 kg/m3), in this study, w/cm ratios (water/(cement+CSS)) of 0.28, 0.32 and 0.40 were selected. Further, large amounts of SP were added to achieve better flow behaviour. CSS powder was used to replace the 5.0%, 7.5% and 10% weights of Portland cement. Mixtures with three different w/cm ratios were prepared for ordinary plain concrete (OPC) and carbon steel slag concrete (CSC), as shown in Table 2; designated as OPC28, OPC32 and OPC40 and CSC28, CSC32 and CSC40, respectively﹝Whitcomb Brent L., Kiousis Panos D.,2009. Kwan Albert K.

> CSS/ cement (%)

OPC28 0.28 0.28 572 -- 757 901 145 15 160 OPC32 0.32 0.32 500 -- 783 932 148 12 160 OPC40 0.40 0.40 400 -- 820 976 152 8 160 CSC28 0.28 0.28 545 5.0 832 820 146 14 160 CSC32 0.34 0.32 465 7.7 861 849 149 11 160 CSC40 0.44 0.40 364 10.0 901 888 153 7 160

Figure 2 illustrates slump vs. different w/cm ratios for both OPC and CSC; the figure indicates that all slump values are greater than 230 mm, the specification for SCC with high flow. Concrete with a lower w/cm ratio than 0.28—implying significantly high binder content—may lead to higher slump and satisfactory flowability. For an identical dosage of SP mixtures, CSC has higher slump than OPC mixtures. Hence, it is clear that the use of CSS might improve the workability of SCC; and this is similar to the test results of the research

Figure 3 shows the effects of CSS on the penetration resistance of concrete, indicating that as the w/cm ratio of OPC or CSC increases, the penetration resistance decreases and the setting time increases. Further, the setting time of CSC is longer than that of OPC irrespective of the w/cm ratio. This is due to the low PAI values of CSS. Thus, the setting time increases with the amount of CSS. This result is similar to the results in a previous study, which showed that the addition of BFS decreased the setting time of SCC [Mihashi H,

papers referenced. Hence, the use of CSS in SCC can lead to high flow properties.

Mix Proportion (kg/m3)

Coarse

aggregate Water SPc Water

+ SP

Fine aggregate

**2.2 Mixture design** 

H.,Ng, Ivan Y. T.,2008.﹞.

a w/c ratio = water/cement

**3.1 Workability of SCC** 

**3.2 Setting time of SCC** 

Yan X, Arikawa S.,1995.].

c SP = Superplasticizer

b w/cm ratio = water/(cement + CSS)

Table 2. Mixture proportion of SCC.

**3. Results and discussions** 

w/c ratioa w/cm ratiob

Cement

Designation of concrete

Fig. 2. Slump vs. various w/cm ratios of concrete.

Fig. 3. Comparison between the OPC and CSC with respect to penetration resistance of concrete.

Carbon Steel Slag as Cementitious Material for Self-Consolidating Concrete 329

CSC28 0.29 42.9 (92) 50.0 (93) 63.2 (91) 71.0 (76) 82.5 (86)

CSC32 0.34 26.5 (106) 34.1 (107) 58.5 (118) 69.4 (121) 78.9 (134)

CSC40 0.44 17.6 (108) 24.1 (109) 40.7 (110) 50.6 (112) 55.3 (121)

Theoretically, the ultrasonic pulse velocity (UPV) of a solid object is higher that of air, and a high-density solid will have high UPV. Therefore, the UPV is a good measure of the soundness of hardened concrete. It is generally acknowledged that the UPV increases with concrete density. In our study, the UPV of all mixtures was greater than 4200 m/s. Table 4 shows the UPV and the difference in UPV between CSC and OPC (as a percentage) at each w/cm ratio, from 3 to 90 days. The UPV of CSC28 at all ages is lower by 1% to 2% than that of OPC28; however, the UPV of CSC32 and CSC40 is higher by 3% than that of OPC32 and OPC40, respectively. This result is similar to trend in compressive strength—the addition of CSS enhances the pozzolanic reaction with high w/c or w/cm ratios, i.e., sufficient water. Figure 5 shows a good linear relationship between the compressive strength and UPV of concrete for both OPC and CSC. In other words, UPV is a good method for evaluating the

CSC28 0.29 4525 (98) 4683 (98) 4828 (99) 4830 (99) 4835 (99)

CSC32 0.34 4466 (102) 4669 (103) 4814 (101) 4821 (101) 4825 (101)

CSC40 0.44 4295 (102) 4361 (102) 4659 (101) 4709 (102) 4728 (102)

Table 3. Compressive strength and percentage compressive strength of concrete.

Compressive strength, MPa (% compressive strengtha)

3 days 7 days 28 days 56 days 90 days

46.5 (100) 53.9 (100) 69.6 (100) 93.7 (100) 95.6 (100)

25.0 (100) 32.0 (100) 49.6 (100) 57.3 (100) 59.1 (100)

16.3 (100) 22.2 (100) 37.1 (100) 45.1 (100) 45.8 (100)

UPV of concrete, m/s (% UPVa)

3 days 7 days 28 days 56 days 90 days

4606 (100) 4787 (100) 4859 (100) 4876 (100) 4890 (100)

4381 (100) 4523 (100) 4760 (100) 4777 (100) 4785 (100)

4211 (100) 4274 (100) 4595 (100) 4635 (100) 4651 (100)

Designation of concrete

OPC28 0.28

OPC32 0.32

OPC40 0.40

w/c ratio w/cm ratio

0.28

0.32

0.40

**3.4 Ultrasonic pulse velocity (UPV) of SCC** 

performance and homogeneity of SCC.

w/cm ratio

0.28

0.32

0.40

a Percentage UPV = (CSC/OPC) × 100 at fixed w/cm ratios

Table 4. UPV and percentage UPV of SCC.

w/c ratio

Designation of concrete

OPC28 0.28

OPC32 0.32

OPC40 0.40

#### **3.3 Compressive strength of SCC**

The compressive strength and percentage of concrete mixtures with different w/cm values at the specified age are shown in Table 3. The compressive strength of each mixture is greater than 41 MPa at 56 days. This satisfies the requirement that SCC must have high strength [Hogan FJ, Meusel JW.,1981.]. The compressive strength of OPC and CSC with a w/cm ratio of 0.28 is either equal to or higher than 83 MPa at 90 days; however, that of CSC28 at any age is lower than that of OPC28. In contrast, the percentage of compressive strength is higher than 90% at 28 days, and it is reduced to 76% at 90 days. This corresponds to a 15% reduction in the PAI at 28 days. The compressive strength of CSC with a w/cm ratio of 0.32 or 0.40 at any age, however, is higher than that of OPC, and the percentages of compressive strength are from 106% to 134% and from 108% to 121%, with respect to the w/cm ratio. This clearly indicates that the addition of CSS improves the strength development of cement paste as long as the water-to cement (w/c) ratio is greater than 0.32 or w/cm is higher than 0.28. This means that the reactions of strength development of cement with CSS will be enhanced with sufficient water contents. It is suggested, however, that the total water content of concrete, including the moisture in liquid admixture be maintained as low as possible to avoid large shrinkage and sedimentation. Figure 4 shows the influence of CSS content on compressive strength: higher CSS content mixtures will cause lower compressive strength.

Fig. 4. Effect of CSS content at different ages on the compressive strength of SCC.


Table 3. Compressive strength and percentage compressive strength of concrete.

#### **3.4 Ultrasonic pulse velocity (UPV) of SCC**

328 Material Recycling – Trends and Perspectives

The compressive strength and percentage of concrete mixtures with different w/cm values at the specified age are shown in Table 3. The compressive strength of each mixture is greater than 41 MPa at 56 days. This satisfies the requirement that SCC must have high strength [Hogan FJ, Meusel JW.,1981.]. The compressive strength of OPC and CSC with a w/cm ratio of 0.28 is either equal to or higher than 83 MPa at 90 days; however, that of CSC28 at any age is lower than that of OPC28. In contrast, the percentage of compressive strength is higher than 90% at 28 days, and it is reduced to 76% at 90 days. This corresponds to a 15% reduction in the PAI at 28 days. The compressive strength of CSC with a w/cm ratio of 0.32 or 0.40 at any age, however, is higher than that of OPC, and the percentages of compressive strength are from 106% to 134% and from 108% to 121%, with respect to the w/cm ratio. This clearly indicates that the addition of CSS improves the strength development of cement paste as long as the water-to cement (w/c) ratio is greater than 0.32 or w/cm is higher than 0.28. This means that the reactions of strength development of cement with CSS will be enhanced with sufficient water contents. It is suggested, however, that the total water content of concrete, including the moisture in liquid admixture be maintained as low as possible to avoid large shrinkage and sedimentation. Figure 4 shows the influence of CSS content on compressive strength: higher CSS content mixtures will

> 4 5 6 7 8 9 10 11 CSS /Cement (%)

Fig. 4. Effect of CSS content at different ages on the compressive strength of SCC.

CSC(7 days) CSC( 28 days) CSC(56 days) CSC(90 days)

**3.3 Compressive strength of SCC** 

cause lower compressive strength.

Compressive strength (MPa)

Theoretically, the ultrasonic pulse velocity (UPV) of a solid object is higher that of air, and a high-density solid will have high UPV. Therefore, the UPV is a good measure of the soundness of hardened concrete. It is generally acknowledged that the UPV increases with concrete density. In our study, the UPV of all mixtures was greater than 4200 m/s. Table 4 shows the UPV and the difference in UPV between CSC and OPC (as a percentage) at each w/cm ratio, from 3 to 90 days. The UPV of CSC28 at all ages is lower by 1% to 2% than that of OPC28; however, the UPV of CSC32 and CSC40 is higher by 3% than that of OPC32 and OPC40, respectively. This result is similar to trend in compressive strength—the addition of CSS enhances the pozzolanic reaction with high w/c or w/cm ratios, i.e., sufficient water. Figure 5 shows a good linear relationship between the compressive strength and UPV of concrete for both OPC and CSC. In other words, UPV is a good method for evaluating the performance and homogeneity of SCC.


a Percentage UPV = (CSC/OPC) × 100 at fixed w/cm ratios

Table 4. UPV and percentage UPV of SCC.

Carbon Steel Slag as Cementitious Material for Self-Consolidating Concrete 331

(a) (b)

(a) (b)

Fig. 6. SEM micrograph of CSC28: (a) at 3 days; (b) at 28 days.

Fig. 7. SEM micrograph of CSC32: (a) at 3 days; (b) at 28 days.

Fig. 5. Relationship between UPV and compressive strength of OPC and CSC.

#### **3.5 Microstructure observation**

Scanning electron microscopy (SEM) observations are conducted with specimens at the ages of 3 and 28 days. The image characteristics of concrete at 3 days are shown in Figs. 6(a)–8(a). As shown in Fig. 6(a), at the early age of 3 days, considerable amounts of hexagonal-shaped calcium hydroxide (Ca(OH)2), spherical-shaped C-S-H gel in CSC28 (w/cm = 0.28) and certain amounts of fine pores (dark zone) exist. Figure 7(a) shows the presence of numerous rosette-shaped mono-sulfoaluminate (AFm) and small amounts of needle-shaped ettringite (AFt) in CSC32 (w/cm = 0.32). Figure 8(a) also shows that there are rosette-shaped AFm in CSC40 (w/cm = 0.40). Here, the w/cm ratio is greater than or equal to 0.32 as a result of increase in CSS amounts and the existence of high Al2O3 and Fe2O3 contents. The primary hydration products are hexagonal-shaped Ca(OH)2, spherical-shaped C-S-H gel and a certain amount of rosette-shaped AFm. This observation confirms the conclusions made for both strength and UPV that the reaction of CSS with cement paste requires sufficient water to aid hydration. At a later age, as shown in Fig. 6(b), the microstructure of CSC28 is extremely dense. Figure 7(b) also shows the presence of numerous rosette-shaped AFm, but no needle-shaped Aft, while Fig. 8(b) shows large pores in CSC40 with a large amount of hexagonal-shaped Ca(OH)2 in the reaction process. While this is advantageous for the hydration reaction of CSS, it also indicates that more pores are observed with higher w/cm ratios.

Fig. 5. Relationship between UPV and compressive strength of OPC and CSC.

Scanning electron microscopy (SEM) observations are conducted with specimens at the ages of 3 and 28 days. The image characteristics of concrete at 3 days are shown in Figs. 6(a)–8(a). As shown in Fig. 6(a), at the early age of 3 days, considerable amounts of hexagonal-shaped calcium hydroxide (Ca(OH)2), spherical-shaped C-S-H gel in CSC28 (w/cm = 0.28) and certain amounts of fine pores (dark zone) exist. Figure 7(a) shows the presence of numerous rosette-shaped mono-sulfoaluminate (AFm) and small amounts of needle-shaped ettringite (AFt) in CSC32 (w/cm = 0.32). Figure 8(a) also shows that there are rosette-shaped AFm in CSC40 (w/cm = 0.40). Here, the w/cm ratio is greater than or equal to 0.32 as a result of increase in CSS amounts and the existence of high Al2O3 and Fe2O3 contents. The primary hydration products are hexagonal-shaped Ca(OH)2, spherical-shaped C-S-H gel and a certain amount of rosette-shaped AFm. This observation confirms the conclusions made for both strength and UPV that the reaction of CSS with cement paste requires sufficient water to aid hydration. At a later age, as shown in Fig. 6(b), the microstructure of CSC28 is extremely dense. Figure 7(b) also shows the presence of numerous rosette-shaped AFm, but no needle-shaped Aft, while Fig. 8(b) shows large pores in CSC40 with a large amount of hexagonal-shaped Ca(OH)2 in the reaction process. While this is advantageous for the hydration reaction of CSS, it also indicates that more

**3.5 Microstructure observation** 

pores are observed with higher w/cm ratios.

Fig. 6. SEM micrograph of CSC28: (a) at 3 days; (b) at 28 days.

Fig. 7. SEM micrograph of CSC32: (a) at 3 days; (b) at 28 days.

Carbon Steel Slag as Cementitious Material for Self-Consolidating Concrete 333

[1] ACI Committee 211. Guide for selecting proportions for high-strength concrete. ACI

[2] Chiang CC, Chenn YY, Lin TY, Hwang CL. The application of electric arc furnace

[3] Esfahani M. Reza, Kianoush M. Reza, Bond strength of glass fibre reinforced polymer

[4] Gregori Amedeo, Ferron Raissa, Sun Zhihui, Shah Surendra, Experimental simulation of

[5] Geiser J. Metallurgical slag-create a future construction material. The International Associate of Metallurgical Slag Recycle and Utilization. Beijing, 1999. p. 1–8. [6] Hogan FJ, Meusel JW. Evaluation for durability and strength development of a ground

[7] Hwang Soo-Duck, Khayat Kamal H., Effect of mixture composition on restrained

[8] Koehler Eric P., Fowler David W., Dust-of-fracture aggregate microfines in self-

[9] Kwan Albert K. H., Ng, Ivan Y. T., Performance criteria for self-consolidating concrete, Transactions Hong Kong Institution of Engineers, v 15, n 2, p 35-41, June 2008. [10] Khalifa AJ, Ramzi T, Mohammed AG. Use of Copper slag and cement by-pass dust as

[11] Luxán MP, Sotolongo R, Dorrego F, Herrero E. Characteristics of the slags produced in the fusion of scrap steel by electric arc furnace. Cem Concr Res 2000;30:517–519. [12] Li G, Zhao X. Properties of concrete incorporating fly ash and ground granulated blast-

[13] Mihashi H, Yan X, Arikawa S. Strength properties and frost damage resistance of high

[14] Monshi A, Asgarani MK. Producing Portland cement from iron and steel slag and

[15] Mihashi H, Yan X, Arikawa S. Strength properties and frost damage resistance of high

Concrete: Material Properties and Design. Germany, 1995. p. 195–204. [16] Paczkowski Piotr, Kaszynska Maria, Self-consolidating concrete for on-site bridge

performance concrete using blast furnace slag and silica fume. In: Schwesinger P, editor. Proceedings of the Fourth Weimar Workshop on High Performance

performance concrete using blast furnace slag and silica fume. High Performance

applications, International Conference organised by the Institution of Civil

granulated blast furnace slag. Cem Concr Aggregates 1981;3:40–52.

Journal, v 105, n 5, p 499-509, September/October 2008.

cementitious materials. Cem Concr Aggregates 2002;24:7–12.

furnace slag. Cem Concr Compos 2003;25:293–299.

Concrete. Germany, 1995. p. 195–204.

Engineers, ICE, p 312-320, 2007.

limestone. Cem Concr Res 1999;29:1373–1377.

Civil and Hydraulic Engineering 2004;16:167–178.

Civil Engineering, v 32, n 3, p 553-560, June 2005.

reduction slags to high performance concrete. Journal of the Chinese Institute of

reinforcing bars in normal and self-consolidating concrete, Canadian Journal of

self-consolidating concrete formwork pressure, ACI Materials Journal, v 105, n 1, p

shrinkage cracking of self-consolidating concrete used in repair, ACI Materials

consolidating concrete, ACI Materials Journal, v 105, n 2, p 165-173, March/April

**5. References** 

2008.

Mater J 1993;90:272–283.

97-104, January/February 2008.

Fig. 8. SEM micrograph of CSC32: (a) at 3 days; (b) at 28 days.
