Preliminary study of the effect of carbonation curing on geopolymers

Estudo preliminar do efeito da cura por carbonatação em geopolímeros

Abstracts

Abstract

Carbonation curing differs from weathering carbonation since it is performed intentionally at the early ages of cement hydration. This cure involves applying different levels of CO₂ (5% to 99%) to concrete for a short period of time, usually followed by conventional hydration. The objective of this article was to evaluate the carbonation curing in metakaolin-based geopolymer concretes, activated with NaOH and Na₂SiO₃, and compare them with Portland cement (PC) concrete. The following tests were applied: determination of pH, carbonation depth, water absorption by immersion, void index, and compressive strength. The results showed that after the carbonation curing, the geopolymer concrete had compressive strength and carbonation depth equivalent to the concrete with PC, but with a lower absorbed CO₂ content. Although this type of cement absorbs less CO₂ but is more sensitive to carbonation. The effect on the void ratio was not remarkable. Furthermore, the alkalinity of concretes can be partially recovered after subsequent curing by water immersion.

Keywords:
carbonation curing; geopolymer; metakaolin; carbon dioxide

Resumo

A cura por carbonatação difere da carbonatação por intemperismo, pois é realizada intencionalmente nas primeiras idades de hidratação do cimento. Essa cura envolve a aplicação de diferentes níveis de CO₂ (5% a 99%) ao concreto por um curto período de tempo, geralmente seguido de hidratação convencional. O objetivo deste artigo foi avaliar a cura da carbonatação em concretos geopolímeros à base de metacaulim, ativados com NaOH e Na₂SiO₃, e compará-los com o concreto de cimento Portland (CP). Foram aplicados os seguintes testes: determinação do pH, profundidade de carbonatação, absorção de água por imersão, índice de vazios e resistência à compressão. Os resultados mostraram que após a cura da carbonatação, o concreto geopolimérico apresentou resistência à compressão e profundidade de carbonatação equivalentes ao concreto com CP, porém com menor teor de CO₂ absorvido. Portanto, este tipo de cimento absorve menos CO₂, mas é mais sensível à carbonatação. O efeito na taxa de vazios não foi notável. Além disso, a alcalinidade dos concretos pode ser parcialmente recuperada após a cura subsequente por imersão em água.

Palavras-chave:
cura de carbonatação; geopolímero; metacaulim; dióxido de carbono

1 INTRODUCTION

Curing by carbonation is a process carried out in the first hours of hydration of cementitious materials and can be applied with different concentrations of CO₂, ranging from 5% to 99.9% [¹]-[⁸]. This process is different from weathering carbonation, which occurs in the hardened cement paste and is associated with the action of CO₂ in the atmosphere, impairing the passivation of the reinforcements. In the case of carbonation curing, CO₂ is intentionally introduced to react with the hydrated and non-hydrated cement phases at early ages. Thus, carbonation curing was developed to accelerate strength increase, promote CO₂ sequestration and convert more stable phases in concrete [⁴], [⁹], [¹⁰].

Carbonation curing is usually performed within the first 24 hours after mixing and casting the cementitious materials, for a short period (on the scale of hours or a few days), followed by conventional curing to complete the chemical reactions. Conventional curing (exposure to humid environments) is usually executed after carbonation curing to allow subsequent hydration of the cement grains [²], [³], [⁷].

Therefore, at least three distinct steps are defined for carbonation curing. The first is the preconditioning of the material for controlled removal of surface-free water. The second step is the actual exposure to CO₂ to form a layer of carbonated phases on the most superficial surface of the materials. The third step is post-conditioning to allow additional hydration of unreacted phases. Despite the reduced pH in the carbonate zone right after the carbonation curing process [⁸], the core alkalinity is kept high to allow passivation and prevent steel corrosion when the technique is applied to reinforced concrete structures. In this situation, the carbonation cure should be limited to the first millimeters of the material, not compromising its protection against corrosion [¹⁰ ]-[¹³]. In general, the main applications of carbonation curing are related to the production of non-reinforced cement-based materials, as precast elements and concrete blocks [¹⁰], [¹⁴], [¹⁵].

Carbon dioxide from carbonation curing can react with different anhydrous and hydrated phases of Portland cement [¹²], [¹³]. In the short term, the reaction of CO₂ with tricalcium silicate (C₃S) and dicalcium silicate (C₂S) in the presence of water produces calcium silicate hydrate (C-S-H) and calcium carbonate (CaCO₃) according to Equations 1 and 2 [⁷]. Carbon dioxide can also react with the hydrated phases of cement, such as Ca(OH)₂, C-S-H, and ettringite [¹], [³], [⁷], [¹⁰] according to Equations 3, 4 and 5. A denser layer with a remarkable amount of CaCO₃ is formed after carbonation curing, with carbonate being less soluble than Ca(OH)₂ [¹]. This technique contributed to the reduction of porosity and permeability of the materials, in addition to increasing the penetration resistance of chloride ions and other durability benefits [¹]-[⁵ ], [¹⁰], [¹⁶], [¹⁷]. Equações estão ok. Apenas a legenda de indicação do “C-S-H” da Equação (1) ficou deslocada, pois deve estar alinhada com o composto 3CaO·2SiO2·3H2O. Igualmente para a equação (4), onde o “Carbonate” e “Gypsum” deve ficar um mais par

$$\begin{array}{l}2\left(3CaO·Si{O}_2\right)+3C{O}_2+3H_{2}O\to 3CaO·2Si{O}_2·3H_2{O}_{}+3CaC{O}_3\\ C_{3}SC-S-HCarbonate\end{array}$$ (1)

$$\begin{array}{l}2\left(2CaO·Si{O}_2\right)+C{O}_2+3H_{2}O\to 3CaO·2Si{O}_2·3H_{2}O+CaC{O}_3\\ C_{2}SC-S-HCarbonate\end{array}$$ (2)

$$\begin{array}{l}Ca{\left(OH\right)}_2+C{O}_2\to CaC{O}_3+H_{2}O\\ PortlanditeCarbonate\end{array}$$ (3)

$$\begin{array}{l}3CaO·A{l}_2{O}_3·3CaS{O}_4·32H_{2}O+3C{O}_2\to 3CaC{O}_3+3CaS{O}_4·2H_{2}O\\ EttringiteCarbonateGypsum\end{array}$$ (4)

$$\begin{array}{l}3CaO·2Si{O}_2·3H_{2}O+3C{O}_2\to 3CaC{O}_3+2Si{O}_2+3H_{2}O\\ C-S-HCarbonate\end{array}$$ (5)

On the other hand, information on carbonation curing is needed for low calcium content materials. The study of the application of this technique in geopolymers is an evident lack in the literature. Few references [¹⁸]-[²¹] on the carbonation of geopolymers are found in the literature.

Geopolymer is an inorganic material obtained from the reaction between aluminosilicates and alkaline activators. The main aluminosilicate precursors used are kaolin, metakaolin, rice husk ash, and volcanic rock dust, depending on the study region. These precursors can also be obtained from industrial waste sources, such as fly ash and slag [²²]. The most commonly used alkaline activators are potassium hydroxides, sodium hydroxides, phosphoric acids, potassium silicates, sodium silicates, or a mixture of these components [²³ ]-[²⁵].

The reactive amorphous silico-aluminous materials dissolve in the tetrahedral form of AlO₄⁴ˉ and SiO₄⁴ˉ rapidly by hydrolysis to form the Si-O-Si and Si-O-Al oligomers when they come into contact with the alkaline solution (pH > 13.0) (Equations 6 and 7). By polycondensation, this material forms a resistant three-dimensional structure, giving rise to a polymeric chain [²³], [²⁵]. The main product of this interaction is the amorphous N-A-S-H gel (hydrated sodium aluminosilicate) (Equation 8) [²⁶], but the presence of zeolites is also indicated in the literature [²⁷].

$$A{l}_2{O}_3+3H_{2}O+2O{H}^{-}\to 2{\left[Al{\left(OH\right)}_4\right]}^{-}$$ (6)

$$Si{O}_2+2O{H}^{-}\to {\left[Si{O}_2{\left(OH\right)}_2\right]}^{2-}$$ (7)

$$\begin{array}{l}A{l}_2{O}_3·2Si{O}_2+2O{H}^{-}+2N{a}^{+}+H_{2}O\to 2NaAlSi{O}_4·H_{2}O\\ N-A-S-H\end{array}$$ (8)

The combination of different aluminosilicate materials, chemical activators, curing procedures, and additives produces different geopolymer cements, but generally with good mechanical properties and adequate durability, in addition to environmental benefits [²²]. The most common applications of geopolymers are in airport floors, marine structures, repair mortars, sidewalks, roof tiles, cycle paths, precast slabs, concrete pipes, railway sleepers, and tunnel segments [²⁵], [²⁸ ]-[³⁰].

Given the particularities of geopolymer materials, this article aimed to evaluate the carbonation curing of metakaolin-based geopolymer concretes, activated with NaOH and Na₂SiO₃. A comparison was made with a reference concrete (ordinary Portland cement), with equivalent compressive strengths. This is a preliminary study, to understand the initial behavior of geopolymer concretes after carbonation curing, focusing on the determination of pH, carbonation depth, water absorption by immersion, void index, and compressive strength. Subsequently, other studies are currently being performed by these authors to evaluate the durability of this material in aggressive environments.

2 MATERIALS AND EXPERIMENTAL PROGRAM

2.1 Materials

Commercial metakaolin composed of quartz (40.03%), illite (27.54%), kaolinite (22.78%), and microclinium (9.66%) was used in this study as a source of aluminosilicate. The alkaline activators were produced using a sodium hydroxide (NaOH ≈ 99%) and sodium silicate (Na₂SiO₃) solution. The proportion of materials for the composition of the geopolymer was 45.28% metakaolin, 4.82% NaOH, and 50.0% Na₂SiO₃. Therefore, a molar ratio (Na₂SiO₃ / NaOH) of 1.60 was used, with a content of 23% Na₂O in relation to metakaolin and alkaline solids-to-metakaolin ratio of 0.62.

Ordinary Portland cement (CEM I) was used in conventional concretes for comparison with geopolymer concretes. Table 1 shows the characteristics of the aggregates (fine and coarse) used in both concretes.

Table 1
Characterization of fine and coarse aggregates.

Property	Fine	Coarse
Specific gravity (g/cm3)	2.64	2.90
Bulk density (g/cm3)	1.64	1.45
#200 sieve fineness (%)	0.63	1.00
Fineness module	2.01	2.64
Maximum size (mm)	4.75	9.50

Cubic specimens (75 mm edge) of concrete were casted for the tests. Some initial parameters were defined for casting the concrete. The compressive strength at 28 days should be around 25 MPa (usual strength in conventional concrete) and the expected carbonation depth after carbonation curing should be 8-15 mm (excessive values can compromise the passivation of reinforcements regarding corrosion). In this way, some concrete mixtures were initially tested, until finding the mixture that met these parameters.

The proportion defined for Portland cement concrete (PC concrete) was 1: 2.03: 2.47: 0.55 (cement: sand: gravel: water). Similar proportions of fine and coarse aggregates were used for the production of geopolymer concrete (GEO concrete). However, the water-to-metakaolin ratio was 0.74.

2.2 Methods

After casting the concrete (GEO and PC), the specimens were cured by two different methods: carbonation curing (C) and conventional curing (N). The carbonation curing consisted of three controlled steps, according to Table 2. In Step 1, the pre-conditioning aimed to partially remove free water from the pores to facilitate the entry of CO₂. In this phase, the thermal curing of the geopolymers (6 h) was necessary to allow the penetration of CO₂, since samples pre-conditioned at room temperature showed a water loss of less than 1.0%, making it impossible for carbon dioxide to enter in GEO concrete, according to Figure 1.

Table 2
Carbonation curing steps.

Concrete	Step 1: preconditioning	Step 2: carbonation	Step 3: additional curing
GEO	4 h in mold (T = 50 ± 5) °C; 2 h out of the mold (T = 50 ± 5) °C; 18 h (T = 25 ± 5) ºC, (RH = 45 ± 5) %	6 h, 15 Psi ≈ 99% CO2 (T = 25 ± 5) °C, ambient RH	27 days (T = 25 ± 5) °C, ambient RH
PC	6 h in mold (T = 25 ± 5) °C; 18 h (T = 25 ± 5) ºC, (RH = 50 ± 5) %	6 h, 15 Psi ≈ 99% CO2 (T = 25 ± 5) °C, ambient RH	27 days (T = 25 ± 5) °C, (RH = 95 ± 5) %

Figure 1
External carbonation (right) and zero carbonation depth in geopolymer concrete (left) without thermal curing.

In Step 2, the cubes were placed in a hermetically sealed steel chamber (500 x 500 mm in size) with a CO₂ cylinder with a pressure regulator. This phase allows the controlled entry of CO₂ into the concrete. Before injecting carbon dioxide, the oxygen inside the chamber was aspirated with a vacuum pump for 5 minutes. Finally, in Step 3 (additional curing), the specimens were stored in a curing chamber to complete the chemical reactions.

Conventional curing (N) of PC concrete was performed for 28 days, similar to Step 3. Conventional curing (N) of GEO concrete was initially composed of thermal curing (50 ± 5) °C for 6 hours, similar to Step 1, followed by curing for 28 days, as per Step 3.

After carbonation (Step 2), some specimens were immediately sectioned into two parts and a phenolphthalein solution was sprayed across the cross-section. The measurement of carbonation depth was performed with a digital caliper, with a precision of 0.001 mm. 24 carbonation depth points were measured per specimen to calculate the average depth. An estimate of the carbonation area was also made using photographs of the specimens treated in software. The percentage of CO₂ uptake in the concretes was expressed in % using as reference the mass of the concrete before carbonation, according to Zhang and Shao [¹⁶]. During the carbonation process, the concrete gains mass with the penetration of CO₂ at the same time that it loses water during the reactions. For this reason, water from the inner walls of the chamber was collected with absorbent paper and the value included in the final concrete mass.

The suspension method called ex-situ leaching [¹⁰] was used to determine the pH of the concretes. Fragments of the specimens were manually ground into granulometry passing through a 0.30 mm sieve. 3 grams of powder were extracted from the samples and mixed with 10 grams of distilled water. A digital pH meter was used in this solution to determine the pH of the concretes.

The water absorption by immersion and the void index of the concretes were determined at 28 days using three specimens per condition. ASTM C140 [³¹] procedures were applied for these tests. Compressive strengths were determined according to ASTM C109 [³²]. This test was performed before (end of Step 1) and after (end of Step 2) the specimens entered the carbonation chamber and after 28 days of additional curing (end of Step 3). Three specimens were selected per condition [⁷]. A universal press with a capacity of 100 tf was used. Finally, the statistical analysis of the experimental data was performed using analysis of variance (ANOVA) and Tukey's test, with a significance level of 5%.

Microstructure analyzes were performed on carbonated and non-carbonated specimens after step 2 (exposure to the carbonation chamber). The concrete samples were manually pulverized and the material that passed the 100 sieve (0.15 mm) was separated for X-ray diffraction (XRD) testing. The test was carried out between 5º and 75º 2θ, with an angular step of 3º per minute in an X-ray diffractometer model Miniflex 600, Rigaku, with X-ray generator tubes (Kα of Cu λ= 0.154 nm) and voltage of 40kV.

3 RESULTS AND DISCUSSIONS

The mass loss of GEO and PC concretes due to the release of free water during the preconditioning phase (step 1) was 13.1% and 32.9%, respectively (Table 3). The smallest loss of water in the geopolymers during the preconditioning phase is mainly associated with the Na₂SiO₃ activator since about 73% of the water used in the mixtures comes from sodium silicate. Thus, the higher density of the alkaline activator (≈ 1.576 g/ml) in relation to water (≈ 1.0 g/ml) may have interfered with the water loss of the GEO concrete at this stage.

Table 3
Results of preconditioning and carbonation in the concretes.

Property	GEO-C		PC-C		Statistical analysis
	Average	%	Average	%	p-value	Significance level
Water loss (g) during step 1	9.00 ± 0.59	13.1	32.77 ± 4.18	32.9	4.64E-14	Significant
Mass increase (g) after 6 hours of carbonation, step 2	9.59 ± 1.01	0.85	22.7 ± 2.23	2.14	1.05E-10	Significant
Carbonation depth (mm)	10.1 ± 2.11	-	9.10 ± 1.54	-	0.074	Non-significant
Carbonated area (mm2)	2556.35 ± 283.94	45.40	2335.17 ± 182.08	41.50	0.319	Non-significant

An increase in mass was observed in the concretes after the period inside the carbonation chamber (6 hours, 15 Psi). PC concrete showed a 2.5 times greater mass increase than GEO concrete. This increase in mass is due to the incorporation of CO₂ into the internal structure of the concrete, forming carbonates (CaCO₃ or Na₂CO₃).

The greater loss of water during Step 1 in PC concrete may have influenced the pore network of the material, facilitating the ingress of CO₂. This gas reacted with the hydrated (Ca(OH)₂ and C-S-H) and non-hydrated (C₃S and C₂S) phases of this cement, resulting in a greater mass increase for PC concrete. Moreover, the higher amount of calcium (Ca²⁺) in Portland cement can also increase the CO₂ uptake [¹⁹], [³³], justifying the greater mass gain observed.

Figure 2 shows the carbonation depth and carbonation area of GEO and PC concretes. The carbonation depth of the geopolymers was around 10.1 mm and the carbonated area was around 45.4%. In Portland cement, a depth close to 9.1 mm and a carbonated area of 41.5% were obtained. Non-carbonated (conventional curing) samples were also used for comparison in Figure 2.

Figure 2
Carbonation of GEO and PC concretes: (a) carbonation depth; (b) carbonated area; (c) non-carbonated (reference) samples.

According to Figure 2, the carbonation depth and the carbonated area in the concretes were similar. Although the colorimetric technique with phenolphthalein is merely qualitative, the results show that concrete with metakaolin-based geopolymers may have a greater capacity to absorb CO₂, measured by mass increase (Table 3), during carbonation curing compared to the Portland cement concrete. Thus, the lower absorbed CO₂ content (by mass) produced an equivalent carbonation depth in the geopolymer compared to the PC concrete. However, other techniques to really measure the amount of products formed by carbonation curing must be performed to quantify this information.

The binding capacity of geopolymer with CO₂ is associated with the type and proportion of alkaline activators used in the geopolymer formulation since carbonation depends on the availability of alkali for the formation of sodium carbonates. The aluminosilicate gel can substantially capture sodium (Na⁺) in the structure and reduce the carbonate content during carbonation, as discussed in Nedeljković et al. [¹⁹ ].

According to Figure 2, the pH of the core (innermost region and therefore not carbonated) of the specimens right after the carbonation step (step 2) was similar to the reference concrete (≈ 12.7). This demonstrates that the core is maintained at a high pH, therefore, there is no risk of steel corrosion if the technique is used in reinforced concrete structures. However, the surface pH showed a slight reduction, being more evident in the GEO concrete (from 12.74 to 10.06).

Figure 3 (XRD results) shows the presence of Na₂CO₃ phases in the GEO-C sample and CaCO₃ in the PC-C sample, both extracted from the concrete surface (0-5 mm). The identification of these peaks is supported by other studies [³⁴ ]-[³⁷]. Therefore, the formation of carbonates (CaCO₃ in Portland cement) and (Na₂CO₃ in the geopolymer) can reduce the pH of the concrete.

Figure 3
Result of X-ray diffraction of the concrete surface after Step 2.

Table 4 and Figure 4 show the results of water absorption by immersion and void index of the concretes after 28 days of carbonation curing (GEO-C and PC-C) and conventional curing (GEO-N and PC-N). These results support the discussions made so far.

Table 4
Results of water absorption by immersion and void ratio of concretes.

Properties	Average (%)
	GEO-N		GEO-C		PC-N		PC-C
Water absorption by immersion	11.09 ± 0.47		10.93 ± 0.43		7.33 ± 0.26		6.33 ± 0.05
Void index	22.86 ± 0.84		22.32 ± 0.67		16.75 ± 0.48		14.59 ± 0.20
Properties		Statistical analysis		p-value		Significance level
		Comparative
Water absorption by immersion		GEO-N - GEO-C		0.7001		Non-significant
		PC-N - PC-C		0.0030		Significant
		GEO-C - PC-C		5,20E-02		Significant
		GEO-N - PC-N		0.0002		Significant
Void index		GEO-N - GEO-C		0.4368		Non-significant
		PC-N - PC-C		0.0004		Significant
		GEO-C - PC-C		4,38E-02		Significant
		GEO-N - PC-N		0.0019		Significant

Figure 4
Water absorption by immersion and void index of the concretes.

The immersion water absorption and void ratio of GEO concrete was significantly higher compared to PC concrete, regardless of the curing process used. Similar behavior was found in the literature. Albidah et al. [³⁸] obtained three times higher water absorption in metakaolin-based geopolymer compared to Portland cement concrete for the same water-to-binder ratio. According to these authors, the porosity of geopolymer concrete can vary greatly depending on the molar ratio between alkaline activators (NaOH and Na₂SiO₃), the ratio between solid activators and precursor (metakaolin), the water-to-binder ratio, and the proportion of aggregates during the formulation of the concrete.

According to Albidah et al. [³⁸], there is an ideal alkali solids-to-metakaolin content to produce a geopolymer concrete with lower water absorption. These authors found water absorption of 6.6% and 7.0% for an alkaline solids-to-metakaolin ratio of 0.21 and 0.41, respectively. Thus, the authors showed that proportions between 0.25 and 0.37 are indicated for absorption values between 5.5% and 4.7%. The present article used an alkaline solids-to-metakaolin ratio of 0.62, which may have favored the greater water absorption of GEO concrete. Also, sodium silicate can re-dissolve the newly formed reaction phases of geopolymers, breaking chemical bonds and weakening the structure of the material, resulting in increased porosity and water absorption, as discussed by Zhang et al. [³⁹].

Also, according to Puertas et al. [³³], the higher content of Na₂SiO₃ used in the activation of activated alkali cement (from slag) resulted in considerable shrinkage during curing and crack formation. Thus, the water absorption observed in the geopolymer concrete of this study may have been influenced by the content of sodium silicate used for alkaline activation. According to Figure 4, the effect of carbonation curing in reducing the permeability of concrete (water absorption and void index) was greater in Portland cement concrete, although a slight reduction in this property was noted for GEO concrete as well, but not statistically significant. This indicates that the formation of sodium carbonate (Na₂CO₃) due to carbonation curing in geopolymer concrete showed difficulty in promoting pore refinement, unlike what occurred in Portland cement-based material, which predominantly forms calcium carbonate (CaCO₃). According to Bernal et al. [⁴⁰], the sodium carbonate resulting from the carbonation of geopolymers binds to a large amount of water, making it more voluminous and with a greater capacity to fill and block pores. Therefore, the low CO₂ absorption may indicate that a smaller amount of Na₂CO₃ was produced during carbonation and, for this reason, the void index of the geopolymer concrete remained close to the reference concrete (non-carbonated).

The significant reduction in water absorption and void index of carbonated PC concrete compared to reference concrete (PC-N) is associated with the filling of voids due to the formation of CaCO₃, partially blocking the pores and reducing the possibility of the entry of substances into the material. This result agrees with other research [²]-[⁵], [¹⁰].

Despite the higher water absorption and void index (Figure 4), GEO concrete had higher compressive strength than PC concrete. Table 5 and Figure 5 shows the compressive strength of the concretes before and after entering the carbonation chamber (Step 2), and after the additional 27 days of curing, totaling 28 days of age (Step 3). The (N) specimens are the reference samples, which have not been carbonated.

Table 5
Results of compressive strength of concrete.

Age	Compressive strength (MPa)
	GEO-N	GEO-C	PC-N	PC-C
24 h	-	25.84 ± 0.66	-	13.25 ± 0.63
30 h	17.74 ± 1.37	26.93 ± 2.32	16.98 ± 0.73	15.50 ± 1.76
28 days	39.08 ± 3.84	26.89 ± 3.05	34.85 ± 2.75	26.44 ± 1.00
Properties	Specimens	Statistical analysis		Significance level
		Comparative	p-value
Compressive strength	(N)	GEO-N-30 h - GEO-N-28 d	0.0008	Significant
		PC-N-30 h - PC-N-28 d	0.0004	Significant
		GEO-N-30 h - PC-N-30 h	0.4418	Non-significant
		GEO-N-28 d - PC-N-28 d	0.1960	Non-significant
	(C)	GEO-C-24 h - GEO-C-30 h	0.4787	Non-significant
		GEO-C-30 h - GEO-C-28 d	0.9876	Non-significant
		PC-C-24 h - PC-C-30 h	0.3647	Non-significant
		PC-C-30 h - PC-C-28 d	0.0047	Significant
		GEO-C-24 h - PC-C-24 h	7.35E-5	Significant
		GEO-C-30 h - PC-C-30 h	0.0068	Significant
		GEO-C-28 d - PC-C-28 d	0.8204	Non-significant
	(N) vs (C)	GEO-N-30 h - GEO-C-30 h	0.0040	Significant
		PC-N-30 h - PC-C-30 h	0.2948	Non-significant
		GEO-N-28 d - GEO-C-28 d	0.0126	Significant
		PC-N-28 d - PC-C-28 d	0.0075	Significant

Figure 5
Compressive strength of concrete at different ages.

According to ANOVA and Tukey's test, there was no significant difference in the compressive strength of GEO concrete before and after carbonation curing and also in relation to the additional 27 days of curing (28 days of age). On the other hand, the variation in compressive strength of PC concrete was considerable after subsequent curing. Despite the differences observed between the GEO-C and PC-C concretes before and after stage 2 of curing by carbonation, at the end of the process (28 days) both mixtures showed equivalent resistance. In addition, there were also important differences in the compressive strength between the carbonated (C) and non-carbonated (N) specimens, except for the PC sample (30 h) as shown in Table 5. The strength of the reference samples (GEO-N and PC-N) was similar and the resistance increased significantly up to 28 days in both studied concretes.

The higher strength of the geopolymer in the initial preconditioning step (24 h) was provided by the thermal curing in the samples that were cured by carbonation - (C) specimens - since the higher temperature accelerates the polymerization reactions and, consequently, increases the initial resistance [²³], [²⁷]. After Step 2 (exposure to CO₂ for 6 hours), a slight increase in compressive strength (about 4.2%) was observed in the geopolymer. The binding capacity of CO₂ can be associated with this behavior, due to the lower absorption of carbon dioxide during carbonation curing (Table 3 and Figure 2).

The increase in compressive strength in PC concrete after Step 2 was 17%. This increase reinforces previous results. The formation of CaCO₃ during carbonation curing clogged the pores and increased the density of the concrete, increasing compressive strength. These results agree with the reduction in water absorption (Figure 4) and the higher CO₂ uptake of PC concrete compared to GEO concrete. Thus, once again the results indicate that the effect of carbonation curing is greater in concretes with Portland cement.

The increase in compressive strength (70.5%) after additional curing occurred only in PC concrete. This increase is associated with the formation of the C-S-H gel mixed with CaCO₃ from carbonation during the 27 days of additional curing [¹], [¹²]. In general, the mechanical strength of concrete is concentrated in the solid part of the material. Compressive strength decreases linearly with increasing porosity [⁴¹]. This means that porosity and resistance are inversely proportional, that is, a reduction in porosity can result in an increase in resistance and vice versa. Therefore, filling the capillary voids with carbonates after the carbonation curing process refines the pores of the concrete and increases the solid part, resulting in increased mechanical strength.

Figure 6 shows the GEO and PC concretes after 27 days of additional curing (step 3). Different shades of pink color were identified on the surfaces of the samples. This demonstrates that the alkalinity of the carbonated region was partially restored after further curing. Partial pH recovery of concretes after the additional curing step during carbonation curing was indicated in other studies with Portland cement [¹⁰], but no similar study was found for geopolymers. Therefore, these results are innovative and visually indicate that the alkalinity of the carbonated region can be partially recovered in the geopolymers in a similar way to Portland cement concrete.

Figure 6
Partial pH recovery after step 3.

According to Figure 6, a slight increase in surface pH value was observed in GEO (≈ 10.71) and PC (≈ 12.17) concretes. Despite the lower pH of the geopolymer concerning the reference concrete, Robayo-Salazar et al. [²¹] discussed that the steel remains passivated with a pH close to 10.7, offering no risk of corrosion of the reinforcement. Values close to 10.5 were also indicated by Pouhet and Cyr [¹⁸]. The increase in pH reinforces that the alkalinity of the carbonated samples was partially recovered, demonstrated by the light pink color obtained by the phenolphthalein colorimetric test. Moreover, pH values around 11.3 to 12.5 have been reported by other studies [¹⁰], [¹⁶] involving Portland cement after additional curing.

4 CONCLUSIONS

The following conclusions were obtained from the execution of this study:

• The metakaolin-based geopolymer concrete showed depth and extent of carbonation similar to Portland cement concrete. However, the GEO concrete requires thermal curing preconditioning to release water from the mixture and unblock the pores for CO₂ penetration. Also, PC concrete showed a 2.5 times greater mass increase than GEO concrete during carbonation curing. Therefore, concretes with metakaolin-based geopolymers may have a greater capacity to bind CO₂ during carbonation curing compared to Portland cement concrete, since less CO₂ was absorbed by the geopolymers and similar carbonation depth was observed.

• During carbonation curing, the surface pH was reduced. Moreover, the alkalinity of the carbonated region was partially restored after further conventional curing. Although other studies have shown that the pH of Portland cement concrete can be partially recovered after the additional curing step during carbonation curing, this article showed that this effect also occurs in geopolymer concrete.

• Carbonation curing in geopolymer concrete showed difficulty in promoting pore refinement. Also, there was no significant difference in the compressive strength of GEO concrete before and after carbonation curing. Even so, the compressive strength of geopolymer concrete was higher than the compressive strength of Portland cement concrete.

• The variation in the compressive strength of PC concrete was significant after the additional curing. The increase in compressive strength in PC concrete after Step 2 was 17%. This increase reinforces that the formation of CaCO₃ during carbonation curing clogged the pores and increased the density of the concrete, increasing compressive strength. These results agree with the reduction in water absorption by immersion and the higher CO₂ absorption of PC concrete compared to GEO concrete. Thus, the results indicate that the effect of carbonation curing is greater in concretes with Portland cement.

ACKNOWLEDGEMENTS

The authors thank the Brazilian Federal Agency for the Support and Evaluation of Graduate Education (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES), the National Council for Scientific and Technological Development (Conselho Nacional de Desenvolvimento Científico and Tecnológico - CNPQ), and the Araucária Foundation (grant number 044/2020) for their support in conducting this study.

• Financial support: None.

• Data Availability:

  The data that support the findings of this study are available from the corresponding author, [N.A.M.B.], upon reasonable request.
• How to cite: N. A. M. Beltrame, R. A. Medeiros-Junior, R. L. Dias, and F.B. Witzke, “Preliminary study of the effect of carbonation curing on geopolymers,” Rev. IBRACON Estrut. Mater., vol. 17, no. 1, e17110, 2024, https://doi.org/10.1590/S1983-41952024000100010

REFERENCES

• ¹
  L. Qin, X. Gao, A. Su, and Q. Li, “Effect of carbonation curing on sulfate resistance of cement-coal gangue paste,” J. Clean. Prod., vol. 278, pp. 123897, Jan 2021, http://dx.doi.org/10.1016/j.jclepro.2020.123897
  » http://dx.doi.org/10.1016/j.jclepro.2020.123897
• ²
  M. Liu, S. Hong, Y. Wanga, J. Zhang, D. Hou, and B. Dong, “Compositions and microstructures of hardened cement paste with carbonation curing and further water curing,” Constr. Build. Mater., vol. 267, pp. 121724, Jan 2021, http://dx.doi.org/10.1016/j.conbuildmat.2020.121724
  » http://dx.doi.org/10.1016/j.conbuildmat.2020.121724
• ³
  Y. Liu et al., “Properties and microstructure of concrete blocks incorporating drinking water treatment sludge exposed to early-age carbonation curing,” J. Clean. Prod., vol. 261, pp. 121257, Jul 2020, http://dx.doi.org/10.1016/j.jclepro.2020.121257
  » http://dx.doi.org/10.1016/j.jclepro.2020.121257
• ⁴
  D. Zhang and Y. Shao, “Enhancing chloride corrosion resistance of precast reinforced concrete by carbonation curing,” ACI Mater. J., vol. 116, pp. 3-12, Jan 2019, http://dx.doi.org/10.14359/51714461
  » http://dx.doi.org/10.14359/51714461
• ⁵
  X. Pan, C. Shi, N. Farzadnia, X. Hu, and J. Zheng, “Properties and microstructure of CO₂ surface treated cement mortars with subsequent lime-saturated water curing,” Cement Concr. Compos., vol. 99, pp. 89-99, May 2019, http://dx.doi.org/10.1016/j.cemconcomp.2019.03.006
  » http://dx.doi.org/10.1016/j.cemconcomp.2019.03.006
• ⁶
  J. Jang and H. Lee, “Microstructural densification and CO₂ uptake promoted by the carbonation curing of belite-rich Portland cement,” Cement Concr. Res., vol. 82, pp. 50-57, Apr 2016, http://dx.doi.org/10.1016/j.cemconres.2016.01.001
  » http://dx.doi.org/10.1016/j.cemconres.2016.01.001
• ⁷
  X. Xian, M. Mahoutia, and Y. Shao, “Production of concrete pipes by carbonation curing in an inflatable enclosure,” Constr. Build. Mater., vol. 363, pp. 129861, Jan 2023, http://dx.doi.org/10.1016/j.conbuildmat.2022.129861
  » http://dx.doi.org/10.1016/j.conbuildmat.2022.129861
• ⁸
  Y. Feng, J. Yang, and P. Zhang, “Effects of carbonation curing regimes on alkalinity of self-compacting concretes for marine artificial reef,” Constr. Build. Mater., vol. 369, pp. 130614, Mar 2023, http://dx.doi.org/10.1016/j.conbuildmat.2023.130614
  » http://dx.doi.org/10.1016/j.conbuildmat.2023.130614
• ⁹
  I. C. A. Esteves, R. A. Medeiros-Junior, and M. H. F. Medeiros, “NDT for bridges durability assessment on urban-industrial environment in Brazil,” Int. J. Build. Pathol. Adapt., vol. 36, pp. 500-515, Oct 2018, http://dx.doi.org/10.1108/IJBPA-04-2018-0032
  » http://dx.doi.org/10.1108/IJBPA-04-2018-0032
• ¹⁰
  D. Sharma and S. Goyal, “Accelerated carbonation curing of cement mortars containing cement kiln dust: An effective way of CO₂ sequestration and carbon footprint reduction,” J. Clean. Prod., vol. 192, pp. 844-854, Aug 2018, http://dx.doi.org/10.1016/j.jclepro.2018.05.027
  » http://dx.doi.org/10.1016/j.jclepro.2018.05.027
• ¹¹
  E. M. Takagi, M. G. Lima, P. Helene, and R. A. Medeiros-Junior, “Self-healing of self-compacting concretes made with blast furnace slag cements activated by crystalline admixture,” Int. J. Mater. Prod. Technol., vol. 56, pp. 169-186, Jan 2018, http://dx.doi.org/10.1504/IJMPT.2018.10009994
  » http://dx.doi.org/10.1504/IJMPT.2018.10009994
• ¹²
  D. Zhang, Z. Ghouleh, Y. Shao, “Review on carbonation curing of cement-based materials,” J. CO2 Util., vol. 21, pp. 119-131, Oct 2017, https://doi.org/10.1016/j.jcou.2017.07.003
  » https://doi.org/10.1016/j.jcou.2017.07.003
• ¹³
  Z. Liu, W. Meng, “Fundamental understanding of carbonation curing and durability of carbonation-cured cement-based composites: A review,” J. CO2 Util., vol. 44, pp. 101428, Feb 2021, https://doi.org/10.1016/j.jcou.2020.101428
  » https://doi.org/10.1016/j.jcou.2020.101428
• ¹⁴
  C. Shi, D. Wang, F. He, and M. Liu, “Weathering properties of CO₂-cured concrete blocks,” Resour. Conserv. Recycling, vol. 65, pp. 11-17, Aug 2012, http://dx.doi.org/10.1016/j.resconrec.2012.04.005
  » http://dx.doi.org/10.1016/j.resconrec.2012.04.005
• ¹⁵
  D. Xuan, B. Zhan, and C. S. Poon, “Development of a new generation of eco-friendly concrete blocks by accelerated mineral carbonation,” J. Clean. Prod., vol. 133, pp. 1235-1241, Oct 2016, http://dx.doi.org/10.1016/j.jclepro.2016.06.062
  » http://dx.doi.org/10.1016/j.jclepro.2016.06.062
• ¹⁶
  D. Zhang and Y. Shao, “Effect of early carbonation curing on chloride penetration and weathering carbonation in concrete,” Constr. Build. Mater., vol. 123, pp. 516-526, Oct 2016, http://dx.doi.org/10.1016/j.conbuildmat.2016.07.041
  » http://dx.doi.org/10.1016/j.conbuildmat.2016.07.041
• ¹⁷
  J. H. Seo, S. M. Park, and H. K. Lee, “Evolution of the binder gel in carbonation-cured Portland cement in an acidic medium,” Cement Concr. Res., vol. 109, pp. 81-89, Jul 2018, http://dx.doi.org/10.1016/j.cemconres.2018.03.014
  » http://dx.doi.org/10.1016/j.cemconres.2018.03.014
• ¹⁸
  R. Pouhet and M. Cyr, “Carbonation in the pore solution of metakaolin-based geopolymer,” Cement Concr. Res., vol. 88, pp. 227-235, Oct 2016, http://dx.doi.org/10.1016/j.cemconres.2016.05.008
  » http://dx.doi.org/10.1016/j.cemconres.2016.05.008
• ¹⁹
  M. Nedeljković, B. Ghiassi, S. Melzer, C. Kooij, S. V. D. Laan, and G. Ye, “CO₂ binding capacity of alkali-activated fly ash and slag pastes,” Ceram. Int., vol. 44, pp. 19646-19660, Nov 2018, http://dx.doi.org/10.1016/j.ceramint.2018.07.216
  » http://dx.doi.org/10.1016/j.ceramint.2018.07.216
• ²⁰
  M. Nedeljković, B. Ghiassi, V. D. S. Laan, Z. Li, and G. Ye, “Effect of curing conditions on the pore solution and carbonation resistance of alkali-activated fly ash and slag pastes,” Cement Concr. Res., vol. 116, pp. 146-158, Feb 2019, http://dx.doi.org/10.1016/j.cemconres.2018.11.011
  » http://dx.doi.org/10.1016/j.cemconres.2018.11.011
• ²¹
  R. A. Robayo-Salazar, A. M. Aguirre-Guerrero, and R. M. Gutiérrez, “Carbonation-induced corrosion of alkali-activated binary concrete based on natural volcanic pozzolan,” Constr. Build. Mater., vol. 232, pp. 117189, Jan 2020, http://dx.doi.org/10.1016/j.conbuildmat.2019.117189
  » http://dx.doi.org/10.1016/j.conbuildmat.2019.117189
• ²²
  N. B. Singh and B. Middendorf, “Geopolymers as an alternative to Portland cement: An overview,” Constr. Build. Mater., vol. 237, pp. 117455, Mar 2020, http://dx.doi.org/10.1016/j.conbuildmat.2019.117455
  » http://dx.doi.org/10.1016/j.conbuildmat.2019.117455
• ²³
  J. Cai, X. Li, J. Tan, and B. Vandevyvere, “Thermal and compressive behaviors of fly ash and metakaolin-based geopolymer,” J. Build. Eng., vol. 30, pp. 101307, Jul 2020, http://dx.doi.org/10.1016/j.jobe.2020.101307
  » http://dx.doi.org/10.1016/j.jobe.2020.101307
• ²⁴
  L. Guo et al., “Sulfate resistance of hybrid fiber reinforced metakaolin geopolymer composites,” Compos. Part B, vol. 183, pp. 107689, Feb 2020, http://dx.doi.org/10.1016/j.compositesb.2019.107689
  » http://dx.doi.org/10.1016/j.compositesb.2019.107689
• ²⁵
  M. Nawaz, A. Heitor, and M. Sivakumar, “Geopolymers in construction - recent developments,” Constr. Build. Mater., vol. 260, pp. 120472, Nov 2020, http://dx.doi.org/10.1016/j.conbuildmat.2020.120472
  » http://dx.doi.org/10.1016/j.conbuildmat.2020.120472
• ²⁶
  Z. Zhang, J. L. Provis, H. Wang, F. Bullen, A. Reid, and Y. Zhu, “Quantitative kinetic and structural analysis of geopolymers. Part 1. The activation of metakaolin with sodium hydroxide,” Thermochim. Acta, vol. 539, pp. 23-33, Jul 2012, http://dx.doi.org/10.1016/j.tca.2012.03.021
  » http://dx.doi.org/10.1016/j.tca.2012.03.021
• ²⁷
  N. Granizo, A. Palomo, and A. Fernández-Jiménez, “Effect of temperature and alcaline concentration on metakaolin leaching kinetics,” Ceram. Int., vol. 40, pp. 8975-8985, Aug 2014, http://dx.doi.org/10.1016/j.ceramint.2014.02.071
  » http://dx.doi.org/10.1016/j.ceramint.2014.02.071
• ²⁸
  A. R. G. Azevedo, C. M. F. Vieira, W. M. Ferreira, K. C. P. Faria, L. G. Pedroti, and B. C. Mendes, “Potential use of ceramic waste as precursor in the geopolymerization reaction for the production of ceramic roof tiles,” J. Build. Eng., vol. 29, pp. 101156, May 2020, http://dx.doi.org/10.1016/j.jobe.2019.101156
  » http://dx.doi.org/10.1016/j.jobe.2019.101156
• ²⁹
  A. Hassan, A. Mohammed, and M. Shariq, “A review of properties and behaviour of reinforced geopolymer concrete structural elements - A clean technology option for sustainable development,” J. Clean. Prod., vol. 245, pp. 118762, Feb 2020, http://dx.doi.org/10.1016/j.jclepro.2019.118762
  » http://dx.doi.org/10.1016/j.jclepro.2019.118762
• ³⁰
  A. H. Mahmood, S. J. Foster, and A. Castel, “Development of high-density geopolymer concrete with steel furnace slag aggregate for coastal protection structures,” Constr. Build. Mater., vol. 248, pp. 118681, Jul 2020, http://dx.doi.org/10.1016/j.conbuildmat.2020.118681
  » http://dx.doi.org/10.1016/j.conbuildmat.2020.118681
• ³¹
  American Society for Testing and Materials, Standard test methods for sampling and testing concrete masonry units and related units, ASTM C140, 2020. https://doi.org/10.1520/C0140_C0140M-22B.
  » https://doi.org/ https://doi.org/10.1520/C0140_C0140M-22B
• ³²
  American Society for Testing and Materials, Standard test method for compressive strength of hydraulic cement mortars (using 2-in. or [50-mm] cube specimens, ASTM C109, 2016. https://doi.org/10.1520/C0109_C0109M-20
  » https://doi.org/10.1520/C0109_C0109M-20
• ³³
  F. Puertas, M. Palácios, and T. Vázquez, “Carbonation process of alkali-activated slag mortars,” J. Mater. Sci., vol. 41, pp. 3071-3082, Oct 2006, http://dx.doi.org/10.1007/s10853-005-1821-2
  » http://dx.doi.org/10.1007/s10853-005-1821-2
• ³⁴
  M. A. Longhi, E. D. Rodríguez, B. Walkley, Z. Zhang, and A. P. Kirchheim, “Metakaolin-based geopolymers: Relation between formulation, physicochemical properties and efflorescence formation,” Compos. B. Eng., vol. 182, pp. 107671, Feb 2020, http://dx.doi.org/10.1016/j.compositesb.2019.107671
  » http://dx.doi.org/10.1016/j.compositesb.2019.107671
• ³⁵
  Y. Wang, S. Tae, R. Lin, and X. Wang, “Effects of Na₂CO₃ on engineering properties of cement-limestone powder-slag ternary blends,” J. Build. Eng., vol. 57, pp. 104937, Oct 2022, http://dx.doi.org/10.1016/j.cemconres.2016.05.013
  » http://dx.doi.org/10.1016/j.cemconres.2016.05.013
• ³⁶
  C. A. Rosas-Casarez et al., “Influence of size reduction of fly ash particles by grinding on the chemical properties of geopolymers,” Appl. Sci. (Basel), vol. 8, pp. 365, Mar 2018, http://dx.doi.org/10.3390/app8030365
  » http://dx.doi.org/10.3390/app8030365
• ³⁷
  S. Y. Kim, Y. Jun, D. Jeon, and J. E. Oh, “Synthesis of structural binder for red brick production based on red mud and fly ash activated using Ca(OH)₂ and Na₂CO₃,” Constr. Build. Mater., vol. 147, pp. 101-116, Aug 2017, http://dx.doi.org/10.1016/j.conbuildmat.2017.04.171
  » http://dx.doi.org/10.1016/j.conbuildmat.2017.04.171
• ³⁸
  A. Albidah, M. Alghannam, H. Abbas, T. Almusallam, and Y. Al-Salloum, “Characteristics of metakaolin-based geopolymer concrete for different mix design parameters,” J. Mater. Res. Technol., vol. 10, pp. 84-98, Jan-Feb 2021, http://dx.doi.org/10.1016/j.jmrt.2020.11.104
  » http://dx.doi.org/10.1016/j.jmrt.2020.11.104
• ³⁹
  Z. Zhang, J. L. Provis, H. Wang, F. Bullen, and A. Reid, “Quantitative kinetic and structural analysis of geopolymers. Part 2. Thermodynamics of sodium silicate activation of metakaolin,” Thermochim. Acta, vol. 565, pp. 163-171, Aug 2013, http://dx.doi.org/10.1016/j.tca.2013.01.040
  » http://dx.doi.org/10.1016/j.tca.2013.01.040
• ⁴⁰
  S. A. Bernal, J. L. Provis, D. G. Brice, A. Kilcullen, P. Duxson, and J. S. J. Van Deventer, “Accelerated carbonation testing of alkali-activated binders significantly underestimates service life: The role of pore solution chemistry,” Cement Concr. Res., vol. 42, pp. 1317-1326, Oct 2012, http://dx.doi.org/10.1016/j.cemconres.2012.07.002
  » http://dx.doi.org/10.1016/j.cemconres.2012.07.002
• ⁴¹
  H. A. Moghadam, A. Mirzaei, and Z. A. Dehghi, “The relation between porosity, hydration degree and compressive strength of Portland cement pastes in the presence of aluminum chloride aditive,” Constr. Build. Mater., vol. 250, pp. 118884, Jul 2020, http://dx.doi.org/10.1016/j.conbuildmat.2020.118884
  » http://dx.doi.org/10.1016/j.conbuildmat.2020.118884

Edited by

Editors: Edna Possan, Guilherme Aris Parsekian.

Data availability

The data that support the findings of this study are available from the corresponding author, [N.A.M.B.], upon reasonable request.

Publication Dates

• Publication in this collection
  16 May 2023
• Date of issue
  2024

History

• Received
  23 Dec 2022
• Reviewed
  29 Mar 2023
• Accepted
  16 Apr 2023
• Corrected
  27 Mar 2024