Influence of Dregs Waste on the Alkali–Silica Reaction: A Comparative Analysis among Different Types of Cement

Abstract

Dregs waste, a byproduct from green liquor clarification in the pulp industry, is increasingly generated as global cellulose production rises. This accumulation, along with its underutilization, presents environmental challenges and opportunities for reuse. The study focuses on the high alkali content in dregs, which can impact cement durability. The main objective is to analyze the effect of dregs on the alkali–silica reaction in mortars. Dregs were incorporated into mortar mixtures at 0%, 5%, 10%, and 15% proportions relative to cement mass, using six types of Brazilian cement and a blend with silica fume. The alkali–silica reactivity was assessed via the accelerated mortar prism test (ABNT NBR 15577-4:2018), with compressive strength tests and scanning electron microscopy evaluating structural integrity and microstructural changes. The results indicated that adding 5% dregs improved compressive strength in certain mortars, such as CP V-ARI RS, CP II-F, and CP IV. However, at 15% dregs, compressive strength significantly decreased, particularly in CP V with 10% silica fume. Cements with high pozzolanic content, such as CP IV and CP III, showed strong potential to inhibit ASR expansion. However, of the 28 mixtures analyzed, only four containing CP III had expansions within the limits set by standards. This study highlights the potential of incorporating dregs as a supplementary material in cement, promoting sustainability in the industry and reducing environmental impact.

1. Introduction

Population growth and increased demand for products of all types have led to an accelerated depletion of the planet’s natural resources. In Brazil, for example, resources were exhausted by August 2023, indicating that consumption exceeded their renewal capacity [1]. This is, unfortunately, a global reality that requires changes in production and consumption patterns.

In this context, Portland cement, widely used in construction, significantly contributes to environmental impact. This is because its production requires high temperatures, which consume a large amount of energy and generate significant CO₂ emissions [1,2]. Additionally, substantial environmental impacts are generated from the extraction of raw materials for cement production [3].

Several initiatives aim to encourage the development and adoption of more sustainable technologies and materials in construction, such as the United Nations’ Sustainable Development Goals (SDGs), which promote sustainable buildings and resilient infrastructure. Another example is related to the Paris Agreement, which sets global goals to reduce greenhouse gas emissions. Therefore, the search for more sustainable alternatives to conventional Portland cement is a way to contribute to achieving these goals [4,5].

There is a significant need for solutions that promote the reduction in polluting gas emissions and the consumption of natural resources. One alternative is using chemically active mineral additions, such as metakaolin, blast furnace slag, and fly ash, which can partially replace cement in the production process [2]. Another alternative is adopting the filler effect in cementitious materials, thereby reducing cement consumption and increasing resistance and durability.

Waste materials can also be a viable option for civil construction to mitigate the environmental impact of raw material extraction and disposal, provided they meet the minimum requirements for their use. Examples include construction and demolition waste, mining waste, and industrial waste.

The cellulose industry has experienced substantial growth in recent years, propelled by increased demand for paper and cellulose-derived products. According to IBÁ [6] data, Brazil ranks as the world’s second-largest cellulose producer, trailing only the United States. In 2022, Brazil’s cellulose production escalated to 25 million tons, marking a 19% increase from the previous year [6]. This expansion is attributed to enhanced production capacities and rising demand from the domestic and foreign markets.

Several industries use the Kraft process to produce cellulose, which is considered an efficient and low-cost method, as it allows the production of large quantities of cellulose from a wide variety of wood. In the Kraft process, the recovery plant is included, which is responsible for recovering chemicals that are used in the process, in addition to producing energy for the factory’s operations [7,8,9]. However, from the reagent recovery processes, some solid waste is generated, the main ones being dregs, grits, and lime mud.

Among the wastes, only the lime mud is, for the most part, primarily destined for the recovery plant, where it is burned and reused as lime in the hydrator. Dregs are generated from the clarification of green liquor and grits from the lime hydration process. Recent data highlight dregs as the main waste generated, compared to grits; by industries, it is noted that generation varies between 11.34 and 54.32 kilos per ton of cellulose produced (

Table 1. Dregs and grits generation.

Company	Cellulose Production (ton/year)	Dregs (kg/ton)	Grits (kg/ton)
A (2021)	1,250,000	11.34	2.15
B (2022)	1,700,000	10.08	1.63
C (2023)	2,063,093	54.32	3.60

Source: sustainability reports from factories except C.

).

Data from a national survey, collected from six of the largest cellulose factories in Brazil, indicate that approximately 17.3 million tons of cellulose are produced annually via the Kraft process. This production corresponds to an estimated 196 million tons of waste annually. When extrapolated to the global cellulose production level of 130 million tons/year [10], waste generation could reach up to 1474 million tons annually. Given this substantial waste generation and its underutilization, significant opportunities exist for its reuse, presenting both challenges and potential.

The use of waste materials has been studied across various fields, such as in the production of geopolymer mortars and bituminous mixtures [9,10]. Also, dregs have been the subject of increasing investigation. Research indicates great potential for using the waste, although there are gaps in understanding the topic.

Martínez-Lage et al. [11] developed a study on mortars with mass replacement of 10, 20, and 30% of cement with dregs residue. In the flexural tensile strength test with 10% replacement, the mortar with waste showed a similar result to the reference, 8.5 MPa. For compressive strength, it presented a result in the order of 4%, higher than the reference (47.7 MPa). As the percentage of residue increases, resistances tend to decrease for both tests.

In the same study, the researchers examined concrete incorporating 10% and 20% dregs residue as a bulk cement substitute. They found that compressive strength decreased by about 20% and 35% for the 10% and 20% substitutions, respectively, compared to a standard reference of 46.3 MPa. The diametral compression strengths were also lower, decreasing by approximately 18% and 23% for the 10% and 20% substitutions. Furthermore, there was a slight decrease in the elastic modulus with higher dregs content.

Santos et al. [12] produced mortars by replacing hydrated lime in percentages of 7% and 14% with dregs residue. It was observed that it is possible to use small amounts (less than 8%) of dregs in mortars. However, they reduced mechanical resistance and increased water absorption in larger quantities.

Torres et al. [13] replaced the clinker in percentages of 2.5, 5, 7.5, 10, and 15% with the residue. The results demonstrated that additions of up to 10% of dregs were satisfactory, within the limits established in Brazilian standards regarding the maximum percentage of carbonate material allowed, up to 25% for CP II-F (Portland Cement Type II with filler) [14].

Oliveira et al. [15] promoted cement replacement in mortars, with 5, 10, 20, and 30% percentages with dregs waste. The results demonstrated a decrease in the consistency index and incorporated air content as replacement increased. The residue content did not significantly influence water absorption by immersion. The presence of dregs reduced the dynamic modulus of elasticity and compressive strength of the reference mortar. However, even with this reduction, the mortars studied remain within the resistance classes of Portland cements commercially available in Brazil [14].

In this paper, the effects of adding dregs to cementitious matrices are studied, focusing on the alkali–silica reaction (ASR) and compressive strength, in addition to a detailed analysis of the microstructure of the samples subjected to the ASR test. The selection of this test is based on the presence of Na₂O in the waste, as reported by several researchers in the literature [8,11,16,17], which is possibly the main challenge for the processing of dregs waste, which had not yet been addressed in current literature.

The alkali–silica reaction (ASR) is triggered by the interaction of alkaline hydroxides (present in cement, mixing water, chemical additives, and additions) with certain minerals present in the aggregates [18,19,20]. ASR is harmful to hardened concrete, causing cracks, differential displacements, chipping, increased permeability, and decreased mechanical resistance due to its expansive characteristics [18,21]. According to Kihara and Scandiuzzi [22], this cracking can be associated with the formation of an expansive gel around the aggregates, with its intensity depending on the amount of available (soluble) alkali, intrinsic characteristics of the aggregates, and environmental conditions (humidity and temperature).

Using dregs can reduce the need for natural mineral addition, encouraging a reduction in cement consumption and the need to dispose of waste in landfills, thus generating positive impacts on the environment. However, due to the high alkali content in dregs, there is a concern that their incorporation into cementitious mixtures could exacerbate the ASR, rather than mitigate it. This study seeks to investigate this specific issue by examining the impact of different proportions of dregs on the ASR in mortars. The goal is to determine whether dregs can be safely utilized in cement production without negatively affecting the durability of concrete, or whether alternative strategies are needed to mitigate the potential adverse effects of this industrial byproduct.

2. Materials and Methods

2.1. Materials Characterization

2.1.1. Dregs

The dregs waste comes from a Brazilian cellulose factory in southern Brazil. Six samples (Figure 1) were collected at different periods to verify homogeneity over time, and characterization tests were conducted.

From Figure 1, it is observed that all the samples exhibit similar visual characteristics, with color variations ranging between light gray and dark gray tones The chemical composition of the samples, obtained by X-ray Fluorescence, is presented in

Table 2. Chemical composition of dregs.

Component (%)	Sample 1	Sample 2	Sample 3	Sample 4	Sample 5	Sample 6
TiO2	0.138	0.303	0.044	0.104	0.109	0.124
Fe2O3	1.919	1.849	2.894	2.268	1.467	0.117
MnO	0.246	0.223	0.349	0.279	0.195	0.143
CaO	49.127	47.356	43.791	41.836	48.846	51.804
MgO	4.052	3.689	5.392	5.168	3.499	2.526
K2O	0.050	0.113	0.135	0.142	0.087	0.066
Na2O	4.187	6.122	6.922	9.870	5.438	3.697
P2O5	0.281	0.345	0.400	0.333	0.360	0.522
Alkaline equivalent (Na2Oeq)	4.219	6.196	7.010	9.963	5.495	3.740

.

The environmental classification of the waste was obtained from leaching [23] and solubilization [24] tests. The elements present in the leached and solubilized extracts were analyzed with ICP (Inductive Coupling Plasma). For this, dilutions of 15× and 150× were carried out; the results of leching and solubilization tests are presented in

Table 3. Results and limits for the leaching test (in mg/L).

Element	Sample 1		Sample 2		Sample 3		Sample 4		Sample 5		Sample 6		Limits
	15×	150×	15×	150×	15×	150×	15×	150×	15×	150×	15×	150×	NBR 10004
Ca	121.62	128.44	1.89	4.29	1.52	4.36	2.88	3.66	3.18	7.59	122.3	163.8	-
Mg	115.19	116.12	38	39.74	18.02	17.58	17.01	10.59	54.38	57.47	189.2	232	-
Al	0	0	0.2	0	0	0	0.06	0	0.03	0	0	0	-
Na	2100	2231	3313	3655	4471	4878	6137	5849	3411	3656	1613	1948	-
Cu	0	0	0	0	0	0	0	0	0	0	0	0	-
Mn	0.03	0.08	0.01	0	0	0	0	0.02	0	0	0.02	0.08	-
Cr	0	0	0	0	0	0	0	0	0	0	0	0	5
Ba	0.21	1.52	0.15	1.98	0.18	1.77	0.17	1.67	0.18	1.9	0.32	2.02	70
Pb	0	0	0	0	0	0	0	0	0	0	0	0	1

and

Table 4. Results and limits for the solubilization test (in mg/L).

Element	Sample 1		Sample 2		Sample 3		Sample 4		Sample 5		Sample 6		Limits
	15×	150×	15×	150×	15×	150×	15×	150×	15×	150×	15×	150×
Ca	0.87	1.82	0.69	1.47	3.44	4.5	1.44	4.4	1.01	1.55	0.9	1.37	-
Mg	0.17	0	0.08	0	0.93	0.36	0.3	0.13	0.1	0	0.86	0.32	-
Al	0.87	0	1.21	0	0.57	0	0.15	0	0.71	0	0.05	0	0.2
Na	3807.8	4288	8321	10,376	12,487	17,261	11,460	16,040	8668	10,657	2073	2102	200
Cu	0.05	0	0	0	0	0	0	0	0	0		0	2
Mn	0	0	0	0	0	0.01	0	0.04	0.01	0.03	0.01	0.01	0.1
Cr	0.08	0	0.04	0	0.1	0	0	0	0.05	0		0	0.05
Ba	0.17	1.73	0.16	1.66	0.16	1.58	0.15	1.81	0.16	1.53	0.17	1.71	0.7
Pb	0	0	0	0	0	0	0	0	0	0	0	0	0.01

, respectively. Considering that the values obtained in the leaching test do not exceed the limits established by NBR 10004 [25], dregs waste is characterized as non-hazardous. However, in the solubilization test, the elements aluminum, sodium, chromium, and barium presented values above the limits established by the standard; therefore, they are characterized as not inert. The final classification of the waste is non-inert class II A.

The mineralogical analysis of the samples was carried out using X-ray Diffraction, with the Philips X-ray diffractometer, model X’Pert MDP (X-ray tube with Cu radiation) in the range of 5–80°. The analysis was carried out using the X’Pert High Score software Version 5.2. Figure 2 shows the diffractograms of the different dregs waste samples. It is possible to observe remarkable similarity in the position and intensity of the peaks. The samples demonstrate that they are mainly composed of calcium carbonate in the form of calcite (CaCO₃). Furthermore, sodium nitrate is present as nitrate (NaNO₃).

The thermal test by thermogravimetry was carried out with equipment model TGA-50 (Shimadzu) with a platinum crucible, using a standard heating rate of 10 °C/min between room temperature and 1000 °C. From the results (Figure 3), it is noted that in all curves, there is a significant mass loss between 600 and 800 °C, which is related to the decomposition of calcium carbonate [26].

Table 5. Mass loss of calcium carbonate in dregs samples.

Compound	Sample
	1	2	3	4	5	6
CaCO3	85.56	82.44	75.64	74.60	86.99	86.77

presents the percentages of mass loss obtained for the products in the dregs waste samples, calculated according to Equation (1), which uses the molar relations between CaCO₃ (100.09 g/mol) and CO₂ (44.01 g/mol).

$$\mathrm{\%}{CaCO}_3={\displaystyle \frac{M_{{CaCO}_3}}{M_{{CO}_2}}}\times \mathrm{\%}{CO}_2$$

(1)

From the characterization tests, it was verified that there were no representative changes between the samples. Thus, the homogeneity of the residue over time was verified.

For preparation, the samples were dried (T: 100 ± 5 °C) and briefly deagglomerated with a porcelain pestle and mortar. The final sample consisted of varying amounts of different samples. Afterward, the residue was subjected to mechanical treatment in a ball mill for 2 h. The time of 2 h was defined in a grinding study, which showed the greatest reduction in particle size. The proportion of waste to porcelain balls was 1:1, and the balls selected were of different sizes to fill the voids between them. After treatment, the residue has a specific mass of 2.45 g/cm³ and D₅₀ of 12.26 μm.

2.1.2. Silica Fume

Table 6. Chemical composition of silica fume.

Component	SiO2	Al2O3	TiO2	Fe2O3	MnO	CaO	MgO	K2O	Na2O	P2O5	L.O.I. 1	A.E. 2
Silica fume	89.49	-	0.04	2.36	0.1	0.51	0.79	2.53	1.19	0.08	2.93	2.85

¹ L.O.I.: loss on ignition; ² A.E.: alkaline equivalent.

presents the chemical composition of silica fume in percentage.

From the concentration of major elements in silica fume, a high alkaline equivalent of 2.85 is noted. Laser granulometry demonstrates a high particle size, with D₉₀ being 69.02 μm and D₅₀ of 25.01 μm. The specific mass of silica fume is 2.20 g/cm³.

2.1.3. Cements

The cements CP V-ARI cement (Portland Cement Type V High Early Strength, similar to type III of C150 [27]), CP V-ARI RS (Portland Cement Type V sulfate-resistant, similar to type II of C150 [27]), CP II-Z 32 (Portland Cement Type II with fly ash, similar to type IP of C595 [28]), CP II-F 40 (Portland Cement Type II with filler, similar to type IL of C595 [28]), CP IV-32 [Portland Cement Type II with fly ash, similar IP of C595 [28]) and CP III-RS 40 (Portland Cement Type II with Blast Furnace Slag, similar to the IP type of C595 [28]). In addition, CP V-ARI was combined with the addition of 10% silica fume in relation to the mass of the cement.

Table 7. Chemical composition of the cements.

Component	CP V-ARI RS	CP V-ARI	CP II-Z	CP II-F	CP IV	CP III
SiO2	13.35	18.43	7.54	14.21	35.81	14.79
Al2O3	10.38	3.53	8.09	3.75	10.17	11.15
TiO2	0.52	0.24	0.56	0.26	0.54	0.77
Fe2O3	4.73	2.76	4.05	2.63	3.59	4.67
MnO	0.16	0.04	0.11	0.06	0.05	0.32
MgO	11.28	2.80	10.78	4.62	4.39	10.51
CaO	54.66	63.18	49.23	63.12	36.16	50.66
Na2O	-	n.a.	0.10	-	0.15	0.11
K2O	1.23	0.86	0.87	1.11	1.33	0.40
P2O5	0.20	0.12	0.17	-	0.06	0.132
L.O.I. 2	n.a.1	3.60	n.a. 1	5.68	3.4	n.a. 1
A.E. 3 (Na2Oeq)	0.80	1.52	0.66	0.73	1.02	0.47

¹ n.a.: not analyzed; ² L.O.I: loss on ignition; ³ A.E.: alkaline equivalent.

presents the chemical composition of the cements used in the research.

Regarding thermogravimetric analyses, exothermic peaks up to 200 °C are generally observed for all. This mass loss can be related to the loss of free and combined water present in the samples. Furthermore, two other high-intensity peaks were identified close to 400 °C and 650 °C. The peak close to 400 °C can be correlated with the decomposition of Portlandite (Ca(OH)₂). The peak around 650 °C can be attributed to the decomposition of calcium carbonate (CaCO₃).

Table 8. Mass losses of products present in cement samples.

Cement	Weight Loss (%)
	Ca(OH)2	CaCO3	Total
CP V-ARI	1.73	12.25	13.98
CP V-ARI RS	1.23	4.23	5.47
CP II-Z	1.19	11.35	12.54
CP II-F	1.73	11.06	12.79
CP IV	0.58	11.06	11.63
CPIII	1.23	7.40	8.63

presents the mass loss percentages obtained for the products present in the cement samples.

Table 9. Physical characteristics of the cements used.

Parameter		CP V-ARI RS	CP V-ARI	CP II-Z	CP II-F	CP IV	CPIII
Specific mass (g/cm3)		3.02	3.01	3.07	3.07	2.67	2.96
Laser granulometry	D10 (μm)	2.53	2.50	1.99	2.02	3.48	2.022
	D50 (μm)	12.37	11.31	11.54	10.63	15.95	11.99
	D90 (μm)	30.11	20.84	28.52	27.70	43.86	29.01
	Dmedium (μm)	15.48	12.00	14.16	14.89	21.13	14.50

presents the physical characteristics of the cements used.

2.1.4. Aggregates

The aggregate used comes from natural extraction from a siliceous origin. The particle size distribution used was that required by NBR 15577-4 [29], as shown in

Table 10. Particle size distribution required by NBR 15577-4 [29].

Mesh Opening (mm)	Percentage of Material Retained (%)
2.36	10
1.18	25
0.6	25
0.3	25
0.15	15

.

2.1.5. Water

The water used went through a deionization process.

2.2. Alkali–Silica Reaction

The alkali–silica reaction test using the accelerated mortar prism method was conducted based on NBR 15577-4 [29], similar to C1260 [30]. For each cement, four mixtures were adopted. The first, without addition, was called REF; that is, the expansion results for these compositions must be solely related to the concomitant action of the cement and the aggregate. In sequence, three percentages of dregs addition were used, namely: 5% (5D), 10% (10D), and 15% (15D). Such adoptions corroborate Dal Molin [31], who highlights that fillers can improve concrete properties when used in proportions of up to 15%. The additions aim to analyze the influence of increasing levels in cement matrices and the mitigation potential of the different types of cement used.

The mixture was 1:2.25, with a water/cement ratio of 0.47. The addition levels were calculated in relation to the cement mass. The amount of water remained constant, and no additives were used to standardize the consistency. This measure was taken to reduce variables in the results.

Table 11. Quantities, by mass, of the materials used for the mixtures.

Composition	Addition Content	Cement (g)	Aggregate (g)	Active Fume (g)	Dregs (g)	Water (g)
CP V-ARI, CP V-ARI RS, CP II-Z, CP II-F, CP IV and CP III	0%	440	990	0	0	206.8
	5%	440	990	0	22	206.8
	10%	440	990	0	44	206.8
	15%	440	990	0	66	206.8
CP V + 10%SA	0%	440	990	44	0	206.8
	5%	440	990	44	22	206.8
	10%	440	990	44	44	206.8
	15%	440	990	44	66	206.8

presents the quantities of materials, by mass, used in the mixtures.

2.3. Compressive Strength

The compressive strength test was used as a control parameter for the mortars tested for alkali–silica reaction (Figure 4). The traces and mixing procedure were identical to those used for ASR. Molding and curing, using test specimens of 5 × 10 cm, followed the guidelines of NBR 7215 [32] similar to ASTM C39 [33].

2.4. Scanning Electron Microscopy (SEM)

After 63 days, the specimens conditioned in a thermal bath with sodium hydroxide solution were fractured. Samples were selected for the test and went through the hydration stop process with isopropyl alcohol, following Scrivener, Snellings, and Lothencach [26] recommendations.

The samples were positioned on supports and subjected to gold metallization for the test. The conventional tungsten filament scanning electron microscope Zeiss EVO 10 (Figure 5) was used, operating at voltages ranging from 0.2 kV to 30 kV. The analyses were performed using secondary electrons. A sample was selected for each composition with 0% and 10% additions that presented an expansion above the limit of NBR 15577-1 [34].

3. Results and Discussion

3.1. Expansion by Alkali–Silica Reaction

Figure 6 shows the expansion results at 30 days for all compositions. The results are classified as potentially innocuous grade R0 (less than 0.19%) and potentially reactive grade R1 (between 0.19 and 0.40%), R2 (between 0.41 and 0.60%), and R3 (higher than 0.60%). The classifications are standardized by Table 2 of NBR 15577-1 [34]. It is important to highlight that these classifications refer to the classification of mortars made with standard cement. For this research, they are also used to classify mortars with different types of cement and dregs waste addition levels as a comparative standard between the results obtained.

From the above, it is noted that mortars are categorized as potentially harmless only for CP III cement. According to NBR 16697 [14], CP III comprises granulated blast furnace slag in the range of 35 to 75%. Slag is characterized as a cementing material resulting from sudden cooling that does not allow the formation of crystals, making it amorphous and highly reactive [31]. Several studies have shown that introducing granulated blast furnace slag into cementitious matrices decreases the concentration of alkaline hydroxides in the pore solution. This effectively promotes the mitigation of expansions [35,36,37]. However, this research shows that it was also influential in mitigating the possible negative impacts generated by the alkalis present in the dregs waste.

Similarly, CP IV has great potential for mitigation with values close to the limit, but the samples are still classified as grade 1 reactive. Like CP III, CP IV is a reference for its potentially ASR mitigating properties due to the high percentage of up to 50% addition of pozzolanic material [14].

CP V–ARI has high purity and lacks pozzolanic additions or granulated blast furnace slag. As a result, it is observed that the mortars reach reactivity level 2. This behavior is expected due to the absence of chemically active additions that help mitigate ASR [20,38]. In this context, several studies find significant advantages in expansion when incorporating pozzolanic additions, in contrast to cement without additions [35,36,39,40,41].

The composition of CP V with the addition of 10% silica fume (CP V+10%SA) demonstrates that, although silica fume is considered a pozzolanic material, in this case, it was not highly effective in mitigating ASR. Therefore, some factors must be considered, including the high alkaline equivalent (2.85) and the considerable size of the silica fume particles (D₅₀ = 25.01 μm), possibly contributing to reducing the silica fume’s chemical and physical action. Due to transportation difficulties, many producers promote the densification of silica fumes [42,43]. However, during the cementitious material mixing process, it is expected that these particles will disagglomerate, which does not always occur, especially in the absence of a superplasticizing additive [42]. As a result, studies indicate that using densified silica fume can significantly lose performance [42,43,44]. The results obtained verified that the mortars of this mixture are classified as potentially reactive grade 1, except for the mortar with a 15% addition of dregs, which is presented as R2.

CP V-ARI RS has advantages over CP V-ARI since the additions tend to react with the alkalis present in the cement, thus reducing their availability to trigger ASR. Mortars with this cement achieved lower expansions than CP V-ARI without addition and were classified as R1. NBR 16697 [14] does not define addition levels for this type of cement. However, it indicates that blast furnace slag or pozzolanic materials can be added. This behavior is related to the presence of additions in the cement; however, they are not in sufficient quantity to inhibit the reaction.

$$\mathrm{A}\mathrm{S}\mathrm{R} \mathrm{m}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{g}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\left(\mathrm{\%}\right)={\displaystyle \frac{\mathrm{R}\mathrm{E}\mathrm{F} \mathrm{m}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{r} \mathrm{e}\mathrm{x}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}-\mathrm{C}\mathrm{o}\mathrm{m}\mathrm{p}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{o}\mathrm{n} \mathrm{m}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{r} \mathrm{e}\mathrm{x}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}}{\mathrm{R}\mathrm{E}\mathrm{F} \mathrm{m}\mathrm{o}\mathrm{r}\mathrm{t}\mathrm{a}\mathrm{r} \mathrm{e}\mathrm{x}\mathrm{p}\mathrm{a}\mathrm{n}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}}}\times 100$$

(2)

CP II–Z, a cement composed of pozzolanic material in the range of 6 to 14%, demonstrated high expansions, indicating that these addition levels cannot mitigate the reaction generated in the mortars. The order of magnitude of the results is similar to that obtained for CP V–ARI. Except for the 15% added content of dregs, classified as R3, the others are potentially grade 2 reactive.

Among all the types of cement used, it appears that the worst performance obtained is related to the use of CP II–F. This cement does not include the addition of pozzolanic materials. However, it involves mass-releasing 11 to 25% clinker with carbonate material. The performance of adding carbonate material is mainly linked to the physical effect [45]. As a result, the expansions exceed those found for other types of cement used, including CP V-ARI. All compositions are classified as R3.

3.1.1. Evaluation of the Different Cements Used

The mitigation percentages obtained with the use of different cements in relation to CP V-ARI and CP II-F, calculated according to Equation (1), are presented in Figure 7. CP V-ARI and CP II-F were chosen as references because they have the lowest content of pozzolanic additions among the cements used.

Comparing CP III with CP V-ARI, it is noted that CP III achieves high percentages of mitigation, reaching 92% for the addition content of 5% of residue. Likewise, the use of CP IV with CP V-ARI presents great advantages due to the high percentages of mitigation, being, again, the highest, in the order of 63%, for the addition content of 5%. High mitigation percentages are also observed when comparing CP III and CP IV cements concerning CP II-F, which for 5D reaches 77% and 95%, respectively. This property is intrinsically linked to the high percentage of additions present in cements.

In line with the results obtained in this paper, Tiecher [20], using different types of cement and aggregates, observed that cement with pozzolanic additions can promote a reduction in ASR. Among the types of cement used, CP IV (with the addition of approximately 27% fly ash) demonstrates a greater mitigating capacity than CP II-Z cement (with the addition of roughly 10% fly ash), as a result of which the CP II-Z addition percentage is not capable of inhibiting harmful expansions in most of the samples analyzed.

A study by Couto [46] demonstrates the positive effects of CP IV for mitigating ASR compared to CP II-F cement, which presents expansions of a greater order. Dutra [47] and Santana et al. [48] also obtained favorable results when cements with a high percentage of addition were used (CP IV and CP III).

Other cements, such as CP V ARI-RS and CP V+10%SA, which have lower percentages of pozzolan, cannot mitigate ASR despite showing notable reductions in comparison with CP V-ARI.

An analysis of variance (ANOVA) was performed to evaluate whether the different factors cause statistically significant differences for ASR, and an analysis of variance (ANOVA) was performed with a 95% confidence interval (

Table 12. ASR expansion variance analysis.

	Source of Variation	Degree of Freedom	Sum of Squares	F	P
Type of cement	4.21446	6	0.70241	176.547	0.000000
Addition	0.06898	3	0.02299	5.779	0.001703
Type of cement × addition	0.04880	18	0.00271	0.681	0.813372
Error	0.21087	53	0.00398

).

Through analysis of variance, it is possible to identify that all controllable factors (addition and type of cement) are statistically significant when analyzed individually in relation to the response variable (alkali–silica reaction). Furthermore, the interaction between controllable factors is also significant in all cases tested.

As shown in Figure 6, the addition of dregs significantly influenced ASR expansion depending on the cement type. For example, mortars using CP II-F with 15% dregs were classified as R3 (highly reactive), indicating a substantial increase in expansion. In contrast, mortars using CP III and CP IV cements, which have higher pozzolanic content, demonstrated lower expansions and were more effective in mitigating ASR, despite the dregs addition. These observations are supported by the ANOVA results in Table 12, which confirm that both the type of cement and dregs addition are statistically significant factors affecting expansion.

Figure 8 demonstrates the effect of the interaction between cement type and the additions’ content on expansion.

This expansive behavior with the addition of materials in cementitious matrices was also observed in papers with the addition of rice husk ash (CCA). Due to the influence of alkaline impurities and the presence of silica content, CCA can react with the sodium ions present in the pore solution and form the alkali–silica gel, leading to the expansion of the mortar. However, it is essential to highlight that the expansive characteristics brought by addition are related to the content and physical characteristics of the particles [49,50,51].

3.1.2. Comparison between Ages 30 and 63 Days

NBR 15577–4 [29] suggests that the alkali–silica reaction test of mortar bars using the accelerated method be completed after 30 days. At this age, the cementitious composite can be classified as potentially harmless or reactive. However, aiming for a more comprehensive analysis of expansions after the initial period of 30 days, the trial was extended to 63 days in this study. This extension allows for a more complete assessment of the behavior of expansions, considering possible variations beyond the standard period established by the standard.

Figure 9 shows the expansion of each cement between the ages of 30 and 63 days.

In most cases, a considerable increase in expansion is noted between the two ages due to the continuity of the alkali–silica reaction, see Hasparyk [52].

Contrary to what was obtained at 30 days, CP II-Z presents lower expansion rates than CP V-ARI at 63 days. The presence of pozzolan in CP II-Z (6–14%) can influence the reduction of expansion, as even in modest amounts, it can continue to react with the cement constituents over time, contributing to dimensional stability. The same does not occur for CP V-ARI, which does not have chemically active additions.

3.2. Compressive Strength

Figure 10 presents the average compressive strength values and the respective standard deviations for the analyzed samples, according to the criteria established by NBR 7215 [32].

The isolated effect of dregs addition on compressive strength highlights that a 5% addition of dregs improved the mechanical properties in mortars with CP V-ARI RS, CP II-F, and CP IV. However, with a 15% dregs addition, a significant decrease in compressive strength was observed, particularly in CP II-Z and the CP V+10% silica fume compositions, indicating a threshold where the dregs content negatively impacts strength performance. In general, there is a loss in compressive strength as the additional content increases.

The ANOVA (

Table 13. Compressive strength variance analysis.

	Source of Variation	Degree of Freedom	Sum of Squares	F	P
Type of cement	4641.4	6	773.6	368.2	0.000000
Addition	69.7	3	23.2	11.1	0.000002
Type of cement × addition	123.1	18	6.8	3.3	0.000060
Error	247.9	118	2.1

) demonstrates a statistically significant difference between all the factors analyzed.

It is observed that as the percentage of dregs increases, there is a reduction in compressive strength, which is more notable with a 15% addition (Figure 11).

Figure 12 demonstrates the interaction between different cement types and dreg additions. Notably, CP V-ARI with 10% silica fume exhibited the highest compressive strength values across all dregs content levels, while CP IV showed lower overall strength compared to other cement types. These trends align with the findings from ANOVA (Table 13), which confirm the statistical significance of both the cement type and dregs content on compressive strength.

In this context, the cements have different resistance classes. Therefore, strengths should only be compared within the same cement group, considering different addition levels.

Oliveira et al. [15], using levels of up to 30% of dregs as a replacement for mortar cement, observed a reduction in compressive strength. For the author, what happened is related to the reduced water/cement ratio and the fixed water/fines ratio that may have provided the dilution effect. Furthermore, there is an increase in resistance for the 5% content, as also observed in some cases in this work.

Martínez-Lage et al. [11] promoted cement replacement by dregs in mortars at percentages of 10, 20, and 30% and observed that for compressive strength, the 10% content demonstrated results superior to the reference.

Several studies already carried out show that high levels of cement replacement with dregs residue cause a considerable attenuation in compressive strength; however, when used in small quantities, it can become a viable alternative [11,12,53].

3.3. Scanning Electron Microscopy

Figure 13 shows alkali–silica gel in most of the samples analyzed. The gel in the solid phase is the most abundant and has a chemical composition with the presence of the elements silicon and potassium. In contrast, the tabular and acicular phases are usually composed of high levels of sodium and sometimes potassium [21]. The SEM micrographs of the studied cement samples provide a vivid illustration of the diverse microstructural responses to alkali–silica reaction (ASR) across different cement formulations. Panels (a) through (f) display a range of degradation patterns associated with ASR, highlighting the significant influence of cement type and additive content on the progression of this deleterious reaction.

Notably, samples with CP V-ARI and CP II-Z exhibit pronounced microcracking and gel formation, indicative of severe ASR activity. These manifestations are also evident in CP V-ARI RS (10D), where the inclusion of 10% dregs appears to exacerbate the ASR, as shown by the extensive network of cracks and the altered texture of the aggregate surfaces. This suggests that while dregs may provide certain economic and environmental benefits as a partial replacement in cement, their reactivity must be carefully evaluated to prevent adverse effects on the long-term durability of concrete.

The micrographs of the sample composed of CP II-F without adding the residue (Figure 14) show the presence of compounds with the appearance of rosettes, while samples with expansions close to the limit of NBR 15577-1 [34] present microstructures with little or no evidence of gel formation, such as CP IV cement.

In the microstructural analysis of CP II-F (Figure 14) without dregs, SEM micrographs clearly illustrate the profound impact of the alkali–silica reaction (ASR) on the cement matrix. As observed, the significant gel formation and expansive void development are indicative of advanced ASR stages. The micrographs at 1000× and 2000× magnifications reveal extensive gel clustering and a rough texture around voids, suggesting ongoing chemical reactions between the alkaline components of the cement and reactive silica present in the aggregates.

4. Conclusions

The main objective of this article was to analyze the effects of adding dregs to the alkali–silica reaction, as well as assess the mitigating capacity of the different types of cement used. Therefore, through the results, it is possible to conclude the following:

• Depending on the cement used, in reference mortars (without the addition of dregs), there are expansions larger than the limit established by Brazilian standards. These expansions are related to the concomitant action of cement and aggregate, indicating the use of a reactive aggregate in the mixtures.
• In most cases, the addition of dregs promoted greater expansions, which was directly proportional to the increase in residue content.
• Cements with high addition levels demonstrate an excellent ability to inhibit the expansion promoted by ASR; this is the case of CP IV and CP III cements.
• The use of a cement composition with 10% silica fume did not demonstrate positive effects in mitigating ASR. This result is possibly related to the agglomeration of silica fume particles during mixing, especially considering the absence of additives or changes in the water/cement ratio.
• The extension of the experiment to 63 days highlighted a significant increase in expansion, demonstrating the continuity of the ASR.
• Contrary to what was obtained at 30 days, CP II-Z demonstrated lower expansion rates than CP V-ARI cement at 63 days. As a result, the pozzolanic addition continues to react with the cement constituents over time, contributing to dimensional stability, which does not occur for CP V-ARI due to the absence of additions.
• The addition of 5% dregs promoted improvements in the compressive strength of some mortars, such as CP V-ARI RS, CP II-F, and CP IV.
• With 15% dregs, there is a significant performance loss in compressive strength, especially in the CP V composition with 10% silica fume.
• Using Scanning Electron Microscopy (SEM)), it is possible to corroborate the presence of alkali–silica gel in a large part of the samples. This gel is predominantly manifested in a massively cracked form but is still identifiable in a sample under the rosette form.

5. Recommendations for Future Research

• Future research should focus on long-term studies that evaluate the performance of cementitious mixtures containing dregs under real environmental conditions. This includes exposure to different climates and moisture cycles to assess the durability and behavior of these materials over time, particularly regarding their resistance to alkali–silica reaction (ASR) and mechanical degradation.
• Further studies are recommended to optimize the proportion of dregs in cementitious mixtures, aiming to find the ideal balance between sustainability and mechanical performance. This research should explore the threshold at which dregs provide environmental benefits without significantly compromising the compressive strength and durability of the concrete, while also investigating the potential of other industrial byproducts in ASR mitigation.
• Advanced microstructural analysis techniques, such as X-ray tomography, should be employed in future studies to provide a deeper understanding of the interactions between dregs and the cement matrix at various proportions. This would allow for a more precise characterization of the mechanisms by which dregs influence ASR mitigation and the overall performance of the concrete.