Upcycling industrial waste into zero-clinker alkali-activated binders for low-carbon construction

Summary

Global decarbonization requires alternative binder that eliminates the high CO₂ burden of Portland cement. This study develops a zero-clinker, cement-free alkali-activated binder produced from two by-products: calcium carbide residue (CCR) and rice husk ash (RHA). Solid sodium silicate and NaOH were incorporated into the dry mixture to provide silica and alkali activators, enabling water-based mixing without calcination. The synergistic availability of Ca²⁺ from CCR and silica from RHA promotes nucleation and calcium silicate hydrate (C–S–H) formation, resulting in a continuous reaction product under mild heat-assisted curing. Microstructural analysis shows a compact C–S–H matrix, while RHA-rich formulations exhibit reduced mass loss at 450°C, indicating enhanced high-temperature stability. Although the strength (∼16 MPa) is suitable for non-structural applications, the system demonstrates a viable pathway for valorizing industrial wastes into functional, low-emission binders. This work supports circular economy strategies and contributes to the development of low-carbon construction materials.

Graphical abstract

Highlights

• •Developed green cement using CCR and RHA as sustainable raw materials
• •Faster setting time compared to OPC with improved thermal stability
• •Strong (Ca+Na)/Si ratio correlation enhances strength and durability
• •Promotes sustainable practices in green construction applications

Industrial chemistry; Materials synthesis; Industrial processing of material

Introduction

The growth of the industrial sector plays a significant role in the stability of Thai economic system. This rapid development has a profound impact on the country’s resource utilization, particularly the energy sector and industrial production processes. These sectors are significant consumers of natural resources and major contributors to carbon dioxide (CO₂) emissions, which are increasing continuously.¹ The emission of CO₂ into the atmosphere is considered as a major global problem that has prompted countries around the world to seek solutions for reducing the emissions.²

The CO₂ emissions are mainly produced from the energy sector and manufacturing processes. In the cement industry, for instance, the production of a ton of cement releases an equivalent amount of CO₂. This industry contributes approximately 8% of global greenhouse gas emissions.³ In 2017, worldwide cement production reached 4.1 billion tons, leading it to the third-largest source of CO₂ emissions.^3,4 By 2023, cement production remained at 4.1 billion tons, resulting in the release of an equivalent 4.1 billion tons of carbon dioxide.⁵ One of the simplest strategies to mitigate CO₂ emissions is to implement reforestation, which helps absorb the CO₂. On average, a single tree can absorb between 10 and 40 kg of CO₂ annually, with values strongly dependent on factors, such as climate, age, and species.⁶ However, this amount is still insufficient when weighed against the rapid expansion of the industry and transport sectors.

Alternative binders have attracted attention as replacements for ordinary Portland cement (OPC), a major contributor to CO₂ emissions. In recent years, low-carbon binder technologies, particularly geopolymers and cementitious materials derived from industrial by-products, have gained significant attention for their environmental and mechanical performance. Geopolymers, synthesized from alumino-silicate precursors, offer a promising pathway to reduce clinker consumption while enhancing durability.^7,8,9 Simultaneously, waste-derived binders produced from materials such as fly ash, slag, and construction waste have been explored as sustainable substitutes in cement-free systems. Recent studies have demonstrated the feasibility and performance of these technologies under various curing regimes and mix designs.^10,11 This study builds on these advancements by developing a zero-clinker, cement-free binder using industrial waste, contributing to global efforts in sustainable construction.

At the 26^th conference of the parties to the United Nations Framework Convention on Climate Change (COP26) in 2021, Thailand announced its commitment to attaining carbon neutrality by 2050 and reaching net-zero emissions by 2065. In support of this goal, the Thai government has established guidelines aimed at reducing CO₂ emissions. In the cement sector, specific policies have been implemented to promote the use of clinker substitutes, develop effective waste management strategies, and encourage the adoption of carbon capture technologies in cement plants. Thailand’s cement production capacity reached 59.25 million tons in 2023,¹² highlighting the sector’s significant contribution to the national carbon footprint.

In addition to its CO₂ capture initiatives, the Thai government has introduced the Second National Waste Management Action Plan, which spans from 2022 to 2027.¹³ This plan underscores the importance of improving recycling efforts and reducing waste generation. Its objective is to maximize resource recovery from waste materials, with a recycling target set at 74% of all waste generated in the country, thereby minimizing reliance on landfills. Furthermore, the nation faces a significant challenge in managing industrial waste. According to the Department of Industrial Works, Thailand produced 19.82 million tons of industrial waste in 2023, underscoring the need for robust and effective management strategies.

The calcium carbide production industry generates significant amounts of CO₂ and solid waste known as calcium carbide residue (CCR), which primarily consists of calcium hydroxide (Ca(OH)₂). Fresh CCR is typically produced in slurry form and may exhibit an unpleasant odor due to residual acetylene and associated impurities; however, this odor is substantially reduced or absent after aging and drying of the material.^14,15 Discharge of CCR slurry into disposed ponds leads to elevated alkalinity in the surrounding soil, which hinders plant growth. Annually, the country produces over 12,000 tons of CCR. In addition, Thailand is an agricultural nation and has numerous industries related to agricultural products, with rice cultivation being particularly prominent. In 2023, Thailand harvested 26 million tons of paddy, resulting in approximately 5.2 million tons of rice husks, which represent 20% of the total yield.¹⁶ These rice husks are frequently utilized as biomass fuel for power plants, in accordance with the National Alternative Energy Development Plan. However, the combustion of rice husk generates rice husk ash (RHA), a by-product that is often disposed of improperly, leading to environmental concerns due to ash dispersal affecting air quality.

Numerous researchers have explored the potential of using CCR and RHA as ingredients in concrete.^10,17 To serve as supplementary cementitious materials, these waste products must undergo grinding to achieve smaller particle sizes. The finer particle size enhances their pozzolanic reactivity, which is crucial for their performance in cementitious systems. It is generally advised that no more than 25% of the cement by weight should be replaced with these materials, as higher substitution rates can compromise the integrity of the concrete, leading to unsoundness and cracking.¹⁷ In addition, the use of RHA in geopolymer systems is limited, as high RHA content can lead to increased matrix elasticity, poor dimensional stability (unsoundness), and relatively low mechanical strength and long-term durability.^18,19 Consequently, there exists a significant volume of CCR and RHA that requires effective utilization. Additionally, it is essential to investigate new cementitious materials that incorporate waste and address concerns related to low CO₂ emissions.

From the cement chemistry, dicalcium silicate (C₂S or 2CaO·SiO₂) and tricalcium silicate (C₃S or 3CaO·SiO₂) compounds within cement interact with water through hydration reactions, leading to the formation of C–S–H and calcium hydroxide (CH or Ca(OH)₂). C–S–H is mainly responsible for the strength of the binder. CH can also contribute to additional strength development trough reaction with pozzolans (i.e., silica-based materials) resulting in the conversion of CH into more C–S–H.^20,21 Among the most prevalent pozzolans are RHA and fly ash generated from coal-fired power plant. This principle has inspired the innovative concept of producing cement using industrial waste ash.

In this study, a zero-clinker green binder was developed by entirely replacing traditional Portland cement with CCR, which is rich in calcium hydroxide, and ground RHA as the primary raw materials. Solid sodium silicate and pearl-grade sodium hydroxide (SH) were deliberately incorporated as dry alkali activators to enable a ready-mix, powder-based binder system. Unlike conventional alkali-activated binders that rely on pre-prepared liquid alkaline solutions, this approach allows all reactive components to be pre-blended in dry form, enabling simple water-only mixing at the point of use. Upon water addition, pearl-grade NaOH provides alkalinity to activate CCR and RHA, while solid sodium silicate supplies soluble silicate species, promoting the formation of calcium silicate hydrate (C–S–H). By utilizing two abundant industrial by-products and eliminating clinker production, this system offers a substantial reduction in CO₂ emissions. Overall, the proposed dry-mix alkali-activated binder presents a practical and scalable pathway for sustainable construction materials, balancing reactivity, simplicity, and environmental benefit, with potential for future development as a ready-mix green cement.

Design

Mix proportion of green cement

The hydration of conventional Portland cement involves the reaction of tricalcium silicate (C₃S, 3CaO·SiO₂) and dicalcium silicate (C₂S, 2CaO·SiO₂) with water, resulting in the formation of C–S–H.²⁰ In the present study, these stoichiometric compositions were used solely as reference Ca/Si ratios for mix design, rather than to imply the formation of clinker mineral phases. Given the similar molecular weights of CaO (56 g/mol) and SiO₂ (60 g/mol), weight-based proportions were employed to approximate CaO/SiO₂ molar ratios of 3:1 and 2:1. Accordingly, the green cement formulations were designed with Ca/Si ratios corresponding to CaO/SiO₂ molar ratios of 3 and 2, denoted hereafter as Ca/Si-3 and Ca/Si-2, respectively.

CCR was used as the source of CaO, while RHA and sodium silicate pentahydrate (SS) served as SiO₂ sources. SH was also used as an additional source of alkali. A SiO₂-to-SH ratio of 2 was chosen for the alkaline activator, based on information from prior research.⁹ Additionally, the water-to-solid particle ratio was adjusted between 0.4 and 0.6 to achieve a fluid paste consistency. Notably, mixtures containing RHA required more water due to its high porosity. The mixing formulation details are presented in Table 1. In the sample notation, Ca/Si-2 and Ca/Si-3 indicate formulations designed with CaO/SiO₂ molar ratios of 2 and 3, respectively, while the suffixes R, RS, and S represent the incorporation of RHA, combined RHA–SS, and SS alone. For example, Ca/Si-2-RS refers to a green cement mixture with a CaO/SiO₂ molar ratio of 2 incorporating both RHA and SS.

Table 1.

Mix proportion of cement pastes

Samples	OPC (g)	CCR (g)	RHA (g)	SS (g)	SH (g)	Water (g)	Water/GCa ratio
OPC	35					17.5	0.5
Ca/Si-2-R	–	20	10	–	5	21	0.6
Ca/Si-2-RS	–	20	5	5	5	17.5	0.5
Ca/Si-2-S	–	20	–	10	5	14	0.4
Ca/Si-3-R	–	30	10	–	5	27	0.6
Ca/Si-3-RS	–	30	5	5	5	22.5	0.5
Ca/Si-3-S	–	30	–	10	5	18	0.4

^aGC = CCR+RHA+SS + SH.

In contrast to previous studies that employed pre-grinding of solid activators with precursor materials, sodium silicate, and NaOH were used in their original commercial form in this study to reduce additional work, energy consumption and avoid additional variability associated with grinding time and particle size. This approach was adopted to assess the feasibility of a practical dry-mix binder system. Although rapid setting may limit complete dissolution during short mixing periods, the activators were sufficiently dispersed for this proof-of-concept investigation.

Results and discussion

Properties of raw materials

The median particle sizes of the CCR and RHA were 42.3 μm and 38.1 μm, respectively. The particle size distribution curve is shown in Figure 1. The chemical compositions of both materials were analyzed using X-ray fluorescence (XRF, Fischerscope, and X-ray XUV 773), and the results are displayed in Table 2. The primary component of the CCR, resulting from the hydration of calcium carbide, was calcium hydroxide. This material has a calcium oxide (CaO) content of 90.4%. Other oxides present in smaller quantities included SiO₂, Al₂O₃, MgO, and Fe₂O₃, which originated from impurities in the coal and the calcium carbide production process.⁹ In contrast, the chemical composition of RHA consisted predominantly of SiO₂, accounting for 93.8% of its content. The X-ray diffraction (XRD, Bruker, and D2 phaser) patterns for CCR and RHA are illustrated in Figure 2, demonstrating that CCR contained a significant amount of portlandite (calcium hydroxide), while RHA was primarily composed of an amorphous phase with some minor crystalline SiO₂ phase in the form of quartz.

Figure 1.

Particle size distribution curve of raw materials

Table 2.

Oxide compounds of CCR and RHA

Oxide compounds (%wt)	CaO	SiO2	Al2O3	MgO	Fe2O3	K2O
CCR	90.4	4.7	2.5	1.7	0.6	0.0
RHA	0.8	93.8	0.5	1.6	0.4	2.8

Figure 2.

XRD patterns of CCR and RHA

Setting time of green cement pastes

The Vicat apparatus plays a crucial role in the field of cement technology, particularly for assessing the setting times of various cement pastes, including cement formulations. In a series of experiments utilizing this apparatus, results showed that the initial setting times for all tested samples of cement paste were approximately 30 min faster than the standard times recommended for OPC, which typically range from 90 to 150 min as shown in Figure 3.

Figure 3.

Setting time of green cement pastes

Data are represented as mean ± SD (n = 3).

The green binder exhibited rapid stiffening within approximately 30 min, accompanied by a noticeable increase in temperature during mixing, indicating a highly reactive alkali-activated system involving calcium- and silica-rich precursors. This behavior is consistent with the exothermic nature of calcium-alkali interactions and early C–S–H formation in calcium-based cementitious systems.²⁰ Although heat evolution and reaction kinetics were not quantitatively measured, the rapid initial setting reflects an accelerated onset of binder formation. Importantly, this behavior does not represent a flash set, as the paste remained sufficiently workable during mixing and casting.

Despite the fast initial set, the final setting time was longer (5–6 h) than that of OPC paste (approximately 3 h). This extended final setting time can be attributed to several factors. Firstly, the chemical reaction pathways in alkali-activated systems often differ from traditional OPC hydration, involving more complex gel formation and polymerization steps that continue over a longer period to develop full structural integrity. Secondly, the presence of solid sodium silicate and SH influences early-age gel formation, leading to the development of an initial reaction network that imparts stiffness to the paste, while the matrix has not yet fully hardened or structurally matured. Lastly, the incorporation of RHA, which contains amorphous silica, may cause a more gradual pozzolanic reaction that extends the setting duration compared to OPC’s faster hydration of clinker minerals.

It is also noted that minor bleeding was observed during the early stages after casting, indicating partial phase separation in the fresh paste. This behavior is attributed to early-age water redistribution in the presence of solid alkali activators and does not reflect hardened-state performance. Although excessive bleeding is generally associated with suboptimal mix design or high water content, the bleeding observed in this study was limited and occurred during proof-of-concept evaluation. In practical applications, such behavior can be mitigated through optimization of the water-to-solid ratio, activator dosage, and particle size distribution, which will be addressed in future studies.

Based on these findings, it is recommended that heat curing be employed for the application of this cement formulation. Heat curing, by maintaining optimal temperatures during the hydration process, can enhance the strength development and durability of the cement paste. This method not only ensures that the reactions reach completion but also promotes the formation of a denser microstructure, leading to improved mechanical properties.

Morphological analysis of hardened green cement pastes

The morphological analysis of hardened cement pastes, evaluated through XRD patterns, provides significant insights into the underlying processes that occur during cement hydration and the resulting structural formations. As illustrated in Figure 4, both Ca/Si-2 and Ca/Si-3 based formulations reveal the presence of C–S–H, which is a crucial component contributing to the strength and durability of cementitious materials. Notably, the XRD patterns indicate that C–S–H is detected at consistent peaks alongside calcium hydroxide, which arises from the interaction of CCR with silica sources such as SiO₂ derived from RHA, and sodium silicate.

Figure 4.

XRD patterns of hardened green cement pastes

(A) Ca/Si-2-based formula.

(B) Ca/Si-3-based formula.

The formation of calcium silicate hydrate (C–S–H) is central to cement chemistry, as it represents the primary binding phase responsible for strength development in cementitious materials. Due to its poorly crystalline nature, C–S–H is not characterized by sharp diffraction peaks in XRD patterns but is typically identified by a broad diffuse feature in the ∼28–34° 2θ region when using Cu Kα radiation. This feature is commonly associated with tobermorite-like short-range order within the C–S–H structure.²² Consistent with previous reports, similar diffuse scattering in this 2θ range has been attributed to C–S–H formation in cement and alkali-activated systems.^23,24,25 Sharp reflections observed at lower angles (e.g., ∼18° 2θ) are instead attributed to crystalline calcium hydroxide rather than C–S–H. The intensity of the XRD peaks can be correlated directly with the quantity of C–S–H produced during the hydration process. This relationship forms the basis for further comparative studies aimed at quantifying the formation of C–S–H, with the peak area between 28° and 34° 2theta serving as a critical metric. By expressing this peak area as a percentage of the total peak areas identified in each sample, researchers can derive a more understanding of the hydration dynamics.^26,27,28

In this study, the C–S–H index was determined by calculating the peak area ratio between each sample and the OPC reference within the 2θ range of approximately 28–34°, using the default integration method in X’Pert HighScore software. This method automatically subtracts the background and integrates the area under the defined peaks. A higher peak area in this range corresponds to a greater amount of crystalline C–S–H phase, indicating enhanced binding performance. The resulting C–S–H index offers a semi-quantitative basis for comparing the effectiveness of formulations incorporating CCR and supplementary cementitious materials such as RHA. The percentage ratios of C–S–H peak areas for all samples are summarized in Table 3 to support performance evaluation.

Table 3.

Percentage ratio of peak area of C–S–H between 28° and 34° 2theta

Samples	Peak area (28°–34° 2theta) (cts∗°2theta)	Total peak area (cts∗°2theta)	Area ratio (%)	C–S–H index (%)
OPC	485.48	921.74	52.67	100
Ca/Si-2-R	583.77	1080.21	54.04	102.60
Ca/Si-2-RS	572.71	1080.21	53.02	100.66
Ca/Si-2-S	724.07	1452.26	49.86	94.66
Ca/Si-3-R	577.52	1163.74	49.63	94.22
Ca/Si-3-RS	657.31	1492.37	44.04	83.62
Ca/Si-3-S	704.88	1432.36	49.21	93.43

According to Table 3, the peak area percentage of C–S–H for the OPC control sample was 52.67%. Notably, the Ca/Si-2 based formula exhibited a marginally higher peak area percentage compared to the OPC, specifically in the Ca/Si-2-R and Ca/Si-2-RS samples. Incorporating supplementary cementitious materials such as sodium silicate with RHA did not yield a dramatic reduction in the C–S–H peak area. This indicates that, although these materials may not significantly alter the immediate formation of C–S–H in the mixtures analyzed, they could still play a crucial role in producing unique microstructural characteristics within the concrete matrix. The presence of sodium silicate, known for its properties as a binder and its ability to augment the pozzolanic reaction, may contribute to improved long-term performance of the concrete, even if initial peak areas do not reflect this. To gain a comprehensive understanding of the effect of these supplementary materials on concrete performance, physical tests such as compressive strength must be conducted. These tests serve to elaborate on the functional properties of the varying formulations beyond merely examining C–S–H formation.

Thermogravimetric analysis of hardened green cement pastes

Thermogravimetric analysis (TGA) and differential thermogravimetry (DTG) are invaluable techniques used to characterize the thermal stability and compositional changes of materials, especially in the field of building materials e.g., cement. In this study, TGA and DTG were conducted on cement pastes, revealing significant insights into their thermal behavior. The results, illustrated in Figures 5A and 5B, highlight the complexities underlying the thermal degradation processes of different cement compositions. The TGA curve primarily illustrates the percentage of weight remaining in the samples as a function of heating temperature under a nitrogen atmosphere. This controlled setting is critical as it prevents oxidation and provides more accurate thermal profiles. Conversely, the DTG curve complements this analysis by showing the specific temperatures at which various volatile compounds within the material decompose, offering a clearer picture of the individual weight loss events.

Figure 5.

TGA/DTG results of hardened green cement pastes

(A) TGA-weight loss.

(B) DTG-derivative weight.

Particularly noteworthy were the DTG curves derived from the C₂S-based formulation, which were chosen as representative due to their similar decomposition temperatures when compared to those produced by the Ca/Si-3-based formulation. This observation indicates a certain degree of thermal uniformity among these types of cement, despite the difference in component compositions.

Focusing on the control OPC sample, significant weight changes were observed at approximately 100°C, 450°C, and 700°C. These temperature transitions correlate closely with specific thermal events: the initial weight loss at around 100°C can be attributed to water evaporation, a common occurrence in the presence of calcium silicate hydrate and pore solutions within the pastes.

Interestingly, all cements analyzed exhibited two distinct peaks on the DTG graph within the 100°C–120°C range, corresponding with the evaporation of water. The DTG peak at around 100°C signifying water evaporation is consistent across various formulations, indicating that the initial physical processes, such as moisture loss, are relatively uniform regardless of the specific cement formulation. This behavior could be influenced by the colligative properties resulting from the incorporation of sodium silicate and NaOH in water during the mixing process. This concentration leads to an elevation in the boiling point of the solution, resulting in an increased temperature at which water from the paste begins to evaporate.²⁹

At a higher temperature around 450°C, the decomposition of Ca(OH)₂ occurs. This transformation is particularly significant, as the presence of Ca(OH)₂ in cement paste and CCR plays a vital role in the early strength development and overall durability of the cement matrix. Additionally, the drop in weight at approximately 700°C marks the decomposition of calcium carbonate (CaCO₃) into CaO and CO₂.^30,31,32

In addition to the decomposition of CaCO₃ occurring at a temperature of 700°C, various studies have indicated that C–S–H begins its decomposition process at a lower threshold of 560°C. However, it is important to note that significant decomposition of C–S–H is observed only when temperatures exceed 600°C. At precisely 600°C, dehydrated calcium silicate undergoes a transformation into the highly reactive α′-C₂S (alite), which possesses superior hydration activity compared to the C₂S (belite) that forms at elevated temperatures beyond this point.³³ As the temperature continues to rise past 600°C, the rate of decomposition for both CaCO₃ and C–S–H increases notably. This accelerated decomposition is a primary factor contributing to the observed loss in strength and structural performance of concrete materials exposed to high temperatures.³⁴ The degradation of C–S–H and the release of gaseous products lead to microstructural changes in the concrete matrix, which manifest as cracks, spalling, or a general reduction in mechanical properties.

Interestingly, experimental findings have demonstrated that Ca/Si-2-S and Ca/Si-3-S samples, which lacked RHA, did not exhibit the decomposition peak at 700°C. The implications of this observation suggest that the inclusion of RHA may influence the thermal stability and decomposition pathways of C–S–H in cements.

In comparing the thermal behavior of cement samples with OPC, a total weight loss of 25%–30% was observed for the cement, while the OPC exhibited a more restrained weight loss of approximately 15% under similar conditions. The data summarized in Table 4 illustrates the weight loss of all samples at key temperatures of 100°C, 450°C, and 700°C, providing valuable insight into the performance characteristics of these materials during thermal exposure.

Table 4.

Weight loss of cement pastes at 100°C, 450°C, and 700°C

Samples	Weight loss at 100°C (%)	Weight loss at 450°C (%)	Weight loss at 700°C (%)	Residue (%)
OPC	9.01	2.94	3.50	84.55
Ca/Si-2-R	15.65	3.20	7.27	73.89
Ca/Si-2-RS	22.65	5.69	0.00	71.66
Ca/Si-2-S	20.23	8.07	0.00	71.70
Ca/Si-3-R	16.05	4.62	8.45	70.88
Ca/Si-3-RS	17.20	5.98	2.33	74.49
Ca/Si-3-S	20.15	9.58	0.00	70.27

The analysis of the weight loss observed at 450°C in the TGA curve reveals some intriguing insights regarding the composition of the Ca/Si-2-R (3.20% weight loss) and Ca/Si-3-R (4.62% weight loss) samples. Notably, these samples exhibited a weight loss pattern that closely resembled that of the OPC sample (2.94% weight loss), while also demonstrating a lower weight loss compared to other samples in the study. This observation suggests that the Ca/Si-2-R and Ca/Si-3-R samples likely contain a significant proportion of ground RHA, an additive known for its high porosity, substantial SiO₂ content, and elevated surface area. The unique characteristics of ground RHA are critical in the context of its chemical reactivity. Due to its high porosity and surface area, RHA can interact more effectively with ground Ca(OH)₂, which is derived from CCR, compared to the reactivity provided by the SiO₂ present in fine granules of sodium silicate pentahydrate. This enhanced reactivity between the RHA and Ca(OH)₂ likely facilitates a more substantial conversion or reaction of the SiO₂ content from the RHA.

Microstructural study of hardened green cement pastes

The microstructure of cement pastes after 30 days of hydration was examined using SEM-EDX. The findings are illustrated in Figure 6, which presents the microstructure for the Ca/Si-2-based formulation, and Figure 7, which is dedicated to the Ca/Si-2-based formulation. For comparison, the microstructure of OPC paste was also analyzed.

Figure 6.

Microstructure of Ca/Si-2-based formulation cement pastes

(A) OPC.

(B) Ca/Si-2-R.

(C) Ca/Si-2-RS.

(D) Ca/Si-2-S.

Scale bar, 2 μm.

Figure 7.

Microstructure of Ca/Si-3-based formulation cement pastes

(A) OPC.

(B) Ca/Si-3-R.

(C) Ca/Si-3-RS.

(D) Ca/Si-3-S.

Scale bar, 2 μm.

Upon close inspection of Figure 6, the OPC paste exhibits a distinctly segmented surface characterized by irregularly shaped gel-like regions. The continuous matrix is attributed to C–S–H, based on its characteristic fibrous morphology, which forms a dense, sponge-like texture. This texture is marked by an intricate network of interwoven fibers that gives it a rough and uneven appearance, particularly when compared to more crystalline phases present in the paste. In contrast, portlandite (Ca(OH)₂) present in the OPC paste exhibits flat, plate-like crystals with smooth surfaces. These features make portlandite readily distinguishable from the fibrous, gel-like morphology of the C–S–H matrix. This stark contrast in texture between C–S–H and Portlandite underscores the differing hydration products formed during the cement hydration process.^35,36

Focusing on the Ca/Si-2-R sample, which incorporates RHA, a similar porous and sponge-like texture of C–S–H is detected, although intermixed with some fragments of ground RHA. The incorporation of this supplementary cementitious material enhances certain properties of the paste. Furthermore, in the samples designated Ca/Si-2-RW and Ca/Si-2-S, the dense texture of C–S–H is more pronounced when sodium silicate is utilized as an additive. Specifically, the Ca/Si-2-S sample, which comprises both Ca(OH)₂ and sodium silicate, reveals the presence of microcracking. The observed microcracking in the Ca/Si-2-S sample may be associated with early-age shrinkage in silicate-rich systems. Similar behavior has also been reported in Ca/Si-3-based cement paste formulations (Figure 7).

Furthermore, in the samples designated Ca/Si-2-RW and Ca/Si-2-S, the dense texture of C–S–H is significantly pronounced when sodium silicate is utilized as an additive. Specifically, the Ca/Si-2-S sample, which comprises both Ca(OH)₂ and sodium silicate, reveals the presence of microcracking. This behavior may be associated with local water redistribution and differential shrinkage during early-age reaction and drying, promoted by the presence of excess soluble silicate. Such shrinkage-induced microcracking has also been reported in silicate-rich cementitious systems, including Ca/Si-3-based formulations (Figure 7).

The elemental composition of cement pastes was analyzed through the entire area of SEM images at 500× magnification using energy dispersive X-ray spectroscopy (EDX). The weight percentage of each element was determined and subsequently converted to mole percentage by dividing by their respective molecular weights. The mole percentages, as well as the calcium-to-silicon (Ca/Si) and sodium-to-silicon (Na/Si) ratios for each sample, are presented in Table 5.

Table 5.

Element analysis of pastes based on full area mode (% mole)

Elements	OPC	Ca/Si-2-R	Ca/Si-2-RS	Ca/Si-2-S	Ca/Si-3-R	Ca/Si-3-RS	Ca/Si-3-S
Si	4.17	4.37	1.81	2.03	4.00	2.28	2.23
Ca	18.18	10.51	5.96	9.44	12.08	9.18	10.72
O	72.58	72.00	80.87	71.99	69.92	75.54	70.78
Na	0.35	12.17	11.11	13.58	11.30	11.28	13.36
Mg	0.46	0.45	0.10	0.35	0.45	0.28	0.64
Al	1.00	0.22	0.09	0.19	0.26	0.18	0.36
S	2.67	0.00	0.00	2.37	1.70	1.15	1.87
K	0.13	0.29	0.05	0.04	0.30	0.09	0.03
Fe	0.46	–	–	–	–	–	–
Ca/Si ratio	4.36	2.41	3.28	4.66	3.02	4.02	4.81
Na/Si ratio	0.08	2.79	6.12	6.70	2.83	4.94	5.99
Suma	4.45	5.19	9.41	11.36	5.85	8.96	10.79

^aSummation of Ca/Si ratio and Na/Si ratio.

The OPC sample exhibited a Ca/Si ratio of 4.36, with low detected Na/Si ratio of 0.08 contributing to a total of 4.45 for the combined ratios. In contrast, the cement preparation incorporated sodium silicate as an external source of SiO₂ and utilized NaOH as the alkali activator, which introduced sodium into the cement paste. The Ca/Si ratios across all cement samples varied from 2.41 to 4.81.

The Ca/Si molar ratio in alkali-activated concrete is a critical parameter influencing concrete performance. A low Ca/Si ratio can diminish the early strength of the concrete due to an insufficient concentration of Ca²⁺ ions, which impedes the necessary hardening reactions and subsequently affects the long-term performance. Conversely, an excessively high Ca/Si ratio leads to elevated levels of alkali and Ca²⁺ ions, potentially resulting in rapid reactions and the degradation of the gel phase, which adversely impacts the mechanical strength and durability of the concrete.^37,38,39 Previous studies have reported that the Ca/Si ratio of C–S–H in neat OPC pastes typically ranges from 1.2 to 2.3, with an average value around 1.75.⁴⁰

While the Ca/Si ratio alone might suggest that the strength of the cement samples should be comparable to that of OPC, the observed strengths are surprisingly lower than those of the OPC samples, as detailed in the following section. This indicates that the Ca/Si ratio is not the sole determinant of the physical properties of these materials, including compressive strength. Thus, it is essential to also consider the Na/Si ratio in conjunction with the Ca/Si ratio.

In the cement pastes, the incorporation of sodium silicate—specifically samples Ca/Si-2-RS, Ca/Si-2-S, Ca/Si-3-RS, and Ca/Si-3-S—resulted in elevated Na/Si ratios. When these ratios are summed, they yield a high (Ca+Na)/Si ratio. The Ca/Si-2-R and Ca/Si-3-R samples demonstrated (Ca+Na)/Si ratios of 5.19 and 5.85, respectively, which are closer to the OPC sample ratio of 4.45. This likely contributes to the superior strength observed in the Ca/Si-2-R and Ca/Si-3-R samples compared to others.

The Na/Si ratio in cement-based materials can significantly influence their properties, especially compressive strength, due to its effects on both the chemical and microstructural characteristics of the material. Sodium can react with silicon during hydration or alkali activation to form reaction products like sodium silicate hydrates or calcium sodium silicate hydrates, which in turn affect the microstructure and overall strength.⁴¹

Excess sodium may be associated with less dense gel structures or reactive phases that are more susceptible to microcracking. A favorable Na/Si ratio generally supports a denser gel phase and a more cohesive matrix by maintaining sufficient silicate availability for network formation. However, in the present dry-mix system using solid activators, microcracking is likely multifactorial: high Na⁺ availability may limit effective silicate cross-linking and increase early-age shrinkage, while incomplete dissolution of solid sodium silicate can act as weak inclusions, and minor bleeding may increase porosity. These factors can reduce matrix integrity and negatively affect strength development.^7,8 Previous studies have reported that maintaining a Na/Si ratio between 0.49 and 0.80 is beneficial for strength development.⁴²

Compressive strength of hardened cement pastes

The compressive strength of hardened cement mortars after 30 days of curing is a crucial metric for understanding their performance and potential applications in construction. As illustrated in Figure 8, the strength values are represented by bars that correspond to the left y axis, while the right y axis details two other critical parameters: the (Ca+Na)/Si ratio obtained from SEM-EDX mapping, and the percent weight loss observed at 450°C as computed from TGA-DTG curves. The interrelationship among these parameters illustrates the complex behavior of cement mortars as a function of their composition and curing conditions.

Figure 8.

Compressive strength of green cement pastes with (Ca+Na)/Si ratio

Data are represented as mean ± SD (n = 3).

The values referring to the percentage of weight loss (Table 4) and the (Ca+Na)/Si ratio (Table 5) show a clear relationship with the compressive strength of the cement mortars. OPC mortar exhibited a compressive strength of 22.8 MPa, while the green cement mortars showed lower strength values. Specifically, the Ca/Si-2-R sample achieved the highest strength within the green cement group (15.9 MPa). In contrast, Ca/Si-2-RS and Ca/Si-2-S registered lower strengths of 9.8 MPa and 4.9 MPa, respectively. A similar trend was observed in the Ca/Si-3 series (Ca/Si-3-R, Ca/Si-3-RS, and Ca/Si-3-S), which also exhibited reduced compressive strengths compared to OPC. From the strength results, the ∼16 MPa formulation is discussed as suitable for non-structural cementitious products, whereas the lowest-strength mixes (∼4–5 MPa) are more appropriately positioned as CLSM/flowable fill-type materials for backfill or void-fill applications where structural capacity is not required.

Although soluble silica is often reported to enhance binder gel formation, the SS- in this study used solid sodium silicate rather than a liquid activator. Under rapid-setting conditions, incomplete or delayed dissolution of solid sodium silicate may limit effective silicate availability for early-age gel formation; partially dissolved SS may also act as weak inclusions, and minor bleeding may increase porosity. These factors can reduce matrix integrity and lead to lower compressive strength. This highlights the importance of activator form and dissolution behavior in one-part systems and warrants further optimization (e.g., particle size control and mixing protocol).

The relationship between the thermal characteristics and the compressive strength can be explained by prior discussions on the decomposition of Ca(OH)₂ at temperatures around 450°C. A lower percent weight loss at this critical temperature indicates that a greater portion of Ca(OH)₂ remains within the cement paste, which suggests reduced interactions among the various source materials. Notably, the Ca/Si-2-R and Ca/Si-3-R mixtures included RHA, which can initiate a pozzolanic reaction between the SiO₂ content of the ash and the Ca(OH)₂ derived from CCR. It plays a significant role in cement hydration through the formation of C–S–H. This interaction facilitates the formation of C–S–H, a crucial component that enhances early strength development and contributes to the overall durability of the cement matrix, as highlighted in previous researches.^43,44

Furthermore, the (Ca+Na)/Si ratio is closely related to the weight loss behavior observed during thermal analysis, with lower ratios corresponding to reduced mass loss at 450°C. For the compositions investigated in this study, compressive strength exhibits a non-linear relationship with the (Ca+Na)/Si ratio, increasing toward an intermediate optimum range before declining at higher ratios. This trend indicates that an appropriate balance between alkaline species and available silica is critical for achieving a dense and mechanically stable binder matrix.

In construction and material science, understanding these relationships can inform the design and optimization of eco-friendly cementitious materials, further emphasizing the need for rigorous research into their interactions and reactions. For instance, as industry trends shift toward sustainability, the utilization of alternative cementitious materials such as RHA could become more prevalent in producing low-carbon concrete. The findings indicate that improving the pozzolanic activity through an adequate (Ca+Na)/Si ratio could potentially elevate the compressive strength of these cements, thereby making them viable substitutes for conventional OPC.

Recommendations for the development of cement

Based on the findings of this study, the incorporation of RHA and solid sodium silicate demonstrates strong potential for developing low-carbon, clinker-free binder systems. Future work should focus on optimizing mix proportions, particularly the alkali-to-silica balance, to improve mechanical performance and minimize early-age cracking. In addition, systematic evaluation of fresh-state properties, long-term durability, and scale-up feasibility is required to support practical implementation. The development of dry-mix, ready-to-use formulations represents a promising pathway toward sustainable cement technologies suitable for broader construction applications.

Economic impact of green cement from industrial waste

Production of OPC is an energy-intensive process and a major contributor to CO₂ emissions, primarily due to the high-temperature calcination and extensive use of fossil fuels. As Thailand pushes toward its bio-circular-green (BCG) economy and net-zero carbon goals, green cement emerges as a compelling and sustainable alternative. By incorporating CCR and RHA as key ingredients, green cement production significantly reduces raw material costs. These by-products, typically discarded as waste, are transformed into valuable inputs, simultaneously lowering environmental management expenses. Furthermore, the innovative dry-mix process eliminates the need for energy-demanding kilns, cutting energy consumption to approximately 3373.45 kJ/kg clinker⁴⁵ and directly reducing production costs.

Beyond cost savings, green cement offers substantial economic value through its low carbon footprint. This qualifies it for both domestic and international carbon credits, which can significantly boost profitability and market appeal. Unlike traditional OPC, which currently commands a higher market price in Thailand, reaching around $83/ton, green cement offers a more sustainable and potentially cost-effective long-term solution. The inherent environmental regulations that internalize the cost of pollution force traditional cement manufacturers to bear the expense of mitigating their environmental impact, further highlighting green cement’s advantage.

Ultimately, this solution strongly supports circular economy by reducing industrial and agricultural waste, conserving valuable natural resources, and decreasing reliance on imported raw materials and fossil fuels. This makes green cement an economically and environmentally sustainable choice for the future construction sector.

Limitations of the study

This study was conducted at the laboratory scale and focused on the early to medium curing-age compressive strength of the cementitious system. Long-term performance aspects, including durability, shrinkage, creep, and behavior under aggressive environmental conditions, were not evaluated. In addition, only a limited range of mix compositions and curing conditions was investigated, and field-scale performance was not considered. Further studies addressing these aspects are required to support practical application.

Resource availability

Lead contact

Further information and requests for resources should be directed to and will be fulfilled by the lead contact, Ubolluk Rattanasak (ubolluk@go.buu.ac.th).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •Data reported in this paper will be shared by the lead contact upon request.
• •This paper does not report original code.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Acknowledgments

The authors gratefully acknowledge the financial supports from the National Research Council of Thailand (NRCT), the National Science and Technology Development Agency (NSTDA), and the Department of Civil Engineering, Faculty of Engineering, King Mongkut’s University of Technology Thonburi (KMUTT) under the High-Potential Research Team Grant Program (grant no. N42A680319). Additional funding was provided by (1) Burapha University (BUU), (2) Thailand Science Research and Innovation, and (3) National Science Research and Innovation Fund (NSRF) (Fundamental Fund: grant no. 2.19/2568). Additionally, the first author would like to acknowledge the “Support by Research and Graduate Studies” Khon Kaen University.

Author contributions

Conceptualization, P.C., C.J., and U.R.; methodology, P.C., K.N., V.S., P.J., and U.R.; investigation, N.S. and U.R.; writing—original draft, U.R.; writing—review & editing, P.C., N.S., V.S., P.J., K.N., and C.J.; resources, C.J.; supervision, P.C. and C.J.; validation, P.J., V.S., and K.N.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
Calcium carbide residue (CCR)	Samut Sakhon, Thailand	N/A
Rice husk ash (RHA)	Central Thailand	N/A
Sodium silicate pentahydrate (commercial grade)	Local supplier	CAS No. 10213-79-3
Sodium hydroxide (micro-pearls, commercial grade)	Local supplier	CAS No. 310-73-2
Ordinary Portland cement	Local supplier	Siam City Cement
Other
X’Pert HighScore software	Malvern Panalytical	https://mystore.malvernpanalytical.com/
Scanning electron microscope (SEM)	Leo 1450VP, gold coating	https://www.zeiss.com
Fourier transform infrared spectroscope (FT-IR)	Perkin Elmer, Spectrum	https://www.perkinelmer.com
X-ray fluorescence (XRF)	Fischerscope, x-ray XUV 773	https://www.helmut-fischer.com/
Thermogravimetric analysis (TGA)	Mettler Teledo, TGA/DSC1	https://www.mt.com
X-ray diffraction (XRD)	Bruker, D2 Phaser	https://www.bruker.com/
Compression testing machine	Local supplier	CB-10M (100 kN)

Method details

Material preparation

CCR was sourced from a domestic acetylene production facility located in Samut Sakhon province, near Bangkok. The collected CCR was dried, ground with a high-speed grinder, and sieved through a mesh sieve with a 100 opening (0.149 mm). The resulting powder had a gray color and a distinct odor due to incomplete reactions of calcium carbide.^46,47 In addition, RHA was collected from a biomass power plant in central Thailand, where it was pulverized using a ball mill.

In this study, commercial-grade sodium silicate pentahydrate (Na₂SiO₃·5H₂O, SS) in fine granule was utilized in its original form. This powder consisted of 28.5% weight of SiO₂ and 28.5% weight of Na₂O, with particle sizes ranging from 200 to 500 microns. Additionally, commercial-grade sodium hydroxide (NaOH, SH), known as caustic soda micro-pearls, was employed also in the original form, featuring a 99% purity and particle sizes between 0.5 to 1.10 mm. Both SS and SH served as alkaline activators. For the mortar mixes, fine aggregate was graded river sand passed through a No. 16 sieve (with 1.18 mm openings) and retained on a No. 100 sieve (with 150 μm openings), yielding a fineness modulus of 2.8. Distilled water was used throughout the experiment. Ordinary Portland cement (OPC) was also used for comparison purpose.

Preparation of green cement pastes and mortars

According to Table 1, the components (i.e. calcium carbide residue, rice husk ash, sodium silicate and sodium hydroxide) were thoroughly blended until homogeneous. Water was then incorporated into this mixture, and the resulting paste was molded. A silicone mold with internal dimensions of 30 mm × 30 mm × 20 mm was used for sample casting. The specimens were cured at 65°C for 24 h to accelerate early-age reaction and strength development for proof-of-concept evaluation of the alkali-activated binder system. After curing, the specimens were demolded and stored in sealed zip-lock bags. This controlled heat-curing regime was adopted for laboratory investigation and does not imply a requirement for practical field application. Optimization of ambient curing conditions suitable for dry-mix and on-site construction applications should be addressed in future studies.

For the mixing of mortar intended for strength testing, sand was first combined with the other dry materials, followed by the addition of water. The fresh mortar was then cast into 5 cm cube molds as per ASTM C109/C109M-20,⁴⁸ with the same heat-curing treatment A sand-to-cement ratio of 2 was maintained. For comparative purposes, mixture with ordinary Portland cement was also prepared.

Testing and characterization

The morphological analysis of hardened cement pastes across various formulations was conducted using X-ray diffraction. Additionally, the samples were subjected to thermogravimetric analysis (TGA-DTG, Mettler Teledo, TGA/DSC1, 30-800°C, 10 C/min heating rate, N₂ gas). The hardened pastes were first ground and then sieved through a 100-mesh sieve (with an opening of 0.149 mm) prior to XRD and TGA evaluations. For a detailed examination of the microstructural properties of the paste surfaces, element mapping was performed using scanning electron microscopy coupled with energy-dispersive X-ray spectroscopy (SEM/EDX, Leo 1450VP, gold coating).

Furthermore, the setting time of the paste was measured using the Vicat needle method in accordance with the ASTM C191-21.⁴⁹ For the cement mortars, compressive strength tests were carried out after a curing period of 30 days. The reported values were the average of three samples.

Quantification and statistical analysis

Quantitative data obtained from repeatable experiments were reported as the average of three independent samples and analyzed at a significance level of p < 0.05. Data obtained from instrumental characterization techniques, including XRF, XRD, TGA, and SEM-EDS, were presented as representative measurements (Tables 2, 3, 4, and 5).

Published: March 7, 2026