1. Introduction
1.1. Research background
Efforts to reduce the carbon footprint of construction materials generally focus on two key approaches: developing innovative low-carbon alternatives and optimizing existing production processes to minimize emissions while maintaining material performance. Reducing emissions during the material manufacturing stage, particularly in cement production, is a central strategy [
1], [
2], [
3], [
4], [
5], [
6]. In addition, integration of CO
2 directly into construction materials is gaining traction as a dual-purpose solution for sequestering CO
2 and reducing its atmospheric release. Emerging technologies in carbon capture, utilization, and storage (CCUS) methods [
7], [
8], [
9], along with direct CO
2 injection methods [
10,
11] exhibit immense potential. However, their practical implementation at construction sites presents technical and logistical challenges that require further research and innovation.
Nanobubble technology has attracted increasing interest as an innovative solution to overcome these limitations [
12], [
13], [
14]. Characterized by diameters less than 500 nm, nanobubbles can remain stable in water for as long as six months after their generation [
15], [
16], [
17]. This remarkable sustainability of nanobubbles arises from the physical and electrostatic characteristics of their interface [
18], [
19], [
20], [
21], [
22]. The contact angle observed in nanobubbles is typically lower than that anticipated according to Young’s law, leading to a decrease in the internal pressure [
18,
19]. In addition, the nanoscale diameter of nanobubbles contributes to high specific surface areas and prolonged stagnation periods, which enhance the efficiency of particle capture [
20]. The electrically charged interface between the liquid and gas in the nanobubbles generates a repulsive force that inhibits bubble coalescence [
21]. Moreover, electrostatic interactions contribute to reducing the internal pressure, further increasing the sustainability of the nanobubbles [
22]. Because of these unique characteristics, a large number of nanobubbles produced in water can encapsulate specific elements and remain stable over extended periods, facilitating diverse applications across various industries [
23], [
24], [
25], [
26], [
27]. Ozone-capturing nanobubbles can eliminate water and soil contamination by delivering ozone to pollutants, thereby facilitating reactions [
23,
24]. According to Fan et al. [
25], the utilization of titanium dioxide for capture purposes can purify wastewater more cost-effectively and expediently. Through the capture and transportation of nitrogen or oxygen, nanobubbles can effectively facilitate crop growth [
26,
27]. Following an identical mechanism, a novel approach has been suggested for capturing and sequestering CO
2 using nanobubbles in the manufacturing of various types of cementitious composites [
28], [
29], [
30]. This research offers a new perspective for overcoming the limitations of conventional CO
2 injection methods.
Even as sustainability has gained momentum in the construction materials sector, strength, ductility, and durability will remain the foremost considerations for concrete. Focusing on these intrinsic properties, there has been a notable surge in interest regarding engineering cement composites (ECCs) and strain-hardening cementitious composites (SHCCs) on a worldwide scale. Over the past two decades, these materials have been regarded as key innovations capable of significantly enhancing the longevity, performance, and durability of various structural applications. Synthetic polymer fibers are known for their exceptional strength and corrosion resistance, making them ideal for use in cementitious composites. These fibers play a critical role in the management of microstructural elements by preventing internal defects and controlling cracks. Polyethylene (PE) fibers are the most frequently used synthetic fibers that disperse effectively within cement mixtures and are compatible with alkaline environments [
31], [
32], [
33], [
34]. SHCCs that primarily incorporate PE fibers can achieve tensile strain capacities exceeding 3%, leading to greater toughness, enhanced ductility, and superior energy absorption compared to ECC [
35], [
36], [
37]. Moreover, SHCCs exhibit exceptional performance and durability when the ratio of the reactive materials, including various cementitious materials, is increased. According to Sangeetha and Joanna [
38], replacing a portion of cement with ground granulated blast-furnace slag (GGBS) leads to significant improvements in the bending and tensile performances of concrete. Kim et al. [
39] also found that GGBS particles enhanced the flowability of the mixture, allowing for improved fiber dispersion within cementitious composites. These advancements in materials technology come at a cost because compromises between economic feasibility and environmental sustainability are necessary. This issue arises from the manner in which cement is used to produce SHCCs. Although efforts have been made to reduce cement use for environmental benefits, strict quality standards for SHCCs require an increase in the cement per unit volume. This conflicts with the main goal of lowering carbon emissions and presents a significant challenge for researchers worldwide. This highlights the need for innovative solutions to balance performance, cost, and sustainability.
1.2. Research gap and objectives
Research on the application of nanobubbles in cementitious composites remains at a nascent stage, and no prior studies have explored their use in high-performance cementitious composites. This study focuses on the development of CO
2-consuming strain-hardening cement composites (CC-SHCCs) designed as sustainable and high-performance materials, highlighting their innovative contribution to the field of construction materials. To achieve this goal, it is necessary to establish effective carbon consumption abilities and improve mechanical performance. Previous studies [
28,
40] have shown that nanobubbles can transfer CO
2 into cementitious materials during the cement paste and mortar stage. The solitary utilization of nanobubbles can lead to an enhancement in the compressive strength owing to a decrease in macro-sized pores, which have the potential to contribute to the increased ductility of cementitious composites. Moreover, the CO
2 encapsulated within the nanobubbles leads to decreased porosity and increased compressive strength through the production of more hydration products. Important factors such as curing duration and temperature also have a significant effect on carbon fixation levels [
41], [
42], [
43], [
44], [
45], [
46], [
47], [
48]. The curing temperature was maintained at 40 °C to achieve optimal CO
2 fixation, whereas prolonged curing at temperatures above 28 days led to reduced CO
2 fixation efficiency [
47]. In reference to the corresponding study, the application of nanobubble water (NBW) during the curing period of CC-SHCCs has been actively explored in this study. The fundamental conclusion of these studies is that CO
2-capturing nanobubbles significantly enhance carbon fixation. This study assessed the impact of NBW and enriched CO
2 concentrations on CC-SHCCs by analyzing their chemical composition, porosity, compressive strength, and tensile performance.
2. Materials and test setup
2.1. Characteristics of materials
In accordance with the ASTM C150/C150M standard [
49], Type I ordinary Portland cement (OPC) was employed as the primary dry material in this study. Silica sand was incorporated as the fine aggregate. The distinctive characteristics of supplementary cementitious materials (SCMs) are fundamental for improving the hardening process of cement. Concrete can be produced with significantly enhanced properties, including reduced permeability and increased compactness, by the incorporation of these materials into cement mixtures. GGBS is recognized as a highly reactive SCM, with its reactivity primarily stemming from its rapid cooling process during production. This reactivity plays a vital role in cementitious composites, where the alkali components generated by the GGBS accelerate the chemical reactions that occur during the hardening process [
38],[
39],[
50], [
51], [
52], [
53], [
54]. Silica fume (SF) is an ultrafine material captured from waste gases released during the production of ferrosilicon or silicon metal. The inclusion of SF in cementitious composites brings notable improvements to the compactness and flowability of the matrix, which results in a significant enhancement of the mechanical performance [
55], [
56], [
57], [
58], [
59], [
60], [
61], [
62]. Detailed information on the physical properties and chemical compositions of the dry materials is presented in
Table 1, and images of the dry materials and the particle size distribution curves of these materials are shown in
Fig. 1.Three different types of water were used to produce the SHCC samples and specimens. Conventional tap water (TW) was used, and NBW was produced using a dedicated device that generates nanosized bubbles with circulating TW. A CO
2 gas injection module installed on the device was used to provide CO
2 capturing NBW (NBW + C). The module was programmed to inject CO
2 gas at a controlled flow rate of 0.3 L·min
−1, resulting in the production of NBW + C with an average CO
2 concentration of 2000 parts per million (ppm) during the entire process. The concentration of CO
2 contained in the NBW + C was measured using a sensor inserted inside the nanobubble generator. A schematic of the manufacturing process of NBW + C and the variation in CO
2 concentration of NBW + C over time are shown in
Fig. 2.A high negative zeta potential and strong interfacial tension act as direct factors that prevent nanobubble extinction or coalescence [
63]. As a result, the high stability of nanobubbles utilized in this study is considered to be due to the zeta potential reaching approximately −20 mV. The average diameter of freshly produced nanobubbles is confirmed to be 450 nm. The incorporation of nanobubbles into cementitious mixtures is expected to result in a decrease in their diameter owing to the increase in pH [
64]. Nanobubbles burst through collisions with reactive particles during mixing process [
65]. During this process, the kinetic energy of the nanobubbles is transferred to the particles, which promotes various reactions, such as the dissolution of CaO or tricalcium aluminate (C
3A) or the nucleation of calcium silicate hydrate (C-S-H) gels [
66]. To promote the self-compacting ability of freshly mixed SHCC, a polycarboxylate-based superplasticizer (SP) was employed as a high-range water-reducing agent. The water-to-binder (W/B) ratio was finely adjusted to a low level of 0.18, considering the water content of the SP.
To achieve optimal strain-hardening behavior, ultrahigh molecular weight (UHMW) PE fibers with a tensile strength of 3000 MPa were prepared for the SHCC mixtures. The fibers had a length of 18 mm, diameter of 0.03 mm, and density of 0.97 g·cm
−3. The physical and geometrical properties of the PE fibers are summarized in
Table 2.2.2. Mixing process and specimen preparations
The mixing procedure was organized into three sequential phases. The initial dry-mixing phase involved only dry materials. The wet mixing phase included the addition of mixing water and SP to enhance the flowability of the mixture. Finally, a fiber mixing phase was executed in which a substantial amount of PE fiber was incorporated into the mixture. To prevent the fibers from clumping together during the mixing process, a total fiber volume content of 2.0% was introduced in multiple small portions. Once the mixing process was completed, the fresh mixture was immediately transferred and poured into a mold specifically designed to shape the material. Additionally, the casting of the tensile specimen was performed with careful consideration of the orientation of the fibers within the mixture. When the mixture is cast along the direction in which the tensile test is conducted, the bridging effect of the fibers is maximized [
67]. Therefore, this casting method maximized the tensile energy absorption capacity. To guarantee a sufficient setting time for the SHCC, each specimen was air-cured for 48 h in a standard room. After completing the initial curing phase, the specimens were gently removed from their molds and placed in a water chamber for curing. The curing process was maintained for a total of 28 d at a constant temperature of 40 °C. These specific curing conditions are critical for achieving superior tensile performance and durability [
68,
69]. The detailed mix design for the SHCCs is shown in
Table 3.Two types of specimens were produced in preparation for mechanical testing. The specimens designated for the compressive strength test were shaped into cubes with each edge measuring 50 mm. In addition, the specimen designated for the direct tensile test was shaped into a dog-bone form to analyze the overall stress-strain behavior of the SHCCs. To promote the concentration of stress and crack formation in the central region of the dog bone specimen, this section was intentionally designed to have dimensions of 80 mm (length) × 30 mm (width) × 13 mm (thickness). A schematic of the tensile test specimen is shown in
Fig. 3(a).
2.3. Mercury intrusion porosimetry
Mercury intrusion porosimetry (MIP) was employed to examine the internal pore characteristics of the SHCCs. The principles established by the Washburn equation indicate that pore size is governed by the indentation pressure of infiltrating mercury [
70]. An Autopore IV 9500 (Micromeritics, USA) mercury porosimeter was used. The process began with nitrogen gas, establishing an initial mercury pressure of 0.003 MPa, which progressively increased to a maximum of 227.5 MPa. The MIP samples were prepared by crushing the SHCC specimens that excluded the PE fibers. At least three test samples were used to ensure reliability of the analysis.
2.4. Compressive strength test
A compressive strength test was conducted according to the procedures outlined in the ASTM C109/C109M standard [
71]. The experimental setup was an advanced universal testing machine (UTM) developed by SALT Co. of Republic of Korea, which could handle a maximum load of 2500 kN and is characterized by a high level of accuracy and precision. To minimize errors and achieve precise data acquisition, the test utilized a steady displacement rate of 0.1 mm·min
−1. To guarantee reliable results, a minimum of six specimens were tested. Additionally, careful efforts were directed toward minimizing the potential eccentricity effects by methodically applying a load to the side of the specimen. The test was performed until the load decreased to 60% of the maximum value, and this point was defined as the failure of the specimen.
2.5. Direct tensile test
In accordance with the recommendations provided by Japan Society of Civil Engineers (JSCE) [
72], a direct tensile test was conducted. UTM ST-1001 (SALT Co., Republic of Korea) with a maximum load capacity of 250 kN was used in the test. The test involved sustaining a displacement rate of 0.4 mm·min
−1, which allowed for the consistent application of a uniaxial load. A minimum of seven specimens were tested to strengthen the reliability of the results. To measure the load exerted on the specimens, a load cell was placed near the crosshead section of the UTM. In addition, linear variable displacement transformers (LVDTs) with a maximum capacity of 10 mm were firmly affixed to the aluminum frame to provide precise measurements of the elongation of the specimen. The criterion for determining the failure of the specimen was identical to that used in the compressive strength test. The setup used for the direct tensile tests is shown in
Fig. 3(b).
After the test, the conditions under which the PE fibers were removed from the tensile specimen were reviewed. Only the fibers pulled out during the direct tensile test were selected to ensure that their surfaces remained undamaged through careful collection. A platinum coating was applied in advance to facilitate accurate focusing and rapid establishment of a vacuum. An SU-70 scanning electron microscope (SEM; HITACHI Co., Japan) was employed to capture detailed microscale images, enabling qualitative analysis of the residual particles present on the fiber surfaces.
2.6. Digital image correlation technique
With the application of digital image correlation (DIC), the deformation of a material surface can be monitored optically. By utilizing real-time deformation tracking and strong synchronization abilities, DIC facilitates the examination of the mechanical characteristics of a wide range of materials [
73], [
74], [
75], [
76]. By employing DIC technology, several shortcomings of current techniques for visually examining cracks in fiber-reinforced concrete have been effectively addressed [
77], [
78], [
79]. Current methods for applying polyurethane coatings encounter difficulties in accurately spotting microscale cracks without specialized tools, leading to concerns regarding their dependability. Furthermore, the existing approach evaluates cracks once the specimen has entirely failed in the tensile test, thus hindering the confirmation of the crack distribution prior to localization or the maximum level of tensile strength. The application of the DIC technique allows for the real-time tracking of crack initiation at critical moments during the tensile test, facilitating the gathering of more extensive information with the quantification of cracks. High-quality images were captured at one-second intervals to continuously monitor the emergence and development of cracks. The photographic setup and coating procedure for the surfaces of the tensile specimens are illustrated in
Figs. 3(b) and
4, respectively.
2.7. Thermogravimetric analysis
Thermogravimetric (TG) analysis was used to measure the production of calcium carbonate (CC) within SHCCs. The chemical composition of the samples was analyzed by measuring the weight loss at specific temperatures to ascertain both the type and quantity [
80,
81]. To increase the purity of the data, fresh paste mixtures excluding the aggregate were prepared separately. After curing, the samples were carefully dehydrated in a vacuum desiccator to extract the remaining moisture from the pores. Subsequently, a powder sample containing particles less than 100 μm was produced using crushing and filtering process. An SDT Q600 (TA Instruments Co., USA) was employed for the analysis, and the internal environment of the equipment was maintained under nitrogen gas to prevent undesired sample reactions throughout the testing process. The process commenced at 10 °C, with the temperature gradually increasing at 10 °C·min
−1 until it reached 900 °C.
2.8. Fourier transform infrared spectroscopy
Fourier transform infrared (FTIR) spectroscopy was employed to qualitatively assess the residual levels of C-S-H. FTIR spectroscopy facilitates the accurate identification and characterization of molecular components by exposing the sample to various infrared frequencies and capturing the resulting absorption spectrum. This approach can effectively overcome the limitations of TG, in which the precise quantification of C-S-H formation is obstructed by the presence of ettringite (AFt) and monosulfate (AFm) [
82]. The spectrometer used was VERTEX 70 (Bruker Co., USA), and the sample was dispersed using prepared pellets of potassium bromide. Sample preparation followed the same steps as those in the TG analysis.
2.9. Life cycle assessment
A life cycle assessment (LCA) was performed to evaluate the environmental sustainability of the SHCC based on the ISO 14040 framework [
83], considering a 75-year service life. The assessment included five major environmental impact categories, each of which was used to quantify and standardize the potential burdens associated with the entire life cycle of the material: production, construction, use, and disposal [
84,
85]. Global warming potential (GWP) is defined as the relative contribution of greenhouse gas emissions to global temperature rise, measured in kilogram of CO
2 equivalent (kg CO
2-eq). The ozone depletion potential (ODP) is assessed based on the emission of substances harmful to the ozone layer, is referenced to chlorofluorocarbon, and expressed in kilogram of trichlorofluoromethane equivalent (kg CFC-11-eq). The photochemical ozone creation potential (POCP) is evaluated using the smog-forming potential of the volatile organic compound (VOC) reactions in kilogram of nitrogen oxides equivalent (kg NO
x-eq). The acidification potential (AP) was determined by the release of acidifying pollutants, given in kilogram of sulfur dioxide equivalent (kg SO
2-eq). Eutrophication potential (EP) is measured by the buildup of nutrient compounds like nitrogen and phosphorus in ecosystems, referenced to phosphate and expressed in kilogram of phosphate equivalent (kg PO
43--eq). Each indicator is formulated using standardized units derived from the relative impact of emissions compared to reference substances, allowing for objective and quantitative comparisons of environmental burdens among various construction materials.
3. Results and discussion
3.1. Porosity
The quantity of mercury intrusion relative to pore size is shown in
Fig. 5. Based on the size, the pores were typically separated into two groups [
86,
87]. Intraparticle (IA) pores were identified as microscopic openings, typically ranging between 0.005 and 5 μm in size. Pores having diameters in the range of 5-1000 μm were classified as interparticle (IE) pores.
The SHCC samples exhibited IE porosity levels ranging from 2.11% to 3.61%. The IE porosity reached a maximum of 3.61% when the curing water was set to TW. However, the use of NBW and NBW + C lowered the IE porosity. In particular, the lowest IE porosity of 2.11% was confirmed in the case where NBW + C was used for both mixing and curing. A comparison of the IA porosities reveals a much more dramatic and noticeable difference. When both the mixing water and curing waters were TW, the IA porosity was recorded at approximately 1.96%. However, using other types of water as curing water instead of TW led to a noticeable increase in the IA porosity, with the highest levels observed when the curing water was NBW, and this was irrespective of the type of water used as mixing water. Particularly, utilizing NBW for both mixing and curing resulted in an IA porosity of 8.95%, which is an exceptionally high value. Upon examining the porosity characteristics of both NBW and NBW + C, it is evident that the presence of nanobubbles contributes significantly to the increase in IA porosity. Kim et al. [
88] reported that the NBW which was utilized as mixing water in cementitious materials resulted in a significant reduction in large pores (> 1 μm) from 40% to 15%, while gel pores (< 10 nm) increased from 6.8% to 12.3%. Liu et al. [
89] reported that nanobubbles crushed the large pores into smaller ones; this improved the initial density of the matrix pore structure in the cement paste. This pore formation trend was evident even in a dense matrix with a remarkably low W/B ratio. In addition, Asadollahfardi et al. [
65] reported that small-diameter nanobubbles significantly enhanced the rate of flotation of cement particles, leading to more frequent particle collisions. This resulted in a filling effect, where additionally generated hydrates fill the pores of various sizes, owing to the promotion of hydration and pozzolanic reactions by the nanobubbles.
3.2. Compressive strength
Fig. 6 presents the compressive strength results for the SHCC specimens prepared using different water sources for the mixing and curing processes.
The introduction of nanobubbles clearly enhanced compressive strength. Utilizing TW as the mixing water resulted in a maximum compressive strength of only 118.4 MPa; however, compressive strengths exceeding 120 MPa were achieved in all cases in which other types of mixing water were incorporated. Although an increase in porosity is directly correlated with a reduction in the compressive strength and durability of concrete [
29], nanosized pores have no significant impact on compressive strength. When comparing the trends of IE porosity and compressive strength, it was confirmed that the two factors were within a close sphere of influence. The maximum compressive strength recorded was 128.5 MPa for the specimens in which NBW + C was utilized for both mixing and curing. This result reflects the beneficial effect of using the CO
2-capturing method in both phases, indicating that this combination may significantly contribute to the development of high-strength cementitious composites. In a CO
2-rich environment, cement composites generate additional CC that increases the density of the matrix [
29]. Some of the unreacted CC generated during this process acted as an inert filler, reducing the porosity and enhancing the compressive strength of the composites [
90]. The introduction of NBW and NBW + C was observed to have a greater effect on the development of compressive strength in the mixing phase than in the curing phase. This result was attributed to the nanobubbles facilitating the pore-filling and the microstructural development during the initial phase of the setting and hardening process of the mixture. Pichler et al. [
91] proposed that the formation of the microstructure was mostly completed in the initial reaction stage after mixing, supporting the above analysis.
3.3. Tensile characteristics
3.3.1. Tensile behaviors
Fig. 7 shows the average tensile stress-strain curves of the SHCCs, and
Fig. 8 shows images of the failed specimens after the direct tensile test.
The width of the initial crack increases with the load; however, cementitious composites reinforced with multiple fibers can delay structural failure by generating additional microcracks. The integration of numerous PE fibers into the SHCC induced strain-hardening behavior by mitigating the formation of macrocracks and promoting the development of multiple microcracks [
92], [
93], [
94]. According to Li and Li [
95], strain hardening occurs when the complementary energy reaches a value higher than the crack-tip toughness, and the maximum fiber bridge strength is greater than the maximum crack strength of the matrix. The ability of PE fibers to induce strain-hardening behavior in the SHCC was markedly more evident than that of steel fibers. Kim et al. [
92] reported that an elevated PE fiber content led to a higher strain capacity and lower tensile strength in SHCC. High-performance cementitious composites exhibit a gradual increase in multiple cracking behavior as the quantity of PE fibers incorporated increases [
93]. However, the integration of polymer fibers into the SHCC reduces the flowability of the mixture. This reduction in flowability can affect the fiber dispersion and potentially result in decreased strength, which requires careful consideration [
94].
The maximum point on the tensile stress-strain curve, where the applied load reached its highest level, was designated as the tensile strength. The strain corresponding to this maximum tensile stress was then labeled as the strain capacity, which measures the elongation of the SHCC specimen under tensile stress. The section below the tensile stress-strain curve that extends to the tensile strength is termed the
g-value and reflects the energy absorption capacity. A summary of the tensile parameters is shown in
Fig. 9.It is evident that the use of NBW adversely affected the tensile strength of the SHCC specimens. By contrast, employing NBW + C during both the mixing and curing processes enhanced the tensile strength. A minimum tensile strength of 10.2 MPa was recorded when both the mixing and curing water were NBW, whereas a maximum strength of 12.7 MPa was attained with the exclusive incorporation of NBW + C during the entire manufacturing process. The observed tensile strength trends corresponded with the compressive strength results, suggesting a consistent relationship between the two properties irrespective of the type of water used. Importantly, the findings indicate a deliberate design focus in SHCC toward achieving a high strain capacity and g-value, even at the cost of tensile strength.
SHCC consistently achieved strain capacities exceeding 2% in all cases. Specifically, the integration of NBW or NBW + C as mixing water facilitated strain capacities greater than 4%. Although the utilization of NBW for mixing water enhanced strain capacity achieved when TW or NBW served as the curing water, the highest strain capacity of 7.79% was achieved in specimens using NBW + C during both the mixing and curing processes. These results imply that the continual utilization of NBW + C in SHCC production processes is a significant factor in improving strain capacity by maximizing the CO2 supply. Consequently, it is inferred that the IA pores formed by the nanobubbles during the mixing and curing processes promote the initiation of microscale cracks, thereby manifesting exceptionally enhanced ductility. Furthermore, the continuous supply of CO2 during curing appears to inhibit macrocrack formation and contribute to the development of multiple cracks in the SHCC. This behavior enhances the overall resilience of the material, suggesting that strategic choice of mixing and curing conditions can effectively improve the performance of SHCC in practical applications.
All the SHCC specimens used in the tensile experiments were able to achieve a very high
g-value exceeding 200 kJ·m
−3. This result significantly surpasses the typical energy absorption capacity of ECC (148 kJ·m
−3) achieved by Ranade et al. [
96]. Although the incorporation of nanobubbles during mixing led to a substantial increase in the
g-value, curing with NBW did not contribute significantly to this enhancement. The primary contributor to achieving a high
g-value was the increase in strain capacity. This was attributed to the broad distribution of strain capacity, ranging from 2% to 8%, compared to the tensile strength of SHCC formed within a narrow range of 10 to 13 MPa. Previously, it was reported that the strain capacity improved owing to a continuous supply of CO
2 accompanied by nanobubbles. Consequently, employing NBW + C in only one of the two processes namely mixing or curing resulted in
g-values of 526 and 353 kJ·m
−3, respectively. In contrast, introducing NBW + C in all the processes resulted in achieving the highest
g-value at 1023 kJ·m
−3, demonstrating that CC-SHCC surpasses existing ultra-high-performance materials [
97].
3.3.2. Micro crack analysis using DIC technique
Fig. 10 illustrates the average and maximum crack widths along with the crack distribution pattern corresponding to the strain levels of the SHCC specimens. A detailed comparison is made by presenting the data and DIC images of the specimens cured using NBW + C.
Prior to crack initiation, cracks of measurable width were not detected, and no indication of strain increase through a color change was observed in the DIC image. Instead, all visible areas maintained a uniform blue color, indicating a strain level of 0%. As microcracks appeared, the color changed from blue to light green and progressively transitioned to red as the crack width increased. Red color indicates greater strain (approximately 20%), whereas the light green color (approximately 10%) indicates relatively less strain.
The specimen utilizing TW as a mixing water exhibited a low strain capacity of 3%, reaching the maximum crack width limit of 300 μm early, thereby inducing failure. In addition, the specimen exhibits a small number of cracks (fewer than 10), leading to an irregular and rapid increase in the average crack width. The specimen utilizing NBW as mixing water exhibited an enhanced strain capacity of 6.5%, and the maximum crack width was observed to increase more gradually compared with the TW specimen. The number of cracks recorded in the NBW specimen was more than 20, which directly contributed to the even distribution of the applied load and enhancement of the strain-hardening region. Consequently, the average crack width remained stable at approximately 150 μm from the moment the first crack appeared until the specimen failed. The specimen with NBW + C as mixing water achieved a notable strain capacity of 8%. The maximum crack width reached 300 μm at a strain level of 7%, and the average crack width was sustained at 150 μm until just before the specimen failed. Similar to the NBW specimen, the number of cracks was greater than 20; however, the NBW + C specimen exhibited enhanced tensile strength and strain capacity. This result indicates that CO2 delivered to the SHCC matrix by the nanobubbles had a minimal effect on the number of cracks but was essential for achieving a high g-value by enhancing the strength and strain.
The results of the crack analysis based on the strain measurements derived from the DIC analysis are summarized in
Fig. 11,
Fig. 12. Owing to variations in the actual strain across different variables, normalization was conducted using the strain at the point of maximum stress to calculate the average value, as defined in Eq. (1):
where εn indicates normalized strain, ε indicates tensile strain, and εu indicates strain capacity at the point of the maximum tensile strength.
Fig. 11 shows the average crack widths in accordance with the normalized strain. At a normalized strain of 1 corresponding to the maximum strain, all tested specimens demonstrated an average crack width close to 150 μm. Previous studies [
98,
99] on the self-healing properties of cementitious composites generally recommended a crack width of less than 50 μm to promote self-healing, with the maximum self-healing limit identified as 200 μm. Yang et al. [
98] reported that when the crack width exceeds 150 μm, the resonant frequency, which reflects microstructural damage in the cementitious composite, remains unchanged after the healing cycle. Accordingly, the SHCCs produced in this study were regarded as having an unstable self-healing potential. The curing water significantly influenced the configuration of the crack width-strain curve. The specimens that utilized TW as curing water exhibited rougher curves than those produced with mixing water TW. A notable increase in the average crack width was observed at a normalized strain of 0.2, with instances of the average crack width in the TW-TW specimen exceeding 200 μm. This trend appears to be fundamentally driven by the presence of only a small number of cracks. However, utilizing NBW or NBW + C as the curing water results in the formation of a more stable curve. When the same mixing water is used, the effects of both types of curing water on the average crack width of the SHCC were similar, with no significant difference in the number of cracks. Consequently, this result suggests that nanobubbles in the curing water are the primary factor contributing to the increase in the number of microcracks.
Fig. 12 shows the maximum crack widths in accordance with the normalized strain. At a normalized strain of 0.2, specimens prepared with TW as the mixing water showed a maximum crack width of approximately 100 μm. In contrast, the use of NBW as the mixing water enabled control of the maximum crack width down to 49.5 μm, whereas the use of NBW + C increased the value up to 167.8 μm. This trend continued until the ultimate strain was reached. At a normalized strain of 1.0, the specimens produced with TW as the mixing water exhibited maximum crack widths in the range of 250-300 μm. When NBW was used as mixing water, the maximum crack width could be reduced to 228.3 μm, while NBW + C as the mixing water increased to as much as 335 μm. These results demonstrate that the composition of the mixing water plays a dominant role in determining the maximum crack width. Furthermore, the curing water served to either complement or amplify this tendency, depending on the specific combination.
3.3.3. Fiber surface investigation
Fig. 13 presents SEM images of the PE fiber surface from the tensile specimens cured using NBW + C. No matrix particles detached from the specimen or large scratches were observed in the fibers extracted from the SHCC specimens fabricated with TW as the mixing water. According to Zhou et al. [
100], PE fibers are highly effective in maintaining interfacial cracks between the matrix and fibers, and promoting multiple cracks with small widths because of their high toughness. Consequently, the PE fibers formed a substantial plastic zone in the matrix layer, which inhibited fiber delamination. This distinct behavior was not observed in the control groups having polyvinyl alcohol (PVA) or steel fibers. Meanwhile, numerous matrix particles were found to be attached to the fiber surface in the specimen made with NBW as mixing water. Matrix particles found on the surface of the fibers formed in small-scale localized areas or accumulated in the chunks. According to Kim et al. [
101], a high concentration of nanobubbles enhances the hydration and pozzolanic reactions by increasing the interactions between water molecules and reactive material particles. Once the hydration reaction is initiated, reactive elements, such as calcium ions, are released, leading to the formation of C-S-H and various hydration products. Finally, the surface of the fiber extracted from the specimen mixed with NBW + C exhibited the formation of large chunks of particles. This fiber was coated thoroughly with chunks that obscured its surface, significantly contributing to the maximized strain capacity and
g-value through enhanced adhesion performance. As confirmed by the porosity analysis, the nanobubbles broke down the macropores of the cement matrix into several nanosized pores. According to Zhang et al. [
102], the nanobubbles cause an increase in the number of nanosized pores, which contribute to a more stable tensile strain process by crushing the matrix more finely during fiber pullout.
3.4. Chemical analysis
3.4.1. CO2 consumption capacities
The derivative thermogravimetric (DTG) curves in
Fig. 14 illustrate the temperature variations between the samples and alumina.
The differential weight loss corresponding to C-S-H and AFt is noticeable throughout the temperature range spanning from 50 to 150 °C. However, the precise quantification of these three chemical components is hindered by closely situated peaks [
81,
103]. The main constituent that significantly influences the porosity and compressive strength of concrete is C-S-H [
28,
104]. Therefore, another primary objective was to investigate the formation of C-S-H within the SHCC, particularly focusing on the effects of the nanobubbles and CO
2 transported by the bubbles through the use of FTIR analysis. Yu et al. [
105] indicated that the remaining quantity of CC in the temperature span of 600-800 °C can be calculated using of Eq. (2):
where rCC (%) indicates residual quantity of CC, m600 and m800 indicate the weight at 600 and 800 °C, msample indicates the inceptive weight of the sample.
Regardless of the mixing water used, a significant increase in CC production was observed when curing involved NBW + C. When NBW and NBW + C were used as the mixing waters, this effect was further intensified. Consequently, the ample CO
2 captured by the nanobubbles during curing significantly increased the CC production in the SHCC samples. In addition, NBW curing consistently produced higher CC than TW curing. Considering that neither TW nor NBW was artificially infused with CO
2, this result suggests that physical effects are the primary factors rather than chemical influences [
101], which implies that NBW curing can partially prevent the leaching of CC [
106] as the curing period increases. Cerrón-Calle et al. [
107] reported that NBW promotes the production of CC and restricts the migration of calcium ions from the matrix to the surrounding curing water through electrostatic interactions between the negatively charged surfaces of the nanobubbles and the positively charged calcium ions, thereby promoting reaction efficiency.
The assumption that CO
2 from both the inside of the cementitious composites and the external environment influenced the CC formation led to the calculation of the amount of CO
2 fixed in each specimen using Eq. (3) [
40]. Additionally, Eq. (4) was employed to calculate the net CO
2 consumption capacity to evaluate the exclusive influence of the various water types utilized in the mixing and curing processes:
where fCO2(%) indicates the CO2 fixing rate, MWCO2 and MWCC indicate the molecular weights of CO2 and CC, CCO2(%) indicates the net CO2 consumption capacity, and PCO2 is the CO2 fixing rate of the control specimen.
A summary of the C
CO2 values for the SHCC samples is shown in
Fig. 15, except for the reference specimen associated with the specified parameters.
The impact of the curing water type on C
CO2 was particularly pronounced. When TW was used as the mixing water, curing with NBW resulted in a C
CO2 of only 0.252%, whereas curing with NBW + C achieved a C
CO2 of 1.202%. This latter value was approximately 93% higher than the 0.622% recorded with NBW + C mixing water and NBW curing water, the highest value excluding those cured using NBW + C. Therefore, it was confirmed that NBW and NBW + C had a more pronounced effect on enhancing CC production during the curing process than during the mixing process. In contrast to TW curing, in which the hydration and pozzolanic reactions are rapidly completed in the initial curing stage, the continuous supply of nanobubbles and CO
2 facilitates the reactivity of the unreacted cementitious particles. C
CO2 exhibited a stronger response to the application of nanobubbles and CO
2 during curing, thereby differing from the trends observed in other parameters. The results confirmed that the formation of the microstructure and mechanical performance were more profoundly affected during the mixing process than during the curing process, although an abundant supply of nanobubbles and CO
2 significantly accelerated CC production during the curing phase. Ultimately, an exceptionally high C
CO2 of 2.333% was achieved when both the mixing and curing water were NBW + C. Comparing this to the C
CO2 of 0.219% for the mortar-type NBW + C sample produced by the same curing conditions (NBW + C curing at 40 °C for 28 d) from a previous study [
47], it is evident that the SHCC sample exhibits a C
CO2 that is more than tenfold higher. However, previous studies where CO
2 was directly injected into the cementitious materials reported higher C
CO2 values. Rostami et al. [
80] suggested that C
CO2 of pure cement paste cured for 2 h in a 99.5% CO
2 concentration environment can reach up to 3.256%. According to Lopez-Arias et al. [
108], the CCO2 of cement paste exposed for 8 h in a chamber where the CO
2 concentration was maintained at 25% was recorded to be 6.15%. Additionally, the C
CO2 of a sample in which 1% of the cement was replaced by nano-TiO
2 under the same conditions improved to 8.16%. Meanwhile, Siddique et al. [
109] claimed that a C
CO2 of 6.12% could be achieved when 30% of the OPC in the cement paste was replaced with belite-rich cement. The corresponding test involved a curing process for 24 h in an environment with 10% CO
2 concentration. However, these studies measured C
CO2 using only reactive materials without any fillers or aggregates. Therefore, the C
CO2 of the high-performance cementitious composites is inevitably lower than C
CO2 of the paste samples. In a study by Dixit et al. [
110] to evaluate the C
CO2 of ultra-high performance concrete (UHPC), the curing process was conducted for 16 h in an environment with a CO
2 concentration of 99%. The C
CO2 of UHPC utilizing only OPC and SF as reactive materials was confirmed to be 0.409%. In contrast, UHPC with 50% of OPC replaced with GGBS exhibited a dramatic increase in C
CO2, achieving a value of 1.07%. Consequently, the C
CO2 value of 2.333% achieved by the CC-SHCC system developed in this study is significantly high, considering several inherent limitations. This is noteworthy for overcoming considerable challenges posed by the extremely low W/B ratio of the mixture. Considering that SHCC is a highly reactive material engineered to maximize tensile performance by reducing the usage of sand to 30% level of the total binder, it is ultimately regarded to have significant potential because the C
CO2 consumption ability and tensile performance are considered simultaneously.
3.4.2. Residual amount of C-S-H
The C-S-H gel structure in the SHCC samples was qualitatively analyzed using FTIR spectroscopy, as shown in
Fig. 16. The wavenumbers of 1084, 796, 697, 522, and 460 cm
−1 correspond to the peaks of quartz [
111]. A notable shift in the major quartz peak was observed as moving from 1084 to 1121 cm
−1 in the SHCC samples, with the peak being distinctly sharp in all samples that excluded NBW + C. This shift is not merely a spectral variation; it signifies potential modifications in the interactions and configurations of the material at the molecular level. Notably, a sharp peak was observed in all the samples that did not utilize NBW + C, suggesting that these conditions led to a more stable and possibly advantageous crystalline structure [
112,
113]. Another shift was observed in the peak linked to the vibrations of O-H stretching, transitioning from 3640 to 3450 cm
−1 [
114,
115]. Independent of the type of mixing water, the intensity of this peak was minimal in the samples cured with TW, whereas it was most prominent with NBW + C curing. Therefore, both the nanobubbles and CO
2 present in the SHCC matrix appeared to promote hydration, aligning with the patterns noted in mortar samples from earlier studies [
28,
40,
47]. On the other hand, the asymmetric stretching of Si-O is represented by a vibration peak located at a wavenumber of 965 cm
−1 [
116], [
117], [
118], [
119], which can be utilized to assess the polymerization characteristics of C-S-H [
114,
116,
118,
120]. Specifically, the elevation of the vibration peak within the wavenumber spectrum signifies an increase in the presence of [SiO
4]
4− chain middle groups integrated into the C-S-H gel configuration of the SHCC [
117,
119]. For the four SHCC samples that had no exposure to NBW + C during mixing or curing, the only sharp peak observed was near 1084 cm
−1, indicating quartz formation. However, in the five samples in which the nanobubbles were applied at least once, the Si-O peak was strongly present, forming two interacting peaks. According to Grzegorczyk-Frańczak et al. [
121], nanobubbles present inside the cement composite accelerate the transfer and accumulation of calcium ions and hydroxide ions, thereby accelerating the formation of C-S-H gel and hydration products. In addition, Asadollahfardi et al. [
65] reported that the very small diameters of the nanobubbles increase the probability of collision with other particles, thereby causing additional reactions of unreacted cement particles. Consequently, comprehensive observations using two chemical analysis techniques confirmed that the introduction of nanobubbles and the increase in CO
2 concentration generated significant amounts of CC and C-S-H.
3.4.3. Relation between chemical products and mechanical properties
To quantitatively evaluate the effect of chemical composition on the mechanical properties of SHCC, the correlations between C
CO2 and the mechanical indicators were analyzed. The results revealed that the correlation between C
CO2 and
g-value exhibited the highest coefficient of determination (
R2 = 0.6259). The corresponding relationships are shown in
Fig. 17.Previous studies [
122], [
123], [
124] have emphasized that the formation of CC enhances the mechanical behavior of fiber-reinforced composites. Wei et al. [
122] reported that the increased tensile performance of high-performance cementitious composites resulted from improved fiber pull-out behavior and stronger fiber-matrix adhesion. Li et al. [
123] further demonstrated that the filler effect of irregularly shaped CC particles enhanced interfacial bonding in composites incorporating polymer fibers. Similarly, Yu et al. [
124] reported that dense CC formation in the interfacial transition zone of polymer fiber-reinforced mortar enhances the post-cracking toughness by strengthening the fiber-matrix interface. Synthesizing these findings indicates that CC not only serves as an inert filler occupying pores within the matrix but also plays an active role in reinforcing the fiber-matrix interface. Notably, SHCC systems incorporating high-strength polymer fibers, such as PE, benefit from abundant CC formation, resulting in enhanced energy absorption and improved crack-bridging performance. These characteristics suggest important directions for optimizing the design of CC-SHCC. Consequently, promoting CC formation through nanobubble techniques can serve as a key mechanism for enhancing the fiber-matrix interactions and improving the mechanical performance of fiber-reinforced cementitious composites.
3.5. Environmental impact
Fig. 18 presents the quantified environmental impact of SHCC, with all the indicators normalized to those of UHPC [
125] for comparison.
SHCC exhibited less than 70% of the environmental impact of UHPC in most categories, including ODP (63.5%), POCP (69.2%), AP (67.4%), and EP (51.6%). However, the normalized GWP of SHCC reached 78.3%, indicating a relatively higher contribution to global warming than the other categories. Material-specific analysis of each impact category identified OPC as the dominant contributor in all the evaluated categories. This result is attributed to the high energy consumption inherent in cement production processes, reaffirming that the OPC content is a critical determinant of the overall environmental impact of SHCC. As shown in
Fig. 18, the relative GWP of SHCC could be reduced to 69.8% when nanobubble technology was applied to enhance CO
2 consumption. These results highlight the potential of CC-SHCC as a promising strategy for the production of sustainable fiber-reinforced composites. Although SHCC has clear environmental advantages over UHPC, its relatively high GWP indicates that further reduction efforts are necessary. In this context, promoting CO
2 consumption using nanobubble technology is an effective method for lowering carbon footprint. This approach provides a practical strategy for achieving superior mechanical performance and enhancing the environmental sustainability of cementitious composites.
4. Conclusions
This study aimed to develop a CC-SHCC with superior carbon consumption capacity and enhanced mechanical properties. Various types of water were employed for mixing and curing, with nanobubbles and CO2 highlighted as the key supplementary factors. The primary conclusions of this study are as follows.
(1) The porosity analysis revealed that SHCC samples exhibited IE porosities in the range of 2.11%-3.61%, with the highest value recorded for TW curing and the lowest (2.11%) for NBW + C in both mixing and curing. In contrast, IA porosity varied significantly, reaching 8.95% when NBW was used throughout the process, demonstrating the role of nanobubbles’ in increasing IA porosity.
(2) The compressive strength results confirmed that NBW and NBW + C positively influenced strength development. The highest strength (128.5 MPa) was achieved with NBW + C for both mixing and curing, highlighting the contribution of CO2 consumption to matrix densification. Additionally, unreacted CC acts as an inert filler, reducing the porosity and further improving the compressive strength.
(3) Although NBW slightly reduced the tensile strength, the use of NBW + C for both mixing and curing significantly improved the strength. The highest strain capacity (7.79%) and g-value (1023 kJ·m−3) were recorded under NBW + C conditions. The specimens incorporating nanobubbles (NBW or NBW + C) exhibited more than 20 microcracks and showed gradual crack widening, promoting distributed cracking and strain hardening.
(4) FTIR analysis revealed stronger Si-O peaks in the NBW + C specimens, indicating enhanced C-S-H formation. TG analysis confirmed significant CO2 capture during curing, increasing the CC production, with NBW + C achieving a CCO2 of 2.333%. A strong correlation was observed between CCO2 and g-value, suggesting that the additional formation of CC through the incorporation of nanobubbles and CO2 contributed to improved fiber-matrix interfacial bonding and enhanced overall tensile performance of the SHCC.
(5) SHCCs exhibited lower environmental impacts than UHPC [
125] for most LCA indicators. OPC was the main contributor to the environmental impact in every indicator owing to its high energy use during production. Applying nanobubble technology reduced the GWP of the SHCC from 78.3% to 69.8%, highlighting that CC-SHCCs are a promising option for combining mechanical strength and environmental benefits.
PDF
(5710KB)
