An experimental study on the influence of fine and coarse aggregates on the strength of polyurethane concrete for highway maintenance

ABSTRACT

This study seeks to assess the impact of the fine-to-coarse aggregate ratio (FA/CA) on the early-age mechanical properties of polyurethane concrete (PUC) designed for quick repair applications. Three sample FA/CA ratios (0.42, 0.54, 1.00) were established based on an optimized particle packing model. The critical characteristics, including compressive and flexural strength, were evaluated after 24 hours. Each test group had at least five specimens, and the results were statistically evaluated using one-way ANOVA and Tukey’s post hoc test (p < 0.05) to verify data reliability. The findings demonstrate that an FA/CA ratio of 0.54 produces enhanced overall strength performance (compressive: 74 MPa; flexural: 8.5 MPa), due to optimized particle dispersion and minimized voids. The FA/CA ratio of 1.00 enhances workability but markedly diminishes mechanical strength. Although this study lacks in-depth discussions of microstructure, long-term durability evaluation, and some transport-related characteristics, the results confirm the potential of PUC. Preliminary results show that PUC is a viable alternative to conventional repair mortars, not only for pavement overlays but also for bridge decks, expansion joints, and other concrete surfaces that require rapid rehabilitation.

Keywords:
Polyurethane concrete; Aggregate optimization; Compressive strength; Flexural strength; Highway maintenance

1. INTRODUCTION

In the context of the rapidly developing construction industry, the need for high-performance and sustainable building materials is becoming more and more urgent. Although conventional concrete has been widely used for decades, it still has limitations in terms of curing time, flexural strength, and resistance to harsh environmental conditions. Phenomena such as cracking, deformation, and degradation of concrete structures under the influence of weather and dynamic loading are significant challenges for engineers and material scientists.

Polyurethane concrete (PUC) has emerged as a promising solution for rapid infrastructure repair, particularly in highway maintenance applications where early strength development and durability under dynamic loading are critical [¹, ²]. Its low curing time, strong chemical bonding, and high mechanical performance under aggressive environments have made it attractive for overlays, bridge decks, and runway rehabilitation [³, ⁴, ⁵]. However, current research has mainly focused on the performance of PUC from the perspective of binder chemistry or thermal resistance [⁶, ⁷]. Meanwhile, the influence of conventional concrete design parameters (e.g., aggregate gradation) on the mechanical behavior of the material has received little attention [⁸].

In particular, the effects of fine aggregate (FA) and coarse aggregate (CA) proportions in PUC remain poorly understood. Although many studies in Portland cement concrete (PCC) and self-compacting concrete (SCC) systems have emphasized the importance of optimizing the FA/CA ratio to improve workability, compressive strength, and fracture toughness [⁹, ¹⁰]. The findings of these studies may not be directly transferable to polyurethane-based systems due to differences in matrix rheology, interface bonding, and curing kinetics. Furthermore, existing studies on PUC rarely isolate the roles of fine and coarse aggregates; most adopt a fixed aggregate blend or vary total aggregate content without a systematic evaluation of gradation effects [¹¹]. This presents a critical research gap, particularly given that mix design “flexibility” is often claimed in the literature [³, ⁴, ¹¹], yet constraints such as formulation cost, binder sensitivity to aggregate absorption, and incompatibility with poorly graded aggregates are seldom addressed explicitly [⁷]. Furthermore, although the FA/CA ratio has been studied in fiber-reinforced concretes [⁹] and SCC [¹⁰], no experimental research to date has quantified its influence in polyurethane systems under controlled conditions. As a result, practical mix designs for PUC in quick-fix applications often rely on empirical adjustments or extrapolations from OPC-based formulations, potentially reducing both mechanical performance and cost-effectiveness in the field.

This study aims to address this gap by conducting a systematic experimental investigation on the independent and combined effects of fine and coarse aggregates on the compressive and flexural strength of polyurethane concrete. Through a series of controlled mix variations, the research seeks to identify aggregate ratios that optimize structural performance while maintaining constructability, thereby contributing practical guidelines for more robust and cost-conscious PUC formulations in highway maintenance contexts.

2. LITERATURE REVIEW

The design of polyurethane concrete (PUC) systems for highway maintenance requires a nuanced understanding of how aggregate gradation affects the mechanical behavior of the composite. Unlike traditional cementitious binders, polyurethane matrices form through rapid chemical crosslinking, often leading to different aggregate-matrix interactions, and potentially different sensitivity to aggregate ratios. Yet, most studies on PUC still borrow assumptions from ordinary Portland cement concrete (OPC), which may not hold under rapid-curing, low-viscosity conditions typical of PU systems [¹²].

Several studies have acknowledged that both fine aggregate (FA) and coarse aggregate (CA) significantly affect the workability and mechanical properties of concrete [¹³, ¹⁴, ¹⁵, ¹⁶]. In self-compacting and fiber-reinforced systems, YARDIMCI et al. [¹⁰] demonstrated that increasing the FA/CA ratio from 0.94 to 1.72 improved flowability and fiber dispersion, leading to enhanced flexural behavior. However, when the ratio was increased further to 2.50, fracture energy declined in low-fiber mixes due to insufficient paste volume. Similarly, IQBAL KHAN et al. [⁹] reported an optimal FA/CA range of 0.9–1.2 in steel fiber-reinforced concrete, where compressive and flexural strength increased by approximately 10–28%, but excessive fine content required large superplasticizer dosages and introduced mix instability.

Contradictions also appear in studies on polyurethane-based systems. JIANG et al. [¹¹] showed that increasing the CA volume fraction up to 60% improved early-age and 28-day compressive strength in PUC, whereas further increases led to performance degradation due to inadequate binder coverage and matrix discontinuity. While their study confirms the relevance of aggregate packing in PUC, it does not separate the effects of FA and CA or systematically assess different aggregate size distributions. Meanwhile, numerical simulations by SHIGANG et al. [¹⁷] revealed that non-uniform aggregate shapes and spacing contribute to stress localization and early crack formation in polyurethane polymer concrete (PPC), highlighting the role of meso-structure, which may interact complexly with gradation ratios.

A critical observation is that although aggregate gradation is often mentioned as a design factor in PUC literature, its optimization remains poorly addressed. For instance, studies like [⁶] and [⁷] explore fillers such as fly ash or slag for improving thermal or chemical properties but provide little guidance on conventional aggregate design. Likewise, while many researchers claim that PUC offers “flexibility in mix design” [³, ⁴, ¹¹], few actually interrogate the associated constraints such as cost sensitivity, proportioning robustness, or workability limits—a point this paper addresses in a separate discussion section.

Moreover, current literature lacks a systematic comparative framework to map how specific FA or CA variations influence strength parameters in polyurethane matrices. Most prior works investigate blended effects, making it difficult to distinguish whether observed performance changes stem from the binder chemistry, aggregate grading, or their interaction.

Table 1 below summarizes key studies addressing aggregate proportions in PUC or related materials, and positions the current work in contrast.

Table 1
Aggregate proportions in PUC.

STUDY	MATERIAL SYSTEM	FA/CA RANGE	BINDER TYPE	HIGHLIGHTS	GAPS REMAINING
YARDIMCI et al. [10]	SFRSCC (fiber-reinforced SCC)	0.94–2.50	OPC	FA/CA affects flow and fracture energy	Not applicable to PU systems
IQBAL KHAN et al. [9]	SFRC (steel fiber RC)	0.8–1.2	OPC	FA/CA ~1.0 optimal for strength and workability	No extension to polymers
JIANG et al. [11]	Polyurethane concrete	Fixed mix	PU	CA up to 60% improves compressive strength	No FA/CA separation; aggregate types unclear
SHIGANG et al. [17]	PPC (numerical model)	Not stated	PU	Aggregate geometry influences stress localization and failure pattern	No empirical validation

As seen, no existing study has independently varied fine and coarse aggregates in a PU matrix to quantify their separate and combined effects on compressive and flexural performance, particularly in the context of highway maintenance where early strength and practical workability are critical. Therefore, the present research directly addresses this gap by isolating and experimentally varying FA and CA contents in polyurethane concrete. The aim is to identify mix combinations that optimize mechanical performance while maintaining practical workability, thus contributing new quantitative insights to aggregate design in PUC systems.

3. AGGREGATE PARTICLE SIZE DISTRIBUTION

In PUC, where the polymeric binder lacks hydration and relies solely on chemical curing, the distribution of aggregate particle sizes plays a more critical role than in traditional cement-based systems. Aggregate typically comprises over 80% of the total concrete volume and directly governs the internal structure, mechanical strength, and resin efficiency during mixing and compaction [¹¹].

Two main aggregate fractions are often involved:

• Fine aggregates (0.15 mm–4.75 mm): responsible for filling voids, improving packing density, and providing a smoother surface texture;

• Coarse aggregates (4.75 mm–19 mm): serve as load-bearing elements and reduce the overall binder requirement [¹⁰].

The ratio of fine aggregate to coarse aggregate (FA/CA) significantly influences workability, strength development, and interfacial bonding quality between aggregate and resin. A well-balanced gradation ensures better coating by polyurethane resin, which, due to its relatively high viscosity and fast curing nature, struggles to wet poorly graded or gap-graded systems [¹⁷]. YARDIMCI et al. [¹⁰] concluded that increasing the FA/CA ratio to 1.72 increased the flowability and cracking energy. However, excessive fine content above this threshold led to increased paste demand and reduced strength. IQBAL KHAN et al. [⁹] emphasized that a balanced FA/CA ratio is essential to maintain both strength and workability, with optimal performance observed around FA/CA = 1.0 in steel-fiber concretes. Yet, these results are drawn from OPC-based systems, and their applicability to PUC remains unvalidated.

In polyurethane systems, the sensitivity to particle size distribution is even more pronounced. SHIGANG et al. [¹⁷] demonstrated that heterogeneous aggregate shape and poorly graded mix contribute to local stress concentration, microcracking, and inefficient stress transfer through mesoscale modeling. Additionally, JIANG et al. [¹¹] confirmed that increasing the coarse aggregate content up to 60% improved the early and 28-day strengths. However, they did not test the interaction between the fine and coarse aggregate ratio or the effect of grading bandwidth.

Thus, the FA/CA ratio is not merely a geometric proportion but a functional determinant of the composite’s internal microstructure and resin interaction. However, most current studies adopt arbitrary or empirical ratios (e.g., 3:7 or 4:6) without evaluating how variation in this ratio quantitatively affects PUC’s mechanical behavior.

To address this, the present research proposes a systematic investigation of three representative FA/CA ratios: 0.45, 0.54, and 1.00, covering a practical range from coarse-dominant to balanced gradations. These values were selected based on:

• Engineering recommendations for optimal packing and reduced voids [¹⁰];

• Observed trends in previous studies showing peak strengths around balanced FA/CA [⁹];

• The need to explore how finer gradations (e.g., FA/CA = 1.0) affect polymer consumption, adhesion, and strength when the binder is polyurethane, not cement.

This approach aims to fill the gap in current literature by establishing quantitative relationships between aggregate gradation and PUC strength - particularly important for highway repair applications where both early strength and practical workability are essential.

4. PUC MIX DESIGN

4.1. Design requirements

Since polyurethane is a rapid-setting thermosetting system, its mix design method differs from conventional asphalt concrete or cement concrete. Instead of using indices such as slump or Marshall, the optimal polymer binder content is determined based on strength and durability tests.

PUC mix design requires careful consideration of material composition, mix proportions, and construction methods to optimize the mechanical properties and durability of the material. The main objective of mix design is to ensure that PUC meets the requirements for compressive strength and flexural strength. These are the two main criteria for the initial evaluation of PU application for high-grade pavement repair.

4.2. Main material

Polyurethane:

• –

  Type of binder: Usually, polyaryl polymethylene isocyanate (Figure 1a) combined with polyether polyol (Figure 1b) is used.

• –

  Optimum ratio: The ratio between isocyanate and polyol is usually maintained at 1:1 by mass, helping to optimize the reaction rate and mechanical strength of concrete.

• –

  Content in the mixture accounts for 13%

Figure 1
Isocyanate, Polyol and key parameters of the two components.

Aggregates:

• –

  Coarse aggregate: Use granite or dolerite with grain size from 5 mm to 9.5 mm, ensuring high density and durability. The density of sand is 2670 kg/m³, and the modulus of elasticity is 3.5.

• –

  Fine aggregate: Natural sand or artificial sand with grain size from 0.15 to 4.75 to increase the density of concrete. The density of sand is 2650 kg/m³ and the modulus of elasticity is 2.5.

• –

  Mineral additives were added in the composition of 4% fly ash and 1% silica fume. The study by AKKOYUN and AKKOYUN [⁷] indicated reinforcement with fly ash can enhance the tensile strength of the polyurethane composites. The incorporation of FA into rigid PU foams significantly influences the morphological, mechanical, thermal, and electrical properties of the composite. Experimental findings by AKKOYUN and AKKOYUN [⁷] indicate that adding FA up to 20 wt% leads to an increase in average cell size and closed-cell content, resulting in a more uniform foam structure with thicker cell walls. In this study, both fly ash (FA) and silica fume (SF) were incorporated as fine supplementary materials in the polyurethane concrete (PUC) mix design. Beyond their conventional role in filling micro-voids and enhancing the packing density of the aggregate skeleton, these mineral additives contributed significantly to the fresh-state and early-age properties of the PUC system. Specifically, the inclusion of FA and SF improved the mixture’s flowability and cohesiveness by reducing internal friction among aggregates and promoting uniform resin dispersion. This enhanced the workability of the mix within the limited pot life window of polyurethane systems. Furthermore, their pozzolanic activity and high surface area contributed to increased matrix stability, reduced bleeding, and improved interface bonding between resin and aggregate particles. As a result, the modified PUC exhibited better rheological behavior and more consistent early strength development, especially under rapid curing conditions.

• –

  The ratio of sand and stone aggregates was investigated at 3 levels: 0.42; 0.54 and 1

Additives added to PU include mineral additives to make fine fillers such as fly ash (4%), silica fume (1%) and epoxy glue (1.5%). The images of aggregate components are shown in Figure 2 with the detailed mix proportions shown in Table 2.

Figure 2
Composition of aggregates used for PUC.

Table 2
Mix proportions for 1 cubic meter of polyurethane concrete.

COMPONETS	DENSITY (kg/m3)	MASS (kg)	% BY VOLUME	MASS (kg)	% BY VOLUME	MASS (kg)	% BY VOLUME
Polyol	1100	70	6%	70	6%	70	6%
Isocyantes	1240	70	6%	70	6%	70	6%
Fine aggregate (FA)	2650	656.78	25%	773.59	29%	1103.88	42%
Coarse aggregate (CA)	2670	1554.38	58%	1436.67	54%	1103.88	41%
Fly Ash (FA)	2100	80	4%	80	4%	80	4%
Fume Silica (FS)	2200	22	1%	22	1%	22	1%
FA/CA		0.42		0.54		1

4.3. Mixing and specimen casting procedure

Before mixing, the raw materials must be quality-controlled to ensure optimal polymerization reaction.

The experimental procedure is illustrated in Figure 3, with the specific steps detailed as follows:

Figure 3
Mixing procedure for Polyurethane Concrete (PUC) [¹⁸].

1. Preparation of Polyurethane Binder: A two-component polyurethane binder was prepared by thoroughly mixing polyol with isocyanate under continuous stirring until a homogeneous mixture was obtained. This step ensures proper pre-polymerization before contact with aggregates. Use a high-speed mixer (1000–1500 rpm) for 30–60 seconds to activate the polymerization reaction without creating unwanted air bubbles.

2. Drying of Aggregates: Standard sand and gravel were oven-dried using an electric blast drying oven at 105 ± 5 °C to eliminate moisture, which could interfere with the polyurethane curing reaction. Optionally, mineral admixtures such as fly ash and silica fume were blended with the dried sand to enhance packing density and microstructural performance.

3. Formulation of PU-Mortar and PU-Concrete: PU-Mortar was produced by gradually adding the dried fine aggregates (sand + mineral admixtures) into the polyurethane binder and mixing until uniform consistency was achieved. PU-Concrete was then obtained by incorporating dried coarse aggregate (gravel) into the PU-mortar and mixing thoroughly to achieve uniform coating of binder around the aggregate particles.

4. Casting and Specimen Preparation: The resulting mixture was poured into pre-cleaned steel molds to fabricate specimens for both compressive and flexural strength testing. A vibrating table was used to eliminate entrapped air and ensure compaction. Specimens were demolded after initial setting and cured under specified ambient conditions before testing.

5. The curing process of PU concrete is fast, helping to shorten the construction time. PU concrete reaches 80–90% of its maximum strength after 24 hours, depending on environmental conditions. Release mold after 3 hours of curing at room temperature (Temperature = 25 ± 2 °C and Relative humidity ≥ 80%). The polyurethane concrete cures quickly and all specimens were ready to be demolded after 24 h of casting. Polyurethane concrete is capable of bearing loads within 24 hours of curing, as it can achieve up to 80% of its final compressive strength in this period, demonstrating its suitability for rapid repair applications in pavement and structural overlays [¹⁹, ²⁰].

Several images captured during the experimental process are shown in Figure 4.

Figure 4
Experimental process.

The compressive strength test was conducted in accordance with the Vietnam standard TCVN 3118:2022 [²¹], using cubic specimens with a side length of 150 mm. Strain gauges were affixed to the specimen surfaces in both longitudinal and transverse directions to record strain responses. The load was applied uniformly at a constant rate of 0.6 ± 0.2 MPa/s until the specimen reached failure. The flexural test was conducted in accordance with the Vietnam standard TCVN 3119:2022 [²²]. Test specimens had dimensions of 100 mm (width) × 100 mm (height) × 400 mm (length). During the test, the load was applied vertically and uniformly at a constant rate of (0,05 ± 0,01) MPa/s until specimen failure occurred.

5. RESULTS AND DISCUSSION

Workability is a critical property in evaluating the field applicability of polyurethane concrete (PUC), particularly given its rapid setting nature and high reactivity compared to traditional cement-based systems. Unlike ordinary concrete, whose workability is typically assessed through slump or flow table tests, PUC exhibits a fundamentally different rheological behavior due to its thermosetting polymer matrix composed of polyol and isocyanate components. Upon mixing, a rapid polymerization reaction occurs, drastically increasing viscosity within minutes, which significantly limits the pot life of the mixture to approximately 5–10 minutes depending on environmental conditions and material proportions (LI et al. [⁶], WANG et al., 2021 [³]).

The sample casting was conducted by preparing five cylindrical mold samples and five beam mold samples for each aggregate mix ratio. Strength evaluations were performed after 5 hours, and the statistical results are presented in Table 3.

Table 3
Experimental results for selected PUC mix designs.

NO	FA/CA	CS (MPa)	MEAN OF CS (MPa)	STDEV OF CS	FS (MPa)	MEAN OF FS (MPa)	STDEV OF FS
M1	0.42	72.42	72.8	0.28	6.87	6.89	0.14
M1	0.42	72.78			6.91
M1	0.42	73.02			6.82
M1	0.42	73.11			7.12
M1	0.42	72.67			6.75
M2	0.54	73.72	74.13	0.39	8.23	8.5	0.3
M2	0.54	73.86			8.75
M2	0.54	74.29			8.21
M2	0.54	74.72			8.42
M2	0.54	74.08			8.87
M3	1.00	70.42	70.4	0.33	8.03	7.53	0.33
M3	1.00	70.78			7.68
M3	1.00	70.02			7.18
M3	1.00	70.11			7.32
M3	1.00	70.67			7.45

Table 2 indicates that experiments were conducted to evaluate the influence of fine-to-coarse aggregate ratios (FA/CA) on the compressive strength (CS) and flexural strength (FS) of polyurethane concrete (PU). Three FA/CA ratios (0.42, 0.54, and 1.00) were studied, with five test specimens prepared for each group to ensure statistical representativeness. The obtained results demonstrate that the FA/CA ratio significantly impacts both CS and FS, showing clear variations between groups.

To assess the distribution and consistency of flexural strength (FS) and compressive strength (CS) of polyurethane concrete (PU), a statistical probability analysis was conducted. Results from the probability plots indicate that both mechanical parameters tend to follow a normal distribution (Figures 5 and 6).

Figure 5
CS normal distribution chart.

Figure 6
FS normal distribution chart.

The mean value of flexural strength (FS) is 7.641 MPa, with a standard deviation of 0.7266 MPa, while the mean compressive strength (CS) is 72.44 MPa, with a standard deviation of 1.629 MPa. The Anderson-Darling (AD) test results for FS and CS are 0.451 and 0.592, respectively, with corresponding P-values of 0.237 and 0.102. Since both P-values exceed 0.05, there is no statistical evidence to reject the hypothesis that the data follow a normal distribution. This confirms that the obtained data are highly reliable and suitable for accurately evaluating the mechanical properties of the material.

Overall, the PUC in this study exhibited high uniformity, as demonstrated by the conformity of data to the normal distribution and the low variability of the measured values. Flexural strength was significantly lower than compressive strength, which is typical of concrete materials. Therefore, enhancing flexural strength could be considered through the incorporation of reinforcing fibers or adjustments in material composition. With stable mechanical properties validated through statistical analysis, polyurethane concrete shows promising potential for application in heavily loaded structures such as bridges, highways, industrial flooring, or structures demanding high durability.

Table 4 and Table 5 show the influence of aggregate gradation on both compressive strength (CS) and flexural strength (FS) in polyurethane concrete (PUC) was systematically investigated through three representative FA/CA ratios: 0.42, 0.54, and 1.00. The results demonstrate that aggregate proportioning significantly governs mechanical behavior in PUC due to its unique polymer-based matrix, which lacks the internal curing mechanisms of traditional cementitious systems and is highly sensitive to void content, resin distribution, and packing efficiency [⁹, ¹⁰, ¹¹].

Table 4
One-way ANOVA of CS versus FA/CA.

ANALYSIS OF VARIANCE
SOURCE	DF	Adj SS	Adj MS	F-VALUE	P-VALUE
FA/CA	2	35.804	17.9019	156.86	0.000
Error	12	1.370	0.1141
Total	14	37.173
MODEL SUMMARY
S	R-sq	R-sq (adj)	R-sq (pred)
0.337826	96.32%	95.7%	94.24%
MEANS
FA/CA	N	MEAN	StDev	95%Cl
0.42	5	72.800	0.277	(72.471, 73.129)
0.54	5	74.134	0.393	(73.805, 74.463)
1	5	70.400	0.334	(70.071, 70.729)

Table 5
One way ANOVA of FS versus FA/CA.

ANALYSIS OF VARIANCE
SOURCE	DF	Adj SS	Adj MS	F-VALUE	P-VALUE
FA/CA	2	6.5046	3.25229	156.86	0.000
Error	12	0.8863	0.07386
Total	14	7.3909
MODEL SUMMARY
S	R-sq	R-sq (adj)	R-sq (pred)
0.271772	88.01	86.01	81.26
MEANS
FA/CA	N	MEAN	StDev	95%Cl
0.42	5	6.8940	0.1397	(6.6292, 7.1588)
0.54	5	8.496	0.301	(8.231, 8.761)
1	5	7.532	0.334	(7.267, 7.797)

Experimental findings confirmed that the intermediate FA/CA ratio of 0.54 consistently yielded the highest performance, with a mean compressive strength of 74.13 MPa and a mean flexural strength of 8.50 MPa. Statistical analysis via one-way ANOVA showed highly significant effects of FA/CA on both CS (F = 156.86, P < 0.001, R² = 96.32%) and FS (F = 44.03, P < 0.001, R² = 88.01%). The narrow confidence intervals associated with the optimal ratio further indicate high consistency and reproducibility of the results.

At FA/CA = 0.42, although coarse aggregates enhance mechanical interlock, the high content of large particles may cause insufficient resin coverage due to the limited flowability and fast curing nature of polyurethane. This can result in interfacial voids and localized stress concentrations under both compressive and bending loads, as previously modeled in meso-scale simulations by SHIGANG et al. [¹¹]. Meanwhile, the FA/CA ratio of 1.00, characterized by a dominance of fine aggregates, increased the overall surface area that needs resin coating. This not only elevates binder demand but also risks forming entrapped air and weak zones due to incomplete wetting—consistent with the findings of YARDIMCI et al. [¹⁰] in high-FA self-compacting concretes.

Comparatively, JIANG et al. [¹¹] achieved compressive strength up to 73 MPa using a PU-based composite with high coarse aggregate content, although they did not examine FA/CA ratios independently. Our findings extend their work by demonstrating that not only the proportion of total aggregate but the internal balance between FA and CA is critical, particularly when optimizing for both strength parameters simultaneously. However, in PUC, our results suggest a narrower optimum centered around 0.54, likely due to the resin’s rheological constraints and fast-setting kinetics.

In contrast to OPC systems that allow delayed strength development and internal hydration, PUC requires precise control of particle size distribution and resin interaction to achieve mechanical integrity. This is further supported by JUNG et al. [⁴] and WANG et al. [¹⁹], who noted the critical impact of resin-aggregate contact in rapid repair materials. Our study contributes to this understanding by offering a quantified comparison across both CS and FS, linking them directly to aggregate gradation strategies.

In conclusion, the FA/CA ratio of 0.54 emerges as the optimal balance point for structural-grade PUC applications, ensuring sufficient compaction, efficient resin distribution, and enhanced mechanical performance. These insights are particularly valuable for highway maintenance and bridge deck repair scenarios, where early strength, durability, and dimensional stability are essential.

To assess statistically significant differences among the groups, the Tukey HSD (Honestly Significant Difference) test was conducted at a 95% significance level.

Figures 7 and 8 present the Tukey Pairwise Comparison results for CS and FS of PUC, evaluated across three fine-to-coarse aggregate ratios (FA/CA = 0.42, 0.54, and 1.00).

Figure 7
Tukey analysis of differences between FA/CA levels for CS.

Figure 8
Tukey analysis of differences between FA/CA levels for FS.

• –

  Figure 6 indicates that the highest mean compressive strength was observed at an FA/CA ratio of 0.54 (74.134 MPa), followed by FA/CA = 0.42 (72.800 MPa), and the lowest was at FA/CA = 1.00 (70.400 MPa). Groups marked with different letters (A, B, C) indicate statistically significant differences between FA/CA levels.

• –

  Figure 7 illustrates that flexural strength was also highest at FA/CA = 0.54 (8.496 MPa), followed by FA/CA = 1.00 (7.532 MPa), and lowest at FA/CA = 0.42 (6.894 MPa).

In Tukey Simultaneous 95% confidence interval graphs, pairs whose confidence intervals do not include zero are considered to have significant differences. The results indicate that differences between the FA/CA ratio of 0.54 and the other groups are statistically significant, whereas differences between FA/CA ratios of 0.42 and 1.00 may not be significant for certain indicators.

5.1. Effect of FA/CA ratio on compressive and flexural strength of PUC

Figures 9 and 10 illustrate the influence of the fine-to-coarse aggregate ratio on the flexural and compressive strengths of polyurethane concrete.

Figure 9
Interval plot of CS (MPa) vs FA/CA.

Figure 10
Interval plot of FS (MPa) vs FA/CA.

Aggregate gradation significantly influences the mechanical behavior of polyurethane concrete (PUC), especially in fast-setting systems where no internal curing occurs. As shown in Figure 8, compressive strength increased from 72.80 MPa to 74.13 MPa when the FA/CA ratio rose from 0.42 to 0.54, then decreased to 70.40 MPa at FA/CA = 1.00. This trend indicates that a moderate increase in fine aggregates enhances packing density and interfacial bonding, while excessive fine content leads to increased surface area demand and entrapped air voids, thereby reducing strength. These observations align with the findings of JIANG et al. [¹¹], who reported strength improvement in PUC when coarse aggregates were appropriately balanced, though they did not investigate FA/CA variations directly.

Flexural strength, a key performance parameter under tensile stress, followed a similar trend. As shown in Figure 9, FS improved from 6.89 MPa at FA/CA = 0.42 to a peak of 8.50 MPa at FA/CA = 0.54, before declining to 7.53 MPa at FA/CA = 1.00. The optimal performance at 0.54 is attributed to a well-graded mix that minimizes voids and maximizes resin coverage efficiency. These results are in agreement with Yardimci et al. [¹⁰], who found similar behavior in self-compacting concretes with balanced FA/CA ratios. In contrast, at FA/CA = 1.00, the high proportion of fines likely caused excessive resin demand and reduced cohesion, as also noted by JUNG et al. [⁴].

While FA/CA = 0.42 provided higher mix stability (with the lowest standard deviations in both CS and FS), it did not yield the highest strengths. This reflects a trade-off between mechanical performance and production consistency. This finding further reinforces Khan’s observation [⁹]. Moreover, meso-scale modeling by SHIGANG et al. [¹⁷] confirmed that coarse-aggregate-dominant mixes increase local stress concentrations, which can initiate early cracking if not properly bonded by the resin.

In conclusion, the present study confirms that an FA/CA ratio of 0.54 offers the most favorable balance between compressive and flexural strength in polyurethane concrete. This can be attributed to improved particle packing and more efficient resin coating of the aggregate matrix, which is particularly critical in PU systems where rapid curing and absence of internal hydration require precise aggregate gradation.

These findings align with JIANG et al. [¹¹] and YARDIMCI et al. [¹⁰], and further expand on them by demonstrating that not only total aggregate content but the fine-to-coarse proportion critically governs mechanical behavior in PUC. Future studies should consider extending the FA/CA range and incorporating microstructural analysis to better understand failure modes and interfacial bonding in such systems.

5.2. Regression-based analysis of aggregate ratio influence

To further quantify the relationship between the fine-to-coarse aggregate ratio (FA/CA) and the mechanical properties of polyurethane concrete (PUC), second-order regression models were developed for both compressive strength (CS) and flexural strength (FS). The results confirm that FA/CA exhibits a nonlinear (quadratic) influence on both performance metrics.

The regression equation for compressive strength is given by:

(1) $$\text{CS(Mpa)=60}\text{.61+42}\text{.95}\cdot \text{FA/CA-33}\text{.16}\cdot {\text{(FA/CA)}}^2$$

This model achieved a high coefficient of determination (R² = 96.32%), with a predictive R² of 94.24%, indicating excellent fit and strong forecasting capability. The positive linear term reflects the initial strength gain with increasing FA/CA, attributed to improved packing density and reduced voids. However, the significant negative quadratic term (P < 0.001) confirms that excessive fine aggregate leads to over-coating, higher resin demand, and reduced internal cohesion, causing a strength decline beyond an optimal range.

Similarly, the flexural strength model:

(2) $$\text{FS (MPa) = -4}\text{.75 + 38}\text{.92 · FA/CA-26}{\text{.63·(FA/CA)}}^2$$

yields R² = 88.01% and predicted R² = 81.26%. Although slightly lower than for compressive strength, these values demonstrate a reliable model fit. The peak FS observed near FA/CA = 0.54 corresponds well with experimental data, reinforcing that this ratio offers the best balance between matrix integrity and interfacial bonding efficiency.

Both models consistently indicate that an intermediate FA/CA ratio around 0.50 to 0.55 is optimal. Ratios lower than this may result in poor interlock and resin segregation due to excess coarse particles, while higher ratios increase surface area and weaken stress transfer across the aggregate-resin interface. These trends align with previous studies by JIANG et al. [¹¹], YARDIMCI et al. [¹⁰], and IQBAL KHAN et al. [⁹], and further validate the critical role of aggregate gradation in PUC mix optimization.

The use of quadratic regression also allows for predictive application in mix design, enabling engineers to estimate expected strength outcomes and reduce trial-and-error in laboratory testing. Future work may extend these models by integrating resin properties, curing temperature, or moisture conditions as additional variables to enhance their robustness and practical applicability.

5.3. Multi-response optimization of aggregate ratio for polyurethane concrete

To identify the optimal fine-to-coarse aggregate ratio (FA/CA) that simultaneously maximizes both compressive strength (CS) and flexural strength (FS) of polyurethane concrete (PUC), a multi-response desirability analysis was performed in Figure 11. The optimization was based on quadratic regression models derived from experimental data and targeted both strength parameters within their observed ranges.

Figure 11
Multi-response optimization of aggregate ratio for polyurethane concrete.

The results reveal that a FA/CA ratio of 0.648 yields the highest combined performance, with predicted values of 9.28 MPa for FS and 74.52 MPa for CS. The composite desirability score reached 0.978, indicating near-ideal optimization for both responses. This finding confirms that an intermediate FA/CA ratio—neither coarse-aggregate-dominant nor overly fine-rich—offers the best mechanical synergy in PUC. The 95% confidence and prediction intervals further validate the robustness of this solution.

This outcome is consistent with the nonlinear trends observed in the regression models, where both CS and FS increased with FA/CA up to a peak and then declined. Unlike conventional cement-based concrete, PUC systems are particularly sensitive to aggregate gradation due to the absence of hydration bonding and the high viscosity of the polyurethane resin. Excessive fine aggregates increase surface area demand and hinder resin flow, while too many coarse aggregates compromise matrix continuity and interfacial adhesion.

When compared to prior studies, this optimal FA/CA ratio for PUC (~0.65) is lower than those reported for OPC or SCC systems. For example, YARDIMCI et al. [¹⁰] and IQBAL KHAN et al. [⁹] reported optimal FA/CA values ranging from 0.9 to 1.2 in self-compacting and steel-fiber-reinforced concretes. These systems benefit from cement hydration and excess paste content, which compensate for the surface area introduced by fine aggregates. In contrast, JIANG et al. [¹¹] and SHIGANG et al. [¹⁷] emphasized the role of particle distribution in early strength development and failure initiation in PU systems, suggesting that PUC performance is more susceptible to gradation imbalance.

Therefore, the FA/CA value of approximately 0.648 identified in this study provides a statistically validated design parameter for high-performance PUC applications. It ensures enhanced resin packing, improved aggregate bonding, and optimized strength characteristics under the rapid-setting constraints of polyurethane binders. This finding also serves as a quantitative basis for engineers aiming to refine mix designs with minimal trial-and-error, especially in time-critical repair scenarios.

5.4. Microstructural interaction between polyurethane and aggregates

Figure 12 presents the microstructural features of polyurethane concrete (PUC), highlighting the interaction between coarse aggregates and polyurethane mortar. The images show partially coated aggregates and areas of visible interfacial debonding between the binder and aggregate particles. This phenomenon reflects a fundamental difference between PUC and traditional cementitious materials: unlike cement paste, polyurethane does not hydrate and lacks self-healing capacity at the interface. As a result, any initial imperfection or mismatch in thermal or mechanical compatibility can lead to persistent defects in the hardened matrix [¹⁷].

Figure 12
Interaction in PUC components.

Observations indicate that interfacial debonding may result from several factors, including surface moisture on aggregates during mixing, high viscosity of the resin hindering complete wetting, and differential shrinkage during early curing. These are especially critical in PUC systems, where the setting time is short and resin mobility is limited. According to JIANG et al. [¹¹] and SHIGANG et al. [¹⁷], localized microcracks may be caused by early-phase stress concentrations brought on by an uneven aggregate distribution or inadequate bonding, especially when heat loading or dynamic stress is present.

Cracks observed within the polyurethane mortar matrix in Figure 10 also suggest limited tensile resistance of the cured resin. Jung et al. [⁴] reported that polyurethane composites experienced a significant reduction - up to 66% – in tensile strength under elevated temperatures, underscoring the vulnerability of the polymer phase in harsh service conditions. These microcracks may propagate rapidly under flexural loading, especially if aggregate-matrix bonding is insufficient or non-uniform.

Importantly, the experimental results discussed earlier support these microstructural findings. At FA/CA = 1.00, both compressive and flexural strength declined significantly, coinciding with greater standard deviation in FS values and likely reflecting less effective resin distribution around fine aggregates. In contrast, the optimal performance at FA/CA = 0.54 corresponds with denser packing, better interfacial coverage, and minimal visible debonding, indicating more effective stress transfer through the matrix.

To mitigate interfacial weakness, several strategies can be adopted. First, pre-treatment of aggregates (e.g., drying, surface activation) may enhance resin wetting and bond strength. Second, incorporation of micro-or steel fibers, as demonstrated by AKKOYUN et al. [⁷], can improve crack resistance and ductility by bridging interfacial voids. Additionally, the application of coupling agents or surface modifiers—such as silane-based primers—can chemically enhance bonding between the polyurethane resin and siliceous aggregate surfaces, improving interfacial shear transfer.

Overall, understanding the microstructural interaction between aggregate and polyurethane is essential for designing durable PUC mixtures. These findings highlight the critical role of both mix design (e.g., FA/CA ratio) and interfacial engineering in optimizing the performance of polyurethane concrete for high-demand applications such as bridge deck overlays and rapid pavement repair.

It is acknowledged that the present study does not include Scanning Electron Microscopy (SEM) or high-resolution micrographs due to equipment limitations. The observations presented in Figure 10 are based on macro- and optical inspection, and are intended as qualitative indicators of interfacial bonding behavior. While the interpretations rely partially on established theoretical insights and previously published SEM-supported studies [¹¹, ¹⁷]. They are also supported by the experimentally observed trends in compressive and flexural strength at different FA/CA ratios.

Furthermore, this study does not quantitatively assess crack density, orientation, or fracture mechanics parameters such as crack mouth opening displacement (CMOD) or fracture energy. These limitations are acknowledged, and future work is recommended to incorporate advanced image-based analyses, such as X-ray CT or SEM imaging, to characterize interfacial zones in greater detail. In addition, controlled fracture tests, including three-point bending with digital image correlation (DIC), could offer deeper insights into crack propagation mechanisms in polyurethane concrete systems. Such approaches would enhance the mechanistic understanding of failure behavior and support more robust design of PUC for structural and repair applications.

6. CONCLUSION

This study explored the influence of the fine-to-coarse aggregate ratio (FA/CA) on the mechanical performance of polyurethane concrete (PUC), with a focus on compressive strength (CS) and flexural strength (FS). Through a combination of regression modeling, statistical validation, and multi-response optimization, it was demonstrated that aggregate gradation significantly affects both strength parameters. An intermediate FA/CA ratio of approximately 0.648 was identified as optimal, yielding the highest predicted values for CS and FS, supported by a composite desirability score of 0.978.

The study confirmed that the mechanical behavior of PUC follows a nonlinear response to FA/CA, due to the unique interaction between polyurethane resin and aggregates. The absence of hydration and rapid setting characteristics of PU binders demand precise control of gradation and interface conditions, especially under ambient curing. The regression models and optimization framework presented herein may serve as practical tools for engineers seeking performance-driven mix designs in time-sensitive infrastructure repair scenarios.

However, several limitations of the current study must be acknowledged. First, the experimental program was limited to early-age strength under laboratory-controlled conditions, without investigation of long-term durability such as freeze–thaw resistance, chemical attack, or dimensional stability. Second, while macro-level observations of resin–aggregate interaction were provided, detailed microstructural characterization (e.g., SEM, X-ray CT) and fracture analysis were not performed. Third, workability behavior was qualitatively assessed, and no rheological or flowability tests were included. Furthermore, the sample size per group was limited to five replicates, and all materials were sourced from a single batch, which may restrict the generalizability of the findings.

To address these limitations, future research will be expanded to include:

• (1)

  Durability testing under aggressive environmental conditions;

• (2)

  Microstructural and fracture-mechanics-based evaluation;

• (3)

  Quantitative workability and polymerization behavior;

• (4)

  Scale-up testing under field mixing and casting scenarios;

• (5)

  Comparative benchmarking with cement-based and HPC systems;

• (6)

  Integration of fiber reinforcement and recycled aggregates to enhance ductility and sustainability.

Despite its constraints, this study contributes a statistically grounded, experimentally validated foundation for understanding the role of aggregate gradation in PUC design. The insights gained may inform both academic investigations and field applications involving fast-setting polymer concretes for structural repair and rehabilitation.

7. BIBLIOGRAPHY

• ^[1] NGUYEN, N.T.T., NGO, T.V., NGUYEN, K.K., et al., “Effects of fly ash and graphene oxide in cement mortar considering the local recycled material context”, Applied Sciences, v. 14, n. 14, pp. 6140, 2024. doi: http://doi.org/10.3390/app14146140.
  » https://doi.org/10.3390/app14146140
• ^[2] NGO, T.V.; BA, V.T.; LE, B.H., et al., “Assessment of the impact of local-recycle cementitious replacement materials on the strength of the Ultra-High-Performance Concrete”, Applied Sciences, v. 14, n. 17, pp. 7484, 2024. doi: http://doi.org/10.3390/app14177484.
  » https://doi.org/10.3390/app14177484
• ^[3] WANG, H., HAI, X.L., CAI, H.Z., “Study on the polyurethane concrete for the rapid repairment of highway pavement”, Applied Mechanics and Materials, v. 193-194, pp. 762-769, 2012. doi: http://doi.org/10.4028/www.scientific.net/AMM.193–194.762.
  » https://doi.org/10.4028/www.scientific.net/AMM.193–194.762
• ^[4] JUNG, K.-C., ROH, I.-T., CHANG, S.-H., “Evaluation of mechanical properties of polymer concretes for the rapid repair of runways”, Composites. Part B, Engineering, v. 58, pp. 352–360, 2014. doi: http://doi.org/10.1016/j.compositesb.2013.10.076.
  » https://doi.org/10.1016/j.compositesb.2013.10.076
• ^[5] HWANG, S.D., KHAYAT, K.H., BONNEAU, O., “Performance-based specifications of self-consolidating concrete used in structural applications”, ACI Materials Journal, v. 103, n. 2, pp. 121–129, 2006. Available: https://www.researchgate.net/publication/268804229_Performance-based_specifications_of_self-consolidating_concrete_used_in_structural_applications, accessed in July, 2025.
• ^[6] LI, B., LIU, H., JIAN, J., et al., “Experimental study on flexural properties of polyurethane–cement composites under temperature load”, Applied Sciences (Basel, Switzerland), v. 12, n. 24, pp. 12799, 2022. http://doi.org/10.3390/app122412799.
• ^[7] AKKOYUN, M., AKKOYUN, S., “Blast furnace slag or fly ash filled rigid polyurethane composite foams: a comprehensive investigation”, Journal of Applied Polymer Science, v. 136, n. 20, pp. 47433, 2019. doi: http://doi.org/10.1002/app.47433.
  » https://doi.org/10.1002/app.47433
• ^[8] DOS REIS, J.M.L., JURUMENHA, M.A.G. “Experimental investigation on the effects of recycled aggregate on fracture behavior of polymer concrete”, Materials Research, v. 14, n. 3, pp. 326–330, 2011. doi: https://doi.org/10.1590/S1516-14392011005000060.
  » https://doi.org/10.1590/S1516-14392011005000060
• ^[9] IQBAL KHAN, M., ABBASS, W., ALRUBAIDI, M., et al., “Optimization of the fine to coarse aggregate ratio for the workability and mechanical properties of high strength steel fiber reinforced concretes”, Materials (Basel), v. 13, n. 22, pp. 5202, 2020. doi: http://doi.org/10.3390/ma13225202. PubMed PMID: 33213071.
  » https://doi.org/10.3390/ma13225202
• ^[10] YARDIMCI, M.Y., BARADAN, B., TAŞDEMIİR, M.A., “Effect of fine to coarse aggregate ratio on the rheology and fracture energy of steel fibre reinforced self-compacting concretes”, Sadhana, v. 39, n. 6, pp. 1447–1469, 2014. doi: http://doi.org/10.1007/s12046-014-0257-2.
  » https://doi.org/10.1007/s12046-014-0257-2
• ^[11] JIANG, Z., TANG, C., YANG, J., et al., “A lab study to develop polyurethane concrete for bridge deck pavement”, The International Journal of Pavement Engineering, v. 23, n. 5, pp. 1404–1412, 2020. doi: http://doi.org/10.1080/10298436.2020.1804063.
  » https://doi.org/10.1080/10298436.2020.1804063
• ^[12] ZHENG, X., CHEN, S., WANG, Y.,“Analysis of mechanical properties of polyurethane concrete and its bond slip characteristics with rebar”, Revista Matéria (Rio de Janeiro), v 30, pp. e20240712, 2025, doi: https://doi.org/10.1590/1517-7076-rmat-2024-0712.
  » https://doi.org/10.1590/1517-7076-rmat-2024-0712
• ^[13] ZHONG, R., PAN, M., WU, H., et al., “Effect of coarse aggregate on the stability and mechanical performance of ultra-high performance concrete (UHPC)”, Composites. Part B, Engineering, v. 297, pp. 112210, 2025. doi: http://doi.org/10.1016/j.compositesb.2025.112210.
  » https://doi.org/10.1016/j.compositesb.2025.112210
• ^[14] ALLUJAMI, H.M., ABDULKAREEM, M., JASSAM, T.M., et al., “Mechanical properties of concrete containing recycle concrete aggregates and multi-walled carbon nanotubes under static and dynamic stresses,” Case Studies in Construction Materials, v. 17, pp. e01651, 2022. doi: http://doi.org/10.1016/j.cscm.2022.e01651.
  » https://doi.org/10.1016/j.cscm.2022.e01651
• ^[15] DHARMARAJ, S.R. et al. “Advances in lightweight concrete: balancing strength and workability”,Revista Matéria (Rio de Janeiro), v. 30, pp. e20240872, 2025. doi: https://doi.org/10.1590/1517-7076-RMAT-2024-0872.
  » https://doi.org/10.1590/1517-7076-RMAT-2024-0872
• ^[16] SAFIUDDIN, MD., ALENGARAM, U.J., SALAM, Md. A., et al., “Properties of high-workability concrete with recycled concrete aggregate”,Revista Matéria (Rio de Janeiro), v. 14, n. 2, pp. 248–255, 2011. doi: https://doi.org/10.1590/S1516-14392011005000039.
  » https://doi.org/10.1590/S1516-14392011005000039
• ^[17] SHIGANG, A., LIQUN, T., YIQI, M., et al., “Effect of aggregate distribution and shape on failure behavior of polyurethane polymer concrete under tension”, Computational Materials Science, v. 67, pp. 133–139, 2013. doi: http://doi.org/10.1016/j.commatsci.2012.08.029.
  » https://doi.org/10.1016/j.commatsci.2012.08.029
• ^[18] GAO, H., LIU, L., YANG, W., et al., “Experimental study on mechanical properties of polyurethane-based composites,” Case Studies in Construction Materials, v. 21, pp. e03907, 2024. doi: http://doi.org/10.1016/j.cscm.2024.e03907.
  » https://doi.org/10.1016/j.cscm.2024.e03907
• ^[19] WANG, H., HAI, X.L., CAI, H.Z., “Study on the polyurethane concrete for the rapid repairment of highway pavement”, Applied Mechanics and Materials, v. 193-194, pp. 762–769, 2012. doi: http://doi.org/10.4028/www.scientific.net/AMM.193-194.762.
  » https://doi.org/10.4028/www.scientific.net/AMM.193-194.762
• ^[20] HUSSAIN, H.K., GUI, W.L., YU, W.Y., “Experimental study to investigate mechanical properties of new material polyurethane–cement composite (PUC)”, Construction & Building Materials, v. 50, pp. 200–208, 2014. doi: http://doi.org/10.1016/j.conbuildmat.2013.09.035.
  » https://doi.org/10.1016/j.conbuildmat.2013.09.035
• ^[21] LUAT VIETNAM, Bê tông - Phương pháp xác định cường độ chịu nén [TCVN 3118:2022, Hardened concrete – Test method for compressive strength], Hanoi City, LuatVietnam, 2022.
• ^[22] LUAT VIETNAM, Tiêu chuẩn TCVN 3119:2022 Bê tông - Phương pháp xác định cường độ chịu kéo khi uốn [Standard TCVN 3119:2022 Concrete - Method for determining tensile strength when bending], Hanoi City, LuatVietnam, 2022. https://luatvietnam.vn/xay-dung/tieu-chuan-tcvn-3119-2022-be-tong-phuong-phap-xac-dinh-cuong-do-chiu-keo-khi-uon-253917-d3.html, accessed in July, 2025.
  » https://luatvietnam.vn/xay-dung/tieu-chuan-tcvn-3119-2022-be-tong-phuong-phap-xac-dinh-cuong-do-chiu-keo-khi-uon-253917-d3.html

Publication Dates

• Publication in this collection
  04 Aug 2025
• Date of issue
  2025

History

• Received
  24 Mar 2025
• Accepted
  24 June 2025