Enabling the High-Volume Utilization of Iron-Rich Sewage Sludge Ash in Cement: The Role of Curing Temperature in Microstructure and Performance Enhancement

Abstract

The massive generation of sewage sludge ash (SSA) poses severe environmental challenges worldwide. Ternary blended cement incorporating SSA and limestone represents a promising green cementitious material that enables high-volume SSA utilization, yet the retarding effect of SSA on cement hydration leads to unsatisfactory early compressive strength, which remains the primary bottleneck restricting its large-scale engineering application, particularly at high replacement levels. In this work, the compressive strength development, hydration heat evolution, phase assemblage, and pore structure of SSA-limestone-cement blends cured at 5 °C, 20 °C, and 40 °C were systematically investigated to reveal the temperature-dependent hydration mechanism. The results show that elevated curing temperature accelerates early hydration and improves the 3-day compressive strength, while curing at 20 °C yields the highest late-age strength. Based on hydration calorimetry, X-ray diffraction (XRD), thermogravimetric analysis, and mercury intrusion porosimetry (MIP), the underlying micro-mechanism is clarified: curing at 5 °C results in the lowest 3-day and 28-day strength due to sluggish hydration reactions and abundant harmful macropores. At 20 °C, the 3-day strength is slightly lower than that at 40 °C owing to the delayed hydration acceleration period, but the substantial formation of ettringite (AFt) and carboaluminate phases at later ages refines the pore structure and gives rise to the maximum 28-day compressive strength. Curing at 40 °C achieves the highest early strength via accelerated hydration, yet the transformation of AFt to AFm and the accumulation of portlandite (CH) generate an unfavorable pore structure and limit late-age strength growth. These findings elucidate the regulatory role of curing temperature in the hydration and microstructural evolution of high-volume SSA ternary blends, providing theoretical guidance for curing regime optimization in engineering practice. This work offers a feasible pathway to promote the high-value utilization of iron-rich SSA in cementitious materials and contributes to carbon reduction in the cement industry.

1. Introduction

In 2023, the world produced 4.072 billion tons of cement, and producing one ton of cement emits about 800–900 kg of CO₂. − And it means cement factories were responsible for releasing approximately 3.7 billion tons of CO₂ in 2023, which is about six times more than in 1990, when emissions were only 0.57 billion tons. , An effective approach to reduce CO₂ emissions is to reduce cement consumption. , Therefore, conventional supplementary cementitious materials (SCMs), such as fly ash and slag, have been effectively utilized as cement replacements. Slag and fly ash are industrial byproducts, and the amount available declines as cleaner fuels are adopted and used more widely. This suggests the need for further research into alternative SCMs to replace portions of cement.

With the accelerated development of China’s urbanization, municipal sewage treatment is increasing greatly year by year. Sewage sludge is the solid precipitated material produced by water and wastewater treatment processes. In 2025, China’s sludge production is expected to exceed 90 million tons, and the market size will exceed 100 billion yuan. Incineration is gaining recognition as an environmentally friendly disposal method and has been attracting growing interest, particularly in Europe and China. − The incineration process decreases the waste mass by approximately 70%, thus producing a large amount of residual sewage sludge ash (SSA) as a final by-product of treatment. , Numerous studies have shown that SSA can be used as a SCM to partially substitute cement because SSA contains SiO₂, Al₂O₃, CaO, and Fe₂O₃ and exhibits some pozzolanic activity. − Studies have confirmed the feasibility of adopting SSA as a filler material for cement concrete. Meanwhile, comprehensive analysis of available experimental results demonstrates that the addition of SSA raises the required mixing water, which causes an elongation of setting time and a deterioration in early mechanical performance, − therefore most subsequent studies aim to control the content of SSA below 20%. − When SSA is used in large quantities (substituting 50% of cement), it may lead to early cracking and deformation of the cement, and the compressive strength of the paste at 28 days decreases by about 20% compared with that of the control group. A review of earlier studies leads to the conclusion that the SSA content should be controlled within 5–10% to satisfy the workability demands of mortar and high-strength concrete. This is primarily due to increased water demand, decreased fluidity, prolonged setting time, and decreased later strength owing to the porous structure and irregular shape associated with high-content SSA.

However, with the rapid increase in the amount of sludge, there have been many recent studies on increasing the proportion of sludge ash. In order to address the problem of increasing water demand, water-reducing agents are mainly added. − On the other hand, to solve the problem of retardation, some studies have addressed it through pre-dephosphorization treatment of SSA. However, this increases the treatment process and cost of SSA, which is also a main problem limiting the utilization of SSA as a cementing material. − Moreover, to address the issue of SSA’s irregular shape and pore configuration, Pan et al. found that the fineness and pozzolanic activity of SSA could be increased by increasing the grinding time, thus increasing the compressive strength. Additionally, some researchers have started to introduce limestone into the sewage sludge ash-cement binary system because the cost of limestone is cheaper than other nanometerials. − For conventional LC3 systems based on highly reactive calcined clay, the correlation between mix proportion, microstructure evolution, and mechanical strength has been systematically clarified. However, recent studies have pointed out that non-kaolinitic mineral phases with low reactivity exert complex effects on the hydration and performance of LC3 systems, and the regulation mechanism is significantly different from that of high-purity calcined clay systems. , Xia Y. et al. reported that the synergistic utilization of SSA and limestone resulted in cement (LC³) replacement levels as high as 50% (2:1 mass ratio), and significantly increased the compressive strength of ternary blended cements with high dosages at 28 days of curing. However, owing to SSA’s limited pozzolanic activity, the early strength of the blend is relatively low, especially when the dosage of SSA is large.

Compared with Portland cement at room temperature, the early hydration products of LC³ generate more ettringite (AFt). However, the formation and stability of AFt are significantly affected by temperatureas the temperature rises, the dissolution rate of ${\mathrm{AlO}}_2^{-}$ ions in cement and SSA increases, and the formation rate of AFt significantly accelerates. When the environmental temperature exceeds 55 °C, AFt gradually transforms into monosulfoaluminate (AFm), resulting in an increase in the porosity of the cement stone and a deterioration in volume stability. Liu C. et al. pointed out that, by comparing standard curing and high-temperature steam curing conditions, a synergistic effect between steam curing and nano-silica was observed, which can improve the early compressive strength. Furthermore, Xu et al. pointed out that the temperature of the paste can be raised by curing limestone calcined clay cement (LC³) using thermal energy generated by a microwave muffle furnace. Moreover, with an increase in microwave power, there is a positive effect on the early compressive strength and pozzolanic reaction of LC³. There is a lack of research on the hydration mechanism of sewage sludge ash-limestone-cement at different curing temperatures.

Therefore, this research concentrates on the advancement of strength and microstructure of ternary blended cements composed of SSA and limestone, designed with different dosages, by investigating their strength and microstructure at multiple curing temperatures. The heat of hydration, XRD, TG, and SEM techniques were used to reveal the hydration process and the products of hydration, and to investigate methods to promote the early strength of this ternary blended cement. What’s more, the MIP test was used to evaluate the pore structure, and the impact of hydration products on the pore structure was explored.

2. Materials and Methods

2.1. Materials

The cement utilized in this research is Portland cement produced by Fushun Cement Group, China, featuring a strength of 42.5 MPa. The sand used complies with Chinese ISO standards and has a particle size range of 0.08 to 2 mm. The sewage sludge used was sourced from the Guangxi Mingyang Sewage Treatment Plant. The sludge was initially oven-dried at 105 °C for 24 h. The dried sludge was placed in a room to cool naturally , ball-milled for 25 min, and then sieved through an 80 μm square-hole sieve. Finally, it was calcined in a tube furnace at 600 °C for 2 h in air at a heating rate of 10 °C min^–1. After calcination, it was naturally cooled to room temperature to obtain SSA. The limestone used was produced by Yuanheng Water Purification Material Factory, Gongyi City, China, with a purity of 98%. The specific surface areas of cement, limestone, and SSA were measured using a Blaine specific surface area tester (FBT-9, China) according to GB/T 8074-2008. Table presents the chemical and physical characterizations of the materials.

1. XRF Composition of Materials.

Components/wt %	SSA	Portland	Limestone
SiO2	10.83	22.15	1.47
Al2O3	4.4	4.39	0.42
CaO	5.08	64.76	57.17
Fe2O3	64.26	3.12	0.58
SO3	1.69	0.65	0.01
BaO	0.08	/	/
Na2O	/	0.57	/
MgO	0.19	/	0.15
K2O	1.04	/	/
TiO2	1.68	/	/
P2O5	7.88	/	0.02
TiO2	/	/	/
MnO	/	/	/
ZnO	/	/	/
f-CaO	/	0.85	/
LOI	2.87	3.51	40.18
Specific surface area (m2 kg–1)	680.2	386	613.5

Figure presents the particle size distributions for cement, SSA, and limestone, as measured with a Malvern Mastersizer 2000 laser particle-size analyzer (UK). The D50 values for cement, SSA, and limestone were 16.63 μm, 25.48 μm, and 22.23 μm.

1.

Cumulative size distribution (a) and differential distribution (b) of SSA, limestone, and Portland cement.

Figure shows the SEM morphology of the SSA, which has many pores and an irregular shape, with various particles interlaced and stacked.

2.

Microscopic morphology of SSA.

The XRD pattern of the SSA is shown in Figure . The main mineral phases in the SSA are quartz (SiO₂), hematite (Fe₂O₃), diopside (CaMgSi₂O₆), anhydrite (CaSO₄), and maghemite (Fe₃O₄).

3.

Mineral phases of SSA.

2.2. Mix Proportions and Sample Preparation

The proportions of the mixtures employed in this research are detailed in Table . For compressive strength testing, mortar specimens (40 × 40 × 40 mm³) were prepared with a binder-to-water ratio of 1:0.5. The coupled substitutions“blends”were made at a total substitution level of 15% to 60%. The proportion of SSA to limestone was kept consistent at 2:1 across all blends. Furthermore, a paste with the same water-to-cement ratio as the mortar was prepared for microscopic characterization.

2. Mix Composition of Mortars (wt %).

Sample Name	Portland	SSA	Limestone	w/b
PC	100	/	/	0.5
S15	85	10	5	0.5
S30	70	20	10	0.5
S45	55	30	15	0.5
S60	40	40	20	0.5

2.3. Methods

2.3.1. Hydration Heat Evolution

A TAM Air eight-channel isothermal calorimeter was used to monitor the early hydration and heat release of cement paste samples. The heat-of-hydration test temperatures were set at 5 °C, 20 °C, and 40 °C each time, and the mixed materials and mixing water were placed in advance in the environment of the test temperature in order to reduce experimental error. The water-cement ratio was 0.5. Typically, the major heat release during hydration occurs within the first 72 h. As a result, isothermal calorimetry testing was performed at 5 °C, 20 °C, and 40 °C for a duration of 3 days. This study employed three curing temperatures5 °C, 20 °C, and 40 °Ccorresponding respectively to cold-weather on-site curing, standard laboratory curing, and moderately elevated temperature curing typical of summer construction. These conditions were chosen to cover common engineering application scenarios and to systematically reveal the temperature-dependent behavior of the ternary blended cement.

2.3.2. Hydration Stoppage

After reaching the specified curing age of curing, the paste was crushed. The samples were placed in centrifuge tubes and soaked in anhydrous ethanol for 48 h, and then dried in a desiccator at 60 °C for 24 h. The sample used for SEM analysis was a powder, and the sample for XRD analysis was ground in an agate mortar and then passed through a 0.075 mm standard sieve.

2.3.3. X-ray Diffraction Analysis

An X-ray diffractometer (XRD; D8 Advance, Bruker, Germany) was used to analyze the phase of the materials. The analysis was performed using an instrument set to 40 mV and 40 mA with Cu Kα radiation. The scanning range was 2θ = 5°–80°, and the scanning speed was 0.5 s step^–1.

2.3.4. Thermogravimetric Analysis (TG-DTG)

TG was performed using a Discovery TGA55 instrument, with the temperature ranging from 50 °C to 900 °C at a controlled heating rate of 10 °C min^–1. The experiment was conducted under a nitrogen atmosphere with a flow rate of 50 mL min^–1. The portlandite content was measured using the tangent method. The content of chemically bound water was determined using eq .

$$W_B=\frac{(M_{105°\mathrm{C}}-M_{600°\mathrm{C}})}{M_{105°\mathrm{C}}}$$ 1

where W_B is the chemically bound water content, M _105 °C and M _600 °C are the masses of the powder samples at 105 °C and 600 °C, respectively.

2.3.5. Scanning Electron Microscopy (SEM)

An SEM (SUPRA 55 Sapphire, Zeiss, Germany) was used to analyze the structures of the hydration products in the cementitious materials. The samples were mounted on the stage with a conductive adhesive, and a gold layer was sprayed onto the surface using a gold sprayer for microscopic analysis.

2.3.6. Mercury Intrusion Porosimetry (MIP)

The 28-day hydrated mortar specimens were cut into small pieces, each measuring less than 1 × 1 × 1 cm³. The MIP test was performed using a Micromeritics AutoPore IV 9510 instrument, with a mercury intrusion pressure of 60,000 psi and a contact angle of 130°.

2.3.7. Compressive Strength

Compressive strength was measured using an automatic cement press machine (DYE-300B, China) at a loading rate of 2.4 kN s^–1 at curing ages of 1, 3, 7, and 28 days. The test was conducted in triplicate, and the average of the measured values was taken as the test result.

2.3.8. Carbon Assessment Method

The carbon footprint of the ternary binders was evaluated using a cradle-to-gate life-cycle assessment approach, consistent with ISO 14040 and ISO 14044. The functional unit was 1 t of binder. The system boundary included raw material extraction, transport, SSA pretreatment (drying at 105 °C, calcination at 600 °C for 2 h, grinding), and blending of constituents. Service-life CO₂ uptake was excluded. Avoided emissions from diverting sludge from landfill were credited.

Emission factors for Portland cement (0.980 kg CO₂-eq kg^–1) and limestone (0.040 kg CO₂-eq kg^–1) were taken from the literature. The SSA emission factor was calculated based on electricity (180 kWh t^–1, 0.5366 kg CO₂ kWh^–1), thermal energy (1.2 GJ t^–1, 0.056 kg CO₂ MJ^–1), and direct process emissions, yielding 0.417 kg CO₂-eq kg^–1. The avoided landfill benefit was 0.161 kg CO₂-eq kg^–1 SSA. Detailed inventory data are provided in the Supplementary Materials (Tables S1–S4).

Total emissions per ton of binder were calculated using eq ,

$$C_{\mathrm{total}}=m_{\mathrm{PC}}\times 0.980+m_{\mathrm{LS}}\times 0.040+m_{\mathrm{SSA}}\times 0.256$$ 2

3. Results and Discussion

3.1. Compressive Strength

As shown in Figure , the compressive strengths of cement paste with different LC3 substitutions are presented. When the curing temperature is 20 °C, as the SSA and limestone substitutions increase, the compressive strength of ternary-blended cement progressively decreases. The compressive strengths of S15, S30, S45, and S60 at 28 days were 83.43%, 66.33%, 48.51%, and 30.16% of those of PC, respectively. This may be due to the low pozzolanic activity and the porous structure of the SSA. Since the activity index of SSA30 was 66.33%, according to GB/T 2847-2022, it meets the compressive strength requirements of GB/T 2847-2022 and can be used as a pozzolanic material in cement production. Therefore, in the subsequent work, a higher dosage of S45 was adopted to further elucidate the influence of temperature on the hydration of the ternary-blended cement at the microscopic level.

4.

Compressive strengths of PC and ternary blended cements.

Figure illustrates the change in compressive strength of S45 at different curing temperatures. The compressive strength of the blends at 1 day did not yield effective values at any curing temperature. After 28 days of curing, S45’s compressive strengths at different curing temperatures were lower than those of the control due to the much lower reactivity of SSA relative to PC. It should be noted that, for mortar specimens, effective strengths could not be obtained at 5 °C at any age due to the extremely low degree of early-age hydration and insufficient structural strength to withstand the loading test. However, paste specimens for microstructural analysis can still be cured and sampled normally at 5 °C, so the corresponding microscopic test results are presented in subsequent sections. The compressive strength of 20 °C S45 is almost the same as that of 40 °C S45 at 3 days, and the compressive strength of 40 °C S45 is much lower than that of 20 °C S45 at 7 days and 28 days. Moreover, a comparison of the compressive strength growth rates at various ages under curing temperatures of 20 °C and 40 °C shows that the compressive strength of S45 at 20 °C increased by 35.9% from 3 to 7 days and by 71.9% from 7 to 28 days. In contrast, at 40 °C, the compressive strength of S45 rose by only 4.7% from 3 to 7 days and by 46.3% between 7 and 28 days. Since both hydration products and pore structure affect the strength of specimens under different curing temperatures, these factors will be analyzed in detail. Therefore, the S45 blend with a higher dosage was adopted for subsequent in-depth investigation. This proportion not only achieves high-volume utilization of SSA with prominent environmental benefits but also retains sufficient mechanical properties to support multi-condition comparative tests. More importantly, the early strength deficiency of S45 is more representative, thereby more clearly revealing the mechanism by which curing temperature regulates the hydration and microstructure of high-volume SSA-blended cement.

5.

Compressive strengths of PC and S45 at different curing temperatures.

3.2. Hydration Heat Evolution

Figure a and b shows the rate of heat evolution and cumulative heat release for the control and S45 at different curing temperatures. Figure a clearly shows that the induction period of S45 is considerably longer than that of PC at curing temperatures of 5 °C and 20 °C. The prolonged induction period may be due to the PO₄ ^3– in SSA retarding the early-age hydration of cement clinker. , Among these conditions, the reaction rate was slower at 5 °C, the induction period lasted longer, the main exothermic peak was lower, and the hydration acceleration period appeared later, delayed by 28 h relative to 20 °C. Unlike the previous two curing temperatures, the exothermic peak of 40 °C S45 appeared about 8 h earlier, and the hydration rate was obviously increased. These results indicate that the samples’ hydration progressed faster with increasing curing temperature.

6.

Heat release during the hydration of the PC and the blends: (a) heat-flow curves; (b) cumulative heat-release curves.

As shown in Figure b, based on the cumulative heat release, it is apparent that the cumulative heat of S45 at different curing temperatures is lower than that of PC, and the cumulative exotherm increases as the curing temperature increases. This result aligns with the findings in Figure a.

3.3. XRD Analysis

Figure shows the XRD patterns of ternary blended cement after hydration for 1, 3, 7, and 28 days at different curing temperatures. Compared with PC, the hydration products of the ternary blended cement are more diverse and include C −S–H, AFt, CH, hemicarboaluminate (Hc), and monocarboaluminate (Mc).

7.

XRD spectra of PC and S45: (a) PC; (b) 5 °C S45; (c) 20 °C S45; (d) 40 °C S45.

The C₃S peak intensity of the 20 °C PC was significantly lower compared with that of S45 at each curing temperature, and, as the hydration reaction proceeded, C₃S continued to dissolve while the portlandite peak was gradually enhanced.

Compared with blends cured at different temperatures, the hydration product peak and intensity of 5 °C S45 were the lowest at each curing age, and the lower curing temperature slowed the hydration process. Notably, 20 °C S45 has the strongest AFt diffraction peak, indicating that a curing temperature of 20 °C is most favorable for AFt formation.

The 40 °C S45 AFt peak is weak, but there is a clear CaCO₃ diffraction peak. In addition, 40 °C S45 has the strongest portlandite diffraction peak and the fastest hydration process, which is in agreement with the results of the heat of hydration. On the other hand, it was also observed that there were obvious AFm peaks at 20 °C S45 and 40 °C S45 at an early curing age, and the diffraction peak intensity was greater at 40 °C, suggesting that increasing the curing temperature may accelerate the generation of AFm.

Carboaluminate products are commonly found in limestone calcined clay cement systems, so the 8°–13° XRD region was compared next, as shown in Figure . The AFt content of S45 at 20 °C was comparable to that of PC at 1 day of hydration and even higher than that of PC at 3 days, which is partly because the pore solution is doped with additional Al³⁺ ions and sulfate from the SSA in the S45 system, thereby promoting the precipitation of AFt. It is worth noting that in blended systems, the diffraction peak of AFt shifts slightly toward a higher angle, which indicates that the dissolution of ferrite phases from iron-rich SSA plays a role in the formation of Fe-AFt in these systems. At 5 °C there were no obvious Hc and Mc peaks, whereas at 20 °C S45, significant Hc hydration peaks were observed at 28 days; however, the Mc and Hc diffraction peaks of S45 at 40 °C were not enhanced by the increase in temperature. Summarizing the above experimental results, the main reason for this is that the formation of carboaluminate occurs only after the sulfate ions in the solution are completely consumed; therefore, at 20 °C a large amount of AFt and carboaluminate formed, but at 40 °C the formation of carboaluminate was hindered, resulting in a large amount of AFm. Furthermore, the C–A–S–H gel will adsorb sulfate ions, and when the reaction temperature is higher, the formation of this gel occurs more rapidly, so more sulfate ions will be adsorbed onto the surface of the cement particles, resulting in a lower amount of AFt.

8.

XRD spectra of PC and S45: (a) 20 °C PC; (b) 5 °C S45; (c) 20 °C S45; (d) 40 °C S45.

3.4. TG-DTG Analysis

Figure presents the TG-DTG curves for PC and S45 at hydration periods of 1, 3, 7, and 28 days. There are three distinct valleys in the figure. The first valley occurs in the temperature range of 50 °C to 200 °C, which includes the evaporation of free water and the decomposition of other compounds, 90 °C is mainly attributed to AFt and C–S–H, and the mass loss around 140 °C is associated with AFm and carboaluminate phases. The next heat absorption peak observed in the temperature range of 400 °C to 500 °C corresponds to CH dehydroxylation. The last peak ranges from 600 °C to 800 °C and represents the decarbonation of CaCO₃.

9.

TG-DTG profiles of PC and S45 at curing periods of 1, 3, 7, and 28 days: (a) 1 d; (b) 3 d; (c) 7 d; (d) 28 d.

As the hydration reaction progressed, the mass loss of each group of samples increased gradually over 1, 3, 7, and 28 days. The mass loss of the control group was significantly higher than that of the S45 samples cured at different temperatures.

At 5 °C, a significant AFt heat-uptake peak appeared only at 7 days because lowering the curing temperature slowed the reaction process. 5 °C, S45 shows a large amount of CaCO₃ production compared with PC. It’s worth noting that at 28 days Ca­(OH)₂ and CaCO₃ mass losses were reduced relative to those at 7 days, whereas carboaluminate mass losses increased significantly. This is owing to the pozzolanic reaction triggered by the addition of a large amount of SSA. One is the reaction of SSA with limestone, which produces carboaluminate, as shown in eq . The other is the pozzolanic reaction of SSA, which produces additional C–(A)–S–H as shown in eq .

$$\mathrm{SSA}+3.5\mathrm{CH}+0.5\mathrm{Cc}+8.5\mathrm{H}\to \mathrm{C}4\mathrm{Ac}0.5\mathrm{H}12$$ 3

$$\mathrm{SSA}+3\mathrm{CH}+6\mathrm{H}\to \mathrm{C}-(\mathrm{A})-\mathrm{S}-\mathrm{H}$$ 4

The same phenomenon is also observed at 20 °C, although both reactions occurred earlier. A decrease in the mass loss of Ca­(OH)₂ and CaCO₃ was detected at 7 days of conditioning. On the other hand, as the age of curing increased, AFt and other hydration products increased significantly relative to 5 °C. It is important to point out that the growth of hydration products from 7 to 28 days was particularly large, with more AFt, HC, and MC formed, which corresponds to the XRD results.

At 40 °C, after 1 day of hydration, a large number of hydration products are formed, and the increase in hydration products in the subsequent hydration period is very small, especially since the total amount of hydration products from 7 to 28 days did not increase much. No obvious AFt peak appeared, which shows that the high temperature leads to the gradual transformation of early-formed AFt into AFm. In addition, large amounts of Ca­(OH)₂ and CaCO₃ were observed at 40 °C S45.

Figure a shows the evolution of chemically bound water with hydration age. As a result of the replacement effect of SSA and limestone on cement, the bound water of S45 samples in each group was lower than that of PC. The amount of bound water was minimal at 5 °C, whereas at 40 °C, although the amount of bound water was greater than at 20 °C at 7 days, the S45 chemically bound water at 20 °C was comparable to that at 40 °C at 28 days. Higher temperatures accelerate hydration, and this aligns with the findings from the heat of hydration. The acceleration of the hydration process occurs mainly in the first 7 days of the reaction, followed by a slowdown in the hydration reaction. Especially at 40 °C, the hydration process is fast at early age, slow at later age, and eventually basically consistent with that at 20 °C. This indicates that more hydration products can be generated after 28 days of hydration at a curing temperature of 20 °C, and raising the curing temperature is conducive to the formation of hydration products at early ages .

10.

(a) Bound water content of PC and S45; and (b) portlandite content of PC and S45.

Figure b shows the relationship between hydration age and calcium hydroxide content, with PC having the highest calcium hydroxide content at each age, followed by 40 °C S45. In addition, we note that the growth rate of calcium hydroxide is greatest under curing at 40 °C, which indicates that a higher curing temperature significantly speeds paste hydration.

3.5. SEM Analysis

Figure shows the microscopic morphology of 28-day PC and S45 at different curing temperatures. Firstly, 20 °C PC contains a large number of C–S–H hydration products, needle-like AFt hydration products, and a large number of plate-like CH, with a compact microstructure and well-connected hydration products. Secondly, the 5 °C S45 structure is loose and the pores are obvious, which is mainly due to the delay in the hydration process caused by the too low curing temperature. Thirdly, in the 20 °C S45, a large amount of AFt and C–S–H can be found. Finally, compared with the 20 °C S45, the structure of the 40 °C S45 is looser, and more pores can be observed. Furthermore, it is worth mentioning that a large amount of CH can be observed, which may be a factor in the compressive strength of the 40 °C S45 being lower than that of the 20 °C sample at 28 days.

11.

Microscopic morphology of paste.

3.6. MIP Analysis

MIP analysis identified four distinct pore types in the cementitious composites, namely fine mesopores (5–27 nm), coarse mesopores (27–50 nm), medium capillary pores (50–100 nm), and large capillary pores (>100 nm). , In this experiment, the effect of different curing temperatures on the porosity and pore distribution of blends was determined. Porosity between 100 and 1000 nm range generally weaken the strength of cementitious materials. As shown in Figure a, the total porosity of all three groups of S45 samples is higher than that of PC samples, and the reason is that the total reactivity of the cementitious material decreases due to the replacement of part of the cement by SSA and limestone, which leads to fewer hydration products filling the pores and higher porosity. Except for 20 °C PC, 20 °C S45 has the smallest total porosity and the greatest compressive strength. In addition, the pore volumes and fractions of the four types of pores in PC and the three groups of S45 are shown in Figure b. It is worth noting that all groups of S45 show a higher pore volume of large capillary pores than that of the control.By further comparing the fractions of the four types of pores in S45 at different curing temperatures, the main type of pore is the large capillary pore, with the fractions at 5 °C and 40 °C being about as high as 45%, far more than the fraction at 20 °C. This is mainly attributed to the slow reaction rate at the beginning of hydration at 5 °C, and, as a result, there is not enough AFt and C–S–H to fill the pores, and large capillary pores and mesopores increase . Conversely, the curing temperature of 20 °C is most conducive to the formation of carboaluminate and needle-shaped AFt, which effectively fill the pores and refine the porous structure of the matrix. Finally, 40 °C S45 contains many AFm clusters and CH, which is detrimental to the development of compressive strength, resulting in a lower compressive strength than 20 °C S45.

12.

Pore structure of PC and S45 after 28 days of curing: (a) cumulative pore-volume curves; and (b) pore volumes and proportions of various pore types in PC and S45.

3.7. Environmental Implications

The partial replacement of Portland cement with sewage sludge ash and limestone offers a direct pathway to reducing the carbon footprint of cementitious binders. To quantify this benefit, the cradle-to-gate CO₂ emissions associated with the production of the S45 binder (55 wt % Portland cement, 30 wt % SSA, 15 wt % limestone) were calculated and compared with those of the reference Portland cement. Detailed inventory data and emission factor derivations are provided in the supplementary material Tables S1–S4. −

For the S45 binder, the total emission of 621.8 kg CO₂ per ton represents a reduction of 358.2 kg CO₂ per ton, corresponding to a 36.6% decrease relative to the Portland cement reference. This reduction is primarily attributed to three factors. First, the displacement of 45 wt % of the clinker directly avoids the substantial process- and fuel-related emissions inherent to cement manufacture. Second, the calcination temperature employed for SSA production in this study (600 °C) is considerably lower than that required for clinker production (≈1450 °C), resulting in a lower thermal energy demand per unit mass of the substitute material. Third, the utilization of SSA diverts sewage sludge from landfill disposal, yielding an additional carbon credit of 0.161 kg CO₂ per kg of SSA. Taken together, these factors account for the observed 36.6% reduction in cradle-to-gate carbon emissions. The results demonstrate that high-volume incorporation of iron-rich SSA in ternary blends with limestone constitutes an effective strategy for mitigating the carbon footprint of cementitious materials. The S45 binder, in particular, offers a favorable balance between mechanical performance and carbon-emission reduction, making it a viable alternative for non-structural and moderate-strength applications.

4. Conclusions

This work systematically investigated the influence of curing temperature on the compressive strength, hydration behavior, and microstructural characteristics of iron-rich SSA-limestone-Portland cement ternary blends. The core scientific insight is that curing temperature regulates the strength development of high-volume SSA-blended cement primarily by modulating hydration kinetics, phase assemblage evolution, and subsequent pore-structure refinement. The main conclusions are summarized as follows:

• (1)Curing temperature exerts a pronounced influence on the hydration kinetics and strength development of the ternary blended cement. At 5 °C, the hydration reaction of S45 is severely retarded, leading to persistently low compressive strength up to 28 days. Curing at 40 °C accelerates early hydration and raises the 3-day cumulative heat release, yet the hydration rate decelerates markedly in later stages. The 3-day compressive strength of S45 at 5 °C and 20 °C is lower than that at 40 °C, whereas the 28-day compressive strength of S45 cured at 20 °C is the highest among the three temperature groups.
• (2)Curing temperature directly determines the type and evolution sequence of hydration products. Distinct AFt formation is observed in blends cured at 5 °C and 20 °C, and the AFt content continues to increase at later ages under 20 °C curing. An elevated temperature of 40 °C is unfavorable for AFt stabilization and promotes its transformation to AFm. No hemicarboaluminate (Hc) or monocarboaluminate (Mc) is detected within the first 7 days at all curing temperatures. From 7 to 28 days, abundant Hc and Mc are formed in S45 at 20 °C, while large amounts of AFm and calcium carbonate accumulate in S45 at 40 °C.
• (3)The pore structure of the blends is jointly governed by hydration degree and phase assemblage, which further dominates the late-age mechanical performance. S45 cured at 5 °C exhibits the loosest microstructure and the highest total porosity due to sluggish hydration. For the S45 specimen cured at 20 °C for 28 days, the continuous formation of AFt and carboaluminate phases effectively fills capillary pores and refines the pore network, with thin mesopores accounting for 32% of the total pore volume, thus resulting in the densest matrix and the highest late-age strength. At 40 °C, the dominance of the AFm phase and portlandite accumulation lead to increased total porosity and a deteriorated pore structure. In general, early-age strength is mainly controlled by the hydration reaction rate, while late-age strength is predominantly determined by the pore structure refinement effect. These findings clarify the temperature-dependent hydration mechanism of high-volume, iron-rich SSA ternary blended cement and provide theoretical guidance for curing regime optimization to balance early and late mechanical performance. This work offers scientific support for the large-scale resource utilization of sewage sludge ash and the development of low-carbon cementitious materials.

Data will be made available upon request.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.6c07009.

• Carbon emission calculation data (Tables S1–S4) and supporting references (PDF)

K.C.: Conceptualization, Formal analysis, Methodology, Writing-original draft. Y.X.: Data curation, Investigation, Methodology, Formal analysis, Validation, Visualization, Writing-original draft. S.L.: Data curation, Investigation, Methodology, Formal analysis, Validation, Visualization, Writing-original draft. Y.R.: Methodology, Conceptualization. W.Z.: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing–review & editing.

The authors declare no competing financial interest.