A Comparative Evaluation of Polymer-Modified Rapid-Set Calcium Sulfoaluminate Concrete: Bridging the Gap Between Laboratory Shrinkage and the Field Strain Performance

Abstract

Rapid pavement repair demands materials that combine accelerated strength gains, dimensional stability, long-term durability, and sustainability. However, finding materials or formulations that offer these balances remains a critical challenge. This study systematically evaluates two polymer-modified belitic calcium sulfoaluminate (CSA) concretes—CSAP (powdered polymer) and CSA-LLP (liquid polymer admixture)—against a traditional Type III Portland cement (OPC) control under both laboratory and realistic outdoor conditions. Laboratory specimens were tested for fresh properties, early-age and later-age compressive, flexural, and splitting tensile strengths, as well as drying shrinkage according to ASTM standards. Outdoor 5 × 4 × 12-inch slabs mimicking typical jointed plain concrete panels (JPCPs), instrumented with vibrating wire strain gauges and thermocouples, recorded the strain and temperature at 5 min intervals over 16 weeks, with 24 h wet-burlap curing to replicate field practices. Laboratory findings show that CSA mixes exceeded 3200 psi of compressive strength at 4 h, but cold outdoor casting (~48 °F) delayed the early-age strength development. The CSA-LLP exhibited the lowest drying shrinkage (0.036% at 16 weeks), and outdoor CSA slabs captured the initial ettringite-driven expansion, resulting in a net expansion (+200 µε) rather than contraction. Approximately 80% of the total strain evolved within the first 48 h, driven by autogenous and plastic effects. CSA mixes generated lower peak internal temperatures and reduced thermal strain amplitudes compared to the OPC, improving dimensional stability and mitigating restraint-induced cracking. These results underscore the necessity of field validation for shrinkage compensation mechanisms and highlight the critical roles of the polymer type and curing protocol in optimizing CSA-based repairs for durable, low-carbon pavement rehabilitation.

1. Introduction

Concrete pavement repair remains crucial for infrastructure resilience, significantly affecting safety, functionality, and economic sustainability by minimizing disruptions to traffic flow. Traditionally, rapid pavement repairs have relied on high-early-strength Type III Portland cement-based concrete (OPC), primarily due to its rapid strength gain, allowing quick reopening to traffic [1,2]. However, OPC-based systems inherently suffer from significant early-age shrinkage, a susceptibility to cracking under restrained conditions, and substantial environmental impacts, notably high carbon dioxide (CO₂) emissions associated with cement production [3]. Currently, cement manufacturing contributes approximately 8% to global anthropogenic CO₂ emissions, posing a critical sustainability challenge [4]. Thus, the construction sector is increasingly shifting toward sustainable alternative binders and admixtures that mitigate these limitations without compromising performance.

The development and application of rapid-set cement-based materials for repair applications have been extensively studied. Research has focused on the material properties, performance characteristics, and field applications of these materials [5,6,7]. These studies have shown that rapid-set materials offer advantages such as faster setting times, a higher early strength, and improved durability compared to traditional cementitious materials. Studies such as Barde et al. (2006) [5] and Yang et al. (2016) [8] have highlighted the effectiveness of utilizing rapid-set materials to significantly reduce the time required for the concrete to set and harden. This allows for the quicker completion of repair projects and minimizes the traffic disruption. Repair materials have also been shown to achieve a high early strength, which means the repaired pavement can be reopened to traffic much sooner compared to traditional concrete [9]. This is particularly important for high-traffic areas where extended closures are not feasible.

One crucial benefit of rapid-set repair materials would be their ability to exhibit enhanced durability, including a better resistance to freeze–thaw cycles and reduced permeability. This leads to longer-lasting repairs and a less frequent need for maintenance [10]. In addition, there have been successful applications in a variety of environmental conditions, including cold weather, where traditional concrete might struggle to set properly [11]. These materials have also exhibited an enhanced bond strength, ensuring a strong adhesion to the existing pavement and reducing the likelihood of repair failures [12]. Lastly, due to their durability and fast setting times, the overall maintenance costs are reduced as the need for frequent repairs diminishes. These advantages make rapid-set cement-based materials a valuable option for efficient and effective concrete pavement repair.

Calcium sulfoaluminate (CSA) cement has gained considerable interest as a sustainable alternative due to its inherently lower carbon footprint (approximately 35–50% lower compared to OPC) [13] alongside its rapid early-age strength development [1,14,15]. CSA cement primarily consists of ye’elimite (C₄A₃S), which hydrates rapidly in the presence of gypsum (calcium sulfate dihydrate, CaSO₄·2H₂O), forming ettringite (AFt) and delivering the rapid strength gain critical for fast-track construction applications [16]. Besides its environmental advantages, CSA concrete demonstrates lower drying shrinkage and enhanced dimensional stability, which are critical for minimizing cracking risks, especially in pavement applications subjected to restrained shrinkage conditions [15,17]. While the initial material cost of CSA cement is typically higher than OPC, its potential benefits in terms of rapid construction, reduced traffic disruption, and enhanced durability may offer significant life-cycle cost advantages, justifying its technical evaluation for such applications [18]. Nevertheless, the comprehensive understanding and empirical validation of the CSA concrete performance under realistic, field-like environmental conditions remain limited. Existing studies predominantly rely on controlled laboratory testing, providing limited insights into the actual pavement performance in carriable outdoor environments, where materials are subjected to environmental fluctuations, including temperature and humidity variations, wind, solar radiation, and restraint-induced stresses [19,20].

The rapid early-age hydration kinetics of CSA concretes, although advantageous for a quick reopening to traffic, inherently pose challenges related to thermal sensitivity, especially at lower ambient temperatures, which can delay hydration and early strength gains [15,21,22]. Recent investigations have also reported complexities in the later-age behavior of CSA concretes, particularly related to the carbonation susceptibility, potential delayed ettringite formation, and shrinkage compensation mechanisms that depend significantly on environmental conditions and mix compositions [17,23,24]. These uncertainties underscore the necessity for detailed comparative investigations under both controlled laboratory and realistic field conditions to ensure reliable performance predictions.

This study aims to directly address this gap by systematically evaluating the early-age shrinkage and temperature behavior of two polymer-modified CSA-based concrete formulations (CSAP with powder polymer additive and CSA-LLP with liquid polymer additive) relative to conventional Type III Portland cement under controlled laboratory and realistic outdoor exposure conditions. Additionally, it will explore the implication for shrinkage compensation mechanisms under realistic curing. As previously highlighted, traditional CSA is primarily composed of ye’elimite (C₄A₃S), which reacts with gypsum to form ettringite, leading to a rapid strength gain [24]. The explicit objectives of this macro-level study are as follows: (1) quantify and compare early-age mechanical properties, shrinkage characteristics, and temperature-induced strain behaviors; (2) investigate the influence of polymer additives on the hydration kinetics, dimensional stability, and early-age performance under variable environmental conditions; and (3) assess the practical viability and performance benefits of polymer-modified CSA concretes for sustainable pavement repairs.

The modification of concrete using polymers and admixtures has been extensively studied. Traditional polymer-modified concrete (PMC) and latex-modified concrete (LMC) have been used in diverse repair applications [25,26]. In this study, a commercially available low-permeability polymer-modified admixture was used to improve the chloride resistance and corrosion protection (therefore, the term polymer-modified should not be confused) [27]. PMC and LMC rely primarily on polymer dispersion to facilitate adhesion and flexibility in concrete mixtures [28,29]. PMC typically incorporates acrylic, vinyl-based, or epoxy polymers, while LMC incorporates styrene–butadiene rubber latex, which improves the bond strength and reduces permeability [28]. The admixture evaluated in this study functions primarily as a corrosion-inhibiting low-permeability admixture, forming a protective film around the embedded steel, while requiring significantly smaller dosages than typical LMC systems [30]. ACI 548.4-11 specifies that conventional LMC systems use approximately 3.5 gallons of latex per 94 lb (one sack) of cement, compared to 10 fluid ounces per 100 lbs. for Liquid Low P used in this study. Studies suggest that LMC improves the bond strength and resistance to bending but does not inherently do much for chloride permeability [31,32]. PMC enhances the compressive strength and shrinkage control, which is subjective to types or dosage amounts. The polymer-modified admixture studied here has been known to limit chloride ion penetration (less than 1000 columbs), making it ideal for bridge deck overlays, pavements, and marine structures [29]. Additionally, while traditional CSA offers rapid setting and reduced shrinkage, its performance can be further tailored using polymer modification. Powdered polymers, integrally mixed, can enhance the workability and bond, while liquid polymers, like the low-permeability admixture used herein, are often designed to reduce permeability and potentially provide internal curing, further mitigating shrinkage [30].

Understanding the early-age shrinkage behavior and thermal sensitivity of polymer-modified CSA concretes under realistic field conditions is essential for optimizing their application in rapid pavement repairs. By bridging laboratory assessments with an outdoor empirical validation, this study provides critical insights into shrinkage compensation mechanisms and the interplay between hydration kinetics and environmental exposure. The findings will support the development of more effective CSA-based repair strategies, reducing premature cracking risks and enhancing long-term durability. Furthermore, this research contributes to advancing sustainable construction materials by demonstrating the viability of CSA mixes in minimizing carbon footprints while maintaining rapid strength gains. The outcomes are expected to inform both material specifications and construction practices, facilitating the broader adoption of CSA-based solutions in infrastructure repair and rehabilitation projects.

2. Materials and Methods

2.1. Materials

This study evaluated two polymer-modified belitic calcium sulfoaluminate (CSA) concrete formulations: CSAP, incorporating powdered polymer blended with cement, and CSA-LLP, using a chemically identical liquid polymer admixture added during mixing. Both polymer types serve to enhance durability, internal curing, and permeability reduction. These CSA mixtures were benchmarked against a conventional high-early-strength Type III Portland cement concrete.

The primary binder materials used included commercial-grade calcium sulfoaluminate cement and ASTM C150 Type III Portland cement [33]. Fine and coarse aggregates were sourced locally, meeting Washington State Department of Transportation (WSDOT) aggregate specifications [34,35]. Both fine and coarse aggregates underwent gradation analyses per ASTM C136 [36] to confirm compliance with standard gradation limits. To enhance the mixture performance, commercially available chemical admixtures were employed, including MasterAir air-entraining admixture (MB Solutions Australia Pty Ltd., Sydney, NSW, Australia), MasterGlenium high-range water-reducing admixture (Master Builders Solutions Canada Inc., Brampton, ON, Canada), citric acid as a hydration retarder, and a proprietary polymer admixture designed specifically for low permeability and corrosion inhibition. Dosages were carefully selected based on manufacturer guidelines and preliminary trials to achieve optimal workability, setting times, and desired durability characteristics for rapid pavement repair applications. Also, note that the specific chemical compositions of the proprietary polymer admixtures and detailed binder mineralogy were not provided by the manufacturers.

2.2. Mixture Proportions

Concrete mixtures were meticulously designed to ensure comparability in mechanical performance and workability, emphasizing the influence of polymer type and environmental conditions. Laboratory mixes used a water-to-cementitious materials (w/cm) ratio of 0.38, while outdoor mixtures utilized a slightly lower w/cm ratio of 0.36. This deliberate adjustment for outdoor conditions was due to differences in mixing methods, volumetric mixer controls, and anticipated ambient temperature impacts on hydration. Additionally, the water-to-cementitious materials ratios were selected based on preliminary trials and manufacturer guidelines to achieve a balance between workability for placement and the low porosity required for durable, high-early-strength pavement repairs.

Table 1. Mixture proportions of mixture samples.

	Mix Type	Cement (lb/yd3)	Coarse Aggregate (¾ in., lb/yd3)	Fine Aggregate (lb/yd3)	Water (w/cm Ratio)	Admixtures per Mass of Cement
Laboratory	CSAP	658	1783	1192	0.38	1% MasterAir, 0.15% Citric Acid
	CSA-LLP	658	1787	1202	0.38	1% MasterAir, 0.15% Citric Acid, 0.40% MasterGlenium, 10 fl oz/cwt Liquid Polymer
	Type III (Control)	658	1783	1192	0.38	0.10% MasterAir, 0.25% MasterGlenium
Outdoor	CSAP	658	1783	1192	0.36	1% MasterAir, 0.15% Citric Acid
	CSA-LLP	658	1787	1202	0.36	1% MasterAir, 0.15% Citric Acid, 0.40% MasterGlenium, 10 fl oz/cwt Liquid Polymer
	Type IL	564	2180	1011	0.44	0.496 lb. Daravair 1000, 1.286 lb. WRDA 64

details the precise mixture proportions, clearly indicating the variations in admixtures between laboratory and outdoor scenarios, facilitating direct comparative analysis. Figure 1a,b provides the gradation of aggregates used in this study.

2.3. Laboratory Tests

Immediately after mixing, the fresh properties of concrete (slump flow (ASTM C143) [37], air content (ASTM C231) [38], and initial concrete temperature) were assessed to ensure consistent workability and air entrainment. These parameters significantly influence the mix’s placeability, finishability, and long-term durability. Mechanical properties, including compressive strength, were assessed using 4 × 8-inch cylinders tested per ASTM C39 [39] at critical early ages (4, 24, 72 h) and standard later ages (7 and 28 days). Flexural strength was measured using beam specimens in accordance with ASTM C78 [40], and splitting tensile strength was determined from cylindrical specimens following ASTM C496 [41]. The comprehensive testing ensured a robust characterization of mechanical performance under varying curing conditions.

Shrinkage characteristics were monitored according to ASTM C157 [42], with measurements at intervals of 1, 4, 7, 14, 28, and 56 days. Specimens were maintained under standard drying shrinkage curing conditions (23 ± 2 °C and 50 ± 5% relative humidity), providing controlled baseline data for comparative analysis. Beams were left in molds (covered with wet burlaps) for 24 h and, thereafter, removed and measured for their initial length followed by consequent evaluations under standard drying shrinkage curing conditions.

2.4. Outdoor Exposure Tests

Recognizing that laboratory conditions may not accurately reflect real-world scenarios, outdoor concrete slabs measuring 5 ft × 4 ft × 12 inches were cast to replicate typical WSDOT pavement dimensions and environmental conditions (see Figure 2a,b). The outdoor slabs were scaled down 1/3 in plan dimension of what is typical in field cast pavements. Despite dimensional scaling due to spatial constraints, typical pavement thicknesses, dowel bar placements, and joint configurations remained realistic. Embedded vibrating wire strain gauges and thermocouples were strategically positioned at mid-depth and near joints/dowel bars to capture comprehensive strain and thermal responses at 5 min intervals over a 16-week exposure period [34,35]. Embedded vibrating wire strain gauges and thermocouples were strategically placed: (1) near the mid-depth of the slab, away from joints, to capture bulk material behavior (see Figure 2b) and (2) adjacent to dowel bar locations (e.g., 2 inches from the dowel bar end) to monitor localized strain concentrations and differential movements near restraints. Recordings were made at 5 min intervals over a 16-week period, capturing the impact of natural environmental variations, including temperature extremes, humidity fluctuations, solar exposure, and wind. Strain gauges were zeroed immediately following concrete consolidation to establish a baseline for subsequent deformation measurements. Also, it is noteworthy that outdoor concrete placement utilized a volumetric mixer truck, providing a more realistic rapid-setting concrete pavement mixing and placement, especially for managing the rapid-setting nature of CSA concretes and accommodating ambient condition variability. For the outdoor samples, freshly cast slabs were covered with wet burlaps immediately after final set to mitigate early-age moisture loss due to environmental exposure (wind, solar radiation). This curing regime, removed after 24 h, aligns with typical DOT pavement panel repair field practices and was essential in maintaining surface moisture balance during the critical hydration window, ensuring realistic assessment of shrinkage and strength development under near-service conditions. The burlaps were removed the next day, and samples were exposed as shown in figure in Section 3.4.1.

The earlier placement of Type IL-OPC slabs by two weeks relative to CSA mixes was intentional, creating adjoining pavement sections to simulate realistic repair scenarios. Although ambient temperature variations between placements were noted, subsequent comparative analyses considered these variations explicitly to ensure validity. Also, OPC panels were not monitored during their placement but monitored and recorded only after the CSA mixes were placed. Overall, this detailed methodological approach, combining rigorous laboratory testing with realistic field conditions, ensures comprehensive evaluation and robust validation of polymer-modified CSA concrete performance, directly addressing previously identified gaps in practical pavement repair applications.

3. Results

3.1. Fresh Concrete Properties

Fresh concrete properties significantly influence the performance and practical usability of concrete, particularly in rapid pavement repairs [28]. The measured fresh properties of the evaluated mixes under both laboratory and outdoor conditions are summarized in

Table 2. Summary of fresh properties.

Mix Type	Initial Slump (in.)	Air Content (%)	Initial Placement Temperature (°F)
CSAP (Lab)	31-inch spread (slump flow)	3.4	87
CSA-LLP (Lab)	8.5	4.0	78
Type III (Control-Lab)	9.0	4.7	84
CSAP Outdoor	8.5	4.5	62
CSA-LLP Outdoor	4.8	4.0	62

.

3.1.1. Workability

Under laboratory conditions, the CSAP mix exhibited a superior initial slump flow (31 inches) compared to the CSA-LLP (8.5 inches) and Type III OPC (9.0 inches). This notably higher slump flow in CSAP is attributable to the combined influence of the powdered polymer and a high-range water-reducing admixture, enhancing the mixture’s plasticity and flow characteristics without segregation [43]. However, under outdoor conditions, the workability of both CSA mixes notably decreased (CSAP: 8.5 inches; CSA-LLP: 4.8 inches), likely due to cooler ambient temperatures (~62 °F), the increased viscosity of the polymer additives, and the faster stiffening associated with rapid hydration kinetics [44,45]. These findings corroborate previous studies indicating polymer-modified CSA concretes exhibit sensitivity to environmental conditions affecting the rheology and setting behavior [46,47].

3.1.2. Air Content

Measured air contents across all mixes ranged between 3.4% and 4.7%, aligning well within the recommended limits (3–7%) for pavement concrete applications to ensure an adequate freeze–thaw resistance [48]. The slightly higher air content in the CSA-LLP (4.0%) relative to the CSAP (3.4%) is likely due to the enhanced surfactant behavior of the liquid polymer admixture stabilizing air bubbles. Outdoor placements showed marginally increased air contents, consistent with limited field compaction practices and shorter working times.

3.1.3. Initial Temperature

Laboratory batches exhibited initial temperatures ranging from 78 °F to 87 °F, reflecting standard indoor conditions, whereas outdoor CSA batches showed notably lower initial temperatures of approximately 62 °F, directly influenced by cooler ambient conditions (48 °F) and the use of cooler aggregates. Lower initial placement temperatures in the CSA concrete offer potential benefits by reducing thermal strains during early hydration and minimizing the risk of early-age cracking. However, these cooler temperatures may also delay hydration kinetics and polymer activation, impacting strength development rates.

The temperature strain gauge reading discussed later in subsequent sections (sub-figures (b) and (c) in Section 3.4) provides a better idea of the temperature dynamics. These variations are indicative of the distinct hydration characteristics of CSA- vs. OPC-based materials. The lower initial temperatures observed in CSA mixes are advantageous for mitigating thermal strains during early-age curing, aligning with the guidelines for acceptable placement temperatures as outlined in the AASHTO and WSDOT guideline for concrete placing temperatures [49,50]. While this may be beneficial for minimizing cracking, it may also delay polymer film formation and hydration kinetics [28].

A rapid initial setting time of approximately 20 min was observed for both CSA laboratory mixes, highlighting their suitability for rapid repair scenarios, enabling swift finishing and early reopening to traffic. This characteristic is consistent with findings in the literature, such as those by Ramseyer and Bescher (2014) [12] and Li et al. (2018) [26], emphasizing the efficiency of CSA-based concretes in minimizing construction disruptions. However, the accelerated hydration kinetics necessitate careful coordination during placement to avoid premature stiffening, particularly in field applications. The inclusion of citric acid proved effective in managing the accelerated setting nature of the typical CSA cement. However, the initial setting time for the outdoor samples was later (about 35–40 min), indicating that the mixtures might be sensitive to colder temperatures. The fast, albeit manageable, setting profile is crucial in field settings to balance rapid traffic reopenings with constructability objectives.

The comparative evaluation further demonstrates the advantages of CSA-based systems in achieving the rapid strength gain while maintaining lower initial temperatures, making them ideal for sustainable pavement repairs. Overall, the evaluation of fresh properties emphasizes that CSA-based mixes possess distinct advantages in rapid pavement repair applications due to their slump balance (albeit optimized with citric acid), rapid initial setting, and appropriate air content. These characteristics underline the critical importance of precise mix handling and placement strategies to exploit their full potential effectively.

3.2. Mechanical Properties

Mechanical properties including compressive strength, flexural strength, and splitting tensile strength are crucial performance indices for pavement repair applications, ensuring the structural capacity to withstand repeated traffic loads. This section assessed each sample under both laboratory and realistic outdoor conditions, with specific attention to the early-age and later-age performance. Figure 3 presents a summary of mechanical property measurements at selected curing intervals.

3.2.1. Compressive Strength

As shown in Figure 3, CSA mixes demonstrated significantly accelerated early-age strength gains compared to the Type III OPC concrete. Both the CSAP and CSA-LLP achieved compressive strengths exceeding 3200 psi within 4 h under laboratory conditions, substantially outperforming the conventional OPC without accelerators, due to the rapid ye’elimite hydration producing ettringite [51]. This performance far exceeds OPC mixes without accelerating admixtures. Note that the Type IL-OPC used in comparison is different from the Type III OPC used for the rapid-set comparison. Type IL-OPC was used for the outdoor pour (see Section 3.4.1), while Type III OPC was produced in the laboratory for the rapid-setting comparison.

However, at 28 days, the compressive strength of CSA mixes (CSAP: 7662 psi, CSA-LLP: 5538 psi) lagged behind the Type III OPC (8462 psi). This behavior aligns with the literature, which reports a typically higher early-age strength but a slower long-term strength development in CSA systems compared to OPC, likely due to distinct hydration phases, including limited belite hydration contributing minimally to long-term strength [52,53].

Hourly compressive strength tests were performed on laboratory-cured specimens. However, for the outdoor-cast specimens, which experienced significantly cooler curing temperatures (ambient ~48–55 °F during first 24 h), a measurable compressive strength (>500 psi, typical threshold for demolding or handling) was not achieved at 2, 3, or 4 h post-casting. This highlights the temperature sensitivity of the CSA cement hydration at very early ages, where the rapid-setting characteristic did not immediately translate to a rapid strength gain under cold field conditions. The first successful breaks for outdoor samples were at 24 h. The observed differences in the lab CSAP vs. the outdoor CSAP at early ages (24 h/72 h) can be attributed to the temperature variations. The outdoor CSA-LLP exhibited a superior early-age compressive strength (3780 psi at 24 h) very close to the CSAP (3737 psi), possibly due to the polymer enhancement [54].

3.2.2. Flexural Strength

Consistent with the compressive strength trend, Figure 4 reveals that the laboratory CSAP and CSA-LLP reached the highest 28-day strength (742; 1005 psi), which was however surpassed by Type III OPC (1237 psi). This behavior is consistent with CSA’s rapid hydration dynamics leading to early-age ettringite development and microstructural stiffening, favoring early flexural resilience. However, OPC’s continued hydration promotes better interfacial bonding and crack bridging over time, enhancing flexural toughness. Studies by Soriano et al. (2019) [44] and Ramseyer & Bescher (2014) [12] confirm similar trends in CSA versus OPC systems, where CSA may provide early merits in flexural stiffness but not necessarily in long-term ductility. CSA-LLP showed a slight drop in strength for the laboratory sample and a lower strength comparatively for the outdoor mixture. This could be due to the film-forming liquid polymer disrupting the uniform matrix cohesion [16,44]. The literature supports that while liquid polymers may improve the shrinkage resistance and chloride ion penetration, their impact on flexural strength is more variable and dosage-sensitive [26,55]. The lower 1-day flexural/tensile strengths of outdoor CSA samples compared to their lab counterparts, despite rapid setting, are consistent with the retarded hydration kinetics due to cold curing (extreme weather) conditions.

3.2.3. Splitting Tensile Strength

The splitting tensile strengths as observed in Figure 5 also reflected the rapid early-age strength characteristics of CSA-based concretes, providing sufficient resistance to tensile stresses induced by early-age shrinkage and thermal gradients [56,57]. These mechanical properties underscore the potential of CSA-based concretes for rapid pavement repair, combining rapid early strength gains with sufficient mechanical properties to ensure an early traffic loading capability, aligning with the objectives of minimizing the infrastructure downtime and enhancing serviceability [58]. At 28 days, however, Type III-OPC regained dominance in tensile strength (571 psi) compared to CSA-LLP (376 psi) and CSAP (402 psi). The difference again is reflective of OPC’s microstructure and continued hydration of C₃S and C₂S. In CSA, early ettringite formation offers a rapid tensile performance, and its dimensional stability may be offset over time by internal shrinkage and the lack of ongoing binder refinement unless assisted by secondary supplementary cementitious materials or fibers.

3.3. Mechanical Properties and Strength Correlations

The relationship between the compressive, tensile, and flexural strength is fundamental to concrete engineering. While compressive strength is the most commonly specified property, tensile and flexural strengths are critical for assessing the cracking resistance, durability, and performance in pavements, slabs, and beams [59]. The analysis in this section will focus on the interplay between compressive strength (f’c), splitting tensile strength (f_sp), and flexural strength (f_r) at both early (1 day) and later (28 day) ages, incorporating a regression analysis and comparative plots to elucidate the behavior of these standard and advanced cementitious systems. Empirical models, such as the one proposed by ACI 318, often relate splitting tensile strength to the square root of the compressive strength (e.g., f_sp ≈ 6√f’c to 7√f’c). However, the values discussed are direct measurements rather than calculations. Figure 6a (scatter plot) shows the 28-day splitting tensile strength versus the 28-day compressive strength revealing a strong positive correlation (R², 0.839), as expected. This indicates that approximately 84% of the variation in the splitting tensile strength can be explained by the variation in compressive strength, which is a strong relationship for this type of data.

Notably, the Type III control mix exhibits the highest strength in both categories, anchoring the upper end of the trendline. The CSA-LLP outdoor mix has the lowest 28-day performance in both metrics. Its significantly lower tensile strength (297 psi) compared to its compressive strength (5538 psi) pulls it slightly below the trendline, suggesting that its field curing may have disproportionately affected its tensile capacity. The CSA mixes generally follow the expected trend, showing that as a class, their tensile-to-compressive strength relationship is not fundamentally different from the Portland cement control at 28 days.

Flexural strength (or Modulus of Rupture, f_r) is also typically correlated with the square root of the compressive strength (e.g., ACI 318: f_r = 7.5√f’c for normal weight concrete). Figure 6b (28-day flexural strength against the 28-day compressive strength) also shows a clear positive correlation (R², 0.768). The R² value of 77% is also strong, indicating that compressive strength is a good predictor of flexural strength for this set of mixes.

The most significant finding here is the performance of the CSA-LLP laboratory mix. Despite having a lower compressive strength than the Type III control (6467 psi vs. 8462 psi), its flexural strength (1005 psi) is much closer to the control’s (1237 psi). This point sits well above the regression line, indicating a superior flexural-to-compressive strength ratio. This is direct evidence that the “liquid polymer” admixture is successfully enhancing the material’s ductility and resistance to bending, which is its intended purpose. The performance gap between the lab CSA-LLP and outdoor CSA-LLP mix is stark. The outdoor mix falls directly on the trendline, showing an average performance. This suggests that the benefits of the liquid polymer were not fully realized in the field-cured slab, likely due to differences in the temperature, moisture, and curing, which can negatively impact the polymer film formation.

For the early-age performance, a key advantage of both Type III and CSA cements is their rapid strength gain. Figure 6c compares 1-day strengths.

As observed, the CSALLP lab mix is the undisputed 1-day champion, achieving a compressive strength of 6018 psi, which is more than double the Type III control. Its flexural strength (1084 psi) is also exceptionally high. This is attributable to the synergistic effect of the CSA cement’s rapid hydration and the MasterGlenium (HRWR), which allows for good workability at a low w/c ratio, accelerating strength development. All lab-prepared CSA mixes outperform the Type III control at 1 day, highlighting their suitability for rapid repair and early trafficking applications. The outdoor mixes, however, show respectable but significantly lower 1-day strengths than their lab counterparts. The lower curing temperature noted in the data (e.g., 62 °F) is a primary cause, as the cement hydration is a thermally activated process.

Moreover, the outdoor CSAP mix interestingly achieved a higher 28-day compressive strength than the lab mix (7662 psi vs. 5794 psi). This is likely due to its lower w/c ratio (0.36 vs. 0.38). However, its 1-day strength and 28-day tensile/flexural strengths were lower, indicating that while the ultimate compaction was good, the early-age curing and overall matrix quality were less ideal than in the lab. For the CSA-LLP mixtures, they showed a dramatic drop in performance from the lab to field across all metrics. The 28-day compressive strength fell from 6467 psi to 5538 psi, and the flexural strength plummeted from 1005 psi to 727 psi. This strongly suggests that this more complex, polymer-modified system is more sensitive to application and curing conditions than the simpler CSAP or Type III mixes. The effectiveness of the liquid polymer, which relies on proper particle coalescence as water leaves the system, can be compromised by a suboptimal temperature and humidity [60].

3.4. Shrinkage and Temperature-Induced Strain Behavior

Drying shrinkage behaviors are summarized in Figure 7a. CSA-LLP exhibited the lowest drying shrinkage in laboratory testing (0.036% at 16 weeks), significantly lower than the OPC and CSAP. Field-cast slabs, summarized in Figure 7b,c, showed an initial expansive phase, transitioning to minimal shrinkage or a slight expansion (~+200 µε at 16 weeks). This notable reduction in shrinkage is attributable to the polymer-induced internal curing and hydration control mechanisms, promoting ettringite formation and stabilizing the pore structure [14,57]. Studies shows that shrinkage-induced strain and temperature-induced variations significantly impact the concrete pavement performance, affecting durability and the early-age cracking potential [61,62,63]. Early-age shrinkage is primarily driven by hydration processes and environmental exposure, while restraint conditions, such as dowel bars and tie bars used at pavement joints, may have an influence on the strain distribution and cracking behavior [61,62,63]. The precise measurement and control of the shrinkage and thermal strains are essential to mitigate structural distress, particularly around these joint restraints, which often become focal points for stress concentrations and early cracking due to differential movements and restraint-induced stresses. These aspects are further elucidated in Section 4.

3.4.1. Laboratory vs. Outdoor Shrinkage

Both CSA mixes, particularly CSA-LLP, exhibited significantly lower shrinkage strains compared to the Type III-OPC (0.0804% at 16 weeks), which was particularly evident at later ages. The reduced shrinkage observed in CSA mixes can be attributed to their rapid hydration reaction, which limits the moisture loss and associated volumetric changes in early-age concrete [8]. The outdoor CSA-LLP showed the lowest shrinkage (0.023% at 28 days and 0.039% at 16 weeks), likely due to the internal curing effect provided by liquid polymer admixtures that help maintain the internal moisture, thereby limiting autogenous shrinkage [14,64]. Laboratory samples for CSA-LLP are very similar (0.019% at 28 days and 0.036% at 16 weeks). Recent studies indicate that binder types significantly affect the thermal and shrinkage behavior; CSA-based binders typically exhibit a reasonable range of heat evolution and reduced shrinkage due to their unique hydration mechanisms and phase formation, resulting in more stable dimensional properties [62,65,66]. CSA cement typically exhibits a higher heat of hydration compared to Type III Portland cement, especially during the early stages of hydration. This is due to the rapid hydration reactions of CSA cement, which are driven by its unique mineral composition, as described in a previous section. This rapid reaction releases significant heat, making CSA cement ideal for applications requiring a high early strength and quick setting [9].

Type III cement, while also designed for a high early strength, achieves this through finer grinding and an altered chemical composition. Its heat of hydration is typically lower than CSA cement but higher than the traditional Type IL Portland cement (OPC) [62,67] used in the outdoor sample, shown in Figure 8, making it suitable for applications where moderate heat generation is acceptable. Zhang et al. (2021) [21] demonstrated that this heat of hydration can be further reduced with supplementary cementitious materials in CSA mixes and still maintain its strength. This characteristic underscore the suitability of CSA for scenarios demanding minimal dimensional changes and high early-age stability [66].

The outdoor exposure testing revealed this consistent trend, with CSA mixes demonstrating an improved shrinkage performance under real environmental conditions, particularly during peak diurnal temperature cycles. The measured temperature-induced strains were also within a reasonable range for CSA-based mixes compared to OPC, further emphasizing their suitability in applications where reduced early-age shrinkage and temperature control are critical. Laboratory-measured shrinkage beams versus outdoor strain measurements have slight differences (0.060% at 16 weeks), possibly because the concrete slabs (though varying restraint conditions) experienced cycles of moisture gain and loss, leading to higher cumulative drying shrinkage (see Figure 7b,c). Nonetheless, CSA-LLP maintained the best dimensional stability even in field conditions, likely due to the film-forming nature of the polymer that restricted excessive water migration [16]. These are consistent with findings by Tortelli et al. (2016) [68], who observed that CSA concretes exhibit rapid hydration and an early strain concentration with outdoor exposure. Note that the positive reading on the figures indicates positive tension. Moreover, Figure 7a,b reveals the critical early-age strain dynamics. For CSA mixes, an initial expansion is observed within the first few hours (reaching ~430 µε), which is characteristic of ettringite formation in CSA cements [57]. This expansion is followed by a noticeable contraction. The OPC slab shows immediate contractive strain. Approximately 80% of the total strain change over the 16-week period for the CSA slabs occurred within the first 24–48 h, dominated by autogenous shrinkage/expansion and plastic shrinkage effects, influenced by the initial exothermic reaction and moisture exchange with the environment (wet burlaps for 24 h.) [69].

Moreover, the slightly elevated strain magnitudes observed in the outdoor samples may be influenced by the presence of solar radiation, wind exposure, and day–night temperature differentials. This reinforces the need to validate laboratory conclusions under field conditions to capture the full extent of volumetric instabilities in early-age concrete. Moreover, the strain trends remained largely stable after week 6, suggesting that CSA mixes rapidly develop sufficient internal resistance to accommodate environmental effects. This observation is further supported by studies conducted by Ke et al. (2022) [14], who report that CSA concrete demonstrates a minimal long-term volumetric drift due to its lower creep and drying shrinkage compared to OPC.

Compared to the laboratory beam shrinkage (Figure 7a), the outdoor slabs (Figure 7b,c) exhibit different net strain behavior. Unlike the OPC slabs, which exhibited net contraction (shrinkage), both the CSAP and CSA-LLP outdoor slabs achieved and maintained a state of net expansion over the two-month monitoring period. Notably in Figure 7b,c, the vast majority of this expansion occurred within the first 24 h, driven by a combination of a high thermal output and the intrinsic chemical expansion of CSA cement. The CSA slabs, after the initial expansion, stabilized at a net positive strain relative to their as-cast state (~+200 µε at 16 weeks), indicating an overall dimensional stability or slight net expansion. The lab results showed all mixes shrinking, while the field results for CSA mixes show expansion. This stark difference is primarily due to the restraint. The small, unrestrained lab prisms are free to shrink, whereas the large outdoor slabs are restrained by the ground and adjacent panels, allowing the expansive potential to build beneficial compressive stress. Lab shrinkage beams (relatively small) may experience more uniform drying and restraint, while large slabs on the plastic sheeting with initial wet curing can capture more of the early expansive potential of CSA. This initial expansion is fundamental to the shrinkage-compensating nature of CSA cements. By expanding before significant drying shrinkage occurs, the net overall contraction is greatly reduced or even offset, leading to a dimensionally more stable material compared to OPC. These aspects are further elucidated in the next section.

The CSA-LLP, with its liquid polymer, consistently showed less initial expansion but also less subsequent contraction than the CSAP in the field, leading to the lowest net positive strain. This may be attributed to the polymer forming a film that moderates moisture exchange and provides internal curing, reducing both autogenous and drying shrinkage components [16,26,69]. Moreover, the outdoor slabs were cast on plastic sheeting (Figure 8), representing a low-restraint condition typical for jointed plain concrete pavements, allowing the free expression of early-age volumetric changes. The field data validates the use of CSA cement for its primary purpose: to counteract drying shrinkage and dramatically reduce the potential for cracking in slabs and repairs.

3.4.2. Temperature-Induced Strain

Temperature sensors embedded around dowels and at the mid-depth (see Figure 8) captured critical insights into thermal behavior. Observations from the outdoor slabs (see Figure 7b,c) indicated minimal differential movements and stresses around the dowel and tie bars for CSA mixes compared to the control, highlighting their ability to effectively mitigate restraint-induced stresses commonly observed at joint interfaces. This behavior underscores the importance of selecting appropriate materials to minimize the cracking potential and extend pavement longevity. CSA’s limited alite content and reliance on ye’elimite hydration result in moderate exothermic profiles, an advantage for volume stability [24]. Note that the early-age data for the outdoor OPC was not captured in this study as the OPC was cast days before the CSA cast, replicating typical repair scenarios. The observed peak internal temperatures in the CSA slabs (Figure 7b), which quickly dropped, although influenced by cooler casting conditions, are consistent with the typically more controlled exothermic reaction of CSA cements compared to high-early-strength OPC [44]. This is particularly beneficial for the reduced risk of thermal cracking, which is especially typical for the restrained scenario.

Sensors located near dowel bars revealed a local strain amplification in all mixes, which was more pronounced in OPC. The CSA mixes exhibited a more uniform strain distribution, indicating better stress accommodation and dimensional compatibility at restrained interfaces. This behavior can be attributed to CSA’s early rigidity and potentially better bonding at joint interfaces due to faster strength development. These observations support the studies on the critical role of early modulus gains in limiting stress concentrations at reinforcement locations [70].

The practical implications of reduced shrinkage include diminished cracking potential, enhanced structural integrity, and reduced maintenance costs associated with early-age distress. These results align with recent research advocating for the use of CSA-based systems for rapid infrastructure repairs, underscoring their benefits in enhancing pavement durability and longevity [12,19,71]. Overall, the shrinkage and temperature-induced strain analysis reinforce CSA-based concretes as a highly favorable alternative for rapid repair applications, providing empirical evidence supporting their adoption to enhance sustainable infrastructure development.

Together, these findings highlight the practical advantages and field responsiveness of rapid-setting CSA systems for sustainable pavement repair applications. In terms of fresh properties, both CSA mixes demonstrated an acceptable workability and air content, which is beneficial for freeze–thaw durability. The CSAP demonstrated superior flow characteristics, particularly in controlled laboratory settings, owing to its blended powder polymer and water reducer. The outdoor placement slightly reduced the slump across mixes, reinforcing the importance of the environmental context in field applications. For mechanical performance, CSA mixes achieved significantly higher early-age strengths than OPC, reaching a structural capacity within 24 h, critical for traffic reopening timelines. Among CSA mixes, the CSAP consistently outperformed the CSA-LLP at later ages, likely due to the better hydration compatibility and polymer–cement synergy.

Regarding the shrinkage and temperature-induced strain, CSA concretes, especially CSA-LLP, showed a superior dimensional stability in both laboratory and outdoor conditions. While outdoor exposure increased shrinkage in all mixes, the CSA-LLP remained the most stable, owing to the internal curing effect of its liquid polymer. The temperature data showed CSA’s lower internal heat evolution and narrower strain cycles, which are beneficial for limiting early-age cracking, particularly around restraints such as dowels. Collectively, these findings confirm that CSA-based concretes, when optimized with polymers and handled with curing care, offer a robust early-age performance, volumetric stability, and environmental resilience, which are key attributes for durable and rapid pavement rehabilitation.

4. Discussion

The performance characteristics of polymer-modified belitic CSA concretes evaluated in this study reveal critical insights into their practical application and underlying hydration mechanisms, specifically under varying environmental conditions. The following discussion interprets and elucidates these findings in the context of the existing literature, highlighting implications for pavement repair applications.

4.1. Fresh Concrete Behavior

The considerable variation in the rheological behavior observed between laboratory and outdoor conditions underscores the complex interaction between polymer additives, CSA cement hydration, and ambient environmental conditions. The significantly enhanced initial slump flow observed in laboratory-prepared CSAP mixes is primarily attributed to the effective dispersion of the powdered polymer in combination with high-range water-reducing admixtures. Previous studies have consistently demonstrated that polymer additives and admixtures in concrete can dramatically improve workability by reducing frictional interactions between cementitious particles and enhancing fluidity [16,26].

In contrast, the rapid loss of workability in outdoor conditions, especially in the CSA-LLP mixture, highlights the sensitivity of liquid polymer admixtures to environmental factors such as lower ambient temperatures and an increased moisture loss due to evaporation. Moreover, the closest probable cause may be due to the slightly lower w/c ratio (0.36) compared to the laboratory mixtures (0.38). Cooler temperatures may decrease polymer dispersion efficiency and polymer–cement particle interactions, directly influencing the concrete viscosity and early stiffening [29,72,73]. These findings emphasize the criticality of controlling the admixture dosage, ambient temperatures, and mixing methods in field applications, aligning closely with the prior literature indicating the temperature-sensitive behavior of polymer-modified cementitious systems [44,74].

4.2. Early-Age Mechanical Performance

The superior early-age mechanical performance of CSA concrete, particularly at 4 and 24 h, can be rigorously explained by the hydration chemistry unique to CSA systems, characterized by rapid ye’elimite hydration forming ettringite crystals. The literature extensively corroborates that early-age strengths in CSA systems originate primarily from the formation of ettringite, whose needle-shaped crystals contribute substantially to early matrix stiffness and strength [57].

At 28 days, the observed strength plateau in CSA mixtures compared to OPC is critically attributable to the relatively limited contribution of belite hydration in the early stages. Belite phases in the CSA cement hydrate much slower than alite phases in OPC, offering a negligible incremental strength gain at early and intermediate ages. This aligns with the literature data reporting a slower long-term strength development in CSA-based concretes, highlighting the necessity for the careful application of CSA mixtures where long-term high-strength properties are less critical than rapid construction and early serviceability [75].

The strength dip at 28 days observed in the CSA-LLP may suggest a destabilization of ettringite or inhibited belite hydration under real-world restraint and thermal conditions. Although SEM and XRD were not conducted, the literature offers corroborative insights: polymer-modified systems may show a delayed ettringite conversion to monosulfate (AFm) and localized microcracks due to the stress concentration around encapsulated hydration products [23,76]. Moreover, the observation may be linked to a combination of other factors; the liquid polymer, while beneficial for shrinkage, may slightly disrupt the cement matrix. Furthermore, the literature suggests that under certain thermal and moisture conditions, the primary strength-giving ettringite (AFt) in CSA systems can convert to monosulfate (AFm), a transformation associated with a loss of solid volume and a potential strength reduction [57,77]. These interpretations merit validation through a targeted microstructural analysis in future studies.

4.3. Shrinkage Behavior and Dimensional Stability

The significantly lower shrinkage observed, particularly in the CSA-LLP mixture, can be rigorously attributed to the internal curing and enhanced hydration uniformity provided by liquid polymer admixtures. Polymer films form on hydration products, modifying pore structures and reducing moisture evaporation, thus directly mitigating shrinkage strains [66,76]. This aligns with multiple previous studies emphasizing polymer-induced pore structure refinement and shrinkage control in cementitious materials [78,79].

Figure 7b, showing the initial 24 h post-casting, is the most critical for understanding the fundamental behavior of these mixes. As observed, the CSA mixes exhibit a massive and rapid temperature spike immediately after casting, reaching a peak temperature of ~100 °F. This is the signature of CSA cement’s rapid hydration chemistry, primarily the exothermic formation of ettringite. In contrast, the adjacent OPC slab, cast two weeks prior, shows no internal heat generation and its temperature tracks the cool ambient air (~45–65 °F). Mirroring this temperature spike, the strain gauges in the CSA slabs show a rapid increase in positive strain (expansion), peaking at approximately +430 microstrain. This initial positive strain is a combination of two powerful effects: (1) the concrete gets hot and expands (thermal expansion) [80], and (2) the formation of ettringite crystals from the hydration of ye’elimite (the primary clinker phase in CSA cement) occupies more volume than the reactants (autogenous effect). This creates an internal, expansive force within the concrete matrix [81].

As the slabs reach their peak temperature and begin to cool back towards the ambient temperature, the strain begins to decrease. This is primarily a thermal contraction. However, as seen in the long-term plot, the slabs do not return to their original volume. The initial chemical expansion has permanently “locked in” a state of positive strain. The initial expansive behavior in outdoor CSA slabs, contrasting with controlled laboratory conditions, is plausibly linked to polymer- and moisture-driven ettringite formation. The ettringite formation in CSA systems typically involves expansion, counterbalancing shrinkage effects, as extensively documented in the existing literature [80,82].

In Figure 7c, spanning approximately two months, reveals the final performance and practical implications. The OPC slabs (OPC_Left and OPC_Right) remain in a state of net negative strain (contraction) for the entire duration. This is classic drying shrinkage; being restrained by the ground, this contraction induces tensile stresses within the slab, making it susceptible to cracking. In stark contrast, the CSAP and CSA-LLP slabs maintain a significant level of positive strain (expansion), settling at around +630 microstrain This means that instead of being in tension, the slabs are in a state of compression. This internal compressive stress must be overcome before any external tensile forces (like those from further drying or thermal cycling) can cause a crack. This is the principle of shrinkage compensation in action, which is typical of the CSA behavior [83]. The peak expansion of ~430 microstrain occurs within the first 12 h. The final stabilized strain after two months is ~630 microstrain. This means that over 80% of the total effective expansion occurs within the first day. This highlights that the success or failure of a shrinkage-compensating system is determined almost entirely by its behavior during the initial curing phase.

These results strongly suggest that the careful management of polymer dosages and curing regimes can effectively exploit these expansive properties to mitigate shrinkage-induced cracking in pavement applications. Field slabs of CSA also showed a slight early expansion followed by stabilization, supporting CSA’s potential in shrinkage-compensating applications. Conversely, OPC slabs lacked expansion phases and typically exhibit higher shrinkage due to their paste content, their finer pore structure, and the absence of expansive mineral phases [84].

Furthermore, the field data appears to directly contradict the lab shrinkage data, which showed all mixes contracting (e.g., CSALLP outdoor at −387 microstrain in the lab vs. +~630 microstrain in the field). This discrepancy is not an error; it is a fundamental lesson in concrete materials science. The lab shrinkage test (ASTM C157/C157M) is performed on small, unrestrained prisms. These prisms are free to move. When they undergo chemical and drying shrinkage, they simply get smaller, and the measurement reflects this as negative strain (contraction). Any expansive potential is simply expressed as the prism getting slightly bigger before it starts to shrink. On the other hand, a large slab cast on the ground is highly restrained. It cannot easily move. When the CSA cement’s chemical expansion occurs, the surrounding ground and adjacent slabs resist this movement. This resistance converts the expansive strain into compressive stress. As ACI 223, the “Guide for the Use of Shrinkage-Compensating Concrete,” explains, this induced compression is the goal. It effectively pre-stresses the concrete, placing it in a compressive state that offsets the tensile stresses that develop later from the drying shrinkage.

In short, the lab measures the free shrinkage/expansion potential. The field measures the result of that potential acting against real-world restraint. The field data proves the material is working exactly as designed. Recent studies, such as those by Xi et al. (2023) [85], have repeatedly shown that the timing and magnitude of the ettringite formation are key to successful shrinkage compensation. The expansion must occur while the concrete’s modulus of elasticity is still relatively low, allowing the expansion to occur without self-induced damage. The field data showing the expansion coinciding with the initial setting and hardening phase perfectly aligns with this principle. Thus, the primary implication is a vastly reduced risk of drying shrinkage cracking. The OPC slab is under constant tension and is at a high risk of cracking. The CSA slabs are under compression and are highly resistant to cracking. This leads to more durable, lower-maintenance structures, especially for industrial floors, bridge decks, and large panel repairs. The data also shows that the CSA slabs expanded and pushed against the OPC slab, effectively tightening the joint. This is ideal for infill panel replacements, as it creates a stable, load-transferring connection without the wide shrinkage gaps that typically form with OPC repairs. The successful outcome in the field, despite the cool ambient temperatures, demonstrates the robustness of the CSA mixes. However, the slightly different behavior between the CSAP and CSA-LLP suggests that admixtures can fine-tune performance, and their interaction should be considered in the mix design phase.

4.4. Thermal Sensitivity and Strain Development

The observed temperature-induced strain behaviors in CSA concretes underscore their favorable dimensional stability compared to OPC under variable thermal conditions. Lower peak internal temperatures and smaller thermal strain amplitudes in CSA mixtures result from the distinct hydration reactions involving ye’elimite and gypsum, generating less heat compared to alite-dominated OPC systems [84,86]. The CSA mixes show a slightly higher peak temperature and expansion. This could be attributed to the presence of the MasterGlenium (a high-range water reducer), which can sometimes accelerate the initial reaction kinetics in CSA systems, leading to a more intense, concentrated heat release [21].

However, the delayed strength development observed at lower ambient temperatures (approximately 48 °F) highlights a critical limitation in CSA concrete applications. Ye’elimite hydration kinetics are notably sensitive to lower temperatures, significantly delaying the early strength gain and compromising the early structural integrity [87,88]. The literature extensively confirms these observations, recommending careful temperature management, insulation, or supplementary curing methods to mitigate adverse thermal impacts on the CSA concrete performance [69,74]. Furthermore, the claim of “thermal sensitivity” for CSA in this study is supported by two key observations: (1) a significant retardation of the initial setting time in cooler outdoor conditions (from ~20 min in the lab to ~35–40 min outdoors) and (2) a dramatic delay in the early strength gain, where a measurable compressive strength was not achieved for several hours outdoors at ~48 °F, in stark contrast to the >3200 psi achieved at 4 h in the lab condition. These macro-level performance indicators are direct evidence of the hydration kinetics’ sensitivity to the temperature.

4.5. Volume Change and Strength Gain

The behavior of any concrete mix, especially in the early stages, can be viewed as a “race” between two competing phenomena: (1) the concrete is trying to expand (due to heat and chemical reactions) and then contract (due to cooling and drying), and (2) the concrete is hardening, developing its modulus of elasticity and tensile/compressive strength. The outcome of this “race” determines whether the slab will be in a state of beneficial compression or detrimental tension. The lab data showed that both CSA mixes, especially the CSA-LLP, gained strength and stiffness extremely rapidly. The CSA-LLP achieved a 1-day compressive strength of 6018 psi and a flexural strength of 1084 psi, far surpassing the Type III cement. Also, the lab prisms showed that these mixes had the lowest intrinsic drying shrinkage potential (CSALLP at −360 microstrain vs. Type III at −803 microstrain after 16 weeks). The rapid strength and stiffness gain is the key. When the chemical expansion from the ettringite formation kicks in (within the first 6–12 h), the concrete is already strong enough to translate this internal expansive force into compressive stress against the ground’s restraint. A weaker, “flimsier” concrete would simply deform or microcrack internally without building significant stress. The high early flexural strength, particularly in the CSA-LLP mix, is critical. This indicates a high tensile strain capacity. It means that the matrix is tough enough to withstand the internal expansive forces without being damaged. This allows for a more efficient conversion of the potential expansion into useful compressive stress. The lower intrinsic shrinkage means that after the initial expansion and thermal contraction, there is less of a driving force pulling the slab back into tension. The initial expansion provides a large compressive “buffer” that the relatively low long-term drying shrinkage cannot overcome.

There exist some nuances in the CSAP vs. CSA-LLP mixes. The CSA-LLP mix showed a slightly higher peak expansion in the field. This aligns with its superior mechanical properties in the lab. The polymer (“LLP”) and high-range water reducer (MasterGlenium) create a denser, more robust early-age matrix that can more effectively harness the chemical expansion and convert it to stress, as evidenced by its superior flexural performance. The CSAP mix, being a simpler system, was still highly effective but slightly less potent. Its robust mechanical properties were more than sufficient to achieve significant net expansion. Its higher-than-the-lab outdoor compressive strength suggests it might be a more “forgiving” mix, less sensitive to curing variations than the polymer-modified CSA-LLP. The Type III mix had a good, but significantly slower, early-age strength gain compared to the CSA mixes (1-day compressive strength of 2565 psi). It also exhibited the highest intrinsic drying shrinkage by a large margin. This is a fundamental characteristic of Portland cement hydration products [89]. The Type III-OPC has no chemical expansion mechanism. Its only early-age expansion comes from the heat of hydration. As it cools, this thermal expansion is completely lost [89]. From that point on, it is subjected to only one dominant force: drying shrinkage. Because it is restrained, this shrinkage potential is immediately converted into tensile stress. The race for OPC is whether its tensile strength gain can keep up with the tensile stress buildup from the shrinkage. Often, especially in adverse drying conditions, the stress builds faster than the strength, leading to the formation of shrinkage cracks. The field data showing the OPC slabs in a state of net contraction (tension) perfectly illustrates this losing race.

5. Conclusions and Practical Implications

5.1. Conclusions

This study provided a comprehensive comparative evaluation of the early-age performance of two distinct polymer-modified calcium sulfoaluminate (CSA) concretes (CSAP and CSA-LLP) against conventional Portland cement (OPC) concrete, under both controlled laboratory and realistic outdoor field conditions. The findings offer critical insights into the suitability of these CSA systems for sustainable and rapid pavement repair applications.

CSA concretes consistently demonstrated superior early-age mechanical properties. In laboratory conditions, compressive strengths exceeding 3700 psi were achieved within 24 h, significantly outperforming OPC. While CSAP generally showed a more consistent strength development, particularly in field conditions, the CSA-LLP also provided a robust early strength. However, outdoor pours conducted under cool ambient temperatures (approx. 48 °F) highlighted the thermal sensitivity of CSA hydration; despite rapid setting times (~20–40 min), the measurable compressive strength gain within the first 4 h was notably retarded compared to laboratory-cured samples, emphasizing the critical role of the ambient temperature in early field strength development. By 28 days, the Type III OPC achieved higher compressive and flexural strengths, consistent with the long-term strength development profile of the Portland cement.

As demonstrated by the regression analyses, strong, positive linear correlations exist between the compressive strength and both splitting tensile and flexural strengths for this suite of mixes (R² > 0.78). This confirms that, as a whole, the mixes behave predictably according to established concrete principles. The CSA-LLP mix proved the effectiveness of the liquid polymer admixture by exhibiting a significantly higher flexural-to-compressive strength ratio compared to all other mixes. This makes it an ideal candidate for applications where toughness and cracking resistance are paramount. The CSA-based mixes, particularly the highly admixtured CSA-LLP, demonstrated a vastly superior early-strength gain compared to the Type III Portland cement control, making them suitable for time-critical construction and repair. This study highlights a significant “lab-to-field” performance gap. The CSA-LLP mix, while the top performer in the lab, was highly sensitive to outdoor conditions, leading to an underperformance relative to its potential. This underscores that the benefits of advanced materials can only be realized with stringent quality control during placement and curing in the field. The simpler CSAP mix appeared more robust and less sensitive to these variations.

For fresh properties, both CSA mixes exhibited acceptable workability and air contents suitable for pavement applications. The CSAP, containing an integral powdered polymer and a high-range water reducer, displayed an excellent flow in laboratory settings. The field placement, utilizing a volumetric mixer, resulted in slightly reduced slumps for all mixes and notably lower initial concrete temperatures for the outdoor batches compared to lab mixes, underscoring the influence of field conditions and mixing methods on fresh properties.

For the shrinkage and volumetric stability, both CSA concretes exhibited substantially lower net shrinkage in laboratory beam tests compared to OPC, with the CSA-LLP (0.036% at 16 weeks) showing the least shrinkage, likely benefiting from the internal curing potential of its liquid polymer admixture. Field-cast slabs revealed a crucial aspect of the CSA behavior not fully captured by standard lab prism tests. The CSA concretes (CSAP and CSA-LLP) exhibited a significant initial expansion within the first 24–48 h, largely attributed to the ettringite formation. Consequently, over the 16-week monitoring period, these slabs often maintained a net positive strain (overall expansion) or minimal net shrinkage relative to their as-cast state. This contrasts with lab prisms, which showed a net contraction for all mixes. This highlights that field conditions, particularly the initial curing (wet burlap) and lower restraint (plastic sheeting), allow for a fuller expression of CSA’s shrinkage-compensating mechanisms. Results also showed the dominance of early-age strain, approximately 80% of the total strain evolution in the outdoor CSA slabs occurred within the first 24–48 h, dominated by the autogenous shrinkage/expansion and plastic deformation, rather than long-term drying shrinkage.

Outdoor temperature profiles showed that CSA mixes, though cast under cooler conditions, experienced lower peak internal temperatures compared to the OPC slab (which was cast on a warmer day). Subsequently, CSA slabs exhibited narrower daily thermal strain amplitudes. This moderated thermal response, combined with early rigidity and dimensional stability, suggests that CSA concretes can better accommodate joint restraint stresses, potentially reducing the risk of early-age thermal cracking. The field instrumentation provides a clear and compelling case study on the efficacy of CSA-based shrinkage-compensating concrete. The data moves beyond simple compressive strength values to demonstrate the functional, in-service behavior that provides potentially enhanced durability and longevity to concrete structures.

5.2. Practical Implications and Future Work

Practical implications include CSA’s potential, particularly polymer-modified variants, as highly promising for rapid pavement repairs due to its rapid strength gain and superior dimensional stability compared to traditional OPC. The choice between the CSAP and CSA-LLP may depend on specific project requirements, with CSA-LLP offering a potentially better net dimensional stability due to its liquid polymer, while CSAP provided more consistent strength under the tested field conditions. Engineers and practitioners must consider the significant impact of the ambient temperature on the very early-strength development of CSA concretes in the field. Rapid setting does not always equate to an immediate high strength under cold conditions. Moreover, standard laboratory shrinkage tests (e.g., ASTM C157 prisms) may not fully represent the net dimensional change in CSA concretes in field applications, as they may not capture the full extent of the initial expansion that contributes to the shrinkage compensation. Field-representative testing or considering the initial expansion is crucial for an accurate performance prediction. Additionally, the observed lower thermal sensitivity and better strain accommodation of CSA concretes can lead to more durable repairs with a reduced cracking potential at joints and restraints.

For crack-sensitive slabs (floors, pavements), the CSA mixes are superior. Their ability to use their rapid strength gain to create a state of internal compression directly counteracts the primary failure mechanism of the large flatwork (drying shrinkage cracking). The lab shrinkage test confirms their low shrinkage potential, but the outdoor strain data proves their functional success. Choosing between the two mixes, if maximum performance and toughness are needed and the field curing can be well-controlled, the CSA-LLP is the premium choice due to its enhanced mechanical properties and expansive potential. For more general applications where the robustness and forgiveness to field conditions are key, CSAP is an excellent and reliable option. This entire analysis is a powerful testament to the fact that standard lab tests (like compressive strength and free shrinkage) only tell part of the story. They measure potential. The true performance is only revealed when the material interacts with real-world boundary conditions like restraint, as perfectly captured by the embedded strain gauge data.

To further advance the understanding and application of polymer-modified CSA concretes, future research should focus on investigating the long-term performance—including the resistance to carbonation, sulfate attacks, freeze–thaw cycles, and abrasion, particularly for these specific polymer-modified formulations—and on a deeper investigation into the specific mechanisms of the interaction between the polymers (powdered vs. liquid) and the CSA hydration products and how these influence the microstructure and long-term properties. Developing and validating field curing protocols tailored for CSA concretes under various environmental conditions, especially to mitigate the impact of cold temperatures on early strength, is also necessary. Extending field monitoring beyond 16 weeks to capture seasonal variations and their impact on long-term strain behavior and durability would also provide expounded insights. Studies investigating the direct measurement of the internal relative humidity in CSA-LLP mixes to quantitatively assess the internal curing effect of the liquid polymer would be useful. Additionally, conducting isothermal calorimetry to precisely compare the heat evolution profiles of the CSA and OPC mixes under controlled conditions would be necessary. Moreover, the study of the chemical and mineralogical composition of the binders used is strongly recommended. Finally, it may be necessary to evaluate the behavior of these mixes under higher degrees of restraint to better understand their cracking resistance in various structural configurations.

Overall, this study affirms the significant potential of polymer-optimized CSA concretes for resilient and low-impact pavement repairs. The integration of field performance data provides crucial empirical validation, bridging a key research gap and supporting the broader adoption of these advanced materials in infrastructure projects.