Influence of recycling in the dynamic mechanical properties of expanded polystyrene foam

Abstract

Expanded polystyrene (EPS) is a cellular material widely used in energy-absorbing systems due to its ability to dissipate kinetic energy through compression. EPS can be recycled, offering a promising avenue for sustainable material use. This study aims to characterize the mechanical response of recycled EPS under dynamic loading conditions. Both external factors (temperature, impact velocity) and internal factors (density, recycling ratio) were investigated, and the behavior of recycled EPS was compared to that of classical material. The results show that recycled EPS exhibits no significant sensitivity to temperature within the range of −20°C to 18°C. However, its mechanical response is strongly influenced by impact velocity and foam density. No statistical difference was observed between recycled and classical EPS at a density of 60 kg/m³. In contrast, at 80 kg/m³, recycling induced a marked shift in the stress–strain curves, indicating that higher densities amplify the effects of the recycling process on mechanical performance.

Keywords

expanded polystyrene recycling temperature impact velocity

Introduction

Expanded polystyrene (EPS) is a polymeric foam extensively used in various industrial applications, particularly in protective systems.^1–4 Owing to its high energy absorption capacity, EPS is commonly integrated as a liner material in helmets to attenuate impact forces and reduce the acceleration transmitted to the brain.⁵ The primary advantage of expanded polystyrene foam lies in its capacity to dissipate kinetic energy through compression without a significant increase in stress. This characteristic is typically manifested as a plateau region on the stress–strain curve.⁶ Moreover, the material is lightweight and easily moldable.

Indeed, EPS exhibits a hysteretic mechanical response. The loading phase is divided into three regions: the initial linear elastic region, the plateau region, and the densification region.^7–9 Upon reaching the densification stage, the material’s capacity to absorb energy without a significant increase in stress is exhausted, resulting in a sharp rise in stress with increasing strain. After attaining maximum compression, the material undergoes a relaxation process during which it experiences permanent deformation; this corresponds to the unloading phase.^10,11

To characterize the mechanical behavior of EPS, several parameters can be extracted from its stress-strain response.¹² The loading phase is typically described by the Young’s modulus, the plateau slope, the yield point, and the maximum point. Concerning, the unloading phase, the parameters are the unloading modulus and the permanent strain. In classical EPS foams, these parameters are influenced by internal factors such as foam density, as well as external factors including impact velocity.¹²

The mechanical properties of classical EPS have been extensively studied and characterized over the past decades. According to Di Landro,⁷ density is the dominant factor governing the mechanical response: increasing EPS density leads to an overall increase in the stress-strain response, including higher Young’s modulus and higher mean plateau stress, as well as decreasing both maximum strain and stress peak values. Impact velocity is also a factor that affect the stress–strain behavior, as shown in a previous study,¹² influencing both the peak response and the unloading phase.

Nowadays, an increasing number of studies focus on material recycling, as reducing waste has become a major environmental challenge. Research has already been conducted on poly (ethylene terephthalate) foams produced from recycled thermoplastics,¹³ as well as on the substitution of classical expanded polypropylene or polyolefin blowing agents with recycled counterparts for foam manufacturing.¹⁴

In addition, several studies have investigated the production of polymers such as polypropylene using recycled or recyclable materials,^15–18 as well as the recycling of these polymers into more complex structures, including Ethylene Propylene Diene Monomer (EPDM) and Polypropylene (PP) blends and recycled polyethylene-based microfibrillated composites.

Since EPS is used in various fields, several studies have proposed methods to recycle it from classical material. Kan et al. (2009)¹⁹ introduced a recycling process based on heat treatment. The authors investigated the effects of this process on density after treatment, thermal conductivity, and compressive strength, showing that all these properties are influenced by the proposed recycling method. However, no stress-strain data were provided to assess the mechanical response before and after recycling. To date, few studies have investigated the mechanical behavior of recycled EPS, and none under dynamic loading conditions or at densities comparable to those used in helmet liners. Acierno et al.²⁰ examined the effect of recycling on selected mechanical parameters, such as the compressive stress at 10% strain; however, their work was limited to low-density EPS (20 kg/m³) and quasi-static loading at 25 mm/min. Barrera et al.²¹ studied the mechanical and thermal behavior of recycled EPS, but their investigation focused on low recycling ratios (<20%) and low densities (25 kg/m³), also under quasi-static conditions (2.5 mm/min). Moreover, Hornberger et al.²² explored the influence of recycling under dynamic conditions using a drop-tower setup, impacting recycled EPS specimens at 3.87 m/s while varying the mass of the impactor from 2.8 kg to 34.7 kg. Although their results cover recycling ratios up to 100%, the density of the EPS used was not reported, making the interpretation and comparison of their findings difficult.

The main contribution of the present work is the investigation of the effect of low temperature on recycled EPS, along with the influence of impact velocity, foam density, and recycling ratio on its mechanical response for high-density EPS (∼60 kg/m³ and ∼80 kg/m³) at elevated recycling levels (50% and 89%). The study also highlights the differences between classical and recycled EPS, both in their stress–strain behavior and in key mechanical parameters.

Material and methods

The dynamic compression tests were performed using a drop tower equipped with a 5.2 kg cylindrical impactor (63 mm in diameter) designed to simulate the weight of a helmet head impacting recycled EPS samples at various impact velocities. An accelerometer (PCB 353B04 of ±500 g range), sampled at 50 kHz, was mounted on the impactor to record acceleration-time signal during foam compression. To accurately measure the impact velocity at the onset of compression, a trigger was positioned above the sample. Compression tests were recorded using a high-speed camera to ensure a plane-to-plane contact between the impactor and the specimen. The samples were cut from a 200 × 200 × 25 mm³ EPS plate into parallelepipeds measuring 35 × 35 × 25 mm³. Each sample was individually measured and weighted. The average densities of the different foam are summarized in Table 1 and a schematic diagram is shown in Figure 1.

Table 1.

Measured densities for recycled EPS.

Named samples	EPS60R50	EPS60R89	EPS80R50	EPS80R89
Measured density (kg/m³)	65.74 ± 1.53	65.00 ± 1.46	88.32 ± 2.31	86.72 ± 1.67

Figure 1.

Experimental setup of dynamic compression tests.

In this study, two EPS densities were investigated (60 kg/m³ and 80 kg/m³), each combined with two recycling ratios (50% and 89%). Regarding the production of these EPS, several key steps are required. 1. Pre-expansion: The process begins with pre-expanding polystyrene beads using steam. This step ensures uniform bead expansion and density control, laying the basis for high-quality foam production. 2. Molding: The pre-expanded beads are then transferred into molds, where heat and pressure treatment are applied to achieve the required shape and density. Precision during this phase is essential to meet product specifications and quality standards. 3. Cooling and Demolding: After molding, the formed EPS products are cooled to stabilize their structure. Once cooled, the products are demolded and inspected for quality assurance before any further processing or packaging. The recycled materials used in this study consist of Polysource 225 polystyrene resin. This material is certified by SCS Global Services as having a minimum of 50% and 89% post-consumer polystyrene content for EPSR50 and EPSR89, respectively. The recycling process conforms to the SCS Recycled Content Standard V8-0, and the material quantification and mass-balance calculations are performed on a dry-weight basis.

For each material configuration, the influence of temperature (−20°C, 0°C, 18°C) and impact velocity (3 m/s, 4 m/s, 5 m/s) was investigated. The selected temperatures were intended to represent cold environmental conditions: −20°C (freezer), 0°C (refrigerator), and room temperature (18°C), in accordance with typical conditioning procedures used for cold configurations in helmet standards.

For the low-temperature tests, samples were conditioned at the target temperature for at least 24 h prior to testing and were removed at the moment of impact. To evaluate the repeatability of the dynamic compression tests, each configuration was tested 10 times, resulting in a total of 360 individual compression tests.

To independently assess the influence of strain rate on the mechanical response of EPS, additional quasi-static tests were conducted at room temperature. The quasi-static experiments were performed using an INSTRON 3345 testing machine with a load capacity of 5000 N. A prescribed displacement was applied until 80% strain was reached. The sampling frequency was set to 5 Hz, 50 Hz, and 500 Hz for strain rates of 0.001 s^-1, 0.01 s^-1, and 0.1 s^-1, respectively. The experiments were repeated four times to assess their repeatability, resulting in a total of 48 additional tests.

To derive stress-strain curves from the raw acceleration-time signals, the data were first filtered using a CFC1000 filter to attenuate high-frequency oscillations caused by noise during dynamic compression tests. The acceleration peaks were then isolated using two thresholds. The onset of each impact was defined as the point where acceleration exceeded 10 m/s², and the end of the signal was set based on a time window with a duration depended on the impact velocity. The filtered acceleration-time curves were then integrated twice using the trapezoidal method to obtain displacement-time curves. Integration constants were set by imposing zero displacement at the beginning of the compression. Strain was then obtained by dividing the displacement by the initial height of the sample, while stress was calculated by multiplying the acceleration by the ratio of the mass to the sample area.

The mean stress-strain curve was computed together with its corresponding standard deviation. To capture variability in both the strain and stress directions, a non-conventional approach was adopted in this study. Instead of relying solely on vertical (stress-wise) deviations, a strain-stress corridor was constructed around the mean curve by considering the maximum deviations observed in both directions. For each point along the curve, the most extreme neighboring value, whether in strain or stress, was identified within the dataset to maximize the deviation and, therefore, the associated uncertainty.

This corridor provides a reference for assessing the robustness of the experimental results. Desquilbet²³ introduced a coefficient of variation, defined as the ratio between the standard deviation and the mean value. To extend this metric along the entire mean stress–strain curve, an integrated form of the coefficient was used. The previously defined corridor was then combined with this metric to quantify the variability throughout the full stress–strain domain.

Denoting $\sqrt{v}$ , as the standard deviation and $μ$ as the mean value, the integrated coefficient of variation is defined as follows:

c_{v} = \frac{\int \sqrt{v}}{\int μ}

(1)

Desquilbet²³ defined a threshold of $c_{v} \leq 10 %$ to validate the repeatability of the experiments and to consider the dataset as statistically robust. This computation was automated using a Python script, which takes as input the 10 acceleration-time curves and outputs the mean stress-strain curve along with the corresponding standard deviation at each point in both stress and strain directions, as well as the integrated coefficient of variation.

Moreover, key mechanical parameters were extracted from the mean stress-train curves. A robust extraction method previously developed was used to extract Young’s modulus, plateau slope, yield point, maximum point, unloading modulus, and permanent strain.¹² The influence of temperature, impact velocity, and recycling ratio on these parameters was then investigated.

To statistically evaluate the influence of density, impact velocity, and temperature on the mechanical response of EPS, Welch’s t-test was employed.²⁴ The test was applied pointwise along the mean stress–strain curves to assess global behavioral differences between configurations. In addition, to evaluate the effect of these factors on specific regions of the curve, Welch’s test was performed on vectors of extracted mechanical parameters (e.g., Young’s modulus parameter at various temperatures). Le $t ({\bar{X}}_{1}$ , ${\bar{X}}_{2}$ ) denote two mean stress-strain curves or two parameters’ vectors of size $N$ , and ( $s_{1}$ , $s_{2})$ their respective variances. The t-statistic is defined as follows:

t = \frac{{\bar{X}}_{1} - {\bar{X}}_{2}}{\sqrt{\frac{{s_{1}}^{2} + {s_{2}}^{2}}{N}}}

(2)

This approach enables the identification of specific regions where mechanical responses differ significantly between experimental conditions, offering both global and local statistical insights. The Welch’s t-test is associated to a p-value ranging from 0 to 1, which quantifies the probability that the observed difference occurs by chance. By adopting a significance level of $α$ = 0.05,²⁵ the test determines whether the difference between two curves or vectors are meaningful. Specifically, if the p-value exceeds this threshold, no statistical difference is observed, indicating that the studied factor does not affect the mechanical behavior of EPS. Conversely, a p-value below 0.05 reveals a statistically significant difference, suggesting that the parameter under investigation influences material response.

Pointwise Welch’s t-tests were performed along the curves. No additional correction for multiple comparisons was applied; however, conclusions are based on consistent trends across multiple points rather than isolated significant points.

For the alignment of the quasi-static curves, a threshold point corresponding to a force of 50 N was selected as a reference. The curves were shifted along the strain axis to align this point, thereby eliminating the initial portion of the signal where contact between the testing device and the specimen is insufficient. Finally, the mean stress–strain curve, along with its standard deviation in both stress and strain, was determined using a procedure similar to that employed for dynamic loading.

Results of recycled EPS characterization

Figure 2 illustrates the different phases of a compression test at an impact velocity of 5 m/s under dynamic loading as compression progresses. State (a) corresponds to the initial moment of contact between the specimen and the impactor. State (b) represents the mid-compression stage, where the EPS reaches the end of its plastic deformation regime. State (c) shows the state of maximum compression at the end of densification, while state (d) depicts the material recovery until a permanent deformation is reached after rebound of the impactor.

Figure 2.

Recycled EPS dynamic compression illustration at 5 m/s impact velocity and at (a) initial stage, (b) middle compression, (c) maximum compression, (d) end of relaxation.

The repeatability of the experiments was assessed using the integrated coefficient of variation, as reported in Table 2. The threshold condition of

c_{v} < 10 %

as proposed by Desquilbet²³ was satisfied in most cases for 50% recycled EPS samples. However, slightly higher values were observed for samples with 89% recycled content, especially for EPS60R89. This increase can be attributed to the inherent inhomogeneity introduced during the recycling process, which can lead to greater specimen-to-specimen variability.

Table 2.

Coefficient of variation for all configurations with values exceeding Desquilbet’s threshold shown in bold.

Impact velocity	Temperature	EPS60R50	EPS60R89	EPS80R50	EPS80R89
3 m/s	18°C	6%	7%	6%	11%
	0°C	7%	12%	6%	6%
	−20°C	6%	8%	8%	5%
4 m/s	18°C	9%	14%	5%	7%
	0°C	10%	12%	7%	6%
	−20°C	6%	16%	8%	6%
5 m/s	18°C	9%	14%	8%	6%
	0°C	11%	16%	8%	6%
	−20°C	12%	12%	12%	9%

Importantly, the variability bounds used in this study do not represent a statistical confidence interval. Instead, a data-driven variability envelope was constructed to capture the full range of experimental dispersion. Stress–strain curves of expanded foams exhibit variability along both axes, meaning that the dispersion is not purely vertical (stress) nor purely horizontal (strain). For this reason, conventional pointwise standard-deviation bands would underestimate the true spread of the data. To overcome this limitation, the adopted approach identifies, at each point along the mean curve, the most extreme neighboring data points in both stress and strain directions. This method provides a conservative, assumption-free representation of experimental variability, particularly suited to nonlinear mechanical responses where the underlying statistical distribution is unknown. Within this framework, the observed variability remains consistent with the expected behavior of recycled EPS, supporting the robustness of the experimental results.

Figure 3 illustrates the typical effect of temperature at an impact velocity of 4 m/s for EPS80R50. Welch’s t-test was conducted along the entire stress-strain curve for each recycled EPS in order to assess if a statistical difference occurs between mean curves at various temperature. More specifically, pairwise comparisons were performed between the mean stress–strain curves at each temperature: 18°C versus 0°C, 18°C versus −20°C, and 0°C versus −20°C. These comparisons enabled a detailed assessment of the temperature influence on EPS behavior across the studied range.

Figure 3.

Temperature influence on EPS80R50 at 4 m/s impact velocity.

The results indicate that temperature generally does not influence the foam’s response, as p-values exceed the threshold of

α

= 0.05, except for EPS80R50 at an impact velocity of 3 m/s, as summarized in Table 3. This isolated sensitivity at low velocity may be explained by the fact that, at lower strain rates, thermal effects on the polymer structure become more pronounced due to reduced strain-rate hardening. Strain-rate hardening refers to the increase in foam resistance at higher deformation rates due to viscoelastic effects in the polymer matrix. In addition, EPS80R50, being both low in density and moderately recycled, might exhibit greater microstructural variability, i.e., natural heterogeneity in cell size, wall thickness, and density distribution, making it more susceptible to temperature-induced changes in stiffness and deformation mechanisms. At higher velocities, inertial and rate-dependent effects may dominate, masking any temperature-related differences. Consequently, the stress–strain curves were grouped by density, impact velocity, and recycling ratio, increasing the number of curves per configuration from 10 to 30. In the following sections, all mean stress–strain curves will therefore refer to the average of these 30 curves, providing a more robust statistical representation of each configuration.

Table 3.

Temperature influence: p-values for all impact configurations. Values below the significance threshold (p < 0.05) are shown in bold.

	Impact velocity	p-value 18°C vs 0°C	p-value 18°C vs -20°C	p-value 0°C vs -20°C
EPS60R50	3 m/s	0.96	0.92	0.93
	4 m/s	0.94	0.84	0.88
	5 m/s	0.85	0.98	0.85
EPS60R89	3 m/s	0.60	0.29	0.54
	4 m/s	0.68	0.72	0.88
	5 m/s	0.81	0.70	0.89
EPS80R50	3 m/s	9e-6	4e-4	0.28
	4 m/s	0.67	0.16	0.34
	5 m/s	0.90	0.63	0.71
EPS80R89	3 m/s	0.49	0.75	0.43
	4 m/s	0.92	0.96	0.88
	5 m/s	0.48	0.75	0.43

Moreover, impact velocity has a pronounced influence on the behavior of EPS. This effect is illustrated in Figure 4 in dynamic loading for EPS80R50, where clear differences in stress-strain response are observed. Table 4 reports the results of Welch’s t-test, which yields p-values below the threshold of $α$ = 0.05, confirming that the behavior of EPS behavior is strongly dependent on the impact velocity.

Figure 4.

Impact velocity influence on EPS60R50 and EPS80R50 in dynamic and quasi-static loading.

Table 4.

Impact velocity influence: p-values for all configurations. Values above the significance threshold (p < 0.05) are shown in bold.

	EPS60R50	EPS60R89	EPS80R50	EPS80R89
3 m/s vs 4 m/s	1e-5	7e-2	8e-17	3e-12
3 m/s vs 5 m/s	1e-2	1e-4	4e-16	4e-12
4 m/s vs 5 m/s	5e-3	2e-2	4e-2	0.24

To assess the influence of strain rate while maintaining it constant throughout compression, additional quasi-static tests were performed. Figure 4 illustrates this effect at three strain rates for EPS60R50 and EPS80R50 and confirms the influence of strain rate under quasi-static loading. Similar trends can be observed for EPS with an 89% recycling ratio.

Regarding the study of stress-strain curve parameters, Welch’s t-test indicates that the parameters, namely Young’s modulus, plateau slope, and yield point, are independent of impact velocity in dynamic loading. In contrast, the maximum point, unloading modulus, and permanent strain exhibit a significant dependency on velocity. These findings are visually confirmed in Figure 4, where the mean stress–strain curves overlap closely up to the densification region, followed by divergence during densification and the unloading phases. Under quasi-static loading conditions, the Young’s modulus is also independent of strain rate. However, the mean plateau stress and the maximum stress both increase with increasing strain rate. Finally, the unloading phase appears to be independent of strain rate. The differences observed between quasi-static and dynamic loading can be explained by the constant strain rate imposed by the Instron device, in contrast to drop tower tests, where the strain rate decreases during compression.

Finally, Figure 5 presents the mean stress–strain curves at different densities for EPS at 50% and 89% recycling ratio, demonstrating that density significantly influences the mechanical behavior of the foam.

Figure 5.

Density influence at 4 m/s impact velocity at 50% and 89% of recycling.

Higher densities are correlated with increased rigidity, impacting all key parameters. Specifically, Young’s modulus, plateau slope, and yield point tend to increase with density due to the greater amount of material per unit volume, which enhances stiffness. Conversely, the maximum point tends to decrease, while unloading modulus and permanent strain increase.

Discussion and comparison with classical EPS

This study aimed to characterize the behavior of recycled EPS under dynamic impact conditions. The experimental setup was specifically designed to simulate the impact of a helmeted head, with the sample height fixed at 25 mm to replicate the typical thickness of EPS liners used in helmets. While this configuration reflects realistic boundary conditions, it does not allow for the investigation of geometric effects. Future work could therefore examine the influence of foam morphology and sample thickness on mechanical behavior. Moreover, since the impact energy varied between tests due to changes in impact velocity at constant mass, further studies could focus on maintaining constant energy by adjusting the impactor mass. This would allow a more targeted assessment of the influence of recycling under fixed energy input and clearly separate the effects of varying energy or strain rate.

Further studies could also be undertaken to investigate the underlying microstructural changes associated with high recycling ratios. Optical and high-resolution imaging methods such as SEM or µCT would make it possible to characterize modifications in cell morphology, density gradients, fracture patterns, and potential defects introduced during reprocessing. Such analyses would provide valuable insight into how the cellular architecture of both classical and recycled EPS evolves with repeated processing and could help explain the mechanical trends observed in the present work.

The effect of temperature on the mechanical properties of EPS was investigated over the range of −20°C to 18°C. Statistical analysis revealed that, within this range, temperature does not significantly influence the foam’s behavior, consistent with observations for classical EPS. However, only cooling effects were only considered in this study, and no tests were conducted at elevated temperatures. Further research is therefore needed to assess the influence of higher temperatures on the mechanical response of recycled EPS.

Regarding the effect of the impact velocity, the mechanical parameters analysis has shown that the parameters (namely Young’s modulus, yield point, and plateau slope) are not significantly affected by impact velocity. In contrast, unloading-related parameters, such as the maximum point, unloading modulus, and permanent strain, exhibit a strong dependency on it. As the impact velocity increases, the input energy also increases, resulting in higher stress levels within the material. This elevated stress drives the foam further into the densification regime, particularly for recycled EPS. Despite these differences, the stress–strain curves obtained at lower velocities remain within the variability corridor of those at higher velocities, indicating consistent material behavior. These findings are in agreement with previous observations on classical EPS in dynamic conditions.

Density appears to be the primary factor influencing the mechanical behavior of recycled EPS. Lower-density foams are associated with reduced stiffness, as reflected by a lower plateau stress and a higher maximum strain. Conversely, higher-density foams exhibit increased rigidity, characterized by a higher mean plateau value and a lower maximum strain. This behavior can be attributed to the fact that denser foams reach the densification regime earlier during compression, resulting in a steeper stress response. Moreover, the increased stiffness contributes to a higher unloading modulus, indicating greater elastic recovery. However, the more constrained cellular structure also leads to increased permanent deformation, as it limits the material’s capacity to dissipate energy during unloading. These trends are consistent with previous findings reported for classical EPS.

The effect of recycling on EPS behavior appears to be dependent on foam density. For low-density foams, no significant statistical difference was observed between classical and recycled EPS, indicating comparable mechanical performance. However, for higher densities (above 80 kg/m³), recycling significantly alters the shape of the stress–strain response. Specifically, an increase in the recycling ratio is associated with a decrease in both Young’s modulus and the mean plateau stress, along with an increase in maximum strain. Although the overall energy absorption capacity remains similar, the distribution of this energy among the elastic, plateau, and densification regions differs. As a result, for a given stress level, recycled EPS exhibits higher strains than classical EPS. Conversely, at a fixed strain, classical EPS sustains higher stress levels, which suggests that it dissipates more energy during compression than its recycled counterpart.

Figure 6 illustrates the influence of recycling on the mechanical behavior of EPS, while Table 5 presents the p-values associated with each test configuration. The results indicate that recycling has no statistically significant effect on EPS60, as all p-values exceed the significance threshold (α = 0.05). At this density, the stress–strain curves of recycled EPS remain within the standard deviation corridor of the classical EPS, suggesting similar mechanical responses. In contrast, for EPS80, recycling is associated with a clear shift of the stress–strain curves toward higher strain values, indicative of a more compliant material behavior.

Figure 6.

Recycling effect: comparison between classical EPS and recycled EPS at 3 m/s impact velocity.

Table 5.

Recycling effect: p-values for each impact configuration. Values below the significance threshold (p < 0.05) are shown in bold.

	EPS60 vs EPS60R50	EPS60 vs EPS60R89	EPS80 vs EPS80R50	EPS80 vs EPS80R89
3 m/s	0.60	0.43	1e-5	1e-6
4 m/s	0.27	0.29	2e-2	2e-2
5 m/s	0.45	0.30	4e-2	7e-2

To quantify the relative change in EPS80 behavior induced by recycling in terms of stress–strain parameters, a dimensionless metric can be introduced to emphasize the influence of recycling on specific parameters, as well as their evolution with increasing recycling ratio. Let $p_{E P S R}$ denote the value of a given parameter for recycled EPS, and $p_{E P S}$ the corresponding value for classical EPS. The relative difference $d$ can then be defined as:

d = (1 - | \frac{p_{E P S R}}{p_{E P S}} |) * 100

(3)

This metric provides a simple and consistent way to evaluate the impact of recycling on key mechanical parameters. Figure 7 et Table 6 provides the mean relative difference depending on the recycling ratio. The mean value of the relative difference increases when the recycling ratio increases, proving that the observed effects is due to the recycling. Therefore, the following conclusions can be expressed.

(1) Young’s modulus decreases when the recycling ratio increases

(2) The yield stress decreases when the recycling ratio increases

(3) The mean plateau value decreases when the recycling ratio increases

(4) The maximum strain increases when the recycling ratio increases

Figure 7.

Recycling effect: relative difference for Young’s modulus, yield stress, mean plateau value, and maximum strain.

Table 6.

Mean relative difference for Young’s modulus, yield stress, mean plateau value, and maximum strain at various recycling ratio for EPS80.

	EPS80R50	EPS80R89
Young’s modulus	34% ± 13%	47% ± 10%
Yield stress	25% ± 7%	47% ± 7%
Mean plateau value	22% ± 6%	37% ± 7%
Maximum strain	9% ± 3%	15% ± 3%

Moreover, impact velocity does not appear to significantly influence the relative difference between classical and recycled EPS, suggesting that the effect of recycling remains consistent across different impact velocities. An exception is observed for the maximum strain at 3 m/s, which exhibits higher values compared to those at higher velocities. This indicates that, under low-velocity impacts, recycled EPS may undergo greater deformation. Nonetheless, the overall trend suggests that the influence of recycling is largely independent of impact velocity.

Since the recycling process affects certain regions of the stress–strain curves, as demonstrated by the parameter analysis, the overall absorbed energy was calculated for both classical and recycled EPS in order to assess whether recycling affects energy absorption capacity. Table 7 presents the absorbed energy for all impact configurations. This quantity was calculated by integrating the area under the loading portion of the stress–strain curve, as proposed by Avalle et al. (2001).⁶

Table 7.

Absorbed energy for classical and recycled EPS at various impact velocities.

Absorbed Energy (MJ/m³)	3 m/s	4 m/s	5 m/s
EPS80	0.61 ± 0.10	1.19 ± 0.15	1.62 ± 0.16
EPS80R50	0.59 ± 0.10	1.12 ± 0.10	1.58 ± 0.17
EPS80R89	0.61 ± 0.07	1.1 ± 0.10	1.47 ± 0.10

As expected, the results indicate that absorbed energy increases with impact velocity, with no significant difference observed between classical and recycled EPS. This suggests that recycled EPS is capable of absorbing a similar amount of energy as classical EPS, although through different deformation mechanisms along the curve.

More specifically, recycled EPS exhibits a reduced ability to dissipate kinetic energy in the early stage of the plateau region, as evidenced by the decrease in mean plateau stress. However, this effect is compensated by an increased deformation capacity compared to classical EPS, as indicated by the higher maximum strain.

Few studies have focused on the mechanical characterization of recycled EPS; however, works by Acierno et al.²⁰ and Barrera et al.²¹ provide some insights into the influence of recycling ratio on stress–strain curve parameters. Acierno et al. conducted quasi-static compression tests on low-density EPS (20 kg/m³) containing 20% to 40% recycled material. Their results showed a decrease in stress at 10% strain with increasing recycling content. In contrast, Barrera et al. reported a linear increase in Young’s modulus with the recycling ratio for EPS with a density of 25 kg/m³ and recycling ratios below 20%.

In the present study, no clear trend was identified in the elastic region for medium-density EPS. However, a statistically significant reduction in the slope was observed at higher densities. These contrasting observations could be attributed to differences in test conditions—namely the dynamic nature of the present tests compared to the quasi-static protocols used by Acierno and Barrera—as well as the very low EPS densities investigated in those earlier studies.

The present study focuses on the mechanical properties of recycled EPS compared with classical EPS. At this stage, the materials have not yet been evaluated under realistic impact or crash scenarios, for example through the evaluation of a helmet’s performance with and without recycled material. Consequently, it is not currently possible to predict the performance of recycled EPS in helmet applications, and this will be the subject of future work.

Future work will focus on evaluating the performance of recycled EPS under realistic impact or crash conditions, for instance through the assessment of helmet behavior with and without recycled material. An additional step will involve validating these results through application-oriented testing to better predict the suitability of recycled EPS for helmet applications.

Conclusion

In this study, the mechanical behavior of recycled expanded polystyrene (EPS) was evaluated in dynamic conditions for two recycling ratios (50% and 89%) and two material densities (60 kg/m³ and 80 kg/m³). The influence of temperature, impact velocity, and density was examined not only through the analysis of full stress–strain curves, but also by extracting key mechanical parameters. Recycled EPS was systematically compared to its classical counterpart.

The results demonstrate that within the investigated range, temperature has no significant effect on the mechanical response of recycled EPS. In contrast, both impact velocity and foam density significantly influence its behavior. Additionally, no meaningful difference was observed between recycled and classical EPS at 60 kg/m³. However, at 80 kg/m³, recycling induces a noticeable shift in the stress–strain response, particularly in terms of stiffness and strain at densification, suggesting that the effect of recycling becomes more pronounced at higher densities.

Footnotes

Acknowledgements

The authors would like to thank Atomic Austria GmbH and the DiMo-NEXT project. The COMET project DiMo-NEXT is funded by the Federal Ministry for Innovation, Mobility and Infrastructure (BMIMI), the Federal Ministry for Economy, Energy and Tourism (BMWET), and the provinces of Salzburg, Upper Austria and Tyrol within the framework of COMET– Competence Centres for Excellent Technologies. COMET is processed by The Austrian Research Promotion Agency (FFG).

ORCID iD

Valentin Rieth

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.*

References

Zhang

Kikuchi

, et al. Constitutive modeling of polymeric foam material subjected to dynamic crash loading. Int J Impact Eng 1998; 21: 369–386. https://doi.org/10.1016/s0734-743x(97)00087-0

Leng

Ruan

Tse

. Recent bicycle helmet designs and directions for future research: a comprehensive review from material and structural mechanics aspects. Int J Impact Eng 2022; 168: 104317. https://doi.org/10.1016/j.ijimpeng.2022.104317

Bhudolia

Gohel

Leong

. Impact performance of innovative corrugated polystyrene foam for bicycle helmets. J Cell Plast 2020; 6: 0021955X20965215.

E Rodríguez-Sánchez

Plascencia-Mora

. Modeling hysteresis in expanded polystyrene foams under compressive loads using feed-forward neural networks. J Cell Plast 2023; 59(Issue 4): 269–292. https://doi.org/10.1177/0021955x231174362

Kroeker

Bonin

DeMarco

, et al. Age does not affect the material properties of expanded polystyrene liners in field-used bicycle helmets. J Biomech Eng 2016; 138: 041005. https://doi.org/10.1115/1.4032804

Avalle

Belingardi

Montanini

. Characterization of polymeric structural foams under compressive impact loading by means of energy-absorption diagram. Int J Impact Eng 2001; 25: 455–472. https://doi.org/10.1016/s0734-743x(00)00060-9

Di Landro

Sala

Olivieri

. Deformation mechanisms and energy absorption of polystyrene foams for protective helmets. Polym Test 2002; 21: 217–228. https://doi.org/10.1016/s0142-9418(01)00073-3

Ouellet

Cronin

Worswick

. Compressive response of polymeric foams under quasi-static, medium and high strain rate conditions. Polym Test 2006; 25: 731–743. https://doi.org/10.1016/j.polymertesting.2006.05.005

Chen

Hao

Hughes

, et al. Static and dynamic mechanical properties of expanded polystyrene. Mater Des 2015; 69: 170–180. https://doi.org/10.1016/j.matdes.2014.12.024

10.

Avalle

Frascio

Monti

. An improved model to describe the repeated loading-unloading in compression of cellular materials. Procedia Struct Integr 2018; 12: 19–31. https://doi.org/10.1016/j.prostr.2018.11.110

11.

Choi

B-H

Chudnovsky

Sehanobish

. Modeling of failure mechanisms using the true stress-strain response of polymeric foam materials. Procedia Eng 2011; 10: 1609–1614. https://doi.org/10.1016/j.proeng.2011.04.269

12.

Rieth

Yazman

Meyer

, et al. EPS foam characterization under compressive dynamic loading: experimental and mathematical approaches. J Cell Plast 2025; 5: 0021955X251357242.

13.

Cadete

Gomes

Carvalho

, et al. Polymeric foams from recycled thermoplastic poly(ethylene terephthalate). J Cell Plast 2021; 57: 609–622. https://doi.org/10.1177/0021955x20948562

14.

Guillén-Mallette

González-Chi

Cruz-Estrada

, et al. Recycling printed polypropylene labels and polyolefins caps as chemical foaming agent to produce foam products. J Cell Plast 2021; 57: 733–756. https://doi.org/10.1177/0021955x20959302

15.

Jolfaei

Gavgani

Jalali

, et al. Effect of organoclay and compatibilizers on microstructure, rheological and mechanical properties of dynamically vulcanized EPDM/PP elastomers. Polym Bull 2015; 72: 1127–1144. https://doi.org/10.1007/s00289-015-1328-1

16.

Kheradmandkeysomi

Salehi

Jalali

, et al. Achieving super-tough high-density polyethylene with promising foamability using silane crosslinked polyolefin elastomer nanofibrils network. Compos Sci Technol 2024; 251: 110576. https://doi.org/10.1016/j.compscitech.2024.110576

17.

Nofar

Yenigul

Ozdemir

, et al. Mechanical and viscoelastic properties of polyethylene‐based microfibrillated composites from 100% recycled resources. J Appl Polym Sci 2021; 138: 50793. https://doi.org/10.1002/app.50793

18.

Kheradmandkeysomi

Salehi

Jalali

, et al. Enhancing mechanical performance of HDPE at different environmental conditions with outstanding foamability through in situ rubber nanofibrillation: exploring the impact of interface modification. IEEE.

19.

Kan

Demirboğa

. A new technique of processing for waste-expanded polystyrene foams as aggregates. J Mater Process Technol 2009; 209: 2994–3000. https://doi.org/10.1016/j.jmatprotec.2008.07.017

20.

Acierno

Carotenuto

Pecce

. Compressive and thermal properties of recycled EPS foams. Polym-Plast Technol Eng 2009; 49: 13–19. https://doi.org/10.1080/03602550903282994

21.

Barrera Castro

Olaya Florez

Ocampo Carmona

. Production and characterization of the mechanical and thermal properties of expanded polystyrene with recycled material. I&U:EfD; 21: 113. https://doi.org/10.11144/javeriana.iyu21-2.pcmt

22.

Hornberger

Might

Mohareb

, et al. Dynamic properties of recycled EPS foam. Polym-Plast Technol Eng 1997; 36: 917–930. https://doi.org/10.1080/03602559708000670

23.

Desquilbet

. Guide pratique de validation statistique de méthodes de mesure: répétabilité, reproductibilité, et concordance. WOAH, 2010.

24.

Yuan

K-H

. Welch’s t test. TDS Archive, 2010, pp. 1620–1623.

25.

Ladner

Abdelaziz

. Série de Fiches Méthodologiques en Recherche et en Rédaction Scientifique Fiche Méthodologique n°1/2021: comment calculer et interpréter la valeur de «p» dans une étude épidémiologique How to calculate and interpret the p-value in an epidemiological study? LA TUNISIE MEDICALE; 99: 1357242.