Optimization and performance of expanded polystyrene concrete for sustainable infill wall construction using response surface methodology

Abstract

The increasing superimposed dead load (SDL) from conventional brick or concrete block infill walls of frame structures necessitate the development of lightweight and sustainable alternatives. During earthquakes, these heavy infill walls attract significant seismic forces due to their mass. This study investigates the use of expanded polystyrene (EPS) concrete panels as a lightweight alternative to traditional infill walls. EPS panels with densities of 1000–1100 kg/m³, using both 2.5% small size (0.71–1 mm) and 2% large size (1–1.66 mm) EPS beads, were developed and evaluated. Mix designs were optimized using statistical modeling techniques, including Analysis of Variance (ANOVA) and Response Surface Methodology (RSM), confirming high model accuracy (R² > 0.80). Experimental tests assessed various properties of EPS concrete panels and shown a 47% reduction in thermal conductivity and 60–70% lower permeability compared to bricks. Water absorption was 50% lower, and despite their 43–48% reduced density, the panels achieved 75% of the compressive strength of conventional materials. Notably, diagonal force stability was approximately 30% greater than brick masonry. These findings suggest EPS concrete panels significantly reduce the non-structural dead load with improved thermal conductivity, water absorption, diagonal shear resistance and other properties, leading to more economical, sustainable construction, thus offering a durable and heat-resistant alternative to traditional infill wall materials.

Introduction

Structural loads are a primary consideration in design of any building because these define the nature and magnitude of external forces or hazards that a building must resist to provide expected performance in terms of safety and serviceability¹. Typically, building structures are designed to withstand effects of both vertical and lateral loads. In addition to self-weight, building structures are subjected to loads from permanent partition walls, fixed permanent fixtures/equipment, floor surfacing materials, roof insulation/water proofing and other finishing materials, all termed as non-structural components². The partition walls are one of the major elements among these that interact with the structural system and during earthquakes, these heavy infill walls due to their greater mass attract larger seismic force. They can mainly be constructed by using unreinforced clay bricks and concrete blocks, reinforced concrete masonry units or drywall systems³. Rapid growth in the construction industry of developing countries, where clay brick/ PCC block remains as one of the essential building materials, has seen advancement in these infill systems⁴. Research studies have shown that the costs related to the failure of nonstructural components like infills in a building may easily exceed the replacement cost of the whole building⁵. The cost of non-structural component losses accounts for approximately 62% in offices, 70% in hotels, and 48% in hospitals⁶. By examining these numbers, it is important that prevention of damage to non-structural components is an important concern for the seismic performance of new buildings. The stiff masonry infill walls restrict the horizontal movement of the RCC frame especially the load bearing columns^7–10. The formation of compression strut in infill masonry results in formation of a short column behavior and get deformed by the full amount over the short height adjacent to the window opening^11,12. On the other hand, regular columns get deformed over the full height. Since the effective height over which a short column can freely bend is small, it offers more resistance to horizontal motion and thereby attracts a larger force as compared to the regular column. Resultantly, the short column sustains more damage¹³.

Researchers have carried out work on developing light weight concrete as replacement of heavy infill walls using different materials and methods. These include EPS particles-based concrete sandwiched in insulating panels, Dry wall technology, Cellular Light Weight Concrete using Glass Fiber^14,15. EPS, also known as expanded polystyrol, is a stable polymeric foam made from polystyrene, characterized by its ultra-low density and discrete air voids embedded within the polymer matrix¹⁶. Cement-based materials incorporating EPS beads are recognized as sustainable options that contribute to environmentally friendly benefits. The cement-based materials containing EPS in various construction applications such as cladding panel, curtain wall, composite flooring system, structural insulated panel (SIP) and composite SIP, load-bearing concrete block, subbase material for a pavement, geogrid, road barrier, and floating marine structure”^17,18.

J. Alexandre Bogas et al.¹⁹. in his experimental work on non-structural lightweight concrete (NSLWC) with volcanic scoria aggregates for lightweight fill in building’s floors analyzed and compared with other conventional solutions with expanded clay aggregates and EPS. His study concluded that compressive strength of NSLWC with natural scoria aggregates was comparable to that of normal weight concrete. The load-deflection relationship is meaningful to reflect the flexural bearing capacity of Improved Composite Structural Insulated Panels (ICSIPs). The average of the results of the five test pieces was calculated to evaluate the mechanical properties of the ICSIPs. Load arrives the peak value at 27.7 KN with a midspan deflection of 8.8 mm. Debonding between recycled aggregate concrete face sheet and EPS core appears. Subsequently, the deflection keeps increasing but the load decreases. It can be observed that the concrete is crushed in this stage. And the ICSIPs reach the ultimate bearing capacity^20,21. The elastic modulus of EPS concrete exhibited an almost linear decrease with the increasing EPS volume fraction. In the field of inorganic non-metallic materials, the elastic modulus E of ceramic materials is estimated as²² given in Eq. (1).

1

Gacia et al. (2024) reported that increasing the EPS bead content in concrete can reduce thermal conductivity by up to 70%. Additionally, sandwich wall panels incorporating EPS beads for thermal insulation exhibited superior mechanical performance compared to those using EPS sheets^7,23–26. Polystyrene aggregate concrete (PAC) exhibited greater drying shrinkage strain as the proportion of polystyrene aggregate increased. Although early-age shrinkage was more pronounced at higher polystyrene content, the differences in drying shrinkage between PAC and control specimens became negligible at later ages^27,28.

EPS, a lightweight artificial aggregate, is commercially available and can be integrated into mortar or concrete to create lightweight insulating concrete. From an engineering perspective, a key advantage of EPS over other lightweight aggregates (LWAs) is its lower water absorption, attributed to its reduced porosity^27,28. The results of absorption tests indicated that as EPS aggregates content is increased, the water absorption percent is increased²⁹. Generally, the permeation of aggressive substances through concrete pores is a key factor influencing various durability properties (Saikia and de Brito, 2012). Ganesh Babu and Saradhi Babu (2004) examined the durability properties of lightweight concrete (LWC) incorporating EPS aggregate in varying proportions from 0 to 95%, with LWC density ranging between 550 and 2200 kg/m³. The permeability of EPS concrete with higher strength and density (e.g., 1500 kg/m³) was reported similar to that of normal weight concrete NWC^18,30.

Sikora and Turkiewicz observed that the sound absorption properties of EPS beads contributed to the reduction of narrowband frequencies within the 500 Hz–20,000 Hz range. Santos et al. (2021) incorporated EPS concrete with a density of 1100 kg/m³ as a subfloor component in social housing projects, identifying it as a cost-effective solution. Additionally, hardened EPS concrete demonstrated significantly enhanced noise reduction compared to conventional concrete, making it a promising material for acoustic barriers in civil engineering applications¹⁹. The walls made of sandwich panels with the EPS core, the EPS core is fully consumed (thermally decomposed) during first 5–10 min of fire conditions. The walls made of sandwich panels with EPS core with the thickness of minimum 100 mm in most cases are classified not less than E 30³¹. A testing program was conducted to assess the in-plane performance of unreinforced masonry (URM) wallettes retrofitted with textile-reinforced mortar (TRM), with a particular focus on the diagonal shear cracking failure mode. The shear strength of TRM-retrofitted wallettes with a two-sided application increased by 446–481% compared to the corresponding as-built wallettes (ABW)^32–35.

Although lightweight concrete is frequently used as an infill material to reduce the superimposed dead load of structures, the potential of Expanded Polystyrene (EPS) concrete panels as sustainable lightweight infill replacement remains underexplored. Furthermore, most existing studies fail to optimize and comprehensively evaluate the interactions between EPS bead sizes and cementitious materials, and the resulting properties of EPS concrete panels compared to conventional infill materials. This study addresses these limitations by statistically optimizing through RSM (Response Surface Methodology) and then experimentally investigating the optimized composition of EPS concrete panels with different bead sizes, fine sand, fly ash, superplasticizer and panel sheet, focusing on achieving superior performance in the category of lightweight infill material.

The primary objective of this research is to first develop an optimized expanded polystyrene (EPS) concrete panel with a target density in the range of 1000–1100 kg/m³ and compressive strength of 4–6 MPa, incorporating different EPS bead sizes and percentages, and then investigate its properties to evaluate the physical, mechanical, functional, and structural characteristics through comprehensive experimental analysis. The performance of the optimized EPS panel is also compared against conventional infill materials to validate its potential as a superior, sustainable alternative. By demonstrating a balance between density, strength, durability, and resource efficiency, this study highlights the significant advantages of EPS concrete panel as a lightweight sustainable material over traditional infill materials, including enhanced resource efficiency and reduced environmental impact. The research provides experimentally validated insights into the enhanced performance of EPS concrete panels, offering a promising solution for lightweight non-structural construction applications, contributing to more sustainable and resource-efficient building practices. The novelty of this research lies in its comprehensive approach in optimizing EPS concrete composition through RSM, integrating EPS beads and Plasticizer to achieve optimized mix design, and providing a statistically validated model for EPS concrete mix design improved performance. This work offers new insights into the sustainable use of EPS beads as aggregates along with plasticizer for EPS concrete as infill material in frame structures.

Materials and methodology

This section discusses the materials that were used in the research work and the different experimental procedures and methods applied to it.

Material

Cement

The Ordinary Portland cement (OPC) grade-53 with 3140 kg/m³ density, was used as binder in this study. The cement complies with Standard Specification for Portland cement³⁶. The chemical composition and physical properties of the cement used are listed in Table 1.

Table 1.

Physical properties of cement.

Items	Properties
Initial setting (minutes)	14637
Final setting (minutes)	23237
Specific gravity	3.14
Consistency (%)	2838
Average particle diameter (µm)	< 3139
28 days compressive strength (MPa)	45.7036
Blaine fineness (m2/kg)	301

Fine aggregates

This research employed sand with a fineness modulus of 2.0 and an absorption capacity of 1.15%. The gradation curve for the fine aggregate is presented in Fig. 1, which shows use of very fine sand for casting of lightweight concrete samples. The sand sample, classified as very fine, met the ASCE lower limits for retention across various sieve sizes. The gradation curve shows a maximum retention of 90–100% for sieve sizes ranging between 0.04 and 0.2 mm, while the retention decreased to 20 − 0% for sieve sizes between 2 and 4 mm.

Fig. 1.

Gradation curve of fine aggregate, ASTMC33/C33M⁴⁰.

Fly ash

Class F, or low-calcium fly ash having less than 10% CaO, was used during the research. Class F ash consists primarily of an alumino-silicate glass (SiO), with quartz and mullite also present in wide range of particle sizes, as shown in the X-ray Diffraction (XRD) spectrum of the used sample in Fig. 2.

Fig. 2.

Chemical composition analysis of fly ash.

The main reason to use Class F was to maximize cement replacement with fly ash by weight to economize the cost, as early strength was not desired, for which Class C fly ash is recommended. Moreover, Class F fly ash reduces permeability very effectively, which is of paramount importance in the case of EPS concrete⁴¹.

Chemical admixture

To improve the workability of the EPS concrete mix, superplasticizer was used. Detailed properties of the admixture used are shown in the Table 2 below.

Table 2.

Chemical properties of super plasticizer.

Composition	Polycarboxylic polyether type polymer based on the latest flake technology (3rd generation)
Appearance / Color	Turbid
pH-Value	4.0–6.0 (at 20 °C)
Dosage	0.5–1.1% by weight of binder
Density	~ 1.05 ± 0.02 kg/L

Panel sheet

For EPS Panels, 6 mm thick cement boards were used to improve heat resistance, water resistance and to provide reinforcement to concrete matrix as a confinement. The properties of the cement board sheet are given in Table 3.

Table 3.

Properties of cement board sheet.

Items	Properties
Water Expansion Rate (24 h soak)	0.12%
Density (kg/m3)	1260
Drying Shrinkage rate	0.04%
Heat Conductivity rate (K)	0.064 W/ MC
Tension Force perpendicular to the surface	1 MPa
Sound Resistance	36 dB

Expanded polystyrene (EPS) beads

Two bead sizes, large and small, with bulk densities of 12–16 g/L and 13–20 g/L, respectively, as shown in Fig. 3 were used. The physical properties of beads are listed in Table 4. This approach was adopted with the idea that different sizes of EPS beads, used in varying ratios, affect the physical, mechanical, structural, and functional properties of EPS concrete panels.

Fig. 3.

(a) Large size EPS beads (EPS LB). (b) Small size EPS beads (EPS SB).

Table 4.

Physical properties of different sized EPS beads (aggregates).

Aggregate type	Specific gravity	Water absorption (%)	Bulk density (kg/m3)	Fineness modulus	Maximum size aggregate
EPS Beads (L)	2.49	0.5	12–16	-	1–1.66 mm
EPS Beads (S)	2.40	0.5	13–20	-	0.71–1 mm

Chemical adhesive

Sika Dur 32 was used as an adhesive with Cement Board Sheet, and the properties are presented in Table 5.

Table 5.

Properties of Sika dur 32.

Items	Properties
Composition	Epoxy Resin
Packaging	5 kg (A + B)
Colour component	A: White Component B: Dark grey Components A + B: Concrete grey
Density	~ 1.4 kg/l (+ 23 °C) (mixed component A + B)
Compressive strength	2 N/mm2 ~ 56 N/mm2 (Depending on Temperature and Curing Day)
Layer thickness	~ 1 mm max

Methodology

The present study deals with the optimization and performance evaluation of EPS concrete panels, containing varying sizes of EPS beads. Figure 4 shows a Schematic diagram of research methodology. The study was carried out in two phases. In the first phase, multiple trials were conducted to determine the optimal values of density and compressive strength basing on the optimal proportions of large-sized and small-sized EPS beads and other constituents to achieve EPS concrete mix with lower density and maximum compressive strength following the ACI mixture design guidelines⁴².

Fig. 4.

Schematic diagram of research methodology.

Response Surface Methodology (RSM), a statistical technique, frequently applied to design, analysis, optimization, and validation of experimental data was also used to validate the optimized mix design, during the phase one. The concrete industry has made extensive use of RSM for experiment design, statistical model development, and correlation analysis between causes (independent variables) and responses (dependent variables). For this, Design Expert 13.0 software⁴³ was utilized. To evaluate the understanding of the model’s performance and optimization of results, the ANOVA (Analysis of Variance) approach inside RSM was adopted⁴⁴. In the second phase, standard EPS concrete mixtures introduced into separate batches were prepared to evaluate their contributions to the physical, mechanical, functional, and structural properties of the Expanded Polystyrene (EPS) concrete panels. Test results of EPS concrete panel were also compared with conventional infill materials.

Mix design optimization, experimental program and test setup

Trial mix

To achieve lightweight material, the density of the EPS concrete panels was set between 1000 and 1100 kg/m³, with a minimum compressive strength of 4–6 MPa. Due to the lightweight nature of the concrete, achieving a stable density of 1000–1100 kg/m³ while maximizing strength proved to be a challenging task. Several factors, such as the speed of the mechanical twin mixer, mixing time, variations in fly ash percentage, EPS bead percentage, and panel sheet thickness, were carefully considered and optimized before finalizing the mix design proportions. The final mixture proportions with large-sized EPS beads and small-sized EPS beads were determined based on laboratory trial batches, following the ACI mixture design guidelines⁴². Trial mixture proportions are provided in Table 6.

Table 6.

Trial mix proportions.

Components	EPS (large beads)	EPS (small beads)
Ordinary Portland cement (kg/m3)	394	394
Fine aggregate (kg/m3)	441	441
Water content (kg/m3)	166	166
w/c ratio	0.42	0.42
Super plasticizer (kg/m3)	0.59	0.59
EPS beads (kg/m3)	2.5	3.0

RSM-Based design of trial mix

The primary goal of this analysis was to assess the effects of key variables, such as plasticizer and EPS beads content on the responses i.e. compressive strength (28 days) and density properties of the EPS concrete to reach the optimum mix design. RSM technique based on the optimal design, was employed on the experimental data of trial mix of phase 1. This design was selected due to its flexibility and precision to fit a model and efficiently explore the response surface with a manageable number of experiments. Design included three factors: small EPS bead content, large EPS bead content, and plasticizer percentage, each varied over three and four levels (low, center, medium-high, and high), respectively. The experimental matrix comprised 44 trials for prediction and optimization of both compressive strength and density responses for water-to-cement (w/c) ratios of 0.5 and 0.42. Factors were transformed into goals (maximize, minimize, target, in range, equal to) to standardize input ranges and improve model stability and interpretability. Similarly, cubic, linear and quadratic polynomial equation models were proposed to forecast the density and compressive strength of EPS concrete depending on the model’s importance. Model selection was carried out based on several statistical criteria. First, Sequential Sum of Squares Analysis was used to determine whether higher-order models (such as cubic models) significantly improved the explanation of variability in the response. These models were retained only if they had statistically significant F-values (p < 0.05), indicating that the added polynomial terms contributed meaningfully beyond random variation. Second, the Adjusted and Predicted R² values were considered. Models were selected to maximize both values, with good agreement between them, to minimize the risk of overfitting and ensure the model’s ability to generalize to new data. Finally, Lack-of-Fit Tests were used to assess how well the model captured the overall data trend. Models with non-significant lack-of-fit F and p-values were preferred, as they indicated a good fit without overfitting. Accordingly, cubic models were used where the data exhibited non-linear, multi-factor interactions, while quadratic or linear models were retained for simpler trends where higher-order terms did not significantly improve the fit.

Following the selection of components at low and high levels (as shown in Table 7), 44 mix design trials of the experiments were produced. The design expert software’s relevant slots were filled using the experimental data of beads and plasticizer. The percentage EPS beads, and percentage plasticizer components were then analyzed, modeled, and optimized. The regression models’ suitability was assessed using the ANOVA’s F-value and p-value. Comparably, the design expert creates a 3D model graph to show how two elements affect the chosen answer using the suggested model.

Table 7.

Design of factors and relevant code in RSM.

Factors	Units	Code	Levels
Plasticizer	%	X1	0.2	0.3	0.4	0.5
EPS Beads	%	X2	1.5	2.0	2.5	-

Statistical assessment of optimum mix design of SB EPS concrete panel

An ANOVA analysis was conducted, and cubic, linear polynomial models were proposed to forecast the density, compressive strength attributes of EPS concrete based on the model’s importance. The model and fit statistics from the ANOVA analysis for both w/c ratio of EPS concrete with SB are shown in Table 8 to determine the relevance of established models.

Table 8.

ANOVA analysis and model validation.

Responses	EPS SB Concrete with w/c ratio 0.5		EPS SB Concrete with w/c ratio 0.42
	Density (kg/m^3)	Compressive strength (MPa)	Density (kg/m^3)	Compressive strength (MPa)
Standard deviation	21.60	0.2011	13.10	0.4040
Mean R2	0.9944	0.9971	0.9930	0.9304
Predicted R2	0.8537	0.8306	0.8089	0.8606
Adjusted R2	0.9721	0.9857	0.9650	0.9130
F-value (Model)	44.62	87.01	35.48	53.45
p-value (Model)	0.0221	< 0.0114	0.0277	< 0.0001
Model	Significant	Significant	Significant	Significant
Lack of Fit (F & p)	Not Significant	Not Significant	Not Significant	Not Significant
Final model type	Cubic	Cubic	Cubic	Linear

The ANOVA analysis shown in Table 8 demonstrates that the chosen models are highly significant, well-fitting, and there is substantial agreement between the variables and responses for all responses, as indicated by the higher F-values and lower p-values (< 0.04). Additionally, the larger R² value (> 0.80) and a difference of less than 0.2 between the predicted and adjusted R² values further supports the validity and significancy of the models, as presented in Table 8. The coefficient of determination, or R², is a measure of the quality and degree of fitness of constructed models. The model is well-fitted and there is a large amount of agreement between the actual and projected responses, as shown by an R² value of > 0.80. For w/c ratio of 0.5 and 0.42, Three-dimensional (3-D) response surfaces diagrams⁴³ have been used as an effective means of illustrating the link between the independent factors (EPS beads and plasticizer) and the related dependent variables (compressive strength and density) as shown in Fig. 5 (a) to (d), respectively. The results clearly show that percentage beads and plasticizer significantly influence the density and compressive strength of EPS concrete.

Fig. 5.

Effect of EPS small beads and plasticizer for w/c ratio 0.5 on (a) density (b) compressive strength and for w/c ratio 0.42 on (c) density (d) compressive strength.

Statistical assessment of optimum mix design of LB EPS concrete panel

The impact of changing bead sizes (Large EPS beads) in varying percentage on density and compressive strength is covered in this section. The ANOVA analysis for the model and fit statistics for both w/c ratio of EPS concrete with LB are shown in Table 9 to gauge the significance of well-established models.

Table 9.

ANOVA analysis and model validation.

Responses	EPS LB concrete with w/c ratio 0.5		EPS LB concrete with w/c ratio 0.42
	Density (kg/m^3)	Compressive strength (MPa)	Density (kg/m^3)	Compressive strength (MPa)
Standard Deviation	17.42	0.1764	17.61	0.4859
Mean R2	0.9967	0.9977	0.9939	0.9515
Predicted R2	0.8069	0.8691	0.8438	0.8928
Adjusted R2	0.9837	0.9885	0.9697	0.9030
F-value (Model)	76.45	108.38	40.97	19.62
p-value (Model)	0.0130	0.0092	0.0240	0.0027
Model	Significant	Significant	Significant	Significant
Lack of Fit (F & p)	Not Significant	Not Significant	Not Significant	Not Significant
Final Model Type	Cubic	Cubic	Cubic	Quadratic

The established models indicated a strong degree of agreement between the variables and answers; the models’ lower p-value and higher F-value indicate their significance. Comparably, more appropriate frequency and better R² (> 0.80) show that the proposed models are well-fit and noteworthy. The numerical difference of less than 0.2 between the predicted and adjusted R²also indicates the significance of the models, as shown in Table 9. of Thus, a link between independent factors (EPS beads and plasticizer) and dependent variables (density and compressive strength) has been established using these models. In addition, three-dimensional (3-D) diagrams⁴³ have been used to best describe the relationship between factors and responses for w/c ratio of 0.5 and 0.42, as illustrated in Fig. 6 (a) to (d), respectively. The results clearly show that EPS beads and plasticizer percentages have a major impact on the density and strength characteristics of EPS concrete.

Fig. 6.

Effect of EPS large beads and plasticizer for w/c ratio 0.5 on (a) density (b) compressive strength and for w/c ratio 0.42 on (c) density (d) compressive strength.

Diagnostics plots of density and compressive strength

To validate the reliability and predictive strength of the developed regression models for both density and compressive strength of EPS small and large bead concrete (at w/c ratio of 0.50 and 0.42), diagnostic plots were carefully examined (Figs. 7, 8, 9 and 10). These plots are essential to assess whether the core assumptions of regression analysis are reasonably satisfied, namely, normality of residuals, linearity, and consistency in variance.

Fig. 7.

(a) Diagnostic plots of density for EPS small beads (at 0.50 w/c). (b) Diagnostic plots of compressive strength for EPS small beads (at 0.50 w/c).

Fig. 8.

(a) Diagnostic plots of density for EPS small beads (at 0.42 w/c). (b) Diagnostic plots of compressive strength for EPS small beads (at 0.42 w/c).

Fig. 9.

(a) Diagnostic plots of density for EPS large beads (at 0.50 w/c). (b) Diagnostic plots of density for EPS large beads (at 0.50 w/c).

Fig. 10.

(a) Diagnostic plots of density for EPS large beads (at 0.42 w/c). (b) Diagnostic plots of density for EPS large beads (at 0.42 w/c).

The Normal Plot of Residuals illustrates how closely the externally studentized residuals align with a theoretical normal distribution. For both responses (density and compressive strength), the residuals cluster around the reference line, with only slight deviations at the tails. This indicates that the error terms are approximately normally distributed, which supports the statistical soundness of the model. While a few outliers exist, they do not appear to exert undue influence or suggest a serious violation of assumptions.

The Predicted vs. Actual plots further confirm the models’ robustness. The predicted values align closely with the actual experimental results, as shown by the near-linear arrangement of data points along the 45-degree reference line. This strong agreement implies that the models not only fit the data well but also possess good predictive ability, capturing the underlying trends accurately with minimal bias or systematic error.

Multi-objective optimization and validation of modeled results

Using the RSM tool, a multi-objective optimization method was applied to get the best result. Seeing how variation of EPS beads size and variation in content of both plasticizer and EPS beads in the mix design affect strength and density qualities of EPS concrete at different w/c ratio, is an intriguing observation. Specifically, the optimization aimed at achieving the goals of maximizing the compressive strength while minimizing the density. Constraints for compressive strength and density were set as, minimum strength ≥ 4–5 MPa (to ensure structural integrity for lightweight construction) and maximum density ≤ 1100 kg/m³ (to maintain the lightweight nature of EPS concrete), respectively. Large, small size beads and Plasticizer percentages were chosen as independent variables. The desirability function approach was used in the RSM software to simultaneously balance these competing objectives. It can be concluded from optimization results that using 2.5% small size EPS beads with 0.5% of plasticizer and 2% large size EPS beads with 0.4% of plasticizer, both for w/c ratio of 0.42 will result in best solution of EPS concrete. These optimized mix combinations derived for water-to-cement (w/c) ratio (0.42) not only satisfied the above constraints but also support the results achieved during the trials for mix design and are therefore, the optimal mix designs.

Sample casting and curing

Mixing of all the proportioned ingredients was done in the different sizes mixer, namely Hobart Mixer Pot and customized mixing box according to the desired size of specimen for casting. Mixing was carried out according to ASTM C192 / C192M-02 (2014)⁴⁵. For EPS concrete panel, first fine aggregates were added to the mixer with Cement and Fly ash. Once dry mix was ready, water was added in parts to prepare the cement mortar followed by mixing of EPS beads directly into cement mortar. Special type of twin mixer was used for mixing of EPS concrete to have higher number of revolutions per minute required for production of such type of lightweight concrete. Mixing time proved to be a critical factor as slight change in mixing time or mixing technique resulted in greater variation of both i.e. strength and density. For casting of EPS concrete panel specimen, adhesive material was applied on cement board sheets which were cut as per mold size and were placed inside molds, before pouring of EPS concrete. The specimens remained in molds for 24 h after pouring, and then were de-molded and moist cured for 14 days at room temperature followed by Air dry, to gain maximum strength against the desired density range of 1000–1100 kg/m³. Total curing period was 28 days. EPS concrete panel, specimens were capped with gypsum at the ends for compression and stress-strain tests, to smoothen the surface and to meet the tolerance according to ASTM C617/C617M⁴⁶.

Tests description

Thermal conductivity test

Specimens of EPS concrete were cast according to the dimensions specified in ASTM E1530⁴⁷. Each sample was tested for a total of 8 h, with exposure to two temperature points: 25 °C (room temperature) and 55 °C (maximum environmental temperature). The samples were subjected to each temperature for 4 h using a DTC 300 Thermal Conductivity meter as shown in Fig. 11, and the thermal conductivity (K) values were recorded for three samples each for EPS LB and EPS SB.

Fig. 11.

Thermal conductivity testing with data automation system (a) EPS LB (b) EPS SB.

Permeability test

Specimens of EPS concrete were cast according to the dimensions specified in EN 12390-8⁴⁸. Figure 12 shows the EPS concrete specimens during the permeability test. Each sample was tested for 72 h under a pressure of 4 bar, maintained using an air compressor. A water tube filled with water was placed at the top of each specimen. After completing the test, the samples were split, and the readings were recorded in millimeters for three samples each for EPS LB and EPS SB.

Fig. 12.

Permeability Testing machine with attached air compressor for (a) EPS LB. (b) EPS SB.

Water absorption test

For both large and small EPS beads, cube specimens of EPS concrete panel with the same size as those cast for the compressive strength test were also cast for the water absorption test, in compliance with ASTM D1621-16⁴⁹. All concrete specimens were oven-dried at 110 °C for 24 h. After removal from the oven, and cooling the specimens for 6 h, their initial weights were recorded. The specimens were then submerged in water for 72 h. Upon removal from the water, the surfaces of the samples were wiped to remove excess water, and their weights were measured again for three samples each for EPS LB and EPS SB.

Drying shrinkage test

EPS concrete specimens for the drying shrinkage test were cast according to the dimensions specified in ASTM (ASTM C596-09e1)⁵⁰. Initial readings for each sample were taken 24 h after casting. Subsequent readings were then recorded on the 1st, 3rd, 7th, 14th, and 28th days. Total shrinkage was calculated relative to the initial reading for three samples each for EPS LB and EPS SB.

Compressive strength test

Compressive test specimens of EPS concrete panel were casted in compliance with ASTM D1621-16⁴⁹ and ASTM C192/C192M-02⁴⁶ for the EPS concrete mix and were tested for compressive strength, as shown in Fig. 13. The specimens were loaded until failure cracks appeared on the surface and the load dropped below 20%. The tests were conducted in displacement-controlled mode. The specimens were attached to a 2000 kN MATEST CTM equipped with a compressometer to measure vertical displacement. The loading rate was adjusted to maintain a constant machine head movement of 1 mm/min and compressive strength values were measured for three samples each for EPS LB and EPS SB.

Fig. 13.

(a) EPS concrete panel specimen. (b) compressive testing machine with data automation system.

Flexural strength test

The flexural strength of the beams was measured in accordance with ASTM C293/C293M⁵¹. A center-point loading setup was used for the testing. Figure 14 shows the EPS concrete panel specimen during the flexure test. Each sample was loaded at a rate of 1 mm per minute until the specimen failed and flexure strength values were measured for three samples each for EPS LB and EPS SB.

Fig. 14.

Center point loading setup for flexural strength of EPS concrete panel beam.

Sound insulation test

Two essential sound characteristics were selected based on previous research on measuring the sound insulation properties of barriers⁵²: the constant rate frequency emitted from a source and the space available for sound waves to create a reverberation effect. The sound absorption was measured in a reverberation room in accordance with ASTM C423⁵³. To evaluate the sound insulation properties of EPS, a miniature reverberation chamber was constructed using six panels of EPS concrete panel, as shown in Fig. 15.

Fig. 15.

Miniaturized reverberation chamber of EPS concrete panels.

Fire rating tests

This test method is designed to evaluate the duration for which EPS concrete panel specimens can contain a fire, retain their structural integrity, or exhibit both properties during a predetermined test exposure⁵⁴. Cube specimens were cast in compliance with ASTM D1621-16⁴⁹ and exposed to a temperature range of 100 °C to 500 °C in an electric furnace. Each specimen was subjected to increasing temperatures for 5 h before undergoing strength tests and visual inspections. Two Type-K thermocouples were embedded in each cube—one in the core and the other on the specimen’s surface and connected to a temperature controller to record real-time temperature variations, as shown in Fig. 16. After exposure, each specimen was tested for compressive strength, and the fire rating was determined for the respective temperature range. For temperature range of 100 °C to 500 °C, against each temperature, three samples each for EPS LB and EPS SB were tested.

Fig. 16.

(a) K- type thermocouples. (b) EPS concrete specimen inside furnace along with temperature recorder.

Wallette test

In this test method, a masonry assemblage is loaded in compression along one diagonal to induce a diagonal tension failure. The specimen splits apart parallel to the direction of loading, allowing the determination of diagonal tensile or shear strength. Wallette test specimens were cast in compliance with ASTM E519/E519M⁵⁵ for the EPS concrete panel mix. After a 28-day curing period, the samples were tested for diagonal shear strength. Each specimen was loaded at a rate of 1 mm per minute until failure, as shown in Fig. 17 and diagonal shear strength values were measured for three samples each for EPS LB and EPS SB.

Fig. 17.

Wallette test for diagonal shear strength.

Results and discussions

The results from physical, mechanical, functional, and structural property tests conducted on EPS concrete panel specimens with a density of 1000–1100 kg/m³ are presented. These results are compared with those of conventional heavy infill materials, such as brick and normal aggregate concrete, which have higher density values of 1920 kg/m³ and 2320 kg/m³, respectively.

Thermal conductivity test

The results from the thermal conductivity test conducted on lightweight EPS concrete specimens, along with a comparison to conventional heavy infill materials such as brick and normal aggregate concrete, are presented in Fig. 18.

Fig. 18.

Thermal conductivity comparison of EPS LB & EPS SB with conventional material.

The primary objective of this experiment was to evaluate the thermal conductivity performance of EPS concrete over time and to analyze the effects of varying bead size and percentage on the heat transfer behavior of the specimens. 25 °C and 55 °C are selected as the range of temperature to check the thermal conductivity performance (K) of EPS concrete LB & SB at room temperature and possible maximum temperature for tropical urban environments, respectively. Thermal conductivity (K) values depend on the passage of heat waves through the concrete. A higher percentage of beads results in more air particles within the material, leading to lower thermal conductivity. Small beads exhibited a lower average K value (0.45 W/mK) at 25 °C and 55 °C compared to large beads (0.45 W/mK), making them more effective thermal insulators for building envelope applications. In contrast, brick and normal aggregate concrete showed higher K values of 0.8 W/mK and 0.5 W/mK, respectively, indicating inferior thermal insulation properties compared to EPS concrete.

Permeability test

The results of the permeability test are presented in Fig. 19. The primary objective of this experiment was to evaluate the permeability of EPS concrete over time and to analyze the effects of varying bead size and percentage on the flow of water vapor through the specimens. The permeability percentage of lightweight concrete is directly related to the porosity of the specimen. More porous concrete, which has a lower density due to a higher number of air voids caused by a greater percentage of beads, allows more water to pass through. EPS panels with a smaller quantity of large beads, being less porous demonstrated lower permeability (0.57%), while those with a higher quantity of small beads showed higher permeability (1.86%). However, both values are significantly lower compared to brick and normal aggregate concrete, which have much higher permeability values of 5.2% and 2.4%, respectively.

Fig. 19.

Permeability comparison of EPS LB & EPS SB with conventional material.

Water absorption test

Figure 20 presents the results of the water absorption test for EPS concrete panel, conventional concrete, and bricks. The results indicate that higher percentages of beads lead to greater water absorption (8.5%). While EPS aggregates themselves absorb very little water, the increased water absorption in the concrete is likely due to the high porosity of the coating materials and the interfacial transition zone (ITZ) between the EPS and the coating. These factors create gaps in the concrete mix, allowing more water to be absorbed. EPS concrete panels with a lower percentage of beads absorbed less water (7.8%), whereas those with a higher percentage of small beads showed higher water absorption (8.5%). However, both values are significantly lower compared to brick and normal aggregate concrete, which have much higher water absorption values of 15% and 9%, respectively.

Fig. 20.

Water absorption comparison of EPS LB & EPS SB with conventional material.

Drying shrinkage test

The results of drying shrinkage as shown in Fig. 21 shows that drying shrinkage value of EPS concrete is increased by increasing EPS aggregate content. The main reason is polystyrene aggregates’ high compressibility and low stiffness, which offer little restraint to the shrinkage process.

Fig. 21.

Drying shrinkage comparison of EPS LB & EPS SB with conventional material.

Higher percentage of beads shows more drying shrinkage (307µɛ) as it gives more space in the concrete matrix to shrink with time compared to EPS concrete with less percentage of beads. EPS concrete with less percentage of beads has shown less drying shrinkage (284µɛ) whereas small beads in greater quantity have shown higher drying shrinkage value (307µɛ), which are very less compared to brick and normal aggregate concrete, having higher drying shrinkage values of (581µɛ) and (500µɛ), respectively.

Compressive strength test

The results of the compressive strength test conducted on lightweight EPS concrete panel specimens, along with a comparison to conventional heavy infill materials such as brick and normal aggregate concrete, are presented in Fig. 22. The primary objective of this experiment was to evaluate the compressive strength of EPS concrete panels and to analyze the effects of varying bead size and percentage on the compressive strength of the specimens. Load and deformation data were collected using a load data acquisition system connected to the Universal Testing Machine and LVDTs, respectively. A stress-strain curve was plotted based on the load-deformation response at room temperature. Due to the hydrophobicity and mechanical properties of EPS aggregates, as well as the weak bond between the EPS surface and the cement paste, the presence of EPS reduces the compressive strength of the concrete. EPS concrete panels with a lower proportion of large beads demonstrated higher compressive strength mainly due to reduced voids and better packing density, while mix designs with a higher proportion of small beads resulted in lower compressive strength. After 28 days of testing, the maximum compressive strength achieved for EPS concrete panels with large beads (LB) was 6.29 MPa, while panels with small beads (SB) reached 5.42 MPa at room temperature. These values are lower compared to brick and normal aggregate concrete, which exhibited higher compressive strengths of 8 MPa and 9 MPa, respectively.

Fig. 22.

Compressive strength comparison of EPS LB & EPS SB with conventional material.

Stress strain curve

The stress-strain response of the concrete mixes containing large and small size EPS beads and panel sheets reveals significant effects of EPS beads percentage and size on both strength and ductility as shown in Fig. 23 (a) and (b).

Fig. 23.

Stress strain curve of (a) EPS LB, (b) EPS SB.

At room temperature, EPS concrete panel with both large and small beads exhibited comparable ultimate strength of 6 to 7 MPa. But EPS concrete with small size beads in greater percentage exhibited more ductile behavior compared to EPS concrete with large size beads which have reflected a sharp stress drop after the peak value indicating lower residual strength and brittle behavior. These findings highlight the distinct roles of EPS beads in higher percentage for improving the ductility of EPS concrete, maintaining the strength over a longer strain range, making them more damage tolerant. The higher ductility observed in the EPS-SB mix can be attributed to several micro mechanisms. First, smaller beads disperse more uniformly, producing a finer pore network that acts as multiple crack “arresters”, increasing propagation path of the micro crack. Second, the larger specific surface area of small beads enhances mechanical interlocking with the cement hydrates which improves stress transfer across the matrix–bead interface and delays debonding. This allows the composite to sustain load over a longer strain range. Also, the yielding of EPS beads creates local stress relief zones. These zones help to interrupt the propagation of crack tips, which in turn increases the post-peak deformation capacity, as observed in Fig. 19. Collectively, these factors explain why the EPS SB mix maintains 80–85% of its peak stress up to strains of ~ 0.013 mm/mm, whereas the larger bead mix shows an earlier stress drop.

Flexural strength test

The results of the flexural strength test are presented in Fig. 24 (a) and (b). Unlike the typical brittle failure mode observed in conventional concrete, the EPS concrete panel specimens (533 mm x 152 mm x 152 mm) did not split in half. Instead, the failure was more gradual. In all tests, cracks primarily developed within the middle third of the span length on the tension surface of the beams. A slight nonlinearity was observed in the load deformation curve of the EPS-LB beam, which occurred when the adhesive bond between the EPS panel sheet and the EPS concrete matrix broke during loading. Compared to EPS concrete panel with small beads (displacement at peak load: 1.1 mm), EPS concrete panel with large beads exhibited a larger displacement at peak load (1.5 mm). EPS concrete panel beams with a lower percentage of large beads carried more load and deformed more, owing to the reduced proportion of beads. In contrast, beams with a higher proportion of beads had a lower load-carrying capacity and deformed less. After 28 days of testing, the maximum load attained by the EPS concrete panel beam with large beads (LB) was 5.30 kN, while the beam with small beads (SB) reached 4.20 kN at room temperature. These values are significantly lower compared to normal aggregate concrete, which demonstrated a much higher flexural load capacity of 50 kN as shown in Fig. 25.

Fig. 24.

Flexure load-deformation curve of (a) EPS LB beam. (b) EPS SB beam.

Fig. 25.

Flexural strength comparison of EPS LB & EPS SB with conventional material beam.

Sound insulation test

The results from sound insulation test are presented in Table 10. Acoustic performance at different frequency ranges was tested up to 1200 Hz frequency and was found as 41–42 dB for both types of EPS beads Panels. Acoustic performance of EPS concrete panel shows that sound waves do not find easy to pass through air medium compared to the dense medium. EPS concrete panels have shown good sound absorption due to their porous nature. Higher the air voids present, greater will be the sound insulation. Large beads in lesser proportion have more compact structure resulting in less air voids (less energy loss of sound waves) and thus lower sound insulation than small beads in higher proportion, who acted as good sound insulator giving less value of 41 dB value. Both sound insulating values of 41dB and 42 dB are very less compared to sound insulating values of brick and normal aggregate concrete, each having value of 50 dB. The use of a miniaturized reverberation chamber, used for sound insulation test have limitations compared to full-scale standardized acoustic testing per ASTM C423. Specifically, scale effects may influence absolute sound absorption values. However, for comparative assessment between samples under controlled conditions, allowing relative evaluation of sound insulation performance of EPS concrete panels, this setup remains suitable as done during the test of sound insulation properties of barriers⁵².

Table 10.

Sound insulation comparison of EPS LB & SB with conventional material.

Specimen type	Sound Insulation (dB)
EPS large beads	42
EPS small beads	41
Brick	50
Concrete block	50

Fire rating test

The results from fire rating test conducted on lightweight EPS concrete panel specimens and their comparison with conventional heavy infill material i.e. brick and normal aggregate concrete, are presented in Figs. 26 and 27, respectively. The basic purpose of this experiment was to observe the fire rating of EPS concrete over time and to determine the effects of changing beads size and percentage, on fire resistance properties, of the specimens. EPS concrete panels have shown high fire rating despite very low burning temperature of beads, mainly due to the confinement and heat insulation effect provided by EPS panel sheet.

Fig. 26.

Fire strength comparison of EPS LB & EPS SB specimen at different temperature.

Fig. 27.

Fire rating comparison of EPS LB & EPS SB specimen with conventional materials.

EPS panels retained structural integrity for over 4 h at 400℃ (compressive strength reduction was around 50% that of the compressive strength of specimens tested at room temperature i.e. 25 °C). For EPS concrete panels with large and small beads, no huge variation of strength was observed, after being exposed to temperature. Both have reflected same range of strength against each temperature range. Bar graph in Fig. 22 shows decrease in strength with increase in temperature. For EPS concrete panel with compressive strength of 4 MPa to 5 MPa at room temperature, about 20% strength reduction was observed at temperature of 150 °C, around 35% strength reduction at 200 °C, 45% strength reduction at 300 °C, 60% strength reduction at 400 °C and 100% strength reduction at 500 °C. All specimens were heated in furnace for 5 h. Once specimens were heated at 500 °C for 2 h, strength reduction was noted as 80%. EPS concrete panel with large and small beads have shown fire rating of 4 h against temperature of 400 °C, whereas brick and normal aggregate concrete, have shown fire rating of 2 h and 5 h against temperature of 400 °C, respectively. Although this test provided valuable insights into the thermal degradation and structural integrity of EPS concrete panels under sustained high temperatures, this testing procedure has the limitation compared to a standardized time-temperature curve such as those prescribed in ASTM E119. While the results effectively demonstrate the relative fire endurance of EPS concrete panels compared to traditional materials like brick and conventional concrete, the absence of a recognized fire curve limits direct applicability for regulatory benchmarking.

Wallette test

EPS concrete panel wallette (609 mm x 609 mm x 101 mm) with large beads in lesser proportion resisted more diagonal force (133.4 KN) compared to small beads (103.2KN) in higher proportion due to more compact structure of the concrete matrix as shown in Fig. 28 (a) and (b). The results of the Wallette test have been compiled in the Table 11. Both type of EPS concrete Panels have Diagonal Shear stress values (1.07 MPa and 0.83 MPa), greater than that of brick masonry (0.7 MPa), which exhibits more stiff behavior. Demonstration of a favorable combination of strength and ductility of EPS concrete panels, particularly with LB is due to the cellular structure and energy absorption characteristics of EPS concrete which contribute to this performance, making it a promising material compared to conventional brick masonry in earthquake-prone regions for improving the seismic performance of infill panels or partition walls under lateral or cyclic loading in seismic zones, where materials must dissipate energy and accommodate inelastic deformations without catastrophic failure.

Fig. 28.

Diagonal load-deformation curve of (a) EPS LB wallette. (b) EPS SB wallette.

Table 11.

Shear stress comparison of EPS LB & SB with conventional material.

Specimen type	Diagonal load (KN)	Shear stress (MPa)
EPS Large Beads	133.4	1.07
EPS Small Beads	103.2	0.830
Brick	72.9	0.7

Conclusions

This study evaluated the effects of varying Expanded Polystyrene (EPS) beads sizes and contents on the performance of EPS concrete panels and EPS concrete, followed by their performance evaluation at optimum concrete mix designs, for replacement of conventional heavy infill walls. At the end, comparison with conventional infill materials was also conducted. Following conclusions were drawn from the research study:

• EPS concrete panels achieved approximately 43–48% weight reduction compared to conventional infill materials like bricks and normal aggregate concrete. This significant weight reduction allows for more economical and lightweight structural designs.

• The compressive strength of EPS concrete panels was observed to be 6.29 MPa for large beads and 5.42 MPa for small beads at room temperature, making them suitable for non-structural lightweight applications. However, small beads led to a slight reduction in strength due to higher porosity.

• EPS concrete panels demonstrated higher resistance diagonal loading, with large beads achieving 133.4 kN and small beads 103.2 kN, translating to shear stress values of 1.07 MPa and 0.83 MPa, respectively, with both surpassing that of brick masonry (0.7 MPa) and higher ductility. This highlights improved performance of EPS panels in seismic prone areas.

• EPS concrete demonstrated excellent thermal insulation, with conductivity values of 0.43 W/mK for small beads and 0.45 W/mK for large beads, significantly lower than bricks (0.8 W/mK), making them effective for energy-efficient construction.

• The flexural strength of EPS concrete panels was lower than conventional concrete, with large bead panels supporting a peak load of 5.30 kN and small bead panels 4.20 kN, reflecting the material’s non-structural application focus.

• EPS concrete panels exhibited significantly lower permeability and water absorption values of 0.57-1.86% and 7.8–8.5%, respectively compared to brick with values of 5.2% and 15%. This demonstrates the enhanced durability and improved resistance of EPS panels against water ingress due to the low porosity of EPS beads.

• EPS concrete panels maintained structural integrity for over 4 h at 400 °C (compressive strength reduction was around 50% that of the compressive strength of specimens tested at room temperature i.e. 25 °C), outperforming bricks with only 2 h of resistance. This highlights the potential of EPS panels for improved fire safety in buildings.

• Sound insulation properties of EPS concrete panels were measured at 41 dB (small beads) and 42 dB (large beads), slightly lower than bricks (50 dB) but adequate for reducing sound.

• transmission in lightweight wall applications.

• A statistical model using ANOVA was created to analyze the compressive strength and density with the variables of EPS beads and plasticizer content. The models were validated and found to have high R² (> 0.80) and adequate precision (> 4.0), as well as lower p-values. This confirms that all the models are significant, well-fitted, and can be used for predicting the responses.

Future work

For future research on EPS Concrete Panels, it is recommended that lightweight EPS concrete should be explored further for structural applications, with an emphasis on enhancing fire safety and seismic performance to assess the scalability of EPS concrete panels for large construction projects. In addition, potential of using demolished EPS concrete as a recycled aggregate also need to be studied, in order to avoid challenges being faced in decomposing the polystyrene beads and to safeguard the environment. Finally, incorporating more advanced modeling techniques and simulation tools could help predict the behavior of EPS concrete in real-world conditions, assisting in the design of more efficient and durable concrete structures.

Author contributions

Conceptualization, M.I.A. and A.Z.; methodology, M.I.A, A.Z, and R.A.; data curation, M.I.A, A.Z and M.I.K.; formal analysis, M.I.A, A.Z and M.I.K.; resources, A.Z, R.A, M.I.K.; Writing - original draft, M.I.A, A.Z,.; writing – review and editing, A.Z, R.A, M.I.K.; supervision, A.Z, R.A, M.I.K; Validation, M.I.A, A.Z, R.A, M.I.K. All authors have read and agreed to this version of the manuscript.

Funding

This work was supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University (IMSIU) (grant number IMSIU-DDRSP2502).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.