Keywords

1 Introduction

Alum sludge is a typical by-product of the drinking water industry. The heterogeneous sludge waste is formed when the aluminum-based coagulant is combined with suspended solids, dissolved colloids, organic matter, and microorganisms in raw water. It is estimated that global sludge production has exceeded 10,000 tonnes per day, and the rapid population growth and economic development may result in a significant increase in its amount in future decades [1]. In Australia, most sludge is disposed of at landfill sites (see Fig. 1), which may cause severe environmental issues because of land wastage and secondary pollution. In view of the transition toward a circular economy, vast-available sludge should be considered as a resource with the potential to be valorized instead of a waste.

Fig. 1
Two photographs of the landfill sites in South Australia with sludge deposits.

Sludge landfill sites in South Australia

Full size image

Most alum sludge has 20–63 wt% Al2O3 and 17–41 wt% SiO2 [2]. Its aluminosilicate nature makes sludge can be recycled as cement replacement, proposing a possible solution to reuse sludge in large quantities. Also, reducing cement usage may contribute to achieving the target of carbon neutrality. Some previous studies have already investigated the feasibility of alum sludge as cement replacement in concrete products [3]. In general, raw alum sludge exhibits no pozzolanic reaction, and the high organic matter in sludge may hinder the cement hydration, resulting in deteriorated mechanical and durability performance of concrete products [4]. Treating sludge with high temperatures, ranging from 600 ℃ to 800 ℃, can efficiently improve the pozzolanic reactivity of sludge due to the fact that crystal phases of silicon and aluminum were dehydroxylated to form disordered phases with high reactivity [5]. However, the optimum temperature to activate sludge activity is still controversial. For the performance of sludge-derived concrete products, a moderate cement replacement (e.g., 10%) with calcined sludge is feasible without compromising the mechanical and durability properties [6,7,8].

Based on the above literature, the ideal temperature (between 600 and 800 ℃) to activate the pozzolanic reactivity of alum sludge needs to be clarified. The reaction degree of sludge, composition of hydration products, and chemo-mechanical properties of sludge-cement binders need to be in-depth discussed. Therefore, in this study, the pozzolanic reactivity of alum sludge under different temperatures was assessed by the strength index test. The reaction degree of sludge was determined by the selective dissolution method. The chemical composition of sludge-cement binders was assessed by x-ray diffraction (XRD) and thermogravimetric analysis (TGA). The chemo-mechanical properties of blended binders, e.g., indentation hardness and modulus, were investigated by an advanced nanoindentation method. Final, the strength of concrete blocks with cement replaced with sludge at weight percentages of 0%, 10%, 20%, and 30% was studied.

2 Methodology

2.1 Materials

Alum sludge is collected from a local water treatment plant in South Australia. First, raw sludge was crushed and milled to less than 75 µm. Then, milled sludge was calcined at 600℃, 700℃, and 800℃, respectively. General-purpose cement is used according to AS 3972. Fine aggregates and coarse aggregates used to manufacture concrete blocks (CB) were concrete sand and crushed limestone. The particle size distribution of the sludge under different temperatures was determined by Mastersizer 3000. The chemical composition of sludge and cement was investigated by X-ray fluorescence.

2.2 Sample Preparation

CB were manufactured with the dry mix method, and the detailed cast procedures are described in a previous study [9]. The pastes were also cast in plastic molds with dimensions of 10 × 10 × 10 mm3, which contained the same water to cement ratio and sludge content as CB, assisting in the study of hydration products by eliminating aggregate interference. Table 1 shows the mix design of the CB.

Table 1 Mixture design of blocks (kg/m3)
Full size table

2.3 Experimental Methods

The pozzolanic reactivity of sludge under different temperatures was determined by the strength index (SAI) test according to ASTM C311. The selective dissolution method was used to assess the reaction degree of sludge based on a mixture of ethylenediaminetetraacetic acid–triethanolamine–NaOH [10]. The compressive strength of the CB was measured according to AS 4456.4. The load was applied at a constant rate of 2 kN/s without any shock until failure.

The hydration products in the cement-sludge binder matrix were characterized by XRD and TGA, and the hydration reaction of samples was stopped by ethanol immersion. An advanced nanoindentation technology (coupling conventional statistic nanoindentation and chemical mapping) was used to characterize the in-situ chemo-mechanical properties. Detailed sample preparation procedures and the data analysis method for the nanoindentation test are described in our previous study [10].

3 Results and Discussion

3.1 Material Characterization

Table 2 shows the chemical composition and the particle size distribution of cement and sludge under different calcination temperatures. The main components in raw sludge were Al2O3, SiO2, and organic matter, and minor contents of Fe2O3, CaO, and K2O were also observed. After calcination, most organic matter content was eliminated, resulting in an increase in the proportions of Al2O3 and SiO2. Because the chemical composition of sludge calcined between 600℃ and 800℃ was similar, only the 800℃-treated one is shown in Table 2. It is worth noting that the sum of the Al2O3, SiO2, and Fe2O3 content in calcined sludge was higher than 70%, which satisfied the composition requirement of natural pozzolan material based on ASTM C618. In Table 3, the particle size distribution of calcined sludge and cement is shown. The Blaine fineness of sludge decreased with increasing calcination temperature, which could be explained by the fact that dehydroxylated particles agglomerate under high temperatures to produce new porous grains, especially fine particles [11]. The cement exhibited finer particle size but comparable Blaine fineness to that of calcined sludge.

Table 2 Chemical composition (LOI, loss on ignition)
Full size table
Table 3 Particle size distribution
Full size table

The results of the SAI test are shown in Table 4. After calcination at 700 ℃ and 800 ℃, sludge exhibited a satisfactory pozzolanic reactivity (SAI ≥ 75%). In contrast, the 600 ℃-treated sludge could not be used as a pozzolan material. Compared with 700 ℃, 800 ℃ is a better temperature to activate the pozzolanic reactivity of sludge, in which the SAI value is up to 113.60%. Therefore, 800 ℃-treated sludge was used to cast the paste and CB samples.

Table 4 SAI test of sludge under different calcination temperatures (SAI, strength index)
Full size table

3.2 Reaction Degree and Hydration Mechanism of Sludge

The selective dissolution method determined the reaction degree of sludge in the blended binders. The paste samples containing 10% sludge by weight exhibited the highest reaction degree, up to 39.0%. Further increasing the sludge content to 20% and 30%, decreased the reaction degree to 25.2% and 24.8%, respectively. The decrease in reaction degree could be attributed to a lack of sufficient portlandite. Figure 2 shows the TGA and XRD. In Fig. 2a, c, a significant endothermic peak occurred ≈ 120º, which was associated with the decomposition of the AFt phase, (e.g., ettringite). The ettringite peak intensity increased with increasing sludge content; thus, sludge might promote the formation of AFt phases. Also, adding sludge resulted in the formation of calcium aluminate hydrate (C–A–H). These results could be related to the high reactive Al content in sludge, leading to the formation of additional Al-bearing phases [12]. In Fig. 2b, d, the XRD patterns of pastes at 28 days and 90 days are shown. The reflection peaks related to C–A–H and stratlingite were only observed in the blended pastes, not the reference ones. The ettringite peak intensity was enhanced with increasing sludge content, which was consistent with the TGA analysis. In both the TGA and XRD analyses, the content of portlandite (CH) decreased with increasing sludge content, confirming the pozzolanic reactivity of sludge. It is interesting to note that the CH content in the paste with 30% sludge was significantly low compared with the reference or 10% sludge pastes. Such a lower concentration of available CH content might result in a decrease in the reaction degree of sludge.

Fig. 2
Four graphs depict 4 fluctuating curves with 0%, 10%, 20%, and 30% sludge in 28 and 90 days. a and c, D T G percent per minute versus temperature degrees Celsius. The curves indicate A F t, C A H, and C H. b and d, Intensity counts versus 2 theta degrees.

a, c TGA and b, d XRD analysis of blended pastes (modified from Liu et al. [13])

Full size image

3.3 Chemo-Mechanical Properties of Pastes and Strength of Blocks

Table 5 shows the chemo-mechanical properties of the pastes and the strength of CB. The homogenized modulus of binders significantly decreased when >10% of cement was replaced with sludge. Such a reduction could be attributed to decreased High-density (HD) C–S–H gel in samples with 30% sludge. However, the total volume of C–S–H gel in the different pastes was almost the same, indicating that the filler effect of sludge might compensate for the cement dilution effect, although the high amount of sludge hindered the transformation of Low-denisty (LD) C–S–H gel to HD C–(A)–S–H gel. Based on the results for Al intensity in the C–S–H gel, with the addition of sludge to 30%, the original “Al-minor” C–(A)–S–H gel in pure cement paste was converted to “Al-rich” C–(A)–S–H gel. There was no significant difference in indentation modulus for C–(A)–S–H gel in the different pastes, indicating that Al incorporation had negligible effect on the mechanical properties of C–(A)–S–H gel.

Table 5 Chemo-mechanical properties of C–A–S–H gel in pastes and the strength of CB
Full size table

At a curing age of 7 days, the compressive strength of CB containing sludge was significantly lower than that of the reference samples. However, after curing for 28 days, the samples with 10% sludge exhibited a comparable or even higher compressive strength. The optimum sludge content in blocks was 10%, which was in agreement with the results of the nanoindentation analysis.

4 Conclusions

In general, the feasibility of reusing sludge as cement replacement was confirmed in this study. 800 ℃ was the best temperature to activate the pozzolanic reactivity of sludge. 10% of cement could be replaced with sludge in concrete blocks without compromising mechanical performance. The reaction degree of sludge in blended pastes could be up to 39%, and the Al content in C–(A)–S–H gel increased with adding sludge. However, the C–(A)–S–H gel modification had no significant effect on the indentation modulus or hardness of the gel. In addition to incorporation into C–(A)–S–H gel, the high content of reactive Al promoted the formation of other Al-bearing phases (e.g., ettringite and stratlingite).