Mechanical properties of an improved grout for cementitious precast beam-column joints

This study proposed a new formulation for an improved grout with superior early strength and ultra-high cured strength; it was designed on the basis of the theory of closest packing. Orthogonal experiments were conducted to analyse the effects of four factors, silica powder content, water reducer content, steel fibre content, and water-cement ratio, on the flowability, compressive strength, and compactness of grout. The criteria for determining whether the grout met the requirements for Code included initial flowability greater than 300 mm, flowability more than 260 mm after 30 min, and compressive strength more than 60 MPa after 12 h of standard curing. The results showed that the performance of the grout satisfied specified requirements for Code with small internal voids and acceptable durability. After the ratio of raw materials was optimized, The grout sleeve test showed that the failure occurred in the steel bars outside the sleeve, and no grout pulling, slipping, splitting, or other behaviour occurred within the sleeve, which meant that the specimens met the design requirements. The development of this grout will greatly reduce construction time for Code and improve the quality of connections in prefabricated components. The results of this study will provide a reference for the design and development of new grouts in the future.


Introduction
Grouts have the characteristics of good fluidity and high strength [1]. They are often used as repair materials to fill gaps of concrete structures for repair and reinforcement [2,3]. They are also used to grout sleeves to connect prefabricated components [4,5]. In addition to the formed quality of the components, the quality of the splicing between the prefabricated components is also related to the positioning accuracy, reliability of the connecting material and other factors [6][7][8]. The key to the overall reliability of prefabricated structures is the connection node technology [9]. However, with the continuous upgrading of engineering technology, the requirements for cement-based grouts for construction projects are becoming increasingly stringent [10]. Traditional Portland cement-based grouts have long setting times and low early strengths, which limit their application in special scenarios such as airport runways, highways, and prefabricated buildings in very cold areas [11][12][13]. Therefore, the development of grouts with excellent performance has become a hot issue for scholars to study [14,15].
Grouts are generally composed of inorganic cementitious materials, mineral admixtures and admixtures, and mixing with water is required [16,17]. A previous study [16] found that the contents of calcium aluminate cement (CAC) and calcium sulfate bone cement (CSC) provided the greatest increase in early strength, but they were unfavourable for later strength. The results of another study [17] showed that fly ash could improve the fluidity of grout and its strength in the middle and late stages, but they were unfavourable to early strength. In a third study [18], a cement-based grout with early strength for semi-flexible pavement was developed by mixing different proportions of cement with fiber. In a fourth study [19], grout with low shrinkage and early strength was prepared using a ternary composite system. The influences of preparation parameters, such as the cement compounding ratio, fineness of the composite admixture and gypsum content, on the performance of the grout were assessed. In yet another study [20], 87 centre pull-out specimens were designed by using the control variable method to explore the influences of compressive strength, steel strength grade, steel diameter, confined stirrup and bond length on bonding between steel bars and grout. The above studies did not achieve a formulation for a grout with closest packing of the solid materials in the system. The compressive strength was low after curing for 12 h and for 28 days. How to develop grout to better meet the needs of superior early strength and sufficient construction time is worthy of further study. The later strength of conventional ultra-high performance concrete (UHPC) is high, but according to current market research, its fluidity cannot meet the requirements, and the strength within 12 h is not high enough, to meet requirements for applications such as rescue and plugging.
This study applied compact packing theory [21][22][23][24][25]. A new grout based on compound cement with ultraearly-strength and ultra-high strength was developed, and performance indices such as fluidity, compressive strength, hydration reaction and compactness were controlled. The mechanism for the influence of the hydration reaction of raw materials, such as steel fibre, fine aggregate, water reducing agent and silicon powder, on the performance of the new grouting material was revealed. The optimal mix ratio was determined to provide a reliable reference for the design and production of grouting materials.

Materials
The raw materials used in the test included P.II.52.5R Portland cement (PC) and type-52.5 fast hardening sulfoaluminate cement (FHSC). The mineral admixtures included fly ash, fumed silica and glass beads. The fine aggregate was quartz sand, and the additives were PCA-I polycarboxylate superplasticizer. The water reduction rate was 35% and a U-type expansion agent (UEA) was used.
(1) Composite cement base material properties Before selecting the final compound cement, four cement mortar tests were conducted to determine the properties of each cement. Figure 1 shows that the compressive strength of PC was better in the later stage, but the compressive strength increased slowly in the early stage, and the flexural strength developed slowly. The compressive strength of the FHSC rapidly increased in the early stage, then slightly decreased in the later stage. The FHSC had good flexural strength and stable performance in the later stage. The water consumption of the two cements at standard consistency was close, at approximately 30%. The specific parameters of the two cements are shown in figure 1, and the physical form is shown in figures 2(h) and (g).
(2) Admixture and fine aggregate The main admixture and fine aggregate parameters are shown in table 1, and the particle size distributions of the admixture and cement are shown in figure 3.

2.2.
Optimization of the ratio of raw materials used in grout 2.2.1. Theoretical analysis Based on the theory of particle close packing, the modified Andreasen and Andersen model was used to design the grout. Formula (1) was used to calculate the particle size distribution of the most closely packed state. The constraints of each component were determined by referring to the research in the literature [8]. By constructing the Lagrange function and using the effective set method, the optimal solution satisfying the constraints was obtained. The solution was converted into the mix ratio of cementitious material and aggregate. In the formula, D is the particle size; P(D) is the cumulative fraction of particles smaller than D (%); D max is the maximum particle size (mm) of the grouting material; D min is the minimum particle size (mm); and q is the partition coefficient. In this study, the value of q was fixed at 0.23 [21], and the particle size distribution is shown in figure 3.
The maximum wet packing density control test method was used to determine the ratio and substitution amounts of cement, mineral admixture and quartz sand in the grout, and the relationship between the amounts of various fillers and the maximum wet packing density of the system was established. The specific calculation method is shown in Formula 2: In the formula, j indicates the packing density, and V represents the mould volume. x represents different cementitious maerials, and r , w r s and r x are the densities of water, quartz sand and cementitious material,  respectively. R , w R s and R x represent the ratio of water, quartz sand and cementitious material to slurry content, and M max refers to the maximum mass of the slurry. The optimal ratios of the raw materials were obtained by theoretical calculation, as shown in figure 4.

Orthogonal test optimization ratio
Through theoretical calculation, the fixed cement-sand ratio in this experiment was obtained: 900:750, FHSC: PC = 675:225, and fly ash:glass beads = 30:30. Factors A, B, C and D were each divided into three levels, 1, 2 and 3. The orthogonal experiment mainly analysed the influence of four factors on the mobility and compressive strength of the grouting material, and there were no interactions between raw materials in the experiment. Therefore, the L9 (34) orthogonal table was selected, as shown in table 2.    The range size obtained from the test determined the priorities of different factors. The range or the significance level of the F test was used as an indicator. The larger the value was, the greater the effect of the factor on the material properties [22]. According to the analysis, the amount of water reducing agent had the most significant effect on the mobility, and the steel fibre content had the least significant effect on the mobility of the slurry. Therefore, the order of influence on the mobility of grout was B (water-reducing agent content) > D (water-cement ratio) > A (fumed silica content) > C (steel fibre content). The analysis showed that the four factors had the following effects on the strength: 32%, 14.5%, 38% and 14.9%, The order of influence on the 12h strength of grout was C (steel fibre content) > A (fumed silica content) > D (water-binder ratio) ≈B (water reducer content).

Mobility test
Mobility is an important index of grouting material. The analysis of figure 9(a) showed that the mobility of the slurry increased gradually with increasing water-binder ratio. With the increase in fumed silica content, the      proportions of small particles in the grout and the specific surface area also increased. These small particles adsorbed more water and water-reducing agent, thus reducing the mobility of grout.
According to the analysis in figure 9(b), the mobility increased with increasing content of water reducer. This was because the hydrophobic groups in the water reducer directionally adsorbed on the cement surface. The hydrophilic groups were oriented to the aqueous solution. Thus, the water reducer adsorbed and formed a single-molecule or multi-molecule film with the same charge on the surface of each cement particle. Under the action of electrostatic repulsion, the cement particles repelled each other and dispersed, and their mobility increased. The influence of steel fibre content on slurry mobility was the same for different contents.

Compressive strength test
The compressive strength test of grouting material referred to the national standard and experimental specification. Grout with measured fluidity was added to the triple mould and smoothed. After samples were prepared, the compressive strength test was conducted, and the compressive strength was measured at 12 h, 1 d, 3 d and 28 d. The compressive strength was calculated according to equation (3): where F c is the failure load on the side of the test block (N); R c is the compressive strength (MPa); and A is the compressive test area (mm 2 ). The greater the compressive strength of the grout was, the greater the connection force between the steel bar and the grout, and the better the connection effect [16][17][18][19][20]. Figure 10 shows that the content of steel fibre and silica fume improved the 12-h compressive strength of the grout, figure 11 shows the failure mode of the test prisms(cubes). When the water-binder ratio was in the range of 0.18-0.2, the water requirement for cement hydration was greater than the total amount of mixed water, and the compressive strength increased with increasing water content. However, when the amount of water continued to increase, excess water separated from the solids, many connected pores formed inside the slurry, and the compressive strength decreased.
Through orthogonal experimental analysis, the mix ratio shown in figure 12 was obtained. Based on the comprehensive analysis of the influence of four factors on fluidity and 12-h compressive strength, the basic mix ratio that satisfied the two requirements was obtained. The initial fluidity of the grouting material mixed under the basic ratio did not reach 300 mm, and the 12-h compressive strength was greater than 60 MPa. To further improve the fluidity of the slurry, the amount of ground fine beads was increased, and the amount of mixed water was increased to 207 g. Considering the good bonding performance between the grout and steel sleeve, a UEA expansion agent was added to increase the vertical expansion rate of the grout so that the grout and the sleeve were closely connected.

Vertical expansion rate test
The vertical expansion rate of the grout for the basic mix ratio was further verified. Three groups of 9 cubes were made with dimensions 100´100´100 mm. The test results showed that the expansion rate of the grout was 0.001% at 3 h and 0.008% at 24 h, and the expansion performance of the grout did not meet the requirements of the specification. The analysis showed that the large proportion of mineral admixtures in the slurry led to an increase in the shrinkage of the grout and offset the expansion of the grout. The mix ratio was further optimized. On the premise of not changing the fluidity and strength, the amount of expansive agent was increased from 0.8% to 1.3%. The optimized mix is shown in figure 13.
The vertical expansion rate of the newly optimized mix ratio was tested. According to table 5, the expansion rate of the grout at 3 h was 0.13%, the vertical expansion rate at 24 h was 0.155%, and the difference between the expansion rate at 24 h and 3 h was 0.025%. The results meet the requirements of the specification.

Self-drying shrinkage test
The optimized grout had a good micro-expansion rate, and the expansion effect of the grout offsets its shrinkage to a certain extent. At the same time, the shrinkage rate of the grouting material for 28 days was very small (as shown in figure 14). The shrinkage rates of the three groups were 0.005%, 0.01% and 0.01%, which met the requirements of the specification at < 0.045%. The measured values of each group are shown in table 6.

Final ratio verification
The mobility experiment, 12-h compressive strength experiment, 3-d compressive strength experiment, 7-d compressive strength experiment, and measurements of 28-d compressive strength, 28-d shrinkage rate, bleeding rate and vertical expansion rate were performed and assessed to verify the performance of the final ratio. The average values of the three groups of results are shown in table 7. Table 7 shows that the initial flow value of the optimized ultra-high strength grout reached more than 300 mm, the flow value after 30 min reached more than 260 mm, the compressive strength at 12 h reached more than 60 MPa, and the compressive strength at 28 days was 102 Mpa. The expansibility and dry shrinkage met the requirements of the specification. It was necessary to further verify whether the connection between the grouting material, sleeve and steel bar was reliable. The common failure modes of the sleeve are steel bar yielding and breaking, sleeve breaking, grout being pulled out and slurry splitting failure.

Sleeve and reinforcement parameters
According to the inspection regulations of grouted sleeve connectors in JGJ355-2015 [26], steel bars with diameters of 16, 20 and 25 mm were selected for the test, and the mechanical properties of sleeve connectors were tested. A full grouted sleeve with ductile iron as the raw material was selected for the test. The sleeve models were GTZQ4-16, GTZQ4-20 and GTZQ4-25, as shown in figure 15. The detailed parameters of the sleeve are shown in table 8. According to JGJ355-2015 [26], the effective anchorage length in the table should not be less than 8 d (where d is the diameter of the connecting steel bar).

Specimen preparation and test loading
The production of the reinforced sleeve grouting specimen was carried out at the material laboratory of the Jinling Institute of Technology. A manual grouting instrument was used. The construction process was as  follows: grout was prepared, grout was pressed from the sleeve grouting port, and the rubber seal ring was plugged when the grout overflowed. The uniaxial tensile test of the grouted sleeve connector was conducted at the structural laboratory of the Jinling Institute of Technology. The load was applied using a 1000 kN electronic hydraulic universal testing machine, as shown in figure 17, and the loading rate was 1 N mm −2 s −1 . According to the provisions of JGJ107-2016 'Technical specification for mechanical connection of steel bars', uniaxial tension testing was conducted. The single tensile loading system was 0 → 0.6 fyk → 0 (measuring residual deformation) → Maximum tension (recorded ultimate tensile strength) → Destruction (determination of total elongation under maximum force). Figure 18 shows the failure mode of the sleeve grouting connector. The yield strength, tensile strength and final failure mode of the sleeve grouting connector were recorded, as shown in figure 15.

Test results
The test results showed that the yield strength of the steel sleeve grouted connection joint was not less than the standard value of the yield strength of the connecting steel bar. The bond strength between the grouting material and the steel bar was greater than the yield strength of the steel bar itself. There was no slip failure between the slurry and the steel bar, and the damage broke outside the joint.
6. Microscopic analysis 6.1. Analysis of grout test block section Figure 19 shows a cross section of the sleeve grout after hardening. The steel fibres were evenly distributed in the cross section. The steel fibre was tightly bonded to the grouting material, and the hydration of the grouting material test block was uniform. With the increase in the critical period, the ettringite formed by the hydration of sulfoaluminate cement overlapped the hydrated calcium silicate formed by the hydration of Portland cement. It filled the gap and improved the compactness and bonding force between the grout and steel fibres so that the steel fibres were not pulled out, which had the effects of strengthening and toughening the specimens.   figure 20), a large amount of hydrated calcium silicate gel was produced inside the grout. The secondary hydration reaction of Ca(OH) 2 with fly ash and fumed silica produced C-S-H gel, and the surfaces of fly ash and fumed silica were wrapped by C-S-H gel. C-S-H gel filled the pores and cracks of the unhydrated cement particles, and the overall structure was relatively dense.

Chloride ion penetration test
The chloride permeability test was conducted by x-ray photoelectron spectroscopy (XPS), as shown in figure 21. X-ray photoelectron spectroscopy is also called electron spectroscopy for chemical analysis, and photoelectrons  are generated by irradiating the surface of the sample with monochromatic x-rays. The species of elements in the sample can be deduced by measuring the electron binding energies of the inner atomic layers, and by analysing the chemical shift of the binding energy, evidence of valence changes of elements or binding to atoms with different electronegativities can be found. The XPS results are shown in figure 21. The sample surface contained chloride ions, and the chloride concentration was calculated according to equation (4). The results are shown in table 9. The test results showed that the chloride ion permeability of the grout was 0.009 0.03, which med the requirements of the specification.
2. Fumed silica improved the 12 h strength of sulfoaluminate cement grout. When the ratio of sulfoaluminate cement to Portland cement reached 3:1, the fluidity of the grouting material was optimal, and the initial fluidity exceeded 300 mm. The fluidity exceeded 260 mm after 30 min and was less dependent on the water     MPa in 28 days. The vertical expansion rate at 3 h was 0.0613%, the difference between 24 h and 3 h was 0.025%, and the 28-day shrinkage rate was 0.0083%.
3. The sleeve grouting process was smooth, and the strength of the prepared connector exceeded the yield strength of the steel bar, which met the requirements. 4. Orthogonal experiments showed that the main factors that determined fluidity were the amount of waterreducing agent and the water-binder ratio. For the 12 h compressive strength, the orthogonal analysis showed that the main factors affecting the strength were steel fibre content and fumed silica content. The most effective way to improve the strength at 12 h was to increase the content of steel fibre, ensure the uniform distribution of steel fibre in grout, increase the content of fumed silica and improve the purity of fumed silica. The addition of the UEA expansion agent increased the extrusion force between the grout and steel sleeve and improved the connection force of the sleeve connector.
5. Observations made with scanning electron microscopy showed that the hydration reaction of the grout was sufficient. The internal gaps were filled with C-S-H gel, and the gaps were dense. The chloride ion content of the grout met the requirements of the specification.

Data availability statement
All data that support the findings of this study are included within the article (and any supplementary files).