Towards Next-Generation Sustainable Composites Made of Recycled Rubber, Cenospheres, and Biobinder

The utilisation of industrial residual products to develop new value-added materials and reduce their environmental footprint is one of the critical challenges of science and industry. Development of new multifunctional and bio-based composite materials is an excellent opportunity for the effective utilisation of residual industrial products and a right step in the Green Deal’s direction as approved by the European Commission. Keeping the various issues in mind, we describe the manufacturing and characterisation of the three-component bio-based composites in this work. The key components are a bio-based binder made of peat, devulcanised crumb rubber (DCR) from used tyres, and part of the fly ash, i.e., the cenosphere (CS). The three-phase composites were prepared in the form of a block to investigate their mechanical properties and density, and in the form of granules for the determination of the sorption of water and oil products. We also investigated the properties’ dependence on the DCR and CS fraction. It was found that the maximum compression strength (in block form) observed for the composition without CS and DCR addition was 79.3 MPa, while the second-highest value of compression strength was 11.2 MPa for the composition with 27.3 wt.% of CS. For compositions with a bio-binder content from 17.4 to 55.8 wt.%, and with DCR contents ranging from 11.0 to 62.0 wt.%, the compressive strength was in the range from 1.1 to 2.0 MPa. Liquid-sorption analysis (water and diesel) showed that the maximum saturation of liquids, in both cases, was set after 35 min and ranged from 1.05 to 1.4 g·g −1 for water, and 0.77 to 1.25 g·g−1 for diesel. It was observed that 90% of the maximum saturation with diesel fuel came after 10 min and for water after 35 min.


Introduction
In the modern world, human civilisation is suffering from many challenges, such as an extensive increase in the generated waste stream by plastic-material pollution and, at the same time, lacking new efficient (lightweight, recyclable, or decomposable, made of biosourced or recycled raw materials) materials.
Among various waste materials, cenosphere (CS) is a low-density (0.25-0.55 g·cm −3 ) [1], inert, nontoxic, nonflammable, powder-like material which is a part of fly ash. Cenospheres along with cenosphere and a natural binder, peat, in developing a composite material. These solutions are in line with the United Nations sustainable development goals by fostering the conversion of waste materials into value-added products.
We describe here the utilisation of devulcanised crumb rubber (DCR), homogenised peat (HP), and cenospheres (CS) for composite-material development with a bio-binder. This research is aimed to answer the question about what effect the main component DCRHP-CS content has on the composite material properties such as density, mechanical properties, and the absorption of water and oil products.

Raw Materials and Compositions
For the manufacturing of sustainable composite material in two forms, blocks and granules, a bio-binder made of HP, DCR, and CS was used.  Tables 1 and 2. For the production of the specimen, the wt.% of HP in wet condition (suspension with water content 85 wt.%) was used, but the real DCR, CS, and HP content after drying is also represented in Tables 1 and 2 for an understanding of the entire composition of the studied materials. Table 1. The composition of block and granules in a raw mixture (wet) and after drying, by wt.% (part I).  Table 2. The composition of block and granules in a raw mixture (wet) and after drying, by wt.% (part II).  For a better understanding, all the studied recipes are represented in a ternary composition diagram in Figure 1. Three groups of composition, classified by a cenosphere (CS) content in the wet composition of 0, 5, and 10 wt.% correspond to the sample series XX-XX-0, XX-XX-5, and XX-XX-10, respectively. content in the wet composition of 0, 5, and 10 wt.% correspond to the sample series X XX-0, XX-XX-5, and XX-XX-10, respectively. Natural peat (deposition Keizerpurvs, Cesis, Latvia) was preliminarily process through a hydrocavitation process for use as a bio-based binder. The raw peat (humid 65-70%) was mixed with water and processed in a high-speed multidisc mixer-disper (HSMD) with cavitation effect for obtaining the homogeneous water-peat slurry with d matter contents of 15 ± 1 wt.%. Raw peat agglomerates before, and peat particles (extrac from the suspension) after treatment by HSMD, are shown in Figure 2. The rotation speed of the HSMD used in the experiments was 8500-9000 min −1 , a the linear velocity of the working teeth was from 70 to 80 m·sec -1 . Therefore, the cavitat conditions required for slurry homogenisation were ensured. The technological sche and HSMD standard view are given in Figure 3. The treatment time by HSMD was 5 m and 45 kg of the total amount of HP was used to ensure a homogenous sludge-like HP Natural peat (deposition Keizerpurvs, Cesis, Latvia) was preliminarily processed through a hydrocavitation process for use as a bio-based binder. The raw peat (humidity 65-70%) was mixed with water and processed in a high-speed multidisc mixer-disperser (HSMD) with cavitation effect for obtaining the homogeneous water-peat slurry with dry matter contents of 15 ± 1 wt.%. Raw peat agglomerates before, and peat particles (extracted from the suspension) after treatment by HSMD, are shown in Figure 2.

Designation of the Composition
olymers 2020, 12, x 4 of 13 content in the wet composition of 0, 5, and 10 wt.% correspond to the sample series XX-XX-0, XX-XX-5, and XX-XX-10, respectively. Natural peat (deposition Keizerpurvs, Cesis, Latvia) was preliminarily processed through a hydrocavitation process for use as a bio-based binder. The raw peat (humidity 65-70%) was mixed with water and processed in a high-speed multidisc mixer-disperser (HSMD) with cavitation effect for obtaining the homogeneous water-peat slurry with dry matter contents of 15 ± 1 wt.%. Raw peat agglomerates before, and peat particles (extracted from the suspension) after treatment by HSMD, are shown in Figure 2. The rotation speed of the HSMD used in the experiments was 8500-9000 min −1 , and the linear velocity of the working teeth was from 70 to 80 m·sec -1 . Therefore, the cavitation conditions required for slurry homogenisation were ensured. The technological scheme and HSMD standard view are given in Figure 3. The treatment time by HSMD was 5 min and 45 kg of the total amount of HP was used to ensure a homogenous sludge-like HP. The rotation speed of the HSMD used in the experiments was 8500-9000 min −1 , and the linear velocity of the working teeth was from 70 to 80 m·s −1 . Therefore, the cavitation conditions required for slurry homogenisation were ensured. The technological scheme and HSMD standard view are given in Figure 3. The treatment time by HSMD was 5 min, and 45 kg of the total amount of HP was used to ensure a homogenous sludge-like HP.
The DCR used for current research is provided by company Rubber Products Ltd. (Riga, Latvia). The DCR is produced using mechanochemical technology [30]. The manufacturing process comprises the processing crumb rubber by grinding at 60-70 • C with devulcanisation agent (urea) addition. The final product represents a sponge-like aggregate of DCR (average devulcanised rubber contents-13.4 wt%). For the DCR milling de-agglomeration, an impact-type disintegrator DESI-15 (Desintegraator Tootmise OÜ, Estonia) at a rotation speed of 3000 min −1 was used. The DCR was milled in direct mode five times (passes). For the present study, a 0.25-2.0 mm fraction was used ( Figure 5). More details about DCR milling, particle size distribution, and morphology are described by Lapkovskis et al. [31].
For the production of the block, the components were manually mixed until homogeneous, then placed into plastic moulds of 140 × 180 × 20 mm 3 . Samples were dried at room temperature for 20 days. After drying, all specimens were demoulded and left for ambient drying for ten days. For removing any residual humidity, samples were dried at 105 • C for 48 h.
facturing process comprises the processing crumb rubber by grinding at 60-70 °C wi devulcanisation agent (urea) addition. The final product represents a sponge-like aggr gate of DCR (average devulcanised rubber contents-13.4 wt%). For the DCR milling d agglomeration, an impact-type disintegrator DESI-15 (Desintegraator Tootmise OÜ, Est nia) at a rotation speed of 3000 min −1 was used. The DCR was milled in direct mode fi times (passes). For the present study, a 0.25-2.0 mm fraction was used ( Figure 5). Mo details about DCR milling, particle size distribution, and morphology are described b Lapkovskis et al. [31]. For the production of the block, the components were manually mixed until hom geneous, then placed into plastic moulds of 140 × 180 × 20 mm 3 . Samples were dried room temperature for 20 days. After drying, all specimens were demoulded and left f For the granules, the components were manually mixed until homogeneous, then placed in a rotary-drum granulator with a drum diameter of 950 mm and rotation speed of 80 s −1 . Samples were dried at room temperature for 2 days. To remove any residual humidity, specimens were dried at 105 • C for 48 h. The standard production scheme of composite blocks and granules is illustrated in Figure 6.
x 6 of 13 ambient drying for ten days. For removing any residual humidity, samples were dried at 105 °C for 48 h. For the granules, the components were manually mixed until homogeneous, then placed in a rotary-drum granulator with a drum diameter of 950 mm and rotation speed of 80 s −1 . Samples were dried at room temperature for 2 days. To remove any residual humidity, specimens were dried at 105 °C for 48 h. The standard production scheme of composite blocks and granules is illustrated in Figure 6.

Liquid Adsorption
Determination of liquid (water and oil products) absorption was performed by immersing specimens in the liquid and checking the weight at a specific interval. The experiments were repeated five times for each composition/liquid, with a margin of error relative to the mean for each experiment. The liquid absorption (W) is calculated according to Equation (1): where m1-the mass of the sample saturated with liquid, g; m0-dry mass (before immersion) of the sample, g; and W-liquid absorption g/g.

Liquid Adsorption
Determination of liquid (water and oil products) absorption was performed by immersing specimens in the liquid and checking the weight at a specific interval. The experiments were repeated five times for each composition/liquid, with a margin of error relative to the mean for each experiment. The liquid absorption (W) is calculated according to Equation (1): where m 1 -the mass of the sample saturated with liquid, g; m 0 -dry mass (before immersion) of the sample, g; and W-liquid absorption g/g.

Used Equipment and Measurement Devices
A high-speed multidisc mixer-disperser with cavitation effect (HSMD) [32][33][34] was used for obtaining a homogeneous water-peat slurry with a dry-matter content of 15 ± 1 wt.%. The moisture content was determined using a moisture analyser Kern MRS 120-3. Measurements were repeated seven times using the standard deviation to determine the standard error from the arithmetic mean. The Clatronic Multi Food Processor KM3350 (Clatronic GmbH, Kempen, Germany) with stainless steel container and a rubber-coated anchor-type mixer was used for the wet-mixture preparation at a rotation speed of 60 min −1.

Morphology of the Obtained Biocomposite Block and Granules
The most characteristic differences of the obtained biocomposites morphology in the form of block and granules are shown in Figure 7. The most significant difference in the appearance of the obtained composites is noted for the block-shaped material with 0, 5, and 10 wt.% of CS. The specimens containing 100 wt.% of HP (composition 0-100-0) were intensely cracked after drying (Figure 7a), demonstrating a high shrinkage. This is attributed to the used HP without any additive containing 85 wt.% of water. Detailed visual inspection of the cracked specimen's parts, using magnification X50 times (Figure 7d) shows a dense non-porous structure with white, crystal-like inclusions-sand particles. After analysis in polarised light, mainly quartz particles and an admixture of limestone were discovered, these being a natural component of the Baltic-region peat. The addition of 5 wt.% of CS and/or 5 wt.% of DCR strongly minimised the shrinkage and cracking. The typical appearance of the 0-95-5, 5-95-0, and 5-90-5 specimens is illustrated in Figure 7b. However, in comparison with highly-loaded composition 20-70-10, its geometry differs from mould shape (Figure 7b,c). Nevertheless, it is necessary to consider that the real content of fillers CS and DCR is much higher ( Table 1, Table 2) because the water loss from HP increases the CS and CDR content in the composite. Specimens 0-95-5, 5-95-0, and 5-95-5 after drying have 0-72.7-27.3, 27.3-72.7-0, and 22.1-55.8-22.1 CDR-HP-CR mass ratio (or weight %), respectively. The shrinkage-ratio decrease has been reported by several works [2,35,36], mainly with a ceramic matrix material where a high shrinkage is usually observed during the drying and firing [2,37].
In contrasts with the block material, the 0-100-0 granules have no significant morphological differences with the other composition specimens (Figure 7g-i). All the manufactured granules are characterised by a near-spherical shape and the particle-size distribution for all composition was: 1-2 mm-7-15%, 2-6 mm-10-20, and 6-10 mm-60-70 wt.%. several works [2,35,36], mainly with a ceramic matrix material where a high shrinkage is usually observed during the drying and firing [2,37].

Mechanical Properties and Density of the Obtained Biocomposite Block and Granules
The obtained composites in the form of blocks were tested for compression strength and apparent density. The results are represented in a combined diagram in Figure 8. It can be seen that the highest compression strength of 79 MPa corresponds to the pure peat-based bio-binder (0-100-0). The second-highest compression strength of 11 MPa corresponds to the 0-100-5 composition with 5 wt.% of CS in the raw wet mixture or 27.3 wt.% in the composite material after drying ( Table 1). The observation of the parts of the cracked specimens 0-100-5 (with 27.3 wt.% of CS) revealed a dense structure without cracks or voids, the same as 0-100-0 (100 wt.% of HP, Figure 7d In all the studied cases, an increase of the CDR in the composites leads to a significant decrease of compression strength, up to 1.5 ± 0.4 MPa, but not less than 1.1 MPa (10-80-10 and 20-70-10). By applying the determined physical-mechanical properties data of the obtained samples to Ashby's [38] compression strength and density summary diagram (Figure 9), it can be concluded that the obtained material demonstrates a relatively low density and relatively high strength, characteristic of biocomposites, which is one of the key aims of this work. Pure bio-binder (0-100-0) composite material in units MPa-kg·m −3 , is characterised by such property combinations that it is located near to the three different types of materials (metals, ceramics, and polymers), which is a unique properties combination and much materials belong to such property's combinations. Compositions 5-XX-XX and 10-XX-XX with 5 and 10 wt.% of DCR content in wet mixture and units MPa-kg·m −3 , belong to the lower zone of the natural material area. In all the studied cases, an increase of the CDR in the composites leads to a significant decrease of compression strength, up to 1.5 ± 0.4 MPa, but not less than 1.1 MPa (10-80-10 and 20-70-10).
By applying the determined physical-mechanical properties data of the obtained samples to Ashby's [38] compression strength and density summary diagram (Figure 9), it can be concluded that the obtained material demonstrates a relatively low density and relatively high strength, characteristic of biocomposites, which is one of the key aims of this work. Pure bio-binder (0-100-0) composite material in units MPa-kg·m −3 , is characterised by such property combinations that it is located near to the three different types of materials (metals, ceramics, and polymers), which is a unique properties combination and much materials belong to such property's combinations. Compositions 5-XX-XX and 10-XX-XX with 5 and 10 wt.% of DCR content in wet mixture and units MPa-kg·m −3 , belong to the lower zone of the natural material area.

Sorption of Liquids in the Structure of the Granulated Biocomposites
The obtained biocomposite granules were used for sorption of water and oil products (diesel). Sorption kinetics were estimated for the developed biocomposite using diesel fuel as a model compound, as demonstrated in Figure 10 and Figure 11. All samples reached a 90% water-sorbent uptake capacity in 25-30 min, with maximal saturation after 35-45 min Figure 10. All the samples' series demonstrated a near 1.0 g·g −1 water-sorption-capacity saturation. A 90% sorbent uptake capacity was noted for the diesel in a shorter time, in 5-10 min, with a maximal saturation after 35-45 min Figure 11. The samples' series demonstrated from 1.0 to 1.5 g·g -1 diesel sorption capacity at equilibrium conditions. The highest adsorption capacity was 1.5 g·g -1 for specimen 30-65-5, which corresponds to a 68.0-20.6-11.3 ratio of the dry composite components. It is necessary to admit that liquid's maximal saturation was for diesel, with maximal saturation reached within 3-5 min. Figure 12 illustrates the water and diesel uptake capacity, in g/g, for granules, and it can be seen that for most cases (except 30-70-0, 5-90-5, 15-80-5, and 20-75-5), there is greater sorption for water. For the composition series XX-XX-0 and XX-XX-10, the water uptake is significantly higher than for diesel, from 10 to 50%, but for the XX-XX-5 series, there is no significant difference between the water and diesel uptake. However, considering the sorption-capacity ratio from the mass ratio [g·g -1 ] of the sorbent mass to the ab-

Sorption of Liquids in the Structure of the Granulated Biocomposites
The obtained biocomposite granules were used for sorption of water and oil products (diesel). Sorption kinetics were estimated for the developed biocomposite using diesel fuel as a model compound, as demonstrated in Figures 10 and 11. All samples reached a 90% water-sorbent uptake capacity in 25-30 min, with maximal saturation after 35-45 min Figure 10. All the samples' series demonstrated a near 1.0 g·g −1 water-sorption-capacity saturation. A 90% sorbent uptake capacity was noted for the diesel in a shorter time, in 5-10 min, with a maximal saturation after 35-45 min Figure 11. The samples' series demonstrated from 1.0 to 1.5 g·g −1 diesel sorption capacity at equilibrium conditions. The highest adsorption capacity was 1.5 g·g −1 for specimen 30-65-5, which corresponds to a 68.0-20.6-11.3 ratio of the dry composite components. It is necessary to admit that liquid's maximal saturation was for diesel, with maximal saturation reached within 3-5 min.    Figure 12 illustrates the water and diesel uptake capacity, in g/g, for granules, and it can be seen that for most cases (except 30-70-0, 5-90-5, 15-80-5, and 20-75-5), there is greater sorption for water. For the composition series XX-XX-0 and XX-XX-10, the water uptake is significantly higher than for diesel, from 10 to 50%, but for the XX-XX-5 series, there is no significant difference between the water and diesel uptake. However, considering the sorption-capacity ratio from the mass ratio [g·g −1 ] of the sorbent mass to the absorbed-liquid volume [cm 3 ·g], the sorbent capacity for diesel is higher by 15%. The diesel density was assumed as 0.85 g·cm −3 .
Polymers 2020, 12, x 11 of 13 Figure 12. Sorbent water and diesel uptake capacity in g/g for granules compositions.

Conclusions
In the current research, a three-phase composite material containing homogenised peat as a bio-binder for water and oil products was produced in the form of blocks and granules for the first time. The obtained material in the form of a block was characterised by the right combination of compressive strength and density.
The obtained granulated sorbent containing 68.0-20.6-11.3 of CDR HP and CS demonstrated up to 1.5 g·g −1 maximal sorption capacity for diesel.
The composite material with CS content of 27.3 wt.% is characterised by the highest value (except for the pure bio-binder) of compression strength of 11.2 MPa and at the same time an apparent density of 0.75 g·cm −3 . HP as a bio-binder and CS as a lightweight filler could become a prospective material for designing lightweight bio-based structures. Further investigations of CS content's influence on the CS-HP biocomposite are foreseen and usage for acoustic and thermal insulation to be explored. Funding: This research was equally funded by two grants: 1) by Riga Technical University's Doctoral Grant programme; 2) This project has been supported by the Latvian Council of Science within the scope of the project "Innovative bio-based composite granules for collecting oil spills from the water surface (InnoGran)" (No. lzp-2020/2-0394). The work resulting within the network collaboration in the frame of COST Actions CA17133 Circular City ("Implementing nature-based solutions for creating a resourceful circular city" and CA18224 GREENERING ("Green Chemical Engineering Network towards upscaling sustainable processes"). COST Actions are funded within the EU Horizon 2020 Programme. The authors are grateful for the support.

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
In the current research, a three-phase composite material containing homogenised peat as a bio-binder for water and oil products was produced in the form of blocks and granules for the first time. The obtained material in the form of a block was characterised by the right combination of compressive strength and density.
The obtained granulated sorbent containing 68.0-20.6-11.3 of CDR HP and CS demonstrated up to 1.5 g·g −1 maximal sorption capacity for diesel.
The composite material with CS content of 27.3 wt.% is characterised by the highest value (except for the pure bio-binder) of compression strength of 11.2 MPa and at the same time an apparent density of 0.75 g·cm −3 . HP as a bio-binder and CS as a lightweight filler could become a prospective material for designing lightweight bio-based structures. Further investigations of CS content's influence on the CS-HP biocomposite are foreseen and usage for acoustic and thermal insulation to be explored. Funding: This research was equally funded by two grants: (1) by Riga Technical University's Doctoral Grant programme; (2) This project has been supported by the Latvian Council of Science within the scope of the project "Innovative bio-based composite granules for collecting oil spills from the water surface (InnoGran)" (No. lzp-2020/2-0394). The work resulting within the network collaboration in the frame of COST Actions CA17133 Circular City ("Implementing nature-based solutions for creating a resourceful circular city" and CA18224 GREENERING ("Green Chemical Engineering Network towards upscaling sustainable processes"). COST Actions are funded within the EU Horizon 2020 Programme. The authors are grateful for the support.