Enhancing mechanical performance of polystyrene carbon composite foams through supercritical CO 2 foaming: An experimental study

Abstract

Conventional polystyrene (PS) foams are widely used in packaging and insulation, but suffer from limited mechanical strength, which restricts their use in load-bearing applications. This study investigates the enhancement of the mechanical performance of PS foam through the incorporation of graphene nanoplatelets (GNP) and flaked graphite (FG), processed via supercritical CO₂ (sc-CO₂) extrusion foaming at two pressures (17.3 MPa and 20.6 MPa). The influence of sc-CO₂ pressure on additive dispersion and mechanical behavior was evaluated using compression testing, dynamic mechanical analysis (DMA), micro-computed tomography (micro-CT). Results showed that increasing the sc-CO₂ pressure significantly enhanced compressive strength from 0.30 MPa to 0.40 MPa for 0.75 wt% GNP foams (34% increase) and from 0.30 MPa to 0.50 MPa for 0.75 wt% FG foams (66% increase). Enhanced storage and loss moduli in DMA confirmed improvements in stiffness and energy dissipation. Micro-CT imaging revealed more well-defined closed-cell structures and uniform carbon particle dispersion at the higher pressure. Overall, these findings emphasize the importance of pressure-optimized sc-CO₂ foaming as an effective strategy for producing lightweight, durable PS-carbon composite foams suitable for structural, insulation, and packaging applications.

Introduction

Polystyrene (PS) foams have long played a critical role in modern materials applications, particularly in packaging, ¹ cushioning, ² thermal insulation, ³ and sound insulation ⁴ due to their lightweight nature, ease of processing, and low cost. ⁵ In particular, the demand for PS foams as insulation in buildings is rising due to the ability to reduce the Greenhouse Gas (GHG) emissions of buildings while reducing costs. ⁶ However, conventional PS foams often exhibit limitations in mechanical strength, restricting their use in demanding environments.^7,8 Enhancing the mechanical performance of PS foams is thus a critical area of research with goals of expanding their applicability while maintaining cost-effectiveness.

PS foams have been reinforced with carbon additives such as graphene nanoplatelets (GNP), flaked graphite (FG), and carbon nanotubes (CNT) to improve both thermal insulation and mechanical performance. These carbon-based additives, often used in combination with talc, serve not only as reinforcements to improve mechanical properties, but also as heterogeneous nucleating agents that influence foam cell morphology. Previous studies have demonstrated that the incorporation of carbon-based additives such as carbon nanofibers, activated carbon, graphene, and graphite into PS foams can significantly enhance their mechanical properties, particularly compressive strength.^9–13 However, these additives tend to agglomerate, causing a significant challenge in achieving uniform dispersion within the polymer matrix. Driven by van der Waals forces, such agglomeration often compromises the expected enhancements in properties. However, when uniformly dispersed, recent studies have shown that these additives consistently improve mechanical strength and thermal insulation in polymer composites.^14–16

Improved dispersion of fillers also has been shown to enhance mechanical performance in polymer foams by promoting uniform cell morphology, increased hardness, and higher tensile strength, modulus, and impact resistance.^17,18 For instance, Zhao et al. ¹⁹ examined the dispersion of CNT in a polypropylene (PP) matrix and found that uniform CNT dispersion significantly enhanced the compressive strength (from 0.18 MPa to 0.57 MPa) and modulus (from 0.52 MPa to 1.74 MPa) of microcellular PP foams by refining cell morphology and improving crystallization and melt strength. Similarly, Zhang et al. ²⁰ investigated the effect of hybrid filler dispersion in polymethyl methacrylate (PMMA) foams using multi-wall carbon nanotubes (MWCNT) and GNP, and found that improved exfoliation and uniform dispersion, facilitated by sc-CO₂ processing, significantly enhanced the tensile strength (up to 34%) and Young’s modulus (up to 57%) of the composite foams due to stronger filler–matrix interactions and synergistic stress transfer. Caglayan et al. ²¹ optimized CNT dispersion in polyurethane (PU) foams and found that 0.1 wt% functionalized CNT improved compressive strength by 13% and enhanced sandwich composite flexural strength by over 30%. Wu et al. ²² investigated the effect of CNT dispersion methods on the mechanical properties of polybutylene succinate nanocomposites and found that using a secondary dispersion technique significantly improved CNT uniformity, leading to enhanced compressive strength and more uniform cell structures in the resulting foams.

While extensive literature exists on the hetero-nucleating role of carbon in PS foams, few studies have explored how sc-CO₂ influences the dispersion of carbon additives and the resulting mechanical properties of these PS foams. While most prior studies have relied on batch foaming to investigate PS-carbon systems, providing valuable mechanistic insights, these methods fall short of capturing the conditions relevant to continuous industrial processing. Our previous work proposed a scalable PS-carbon composite foam synthesis strategy that demonstrated significant improvements in cell morphology and thermal insulation by leveraging sc-CO₂-assisted extrusion foaming. ²³ Building on this, the current study extends that approach to systematically investigate how sc-CO₂ pressure and carbon additive type affects the mechanical performance, specifically compressive and viscoelastic properties of PS carbon composite foams produced on a pilot-scale extrusion line under industrially relevant conditions.

In this study, we investigate the combined effect of sc-CO₂ pressure and carbon additive incorporation on the mechanical performance of PS foams. Specifically, we examine the influence of two sc-CO₂ pressures (17.3 and 20.6 MPa) on the dispersion of GNP and FG within the PS matrix, and their impact on compressive strength and energy absorption. A comprehensive characterization approach combining compression testing, dynamic mechanical analysis (DMA), and high-resolution micro-computed tomography (micro-CT) is employed to assess the additive distribution and compression performance. The novelty of this work lies in the systematic evaluation of pressure-driven dispersion effects with two distinct carbon additives, supported by experimental analysis on a pilot foaming extruder. Furthermore, the study provides quantitative insight into how optimized sc-CO₂ pressure can simultaneously enhance additive distribution, mechanical and viscoelastic performance. These findings contribute to the advancement of high-performance, lightweight PS-based composite foams and offer a scalable, environmentally acceptable approach for tuning foam properties via sc-CO₂ extrusion foaming.

Experimentation

Materials

Polystyrene PS 595T with a Melt Flow Index (MFI) of 1.6 g/10 min (200°C, 5 kg) and a density of 1.04 g.cm⁻³ was purchased from TotalEnergies Petrochemicals and Refining USA. The blowing agent CO₂ (99.9% pure) was supplied by Linde Canada. Talc powder (JetWhite 1HC, median particle size 1.1 micron) was kindly donated by Magris Talc USA. Graphene nanoplatelets (GNP) XFQ021 (diameter 5-10 µm, thickness 3 – 10 nm and number of layers 10 -15) was supplied by Xi’an Lyphar Biotech Co., Ltd, Xi’an, China. Flaked Graphite (FG) with a median diameter 7 -10 µm and 99% carbon basis was purchased from Alfa Aesar, Canada.

Extrusion foaming

Extrusion foaming of PS or PS-carbon particles was carried out using a pilot twin-screw extruder (Feininger Model: SHJ-Z36*25, D = 36 mm, L/D = 25, output = 3 kg/hr) by pumping the blowing agent (sc-CO₂) through a gas/liquid injection port located at L/D = 10 from the hopper of the extruder. The extruder consisted of nine zones, with zone 1 starting from after the hopper and ending with zone 9 at the slot die. The extrusion foaming was done in two steps. In the first step (primary extrusion), compounding of PS carbon and talc occurred at high temperatures (feed zone: 190°C, compression zone: 195°C, metering zone: 190°C). Foaming was carried out in the second step (secondary extrusion) at low temperatures (feed section: 126°C, compression section: 195°C, metering section: 140°C). During the extrusion foaming process, the screw speeds of the feeder and extruder were both kept at 50 rpm. Sc-CO₂ was injected into the extruder at two pressures (17.3 and 20.6 MPa) using a syringe pump (Model 260 D) to maintain the supercritical conditions inside the extruder. 1 wt% talc powder was added in all the foam composite samples to enhance the dimensional stability and achieve maximum extruder throughput. ²⁴ Virgin PS was extruded at the sc-CO₂ pressures to produce pristine PS foam samples. GNP and FG were mixed with PS at 0.5, 0.75, 1 and 1.5 wt% (weight/weight).

Characterization

Density and morphological analysis

Foam density was calculated using ASTM D1622. Three specimens were tested for each composite foam sample, and the average value is reported. The test specimens were rectangular in shape (23 x 8 x 7 mm). The foam skin was removed prior to the measurements, and an average of three readings was considered for the final reported value. Total porosity (void fraction) of the foam was calculated from

$\text{Total}\hspace{0.17em}\text{porosity}=1-\frac{{\rho }_{foam}}{{\rho }_{solid}}$

(1)

where ${\rho }_{foam}$ is the density of the foam and ${\rho }_{solid}$ is the density of solid polymer composite.

Morphological analysis of the compressed PS composite foams was performed using a Scanning Electron Microscope (SEM, Hitachi Flex SEM 1000 II) operating at a voltage of 5 kV. The morphologies of the foam were observed along the direction perpendicular to the extrusion at the center of foam width after the removal of skin. Fiji (ImageJ) software was used for cell size analysis, utilizing SEM pictures with a magnification ratio of 50 times.

Mechanical properties

Viscoelastic properties of PS-carbon composite foam (storage, loss modulus and tan δ) were studied using Dynamic Mechanical Analyzer (TA-Q800) in the temperature range of 40 to 140°C at a heating rate of 3.0°C min⁻¹ and a fixed frequency of 1 Hz. The sample dimensions were 17 x 11 x 5 mm.

Strain-controlled compressive strength tests of the foams were studied using a vertical compression setup (Figure 1), consisting of a load cell (1 kN), controller panel, and upper/lower platens (Instron Model 5943). The compressive modulus was calculated from the slope of the curve in the linear elastic region of the stress–strain curve and the compressive strength at 10% strain. Seven specimens were tested for each composite foam sample and the average value is reported. The test specimens were rectangular in shape (16 x 16 x 7 mm). Each specimen was subjected to an axial compression pre-force of 2N. The composite foam samples were placed between the parallel compression platens of the machine and compressed at a constant crosshead speed of 0.7 mm min⁻¹ until a compression strain of 50% was reached. The tests closely followed ASTM D1621, except for the dimensions of test specimens required for the compression testing; dimensions of test specimens were smaller than recommended due to limitations of the thickness of the extruded foams. All stress-strain data represent true stress and true strain values calculated according to ASTM D1621. True stress and true strain were calculated from the following relations:

$\text{True}\hspace{0.17em}\text{stress}=\frac{P}{A_{\text{inst}}}$

(2)

${\hspace{0.17em}A}_{\text{inst}}=A_0.\frac{h_0}{h}$

(3)

$\text{True}\hspace{0.17em}\text{strain}=\ln \left(\frac{h_0}{h}\right)$

(4)

where $P$ is the load, $A_{\text{inst}}$ is the instantaneous cross-sectional area of the specimen at a given deformation level, $A_0$ and $h_0$ are the initial cross-sectional area and height of the specimen, and $h$ is the instantaneous height obtained from crosshead displacement. Specific compressive modulus or strength were calculated as the ratio of compressive modulus or strength to the foam density.

Figure 1.

Schematic of the compression setup used for evaluating PS composite foam specimens

X-ray micro-computed tomography (micro-CT)

The samples were trimmed into rectangular pieces that were suitable for micro-CT analyses. The micro-CT was a Zeiss Xradia 410 Versa and the acquisition parameters are listed in Table 1. Micro-CT involves passing X-rays through a sample volume and detecting the intensities of the transmitted X-rays. Variations in the sample volume density, such as voids or cracks, cause variations in X-ray absorption and are expressed as differing greyscale values in the projected image (X-ray radiometric projection image). Multiple projection images are captured at different angles and are subsequently reconstructed into a 3D dataset. Following the acquisition of the datasets, each was processed in ORS’s Dragonfly Pro software and presented as 2D slices. Table S1 lists all the conditions used for micro-CT analysis of each sample.

Table 1.

Effect of sc-CO₂ pressure and carbon loading on the apparent density and void fraction of PS composite foams.

Loading (wt%)	Apparent density (kg/m3) ASTM D1622
	GNP		FG
	17.3 MPa	20.6 MPa	17.3 MPa	20.6 MPa
0	101 ± 10	113.9 ± 12	101 ± 10	113.9 ± 12
0.5	124 ± 9	142 ± 10	94 ± 9	115 ± 10
0.75	98 ± 8	137 ± 11	96.7 ± 9	160 ± 10
1.0	166 ± 10	142 ± 10	106.7 ± 10	108 ± 10
1.5	166 ± 10	146 ± 11	147 ± 11	122 ± 9

Results and discussion

Foam density studies

Density is the most important parameter that affects the mechanical properties of polymer foams. ²⁵ Apparent densities were measured according to ASTM D1622 for PS composite foams containing varying loadings of GNP and FG at sc-CO₂ pressures of 17.3 MPa and 20.6 MPa. The role of sc-CO₂ pressure in the dispersion of carbon particles (GNP and FG) in the PS matrix can be inferred from the foam density results (Table 1). The values of apparent density are represented as mean ± standard deviation. For low to moderate carbon loadings (0.5–0.75 wt%), the foam density results suggested that, at lower pressure (17.3 MPa), lower foam density was obtained due to better cell expansion. However, at a higher pressure (20.6 MPa), a significant increase in foam density, particularly in FG-filled foams (160 kg/m³ at 0.75 wt% FG), suggested that the higher pressure may have induced localized particle agglomeration, which then restricted bubble growth resulting in a denser foam structure. ²⁶ As shown in Table S2, the total porosity remained high (0.84 - 0.90) across all carbon loadings and both the sc-CO₂ pressures. Measurements from multiple regions of each specimen varied by less than 5 %, confirming that the foams are homogeneous without detectable porosity gradients.

At higher GNP loadings (1-1.5 wt%), the increased foam density at both pressure levels suggested that excessive additive content, combined with high pressure, may have led to particle agglomeration, limiting the polymer’s ability to expand. Nevertheless, at the optimal loading of 0.75 wt% under high pressure, both GNP and FG samples exhibited superior compression properties. In GNP-reinforced samples at 17.3 MPa, density remained high (166 kg/m³) suggesting that GNP particles might be forming percolated networks, consequently reducing overall foam expansion. Meanwhile, FG-reinforced foams showed a less dramatic density increase at higher loadings, which indicated better dispersion stability under elevated pressure conditions. ²⁷ Other researchers have observed an enhancement in the mechanical performance as the foam apparent density increased.^25,28

Compression properties

Thermal insulation materials are often subjected to compressive loads in building applications and therefore must be tested for their elastic response under compression. ¹⁹ To evaluate the relationship between cellular structure and compressive performance, compression tests were conducted on the foamed composites. Figure 2 shows the compressive stress–strain curves for PS composite foams with varying carbon additive loadings at the two sc-CO₂ pressures. According to Ashby et al., ²⁹ a typical compressive stress–strain response of polymeric foams includes three distinct regions: (1) a linear elastic region (cells deform elastically), (2) a plateau region where cell collapse occurs (plastic deformation), and (3) a densification region marked by a steep rise in stress. These regions are clearly distinguishable in the curves in Figure 2 with the transition from linear elastic to plateau indicating the compressive strength measured at 10% strain. The slope in the initial region corresponds to the compressive modulus.

Figure 2.

Stress-strain curves for PS composite foam with varying carbon loadings at two sc-CO₂ pressures in the extruder.

For both GNP- and FG-reinforced foams, an increased resistance to deformation was observed at lower additive loadings (0.5 and 0.75 wt%) under high-pressure foaming conditions (20.6 MPa). This enhanced stiffness is likely due to improved dispersion of carbon additives, which in turn contributed to higher apparent densities (as shown in Table 1). Our previous findings demonstrated that the carbon additives (GNP/FG) and elevated sc-CO₂ pressure had a pronounced influence on foam morphology. Specifically, 0.75 wt% GNP or FG at 20.6 MPa led to up to a two-fold increase in cell density due to the generation of smaller and more uniform cells. ²³ This enhanced microstructure is critical for improving load-bearing capacity, as uniform and finer cell structure facilitated more efficient stress transfer and contributed to the observed improvements in stiffness and compressive strength.^30,31

Figure 3 compares compression moduli and strength of PS composite foams at varying loadings of carbon and the two sc-CO₂ pressures. For both the GNP and FG, modulus and strength improved with increasing loadings at both pressures. This is due to the presence of carbon particles in the PS matrix and a reduction in the cell size with an increase in carbon loading due to its hetero-nucleating effect. ¹⁰ The compression properties increased with the loadings up to 1 wt% and then decreased at 1.5 wt% for GNP, while for FG the properties were enhanced linearly up to 1.5 wt%. The absolute compressive properties in Figure 3 oscillated with increasing additive loading rather than showing a strictly monotonic increase. This trend can be rationalized using simple micromechanics of cellular solids. According to the Gibson-Ashby model, ³² the modulus of a foam scales with its relative density as:

$\frac{E^{*}}{E_s}\propto {\hspace{0.17em}\left(\frac{{\rho }^{*}}{{\rho }_s}\right)}^n$

(5)

where $E^{*}$ is the foam modulus, $E_s$ is the modulus of the solid polymer, ${\rho }^{*}$ is the foam density, ${\rho }_s$ is the solid density, $C$ is a constant of order unity, and $n$ is typically between 1.5 and 2 depending on the deformation mechanism. The exponent $n$ was obtained from the slope of a linear regression of $\ln \left(E^{*}/E_s\right)$ versus $\ln \left({\rho }^{*}/{\rho }_s\right)$ (Figure S1). A larger $n$ indicated greater sensitivity of modulus to relative density, meaning that small differences in foam porosity produced proportionally larger changes in modulus. For foams processed at 17.3 MPa sc-CO₂ pressure, $n$ was approximately 1.3-1.5, consistent with an intermediate stretch-bending dominated mechanism. In contrast, foams processed at 20.6 MPa exhibited a higher $n$ value of about 2.6, suggesting a likely bending-dominated deformation and stronger density dependence. Thus, even minor variations in relative density at 20.6 MPa could lead to noticeable oscillations in absolute compressive modulus. After normalization by density, however, the specific modulus and strength displayed a more monotonic trend, revealing the intrinsic reinforcing effect of GNP and FG fillers. Therefore, the apparent oscillations in absolute properties were attributed primarily to small variations in foam density and microstructural heterogeneity rather than the absence of filler reinforcement.

Figure 3.

Compression properties of PS composite foam at varying carbon loadings at two sc-CO₂ pressures in the extruder.

The PS- 0.75 wt% GNP and FG foams produced at 20.6 MPa sc-CO₂ pressure exhibited a significant rise (34% and 66% for GNP and FG respectively) in absolute compressive strength as compared to 17.3 MPa due to the presence of more carbon particles in the PS matrix as well as smaller mean cell sizes. The PS-0.75 wt% GNP and FG foams, processed at 20.6 MPa sc-CO₂ pressure, showed a substantial increase in specific compressive modulus, approximately 40% for GNP and 22% for FG compared to 17.3 MPa. The reduction in the cell size for the foams produced at high sc-CO₂ pressure ²³ induced more resistance in deforming the cell walls and struts, which led to an increase in compression modulus and strength. ³³

The composite foam with GNP showed higher compression properties than the foam with FG because the high aspect ratio of GNP led to better interfacial bonding with the polymer matrix to improve the stiffness.^34,35 At higher GNP loading (1.5 wt%), a reduction in compressive strength at 20.6 MPa may be attributed to the formation of additive aggregates which act as stress concentration sites compromising structural integrity. ³⁶ Conversely, samples produced at 17.3 MPa with higher additive content (1.5 wt%) demonstrated increased compressive strength, likely due to their higher apparent densities (Table 1) compensating for any dispersion limitations. Other researchers also observed a decrease in the compression modulus at high loadings of additives and attributed this to a less uniform dispersion in the polymer matrices.^37,38 Thus, PS composite foam with enhanced compression properties can be obtained at a high sc-CO₂ pressure in the extruder. This synergistic approach is promising for developing lightweight structural foams with enhanced mechanical properties.

Compression deformation mechanisms

After compression testing, foams were cut in the axial direction and SEM was used to observe the failed microstructures to provide critical insights into their deformation mechanisms. Under compressive load, PS foams typically exhibited a sequence of deformation processes including cell wall bending, buckling, and eventual fracture. The post-compression SEM micrographs (Figure 4) reveal plastically deformed struts with characteristic hinge formations and fractured regions, marking the transition from elastic deformation to plastic collapse.

Figure 4.

The axial section of SEM images of cell morphology of PS composite foams at two sc-CO₂ pressures after compression testing.

More uniform bending and deformation in 0.75 wt% GNP and FG foams indicated efficient stress transfer due to better additive dispersion and cell connectivity.^39,40 At 20.6 MPa, foams showed more coordinated deformation compared to 17.3 MPa, where irregular and localized collapse was observed. In contrast, at 1.5 wt% loading, higher compressive strength and modulus were achieved at 17.3 MPa, where homogeneously bent cells supported better energy absorption, while high-pressure samples exhibited more intact cells likely due to stress buildup from agglomeration. ⁴¹ These findings highlight that both the additive content and sc-CO₂ pressure critically influenced mechanical reinforcement and deformation mechanisms in carbon-reinforced PS foams.

Energy absorption

The energy absorption study of PS foams is of utmost importance for cushioning applications. For PS foams, most of the energy is absorbed in the plateau region (Region (2) due to cell walls and struts deformation leading to its irreversible crushing. ⁴² The energy absorbed (at 42% strain) by PS composite foams with different loadings of carbon and sc-CO₂ pressure is shown in the Figure S2. Energy absorption of the sample was evaluated by calculating area under the stress-strain curve from the compression test equation (2):

$E={\int }_0^{\epsilon }\sigma \left(\epsilon \right)\hspace{0.17em}d\epsilon$

(6)

where E is the energy absorbed per unit volume, ε is the compressive strain, σ is the compressive strength at 50% strain: (here ε = ε_42%). To compare foams with different densities, the specific energy absorption (SEA) was calculated by normalizing the absorbed energy by the foam density $\left(\rho \right)$ .

$SEA=\frac{E}{\rho }$

(7)

The SEA was found to be sensitive to both cell morphology and the local effectiveness of filler - matrix stress transfer. From Figure 5, two samples: 0.75 wt% GNP at 20.6 MPa and 0.5 wt% FG at 20.6 MPa, exhibited notably higher SEA. This was interpreted as a synergistic effect of (i) cell structure with small cells and higher cell density that was generated at the higher sc-CO₂ pressure as reported in our previous study ²³ and (ii) efficient reinforcement and stress transfer at those particular loadings. High-cell-density foams provided a larger total area of cell walls/struts per unit volume, which increased the number of deforming elements that absorbed energy through bending, buckling and progressive crushing; they also delayed densification so that the plateau regime (where most energy was absorbed) was extended, increasing the integrated energy. At the same time, well-dispersed carbon fillers (GNP or FG) stiffened cell walls and improved load transfer from the polymer to the filler, which raised the plateau stress and therefore SEA. Conversely, at higher loadings where agglomeration was more likely, local inhomogeneity reduced effective stress transfer and sometimes promoted premature local collapse, lowering SEA despite higher nominal filler content. Thus, the observed peaks at 0.75 wt% GNP and 0.5 wt% FG at 20.6 MPa were consistent with an optimal combination of smaller cells and uniform filler dispersion that maximized the area under the plateau region per unit mass. Other researchers have also reported improved toughness and energy absorption of polymer foams with the addition of various carbon additives^43,44

Figure 5.

Specific energy absorption (SEA) of (a) PS-GNP and (b) PS-FG composite foams processed at 17.3 MPa and 20.6 MPa sc-CO₂ pressure.

Viscoelastic properties

Dynamic Mechanical Analysis (DMA) of PS foams reinforced with GNP and FG processed at the two pressures is represented in Figure 6. This study revealed significant variations in viscoelastic behavior as a function of sc-CO₂ pressure and carbon additive content. An evident trend observed across all samples is the enhancement in storage modulus (E′) with increasing sc-CO₂ pressure at optimum carbon loading. Specifically, 1.5 wt% GNP and 0.75 wt% FG foam processed at 20.6 MPa exhibited higher E′ values compared to their 17.3 MPa counterparts. This enhancement is attributed to a better interfacial adhesion between uniformly dispersed carbon particles and the PS matrix thereby enhancing the stiffness of the composite material.^45–47 Furthermore, the variation in the loss modulus (E″) peak height across different samples provided additional insight into their damping behavior. At 17.3 MPa, the highest loss modulus (E″) was observed at 1.5 wt% loading for both GNP and FG. Increasing the pressure to 20.6 MPa led to a marked enhancement in E″ for 1.5 wt% GNP. A similar trend was observed for FG, where 0.75 wt% at 20.6 MPa showed higher E″ compared to the same loading at lower pressure. The broader and higher E″ peaks in FG-reinforced foam samples at elevated pressure suggest enhanced energy dissipation due to improved polymer–additive interaction and interfacial friction.^48–50

Figure 6.

DMA results of PS-GNP and PS-FG composite foams at two sc-CO₂ pressures in the extruder.

Tan δ curves of the foam samples (ratio of E″/E) were plotted to understand the effect of carbon additives on the glass transition temperature (T_g); an important observation was the upward shift in the T_g. The higher glass transition temperatures for foams produced at the higher pressure reflected the restricted molecular mobility of the PS chains. This shift is associated with more efficient dispersion of carbon particles and increased polymer-additive interactions, which hinder segmental motion and raise the thermal stability of the composites. ⁵¹ The improved storage modulus, elevated T_g, and enhanced loss modulus at 20.6 MPa confirmed the potential of high-pressure sc-CO₂-assisted extrusion in producing mechanically robust and thermally stable polymer foams.

Microstructural analysis

Micro-CT imaging was utilized to analyze the internal cell wall structure, porosity, pore size distribution and carbon additive distribution in 1 wt% GNP and FG reinforced PS foams processed at 17.3 and 20.6 MPa. From Figure 7, the GNP sample processed at 20.6 MPa exhibited smaller mean cell size (127 μm) and more homogeneous closed-cell morphology with well-defined cell walls and uniformly dispersed GNPs (white arrows), suggesting even distribution and integration of the additives into the polymer matrix. In contrast, the 17.3 MPa GNP foam displayed irregular cell sizes (ranging from 50 to 245 μm) and visible additive agglomerates (white circles), which can act as stress concentrators and compromise structural integrity. A similar trend was observed in FG foams, where the 20.6 MPa sample showed elongated, aligned cells and better-distributed FG particles. Conversely, the 17.3 MPa FG foam revealed more isotropic cells with multiple regions of FG agglomeration. These clusters can disrupt foam structure and hinder uniform stress transfer, leading to decreased mechanical performance. Micro-CT analysis provided quantitative information on the porosity and pore size distribution of the foams (Figure 8). The total porosity values obtained from micro-CT were consistent with those estimated from density measurements, differing by less than ±10%. Increasing the sc-CO₂ pressure from 17.3 MPa to 20.6 MPa resulted in higher overall porosity and a narrower pore size distribution, indicating the formation of smaller and more uniform cells. ⁵² However, the open-cell content decreased at higher pressure (e.g., from 37.5% to 33.3% for pristine PS and from 23.6% to 18.3% for 1 wt% GNP foams), suggesting that enhanced gas saturation and rapid solidification at elevated pressure could have prevented the pore rupture, possibly due to increased melt viscosity of PS.^53,54 Quantitative directionality analysis was performed using the directionality plugin in Fiji/ImageJ ⁵⁵ to assess the predominant orientation of cell walls relative to the foaming direction. In this context, directionality was defined as the predominant orientation of cell walls relative to the foaming direction and was quantified from orientation histograms generated in ImageJ. The directionality analysis of the foam microstructures was performed using Fourier-based orientation distributions of the cell walls to quantify the degree of morphological anisotropy.

Figure 7.

Micro-CT images of 1 wt% GNP and FG PS foams at 17.3 and 20.6 MPa sc-CO₂ pressure.

Figure 8.

Pore size distribution (from micro-CT analysis) of PS composite foams: (a) 1 wt% GNP at 17.3 MPa, (b) 1 wt% GNP at 20.6 MPa, (c) 1 wt% FG at 17.3 MPa, and (d) 1 wt% FG at 20.6 MPa sc-CO₂ pressure.

The directionality plugin produced normalized histograms showing the fraction of cells oriented between 0° and 180° with a bin size of 1°. This analysis was described in detail by Tinevez et al. ⁵⁵ The extracted parameters (i) direction-representing the center of the orientation peak (ii) dispersion-indicating the standard deviation, (iii) amount-indicating the fraction of features concentrated near the peak, and (iv) goodness-representing the quality of the Gaussian fit. These parameters were used to characterize the dominant cell alignment and its statistical distribution. A higher amount and goodness, combined with a lower dispersion, denoted a well-aligned, anisotropic structure, whereas lower values indicated a more isotropic morphology with randomly oriented cells. Figure S2 shows the orientation histograms obtained from the directionality analysis of PS foams. For the 1 wt% GNP foams, a distinct orientation peak was observed at approximately −12° for the 17.3 MPa sample, accompanied by a low dispersion (25.7°), high amount (0.68), and strong goodness of fit (0.94). This combination signified a highly anisotropic cellular network with preferential alignment near the loading direction. At 20.6 MPa, the GNP foam maintained moderate anisotropy (amount = 0.60, goodness = 0.83) but exhibited a shift in dominant orientation to approximately 52°, suggesting partial cell-wall reorientation or distortion under higher sc-CO₂ pressure. In contrast, the 1 wt% FG foams exhibited much lower orientation parameters, with amount values of 0.25 and 0.23 and goodness values of 0.12 and 0.38 at 17.3 MPa and 20.6 MPa, respectively. The absence of a sharp orientation peak confirmed that both FG foams were largely isotropic. Increasing the sc-CO₂ pressure from 17.3 MPa to 20.6 MPa produced only a slight enhancement in orientation detectability (reflected in the increased goodness value), suggesting minor cell-wall alignment or shape anisotropy at higher pressure. Overall, the structure of the FG foams remained predominantly isotropic compared with the GNP system. Clemons et al. ⁵⁶ similarly reported that flat orientation histograms corresponded to isotropic microstructures, whereas well-defined peaks indicated preferred orientation. Hadad et al. ⁵⁷ further observed that low goodness values in directionality analysis of micro-CT images were characteristic of isotropic structures, consistent with the current findings. These results demonstrated that the incorporation of GNP promoted pronounced cell alignment and anisotropic morphology, whereas FG yielded a more uniform, isotropic cellular structure even under elevated sc-CO₂ pressures. The enhanced directional order in the GNP foams likely could have contributed to their superior mechanical performance.

A mechanism proposed in our earlier work explains that higher sc-CO₂ pressure improves the dispersion of carbon additives by enabling CO₂ to adsorb on the defect sites on GNP/FG, promoting layer delamination and thus ensuring more uniform carbon distribution during foaming. ²³ George et al. ⁵⁸ also observed that increased filler loading led to agglomeration of biocarbon particles in the polypropylene matrix, as revealed by X-ray micro-CT, which corresponded with a reduction in mechanical properties such as tensile strength and ductility. Collectively, the results underscore the crucial role of optimized sc-CO₂ pressure in promoting additive dispersion, thereby enhancing both foam morphology and additive–matrix interaction.

Table 2 compares the compressive strength, modulus, and specific properties of the FG- and GNP-reinforced PS foams developed in this study with values reported in the literature and for selected commercial PS-based foams. The compressive strength (0.40–0.50 MPa) and modulus (6-8 MPa) obtained for the PS-1 wt% FG and PS-1 wt% GNP foams are comparable to, or in several cases superior to, those reported previously for similar systems, even at equivalent or lower additive loadings. Notably, the experimental foams demonstrate higher absolute compressive strengths than the commercial PS foams (e.g., SOPRA-XPS 35, Greenguard GGS 25-LG, Isolofoam R Plus, and DuPont Styrofoam), highlighting the reinforcing influence of both FG and GNP fillers on the PS matrix. While some commercial foams exhibit higher specific compressive strengths due to their lower densities, the FG- and GNP-reinforced foams produced in this work show competitive performance despite their relatively higher apparent densities.

Table 2.

Comparison of the mechanical properties of PS-GNP and PS-FG foams developed in this work with values reported in academic literature and commercial PS foam insulation products.

Product	Compressive strength (MPa)	Compressive modulus	Density (kg/m3)	Specific compressive modulus (MPa/g.cm−3)	Specific compressive strength (MPa/g.cm−3)	Porosity (%)
PS-1 wt% FG (This work)	0.40	8	108	55	3.80	90
PS- 1 wt% GNP (This work)	0.50	6	142	40	3.52	86
PS - 1 wt% Graphene 13	-	1.65	70	20.20		89
PS- Graphite foam 59	0.067	-	30	-	2.23
PS-Graphite foam 60	0.08	-	20	-	4
PS- 30 wt% Expandable graphite 61	0.25	0.1	123	0.813	2.03	90
SOPRA-XPS 30 62	0.20		32.5		6.25	-
Greenguard GGS 25-LG 63	0.17		32.5		5.31	-
Islofoam R plus 64	0.10		32.5		3.12	-
Dupont styrofoam 65	0.10		32.5		3.12	-

A further reduction in foam density, without compromising cell integrity, would likely enhance the specific compressive strength and modulus values, advancing these materials toward the performance range of lightweight commercial PS foam products. Overall, these results underscore both the reinforcing efficiency of the carbon additives and the importance of density optimization in tailoring mechanical performance.

Conclusions

This study investigated the mechanical behavior of PS composite foams reinforced with GNP and FG, processed via sc-CO₂ extrusion foaming at pressures of 17.3 MPa and 20.6 MPa. The results demonstrated that increasing the sc-CO₂ pressure significantly enhanced the compressive strength and stiffness of the composite foams. Specifically, 0.75 wt% GNP foams showed a 34% increase in compressive strength, while 0.75 wt% FG foams exhibited a 66% improvement at the higher pressure. These improvements were supported by micro-CT observations, which revealed more uniform closed-cell structures and evenly distributed additives at 20.6 MPa. Additionally, specific energy absorption increased by approximately 14% for at 0.75 wt% GNP foams and 43% for 0.5 wt% FG foams at 20.6 MPa compared to 17.3 MPa, indicating enhanced impact protection for cushioning applications. DMA studies further confirmed superior viscoelastic behavior in high-pressure foams, with increased storage and loss moduli indicating improved stiffness and energy dissipation under loading.

The combined evidence from compression testing, viscoelastic performance and micro-CT imaging highlights the strong influence of sc-CO₂ pressure on both carbon additive dispersion and compressive performance. Higher pressure not only facilitated improved cell uniformity, but also reduced additive agglomeration, resulting in more efficient stress transfer and enhanced compressive response. These findings emphasize the importance of optimizing sc-CO₂ pressure for improving compressive performance in PS foams. Furthermore, the study demonstrates that high mechanical performance can be achieved even at relatively low additive loadings, supporting the development of lightweight, cost-effective PS-carbon composite foams for structural, insulation, and packaging applications.

Supplemental Material

Supplemental Material - Enhancing mechanical performance of polystyrene carbon composite foams through supercritical CO₂ foaming: An experimental study

Supplemental Material for Enhancing mechanical performance of polystyrene carbon composite foams through supercritical CO₂ foaming: An experimental study by Apurv Gaidhani, Stephan Edwards, Lauren Tribe, Paul Charpentier in Journal of Thermoplastic Composite Materials.