A study on the mixed properties of green controlled low strength cementitious

In this study, by-products with CFA and GGBS were used to replace cement (53–100 wt.%) and (50–100 vol.%), respectively. Two types of super-plasticizers, one type of accelerator, two types of methylcellulose, and one type of adhesives were used to adjust the properties of the fresh mix to meet the self-consolidating properties. For the cementing material, CFA and GGBS were used to replace the cement with high content, with the hope of achieving the recycling economic benefits of waste reuse and replacing the cementing of up to 78% by volume in the concrete.

The coarse aggregates were mined from the Lanyang River. After taking into account the requirements of fluidity, a gradation with a maximum particle size of 13 mm was selected. The specific gravity of the coarse aggregates was 2.84 and the water absorption was 1.24%. The cement used in this study was Portland Type I cement from Taiwan Cement Co., Ltd. The main ingredients of the cement were 61.96% CaO and 20.42% SiO₂. The specific gravity of the cement was 3.15 and the specific surface area was 345 m²/kg.

The composition of the cement met the requirements of the American Society for Testing and Materials (ASTM) C150. According to the chemical properties of the CFA in Table 2, the main chemical components were CaO, Al₂O₃, SiO₂, and a small amount of SO₃. The scanning electron microscope (SEM) photo is shown in Fig. 3, and the X-ray Diffraction (XRD) component analysis is shown in Fig. 4. The CFA that was studied in this experiment was introduced by Yongfengyu Group’s Shinwu Plant, and it was produced by the combustion of a CFBC at Ahlstrom Company. The specific gravity of the CFA was 2.73 and the specific surface area was 280 m²/kg.

Fig. 3

Fig. 4

The GGBS was selected by following ASTM C989 [38], and it was made by smelting molten iron in the steelmaking plant of CHC Resources Co., Ltd. The molten slag produced with iron ore, limestone, coke, and flux was produced by high pressure. The water quenching blast furnace grate was produced by the water quenching process, the main chemical components of which were CaO, SiO₂, and Al₂O₃. Its properties are shown in Table 3. It had the properties of hydration and cementation that were caused by the reaction of the pozzolans. The surface area was 419 m²/kg and the activity was 112% in 28 days. The SEM micrograph is shown in Fig. 5.

Fig. 5

In this study, the reference specification American Concrete Institute (ACI) 229 [39] was used for the CLSM proportioning design. First, it was necessary to confirm whether the characteristics of the pozzolans were taken as the priority consideration and whether to replace the cementitious or aggregates. The materials were selected to reach the standard workability and compressive strength. Table 4 shows the mix design of the activity index test in cementation material. Besides, three specimens were tested for each mixture in our test, and then the results were averaged and compared. A standard deviation was controlled less than 5% for the tested results.

The recommended dosage of the CLSM according to the recommendations of ACI 229R is shown in Table 5. The recommended dosage of the cement was in the range of 30–120 kg/m³. In the CLSM, the cement content was very small compared with that of ordinary concrete. However, the reason why the recommended dosage did not use cement was that cement replaced materials would have affected the material in terms of the cementing ability and the setting time. Therefore, the addition of a small amount of cement was an irreplaceable factor in the construction control of the proportioning design, followed by water. In terms of volume, the cement replaced materials were used in large quantities. If certain workability was required, more water was used than in ordinary cement concrete. Therefore, the water consumption was as high as 193 kg/m³ to 344 kg/m³, which accounted for 20%–35% by volume of CLSM. If a finer material was used, a higher content of water was used due to the effect of a specific surface area.

For the CLSM, the trial and error method was mainly adopted when carrying out the proportioning design. The recommended amount of CFA as cement-replaced material was used in a large range. The design was based on fluidity requirements. The amount of fine aggregate filled after the cement, CFA, water, and air content were determined. For the filling, the general dosage was about 1500–1800 kg/m³, explained in Appendix A.4 of the National Standards of the Republic of China (CNS) 15865 in Taiwan. A typical CLSM contains about 5%–10% of Portland cement, and soil containing an upper limit of 20% can be used. In total, three batches of CFA and six mixing series were used to test the materials before and after the use of CFA and GGBS. The compression values that were obtained after mixing were too high to meet the compression value requirements of the CLSM specification.

In the proportioning design and concrete trial mixing, the proportioning design research that complied with the CLSM and self-consolidating concrete (SCC) regulations was completed. The proportioning design is shown in Table 6 and Table 7. Methylcellulose was used as the adhesive and the aluminate-based accelerators and carboxylic acid-based superplasticizer were used in the mixtures. SCC was defined according to ACI 237 [40] as highly self-consolidating ability with non-segregation. Without any mechanical tamping, the concrete can be filled into the formwork and wrapped with reinforcement in the filling pouring area. Table 8 is the evaluated properties of SCC self-consolidating capacity grade, which is the evaluation of CLSM performance in this study.

The fluidities for the strength activity index testing of the Ordinary Portland Cement (OPC), CFA, and GGBS specimens are shown in Table 9. According to the standard of ASTM C311 [41], the fluidity of the OPC was 17 cm and the water amount was 242 g, as marked in Table 5. The CFA and GGBS had the same fluidity as the OPC group. The final amounts of water used were 252 and 240 g, respectively. It could be known that when the proportion of CFA was used, if the same fluidity needed to be maintained, either the water consumption had to be increased or a water reducing agent had to be added at the same time.

The results of the compressive test of the strength activity index at 7 and 28 days are shown in Table 10. It is worth knowing that the compressive strength of the CFA at 7 days was the same as that of the OPC group, but the compressive strength of the CFA was only 29% at 28 days. The improvement showed that the strength could be maintained in the early stage, and it had lower strength in the later age, which could meet the requirement of CLSM containing CFA.

As shown in Table 10, the calculated strength activity index of the GGBS was similar to the results of the CFA test at seven days, but it increased slightly to 112% at the age of 28 days. Considering the increase in the amount of GGBS for fluidity, it should be noted that the compressive strength was increased at the same time to avoid exceeding the upper limit of the CLSM requirements. The strength activity index of CFA decreases as the number of days increases may be because the sulphide of CFA has a micro-expansion effect on cement hydrates [42], resulting in a slight decrease in compressive strength, but it is still 75% higher than the ASTM C311 requirement [41].

The freshly mixed properties according to Table 7 are shown in Table 11. These properties indicated that the T10 specimens could meet the requirements of the fresh properties of the CLSM. The T10 specimens could meet the requirements of flowability (50.0 cm) and tube fluidity (20.0 cm). In addition, the mixture of T10 specimens had higher CFA to replace large quantities of cement in the CLSM. According to the strength of each recommended load in Table 1, the recommended lower limit of the compressive strength was 0.35 MPa, which showed good compaction for the CLSM. The suitability for the load application of the T10 specimens using a ball drop test could reach its performance 7.6 cm. After the actual measurement, the material could withstand the weight of a 75 kg heavy plate without deformation after 21 hours of mixing. According to the standard penetration tests, the cemented material could reach its recommended value of 50 psi. Therefore, mould removal and curing could be performed within 24 hours without damage to the concrete. According to the results of the compressive strength at seven days, the strengths in the indoor dry curing, saturated lime water curing, and atmospheric exposure curing methods were about 2.45–3.92 MPa. This showed that the standard penetration test resistance of the cementitious material in accordance with ASTM C403 met the CLSM requirement, as shown in Fig. 6. According to Table 8, T5 and T10 specimens meet the requirements of SCC grade 3 of V type test, box test and flowability. The best mix proportion of CLSM is 11% cement, 34% CFA and 55% GGBS, which can achieve the self-consolidating ability.

Fig. 6

In this test, a 100% replacement of the cement in the paste specimens was used to confirm the feasibility of the cementing materials in the CLSM. The mixing design is shown in Table 6. A water/cement ratio of 0.70 and five proportions were used for the paste specimens. After mixing, the specimens were demoulded for four days (as shown in Fig. 7) and then cured in the saturated lime water for seven days. This indicated that the F100 specimens could be demoulded, but the strength was extremely low, and the F100 specimens became a loose structure after being placed in the curing tank.

Fig. 7

The demoulding of the F0 specimens (100% GGBS replacement of cement), as shown in Fig. 8, failed to have the ability to coagulate. The compressive strength at seven days was as illustrated in Table 12. The results showed that the compressive strength increased with the increase of the GGBS and the compressive strength of the F25 specimens increased by up to 9.80 MPa. This suggested that the CLSM specimens should use the higher water/cement ratio or the higher amount of CFA to meet the requirements of the CLSM standard.

Fig. 8

The compressive strength results are summarized in Table 13 and Fig. 9. The compressive strength was between 2.84–6.18 MPa, which was the compressive strength of the CLSM. The compressive strength at 7 days was about 2.94 MPa. There were between 4.41–4.90 MPa at 14 days and between 5.39–5.88 MPa at 28 days. The curing methods were mainly divided into indoor dry curing (E), water curing (W), and atmospheric exposure (A). For indoor dry curing conditions, the test specimen was mainly placed in a dry indoor space to prevent water and gas intrusion. For atmospheric exposure, the site was mainly simulated after the grouting is completed, and the site was put directly into the atmosphere for curing operations. For saturated lime water curing, the test specimen was mainly placed in a lime-filled area. For different curing methods, the maximum differences between the testing and control groups were 1.78 MPa for 7 days, 2.16 MPa for 14 says, and 1.27 MPa for 28 days. At the age of 28 days, the strength growth tended to slow down and all three curing approaches neared 5.88 MPa.

Fig. 9

The specimens at 7 days had the highest indoor dry curing and the lowest curing was in the atmospheric exposure group. Compared with the weather observation data (Table 14), the precipitation was large during the 1-day to 10-day curing age period. Although the temperature was high, the specimen was in a humid state for a long time. From 11 to 28 days, the test specimens were in a dry and high-temperature state. In the atmospheric exposure group, the values were lower than the other two groups at 7 days, and they had slowed down at fourteen days. It was inferred that the test specimens may have been dry and they may have had high temperatures. This made the intensity increase more quickly.

The accelerated chloride migration test (ACMT) according to the results of the initial current generated using 30 V is shown in Table 15. The current value generated by the 14-day age was about 70–140 mA, which was low when compared with the current value generated by the 7-day age (150–220 mA). The results showed that there was a downward trend in the diffusion coefficient at 7 days for atmospheric exposure, and the chloride ion diffusion coefficient of the specimens for indoor dry curing at 28 days was the highest than the other curing methods. The chloride diffusion coefficient decreased as the increase of curing age increased, which was consistent with the compressive strength. For indoor dry curing at the age of 28 days, the chloride diffusion coefficient was the lowest of the three curing methods, which was in line with the trend of higher compressive strength. The specimens that were cured in water could be reflected in the better growth of the capillary pores, resulting in a lower diffusion coefficient. The specimens in atmospheric exposure conditions were also reflected in the low diffusion coefficient due to the random environmental dry and wet exposure. To keep the durability and to protect the infrastructures, the curing condition of CLSM in site construction should be considered.

The microscopic property observations and EDS analysis at 7 days for the atmospheric exposure are shown in Figs. 10 and 11. In the test results, the size of the main hydrated products was 2 μm. The characteristics could be judged to be those of ettringite based on the length and shape of the needle, and a small amount of C-S-H or C-A-S-H colloid could be observed, which as verified by EDS analysis. For 5000 magnification observations, the fracture of 1–5 μm could still be observed due to the random environmental dry and wet exposure. The EDS test results were mainly composed of oxygen, carbon, silicon, calcium, and aluminium. It was confirmed that the hydrated products might be C-S-H, C-A-S-H, ettringite, and monosulfoaluminate.

Fig. 10

Fig. 11

The microscopic property observations and EDS analysis at 28 days under atmospheric exposure are shown in Figs. 12 and 13. The observations and the analysis indicated that the filled hydrated products made the surface of the specimen denser, which was consistent with the results of the compressive strength and diffusion coefficients. The SEM photos at 28 days under the water curing and indoor dry curing are shown in Fig. 14. These photos indicated that there was evidence of further hydration with the pronounced formation of ettringite with needle-shaped structures and that a smooth surface was observed in the micro-structures, especially for the specimens under the water curing conditions. The surface of the specimens with indoor dry curing could be seen to have more failure surfaces, an interfacial transition zone, and larger pores. It was also confirmed that the use of CFA and GGBS to replace large quantities of cement in the CLSM could be cemented and hydrated at a 28 day curing age under various curing conditions.

Fig. 12

Fig. 13

Fig. 14

MIP is an analytical index for capillary pores and gel pores. The distributions of the cumulative pore size diagrams are shown in Fig. 15. According to the experimental results, the pores sizes of the specimens at all ages were mainly distributed between 10 nm and 4000 nm. The pore size distribution curves for three conditions were similar and the cumulative intrusions of the specimens were approximate between 0.13 ml/g and 0.15 ml/g. Fig. 16 demonstrates the relationship between a gel pore and a capillary pore for three curing conditions. The pores in hydrated cement paste could be categorized as either capillary (10–10000 nm diameter) or gel pores (3–10 nm diameter). Capillary pores are usually determined in cement-based composites using their compressive strength, permeability, and diffusivity. Gel pores are considered an intrinsic part of calcium-silicate-hydrate (C-SH), which mainly affects concrete shrinkage and creep. A capillary pore system consists of many interconnected large pores, through which water, ions, carbon dioxide, and oxygen penetrate or diffuse into a rebar surface, deteriorate protective passive films, and cause corrosion. The capillary pores significantly increased in wet conditions. This may have been due to the leaching of calcium ions. For both the air and environmental conditions, the specimens had much fewer capillary pores than that for wet conditions. However, the specimens for three conditions still had a lower amount of gel pores due to the higher w/c ratio of 0.85.

Fig. 15

Fig. 16

The effects of the fresh properties in this study are summarized in Table 16. The CFA could effectively reduce and maintain a specific strength for the CLSM and meet the compressive performance requirements of the CLSM.

The optimal mixtures that were obtained through the test results are summarized in Table 17. Table 15 shows that the mechanical properties had the lowest strength at 7 days from the environmental simulation. Through chloride ion diffusion tests and microscopic property observations, it was determined that the water absorption capacity was higher than the water absorption capacities of the water curing and indoor dry curing were, the group was high, and the chloride ion transmission rate was high. According to the observation of the microscopic properties, the width of pore cracks was about 5 μm, which is 1 to 2 times that of the other groups.