**Contents**



### **About the Editor**

**Vincenzo Fiore** is an Associate Professor in Technology and Material Science at the Engineering Department of the University of Palermo. He was a Research Fellow at the University of Palermo from April 2012 to August 2020.

He graduated with honors in "Material Engineering" from the University of Messina in July 2004, and he then earned his PhD in "Economic analysis, technological innovation and management for territorial development policies" from the University of Palermo in April 2008, presenting a thesis on "Innovative technologies for the manufacturing of composite structures in the nautical field".

His research interest is focused on fiber-reinforced composite materials, with an emphasis on the following topics:

—Manufacturing and testing of fiber-reinforced composite materials;

—Extraction and characterization of new lignocellulosic fibers to be used for the reinforcement of polymeric matrices;

—Manufacturing and testing of adhesive, mechanical and mixed joints between similar and dissimilar materials (i.e., metal to composite, glass to metal, glass to composite);

—Manufacturing and testing of new eco-friendly materials with enhanced insulating properties;

—Evaluation of the aging resistance of traditional and innovative materials in hostile environments; —Analysis of the viscoelastic behavior of metal, glass, composite structures and natural materials

He is the author/co-author of more than 50 publications in peer-reviewed journals, 2 patents, 4 book chapters and more than 30 conference presentations, seminars and invited lectures. He has supervised/co-supervised more than 60 master's theses and has more than 10 years of teaching experience.

### **Preface to "Natural Fibres and their Composites"**

Natural fibers have several promising characteristics, such as low density and other specific properties, low price, easy processing, health advantages, renewability and recyclability. As a result, natural fibers have received increasing attention over the last several years as alternatives to their synthetic counterparts for the reinforcement of polymer-based composites.

However, it is well known that the hydrophilic nature of natural fibers renders them highly susceptible to moisture absorption and low resistance to humid and wet environments. Moreover, these fibers exhibit limited and highly variable mechanical properties, as well as weak adhesion to hydrophobic polymers. For these reasons, the use of natural fibers in industrial applications, such as those in the automotive, marine and infrastructure industries, are often limited to non-structural or semi-structural interior components.

To overcome these drawbacks, natural fibers can be pre-treated or used in combination with specific additive and/or synthetic fibers. These approaches have been widely exploited in the literature, and the resulting composites have shown a suitable balance of mechanical properties, thermal stability, aging tolerance in humid or aggressive environments, cost and environmental friendliness.

In this context, the present Special Issue comprises 14 peer-reviewed original research articles about polymer composite materials reinforced with natural fibers.

In particular, the main topics include

—Investigation of the effect of novel natural reinforcements and nano and/or micro additives on the performance of thermoplastic or thermoset-based composites;

—Use of natural fibers in hybrid composites for different applications;

—Evaluation of basalt fiber as an eco-friendly alternative to glass fiber;

—Analysis of the mechanical and rheological responses of composites reinforced with different natural fibers;

—Use of recycled natural fibers for the reinforcement of polymeric composites;

—Theoretical modeling of the mechanical behavior of natural fiber-reinforced composites.

All of the papers included in this Special Issue provide some very valuable insights into the use of natural fibers for the reinforcement of composite materials. Thus, this volume will be useful for students, designers and engineers who aim to develop a deeper understanding on this important and emerging research subject.

### **Vincenzo Fiore** *Editor*

ix

### *Editorial* **Natural Fibres and Their Composites**

### **Vincenzo Fiore**

Department of Engineering, University of Palermo, Viale delle Scienze, Edificio 6, 90128 Palermo, Italy; vincenzo.fiore@unipa.it

Received: 28 September 2020; Accepted: 10 October 2020; Published: 15 October 2020

Due to several promising properties, such as their low density and specific properties, low price, easy processing, health advantages, renewability and recyclability, increasing attention was paid in the last years to natural fibres as alternatives to synthetic counterparts for the reinforcement of polymeric based composites.

On the other hand, it is well known that the hydrophilic nature of natural fibres leads to high susceptibility to moisture absorption and low resistance to humid and wet environmental conditions. Moreover, these fibres evidence limited and highly variable mechanical properties as well as weak adhesion with hydrophobic polymers. For these reasons, the use of natural fibres in industrial applications such as automotive, marine and infrastructure, are often limited to non-structural or semi-structural interior components.

To overcome these drawbacks, natural fibres can be pre-tread or used together with specific additive and/or synthetic fibres. These approaches have been widely exploited in literature, and the resulting composites have shown a suitable balance of mechanical properties, thermal stability, ageing tolerance against humid or aggressive environments, cost and environment care.

In this context, the present Special Issue comprises fourteen peer-reviewed original research articles about polymer composite materials reinforced with natural fibres.

The main topic includes the investigation of the effect of novel lignocellulosic reinforcements and nano or micro additives on performances of thermoplastic or thermoset based composites.

The use of novel lignocellulosic fibres is studied in several contributions to this Special Issue. The team of Andrea Zille [1] analyzes the effect of adding dog wool fibres on the properties of polyurethane (PU)-based eco-composite foam. They show that tensile and compression strengths, hydration capacity and thermal capacity are improved whereas the foam dilatation with heating decreased with increasing the amount of dog wool microparticles, thus demonstrating the potential of this animal-derived waste for insulation applications, with a low cost and minimal environmental impact. Mina Hernandez et al. [2] develop a fully bio-based composite using a natural resin obtained from Mopa-Mopa (Elaeagia Pastoensis Mora) plant as matrix and fique fibres as reinforcement. Thanks to easy processing and good physicochemical and mechanical characteristics, it is shown that the bio-based composite can be used as wood–plastic for the replacement of plastic and/or natural wood products widely used today in several applications. Similarly, Pompei et al. [3] show that the introduction of leather fragments in a self-produced thermoplastic starch (TPS) based on starch plasticized with glycerol and cross-linked using citric acid proved to be promising. The composite biodegradability allows its possible application in products where contact with soil and progressive non-toxic degradation is required, such as it is the case for the on-site production of mulching films.

Wang et al. [4] studied the relationships between the working fluids, process characteristics and products from the modified coaxial electrospinning of zein focusing their attention on the control of the processing process during the manufacturing process as well as on the prediction and maintenance of the nanofibre quality. In particular, using an electrospinnable zein solution as the core fluid and LiCl solutions as the sheath working liquids, a series of modified coaxial electrospinning processes are performed, thus successfully preparing a number of zein nanoribbons. In a further paper [5] concerning the use of novel lignocellulosic fibres, a mixture of thermoplastic polybutylene succinate (PBS), tapioca starch, glycerol and empty fruit bunch fibre was prepared by a melt compounding method and characterized by means of mechanical, thermal and immersion tests.

The effect of the addition of nano or micro additive as filler in natural fibre reinforced composite is investigated in several papers [6–9].

In this context, the team of Andrea Lazzeri [6] prepared plasticized poly(lactic acid) (PLA)/poly(butylene succinate) (PBS) blend-based films containing chitin nanofibrils and calcium carbonate by extrusion and compression moulding methods. The diffusion coefficient experimental data shows that the adding of chitin nanofibrils can slow the plasticizer migration. However, the best result was achieved with micrometric calcium carbonate while nanometric calcium carbonate results less effective due to bio polyesters' chain scission. On the other hand, the use of both micrometric calcium carbonate and chitin nanofibrils was counterproductive due to the agglomeration phenomena that were observed. Russo et al. [7] evaluate the effects of two carbon nanostructures (graphene nanoplatelets (GNPs) and carbon nanotubes (CNTs)), of a chemical modification with a fatty acid and of maleated polypropylene, with the aim of mitigating the highly hydrophilic nature of flax fibres thus increasing their compatibility with apolar polypropylene. On the bases of the experimental data, the authors state that these simple treatments, potentially prone to further optimization, can represent a step toward producing natural fibre composites with mechanical profiles compatible with semi-structural applications. Similarly, Wang et al. [8] improved the hygrothermal resistance of flax fibre reinforced epoxy composites through grafting flax fabric with nanoclay. In more detail, the introduction of nano-clay reduces both moisture uptake and the coefficient of diffusion of composites that show better dimensional stability than the untreated ones.

In the paper by Costa et al. [9], the performance of graphene oxide coated curaua fibre reinforced epoxy composites in a multilayered armour system intended for ballistic protection is evaluated showing that this innovative composite attends the standard ballistic requirement remaining intact, differently from the non-coated curaua fibre similar composite.

Another topic addressed in this special issue deals with the use of natural fibres in hybrid composites useful for different applications. Dhakal et al. [10] present an interesting study on the evaluation of the low-velocity falling weight impact behaviour of flax-basalt vinyl ester (VE) hybrid composites, showing that the hybrid system possesses higher impact energy and peak load than flax fibre reinforced composite. Hence, the experimental results indicate that the hybridization is a promising strategy for enhancing the toughness properties of natural fibre composites. Flax, Basalt, E-Glass FRP and their hybrid combinations are used to strengthen wood beams in the paper by Wang et al. [11]. The bending tests performed on strengthened wood beams shows that all hybrid FRPs exhibit no significant enhancement in load carrying capacity but larger maximum deflection compared to the single type of FRP composite.

The last three papers belonging to this special issue deals with the investigation of flexural properties and impact damage behaviour of basalt fibre reinforced polypropylene composites [12], the mechanical and rheological behaviour of composites reinforced with different natural fibres [13] and the theoretical modelling of the stiffness of recycled cotton fibres reinforced polypropylene composites [14]. In particular, the paper by Serra et al. [14] highlight the opportunity of recovering textile cotton fibres, useless for the textile industry, to obtain composite materials with promising performances. Moreover, the use of two different micromechanics models allowed evaluating the impact of the morphology of the fibres on Young's modulus of a composite.

In summary, the papers contributed to this Special Issue give some very nice insights on the use of natural fibres as reinforcement of composite materials thus making this volume useful for students as well as for designers and engineers that would like to develop a deeper understanding on the use of this important and emerging research subject.

**Funding:** This research received no external funding.

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

### **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **Development of Thermoplastic Starch (TPS) Including Leather Waste Fragments**

### **Silvio Pompei 1, Jacopo Tirillò 2, Fabrizio Sarasini <sup>2</sup> and Carlo Santulli 3,\***


Received: 6 May 2020; Accepted: 11 August 2020; Published: 12 August 2020

**Abstract:** A thermoplastic starch (TPS) material is developed, based on corn starch plasticized with glycerol and citric acid in a 9:3:1 ratio and further bonded with isinglass and mono- and diglycerides of fatty acids (E471). In TPS, leather fragments, in the amount of 7.5 15 or 22.5 g/100 g of dry matter, were also introduced. The mixture was heated at a maximum temperature of 80 ◦C, then cast in an open mold to obtain films with thickness in the range 300 ± 50 microns. The leather fragments used were based on collagen obtained from production waste from shoemaking and tanned with tannins obtained from smoketree (*Rhus cotinus*), therefore free from chromium. Thermogravimetric (TGA) tests suggested that material degradation started at a temperature around 285 ◦C, revealing that the presence of leather fragments did not influence the occurrence of this process in TPS. Tensile tests indicated an increase in tensile properties (strength and Young's modulus) with increasing leather content, albeit coupled, especially at 22.5 wt%, with a more pronounced brittle behavior. Leather waste provided a sound interface with the bulk of the composite, as observed under scanning electron microscopy. The production process indicated a very limited degradation of the material after exposure to UV radiation for eight days, as demonstrated by the slight attenuation of amide I (collagen) and polysaccharide FTIR peaks. Reheating at 80 ◦C resulted in a weight loss not exceeding 3%.

**Keywords:** leather waste; thermoplastic starch; mechanical characterization; thermal characterization

### **1. Introduction**

Leather is obtained from animal hides, which are based on collagen, a protein molecule constituted by sequential chains of amino acids, twisted and bound in a strong molecular structure made of fibers. The process of leather fabrication does require its conservation for manipulation through tanning, which involves the fixation of tannins to the collagen matrix. The effect of tannins is enabling preservation, making leather imputrescible, therefore treatable with chemicals, while improving its hardness and strength. If trivalent chromium salts are used for this purpose, leather waste is classified as a special refuse, which requires specific methods for handling to prevent chromium from penetrating into the soil or underground water [1]. To improve leather sustainability and ease its disposal, vegetable tannins can be employed, such as those obtained from smoketree (*Rhus cotinus*). This was a traditional use in some regions, such as Marche and Tuscany, of Central Italy, and has currently a reprise, other than on sustainability grounds, also in connection with the revived use of natural colors, of some of which tannin is an essential element [2]. In this way, the structural and functional advantages of chromium tanning are preserved, while, on the other side, leather waste can be disposed of e.g., in other biodegradable materials [3]. The waste from leather treated with vegetable

tanning proved to be microporous and functional for sustainable use, as it was proposed after chemical activation with alkali, for the removal of volatile organic compounds (VOCs), such as toluene [4].

Starting from these considerations, the waste from vegetable-tanned leather could represent a candidate for possible inclusion into biopolymers such as thermoplastic starches (TPS) based on starch-glycerol. This procedure demonstrated to be suitable for the introduction of fillers in powder form, such as clay, even with limited control of their dimensions [5], or garden waste, such as the one from Opuntia, in fibrous form [6]. In general terms, this waste could be introduced in the matrix without any further treatment, although obviously with quite limited performance. As a consequence, these were able to effectively include waste, mainly from the agricultural food and non-food sector [7,8]. The aforementioned limited performance, in the case of film fabrication, can derive also from the difficult control of flowability: in this sense, citric acid, a cheap and easily available chemical, can provide some better compatibility with starch, therefore easing film processing [9]. The limitation of TPS is given by their glass transition temperature Tg, which is usually around 80 ◦C, yet strongly dependent on the amount of water added. Above Tg, thermoplastic starch loses its mechanical properties, and considerably swells in an irreversible way [10]. This phenomenon requires putting some attention into the creation of a suitable mixture for inclusion of other materials, especially when, as it is the case with films, the dimensional tolerance is very limited. The main waste constituting leather fragment is collagen, which has been added with starch in applications linked to food preservation (e.g., sausage casing): the addition of not exceedingly large amounts of starch (less than 50 wt%) to collagen was demonstrated as being effective for enhancing film strength and improving durability [11].

This study aims particularly to disclose the possibility of re-using environmentally friendly waste material, such as vegetable-tanned leather, which is currently disposed in general waste, because the limited and local characteristics of the productive system do not suggest transportation for recycling as an economically viable option. The application proposed involves its inclusion in self-developed thermoplastic starch (TPS), with the idea to optimize the composition of TPS, in view of its compatibility with waste, to be used as filler. The biodegradability of the composite obtained allows its possible application in products where contact with soil and progressive non-toxic degradation is required, such as it is the case for the on-site production of mulching films.

### **2. Materials and Methods**

### *2.1. Material Development*

### 2.1.1. Raw Materials Used

The mixture used for the production of the film included commercial corn starch for alimentary use, glycerol of 99.5% purity (E422), citric acid from lemon juice and isinglass to ease bonding, in the respective proportions 9:3:1:0.7. Moreover, 0.15 parts of mono- and diglycerides of fatty acids (E471) were added to prevent the formation of mold and to ease the manufacturing of a flat film of material during laying up. In this way, a thermoplastic starch (TPS) has been obtained. Smoketree-tanned leather in irregular fragments having their largest dimension between 10 and 500 microns approximately, was added to the material in the proportions of 7.5, 15 and 22.5 wt% over the dry weight of the mixture. The three composites obtained were defined ex-post as TPS\_ leather 1/2, TPS\_leather 1 and TPS\_leather 3/2, considering 1 as the ideal proportion for the production of the final material, which requires some deformation to be retained in the flat sheet obtained. These proportions were set after a large number of experiments carried out starting from October 2015 in the Course of "Sperimentazione di Materiali Innovativi per il Design" (Experimentation on Innovative Materials for Design) in University of Camerino and completed during 2018 with the experimental aims that are illustrated in Figure 1: the last image of Figure 1 shows the material developed, which is here reported as "TPS\_Leather 1", further developed in sheets with dimensions 300 × 200 mm. The contents of the different raw materials in percentage over dry weight of the film are reported in Table 1. The three composites are compared between them, and to the pure TPS developed with the aforementioned proportions.

**Figure 1.** Experimental process leading to the development of the film.


**Table 1.** Different components of the film.

### 2.1.2. Film Production

Film production was obtained by adding the mixture with some amount of water to allow obtaining a sufficient fluidity, then heating it at a temperature not exceeding 80 ◦C, reached by its continuous mixing in an uncovered container in 10 min. The mixture was then cast on a silicon plate, 1 mm thick, with the assistance of a steel lamina, again 1 mm thick, to level it inside a steel frame with a thickness of 1.5 mm. The frame is removed just after the cast process and the film is left cooling naturally on the silicon sheet. The phases of the production process are reported in Figure 2.

**Figure 2.** The different phases of the film casting.

The natural process of film drying, which typically lasted 5 days, made it possible to finally obtain a film of rather constant thickness, in the order of 300 ± 50 microns. From Figure 3, it is possible to observe the interfacial adhesion of leather fragments to the substrate, despite the irregularity of the filler. The films, cut into rectangular strips with maximum dimensions around 150 × 100 mm, are supposed to be coupled in the way reported in Figure 4, therefore partially superimposed to each other. They would serve as the support for texturized lawn modules, able to support small seeds for the growth of plants, therefore, the requirement is more on the effective integration of leather filler than in the creation of larger pieces.

**Figure 3.** Macrograph of the material with fragments of leather waste.

**Figure 4.** Coupling of the different films strips with lawn seeds prepared for application.

### *2.2. Experimental Methods*

### 2.2.1. Ageing Tests

Ageing tests were carried out by exposing materials to a gallium-doped Helios (Helios Quartz Group SA, Novazzano, Switzerland) medium pressure mercury lamp, model HMPL, which has emission peaks in the region between 400 and 430 nm, therefore in the IR range, yet it also has considerable emissions in the UV range between 200 and 400 nm. The light exposition was continuative for a period of 192 h. FTIR analysis was carried out using a Perkin Elmer (Milan, Italy) Spectrometer 100 in attenuated total reflection (ATR) mode. A spectral resolution of 3 cm−<sup>1</sup> in the range (4000–600 cm<sup>−</sup>1) with 512 scans was adopted to record the spectra. Two spectra were carried out, to compare those obtained from new and aged materials.

Tests with Radwag (Radom, Poland) MA 110.R thermobalance involved a program of heating with 5 min at 40 ◦C, 5 min at 60 ◦C and 8 min at 80 ◦C, therefore for a total duration of 18 min. Heating was applied on square samples of 20 mm side, removed by cutting from the rectangles of materials, on which mass was measured every 10 s with an accuracy of ±1 mg.

### 2.2.2. Mechanical Characterization of Films

Specimens for the mechanical characterization were cut from the films in accordance with UNI EN ISO 527-2 (Type 1BA samples with a gauge length of 30 mm). Tensile tests were performed at room temperature in displacement control with a crosshead speed of 2.5 mm/min by using a Zwick/Roell Z010 (Ulm, Germany) universal testing machine. The results are the average of five tests.

### 2.2.3. Thermal Characterization of Films

The thermal stability of the films was investigated by thermogravimetric analysis (TGA). To this purpose, a SETSYS Evolution system by Setaram (Caluire, France) was used, heating the samples from 25 ◦C to 800 ◦C, with a heating rate of 10 ◦C/min in a nitrogen atmosphere.

### 2.2.4. Morphological Characterization by SEM

The fracture morphology of samples failed in tension was investigated by field emission scanning electron microscopy (Mira3 by Tescan, Brno, Czech Republic). Specimens were sputter coated with gold prior to analysis.

### **3. Results and Discussion**

The tensile test results, reported in Table 2, indicated in general a limited variation in terms of stress, strain and stiffness, with respect to other typical materials including organic waste. A comparison was also possible with similar thermoplastic starches with no fillers e.g., those examined in [12]. This reference suggests that the plasticization effect is in the present study much more contained, as observed by the elongation at break for the pure TPS, which just exceeds 50%, making the material quite controllable and with predictable properties during use. In particular, a considerably lower deformation was measured with respect to what was observed in other studies, such as [13], on glycerol and citric acid on corn starch tensile properties.


**Table 2.** Summary of tensile test results.

The mechanical properties of the developed films compare quite favorably with those reported in other studies, especially for what concerns the Young's modulus. In particular, in [14], for materials based on cornstarch and glycerol/water as plasticizer reinforced with bacterial and vegetable cellulose (1% and 5% *w*/*w*), a strength in the range 0.5–3.5 MPa and a Young's modulus in the range 1–20 MPa was reported, while in [15,16], for thermoplastic starch/clay nanocomposites, strength spanned the range from 2 to 28.1 MPa and the modulus from 7.5 to 196 MPa. As far as the introduction of leather fragments is concerned, the effect of 7.5 wt% over 100 parts of TPS (TPS\_Leather 1/2) appears limited on the modification of TPS properties, whereas the material results considerably stronger and stiffer when higher amounts of leather fragments are used. However, as inferred from the typical tensile curves in Figure 5, the introduction of 22.5 wt% of leather fragments resulted in a material failing basically with no plastic behavior, which limits its applications, therefore the intermediate solution of 15 wt% of leather fragments appeared the most suitable for the use proposed.

**Figure 5.** Typical tensile stress-strain curves.

The onset of material degradation started around 285 ◦C, as determined from the graphical method suggested in [17], for corn starch-glycerol materials, and indicates a typical value for thermoplastic starches. This can suggest that the effect of leather fragments on material degradation is likely to be quite limited. However, the higher amounts introduced offered some reduction of the degradation rate in the region around 250–300 ◦C, as observable specifically from the inset in Figure 6b. Studies on collagen indicate water loss taking place up to around 125 ◦C, while the degradation onset of leather is around 300–320 ◦C [18]. In this case, it is suggested that water loss occurs seamless with desorption of water as the effect of softening of thermoplastic starch: as a matter of fact, the trend appears to be linear up to over slightly 200 ◦C. Regarding the residual mass at the end of thermal degradation process, a previous study on acrylonitrile-butadiene-styrene (ABS) resin—leather waste composites suggested that leather powder alone was leaving just below 20% of the initial mass at 800 ◦C [19]. The data found here are basically in line with this indication, suggesting that the presence of leather increases the amount of material normally left after the degradation of thermoplastic starches obtained using corn starch with similar amounts of glycerol [20].

**Figure 6.** Thermogravimetric curve (TGA) (**a**) and first derivative of the thermogravimetric curve (DTG) (**b**) of thermoplastic starch (TPS) and the different composite materials.

The microscopic characterization of the fracture surfaces indicates for pure TPS a variable fragmentation with loss of material and creation of voids, which is attributable to plasticization effect (Figure 7). In contrast, in the case where leather fragments are introduced, the occurrence of some step-like regions is detected (Figure 8a), which is often encountered in the presence of protein-based structures and likely to be related to the internal layered structure of leather particles [21]. Some pulled-out particles were also observed, as shown in Figure 8b, which suggests the need to improve the interfacial adhesion, because the subsequent detachment of few leather fragments from the matrix would result in the coalescence of voids and sudden failure. On the other side, in most cases the interface proves effective, such as in Figure 9, though its performance might be variable, due to the random distribution of waste particles: in practice, with a lower amount of leather fragments, their boundaries are more recognizable in the TPS matrix (Figure 9a,b), whereas, with their increasing amount, the deformation of the matrix due to the insertion of the filler and the subsequent loading would rather conceal them (Figure 9c). Looking in more detail at the structure of the TPS matrix, some typical situations are recognized during loading, in particular differential deformation, leading in the most critical cases to the widespread formation of cracks (Figure 10). However, it is promising that, in most cases, leather fragments have been shown to provide a strong adhesion to the matrix, although the margins for improvement are surely to be recognized.

**Figure 7.** SEM images of the pure TPS fracture surface.

**Figure 8.** Fracture surface of the composite, without (**a**) and with (**b**) pulled-out leather particles at failure.

**Figure 9.** Leather particles embedded in the TPS matrix, 1/2 (**a**), 1 (**b**) 3/2 (**c**).

**Figure 10.** Occurrence of differential deformation in the various composites.

In general terms, the material proved suitable for the application envisaged, despite the fact that further mechanical fragmentation of leather waste e.g., by milling, had been purposely avoided. As expected, the dimensional variation of the particles resulted in some further limitation of the plasticization effect provided by the corn starch-glycerol-citric acid interaction, though reduced already by isinglass, as common in TPS [22]. It needs to be considered in any case that the mechanical properties obtained and the thermal stability offered can be considered acceptable: further refinements of the production process and of the film drying uniformity, leading to a stricter dimensional tolerance, would follow in future investigations.

Regarding the material that proved to be the most suitable for the envisaged application, hence TPS\_Leather 1, further characterization was carried out. In particular, a number of peaks were observed from FTIR analysis, whose attribution is reported in Table 3, mainly from the comprehensive and reference study of FTIR spectra on starch, carried out in [23]. Some of these are also corroborated by other information offered in the respective references, and also the peaks referring to the other non-polysaccharide components in the materials, hence glycerol, citric acid and collagen from leather are attributed according to the other references quoted in Table 3. The investigation did not allow the attribution of FTIR peaks to vegetable tannins, despite some studies on this also being available, such as [24], probably due to their amount of leather wastes being too low.

Moreover, FTIR analysis suggested that peaks correlated to polysaccharides (starch), to interaction between starch and glycerol, in particular the 1104 cm−1, to amide I (collagen) and amino-acid, namely the 998 cm<sup>−</sup>1, only showed, as from Figure 11, a slight degradation after ageing for eight days to simulated exposure to the sunlight. Post-ageing, a re-heating at 80 ◦C of the material did lead to a water loss not exceeding 3%, as indicated in Figure 12, which demonstrates that most water was linked to the polymeric structure, which could suggest the material is sufficiently strong and stable during storage and application, possibly with limited shrinkage. In fact, coming back to thermogravimetric tests reported in Figure 6, they also suggested a very limited weight loss up to around 80 ◦C, indicating that the components are hydrated in a stable form.


**Table 3.** Attribution of FTIR peaks observed during the scanning.

**Figure 11.** FTIR spectra carried out on the material newly produced (darker curve) and then subjected to UV ageing (lighter curve).

**Figure 12.** Weight loss of the material during heating up to 80 ◦C.

### **4. Conclusions**

The introduction of leather fragments in three different quantities in a self-produced thermoplastic starch (TPS) based on starch plasticized with glycerol and cross-linked using citric acid proved to be promising. This can be regarded as a sustainable procedure for introducing in soil a biodegradable composite e.g., for lawn growth support, since leather is chrome-free, being smoketree-tanned. The material with intermediate amount of leather fragments did not suffer any significant degradation below 80 ◦C, which makes it basically acceptable for the application envisaged.

The positive aspects are also in the formation of a sound matrix-filler interface, though with possible improvements based on a more accurate selection of particle distribution. As expected, collagen-based reinforcement does result in hindering the plasticization of TPS, whereas it also provides more controllable mechanical properties. As far as thermal properties are concerned, the introduction of leather fragments does not change to a significant extent the degradation patterns of TPS, which take place just below 300 ◦C. The amount of leather waste introduced could be possibly increased by working on particle size with more accurate fragmentation methods, while, in the present work, only basic cutting operations have been performed.

**Author Contributions:** Conceptualization, S.P., F.S., and C.S.; methodology, S.P., J.T., and F.S.; measurements, S.P., J.T., and F.S.; investigation, S.P. and C.S.; writing—original draft preparation, S.P. and C.S.; writing—review and editing, F.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

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

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

### *Article* **E**ff**ect of Fique Fibers in the Behavior of a New Biobased Composite from Renewable Mopa-Mopa Resin**

### **José Herminsul Mina Hernandez 1,\*, Edward Fernando Toro Perea 2, Katherine Caicedo Mejía <sup>1</sup> and Claudia Alejandra Meneses Jacobo <sup>1</sup>**


Received: 17 May 2020; Accepted: 13 July 2020; Published: 16 July 2020

**Abstract:** A fully biobased composite was developed using a natural resin from the *Elaeagia Pastoensis Mora* plant, known as *Mopa-Mopa* reinforced with fique fibers. Resin extraction was through solvent processing reaching an efficient extraction process of 92% and obtaining a material that acted as a matrix without using any supplementary chemical modifications as it occurs with most of the biobased resins. This material was processed by the conventional transform method (hot compression molding) to form the plates from which the test specimens were extracted. From physicochemical and mechanical characterization, it was found that the resin had obtained a tensile strength of 15 MPa that increased to values of 30 MPa with the addition of 20% of the fibers with alkalization treatment. This behavior indicated a favorable condition of the fiber-matrix interface in the material. Similarly, the evaluation of the moisture adsorption in the components of the composite demonstrated that such adsorption was mainly promoted by the presence of the fibers and had a negative effect on a plasticization phenomenon from humidity that reduced the mechanical properties for all the controlled humidities (47%, 77% and 97%). Finally, due to its physicochemical and mechanical behavior, this new biobased composite is capable of being used in applications such as wood–plastic (WPCs) to replace plastic and/or natural wood products that are widely used today.

**Keywords:** Mopa-Mopa resin; biobased composite; fique fibers; wood–plastic

### **1. Introduction**

Due to the environmental impact caused by the conventional synthetic polymers when they are not properly disposed of at the end of their life cycle, research studies are currently being carried out in the field of polymeric materials that focused on the development of polymers characterized by being biobased, generated from renewable sources such as starches, proteins, hydroxy alkanoates, among others, and by presenting complete biodegradability under composting conditions [1–8]. Unlike traditional synthetic polymers, these materials are not oil-dependent; therefore, they have an added value as a potential alternative to produce eco-friendly materials. Among this family of polymer materials, the natural resins extracted from plants stand out due to the potential use of a wide variety of plants in the ecosystem that constitute a renewable source for polymer obtention [6,7]. These resins are currently being employed for the development of biobased composites, which, in most cases, are used as a partial substitute of reagents on the synthesis of polymers, such as canola oil for the obtention of polyols in order to react with isocyanates for the production of polyurethane adhesives and foams [8], tannin-furfuryl alcohol for thermoset resins [9], soya as polyol for polyurethanes [10], starches, wood, and other natural materials as sources for synthesizing reagents to produce epoxy bio-resins [11], such as modified vegetable oils, sugars, polyphenols, terpenes, colophony, natural rubber, and lignin for chemical synthesis of resins and curing agents for epoxy polymers [12], among others. The Mopa-Mopa resin forms the base of the varnish that is extracted from the *Elaeagia Pastoensis Mora* wild shrub, which belongs to the *Rubiaceae* family and grows in the Department of Putumayo, Amazon region in the Colombian jungle. Twice a year, the plant produces a gelatinous paste, which, through an artisanal process, is transformed into a thin sheet that can be molded to make decorative drawings on pre-painted wood [13]. The resin has been extracted and used by generations of farmers to commercialize it as a raw material mainly for manufacturing and/or restoring handicrafts [14,15]. Although the knowledge of the Mopa-Mopa has been limited to the development of new techniques and products on an artisanal level, there has been a growing interest within the scientific community to analyze and expand the understanding of this polymeric material through research about its physical, mechanical, thermal and chemical properties. Among those studies, Insuasty et al. [16] started the chemical characterization of the Mopa-Mopa resin from solubility tests and several spectroscopic techniques and identified that ethanol and methanol were the best solvents for the resin without losing its physical properties, especially its elasticity. Other studies [17,18] have shown both some physicochemical and thermal properties of the Mopa-Mopa resin and the effect of add polycaprolactone on the properties of binary mixtures using Differential Scanning Calorimetry (DSC), X-ray Diffraction (XRD), Thermogravimetry Analysis (TGA), and Fourier Transform Infrared Spectroscopy (FTIR) tests as well as in determining the semi-crystalline nature of the resin.

Considering the aforementioned aspects, this project developed an efficient extraction process of Mopa-Mopa resin to be implemented as a fully biobased matrix for the production of a biobased composite reinforced with short fibers of fique (25 mm); fique fibers are available in Southwestern Colombia, they are mainly used in packaging and cordage [19]. It is also noted that, due to their good mechanical properties and the technological development associated with its extraction process, fique fibers have also been considered as reinforcement in plastic matrix composite materials. [20–23]. In the same manner, this work contemplated the obtained biobased composite of Mopa-Mopa resin with 10 and 20% (m/m) of fique fibers, the evaluation effect of the fiber superficial modification process (by alkalization) and the exposure of the material to three different relative humidities (47, 77 and 97%), in its physicochemical and mechanical properties. Furthermore, the new material developed was characterized because it can be processed through conventional transformation methods and because its behavior means that it can also can be used in other applications, such as wood–plastic.

### **2. Materials**

The Mopa-Mopa resin used in this research was extracted from the buds of the *Elaeagia Pastoensis Mora*, a plant native to the Department of Putumayo (Mocoa, Colombia) (typical taxonomic classification is provided in Table 1). Furthermore, the obtained resin showed some physical characteristics, such as ρ = 1.108 g/cm3, Tm = 117 ◦C, and Tg = 34 ◦C. The fique fiber used belongs to the *Furcraea* genus of the *Uña de Águila* (white variety plant), which was provided by the Empaques del Cauca Company located in the city of Popayán, Colombia. The raw material was used to make a short reinforcement with an average length of 25 mm randomly located, and with and without an alkaline treatment inside a Mopa-Mopa resin matrix. Finally, sodium hydroxide used in the alkaline treatment of the fibers and ethanol employed in obtaining Mopa-Mopa resin [24] were reactive grade acquired by the company Técnica Química S.A. (Cali, Colombia).


**Table 1.** Taxonomic classification of the Mopa-Mopa resin.

### **3. Experimental Procedure**

### *3.1. Obtaining the Mopa-Mopa Resin*

Through the use of a manual disk coffee milk, a physical treatment of comminution was carried out to reduce the buds of the Mopa-Mopa tree that were initially agglomerated (Figure 1a), thus achieving an efficient size reduction (Figure 1b). Subsequently, for the obtention processing, 300 mL of a solution of the Mopa-Mopa resin in ethanol was prepared at a concentration of 20% (*m*/*v*) by using a flat-bottom three-necked flask with a capacity of 500 mL and set up in a closed distillation system, keeping the solution under heating at a temperature of 75 ◦C for 25 min. The remaining fluid was cooled to approximately 40 ◦C and the solid residues were separated from the mixture, which corresponds to the remains of leaves, seed husks, and stems, among others, by using a vacuum filtration system composed of a porcelain funnel, an Erlenmeyer that received the filtered solution and a vacuum pump Welch model BS-8000 Fisher Scientific (Gardner, MA, USA). This procedure was carried out in two stages, the first stage with absorbent towels, eliminating the larger residues, and the second stage, with qualitative filter paper, to ensure cleaning of the solution. The already filtered mixture was heated again at a temperature between 76 and 78 ◦C for 45 min, to evaporate as much as possible, that is to say, to concentrate the Mopa-Mopa, taking special care to not degrade the resin; this procedure was conducted in aid of a fractional distillation system to condense the solvent and to regain around 80% of the initial ethanol. Simultaneously to the previous procedure, a volume of 100 to 150 mL of distilled water was heated to a temperature of 100 ◦C, in which it was added to the Mopa-Mopa concentrated solution. In the making of this mixture, the resin kept suspended on the water surface due to the immiscibility of these two phases (Figure 1c). Subsequently, constant agitation was maintained using a spatula lab until the resin precipitates and can be separated from the liquid. This step corresponded to the 20 min after mixing with distilled water. Similarly, it is important to specify that in this step the remaining solvent that was left in the previous stage is evaporated. The Mopa-Mopa resin obtained was deposited in a watch glass, where it was left to cool for 5 min until reaching a temperature of 25 ◦C to register the mass. (Figure 1d). A two-stage size reduction process was necessary to carry out the elaboration of test specimens. In a first stage, the material was crushed in a low-speed granulator of the SG-16/20 series until obtaining a particle size of 3 mm, in a second stage, using a manual coffee grinder, until reaching a particle size between 0.3 and 0.5 mm. It is important to note that this resin Mopa-Mopa is characterized by containing a variety of metabolites, including alkaloids, flavonoids, and cardiotonic aglycones [16]. The test specimens were formed in a Carver press model MH 4389-4021 (Wabash, IN, USA) equipped with heating plates and a forced circulation water system) using a pressure of 25,000 lb, and a temperature of 165 ◦C for 30 min; after that, the obtained plates were die-cut following the shape of the Type IV probes in accordance with ASTM D 638 [25].

**Figure 1.** Mopa-Mopa resin: (**a**) in the raw state; (**b**) after the grinding process; (**c**) during the separation process; (**d**) after the extraction process.

### *3.2. Surface Modification by Alkalization of Fique Fibers*

Following a similar methodology to that proposed by Valadez et al. [26] for henequen fibers, the alkalization of the fique fibers consisted of pre-drying the material at 100 ◦C for 24 h, to then carry out a surface treatment by immersing the fibers in an aqueous solution of NaOH (2% *m*/*v*) for 1 h at 25 ◦C; later, the fibers were washed with distilled water acidified with acetic acid, until reaching a neutral pH, that is, until the washing water had no residual alkaline solution. Finally, the fibers were dried at 60 ◦C for 24 h.

### *3.3. Preparation of the Biobased Composite*

For the elaboration of the biobased composite, the components, the Mopa-Mopa resin and the native and alkaline fique fibers, were pressed separately and 25,000 lb of pressure was applied, forming nonwoven fique mats (Figures 2a and 3b). Finally, the components were molded using a sandwich-type layout (Mopa-Mopa sheet–fiber mat–Mopa-Mopa sheet) at a temperature of 170 ◦C, under the following pressure scheme: 10,000 lb for 3 min, 20,000 lb for 3 min and 25,000 lb for 30 min (Figure 3c,d); reinforcing proportions of 10 and 20% concerning the total mass were maintained.

**Figure 2.** Biobased composite preparation: (**a**) Mopa-Mopa compacted sheet; (**b**) nonwoven fique mats; (**c**) Mopa-Mopa/nonwoven fique mats/Mopa-Mopa laminate; (**d**) molding the biobased composite.

**Figure 3.** Fourier Transform Infrared Spectroscopy (FTIR) for the fique fibers: (**a**) native; (**b**) alkalized.

### *3.4. Fourier Transform Infrared Spectroscopy (FTIR)*

For the analysis of the materials, a Fourier Transform Infrared Spectrum 100 model was used. In the case of fique fibers with and without alkaline treatment, the Attenuated Total Reflectance (ATR) technique was followed, working at 100 sweeps and a resolution of 2 cm<sup>−</sup>1. For the Mopa-Mopa resin, the infrared analysis was performed on conditioned specimens at three relative humidity ranges (47, 77 and 97%), as well as on an additional sample that was dried in an oven at 60 ◦C for 2 h. The Diffuse Reflectance (DRIFT) technique was used in these materials, working at 200 scans and a resolution of 2 cm<sup>−</sup>1.

### *3.5. Moisture Adsorption*

Salts of potassium carbonate, sodium chloride, and distilled water were added to desiccators for maintaining constant the relative humidity at 47%, 77%, 97%, respectively, in accordance with ASTM E 104 [27]. Humidity meters were placed in the desiccators to monitor such relative humidity. For the evaluation of the fique fibers, three bundles of the fibers were cut, with masses between 3 and 6 g, both for the native and those that had the alkaline treatment and for each relative humidity. Subsequently, the samples were dried in an oven at 100 ◦C for 24 h and placed in the respective desiccators, registering their mass. In the case of the Mopa-Mopa resin and the biobased composite, three type IV test specimens were prepared according to ASTM D 638 [25] for each relative humidity and each condition of the composite (proportions of 10 and 20%, with and without treatment alkaline), to then analyze the effect of relativity humidity and exposure time on mechanical properties by applying a tensile test. The data of the mass gain as a function of time (Mt) were taken in each case and the percentage of moisture adsorption (H) was determined based on the model presented in Equation (1), considering the mass after drying in the oven (Ms).

$$\text{pH} = \left(\frac{\text{Mt} - \text{Ms}}{\text{Ms}}\right) \times 100\tag{1}$$

### *3.6. Density Determination*

This test was carried out on a Mettler Toledo AG245 equipment with a maximum capacity of 210 g and a deviation of 0.001 g. The density value for the Mopa-Mopa resin and the biobased composite in all the study conditions (proportions of 10 and 20% with native and alkalinized fiber) was determined by method A of the ASTM D 792 [28]. In the case of fique fibers with and without alkaline treatment, the density was estimated following method B of the same standard. The theoretical density was estimated with the aid of the rule of mixture model shown in Equation (2)

$$
\rho\_b = \rho\_f \nu\_f + \rho\_m (1 - \nu\_f) \tag{2}
$$

where ρ is the density, and the subindices *b*, *f*, and *m* refer to the biobased composite, the fiber, and the matrix, respectively; ν*<sup>f</sup>* is the volume fraction of fiber incorporated into the biobased composite.

### *3.7. Tension Test*

The determination of the tensile mechanical properties of the fique fibers with and without alkaline treatment, the Mopa-Mopa resin, and the biobased composite was carried out by a Tinius Olsen model H50KS universal testing machine. An experimental setup was used in the fique fibers, which consisted of the fixing of filaments in cardboard frames, allowing for a better grip of these to the clamp, in accordance to the ASTM D 3822 [29] standard, with a clamp displacement rate of 3 mm/min and taking into account a previous analysis of the distribution of average diameters in each fiber tested. A sample of 140 fibers (70 native and 70 alkalized) was selected at random and estimating about 8 diameter values set for each of the filaments, which generated a standard deviation of 0.03 for the native fibers and 0.02 for the alkalized fibers. For Mopa-Mopa resin and biobased composite, ASTM D638 [25] was followed with type IV specimens and a speed displacement rate of 5 mm/min.

### *3.8. Scanning Electron Microscopy (SEM)*

The morphological characterization of the fique fibers, the Mopa-Mopa resin, and the biobased composite were carried out by a Electron Microscope model JMS 6490 LV (Jeol, Mexico D.F., Mexico), in the secondary electrons mode (SEM) and under an accelerating voltage of 20 kV. The chemical microanalysis was performed on several inspection areas using an Energy Dispersive X-ray Spectroscopy (EDS) model IncaPentaFETx3 (Oxford Instrument, Belfast, UK). This technique allowed us to qualitatively know the surface effect of the treatment of the fique fibers, the ductile nature of the Mopa-Mopa resin, and the interface of the biobased composite. The samples were previously bonded on a carbon tape and then metalized with a gold layer between 10 and 50 nm (thin film deposition equipment Model Desk IV at a pressure of 50 m Torr and a time of 60 s, (Denton Vacuum, Moorestown, NJ, USA) in order to generate a conductive surface, analyzing the cross-section of the resin, the biobased composite and the contour of the fique fibers.

### **4. Results and Discussion**

### *4.1. Fourier Transform Infrared Spectroscopy (FTIR)*

The infrared spectroscopy test of the Mopa-Mopa resin was carried out carefully on four conditioned samples: one dried sample in the oven at 60 ◦C for 2 h (Reference sample), the remaining samples, at three relative humidities of 47, 77 and 97%. The purpose of this procedure was to learn the effect of water adsorption on the molecular interactions of the resin. Consequently, the following bands for the dry sample could be observed: an intense signal in the region below 3000 cm<sup>−</sup>1, with two bands at 2979.6 and 2951.0 cm−<sup>1</sup> assignable to the vibration in tension (asymmetric and symmetrical) of the CH bond; at 1751.0 cm−1, an intense signal corresponding to the vibration to a tension of the carbonyl group C=O; at 1656.6 and 1474.9 cm<sup>−</sup>1, associated with the vibration in tension C=C, between 1400 and 1000 cm−1, other slightly widened bands were observed, most likely related to CO bonds and the deformation vibration of CH bonds. The results found were similar to those reported for the Mopa-Mopa resin in other studies [17,18]. On the other hand, the samples conditioned at 97%, 77%, and 47% relative humidity presented a new band associated with interactions of the hydroxyl group with water molecules at 3284.3 cm−1, but with a difference in the peak's intensity seen in the

axis of ordinates that varied in values of 0.017, 0.009 and 0.007, respectively, being greater for the higher humidities. This fact can be correlated with a possible plasticization of the Mopa-Mopa by the influence of the humidity. This phenomenon is similar to the one that happens in the thermoplastic starch [30], whereby incorporating a plasticizer to the cassava starch means that the interactions of the hydroxyl groups are modified within the material and new second-order intermolecular associations (hydrogen bonds) are established with fewer steric hindrances. Table 2 shows the spectral bands for each sample used with the characteristic link type.


**Table 2.** Characteristic bands and type of bond for the Mopa-Mopa resin.

Figure 3 shows the infrared spectra obtained in the native and alkalized fibers; in these spectra, the representative bonds of the fundamental constituents of the fiber, such as cellulose, lignin, and hemicellulose, stood out. With the treatment of the fibers in the alkaline solution, it was observed that the peak corresponding to the tension stretch of the carbonyl C=O, which occurred at a wavenumber of 1736 cm<sup>−</sup>1, disappeared from the spectrum, this peak is associated with Ester-type bonds that usually appear in the hemicellulose structure; therefore, the loss of this band indicated that, at least at the surface level, the removal of this component in the fiber was generated. Similarly, it showed a decrease in peaks at 1505 and 2862 cm−<sup>1</sup> related to the aromatic skeleton and stretching of the -OCH3 bond of lignin, respectively, indicating that the lignin surface concentration decreased with alkaline treatment. Similar results were reported by Mina [31] who additionally included the chemical composition for the fique and other natural fibers, finding that the percentage related to lignin and cellulose in the fique was under 14 and 61.2%, respectively.

### *4.2. Moisture Adsorption*

This study was carried out considering the conditioning of the material at relative humidities of 47, 77 and 97%, finding in specimens a reduced increase in mass due to low water adsorption in all the different relative humidities controlled. The humidity adsorption values of the equilibrium corresponded to 1.9, 0.64, and 0.09% for the humidity of 97, 77, and 47%, respectively. These data indicated that the Mopa-Mopa resin presented low humidity adsorption compared to other natural polymers, such as thermoplastic starch, that, according to reported results [30], presents about 7% adsorption when reaching equilibrium; for a relative humidity of 43% and 25 ◦C, these conditions were lower than those used in the present study and with very high adsorption results that were not reached by the Mopa-Mopa resin even at the highest relative humidity studied (97%). The general behavior of the humidity adsorption of the Mopa-Mopa resin in the different atmospheres is shown in Figure 4a, where it becomes evident both the relationship of proportionality between the relative humidity to which the resin was exposed and its adsorption percentage for the humidity of 97 and 77%. Moreover, it must be considered that, during the initial times, the adsorption of the resin occurred at a faster rate (first stage of the curve), followed by an almost constant behavior defined as balance. In a particular case, the curve corresponding to the humidity of 47% showed a decrease in the humidity adsorption

because the specimens were not completely dry before being introduced into the desiccator since the resin is very vulnerable to oxidation in prolonged healing periods; therefore, when these presented a higher humidity than the exposure medium, the salts generate a drying effect to balance their humidity, represented in the decreasing behavior of the curve. Fique fibers' ability to absorb moisture could be influenced by the superficial modification generated by the alkalization treatment, since, with it, some hydrophobic groups are removed from the fiber, such as waxes and pectins, generating a concentration of cellulose on its surface that has hydroxyl groups and a great affinity with water, forming hydrogen bonds and thus increasing its adsorption moisture. This behavior was reflected in the increase in the mass of the fibers when they were exposed to different relative humidities (97, 77, and 47%), with higher adsorption at higher humidities compared to native fibers. Figure 4b shows the curves corresponding to the adsorption isotherms of native fique fibers and of the alkalized fibers, conditioned to the three relative humidities of 47, 77 and 97%. Due to the type of bonds it has, Mopa-Mopa is a material with the proper functionality to form hydrogen bonds with water molecules present in the humidity of the environment, behavior that is similar to that of the fique fibers used as reinforcement and that turn to the increase in the level of moisture adsorption of the Mopa-Mopa/Fique biobased composite concerning the individual behavior of the components, as can be seen in Figure 5a,b. The increase in mass due to the adsorbed water is directly affected by the content of the fibers used, being greater in those with 20% of them. Besides, the alkaline treatment generated a greater absorption of moisture on the part of the biobased composite; this was attributed, as mentioned above, to the fact that the treatment caused an increase in cellulose on the surface of the fiber, which has hydroxyl groups and a great affinity with water, forming hydrogen bonds and increasing their ability to absorb moisture. According to the above, the increase in the mass by water adsorption in the compounds with alkalized fibers could be produced through a diffusion mechanism, due to the presence of micro-spaces between the fiber and the matrix that allowed the water filtration in the material [32]. This phenomenon becomes more important when the amount of fiber increases in the matrix. It is important to mention that, as in the mechanical characterization of the Mopa-Mopa resin, the adsorption of humidity in the same three exposure times was studied in the biobased material, showing a proportional increase in the adsorption of water throughout the exposure time and generating a negative effect on the mechanical performance of the material.

**Figure 4.** Adsorption isotherms at a relative humidity of 97, 77, and 47% for (**a**) Mopa-Mopa resin; (**b**) native and alkalized fique fibers.

**Figure 5.** Adsorption isotherms at relative humidities of 97, 77, and 47 for (**a**) biobased composite with 10% native and alkalized fique fibers; (**b**) biobased composite with 20% native and alkalized fique fibers.

### *4.3. Density Estimation*

This test was performed on all the manufactured Mopa-Mopa/Fique biobased composite and the Mopa-Mopa resin matrix to establish the effect of fiber incorporation and the surface treatment of the fique on the density of the material, the used values for the theoretical determination of the different compounds were <sup>ρ</sup>*<sup>m</sup>* = 1.108 <sup>±</sup> 0.03 g/cm3, <sup>ρ</sup>*<sup>f</sup>* = 1.393 <sup>±</sup> 0.04 and 1.308 <sup>±</sup> 0.06 g/cm<sup>3</sup> for the native and alkalize fibers, respectively. Figure 6 shows the results obtained and that was compared with the estimated values employing the rule of mixture in Equation (2). By incorporating 10% of fique fibers with and without alkaline treatment, it was found that the density of the composite decreased concerning the density of the matrix. However, the density of the composite with the alkalized fibers was higher than the one with the native fibers. This positive resulting effect from the treatment of the fique fibers in the density of the composite is due to the reduction of voids and/or cavities between the fiber and the matrix that also affects the improvement of the interfacial zone as will be discussed later with the help of SEM images. On the other hand, in the composite made with 20% of native fique fibers and previously treated with NaOH, it was observed that the density increase regarding that of the Mopa-Mopa resin and the compounds with 10% of the fibers is greater than that of the material with 20% alkalinized fibers, presenting a similar trend to the one estimated from the theoretical values. The materials that presented higher values of density also showed greater resistance in tension because they had a better mechanical anchorage; the relationship between the density, the interfacial zone, and the properties of the compounds has been documented in the literature [33]. It is also important to highlight that the dispersion of data decreased with the increase in the volume of fiber, being closer to the theoretical values than the density of the biobased composite reinforced with alkalized fiber.

**Figure 6.** Density data for Mopa-Mopa resin and biobased composite with 10 and 20% native and alkalized fique fibers.

### *4.4. Tensile Strength*

The specimens used to determine the mechanical properties were tested at three time points (t1: before exposure to the desiccator; t2: 3 days of exposure; t3: 15 days) of conditioning, and a relative humidity of 47, 77 and 97%. Table 3 shows the results of the tensile properties for the natural Mopa-Mopa resin at the different relative humidities studied. Here, it can be seen that the material is characterized by having a tensile strength of 10.42 MPa for the conditioning time 1. This value turns out to be low when compared to some conventional synthetic polymers that show strengths around 22 and 30 MPa, as in the case of hight density polyethylene (HDPE) and polypropylene (PP), respectively [34]. However, it has an interesting behavior when compared to materials such as low density polyethylene (LDPE), for which there had been reported values from 5 MPa [35] and some bio-based polymers such as thermoplastic starch that reaches values between 0.23 and 5.5 MPa [31]. On the other hand, it is important to consider that the mechanical properties can be increased with the use of reinforcing materials such as natural fibers, in this case, those of fique. The mechanical properties of the Mopa-Mopa resin varied with the relative humidity to which it was exposed, mainly due to the phenomena of plasticization by water that the material undergoes and which allowed decreases of 69.78, 62.57 and 39.05% for the relative humidities of 97, 77 and 47%, respectively. Such a phenomenon influenced the secondary interactions that were previously discussed through an infrared analysis that is also a condition that occurs in biobased polymers such as thermoplastic starch, for example [30–36]. When incorporating the fibers in the matrix based on the Mopa-Mopa resin, an increase in the tensile strength of 43.35% for the composite with native fibers and 53.95% for the composite with alkalized fibers was evident, while in the case of the Young s modulus of the biobased composite, the increase was 34.87 and 37.88% for the fibers without and with alkaline treatment, respectively. These values were a direct function of the content of incorporated fibers, corroborating that the use of these had a positive impact on the mechanical properties of the material. In return, the alkaline surface treatment on the fibers improved the mechanical behavior of the biobased composite by 24.75 and 38.32% compared to that which was reinforced with 10 and 2% fibers without any treatment (native fibers), respectively, due to the increase in roughness and generation of mechanical anchorage. It is important to highlight that the mechanical properties were negatively affected by factors such as humidity and conditioning time, causing a considerable decrease in tensile strength, as can be seen in the data reported in Table 3. This behavior was attributed to the plasticizing effect due to the humidity that the biobased composite

adsorbs due to the hydrophilic character of both the matrix and the reinforcement, as previously mentioned. The best mechanical performance in biobased composite associated with the material that was reinforced with 20% alkalized fibers, for which a tensile strength and Young s modulus of 32.73 and 2128.36 MPa was obtained, respectively; the above for the first evaluation time that corresponded to the specimens before conditioning.



These results were superior to those reported by Delgado et al. [37], who developed wood–plastic composites from a low density polyethylene (LDPE)/hight impact polystyrene (HIPS) matrix reinforced with natural fibers, reaching values below 6 and 70 MPa for the parameters of interest. Additionally, the results reported by Fajardo et al. [38], who developed a PP matrix reinforced with bamboo fibers, that contained values of 12 and 1400 MPa in a tensile strength and Young s modulus, respectively. In the case of the biobased composites studied in the present study, the mechanical properties achieved were greater, independent of the humidity and conditioning times employed.

### *4.5. Scanning Electron Microscopy (SEM)*

From SEM micrographs, it was possible to evaluate the quality of the fiber–matrix interface for the biobased composite that was reinforced with the fibers with and without surface treatment. In Figure 7 it can be seen some gaps in the areas located between the matrix and the fiber for the two case studies and that could give an indication of the formation of a weak interface between the phases of the compound. However, thanks to the intrinsic adhesion exhibited by the matrix, the mechanical properties showed increases in the resistance and modulus with the incorporation of the fibers and performing greater values with the fiber alkalized by the mechanical anchorage promoted by the roughness generated by the treatment, as mentioned in other studies [32]. The interface and

the mechanical anchorage of the biobased composite are themselves affected by the relative humidity in which the material was exposed (47%, 77% and 97%). It can be seen in Figure 7 that at a higher relative humidity the mechanical anchorage decreases, due to the plasticization of the material caused by the adsorption of water. The lack of adherence between the fibers and the matrix can be seen in the reduction of the amount of matrix on the fibers and the increase in the empty spaces between them (see red circle), being observed a greater intensity at the higher relative humidity. This fact was also visible in the decrease in the mechanical properties of the biobased composites.

**Figure 7.** Scanning Electron Images of fracture surfaces of the biobased composite with native and alkaline fibers conditioned at (**a**) a relative humidity of 47%; (**b**) a relative humidity of 77%; (**c**) a relative humidity of 97%. The red circles show voids between the fiber and the matrix.

### **5. Conclusions**

It was possible to develop a new composite material in which the matrix corresponded to the Mopa-Mopa resin, extracted from the bud of the *Elaeagia Pastoensis Mora* plant, characterized by being a fully biobased composite and, unlike most materials reported in the literature, only requiring

a physical process of extraction through a closed distillation system that provided a significant yield (92%). In addition, this biobased material could be processed by hot compression molding, which is a conventional transformation process, through which it is possible to manufacture plates of fique fiber-reinforcement and from which test specimens for physical-chemical and mechanical characterizations were produced.

The biobased composite presented a positive synergy between the Mopa-Mopa resin and the fique fibers that were evidenced in the quality of the interfacial zone and from a macro-mechanical perspective, in an increase of the tensile strength and the Young's modulus as a function of the content of fibers in the material; much more significant increase in those properties was shown when a 20% of fibers superficially modified by an alkalization treatment were incorporated.

On the other hand, it was found that the relative humidity and the conditioning time of the material played an important role in a plasticization phenomenon that generated important reductions in the mechanical properties, especially to 97%, albeit the amount of moisture absorption in the equilibrium was low compared to what was reported for other biocomposites based on natural polymers such as thermoplastic starch. Because of the ease of processing and the physicochemical and mechanical characteristics evaluated that are evident when comparing the results achieved in this new material with those reported LDPE, HIPS and PP biocomposites reinforced with natural fibers, it is possible that the Mopa-Mopa resin/fique fibers biobased composite developed can be used as wood–plastic for the substitution of plastic and/or natural wood, increasing this viability if in further research the effect of integrating additives (lubricants, fillers, pigments, among others) in the composition of the material is studied.

**Author Contributions:** Conceptualization, J.H.M.H., E.F.T.P., K.C.M., and C.A.M.J.; Formal analysis, J.H.M.H., and E.F.T.P; Investigation, J.H.M.H., E.F.T.P., K.C.M., and C.A.M.J.; Methodology, J.H.M.H., E.F.T.P., K.C.M., and C.A.M.J.; Writing—original draft, J.H.M.H., and E.F.T.P.; Writing—review & editing, J.H.M.H., and E.F.T.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors acknowledge the Escuela de Ingeniería de Materiales of the Universidad del Valle, for funding this research.

**Conflicts of Interest:** The authors declare no conflicts of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
