The origin of the recycled coarse aggregate used in this study was the concrete waste from structures in Harbin (the capital city of Heilongjiang Province in northeastern China, an area with a very cold climate), where it was subject to countless freeze-thaw cycles prior to crushing, grinding, screening and processing. Since the components in the waste concrete were laboratory-processed into RCA, they contained few impurities and no hazardous materials. The resulting aggregate proved to be apt for use in the preparation of high-strength concrete (12). The micro-cracks that appeared on the surface of the aggregate under the heavy pressure applied during processing detracted from the mechanical performance of the material. To ensure the quality of high-strength recycled concrete, only RCA whose abrasion test results are comparable to the findings for natural aggregate should be used. In the present study, abrasion tests were conducted on coarse aggregate sourced from 40‑MPa concrete, 60‑MPa concrete and concrete of unknown compressive strength, as well as from natural stone. The results are shown in Figure 1.

At 32.1%, the mean abrasion loss in RCA was greater than in the natural material (24.6%). Abrasion loss was 28.5% in 60‑MPa concrete aggregate and 33.4% in the material processed from 40‑MPa concrete waste. In other words, the higher the pre-crushed concrete strength, the lower the abrasion loss in the aggregate. The highest abrasion loss, 34.3%, was recorded for the waste concrete of unknown strength. That finding was attributed to the fact that it had been in service longest and the initial quality was the lowest. Since abrasion loss was under 50% in the three types of RCA studied (Figure 1), all three were regarded as apt for use in structural concrete (13). The recycled coarse aggregate sourced from crushed 60‑MPa concrete was chosen for preparing the high-strength recycled concrete used here, inasmuch as its abrasion loss was lower than the other two types of RCA tested.

The data on physical properties given in Table 1 confirm the higher quality of natural than recycled coarse aggregate.

Cement: Swan P·O 42.5 ordinary portland cement supplied by the Yatai Group cement plant at Harbin.

Fine aggregate: natural standardised sand (14) with an apparent density of 2640 kg/m³.

Natural coarse aggregate: crushed stone, ≤31.5 mm.

Water: runningwater.

The natural and recycled aggregate were mixed at replacement ratios of 0, 30, 50, 80 and 100%. Four water/cement (w/c) ratios were used to prepare the concrete: 0.30, 0.35, 0.40 and 0.45. The batching design called for 195 kg/m³ of water and 1130 kg/m³ of coarse aggregate, while the fine aggregate content was adjusted in keeping with the w/c ratio. The mix was enhanced with an LQ-polycarboxylate high-range water-reducing admixture dosed at 1 wt% of the cement. The recycled concrete mix was moulded into 150‑mm³ cubic specimens and cured under standard conditions for 28 days, when compressive strength was measured as described in Chinese specification GB/T50081-2002 Standard for test method of mechanical properties on ordinary concrete (15). Table 2 lists the concrete batching used and the 28‑day compressive strength values for the specimens. As the specimens failed at around 70 MPa in all cases, this was the ultimate limit strength defined for high-strength recycled concrete.

The design strength established for the recycled concrete studied was 50–70 MPa. The concrete was batched to a design for concrete with natural aggregate (16) compliant with Chinese standard JGJ 55 - 2000 Mix proportion design of ordinary concrete rules (17) (Table 2). The graph in Figure 2 plots compressive strength against the w/c ratio in the recycled concrete. Adjusting the w/c ratio during concrete preparation to attain the design strength minimised the effect of the RCA on that parameter. The procedure yielded satisfactory results for the concretes in which the RCA accounted for 50% or less of the total. These trials also showed that in high-strength recycled concrete, the suitable range for the w/c ratio is 0.30 to 0.40.

Multiple samples taken from different parts of the high-strength recycled concrete were tested for slump 5 minutes after mixing. The graph in Figure 3 relates slump values to the aggregate replacement ratio.

Concrete slump gradually declined with rising replacement ratio, very significantly when the natural aggregate was wholly replaced by the recycled material.

The explanation for that finding is that the recycled coarse aggregate surface was rougher and more cracked than the crushed stone surface, inducing higher water absorption in the former. The absorption of much of the mixing water by the RCA reduced the amount actually available for mixing per se. The result was greater friction during mixing and pouring and consequently a smaller slump.

Workability of high-strength recycled concrete can be improved by adopting two measures. During processing, the cement mortar attached to the recycled coarse aggregate should be removed to reduce absorption and impurities. Workability can also be enhanced by blending a suitable dose of water-reducing admixture with the other components.

Cylindrical specimens 100 mm in diameter by 300 mm high were tested for creep. The 28-day specimens were compressed on a test frame at a constant pressure equal to 40% of the compressive strength of the cubic specimens. Shrinkage was tested on prismatic specimens measuring 75‑mm square by 300 mm deep, stored at ambient temperature and 55% relative humidity. Specimen shrinkage was determined with an unrestrained method, described, like the procedure for measuring creep, in Chinese standard GB/T 50082-2009 Standard for test methods of long-term performance and durability of ordinary concrete (18). The elastic modulus, creep and shrinkage values recorded are given in Table 3.

Although the elastic moduli for the concretes with 0 and 30% RCA were nearly identical, at higher replacement ratios the values declined steadily. Shrinkage and creep were also affected by the replacement ratio, for two reasons. Firstly, the downward adjustment to the w/c ratio to ensure the concretes prepared would attain the design strength raised shrinkage and creep in the resulting material. Secondly, the mortar clinging to the RCA also intensified creep and shrinkage.

Durability in a material is its ability to remain intact in the long run. The greater the durability, the longer the service life of the material at issue. The cement mortar attached to the RCA studied increased its void ratio and water sorptivity substantially, inducing the formation of micro-cracks on the surface of the aggregate, with the concomitant impact on concrete durability.