Eco-friendly disposable porous absorbents from gluten proteins through diverse plastic processing techniques

Journal of Cleaner Production 459 (2024) 142419 Available online 3 May 2024 0959-6526/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). Eco-friendly disposable porous absorbents from gluten proteins through diverse plastic processing techniques Mercedes A. Bettelli a , ** , Emmanuelle Traissac a , Athanasios Latras a , Mercedes Jimenez- Rosado b , Antonio Guerrero c , Richard T. Olsson a , Mikael S. Hedenqvist a , *** , Antonio J. Capezza a , * a Department of Fibre and Polymer Technology, Polymeric Materials Division, School of Engineering Sciences in Chemistry, Biotechnology and Health. KTH Royal Institute of Technology, Stockholm, 10044, Sweden b Departamento de Química y Física Aplicadas, Universidad de Le´ on, Campus de Vegazana, 24007, Le´ on, Spain c Departamento de Ingeniería Quimica, Universidad de Sevilla, 41012, Sevilla, Spain ARTICLE INFO Handling Editor: Maria Teresa Moreira Keywords: Gluten Proteins Plastic-free foams Thermoforming Sustainable absorbents ABSTRACT The production of biodegradable gluten-based protein foams showing complete natural degradation in soil after 26 days is reported, as an alternative to commercial foams in disposable sanitary articles that rely on nonbiodegradable materials. The foams were developed from an extensive evaluation of different foaming methodologies (oven expansion, compression moulding, and extrusion), resulting in low-density foams (ca. 400 kg/ m 3 ) with homogenous pore size distributions. The products showed the ability to absorb 3–4 times their weight, reaching ranges for their use as absorbents in single-use disposable sanitary articles. An additional innovative contribution is that these gluten foams were made from natural and non-toxic wheat protein, glycerol, sodium and ammonium bicarbonate, making them useful as fossil-plastic-free replacements for commercial products without the risk of having micro-plastic and chemical pollution. The impact of different processing conditions on forming the porous biopolymer network is explained, i.e., temperature, pressure, and extensive shear forces, which were also investigated for different pH/chemical conditions. The development of micro-plastic-free foams mitigating environmental pollution and waste while using industrial co-products is fundamental for developing large-scale production of single-use items. A sanitary pad prototype is demonstrated as an eco-friendly material alternative that paves the way for sustainable practices in manufacturing, and contributes to the global effort in combating plastic pollution and waste management challenges, Sustainable Development Goals: 12, 13, 14, and 15. 1. Introduction Wheat gluten (WG) is an abundant plant protein available as an agricultural by- or co-product currently used in food and non-food applications (Blomfeldt et al., 2011; Cho et al., 2010; Lens et al., 2003; van der Schaaf et al., 2020; Veraverbeke and Delcour, 2002). The global gluten production by 2020 was reported to be as high as ca. 1 million tonnes (Platform, 2021). WG consists mainly of two major protein fractions, i.e., gliadins and glutenin. Gliadins are monomeric protein fractions (30–80 kDa)(Kontogiorgos et al., 2007), representing the more viscous component of the gluten matrix (Kuktaite, 2004), whereas glutenins are the high molecular weight fraction (30–160 kDa) responsible for the extensive 3D network formation due to intermolecular disulphide bonds, which contribute to the elastic properties of the gluten matrix (Gao et al., 2010; Hern´ andez-Mu˜ noz et al., 2003; Veraverbeke et al., 1998). The formation of the WG network involves two crucial reactions: the oxidation of free sulfhydryl (SH) groups into disulfide bonds (SS) and the SH-SS exchange resulting from the rearrangement/cleavage of SS bonds. (Yue et al., 2019). The formation of WG networks confers gluten with high viscoelastic properties, which are relevant in the food industry to enhance the breadmaking performance of wheat flour (Day et al., 2006; Garg et al., * Corresponding author. ** Corresponding author. *** Corresponding author. E-mail addresses: [email protected] (M.A. Bettelli), [email protected] (M.S. Hedenqvist), [email protected] (A.J. Capezza). Contents lists available at ScienceDirect Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro https://doi.org/10.1016/j.jclepro.2024.142419 Received 29 November 2023; Received in revised form 6 March 2024; Accepted 30 April 2024 Journal of Cleaner Production 459 (2024) 142419 2 2019). However, the concern regarding celiac disease and protein allergy will likely reduce its use in food but open up for its consideration in non-food applications, e.g., as a sustainable raw material in the production of bioplastics (Kontogiorgos et al., 2007; Liu et al., 2016; Majzoobi and Abedi, 2014). Recent studies have shown that WG can be thermally processed as solid gluten-products with tailored properties and functionalities (oxygen barrier, electrical conductivity, etc.) using traditional polymer processing techniques (extrusion, injection moulding, compression moulding) (Bettelli et al., 2022; Capezza et al., 2020a, 2020b; Cho et al., 2011; Jim´ enez-Rosado et al., 2019; Joyce et al., 2021; Liu et al., 2023; Tuntachon et al., 2019; Veraverbeke and Delcour, 2002; Wu et al., 2014). Extrusion is interesting since it allows producing continuous bio-based materials in large quantities (Bettelli et al., 2022; Capezza et al., 2020b; Georges et al., 2018; Türe et al., 2011). Beyond the use of gluten in food and bioplastics, an emerging application lies in producing porous materials, which have been extensively studied using batch-wise free expansion, freeze-drying/solvent exchange, and supercritical drying techniques (Georges et al., 2018; Pietsch et al., 2017, 2019). The results are promising in application areas where synthetic foams are used, such as cushioning and absorbents in consumer products. Novel sustainable production methods of polymeric foams have been reported where greener polymer synthesis using traditional matrices (polyurethane) with recycled plastics are combined (Atta et al., 2013). Only a few studies that focus on WG and methods to produce porous structures on a large scale have been reported, as the materials have not shown the capability to form homogenous and stable porous structures when using non-toxic foaming agents (Bettelli et al., 2022). This is unfortunate as it limits the development of these types of materials into higher technical readiness levels (TRL). The high molecular weight of the WG, heat, and/or shear during extrusion processing alters the protein structure and properties in different ways, e.g., causing important conformational changes in the structure (Bettelli et al., 2022; Rahaman et al., 2015) thereby interfering with the foaming process and possibly hinders the formation of stable and continuous foams (van der Schaaf et al., 2020). Additionally, previous reports have shown that the polymerization of WG depends strongly on the pH of the mixture, the extent of hydrogen bonding, and hydrophobic interactions and can lead to a highly aggregated and cohesive material with strong non-covalent interactions that influence the final properties of extruded gluten materials (Bruyninckx et al., 2015; Kuktaite et al., 2004; Ogawa and Matsumura, 2021; Türe et al., 2011). However, no reports have considered the influence of the pH nor correlated how the processing with the same formulation alters the microstructure/physicochemical interactions for producing porous gluten-based materials. This article aims to evaluate the structure and properties of glycerolplasticized gluten foams, shedding light on the challenges and opportunities in utilizing wheat gluten for eco-friendly and cost-effective porous materials, giving an important contribution towards innovative plastic-free, zero-waste biopolymer foam manufacturing. Oven expansion, compression moulding, and extrusion were used as thermal processing routes and sodium and ammonium bicarbonate were used as foaming agents at different pHs. The results showed that shear forces combined with temperature during the foaming are key parameters impacting the gluten’s ability to form stable pores. Despite the challenges during gluten processing, porous absorbent prototypes, fully degradable, were produced with competitive prices and performance compared to synthetic commercial foams. Understanding the gluten matrix’s behaviour allows for designing protocols that allow for largescale production of gluten foams using continuous extrusion without compromising the structural porosity and foam properties. 2. Experimental 2.1. Materials The wheat gluten (WG) powder was supplied by Lantm¨ ammen Reppe AB, Lidk¨ oping, Sweden. According to the manufacturer, the composition was 85.2 wt% of the wheat gluten protein, 5.8 wt% of wheat starch, 1.2 wt% lipid, 0.9 wt%, and 6.9 wt% of ash and water. Glycerol (G,ACS reagent ≥99.5%), Sodium bicarbonate (SBC, NaHCO 3 , ACS reagent ≥98%), ammonium bicarbonate (ABC, NH 4 HCO 3 , ACS reagent ≥98%), and sodium hydroxide (using 1M, NaOH, ACS reagent ≥98%), were purchased from Sigma-Aldrich (Sweden). Hydrochloric acid (1M, HCl≥98%) was purchased by VWR International (Sweden). The defibrinated sheep blood was purchased from Håtunalab (Sweden). The water used was Milli-Q water (MQw, 18.2 MOhm.cm at 25 ◦C). 2.2. Foam preparation The sample preparation process is shown in Fig. 1. For every 50 g of plasticized gluten mixture (i.e., 35 g of WG powder homogenously mixed with 15 g of glycerol to form a 70/30 WG/G mixture), 2.5 g of SBC or ABC (5 g foaming agent/100g WG/G) were added. The ingredients were mixed until a homogeneous dough was obtained (see Fig. 1). The 70/30 gluten/glycerol weight ratio was selected based on previous work on gluten film extrusion (Türe et al., 2011). Moreover, for the samples with a foaming agent and pH pre-treatment, the preparation was the same as described above for WG/G, with the difference being that water (MQw) and buffer were added together before being mixed with the WG/G. The buffer/water ratios used in the gluten mixture before foaming to adjust the pH are described in Table 1. The MQw with pH adjusted towards more basic (B) and acidic (A) values were obtained using 1M NaOH and 1M HCl, respectively. Table 1 summarizes the final material compositions after several pre-trials where the appearance, densities, and foaming process were evaluated. The pH of the WG mixtures was monitored by using a pH meter equipped with a probe to measure semi-solid materials (Pro-ISM, Mettler Toledo) and corrected when needed. Finally, the mixtures were foamed through oven expansion (OE), compression moulding (CM), or foam extrusion (E). 2.2.1. Oven expansion (OE) The plasticized gluten mixture was placed into a rectangular silicone mould having twelve individual circular cavities (i.e., 3.5 g gluten mixture/cavity) with a diameter of ~26 mm and 15 mm depth (Fig. 1, Oven Expansion - OE). The mixture was heated in a Memmert UF 260 oven (Germany) (Fan: 50 % and Flap: 100 %) at 120 ◦C (with the oven already pre-heated) for 10 min. The process is shown in Fig. 1. 2.2.2. Compression moulding (CM) Compression moulding was performed in a platen hot press (Fontijne TP-400, The Netherlands) to simulate the material behaviour under pressure and elevated temperature, but in contrast to extrusion without extensive shear deformation. A 10 mm thick rectangular mould with twelve cavities having a diameter of ~25 mm was used (see Fig. 1, Compression Moulding- CM). Polytetrafluoroethylene (PTFE) sheets were placed on both sides of the sample to prevent the material from sticking to the metal during pressing, and 1 mm thick plates were placed between the press plates and the PTFE sheets to distribute the pressure in the mould equally. Ca. 3.5 g of the gluten mixture was placed in each cavity, and the subsequent pressing was performed using a force of 250 kN for 1 min at 120 ◦C. Subsequently, the samples were cooled to room temperature while remaining under pressure in the press. The total cycle until the removal of the sample from the press took ca. 5 min. 2.2.3. Extrusion (E) 50 g of gluten mixture was processed in a conical, fully intermeshing and co-rotating double-screw micro-compounder (DSM Xplore 5 cc, The M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 3 Netherlands). The screw L/D ratio was 8, the compression ratio was 3.3, and the circular die had a diameter of 2.8 mm. A screw speed of 90 rpm was used, and all heating zones were set to 120 ◦C for the sodium bicarbonate (SBC) systems and 70 ◦C for the ammonium bicarbonate (ABC) systems. All the herein processed samples (OE, CM, E) were dried overnight at 50 ◦C in a ventilated oven and stored in a desiccator containing silica gel at least 1 week before any test. The full description of the samples is given in Table 1. The reference samples prepared with glycerol were named WG, and those containing the foaming agents’ sodium and ammonium bicarbonate were labelled WG/SBC and WG/ABC, respectively. 2.3. Apparent density The apparent foam density ( ρ a ) was calculated by dividing the weight (m) of the dry samples in kg by their respective volume (V) in m 3 . Three samples from each formulation were measured, and their volume was calculated assuming a cylindrical shape. The dimensions of each sample were measured at three places with a Vernier caliper. The porosity of the foams was estimated with equation (1), using the determined WG/G solid material density (1280 ±20 kg/m 3 ) as reference. Porosity (%) = [1− ( ρ a/1280)]⋅100 (1) where ρ a is the apparent density of each material (in kg/m 3 ). 2.4. Scanning Electron Microscopy (SEM) The samples’ morphology and biodegraded samples microstructure in soil was observed using a Hitachi Tabletop SEM (TM-1000, Japan) at a 10 kV voltage. For the sample preparation, the specimen was frozen by immersing it in liquid nitrogen for 1 min and subsequently broken into pieces. These cryo-fractured pieces were placed onto a specimen holder using conductive carbon tape. The pore size was measured using the Image J software, with at least 50 measurements per sample, and the results were reported as the average value and standard deviation, including the pore size distribution. Pore sizes above 1 mm were not included in the average but are displayed in the pore size distribution. The images of the swollen materials in saline and lyophilized were obtained by using a Hitachi field emission scanning electron microscope (FE-SEM S-4800, Japan) with a voltage of 3 kV and sputtered for 45 s with a palladium/platinium (Pt/Pd) target in an Agar High-Resolution sputter coater (model 208RH). 2.5. Liquid uptake capacity The WG samples’ swelling/uptake capacity (FSC) was measured by first placing the sample into an empty plastic teabag, and then immersing it in a saline solution (0.9 wt% NaCl) or in defibrinated sheep blood during different periods of time: 1, 5, 10, 30 min and 24 h. The plastic teabag was filled with ca. 200 mg of the dry samples (W d ) before immersion, and then it was hooked to a glass rod and placed into beakers containing saline solution. The bags with the material were hung for 10 s and then removed from the liquid and touched gently against tissue Fig. 1. Experimental protocol used for the foam preparation following the different methods. M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 4 paper for 10 s to remove excess liquid. The wet sample was removed intermittently from the tea bag and weighed (W i ). The FSC results were calculated according to equation (2), and the results are reported as the average of triplicates. FSC (g/g) = Wi−Wd Wd (2) Once the longest FSC was determined (30 min), the teabags containing the swollen wheat gluten foams were centrifuged at 1230 rpm (250xG) for 3 min, and the bags’ weight was recorded (W c ). The FSC and centrifuge retention capacity (CRC) tests at 30 min are significant for assessing the properties of materials intended for sanitary articles. The CRC of the samples was calculated according to: CRC (g/g) = Wc−Wd Wd (3) 2.6. Fourier-Transform Infrared Spectroscopy (FTIR) The infrared spectra were obtained with PerkinElmer Spectrum 100 (USA), equipped with MKII Golden GateTM (Single-reflection ATR, Graseby Specac LTD, England). The scan resolution was 4.0 cm −1 with a scanning step of 1.0 cm −1 . 16 consecutive scans were used between 4000 and 600 cm −1 . 2.7. Degree of crosslinking 1–2 mg of a WG foam was immersed in 40 mL of a denaturing solution (0.086 mol/L Tris base, 0.045 mmol/L glycine, 2 mmol/L Ethylenediaminetetraacetic acid EDTA and 10 g/L sodium dodecyl sulphate (SDS) solution (pH 6 buffer)) for 2 h to denature the un-crosslinked proteins. Subsequently, the solutions were centrifuged at 10000xG for 10 min to obtain the protein extractability of each system. The soluble protein fraction contained in the denatured solution was determined by Lowry’s method (Markwell et al., 1978). Briefly, the soluble protein was mixed with the Folin-Ciocalteau reagent in an alkaline medium. This mixture produces a reaction and generates a blue-green colour that is measured using a Genesys-20 spectrophotometer (Thermo Scientific) at a wavelength of 750 nm. The amount of soluble protein is obtained using a calibration curve, and the crosslinking density was estimated by comparing the reference sample from each manufacturing system (i.e., WG-Nm-OE, WG-Nm-CM, and WG-Nm-E, 0% crosslinking) and the denaturing-agent solution (100 % crosslinking, no protein dissolved in the solution). 2.8. Dynamic mechanical properties Dynamic thermomechanical compression (DMA) tests were performed to assess the viscoelastic properties of the foam samples (Groenewoud, 2001; Menard and Menard, 2020). These were evaluated using a DMA805 (TA Instruments, USA) with a cylindrical plate-plate compression geometry (ø: 15 mm). First, strain sweep tests from 0.002 to 2 % (at 1.0 Hz at 25 ◦C) were run to determine the linear viscoelastic strain amplitude region, in which elastic and viscous moduli remain constant while a deformation is applied. Thus, this zone implies that the system is not altering its structure. This interval is defined by the critical strain, the last strain that meets the range criterion. Frequency sweep tests were also performed (at constant strain within the linear viscoelastic range (65% of critical strain), from 0.2 to 20 Hz at 25 ◦C). The storage (E ′ ) and loss (E ″ ) moduli were evaluated in the entire frequency range, and the damping (tan δ =E’’/E ′ ) was evaluated. The values of E’ and tan δ at 1.0 Hz and the critical strain were used to compare the different systems. All measurements were carried out in triplicates. 2.9. Compression and tensile mechanical properties The compression mechanical properties of the oven-expanded (OE) and compression moulded (CM) foams were measured using an Instron 5944 universal testing machine with a load cell of 500 N. The compression rate used was 10 %/min and five replicates were used for each sample. The elastic modulus (E) was determined from the slope below 5 % strain, according to the ASTM D1621-16 standard. The compression strength ( σ s ) was obtained as the stress yield when the yield occurred below 10% specimen deformation. When no deflection on the compression curve was observed below 10% of compression (densification of the porous foam), the stress was reported as the compression stress at 10% strain ( σ 10 ), including the compression stress at 50% strain ( σ 50 ). The toughness (area under the stress-strain curve) was also reported using 50 % strain as the limit (U 50 ). Tensile testing of the extruded samples was performed in the same Instron machine as the compression tests. The extension rate was 10 mm/min, and five specimens of each sample were tested. The stressstrain curve of each specimen was measured, and the elastic modulus (E) was obtained from the slope of the linear region (below 5 % strain). The tensile stress ( σ b ) and elongation at break ( ε b ) were measured as the highest value of stress and the strain at the last recorded data point before the break, respectively, and the toughness (U) was calculated as the area under the tensile curve. The specimens were conditioned at 50 % RH and 23 ◦C for at least 72 h before the compression and tensile testing. Table 1 Sample nomenclatures. Sample Name MQw (Wt.%) NaOH 1M (Wt. %) HCl 1M (Wt. %) pH † T processing (◦C) Process WG-Nm- OE – – – 6.7 120 Oven expansion (OE) WG/SBC- Nm-OE 5 – – 8.7 WG/SBC- B-OE 2 3 – 8.9 WG/SBC- A-OE – – 5 7.9 WG-Nm- CM – – – 6.7 120 Compression moulding (CM) WG/SBC- Nm-CM 5 – – 8.7 WG/SBC- B-CM 2 3 – 8.9 WG/SBC- A-CM – – 5 7.9 WG-Nm- E – – – 6.7 120 Extrusion (E) WG/SBC- Nm-E 5 – – 8.7 WG/SBC- B-E 2 3 – 8.9 WG/SBC- A-E – – 5 7.9 WG-Nm- E (70) – – – 6.7 70 WG/ABC- Nm-E (70) 5 – – 8.7 Note: OE: Oven expansion conditions: 120 ◦C for 10 min. CM: Compression Moulding conditions: force of 250 kN for 1 min at 120 ◦C. E: Extrusion conditions: 120 ◦C (±3 ◦C). The formulations were adjusted to basic (B), non-modified (Nm) and Acidic (A) values. The extrudate samples were extruded at a screw rotational rate of 90 rpm. † pH of the mixture before being processed. All formulations have a WG/G ratio of 70:30. The WG formulation with foaming agent contains 5 g foaming agent/100g WG/G. The ratios between buffer and water were selected based on previous optimization processes. M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 5 2.10. Mould resistance test A mould growth test was performed by placing approximately 2 g of the sample in a Tissue Culture Plate and sealing it in an air-tight plastic container with MQw on the bottom to obtain ~100 % relative humidity. The test was performed at 25 ◦C, and the appearance of the foams was recorded as a function of time. 2.11. Soil biodegradation test A study on the impact of wheat gluten foams on the environment was assessed using a soil degradation test. Approximately 1 cm of the WG foams were buried in an opaque plastic glass with a composting soil medium (71% calcium, 14% potassium, 10% nitrogen, and 2% phosphorus). After burying the samples, they were unearthed after specific periods, and excess soil was removed carefully from the sample surface. The visual appearance was captured, and the samples were buried again in the same position. According to ASTM D6400, the test finishes when the pieces of samples cannot be observed (<1 mm), and the material is considered biodegradable/compostable if it starts disintegrating within 12 weeks and fully degrades within 180 days (6 months). 2.12. Manufacturing an absorbent pad as a proof of concept The possibility of developing scalable and already-made porous pads following the methods described above was investigated. The samples were elaborated through oven expansion conditions (OE), and the sample composition was selected after observing the best foaming process. The mould used to shape the structures is shown in Fig. 2, with a thickness of 2.7 mm. The pad mould filled with 20 g of the material was heat treated in a Memmert UF 260 oven (Germany) (Fan: 50 % and Flap: 100 %) at 120 ◦C for 20 min and then dried at 40 ◦C for 24 h (Fig. 2). The microstructure of the WG pads was observed using SEM (as described in section 2.4), and the free swelling/uptake capacity in saline and blood and centrifuge retention capacity was also assessed (as mentioned in section 2.5). The liquid spreading properties of the pad samples in blood, were assessed by a visual absorption test (VAT), as previously described (Capezza et al., 2020a). Briefly, 1x1 (cm x cm) of the dry sample was placed on a Petri dish, and a droplet of defibrinated sheep blood (100 μ l) was gently placed on the specimen. The droplet was then allowed to be soaked in before the next droplet was added. The test stopped when a complete liquid saturation was visible. The VAT was calculated as grams of liquid required to reach saturation per gram of dry material and reported as the average of duplicates (g/g). The produced pad as proof of concept was labelled as PADS. Compression tests were performed on the PAD-WG materials using an Instron 5566 universal testing machine with a 10 N load cell (compression rate of 10%/min). A commercial sanitary pad containing polyurethane foam (PUR) as an absorbent core was used as a reference. Here, the compressive stress at 20% strain was reported to be able to compare the properties of the commercial pad with the WG pad. The tensile properties were assessed as described in section 2.9, with the difference here that dumbbell shapes were die-cut at random places on the gluten and commercial PUR pads. 2.13. Statistical analysis Statistical analyses were performed with the least significant difference (LSD) in Fisher’s procedure, evaluating the significance of the measurements (p <0.05, 95% confidence level). These tests were analysed by using the software Statgraphics 18. The results were represented as mean values with standard deviations, showing the significant differences from different superscript letters (P ±SD x ). At least triplicates were used in the analysis. 3. Results and discussion 3.1. Effect of the processing on the microstructure Fig. 3 and Fig. S1 show the different samples produced and their respective pore structure. The materials manufactured by oven expansion (OE), with and without the sodium bicarbonate (SBC) foaming agent, display a homogenous cylindrical structure without any apparent cell collapse after cooling. The reference sample manufactured without a foaming agent (i.e., WG-Nm-OE, Fig. 3a) showed only a porosity of 33 ± 5% and, consequently, a relatively high density (851 kg/m 3 ) and with an average pore diameter of 74 μ m (Table 2). The formation of small pores has been previously identified as a consequence of rapid moisture uptake by the dry hygroscopic WG powder just before the thermal processing (Bettelli et al., 2022). When SBC was added, the gluten mixture without pH adjustment (WG/SBC-Nm-OE) displayed a highly porous structure with a significantly lower density than WG-Nm-OE (Fig. 3, Table 2). This sample had the lowest density (410 kg/m 3 ) and, consequently, the highest porosity (68 ±1%) with a large average pore size of 236 μ m. Such density falls within the classification range of high-density foams (Alander et al., 2018; Liu and Chen, 2014). Overall, the alteration of the pH resulted in a slight increase in density with the smallest and more narrow pore size distribution more noticeable at basic pH conditions (Table 2, Fig. S2 a-d). Besides possible protein interactions at different pH, the result also indicates that the presence of NaOH and HCl could impair the foaming capacity of the bicarbonate and thereby the material’s density. The compression moulded (CM) materials became denser than the OE samples (Table 2), but the cylindrical sample shape was maintained with both methods (Fig. 3). The reference sample (WG-Nm-CM) had a density of 1120 kg/m 3 and porosity of 12 ±2%, which is significantly higher than the density or the sample produced by oven expansion (WG- Nm-OE) with 851 kg/m 3 (porosity of 33 ±5, Table 2). Moreover, the CM foams containing SBC (WG/SBC-Nm-CM, WG/SBC-B-CM, and WG/SBC- A-CM) resulted in higher densities ranging from 600 to 800 kg/m 3 with smaller pore sizes (141-126 μ m) and narrower pore size distributions compared to those of OE (Fig. S2 f-h). In addition, the porosity obtained in the CM process was more irregular than for OE, indicating that the pressure induced during the compression impacts the pore microstructure despite the short compression cycle (only 1 min). Fig. 3 f-h show that the CM resulted in denser structures than with OE. According to the SEM images, large pores in the CM samples (above 1 mm) are also Fig. 2. Experimental protocol used for the absorbent pad preparation as proof of concept. The scale bar in the final product (PADS) is 5 cm. M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 6 observed, but to a lower extent in the SBC-OE samples (Fig. S2 e-h and S2 b-d, respectively). The formulations without SBC (WG-Nm-OE, WG- Nm-CM, and WG-Nm-E) also showed a porous cross-section (Fig. 3 a, e, i, respectively). Fig. 3 i-l shows that the extruded samples (E), with and without SBC, behaved differently from those produced using OE and CM. The SEM images reveal that the extrusion of the material resulted in a more uneven structure, as have been shown in previous work (Li et al., 2018). The microstructure of the extruded WG/G material was non-uniform, with less easily observable cells. This was also the case with the extruded samples containing the SBC and at different pH’s (Fig. 3j-l). The collapsed and non-uniform microstructure obtained for the extruded samples containing SBC is suggested to stem from the high pressure/shear forces combined with the elevated temperature and high alkalinity induced by the presence of SBC. It should be noted that SBC thermally decomposes forming CO 2 , water, and sodium carbonate, the latter yielding increased alkalinity (Bettelli et al., 2022; Chen et al., 2018). Alkaline conditions have been previously shown to oxidize thiol groups in gluten, leading to more inter-molecular disulphide bonds and a more aggregated/crosslinked structure (Bettelli et al., 2022; Capezza et al., 2020b; Chen et al., 2018). Here, the production of a homogenous foam with a fully developed cell structure relies on a fine balance between an effective expansion by the foaming agent (without the collapse of the foam structure after the extruder’s die) and the avoidance of a too-aggregated protein (Jug´ e et al., 2023). The material was also extruded with ammonium bicarbonate (ABC) at a lower temperature to demonstrate the discussed effect of the SBC and the high temperature on the gluten network formation (sample WG/ ABC-Nm-E (70)). ABC does not increase the foam alkalinity since there are no residual reaction products in the foam; it thermally decomposes into carbon dioxide (CO 2 ), ammonia (NH 3 (g) ), and water at the temperature used here (70 ◦C). Extruding the material with ABC results in a larger expansion ratio (ER: 2.0, Fig. 3n, Table 2) and significantly lower density (553 kg/m 3 ) than the SBC foam (WG/SBC-Nm-E) (ER: 0.6 and 1000 kg/m 3 , Fig. 3j). Furthermore, Fig. S1 reveals that all formulations including SBC contained residual crystals from sodium carbonate (bright particles in Fig. S1a-l). Such crystals were not observed in the sample extruded with ABC, suggesting that the thermal degradation process is more efficient for ABC than SBC. With the current state-of-the-art, pores can only be generated in extruded gluten networks if a foaming agent decomposes into non-alkaline products and at low temperatures, thereby minimizing the possibilities of forming a highly aggregated/ crosslinked gluten network triggered by the combination of temperatures above 100 ◦C, shear forces and alkalinity. It was concluded that the most porous WG foams are obtained through free foaming (OE). The foaming efficiency and sample homogeneity were reduced significantly in the presence of large compressive and/or shear forces during the compression and extrusion processes, respectively. It is obvious that non-free expansion limits foamability, but the complex inter and intramolecular network reactions in gluten are also hindering the expansion and possibly promoting cell collapse at the exit of the extruder die. Fig. 3. Images of the WG foam samples and their SEM cross-sections: Oven expansion (a) WG-Nm-OE, (b) WG/SBC-Nm-OE, (c) WG/SBC-B-OE, (d) WG/SBC-A-OE; Compression moulding (e) WG-Nm-CM, (f) WG/SBC-Nm-CM, (g) WG/SBC-B-CM, (h) WG/SBC-A-CM and Extrusion (i) WG-Nm-E, (j) WG/SBC-Nm-E, (k) WG/SBC-B- E, (l) WG/SBC-A-E, (m) WG-Nm-E (70) and (n) WG/ABC-E (70). M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 7 3.2. Liquid swelling and retention capacity Fig. 4a shows that the oven-expanded samples (WG/SBC-Nm-OE and WG/SBC-A-OE) had the highest uptake of saline, with ca. 3.5 and 3.0 g/g after 24 h, respectively. The lowest uptake was observed for the samples without SBC and without pH adjustment (WG-Nm-OE and WG-Nm-CM), both showing ca. 0.5 and 1.25 g/g after 30 min and 24 h, respectively. The high uptake by the porous samples obtained from the OE method correlates well with their low densities and large pore sizes (Table 2), while the low uptake matches those being the densest (Fig. 3 and Table 2). Fig. 4a shows also the correlation between the manufacturing process and the liquid uptake. With the same formulation, the uptake decreased in the following order: oven-expansion, compression moulding and extrusion. For the sample with SBC without pH adjustment, the uptake went from 3.5 to 2.25 and finally to 1.25 g/g. The inset in Fig. 4a shows the microstructure of the WG/SBC-Nm-OE sample swollen in saline for 24 h, freeze-dried, and cryo-fractured. Saline salt crystals of ca. 0.8 μ m in size are visible and embedded in the gluten cell wall. This agrees with previous work and demonstrates that the liquid penetrates the pores and cell walls (Capezza et al., 2020a). Fig. 4b shows the different samples’ centrifuge retention capacity (CRC) values after 30 min of saline swelling and 3 min centrifugation (250xG). The WG/SBC-Nm-OE sample had the highest CRC (1.3 g/g). The reference sample had a CRC of 0.43 g/g, independent of the processing method. The decrease in porosity affected the CRC, which decreased in the following order: WG/SBC-Nm-OE (1.3 g/g), WG/SBC- Nm-CM (0.7 g/g) and WG/SBC-Nm-E (0.42 g/g). This demonstrates that the pore structure significantly affects the size of the CRC of proteinbased materials. Fig. 4c displays that all samples showed a rapid blood uptake, with the maximum uptake reached within 1 min with no sizeable further uptake up to 30 min. 30 min was chosen since it represents a relevant absorption time for menstruation sanitary products (Jug´ e et al., 2023). Here, the highest blood uptake was observed for the WG/SBC-Nm-OE sample (2.1 g/g) and the lowest for WG-Nm-OE (0.2 g/g). The results Table 2 Physical properties of the samples. Sample Expansion Ratio (ER) Apparent Density (kg/ m 3 ) Porosity with density (%) Pore size ( μ m)† WG/G solid material NM 1280 ±20 h – – WG-Nm-OE 851 ±70 f 33±5 c 74 ±57 a WG/SBC-Nm- OE 410 ±20 a 68±1 g 236 ± 147 a WG/SBC-B- OE 445 ±1 ab 65±0 f 123 ±85 a WG/SBC-A- OE 440 ±20 ab 66±1 fg 226 ± 179 a WG-Nm-CM 1120 ±20 h 12±2 a 50 ±39 a WG/SBC-Nm- CM 716 ±60 de 44±4 d 125 ± 110 a WG/SBC-B- CM 600 ±50 cd 53±4 e 141 ± 135 a WG/SBC-A- CM 800 ±70 ef 37±5 cd 126 ± 112 a WG-Nm-E 0.80 820 ±70 ef 36±5 c NM WG/SBC-Nm- E 0.60 1000 ±40 g 22±3 b NM WG/SBC-B-E 0.70 1140 ±90 h 11±7 a NM WG/SBC-A-E 0.70 1160 ± 100 h 15±2 a NM WG-Nm-E (70) 1.80 903 ±94 fg 29±7 bc 63 ±26 a WG/ABC- Nm-E (70) 2.10 553 ±96 bc 57±5 e 247 ± 106 b † The pore size distribution was obtained from the size measurements on the SEM images; values and standard deviations (±values) were based on a minimum of 50 measurements on each sample. The porosity was estimated using Equation (1). Pore sizes above 1 mm were not included in the average but are displayed in the pore size distribution. NM: Not measured or difficult to resolve the pores. Note: Different letters mean the values are significantly different (P <0.05) for one column. Fig. 4. (a) FSC results in saline solution (0.9 wt% NaCl in MQw) and the FE-SEM image of WG/SBC-Nm-OE foam structure after 24 h immersion in saline solution with an embedded salt crystal is shown in the inset (the scale bar is 1 μ m), (b) CRC (after 30 min swelling in saline), and (c) FSC results in defibrinated sheep blood. Note: Different letters (a*-d*) in Fig. 4b means that the values are significantly different (P <0.05). M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 8 align with the fact that the former sample has one of the largest pore sizes (236 μ m; see Table 2), while the latter is a high-density sample. Large pore sizes are relevant for a high uptake of denser fluids, such as blood. Among the CM samples, the more acidic WG/SBC-A-CM had the largest blood uptake (ca. 1.1 g/g), whereas the lowest uptake was observed for the sample without pH adjustment (WG-Nm-CM, 0.25 g/g). No significant difference between extruded samples could be determined from the uptake (Fig. 4c). 3.3. Effect of the processing on the aggregation of the protein structure The FTIR spectra of the samples manufactured by the different processes are shown in Fig. 5. All samples display a broad band at 3280 cm −1 ascribed to the protein’s O–H and N–H bond stretch vibrations, as well as those of the glycerol O–H (Parveen et al., 2019). The two absorption bands at around 2921 and 2870 cm −1 originated from C–H symmetric and asymmetric stretching vibrations, whereas the peaks at 1650, 1535 and 1245 cm −1 correspond to C – – O (amide I) stretching, N–H bond bending (amide II) and C–N and N–H stretching (amide III), respectively. (Deng et al., 2019; Dousseau and Pezolet, 1990). Bands originating from vibrations associated with C–C and C–O bonds/O–H deformation in glycerol were observed in the 800–1150 cm −1 region. (´ Avila-Martín et al., 2020; Azevedo et al., 2017). All the thermally processed samples, irrespective of the type of processing used (i.e., OE, CM, E), showed a change in the amide I profile compared to the protein raw material, suggesting an alteration of the secondary structure (Fig. 5a.1, b.1, c.1). Fig. 5a.1 shows that the oven-expansion process (OE), combined with SBC, resulted in a high degree of aggregation associated with the amide I peak in the β-sheet band region (Bettelli et al., 2022; Capezza et al., 2020b; Newson et al., 2015; Ullsten et al., 2009). However, when SBC was not included, the profile peaked in the α -helix/random conformation region, indicating a less aggregated structure (WG-Nm-OE in Fig. 5a.1). The degree of crosslinking of the materials obtained by using different processing techniques is shown in Table S1. The samples containing SBC obtained a lower crosslinking degree than the respective references despite their high content of β-sheets, suggesting a more soluble protein network (with fewer intermolecular crosslinks). The samples manufactured by compression moulding (CM) displayed FTIR spectra in the amide I region that all peaked in the α -helix/random conformation region, indicating that all samples had a lower degree of aggregation compared to the as-received WG powder (Fig. 5b.1). The results also indicate that the use of pressure led to a lower degree of aggregated gluten in the presence of SBC (compare the OE and CM curves in Fig. 5). As in the case of the OE samples, the use of SBC also yielded a lower degree of crosslinking compared to the SBC-free sample. It is important to mention that the formation of a less aggregated gluten network in CM, compared to the OE, can be explained partly by the shorter time at 120 ◦C (2–3 min, including cooling time) in the compression process. The short CM time matched the time at elevated temperature in the extrusion process. Fig. 5c.1 shows that the materials extruded at 120 ◦C were more aggregated, as suggested by a more intense and skewed absorbance towards lower wavenumber (due to increasing content of β-sheets) than for the CM samples. In contrast to the OE and CM processes, the extrusion yielded a more crosslinked material in the presence of SBC than without it (Table S1). This showed the strong effect of the combination of high temperature, high shear force, hydrostatic pressure and the alkalinity of SBC, leading to an increase of aggregated gluten with an intermolecular crosslinked network despite the short residence time of the mixtures in the extruder barrel (ca. 30 s). These SBC and temperature effects are confirmed when comparing the samples extruded at a lower temperature (at 70 ◦C) with ABC, which displayed a less aggregated and crosslinked system (Fig. 5c.1 and Table S1). The shape of the amide I region remained essentially the same at different pH, suggesting that the change in pH did not affect the conformation of the proteins to a large extent. Notably, the foams preadjusted to acidic pH and processed through CM and E showed the lowest crosslinking degree (Table S1). As reported in previous work, the result suggests protein hydrolysis occurs when elevated temperature is combined with pressure and acidic media (Garg et al., 2019). The protein denatures and exposes sulfhydryl and hydrophobic groups at pH values outside the WG isoelectric region (i.e., 6–9). These groups generate strong repulsive forces between negative (extreme alkaline pH >12) or positive (extreme acidic pH <1) charges along the protein chains, hindering protein molecules from associating and forming aggregated networks/crosslinks (Dousseau and Pezolet, 1990; Garg et al., 2019). 3.4. Effect of the processing on the mechanical properties The reference sample produced by oven expansion (WG-Nm-OE, Fig. 6a) had the highest compression Young’s modulus (0.36 MPa) and compression strength at 10 and 50 % strain of σ 10 : 0.039 and σ 50 : 0.57 MPa, respectively, followed by the SBC-foams produced at acidic/basic conditions (E: 0.33–0.34 MPa, σ 10 : 0.029–0.031 MPa and σ 50 : 0.019–0.23 MPa), as shown in Table 3. The increase in the shift of the densification region to higher strains in the SBC oven-expanded foams Fig. 5. Full FTIR spectra of the WG foams: (a) oven expansion, (b) compression moulding, and (c) extrusion, and their respective amide I region (a.1, b.1, c.1). M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 9 correlates well with their higher porosity than the reference sample (WG-Nm-OE) (Fig. 6a and Tables 2 and 3). Fig. 6b shows that the gluten material prepared by compression moulding (CM) without SBC had the highest compression modulus and compression strength at 10 and 50%, in line with this sample having the highest density (Tables 2 and 3). The reference WG-Nm-CM material also had a lower slope in the densification region than the SBC-foamed samples, suggesting that the reference’s solid phase was slightly weaker than the cell walls of the samples foamed with SBC (see Fig. 6b and Fig. S3). Note the low compression modulus of the WG/SBC-A-CM sample (0.04 MPa) compared to the other CM and OE samples (Table 3), which is probably due to this sample having the lowest degree of crosslinking among the samples tested (Table S1). This, in turn, could indicate excessive protein hydrolysis and weakening of the cell wall. The shape of the extruded (E) samples did not allow for accurate testing in compression. Hence, these were tested in tension instead to estimate differences in mechanical properties for the different extrusion grades and conditions. Fig. 6c shows that the tensile behaviour for the formulations, including those with SBC, had a considerably lower strain at break (5–10 %) than the reference sample (WG-Nm-E, >25%, Table 3). The trend also revealed a higher elastic modulus for the SBC- foamed materials than for the reference (Table 3 and Fig. 6c). This could indicate that unreacted sodium bicarbonate/sodium carbonate particles in the gluten cell walls (Fig. 3) acted as reinforcing/hard fillers in the polymer matrix, (Monta˜ no-Leyva et al., 2013) increasing the modulus and decreasing the strain at the break of the solid phase. Additionally, the materials that were extruded using a mixture with the pH modified to basic conditions (i.e., WG/SBC-B-E) had the best mechanical properties (high modulus, yield strength and strain at break), followed by those extruded with the pH-modified to acidic conditions (i.e., WG/SBC-A-E). In previous work, extruded solid WG films modified to pH values above the protein’s isoelectric region (9–13) increased the stress at break, contrasting those produced at acidic pH (2–4). The effect was ascribed to strong repulsive forces among positively charged WG chains at low pH, resulting in materials with lower strength and less intermolecular crosslinking (Gennadios et al., 1993; Jansens et al., 2013). For the foams made with ABC (Fig. 6d), both the reference WG/-Nm- E (70) and WG/ABC-Nm-E (70) presented similar mechanical properties, with high elongation at break. The ABC extruded sample behaved quite differently from those produced with SBC (Fig. 6c and d), with the largest elongation at break and toughness among all samples (see Table 3). However, it should be noted that U depends on several external factor such as mechanical stress situation and can only be compared among the herein samples (Brostow et al., 2011). A possible reason is that the temperature used for ABC was below the temperature region where self-crosslinking of gluten occurs (Fig. 5c.1) (Wu et al., 2014). It is worth mentioning that extrusion at 70 ◦C with SBC was performed without any successful pore formation (results not shown). The reason is that temperatures below 100 ◦C are too low for promoting thermal decomposition of NaHCO 3 in contrast with ABC (Jug´ e et al., 2023). Furthermore, it is well-known that SBC acts as an alkaline foaming agent, which could induce further crosslinking in the protein, reducing its elongation at break while increasing the Young modulus (Table 3). Fig. 6. Mechanical properties of WG foams: Compressive stress-strain curves of the foams manufactured by (a) oven expansion and (b) compression moulding, and representative tensile stress-strain curves of the extruded foams using foaming agents: (c) sodium bicarbonate and (d) ammonium bicarbonate. M.A. Bettelli et al. Journal of Cleaner Production 459 (2024) 142419 16 Kuktaite, R., Larsson, H., Johansson, E., 2004. Variation in protein composition of wheat flour and its relationship to dough mixing behaviour. J. Cereal. Sci. 40 (1), 31–39. https://doi.org/10.1016/j.jcs.2004.04.00. Lens, J.-P., De Graaf, L., Stevels, W., Dietz, C., Verhelst, K., Vereijken, J., Kolster, P., 2003. Influence of processing and storage conditions on the mechanical and barrier properties of films cast from aqueous wheat gluten dispersions. Ind. Crop. Prod. 17 (2), 119–130. https://doi.org/10.1016/S0926-6690(02)00092-4. Li, T., Guo, X.-N., Zhu, K.-X., Zhou, H.-M., 2018. Effects of alkali on protein polymerization and textural characteristics of textured wheat protein. Food Chem. 239, 579–587. https://doi.org/10.1016/j.foodchem.2017.06.155. Liu, B.Y., Zhu, K.X., Peng, W., Guo, X.N., Zhou, H.M., 2016. Effect of sequential hydrolysis with endo- and exo-peptidase on bitterness properties of wheat gluten hydrolysates. The Royal Sociaty of Chemistry Advances 6 (33), 27659–27668. https://doi.org/10.1039/C5RA28171G. Liu, P.S., Chen, G.F., 2014. General introduction to porous materials. In: Liu, P.S., Chen, G.F. (Eds.), Porous Materials. Butterworth-Heinemann, Boston, pp. 1–20. Liu, S., Bettelli, M.A., Wei, X., Capezza, A.J., Sochor, B., Nilsson, F., Olsson, R.T., Johansson, E., Roth, S.V., Hedenqvist, M.S., 2023. Design of hygroscopic bioplastic products stable in varying humidities. Macromol. Mater. Eng. 308, 2200630 https:// doi.org/10.1002/mame.202200630. Majzoobi, M., Abedi, E., 2014. Effects of pH changes on functional properties of native and acetylated wheat gluten. Int. Food Res. J. 21 (3), 219–1224. Markwell, M.A.K., Haas, S.M., Bieber, L., Tolbert, N., 1978. A modification of the Lowry procedure to simplify protein determination in membrane and lipoprotein samples. Anal. Biochem. 87 (1), 206–210. https://doi.org/10.1016/0003-2697(78)90586-9. Menard, K., Menard, N., 2020. Dynamic mechanical analysis. In: Menard, K., Menard, N. (Eds.), Basic Rheological Concepts: Stress, Strain, and Flow. CRC Press, Boca Raton. Monta˜ no-Leyva, B., Ghizzi, D., da Silva, G., Gastaldi, E., Torres-Ch´ avez, P., Gontard, N., Angellier-Coussy, H., 2013. Biocomposites from wheat proteins and fibers: structure/mechanical properties relationships. Ind. Crop. Prod. 43, 545–555. https://doi.org/10.1016/j.indcrop.2012.07.065. Mugita, Y., Koudounas, S., Nakagami, G., Weller, C., Sanada, H., 2021. Assessing absorbent products’ effectiveness for the prevention and management of incontinence-associated dermatitis caused by urinary, faecal or double adult incontinence: a systematic review. J. Tissue Viability 30 (4), 599–607. https://doi. org/10.1016/j.jtv.2021.07.002. Newson, W.R., Rasheed, F., Kuktaite, R., Hedenqvist, M.S., G¨ allstedt, M., Plivelic, T.S., Johansson, E., 2015. Commercial potato protein concentrate as a novel source for thermoformed bio-based plastic films with unusual polymerisation and tensile properties. Royal Soc. Chem. Adv. 5 (41), 32217–32226. https://doi.org/10.1039/ c5ra00662g. Ogawa, T., Matsumura, Y., 2021. Revealing 3D structure of gluten in wheat dough by optical clearing imaging. Nat. Commun. 12 (1), 1–10. https://doi.org/10.1038/ s41467-021-22019. Parveen, S., Chaudhury, P., Dasmahapatra, U., Dasgupta, S., 2019. Biodegradable protein films from gallic acid and the cataractous eye protein isolate. Int. J. Biol. Macromol. 139, 12–20. https://doi.org/10.1016/j.ijbiomac.2019.07.143. Perez-Puyana, V.M., Capezza, A.J., Newson, W.R., Bengoechea, C., Johansson, E., Guerrero, A., Hendeqvist, M.S., 2023. Functionalization routes for keratin from poultry industry side-streams— towards bio-based absorbent polymers. Polymers 15 (2), 351. https://doi.org/10.3390/polym15020351. Pietsch, V.L., Emin, M.A., Schuchmann, H.P., 2017. Process conditions influencing wheat gluten polymerization during high moisture extrusion of meat analog products. J. Food Eng. 198, 28–35. https://doi.org/10.1016/j.jfoodeng.2016.10.027. Pietsch, V.L., Werner, R., Karbstein, H.P., Emin, M.A., 2019. High moisture extrusion of wheat gluten: relationship between process parameters, protein polymerization, and final product characteristics. J. Food Eng. 259, 3–11. https://doi.org/10.1016/j. jfoodeng.2019.04.006. Platform, I., 2021. Global wheat gluten production reduces slightly but exports remain robust. https://www.globaltrademag.com/global-wheat-gluten-production-redu ces-slightly-but-exports-remain-robust/. (Accessed 22 August 2023). Rahaman, T., Vasiljevic, T., Ramchandran, L., 2015. Conformational changes of β-lactoglobulin induced by shear, heat, and pH-Effects on antigenicity. J. Dairy Sci. 98 (7), 4255–4265. https://doi.org/10.3168/jds.2014-9010. Rozali, S.N.M., Milani, E.A., Deed, R.C., Silva, F.V.M., 2017. Bacteria, mould and yeast spore inactivation studies by scanning electron microscope observations. Int. J. Food Microbiol. 263, 17–25. https://doi.org/10.1016/j.ijfoodmicro.2017.10.008. Singha, S., Mahmutovic, M., Zamalloa, C., Stragier, L., Verstraete, W., Svagan, A.J., Das, O., Hedenqvist, M.S., 2021. Novel bioplastic from single cell protein as a potential packaging material. ACS Sustain. Chem. Eng. 9 (18), 6337–6346. https:// doi.org/10.1021/acssuschemeng.1c00355. Slezak, R., Krzystek, L., Puchalski, M., Kruci´ nska, I., Sitarski, A., 2023. Degradation of bio-based film plastics in soil under natural conditions. Sci. Total Environ. 866, 161401 https://doi.org/10.1016/j.scitotenv.2023.161401. Trust, N.F., 2020. Heavy periods (Menorrhagia) and Dysfunctional uterine bleeding. htt ps://www.wuth.nhs.uk/media/18090/pl00829-heavy-periods.pdf. (Accessed 16 November 2023). Tuntachon, S., Sukolrat, A., Numnuam, A., Kaewtatip, K., 2019. Effect of kaolin content and sonication on the properties of wheat gluten composites. Powder Technol. 351, 66–70. https://doi.org/10.1016/j.powtec.2019.04.007. Türe, H., G¨ allstedt, M., Kuktaite, R., Johansson, E., Hedenqvist, M.S., 2011. Protein network structure and properties of wheat gluten extrudates using a novel solventfree approach with urea as a combined denaturant and plasticiser. Soft Matter 7 (19), 9416–9423. https://doi.org/10.1039/C1SM05830D. Türkkan, M., ¨ Ozcan, M., Erper, ˙ I., 2017. Antifungal effect of carbonate and bicarbonate salts against Botrytis cinerea, the casual agent of grey mould of kiwifruit. Akademik Ziraat Dergisi 6 (2), 107–114. https://doi.org/10.29278/azd.371066. Ullsten, N.H., Cho, S.-W., Spencer, G., G¨ allstedt, M., Johansson, E., Hedenqvist, M.S., 2009. Properties of extruded vital wheat gluten sheets with sodium hydroxide and salicylic acid. Biomacromolecules 10 (3), 479–488. Van der Schaaf, U.S., Schreck, J., Pietsch, V.L., Karbstein, H.P., 2020. Wheat gluten stabilized emulsions: influence of homogenization process, pH, and ethanol concentration on droplet breakup and stabilization. J. Food Eng. 287, 110136 https://doi.org/10.1016/j.jfoodeng.2020.110136. Veraverbeke, W.S., Delcour, J.A., 2002. Wheat protein composition and properties of wheat glutenin in relation to breadmaking functionality. Crit. Rev. Food Sci. Nutr. 42 (3), 179–208. https://doi.org/10.1080/10408690290825510. Veraverbeke, W.S., Verbruggen, I.M., Delcour, J.A., 1998. Effects of increased high molecular weight glutenin subunits content of flour on dough mixing behavior and breadmaking. J. Agric. Food Chem. 46 (12), 4830–4835. https://doi.org/10.1021/ jf980636y. Wei, X.-F., Ye, X., Hedenqvist, M.S., 2022. Water-assisted extrusion of carbon fiberreinforced wheat gluten for balanced mechanical properties. Ind. Crop. Prod. 180, 114739 https://doi.org/10.1016/j.indcrop.2022.114739. Wu, Q., Andersson, R.L., Holgate, T., Johansson, E., Gedde, U.W., Olsson, R.T., Hedenqvist, M.S., 2014. Highly porous flame-retardant and sustainable biofoams based on wheat gluten and in situ polymerized silica. J. Mater. Chem. A 2 (48), 20996–21009. https://doi.org/10.1039/c4ta04787g. Youssef, K., Sanzani, S.M., Ligorio, A., Ippolito, A., Terry, L.A., 2014. Sodium carbonate and bicarbonate treatments induce resistance to postharvest green mould on citrus fruit. Postharvest Biol. Technol. 87, 61–69. https://doi.org/10.1016/j. postharvbio.2013.08.006. Yue, Y., Liu, X., Wang, J., Jia, F., Wang, Q., Zhang, X., 2019. Change in physicochemical characteristics and molecular weight distribution of glutenin macropolymer induced by postharvest wheat maturation. Qual. Assur. Saf. Crop Foods 11 (8), 789–798. https://doi.org/10.3920/QAS2019.1658. Zhou, Y., Lin, X., Xing, Y., Zhang, X., Lee, H.K., Huang, Z., 2023. Per- and polyfluoroalkyl substances in personal hygiene products: the implications for human exposure and emission to the environment. Environ. Sci. Technol. 57 (23), 8484–8495. https:// doi.org/10.1021/acs.est.2c08912. Zumstein, M.T., Schintlmeister, A., Nelson, T.F., Baumgartner, R., Woebken, D., Wagner, M., Kohler, H.-P.E., McNeill, K., Sander, M., 2018. Biodegradation of synthetic polymers in soils: tracking carbon into CO 2 and microbial biomass. Sci. Adv. 4 (7), eaas9024 https://doi.org/10.1126/sciadv.aas9024. M.A. Bettelli et al.