Thermal,rheological and mechanical behavior of PLA and PP biocomposites filled with Brewer’s spent grain (BSG): A comparative analysis of polymer-filler interactions

Abstract

This work is focused on composites based on Brewer’s spent grain (BSG), a by-product from beer production, with two thermoplastic polymers with different properties, Polylactic Acid (PLA) and Polypropylene (PP). The effects of BSG concentration were evaluated using Thermogravimetric analysis (TGA), Differential Scanning Calorimetry (DSC), melt dynamic rheology, tensile testing, and Scanning Electron Microscopy (SEM). TGA showed that BSG tend to interact with PLA, particularly at high temperatures during compounding, thereby accelerating PLA degradation and promoting chain cleavage as evidenced by size exclusion chromatography. This interaction affects the thermal transitions of the composites and mechanical properties, as evidenced by the ductile behavior observed in the micrographs. Conversely, BSG seems to act as an inert filler in the case of PP matrix. Rheological analysis suggested particle flocculation when PLA is mixed with above 30% (wt./wt.). In contrast, PP demonstrated consistent flow behavior with increasing BSG content, which is highly advantageous for extrusion and injection molding. PP appears to promote a good dispersion of the BSG particles, likely due to its sufficiently high viscosity. The tensile strength of both polymers was negatively affected by BSG, likely due to poor interfacial adhesion, as indicated by the microvoids in the SEM images. However, a steady elastic modulus with increasing BSG content for both polymers demonstrated their potential to maintain mechanical integrity when subjected to stress. This study provides a better understanding of BSG-filled PP and PLA composites, with opposite behavior, also contributing for the development of sustainable materials for packaging, home furniture, or automotive applications.

Keywords

bio-based composites particulate composites interfacial adhesion biomass valorization

Introduction

The rising need for sustainable materials has led the composite industry to consider lignocellulosic fillers as alternatives to inorganic fibers – and a search to find cost-effective, performant, and sustainable substitutes from biomass. Indeed, various lignocellulosic materials such as rice husks, wheat straw fibers, and tea mill waste are under investigation.^1–3 Using these by-products in composite manufacturing both reduces waste and increases the value of underused materials. However, while the use of long fibers in composites is stablished, there is less research on smaller particles from biomass, despite their potential as non-competitive food production resources. These particles can offer benefits such as better dispersion within matrices, the potential to enhance specific material properties, and low costs.

The beer production is spread worldwide. The brewer’s spent grain (BSG) is a by-product from beer industry, mostly made up of barley grain husks, and corresponding 85% of the waste generation. In 2023, world beer production reached 1.88 billion hectoliters⁴ which tends to further increase in the following years, highlighting the relevance of valorising this byproduct. The conventional disposal of this agricultural waste includes composting or animal feeding. However, with approximately 70% fibers and 20% protein,⁵ BSG is considered a lignocellulosic material potentially suitable as fillers in polymer composites. However, only a few studies have examined BSG as a composite filler, with varying success, highlighting the need for optimized processing conditions to overcome aggregation and poor filler-matrix adhesion.^6–10 Comprehensive investigation on BSG composites is necessary to address these challenges.

Thermoplastic polymers are a good match for this type of composite, since they offer ease of processing, which includes the ability to be reshaped multiple times without significant degradation, making them suitable for various manufacturing techniques. Among different possibilities, polylactic acid (PLA) and polypropylene (PP) are interesting alternatives, from different sources and with different behaviors. Consequently, this study focuses on the development and characterization of environmentally friendly BSG thermoplastic composite materials using PLA and PP. Although both polymers have been extensively studied as matrix in biocomposite materials, comparative studies specifically involving BSG as a filler are absent. This lack of comparative research involving presents a gap in understanding how this abundant agro-industrial by-product performs in different polymer matrices, its potential and limitations. Therefore, this study aimed to understand the interaction of PP or PLA with BSG particles, valorize BSG and contribute to the ongoing efforts to develop more sustainable composite materials. The effect of the matrix and filler loading on mechanical, physical, thermal and morphological properties were evaluated to determine which of the matrices and the loading offer the better balance between performance and sustainability.

Material and methods

Materials

PLA (PLE001) was supplied by NaturePlast with a density of 1.24 g/cm³ and melt flow index of 6 dg/min (ASTM D1238, 210°C/2.16 kg) and PP (PP H357-09RSB) was supplied by Braskem with density of 0.900 g/cm³ and melt flow index of 9.5 dg/min (ISO 1133, 230°C/2.16 kg).

Brewer’s spent grain used in this study was kindly provided by Senses Brewing (Reims, France). It was a waste from the production of light beer consisted of 9% of Best Caramel Pils, 45.5% of Pilsen and 45.5% of Pale Ale barley malts. To make it suitable for composite manufacturing, BSG was dried at 70°C for 44 h (MMM Medcenter Venticell 55 - ECO line – drying oven) from 74% to 3% moisture, then grinded (RETSCH Ultra Centrifugal Mill ZM 1000) for 5 minutes at 15000 rpm (Figure 2). BSG cellulose (22%), hemicellulose (18%), and lignin (22%) contents (dry basis) were determined using National Renewable Energy Laboratory (NREL) standard procedure.¹¹ True density (1.35 ± 0.02 g/cm³) was determined using a pycnometer and isopropanol. Particle size distribution (PSD) is presented in Figure 1. PSD was conducted using a Malvern 3000 laser particle sizer apparatus, operating within a range of 0.01-2000 µm. PSD is shown on Figure 1. The majority of the particles fall within 100-250 µm (37.6 ± 1.5%) and within 0-100 µm (35.4 ± 0.8%). Larger particles are mostly in the 250-500 µm range (20.2 ± 1.5%) and 6.3 ± 0.7% fall within 500-1000 µm. Only 0.5 ± 0.2% exceed 1000 µm.

Figure 1.

Particle size distribution of dried grinded BSG particles.

Preparation of composites

Blends of 50 g of polymer pellets (PLA or PP) and dry grinded BSG with different ratios (0, 10, 20, 30, 40, 50% (wt./wt.)), Table 1) were mixed at 180°C for 5 min using a Rheomix 600 QC (Thermo Fisher Scientific) equipped with two counter roller rotors R600. Mixing was conducted at 50 rpm screw speed to compound all the blends. Specimens were prepared by injection molding. After internal mixing, the blends were milled to obtain pellets with the Holishred (Holimaker). The injection was processed at 180°C for PLA composite materials and 200°C for PP composite materials to ensure optimal flow and molding using the Holipress (Holimaker). The pellets were melted in the barrel of the injection press for 6 min before injection. Three shapes of specimens were used: a circular one (diameter of 26 mm and a thickness of 2 mm) used for rheological tests; a rectangle one (length of 60 mm, thickness of 2.5 mm and 13.5 mm width); a dog-bone shape ISO527-2 type 5A¹² used for tensile tests. The circular-mold temperature was set at 80°C, the rectangle-mold temperature was set at 120°C. The dog-bone shape specimen was cut from the previously injected rectangle-shaped specimen which was heated on a heated plate for 3 minutes at a temperature of 60°C, after which it was die-cut, using the dog-bone shape die (Figure 2).

Table 1.

PP and PLA - BSG composites formulations and sample codes.

Polymer	Sample code	Weight ratio of polymer (%)	Weight ratio of BSG (%)	Volume ratio of BSG (%)^a
Polypropylene (PP)	PP	100	0	0
	PPBSG10	90	10	7.0
	PPBSG20	80	20	14.4
	PPBSG30	70	30	22.4
	PPBSG40	60	40	31.0
	PPBSG50	50	50	40.2
Polylactic acid (PLA)	PLA	100	0	0
	PLABSG10	90	10	9.3
	PLABSG20	80	20	18.8
	PLABSG30	70	30	28.5
	PLABSG40	60	40	38.2
	PLABSG50	50	50	48.1

PP density = 0.9 g/cm3; PLA density = 1.24 g/cm³; BSG density = 1.35 g/cm³.

^aVolume obtained with BSG density using a pycnometer and isopropanol.

Figure 2.

Flow diagram showing the composite fabrication process, with key steps in material preparation, mixing, molding, and characterization.

Characterization methods

Moisture content

The moisture content of both BSG and PLA was determined using a Kern DAB moisture analyzer (Kern & Sohn, Germany), using three replicates. The obtained moisture values were 3.2 ± 0.3%w.b. (g water/g product) for BSG, 1.0 ± 0.1% w.b. (g water/g product) for PLA, and 0.7 ± 0.0% w.b. (g water/g product) for PP.

Thermogravimetric analysis (TGA)

TG analyses were conducted using a thermogravimetric analyzer (TG209F3, Netzsch) in nitrogen atmosphere. A three-step heating rate was applied: 40 K/min from 20°C to 100°C, an isotherm for 30 min at 100°C to remove the residual water, then 10 K/min from 100°C to 600°C. Samples, weighing approximately 10 mg, were loaded into alumina (Al₂O₃) pans. Key parameter derived from TGA curves was the derivative temperature peak, which was indicative of the temperature reached for the highest degradation rate. Control experiments were performed for all experimental procedures, with each test replicated three times. To eliminate the influence of the crucible mass on heat capacity measurements, the results were adjusted by deducting data from a control test conducted using identical protocols and crucibles. This calibration process ensured the accuracy of the raw data.

Gel permeation chromatography (GPC)

Given our hypothesis regarding potential PLA degradation during compounding, molecular weight of PLA samples was measured using Shodex K-804 L mixed gel styrene-divinylbenzene copolymer column (Shimadzu, USA) (8 mm internal diameter, 300 mm length, 7 µm particle size). Prior to analysis, the PLA composites were dissolved in chloroform at a concentration of 3-4 mg/mL and filtered with 0.45 µm PTFE filters to separate the filler phase from polymeric phase. The injection volume was 20 µL and chloroform was used as mobile phase at a flow rate of 1 mL/min. Column temperature was maintained at 50°C. A refractive index detector was used to detect the studied samples. The system was calibrated using polystyrene standards with molecular weight ranging from 12 630 to 388 500 g/mol.

Differential scanning calorimetry (DSC)

DSC analyses were conducted using a DSC Q20 (TA instrument) using three ramps at 10 K/min: an first heating ramp from −60°C to 200°C, a cooling ramp, and a second heating ramp. Samples, weighing approximately 12 mg, were loaded into aluminium pans. The DSC analysis was performed under nitrogen purge with a flow rate of 50 mL/min. Temperatures of different thermal events such as glass transition (T_g), crystallization (T_c), cold crystallization (T_cc) and melting (T_m) were recorded. The degree of crystallization (χ_c) was calculated using the following equation¹³:

χ c = \frac{Δ H_{m -} Δ H_{c c}}{Δ H_{m}^{0}} x \frac{100}{w}

(1)

ΔH_cc is the enthalpy for the cold crystallisation in J/g, ΔH_m is the enthalpy for melting transition in J/g, ΔH⁰_m is the enthalpy of melting for a 100% crystalline PLA or PP sample in J/g, taking ΔH⁰_m = 93.6 J/g for PLA¹⁴ and ΔH⁰_m = 209 J/g for PP.¹⁵ w is the weight fraction of PLA or PP in the sample.

Melt rheology

Oscillatory shear rheological properties of each blend and neat polymers were measured using a smooth parallel plate (diameter = 25 mm) in a Rheometer AR2000ex (TA Instrument). Tests were performed at 180°C for all composites, the gap was adjusted to 2200 µm 2 minutes after the sample has been loaded to let it melt. Strain sweeps were conducted from 0.01 to 100% at angular frequency of 10 rad/s and of 0.1 rad/s to determine the linear viscoelastic region (LVR) of the samples. Frequency sweeps were conducted at 180°C from 200 rad/s to 0.1 rad/s and at the maximum strain the material can withstand in the LVR. Thus, the selected strain was 10% for the pure polymers and composites with BSG fractions 10 and 20% (wt./wt.), while it was 0.5% for more highly filled composites (30, 40, and 50% (wt./wt.)) in order to remain in the LVR. Frequency sweep tests were performed at least three times to obtain a mean curve and standard deviation.

Tensile tests

Samples were prepared according to the ASTM D638 standard,¹⁶ cut into dog-bone shape with a total length of 50 mm and a working length of 17 mm, and a width of 4 mm. Samples were conditioned for at least 48 h at ambient temperature (20 - 25°C) and 55 ± 5% relative humidity before tensile testing. Tensile tests were performed using an Instron K1188 5544 universal testing machine equipped with a loadcell of 2 kN. Prior to testing, the specimen was securely mounted onto the testing machine’s grips, ensuring proper alignment. A preload of 10 N was applied at a speed of 1 mm/min before the test to eliminate any slack in the specimen and ensure proper contact between the grips and the specimen. Then the cross-head speed was 2 mm/min. The test was conducted until rupture, and data, including maximum load, load at rupture, and elongation at rupture, were recorded. Each test was conducted five times to assess the reproducibility of results.

Water contact angle measurement

The surface affinity of PP and PLA toward water was assessed through static contact angle measurements, which provide an estimate of the hydrophilicity or hydrophobicity of the polymer surfaces. Measurements were performed using a fully automated goniometer (OCA25, DataPhysics, Germany) equipped with a diffuse light illumination system to minimize shadowing effects and a high-resolution CCD camera operating at 20 frames/s to precisely capture the droplet profile over time. Prior to analysis, all PP and PLA samples were prepared under identical conditions to avoid surface variability. The materials were first compounded by internal mixing and subsequently shaped by injection molding. For each measurement, an ultrapure water droplet (≈3.5 µL) was deposited onto the surface using a precision micro-syringe fitted with a Teflon needle. The dispensing rate was carefully controlled at 0.058 µL/s to avoid introducing kinetic effects that could distort the droplet shape. The static contact angle was determined immediately (after 18 to 20 frames) after droplet deposition using the instrument’s built-in image analysis software based on the Sessile Drop method. For each formulation, at least five droplets were deposited at different surface locations to ensure statistical reliability.

Scanning electron microscopy (SEM)

Scanning electron microscopy was used to observe the fracture surfaces after tensile testing. The rupture surface of the samples was sputter-coated with a 13 nm layer of gold/palladium. The sputtering was generated with a Leica EM SCD050 under Argon atmosphere at 60 mA during 30 s. The microstructures of composites fractures were analyzed using a Hitachi TM3030Plus tabletop microscope under mix mode (backscattered and secondary electrons) at accelerating voltage of 15 kV.

Experimental design and statistical evaluation

The mean and standard deviation were calculated, and the data were statistically analyzed for comparison using R Studio software (version 3.6.3). To evaluate differences between the sample groups, a pairwise t-test was performed. Statistical significance was determined using a p-value of 0.05. The results were annotated with letters to indicate statistical groupings. The results sharing the same letters were not significantly different from one another and results with different letters were considered to show significances differences in their means. GPC was performed in duplicates. TGA, rheology and DSC measurements were performed in triplicates. Tensile tests were performed 5 times per composites formulations.

Results and discussion

Thermal degradation behavior

To obtain information on composition and thermal resistance to degradation, thermogravimetric analyses were carried out on all formulations.

Figure 3 depicts the first derivative weight loss while heating the samples and BSG to 600°C under nitrogen atmosphere. TG curves can be found in supplemental material. Table 2 gathers the temperatures reached at maximum mass loss rates.

Figure 3.

Thermogravimetric Analysis of biocomposites made with BSG and PP or PLA. DTG curves showing decomposition rate as a function of temperature (10 K/min under nitrogen flow). Average of three experiments.

Table 2.

Temperatures reached at maximum degradation rates for biocomposites made with BSG and PP or PLA.

Material	Sample code	Temp 1st peak (°C)	Temp 2nd peak (°C)	Temp 3rd peak (°C)
BSG	BSG	298^a ± 1	347^a ± 6	-
Polypropylene (PP)	PP	-	-	445^a ± 10.
	PPBSG10	294^b ± 1	-	453^ab ± 3
	PPBSG20	294^b ± 1	355^a ± 1	462^bc ± 2
	PPBSG30	296^ab ± 1	355^a ± 2	461^bc ± 3
	PPBSG40	295^ab ± 1	355^a ± 1	460^bc ± 3
	PPBSG50	294^b ± 1	356^a ± 1	468^c ± 1
Polylactic acid (PLA)	PLA	-	-	366^a ± 2
	PLABSG10	-	-	365^a ± 1
	PLABSG20	-	-	364^a ± 1
	PLABSG30	290-300	-	363^ab ± 1
	PLABSG40	290-300	-	359^b ± 1
	PLABSG50	299^a ± 2	-	359^b ± 1

The values represent the mean of triplicate measurements, with standard deviations indicated. Different letters (a, b, c) indicate statistically significant differences. For the treatments PLABSG30 and PLABSG40, a temperature range is provided since they do not exhibit a clear minimum to reliably identify the first peak.

Regarding pure BSG thermal degradation profile, the initial pyrolysis step begins at around 175°C and is characterized by a shoulder corresponding to the lightest volatile compounds in BSG. Sobek et al. conducted research on BSG pyrolysis kinetics and observed a similar profile attributed to the lightest volatile fractions of BSG.¹⁷ Using a deconvolution procedure, these authors identified three pseudo-components, attributed respectively to hemicelluloses, cellulose and lignins, from the most volatile to the least volatile. These results on BSG are perfectly consistent with the pyrolysis of the three primary components of lignocellulosic biomass.^17–19 All these works suggest then that the first shoulder of Figure 3 represents the volatilization of lightest components. This shoulder extends until the first sharp peak, which represents the maximum mass loss rate reached at 298 ± 1°C, clear signature of the fast decomposition of hemicellulose.^18,19 Following this, a second sharp peak detaches, attaining its optimal point at 347 ± 6°C, related to cellulose.¹⁷ The significant overlapping of the peaks highlights the intricacy of BSG composition. Nevertheless, BSG starts to degrade at a low temperature regarding the temperatures reached during thermoplastic material processing. Indeed, we observed 1% mass loss at 180°C after 40 min and 3% at 210°C after 43 min.

Only single degradation peak, consequence of the homogeneous nature of the polymers, was observed for PP and PLA. PP reaches its maximum degradation rate at 444 ± 10°C while PLA degradation peak occurs at 366 ± 2°C which is consistent with published studies.^20–22 It can be observed that PP exhibits a highly variable degradation temperature, as indicated by its large standard deviation compared to PLA. On the contrary, PLA shows more homogeneous thermal behavior. A similar dispersity between PP degradation temperatures was also observed by Stawski²³ who did not identify this variability as unusual for polypropylene.

Further analysis of Figure 3 reveals an interesting degradation pattern of PP-composites. The degradation temperatures of BSG exhibit relative consistency across the composites, indicating that PP does not significantly influence the thermal stability of BSG. Indeed, the first peak occurs at similar temperatures, irrespective of the presence or content of PP. The hemicellulose of BSG appears to degrade marginally earlier in the composites compared to pure BSG. However, this difference is minimal and may be attributed to the composite manufacturing process, which could have weakened the BSG. In contrast, the degradation behavior of PP is substantially affected by the presence of BSG. As observed in the third peak, the degradation temperature of PP increases with higher BSG content, suggesting that BSG provides a protective effect. Specifically, the PPBSG50 composite exhibits a maximum degradation temperature of 468°C, which is significantly higher than the 445°C of pure PP. The DTG curves reveal that BSG degrades at a lower temperature than PP, as expected, which can affect the thermal stability of the biocomposite. According to literature, the biomass thermal degradation can potentially form a char layer which could interact with PP and shift its degradation peak to higher temperatures. This behavior has been observed with various biomasses like bamboo,²⁴ hemp fibers,²⁵ and lignin.²⁶ The BSG-PP behavior observed in the present work can be attributed to such mechanisms, namely the adsorption of PP volatiles by BSG char likely to trigger secondary charring and delay mass loss.²⁷ Consequently, the BSG-PP behavior observed in the present work can be attributed to this mechanism. In fact, for the PP-based composites, this statistical treatment clearly confirms that the increase in PP degradation temperature with BSG content is significant (Table 2). Furthermore, the standard deviations associated with the third peak for PP degradation demonstrate a marked decrease as the BSG content increases (Table 3). For pure PP, the degradation temperature exhibits relatively high variability (±10°C, 2.3%), whereas in the composites with higher BSG content, the variability reduces significantly (±1°C, 0.1% for PPBSG50). The statistical grouping confirms this stabilization trend: despite differences in mean temperature, all PPBSG composites fall into narrower statistical groups (“ab”- “c”), reflecting that PP degradation becomes more uniform as BSG content increases. Such homogenization suggests that BSG contributes to creating a more consistent thermal environment within the composite during heating, thereby limiting local fluctuations in temperature or degradation pathways. This dual effect (higher stability and lower variability) indicates that the material decomposes in a more uniform and predictable manner, making it more reliable under thermal stress and improving processing stability in applications such as injection molding.

Table 3.

DSC data of PP, PLA and their composites after internal mixing and injection molding.

Material	Sample code	T_g (°C)	T_cc (°C)	T_c (°C)	T_m (°C)	$χ c$ (%)
BSG	BSG	-	-	-	-	-
PP	PP	-	-	117^a ± 1	167^a ± 1	40.9^a ± 1.1
	PPBSG10	-	-	118^ab ± 0	166^a ± 1	42.8^a ± 0.7
	PPBSG20	-	-	116^b ± 1	166^a ± 1	40.3^a ± 6.3
	PPBSG30	-	-	115^c ± 0	166^a ± 1	40.9^a ± 2.4
	PPBSG40	-	-	113^c ± 0	166^a ± 1	41.7^a ± 4.3
	PPBSG50	-	-	114^c ± 0	164^a ± 1	43.0^a ± 2.0
PLA	PLA	60^a ± 1	116^a ± 1	-	150^a ± 2	0.6
	PLABSG10	57^b ± 0	113^b ± 0	-	147^a ± 1	∼0
	PLABSG20	57^bc ± 0	125^c ± 0	-	150^a ± 0	∼0
	PLABSG30	56^cd ± 0	121^d ± 0	-	149^a ± 0	∼0
	PLABSG40	54^e ± 0	116^a ± 1	-	147^a ± 1	∼0
	PLABSG50	55^de ± 1	116^a ± 1	-	148^a ± 1	∼0

Different letters (a,b,c,d,e) indicate statistically significant differences between samples. p-value = .05.

PLA composites reveal a different trend of degradation. In the composite samples, particularly at lower content of filler, the first peak of BSG degradation appears as a shoulder that tends to overlap and partially fuse with PLA degradation peak (Figure 3). As the BSG content increases in the composites, the first peak of BSG degradation becomes more pronounced and starts to separate from the main PLA degradation peak. For example, in PLABSG50, BSG degradation 1st peak becomes distinct and appears at 359°C. For PLABSG30 and PLABSG40, although the curves show a decrease in the signal, they do not exhibit a clear minimum that would allow us to reliably identify the temperature of the first peak. The transition is gradual and does not present a well-defined minimal point. Therefore, it is not possible to assign a precise temperature value without over-interpreting the data. Accordingly, in Table 2 we report a temperature window rather than a single value, representing the range within which the first degradation peak of PLABSG30 and PLABSG40 is expected to occur. The standard deviations for PLA degradation peak (3^rd peak) are relatively small and consistent across the samples. Furthermore, the PLA degradation temperature decreases as the BSG content increases. In the PLABSG50 composite, PLA degrades at 359°C, compared to 366°C for pure PLA. This behavior has been observed in literature where it was observed that potassium present in the filler acted as catalyser for PLA degradation.²⁸ It is possible that BSG acts as a catalyst, facilitating the breakdown of PLA during thermal degradation by providing reactive sites or transferring heat more effectively. This behavior might be exacerbated under air atmosphere for example during internal mixing. This degradation has been highlighted through SEC analysis, which reveals clear evidence of PLA chain scission when mixed with BSG. Indeed, upon adding 10% (wt./wt.) of BSG, the number-average molecular mass (Mn) of PLA decreased by 27%, and at 50% (wt./wt.) incorporation, the reduction in PLA Mn reached 48% (Figure 4).

Figure 4.

Detection of PLA molecules using refractive index as function of elution time and comparison of the number average molecular weight (Mn) of PLA composites after internal mixing and injection molding. Different letters (a,b,c,d) indicate statistically significant differences between samples.

To conclude, while PP seems to be preserved by the char layer formed by BSG, PLA degradation seems to be enhanced by some thermal degradation products of BSG. This is likely due to the very different degradation temperature of both polymers. Moreover, due to its biodegradable nature, PLA presents reactive sites subjected to scission. Thus, it can easily decompose into carbon dioxide, water or methane when stimulated be inducers and temperature.

Thermal transition behavior

DSC analyses were carried out to determine characteristic temperatures and obtain information on the microstructure of the different formulations. Figure 5 shows the DSC curves of PP and PLA composites. For PP, both first cooling and the second heating curves were analyzed to show its crystallization and melting behavior, respectively. During the heating cycle, the melting temperature (T_m) was determined. The cooling curve was used to determine the crystallization temperature (T_c) and the degree of crystallinity (χ_c). In contrast, PLA exhibits a different thermal behavior characterized by the absence of crystallization under typical cooling rates. Thus, only the second heating curve was used to assess its glass transition temperature (T_g), its melting temperature (T_m), and its cold crystallization temperature (T_cc). Table 3 presents the results of measured T_c, T_m, T_cc, T_g and χ_c for the pure polymers and BSG composites.

Figure 5.

Differential scanning calorimetry curves of pure BSG, injected PP and PLA and composites at 10 K/min.

During cooling, the neat PP exhibits an exothermic peak corresponding to its crystallization occurring at 117°C which is consistent with previous studies.²⁹ With the addition of fillers, the crystallization temperature tends to decrease, and for PP with 50% (wt./wt.) BSG, crystallization occurs around 114°C, indicating a delayed crystallization. The crystallization process of PP might be hindered by an increased BSG content, causing a shift to lower temperatures. This phenomenon may be attributed to the restriction of polymer chain mobility by BSG.^30,31 Although the comparison of T_m did not reach statistical significance, a noticeable trend is observed for the temperature of melting, where fillers seem to have a negative impact on T_m. Pure PP melts around 167.0°C, while PP with 50% (wt./wt.) fillers melts around 164°C, indicating lower interactions between PP chains due to the presence of BSG. In the end, crystallization rate of PP and its composites do not appear to differ significantly.

Regarding PLA thermal transitions, the accurate determination of crystallization rates from DSC data needs distinct and well-separated crystallization and melting peaks. In the current study, the thermal events corresponding to cold crystallization and melting were closely overlapped, making it difficult to accurately define the onset and end temperatures for these transitions. Consequently, the integration limits for calculating the enthalpy of crystallization were not well-defined, leading to imprecise quantification of the crystallization rate. Thus, it led to anomalous crystallization rates close to 0 but yielding slightly negative values which is physically impossible. Therefore, these values were set to zero. The T_g for neat PLA was observed at 60°C, whereas with the incorporation of BSG, the T_g decreased, reaching 55°C for PLABSG50. This reduction in T_g suggests that BSG has a possible plasticization effect on PLA matrix, by spacing PLA chains apart, potentially increasing the polymer chain mobility of the amorphous regions.

An additional factor contributing to the lower T_g could be that early thermal decomposition products of BSG could have initiated PLA chains hydrolysis during internal mixing process.³² This hypothesis has been confirmed through SEC analysis, which reveals clear evidence of chain scission (Figure 4). Moreover, residual moisture in BSG (3.2% before compounding) may have also induced hydrolytic degradation of the PLA chains, resulting in chain scission and a consequent reduction in the molecular weight of the PLA. Decreased molecular weight typically leads to enhanced chain mobility and a corresponding decrease in T_g. Indeed, a higher number of chain ends causes the existence of a higher free volume enhancing chains mobility.³³ This effect is particularly pronounced at higher BSG loadings, as observed in the PLABSG40 and PLABSG50 composites. As the decrease in T_g may be attributed to BSG thermal degradation products enhancing PLA chain scission, higher BSG concentrations could amplify this effect. Along the same lines, Zhang et al. showed the molecular weight dependence of glass transition temperature in their study on amorphous Poly-L-Lactic Acid system.³⁴ This correlation has also been highlighted by others.³⁵

The T_cc of PLA composites also varied with BSG content (Table 3). Indeed, for pure PLA, the T_cc was observed at 116°C. In the PLABSG10 composite, the T_cc decreased to 113°C, but surprisingly, it increased to 125°C in PLABSG20, before slightly decreasing again for composites with higher BSG content, such as PLABSG50, where T_cc was recorded at 116°C. There is likely a dual effect of BSG on the crystallization behavior of PLA. First, BSG’s thermal degradation products tend to reduce the molecular weight of PLA during processing, which in turn lowers the T_cc.³⁵ Second, the presence of BSG particles may physically interfere with the growth and propagation of PLA crystallites, which can shift the cold crystallization temperature (T_cc) to higher values. At low BSG content, the chain scission effect seems to dominate, leading to a probable reduction in molecular weight and lowering T_cc. At moderate BSG content, the physical barrier hindering crystals propagation becomes more prominent, causing delays in T_cc. At high BSG content, the composite system might experience a phase separation with the agglomeration of the excess particles which could lead to localized regions where PLA crystallization behavior resembles the behavior seen in PLABSG10. This highlights the dual effect of BSG on thermal events of the composites. Mofokeng et al. also observed a dual effect of sisal on PLA chain mobility and stated that fibres can contribute to both nucleation and immobilization of PLA chains.³¹

Turning now to the melting transition, Figure 5 shows the presence of double melting peaks for some PLA composite formulations. This indicates that the presence of BSG may induce the formation of a second type of crystal within the PLA matrix. The first melting peak (mP_I) is associated with the melting of less stable crystals, while the second melting peak (mP_II) corresponds to more stable crystals that likely form through recrystallization (corresponding to exothermic peak between mP_I and mP_II).¹⁴ However, in PLABSG20, where cold crystallization is delayed (as evidenced by higher T_cc values), a different behavior is observed and highlighted by the appearance of a single melting peak. In these cases, the mP_I overlaps with the mP_II. A crystallization happening at higher temperature suggests the formation of more stable crystal forms named α- crystals.³³ In PLABSG30, the double melting peak is present but less distinct, indicating that the formation of the second type of crystal is partially hindered, possibly due to the overlapping melting transitions of the two types of crystals. In their study on Poly-L-Lactic Acid crystallization behavior, Lorenzo highlighted the dependence between T_cc and the presence of the first melting peak.³⁶ They stated that T_cc > 120°C (as it is the case for PLABSG20 and PLABSG30) induces a merge between the two melting peaks and recrystallization peak due to the lower extend of the recrystallization process. Moreover, they found that a T_cc < 120°C leads to imperfect crystals that show tendency to reorganize through an exothermal peak between the two melting transition peaks.³⁶

While we did not observe a significant crystal rate in PLA composites (Table 3), it is possible to conclude that crystallization behavior may have consequences on its mechanical properties. Indeed, crystal structure can affect the overall material properties. As we observed, some composites showed the presence of two types of crystals highlighted by the two melting peaks. Less stable crystals indicate a looser and more imperfect chain packing, which typically leads to lower stiffness and reduced barrier performance of the material.³³ Therefore, while increasing the mobility of the chains as evident in the T_g reduction, BSG might initiate the nucleation of less stable crystals which could have consequences on PLA mechanical properties. As for PLA the influence of BSG content on crystallization rate of PP composites was negligible. However, PP composites appeared to show an increasing delay in their transition temperatures with increasing BSG content suggesting a physical effect of BSG which could act as a barrier hindering polymer chain mobility.³¹ Special care must be taken when choosing the process conditions especially the cooling rate to optimize processing parameters for desired material performance.

Rheological behavior

To better understand the interactions between BSG and the polymer chains in our biocomposites, it is very useful to observe the behavior of their elastic moduli and complex viscosity in the molten state at low frequencies, in the so-called flow zone. Figure 6 presents the evolution of G′ and complex viscosity with frequency for PP-BSG and PLA-BSG composites at different filler contents.

Figure 6.

Effects of BSG ratio on storage modulus (Pa) and complex viscosity (Pa.s) versus angular frequency (rad/s) in plate rheometer at 180°C for Polylactic Acid (PLA) and Polypropylene (PP) composites processed at 180°C in internal mixer and injected at 180°C and 200°C respectively Փ= volume fraction of BSG, plate-plate gap = 2200 µm, plate diameter = 25 mm, strain = 10% for PP, PLA, PPBSG10, PPBSG20, PLABSG10, and PLABSG20, 0.5% for composites with 30, 40 and 50% (wt./wt.) BSG. Average of three experiments.

For the PP-BSG composites, G′ systematically increased with increasing filler fraction, indicating enhanced stiffness due to the presence of BSG. This behavior has also been observed by Guo et al. in similar filler loadings for glass fibers reinforced LLDPE.³⁷ Looking at the complex viscosity, a shear-thinning behavior is observed.^38–41 In all PP composites, irrespective of the BSG content, the onset of shear thinning occurs at approximately the same frequency (in the range 0.2 to 0.5 rad/s). It was shown in previous studies that the incorporation of particles systematically increases the storage modulus, as well as the loss modulus and complex viscosity.^38,39,42,43 This increase is attributed to the particles acting as obstacles, resisting the recovery of the polymer chains under strain and thus enhancing the overall stiffness of the system. Lastly, the incorporation of BSG does not alter the overall shape of the G′ and viscosity curves of PP composites, further supporting the absence of significant specific interactions between BSG and PP.

In contrast, the behavior of PLA-BSG composites differs significantly. While a decreasing G′ with decreasing frequency is still observed, neat PLA, PLABSG10 and PLABSG20 exhibit a two-stage behavior with a plateau at intermediate frequencies (1–10 rad/s). This unusual trend has been reported, but rarely discussed, in the literature.⁴⁴ It may be related to temporary entanglements and relaxations of the chains. Indeed, due to the broad polydispersity of PLA, chains with different molecular weights tend to relax at different characteristic frequencies.⁴⁵ As a result, the addition of BSG particles can amplify this effect by modifying the mobility and relaxation dynamics of different chain fractions. With increasing BSG content the plateau shifts to lower frequencies which may indicate the formation of a weakly connected particle network as the concentration approaches the percolation threshold.^38,43,46 This network partially restricts chain mobility at long relaxation times. A pronounced change in the G′ slope at high frequencies between low-filled (PLA, PLABSG10, PLABSG20) and high-filled (PLABSG30, PLABSG40, PLABSG50) composites further suggests a transition in the microstructural organization of the melt. Contrary to PP composites, PLA-BSG composites exhibit markedly different complex viscosity curves. At low filler concentrations (<30% (wt./wt.)), the composites demonstrate quasi-Newtonian behavior, wherein the viscosity remains almost constant over the entire frequency range, with no discernible shear-thinning effect. This suggests that at particle concentrations under 30% (wt./wt.), the filler may interact with PLA which restricts its chain mobility extending the Newtonian plateau and delaying a possible shear-thinning effect. However, the content is insufficient to significantly alter the flow properties of the PLA matrix which keeps its fluid-like behavior. We observe that the overall curves of PLABSG10 and PLABSG20 fall below that of pure PLA, which may be related to a reduction in the molecular weight of PLA, knowing that viscosity is strongly dependent on polymer molecular weight. Moreover, similar effects have been reported with other natural particles-reinforced PLA composites. For instance, in PLA composites, the addition of abaca fibers was found to reduce less than expected the melt flow rate of the composite. This effect was attributed to a reduction in PLA molecular weight due to chain scission happening during material processing. Indeed, the thermal decomposition of the natural fibers during processing releases residues such as water, waxes, and other volatiles. These decomposition byproducts likely interacted with the PLA matrix, initiating hydrolysis and scissions within the polymer chains and leading to a reduction in molecular weight.³² This degradation process in PLA-BSG composites could similarly contribute to the observed reduction in complex viscosity. SEC analysis allowed us to confirm that PLA after internal mixing and injection molding showed a Mn of 76604 ± 2053 g/mol while Mn of PLABSG10 and PLABSG20 were respectively reduced to 55899 ± 5807 g/mol and 57459 ± 1510 g/mol (Figure 4).

However, at higher BSG content, the complex viscosity begins to follow a power-law behavior at low frequencies. Such a behavior indicates molten composite system reaching the percolation threshold which we could also observe on G′ curve. At this stage, the rheological response reflects the presence of a particle network that resists flow, causing the material to exhibit solid-like behavior at low shear. Moreover, these results confirm that particle-particle interactions dominate the behavior of composite at low frequencies.^43,46–48

In both polymer systems BSG seems to exacerbate matrix flow behavior. The difference in behavior between PLA and PP composites highlights the stronger influence of particle-particle and particle-matrix interactions in the PLA system. These interactions significantly shape the curves across frequencies, reflecting their impact on composite’s response to shear stress. Although particle-particle interactions and aspect ratio are inherently similar in both PP and PLA systems, PLA presents a lower viscosity than PP under the same processing conditions. As a result, PLA provides less hydrodynamic hindrance to particle movement. BSG particles in the PLA melt remain more mobile and can interact more freely with each other during oscillatory shear, whereas in the more viscous PP matrix, the particle motion is more restricted, keeping particles apart. These analyses highlight that the differences in rheological behavior between PLA- and PP-based composites arise primarily from the matrix viscosity and its influence on particle mobility.

Mechanical behavior

Young’s Modulus

The Young’s modulus values of PP and PLA composites calculated from the slope of stress versus small strain (0.5-1.5%) plotted against BSG mass content are gathered in Figure 7, strain-stress curves are given in supplemental material. As can be observed, the elastic modulus of PP and PLA composites remains steady with increasing volume fraction of BSG. Pure PP which was mixed by internal mixing and injected shows a tensile modulus of 1497 ± 140 MPa which is consistent with previous studies^3,49,50 whereas PP composites with 50% (wt./wt.) have a modulus of 1556 ± 122 MPa. Pure PLA shows a Young’s modulus of 2301 ± 442 MPa aligning with previous studies^51,52 whereas PLA composites with 50% (wt./wt.) have a modulus of 2169 ± 139 MPa.

Figure 7.

Comparison of Young’s modulus for different BSG mass fraction in PLA and PP composites. Different letters (a,b) indicate statistically significant differences between samples.

Although the results deviate from those in literature demonstrating the incorporation of fillers leads to an increase of the Young’s modulus,^3,53,54 they are similar to studies reporting biomass fillers may not influence or may even weaken the mechanical properties of biocomposites.⁵⁵

For instance, the addition of wheat straw fibers to polypropylene increases the modulus of the composites from 1.21 GPa for PP to 2.99 GPa for 30% reinforced PP, the fibers were mechanically defibrillated and sieved (500 µm).³ Similarly, the addition of rice husk flour to PP have been found to increase the Young’s modulus continuously with increasing flour loading (0% to 40% (wt./wt.)).⁵⁶ Moreover, 50% (wt./wt.) peach stone particles with small particles size (<80 mesh) reinforced composites have demonstrated superior tensile modulus (2.49 GPa) compared to the pure PP (1.35 GPa).⁵⁰ However, in the same study, Nunez-Decap et al., observed that 50% peach stone particles composites with large particles size (80-140 mesh) demonstrated a tensile modulus which was not significantly different from the modulus of pure PP.⁵⁰ For PLA, the addition of 30% pineapple skin flour was found to increase tensile modulus of the composites.⁵¹ Similarly, wood fibers reinforced PLA composites exhibited increased tensile modulus, with 40% (wt./wt.) of wood fibers reaching a Young’s modulus of 6.3 GPa against 2.7 GPa for the pure PLA.⁵⁷ Finally, Sutivisedsak et al. reported a decrease in elastic modulus of cotton burr/PLA composite, after extrusion and punch press cutting, from 0.609 GPa for pure PLA to 0.239 GPa for 10% cotton burr-filled PLA.⁵⁵

This less influence or weaking of biomass in the composites can be attributed to several factors. First, the size of our filler can negatively impact the Young’s modulus of our composites. Indeed, bigger filler could result in a reduced tensile section inducing more stress concentration at the voids. Secondly, the steady modulus with increasing BSG content may be indicative of a limited stress transfer between BSG and the polymer. Our SEM imaging (Figure 9) revealed clear debonding at the BSG-matrix interface, confirming the absence of adhesion. Under such circumstances, BSG particles behave as mechanically inert inclusions, and classical micromechanical models predict no increase in stiffness because the filler cannot participate in load transfer. Indeed, poor wetting or increased porosity at high mass loadings reduces effective stress transfer and often compensates the stiffening effect of the filler — resulting in a plateau or marginal change of E.^58,59 This explains why the Young’s modulus remains unchanged even at high BSG contents. Furthermore, interfacial microvoids created by poor wetting act as local defects and stress concentrators, further preventing any modulus enhancement and potentially counteracting reinforcement that could otherwise be expected from rigid particles. As shown by Hashin and Shtrikman,⁶⁰ the effective elastic moduli of a multiphase body are rigorously bounded by simple functions of the constituent phase moduli and their volume fractions. Hence, for a thermoplastic matrix containing rigid particulate inclusions, classical variational bounds predict that — depending on filler stiffness, filler volume fraction and phase contrast — the macroscopic Young’s modulus can remain nearly unchanged (or lie within narrow bounds) even at high filler loadings. This provides a theoretical basis for the experimental observation of an essentially constant E for our PP/BSG composites. Lastly, while fillers extracted from biomass like BSG are often reported to act as nucleating agents, promoting crystallization and subsequently enhancing the stiffness et reducing elasticity of materials^49,61 our DSC results suggest that the addition of BSG did not significantly affect the crystallization rate of neither PP nor PLA matrix. This may explain the lack of reinforcement observed in the mechanical properties of our BSG composites.

Tensile strength

To quantify the adhesion at the interface, Nicolais and Narkis proposed an equation for a lower-bound strength of the composite, defining the tensile strength of the composites as function of volume fraction for the filler, based on the hypothesis that there is no adhesion between filler and polymer and that the load is sustained solely by the polymer. An upper-bound could be obtained by considering that for perfect adhesion, the strength of the composite is simply equal to the strength of the polymer matrix.⁶² Nicolais and Narkis created a simple model that demonstrated the relationship between the filler volume fraction, the tensile strength of the pure polymer and tensile strength of the composite in the absence of adhesion. Nicolais and Narkis assumed that the filler was uniformly distributed and of spherical shape demonstrating the tensile strength decreased substantially with an increase in bead concentration. This is consistent with previous research on elastomeric PVC composites.⁶³ According to their model, the strength can be described by:

σ_{c} = σ_{m} (1 - 1.21 {ϕ_{p}}^{2 ∕ 3})

(2)

σ_{c}

: composite tensile strength,

σ_{m}

: matrix tensile strength (pure polymer),

Φ_{p}

: particle volume fraction

The constants are associated with stress concentration and the filler’s geometric properties. They are equal to 1.21 and 2/3 for fillers that are spherical and uniformly dispersed into the matrix.⁶³ These constants and the particle volume fraction reflect the effective cross-sectional area “lost” for load bearing due non-adhering filler inclusions. This model has only been established for this type of filler. This assumption is acceptable in our case because the BSG particles are relatively large, randomly oriented, and with a moderate and broadly distributed aspect ratio.

Figure 8 displays the tensile strength of PP and PLA composites as a function of BSG volume fraction. For reference, a conversion table between mass and volumic content is provided in 2.2 (Table 2). Unlike the elastic modulus, tensile strength was strongly and negatively affected by the filler for both polymer matrices. For instance, by reinforcing PP with 40.2 vol% of BSG, the tensile strength is halved. For pure PLA the tensile strength equals 57.4 ± 17.8 MPa which is consistent with the literature^55,64 whereas for 48.1 vol% BSG reinforcing PLA, the tensile strength was dropped to 30%. This trend was also reported by several researchers in both polymer systems,^51,55,64,65 being connected with the poor adhesion of the BSG filler to both thermoplastic matrices.^66–68 This weak interfacial adhesion suggests that stress cannot be efficiently transferred from the polymer matrix to the filler. This behavior is fully consistent with SEM observations showing poor interfacial adhesion and microvoids at the BSG-matrix interface. Under tensile loading, BSG behaves as mechanically inert inclusions or stress concentrators, preventing efficient stress transfer (Figure 9).

Figure 8.

Comparison of maximum tensile strength predicted form Nicolais-Narkis model for BSG reinforced PP and PLA composite. Different letters (a,b,c,d,e,f) indicate statistically significant differences between samples.

The Nicolais-Narkis model (equation (2)), describes the tensile strength of composites under the assumption of no adhesion between filler and matrix fits our data well (Figure 8). Upon comparison of the two matrices, PLA aligns more closely with the model of no adhesion, whereas PP exhibits tensile strengths slightly above the model’s predictions. This slight deviation suggests marginally better wetting or mechanical anchoring of BSG particles within the PP matrix compared to PLA. Although both systems exhibit poor adhesion, the PP data indicate a slightly stronger interfacial interaction. This observation is consistent with our water contact angle measurements, which show that PLA surfaces (water contact angle of 99.6° ± 12) are more hydrophobic than PP (water contact angle of 78.8° ± 4) under our processing conditions, limiting PLA’s ability to wet hydrophilic BSG particles. Although it is rather unusual for PLA to exhibit a higher water contact angle than PP, our measurements align with the findings of Simon et al. who also reported a contact angle of approximately 100° for PLA.⁶⁹ This difference in interfacial wetting may explain the marginally better apparent adhesion observed in PP-BSG composites. Considering the adhesion between BSG and both matrix remains low, future research may focus on improving the interfacial interaction between BSG and the matrix. However, BSG thermoplastic composites can still perform well in injected applications where the strength is not a critical requirement or where the mechanical solicitation remains in its linear elastic region, offering a sustainable alternative for less mechanically demanding usages.

Interface morphology

By evaluating the interface between BSG and the polymer matrices, the tensile-fractured sections of both PLA and PP composites reveal differences in fracture behavior. Pure PLA exhibits a predominantly brittle fracture profile (Figure 9(b)), whereas the incorporation of BSG introduces stress concentration points characterized by ductile deformation around the filler particles (Figure 9(d) and (f)). These points of ductility increase with BSG content, suggesting that BSG reduces the brittleness of PLA which is a favorable improvement since PLA applications is limited by its high brittleness. SEM micrographs of PP composites show a similar shift in fracture morphology, from the uniform stretching seen in neat PP (Figure 9(a)) to shorter and focused stretching points around the fillers (Figure 9(c) and (e)). This shift reflects stress concentration at the filler-matrix interface, which results in localized weaknesses and indicates that BSG may disrupt the propagation of tensile stress.

Figure 9.

SEM Micrographs of samples fractured cross-section of tensile specimens: (a) Neat PP; (b) Neat PLA; (c) PPBSG30x200; (d) PLABSG30x200; (e) PPBSG50x500; (f) PLABSG30x500.

Additionally, both PLA and PP composites exhibit voids at the interface between BSG and the matrix, as well as particles pull-out under tensile stress, indicating poor adhesion, which is further worsened by the low elongation ratio of BSG particles. This lack of adhesion, likely due to the hydrophilic nature of BSG and the hydrophobic nature of the matrices, exacerbates the formation of microvoids around the interface. These microvoids seem to act as crack-blunting sites, allowing the material to deform more plastically before failure, as seen in the micrographs. In PLA, this enhanced ductility is supported by DSC results, where a reduction in T_g with BSG addition suggests increased flexibility in the polymer. Together, these observations underscore that BSG may alter the mechanical behavior of PLA and PP by enhancing ductility rather than stiffness, with microvoids and stress concentration points contributing to plastic deformation before fracture. The absence of adhesion likely contributes to a geometrical deformation of the polymer matrix due to the reduction of the polymer section. Moreover, the SEM observations reveal microvoids at the interface, which clearly indicate the absence of interfacial adhesion between BSG particles and both matrices. Under these conditions, load transfer from the matrix to the filler is essentially inhibited. This is fully consistent with rheological evidence showing that, in the molten state, the percolated structure corresponds only to a contact network created by particle–particle proximity rather than by true adhesive bonds. SEM images further confirm that the particles do not adhere either to one another or to the PLA matrix. Consequently, the percolated structure that may exist in the melt becomes purely frictional in the solid state and is not mechanically load-bearing. Under tensile loading, such loose particle assemblies act as defect zones rather than reinforcement, promoting debonding and premature stress concentrations. As a result, the BSG particles do not contribute to the composite’s stiffness and behave mechanically as inert inclusions rather than reinforcing elements. Therefore, the absence of an increase in Young’s modulus at high BSG loadings is fully consistent with the poor interface quality observed in the images.

Conclusions

The aim of this study was to conduct a comparative approach to explore the potential of BSG as a filler in thermoplastic composites. Therefore, the study focused on the description of the rheological, thermal, and mechanical behavior of the BSG-PLA and BSG-PP composites containing 0 to 50% (wt./wt.) of BSG. The findings were correlated with SEM images of cross-sections of the samples revealing the interface of the filler and matrix. One key result was that BSG impacted thermal stability of both composites, but with complete opposite effects. Indeed, whereas BSG acted as a degradation buffer probably through the formation of a char layer around PP, standardizing and delaying its degradation, BSG tended to provoke PLA degradation. As a result, the flow behavior of PLA composites and their thermal transition temperatures were affected. It was shown that this also affected the mechanical properties of the composites, as evidenced by the ductile behavior observed in the micrographs. Rheological analysis showed that highly filled PLA composites were challenging to obtain. Indeed, their rheological profile under low shear rates might be the sign of particle flocculation above 30% (wt./wt.) of BSG. In contrast, PP demonstrated consistent flow behavior with increasing BSG content, suggesting the potential for higher filler loadings without compromising processability. Finally, the tensile strength of both polymers was negatively affected by BSG, likely due to poor interfacial adhesion, as indicated by the microvoids in the SEM images. Therefore, future research may focus on improving the interfacial interaction between BSG and the matrix in order to fully develop the potential of BSG as a filler. However, a steady elastic modulus with increasing BSG content for both polymers demonstrated their potential to maintain mechanical integrity when subjected to stress. Despite, their mechanical limitations, BSG-composites offer advantages such as eco-friendliness and low price which make them attractive for specific fields where the strength requirements are moderate, but the environmental impact of the material is of significant concern. In conclusion, this study sheds light on the behavior of BSG-filled PLA and PP composites, particularly regarding their mechanical and rheological properties. The composites were successfully processed via injection molding, highlighting their suitability for applications where high strength is not essential. This work could be used as a step toward the valorization of BSG as a filler in sustainable materials, building a path for future research to address filler-matrix interactions and expand the range of applications.

Supplemental material

Supplemental Material - Thermal, rheological and mechanical behavior of PLA and PP biocomposites filled with Brewer’s spent grain (BSG): A comparative analysis of polymer-filler interactions

Supplemental Material for Thermal, rheological and mechanical behavior of PLA and PP biocomposites filled with Brewer’s spent grain (BSG): A comparative analysis of polymer-filler interactions by Adele Poirier, Pedro E. D. Augusto, Patrick Perré, Sébastien Alix, Catherine Lacoste in Journal of Thermoplastic Composite Materials.

Footnotes

Acknowledgements

Communauté urbaine du Grand Reims, Département de la Marne, Région Grand Est and European Union (FEDER Champagne-Ardenne 2014-2020, FEDER Grand Est 2021-2027) are acknowledged for their financial support to the Chair of Biotechnology of CentraleSupélec and Centre Européen de Biotechnologie et de Bioéconomie (CEBB).

Author contributions

Conceptualization: Adèle Poirier, Pedro E. D. Augusto, Patrick Perré Sébastien Alix, Catherine Lacoste; Methodology: Adèle Poirier, Pedro E. D. Augusto, Patrick Perré Sébastien Alix, Catherine Lacoste; Validation: Adèle Poirier; Formal analysis: Adèle Poirier; Investigation: Adèle Poirier, Patrick Perré, Catherine Lacoste; Data curation management: Adèle Poirier, Patrick Perré, Catherine Lacoste; Writing – original draft: Adèle Poirier; Writing – review and editing: Pedro E. D. Augusto, Patrick Perré, Sébastien Alix, Catherine Lacoste; Visualization: Adèle Poirier, Pedro E. D. Augusto, Catherine Lacoste; Supervision: Pedro E. D. Augusto, Patrick Perré, Sébastien Alix, Catherine Lacoste; Project administration: Pedro E. D. Augusto, Patrick Perré, Sébastien Alix, Catherine Lacoste; Funding acquisition: Pedro E. D. Augusto, Patrick Perré, Catherine Lacoste.

Declaration of conflicting interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is supported by Communauté urbaine du Grand Reims, Département de la Marne and Région Grand Est and European Union (FEDER Champagne-Ardenne 2014-2020, FEDER Grand Est 2021-2027).

ORCID iDs

Adele Poirier

Pedro E. D. Augusto

Patrick Perré

Data Availability Statement

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

Supplemental material

Supplemental material for this article is available online.

References

Dönmez Çavdar

Kalaycioǧlu

Mengeloǧlu

. Tea mill waste fibers filled thermoplastic composites: the effects of plastic type and fiber loading. J Reinforc Plast Compos 2011; 30: 833–844.

Yussuf

Massoumi

Hassan

. Comparison of polylactic Acid/Kenaf and polylactic Acid/Rise husk composites: the influence of the natural fibers on the mechanical, thermal and biodegradability properties. J Polym Environ 2010; 18: 422–429.

Panthapulakkal

Zereshkian

Sain

. Preparation and characterization of wheat straw fibers for reinforcing application in injection molded thermoplastic composites. Bioresour Technol 2006; 97: 265–272.

BarthHaas . BarthHaas Report.

Mussatto

Dragone

Roberto

. Brewers’ spent grain: generation, characteristics and potential applications. J Cereal Sci 2006; 43: 1–14.

Berthet

Angellier-Coussy

Machado

, et al. Exploring the potentialities of using lignocellulosic fibres derived from three food by-products as constituents of biocomposites for food packaging. Ind Crops Prod 2015; 69: 110–122.

Cunha

Berthet

Pereira

, et al. Development of polyhydroxyalkanoate/beer spent grain fibers composites for film blowing applications. Polym Compos 2015; 36: 1859–1865.

Hejna

Haponiuk

Piszczyk

, et al. Performance properties of rigid polyurethane-polyisocyanurate/brewers’ spent grain foamed composites as function of isocyanate index. E-Polymers 2017; 17: 427–437.

Zedler

Colom

Saeb

, et al. Preparation and characterization of natural rubber composites highly filled with brewers’ spent grain/ground tire rubber hybrid reinforcement. Compos B Eng 2018; 145: 182–188.

10.

Bledzki

Mamun

Volk

. Barley husk and coconut shell reinforced polypropylene composites: the effect of fibre physical, chemical and surface properties. Compos Sci Technol 2010; 70: 840–846.

11.

Sluiter

Templeton

. Determination of structural carbohydrates and lignin in biomass: Laboratory Analytical Procedure (LAP) (Revised July 2011). https://www.nrel.gov/biomass/analytical_procedures.html

12.

ISO 527-2:2012 . Plastics-determinations of tensile properties - part 2: test conditions for moulding and extrusion plastics, 2012. https://www.iso.org/

13.

Zhang

Shi

Nie

, et al. Study on poly(lactic acid)/natural fibers composites. J Appl Polym Sci 2012; 125: E526–E533. DOI: 10.1002/app.36852, Epub ahead of print 25 September.

14.

Wang

Huang

, et al. Heat resistance, crystallization behavior, and mechanical properties of polylactide/nucleating agent composites. Mater Des 2015; 66: 7–15.

15.

Liu

Guan

, et al. Rheological/crystallization behavior of PP/graphite nanosheet composites and performance of microcellular foaming plastics. Compos Commun; 32: 101133. DOI: 10.1016/j.coco.2022.101133, Epub ahead of print 1 June 2022.

16.

Test Method for Tensile Properties of Plastics . Epub ahead of print 15 December 2014. DOI: 10.1520/D0638-14.

17.

Sobek

Zeng

Werle

, et al. Brewer’s spent grain pyrolysis kinetics and evolved gas analysis for the sustainable phenolic compounds and fatty acids recovery potential. Renew Energy 2022; 199: 157–168.

18.

Almeida

Perré

. TGA-FTIR analysis of torrefaction of lignocellulosic components (cellulose, xylan, lignin) in isothermal conditions over a wide range of time durations. Bioresources 2015; 10: 4239–4251.

19.

Yang

Yan

Chen

, et al. In-depth investigation of biomass pyrolysis based on three major components: Hemicellulose, cellulose and lignin. Energy Fuels 2006; 20: 388–393.

20.

Inseemeesak

Siripaiboon

Somkeattikul

, et al. Biocomposite fabrication from pilot-scale steam-exploded coconut fiber and PLA/PBS with mechanical and thermal characterizations. J Clean Prod; 379: 134517. DOI: 10.1016/j.jclepro.2022.134517, Epub ahead of print 15 December 2022.

21.

Rahman

Hassan

Yahya

, et al. Micro-structural, thermal, and mechanical properties of injection-molded glass fiber/nanoclay/polypropylene composites. J Reinforc Plast Compos 2012; 31: 269–281.

22.

Mandal

Bhunia

Bajpai

. Thermal degradation kinetics of PP/PLA nanocomposite blends. J Thermoplast Compos Mater 2019; 32: 1714–1730.

23.

Stawski

. The effect of sample weight in thermogravimetric analysis of low viscosity polypropylene in air atmosphere. Polym Test 2009; 28: 223–225.

24.

Pǎrpǎriţǎ

Nistor

Popescu

, et al. TG/FT–IR/MS study on thermal decomposition of polypropylene/biomass composites. Polym Degrad Stabil 2014; 109: 13–20.

25.

Wladyka-Przybylak

Wesolek

Rojewski

, et al. Synergistic effect of modified natural fibres with halogen-free fire retardants in reducing flammability of composites. J Biobased Mater Bioenergy 2015; 9: 115–127.

26.

Canetti

Bertini

De Chirico

, et al. Thermal degradation behaviour of isotactic polypropylene blended with lignin. Polym Degrad Stabil 2006; 91: 494–498.

27.

Tian

Perré

. Effects of particle size on the pyrolysis of spruce and poplar: thermogravimetric analyses, DAEM modelling, validation, and prediction of secondary charring. Biomass Bioenergy 2023; 176: 106913.

28.

Zouari

Devallance

Marrot

. Effect of biochar addition on mechanical properties, thermal stability, and water resistance of hemp-polylactic acid (PLA) composites. Materials; 15: 2271. DOI: 10.3390/ma15062271, Epub ahead of print 1 March 2022.

29.

Rjeb

Labzour

Rjeb

, et al. TG and DSC studies of natural and artificial aging of polypropylene. In: Physica A: Statistical Mechanics and its Applications. Elsevier, 2005, pp. 212–217.

30.

Martí-Ferrer

Vilaplana

Ribes-Greus

, et al. Flour rice husk as filler in block copolymer polypropylene: effect of different coupling agents. J Appl Polym Sci 2006; 99: 1823–1831.

31.

Mofokeng

Luyt

Tábi

, et al. Comparison of injection moulded, natural fibre-reinforced composites with PP and PLA as matrices. J Thermoplast Compos Mater 2012; 25: 927–948.

32.

Jaszkiewicz

Meljon

Bledzki

, et al. Gaining knowledge on the processability of PLA-based short-fibre compounds - a comprehensive comparison with their PP counterparts. Compos Part A Appl Sci Manuf 2016; 83: 140–151.

33.

Saeidlou

Huneault

, et al. Poly(lactic acid) crystallization. Prog Polym Sci 2012; 37: 1657–1677.

34.

Zhang

Liang

Yan

, et al. Study of the molecular weight dependence of glass transition temperature for amorphous poly(l-lactide) by molecular dynamics simulation. Polymer (Guildf) 2007; 48: 4900–4905.

35.

Pan

Kai

Zhu

, et al. Polymorphous crystallization and multiple melting behavior of poly(L-lactide): molecular weight dependence. Macromolecules 2007; 40: 6898–6905.

36.

Di Lorenzo

. Calorimetric analysis of the multiple melting behavior of poly(L-lactic acid). J Appl Polym Sci 2006; 100: 3145–3151.

37.

Guo

Azaiez

Bellehumeur

. Rheology of fiber filled polymer melts: role of fiber-fiber interactions and polymer-fiber coupling. Polym Eng Sci 2005; 45: 385–399.

38.

Hausnerova

Honkova

Kitano

, et al. Superposed flow properties of ceramic powder-filled polymer melts. Polym Compos 2009; 30: 1027–1034.

39.

Chae

Kim

. Thermal and rheological properties of highly concentrated PET composites with ferrite nanoparticles. Compos Sci Technol 2007; 67: 1348–1352.

40.

Huq

AMA

Azaiez

. Effects of length distribution on the steady shear viscosity of semiconcentrated polymer-fiber suspensions. Polym Eng Sci 2005; 45: 1357–1368.

41.

Poslinski

Ryan

Gupta

, et al. Rheological behavior of filled polymeric systems I. Yield stress and shear‐thinning effects. J Rheol (N Y N Y) 1988; 32: 703–735.

42.

Ramazani

SAA

Ait-Kadi

Grmela

. Rheology of fiber suspensions in viscoelastic media: experiments and model predictions. J Rheol (N Y N Y) 2001; 45: 945–962.

43.

Carrot

Majesté

Olalla

, et al. On the use of the model proposed by Leonov for the explanation of a secondary plateau of the loss modulus in heterogeneous polymer-filler systems with agglomerates. Rheol Acta 2010; 49: 513–527.

44.

Anuar

Rahman

NAA

Manshor

, et al. Novel soda lignin/PLA/EPO biocomposite: a promising and sustainable material for 3D printing filament. Mater Today Commun; 35: 106093. DOI: 10.1016/j.mtcomm.2023.106093, Epub ahead of print 1 June 2023.

45.

. Polymer molecular-weight distribution from dynamic melt viscoelasticity. Polym Eng Sci 1985; 25: 122–128.

46.

Olalla

Carrot

Fulchiron

, et al. Analysis of the influence of polymer viscosity on the dispersion of magnesium hydroxide in a polyolefin matrix. Rheol Acta 2012; 51: 235–247.

47.

Kaully

Siegmann

Shacham

. Rheology of highly filled natural CaCO3 composites. II. Effects of solid loading and particle size distribution on rotational rheometry. Polym Compos 2007; 28: 524–533.

48.

Feger

Gelorme

McGlashan-Powell

, et al. Mixing, rheology, and stability of highly filled thermal pastes. IBM J Res Dev 2005; 49: 699–707.

49.

Pǎrpǎriţǎ

Darie

Popescu

, et al. Structure-morphology-mechanical properties relationship of some polypropylene/lignocellulosic composites. Mater Des 2014; 56: 763–772.

50.

Núñez-Decap

Wechsler-Pizarro

Vidal-Vega

. Mechanical, physical, thermal and morphological properties of polypropylene composite materials developed with particles of peach and cherry stones. Sustain Mater Technol; 29: e00300. DOI: 10.1016/j.susmat.2021.e00300, Epub ahead of print 1 September 2021.

51.

Kim

Lee

Kim

, et al. Thermal and mechanical properties of cassava and pineapple flours-filled PLA bio-composites. J Therm Anal Calorim 2012; 108: 1131–1139.

52.

Huda

Drzal

Misra

, et al. Wood-fiber-reinforced poly(lactic acid) composites: evaluation of the physicomechanical and morphological properties. J Appl Polym Sci 2006; 102: 4856–4869.

53.

Radzi

MKFM

Muhamad

Akhtar

, et al. The effect of kenaf filler reinforcement on the mechanical and physical properties of injection moulded polypropylene composites. Sains Malays 2018; 47: 367–376.

54.

Arib

RMN

Sapuan

Ahmad

MMHM

, et al. Mechanical properties of pineapple leaf fibre reinforced polypropylene composites. Mater Des 2006; 27: 391–396.

55.

Sutivisedsak

Cheng

Dowd

, et al. Evaluation of cotton byproducts as fillers for poly(lactic acid) and low density polyethylene. Ind Crops Prod 2012; 36: 127–134.

56.

Yang

Kim

Son

, et al. Rice-husk flour filled polypropylene composites; mechanical and morphological study. Compos Struct 2004; 63: 305–312.

57.

Huda

Drzal

Mohanty

, et al. Chopped glass and recycled newspaper as reinforcement fibers in injection molded poly(lactic acid) (PLA) composites: a comparative study. Compos Sci Technol 2006; 66: 1813–1824.

58.

Revert

Reig

Seguí

, et al. Upgrading brewer’s spent grain as functional filler in polypropylene matrix. Polym Compos 2017; 38: 40–47.

59.

Pickering

Efendy

MGA

. A review of recent developments in natural fibre composites and their mechanical performance. Compos Part A Appl Sci Manuf 2016; 83: 98–112.

60.

Hashin

Shtrikman

. A variational approach to the theory of the elastic behaviour of multiphase materials. J Mech Phys Solid 1963; 11: 127–140.

61.

Hoffmann

Huber

Mäder

. Nucleating and clarifying agents for polyolefins. In: Macromolecular Symposia. John Wiley and Sons Ltd, 2001, pp. 83–92.

62.

Feng

Lauke

, et al. Effects of particle size, particle/matrix interface adhesion and particle loading on mechanical properties of particulate-polymer composites. Compos B Eng 2008; 39: 933–961.

63.

Nicolais

Narkis

. Stress-strain behavior of styrene-Acrylonitrile/Glass bead composites in the glassy region. Polym Eng Sci 1971; 11(3): 194–199.

64.

Cheng

Lau

Liu

, et al. Mechanical and thermal properties of chicken feather fiber/PLA green composites. Compos B Eng 2009; 40: 650–654.

65.

Taflick

Maich

ÉG

Dias Ferreira

, et al. Acacia bark residues as filler in polypropylene composites. Polimeros 2015; 25: 289–295.

66.

Bajpai

Singh

Madaan

. Comparative studies of mechanical and morphological properties of polylactic acid and polypropylene based natural fiber composites. J Reinforc Plast Compos 2012; 31: 1712–1724.

67.

Bax

Müssig

. Impact and tensile properties of PLA/Cordenka and PLA/flax composites. Compos Sci Technol 2008; 68: 1601–1607.

68.

Lim

. Fabrication and mechanical properties of completely biodegradable hemp fiber reinforced polylactic acid composites. J Compos Mater 2007; 41: 1655–1669.

69.

Simon

Klassen

Mottoul

, et al. Long-lasting hydrophilicity induced by ultraviolet light on surface modified hydrophobic polylactic acid. J Appl Polym Sci 2025; 142: e57009 DOI: 10.1002/APP.57009, Epub ahead of print 20 June 2025.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.57 MB