BRPI1103362B1 - ecological composite, ecological composite production process and use of ecological composite in coatings - Google Patents

Abstract

ecological composite, ecological composite production process and use of ecological composite in coatings. the present invention describes highly resistant ecological composites comprising parts of ubuçu, more specifically, ubuçu fibers that can be mixed with resins derived from vegetable oil. ecological composites allow for various uses, including in the manufacture of floors and surface coating

Description

Invention Patent Descriptive Report

Ecological Composite, Ecological Composite Production Process and Use of Ecological Composite in Coatings.

Field of the Invention

The present invention describes highly resistant ecological composites comprising parts of ubuçu, more specifically, ubuçu fibers that can be mixed with resins derived from vegetable oil. Ecological composites allow for various uses, including in the manufacture of floors and surface coating.

Background of the Invention

In recent decades, the process of environmental degradation and ecological movements have made environmental issues somewhat popular. In order to preserve the environment, the use of materials derived from biomass has been intensifying. In this sense, the area of polymeric composite materials reinforced with vegetable fibers represents a good alternative in the field of materials for engineering and design applications.

Currently, environmental issues bring up a series of problematic conditions that surround us, mainly the degradation of the environment. OLIVEIRA et al., (2007) affirm that this process of degradation can be cited as a major cause the increase in production in industrial activities started after the Industrial Revolution and intensified in the second half of the 20th century. Thus, in view of the exacerbated use of raw materials from non-renewable sources and which already show signs of shortages, it is necessary to produce more environmentally friendly materials using renewable sources.

In this context, the area of composite materials currently has as one of its biggest areas of interest, the development of polymeric composites reinforced by vegetable fibers. This interest stems from the fact that

2/63 that polymeric composites are mostly reinforced by synthetic fibers, however, due to their low biodegradability and non-renewable origin (oil), it is necessary to replace them with ecological and human-friendly materials.

Among the ecological materials, polyurethane derived from papaya oil in (Ricínus communis L} and ubuçu fiber (Manicaria saccífera) are configured as objects of study in this work proposal. These materials were used in the manufacture of a polymeric composite for its application in floors and coatings Parallel to the development of this composite, an ecological feasibility study was also carried out through an Ecodesign research, involving a Life Cycle Analysis (LCA).

The ubuçu is a palm tree found in Central and South American countries, and in Brazil the fiber is found mainly in the North Region. The use of tururi fiber (obtained from ubuçu) by small 15 companies in Amapá and Santa Catarina (textile sector) causes it to be disseminated nationally and internationally. Despite the widespread use, more in-depth studies have not yet been carried out regarding the physical-mechanical potential of the ubuçu fiber with a view to its application in various composite materials.

Polyurethane derived from castor oil is a polymer considered ecological, since it does not originate from petroleum sources. This material has been used in several areas, such as the telecommunications sector, telephone networks, electronics, automobiles, paints and varnishes, adhesives, foams and elastomers. According to one of its manufacturers, 25 PROQUINOR, castor resin has a renewable chemical, non-aggressive to man or his environment, consciously developing green technology for the benefit of society. Thus, the union of the ubuçu fiber with the castor resin represents what is known as ‘ecocomposite”. SCHUH & GAYER (1997) mention that when natural vegetable fibers are associated with resins derived from vegetable oils, a class of materials called “ecocomposite” forms. Ecocomposites are ί

3/63 consisting of at least one biodegradable phase and / or phases that require less energy consumption for its production. Ecocomposites developed with materials from plant biomass have low cost, lower density, reduced carbon dioxide emission, when compared to composites with inorganic fibers (BOGOEVA et al., 2007).

The inherent hydroxyl-rich nature of plant fibers suggests that they are particularly useful in thermoset systems, such as polyurethane, where the hydroxyl group of the fibers can react with the polyurethane isocyanate group. Polyurethanes are very versatile polymers that, due to the proper choice of their elements, can be prepared as a thermoplastic, thermoset, elastomer, rigid foam or adhesive. It can be derived either from petroleum or vegetable oils, which makes it a great attraction in the fields of the so-called “eco composites” (VILARIM, 2003),

The use of renewable natural resources in the manufacture of the 15 so-called ecocomposites represents an alternative of great economic potential for industries that want to stand out in the market by making eco-efficient products. Thus, the development of this research meets these economic as well as ecological interests, since castor resin and ubuçu fiber are raw materials from biomass, 20 therefore being renewable resources.

This study presents its originality based on the use of a little-known vegetal fiber (ubuçu), as well as its union with castor resin, obtaining a composite with sustainable characteristics. Another factor to be mentioned is the study of the ecological viability of the composite, through 25 a powerful Life Cycle Analysis tool, which is currently used by companies and organizations that seek to present products with less environmental impact to the consumer market, that is, , eco-efficient.

The search in the patent literature pointed out some relevant documents citing ecological composites, which will be described below.

US 20100233146 describes coatings and surface treatments comprising active enzymes and peptides, where the active enzyme i

1*

4/63 can be a lipolytic enzyme obtained from Rícinus communis. The present invention differs from that document in that it is not a surface treatment, but rather a composite per se, comprising oil of Rícinus communis and fibers from Manicaria saccifera, not described in the referred document.

From what can be inferred from the researched literature, no documents were found anticipating or suggesting the teachings of the present invention, so that the solution proposed here has novelty and inventive activity in view of the state of the art.

Summary of the Invention

In one aspect, the present invention provides an ecological, highly resistant composite comprising ubuçu, more specifically a mixture between ubuçu fibers and castor resin. The present invention 15 also presents the production process of this new and inventive plant composite and its use in coatings, more specifically, in floors and the like.

Therefore, an object of the present invention is an ecological composite comprising parts of ubuçu and resins derived from vegetable oil.

In a preferred embodiment, resins derived from vegetable oil 20 comprise castor oil resins.

In a preferred embodiment, ubuçu fiber and plant-based polyurethanes are found in a 1: 1 ratio of the matrix components.

In a preferred embodiment, ubuçu fiber and plant-based polyurethanes are found in a 2: 1 ratio of the matrix components.

It is, therefore, an additional object of the present invention to produce ecological composites, comprising the steps of:

a) obtaining parts of ubuçu;

b) mixing the parts of a) with resins derived from vegetable oil.

In a preferred embodiment, resins derived from vegetable oil comprise castor oil resins.

An additional object of the present invention is the use of composites

5/63 ecological products comprising ubuçu fibers for the preparation of coating surfaces.

In a preferred embodiment, the ubuçu fibers are mixed with resins derived from vegetable oil, more specifically, castor oil resins.

In a preferred embodiment, the coating surfaces comprise floors, walls, among others.

These and other objects of the invention will be immediately valued by those skilled in the art and by companies with interests in the segment, and will be described in sufficient detail for their reproduction in the following description.

Brief Description of the Figures

Figure 1 shows the outer face of the ubuçu fiber (A) and the inner face (B) seen by SEM.

Figure 2 shows the flexural strength (MPa) of the studied composite materials, where M1 (Material 1) has a MPa of 36.7; M2 (Material 2) has a MPa of 23.2; M3 (Material 3) has a MPa of 53.1.

Figure 3 shows the average of the values of the modulus of elasticity (GPa) for the analyzed composites, where M1 (Material 1) presents GPa of 2.2; M2 (Material 2) has a GPa of 0.9; M3 (Material 3) has a GPa of 6.5.

Figure 4 shows the average values of compressive strength (MPa) for the three materials analyzed, where M1 (Material 1) has a MPa of 59.8; M2 (Material 2) has a MPa of 27.3; M3 (Material 3) has a MPa of 65.5.

Figure 5 shows the maximum force (N / mm) for each material, where M1 (Material 1) presents N / mm of 423.1; M2 (Material 2) presents N / mm of 539.4; M3 (Material 3) has an N / mm of 558.8.

Figure 6 shows the abrasive wear (%) of the analyzed materials, where M1 (Material 1) presents 24.52%; M2 (Material 2) presents 25.48%; M3 (Material 3) presents 20.13%.

6/63

Detailed Description of the Invention

The examples shown here are intended only to exemplify one of the countless ways of carrying out the invention, however, without limiting its scope.

Ecological Composite

The present invention features an ecological composite comprising parts of ubuçu and resins derived from vegetable oil.

In a preferred embodiment, resins derived from vegetable oil 10 comprise castor oil resins.

In a preferred embodiment, ubuçu fiber and plant-based polyurethanes are found with a proportion of 1; 1 of the matrix components.

In a preferred embodiment, ubuçu fiber and plant-based polyurethanes are found in a 2: 1 ratio of the matrix components.

Production process

The present invention additionally presents the process of producing ecological composites comprising the steps of:

a) obtaining ubuçu fibers;

b) mixing the fibers of a) with resins derived from vegetable oil.

In a preferred embodiment, resins derived from vegetable oil comprise castor oil resins.

Use of composites

The present invention furthermore presents the use of ecological composites comprising ubuçu fibers for preparing coating surfaces.

In a preferred embodiment, the ubuçu fibers are mixed with resins derived from vegetable oil, more specifically, castor oil resins.

In a preferred embodiment, the coating surfaces comprise floors, walls, among others.

7/63

Example 1 ~ Preferred Realization

The research with the ubuçu fiber composite and castor resin as an element for floors and coverings, aims to promote a gradual replacement of the materials currently used, such as some wood, which are in short supply. However, for the use of these materials on a reasonable scale, as economically viable engineering materials with the possibility of industrialization, a scientific study of their properties and characteristics is necessary, as well as special care with the management of these cultures in their places of production. production, aiming at sustainable production. Join this, the union of knowledge of engineering and design, since the latter is an activity responsible for the projection of the objects that surround us. Therefore, the importance of extending this research to the field of sustainable design, which is the activity that projects objects of use with total commitment to man and the environment.

Fiber from Ubuçu

The fiber used in this work was extracted from the ubuçu palm (through a venom (Embira alloys in which the feet are attached to climb the trees without branches) by the riverside population of Amapá, subsequently undergoing a softening process, which consists of nothing more than washing the casing under running water to remove impurities. This is, therefore, a totally ecological process, as it does not generate chemical residues in the environment. Therefore, for this work the fiber was received in the form of the casing (bag) in natura, measuring on average, 60 cm long and 12 cm wide.

The material was collected in September 2008, in the city of Macapá, located at 00 ° 22’0 ”in latitude, 51 * Ό3’59” in longitude and 16 meters in altitude. The climate of the municipality of Macapá is equatorial hot-humid, with a maximum temperature of 32.6 ° C and a minimum of 20 ° C, and a rainfall of no

Maximum 8/63 3.000mm. The soil is classified as broad yellow soil in tertiary dendritic-clay soils.

Castor Resin

Polyurethane comes in two-component form, consisting of a polyol and a prepolymer. The polyol was synthesized from castor oil. The prepolymer was synthesized from diphenylmethane diisocyanate (MDI) and prepolymerized with a polyol, also derived from castor oil.

The company ZORITE - Resinas Ecológicas, located in the city of Rio de Janeiro, supplied the prepolymer and the polyol for this work, with ίο polyurethane prepared by mixing these two components. Two ratios were used for mixing the components, that is, 1: 1 and 2: 1 (polyol and prepolymer, respectively). According to the manufacturer the technical data of the polyurethane are:

Density: 900 Kg / m ³

Impact Resistance: 2.36 kJ / m ²

Bending Stress: 19.1 x 10 ' ⁶ MPa (19.1 N / m ² )

Tensile Strength: 12.9 x 10 * ⁶ Mpa (12.9N / m ² )

Break Elongation: 30%

Shore Hardness (ASTM D2240): 70 to 80

The preparation of the polyurethane with the mixture of the two components, polyol and prepolymer, and ratios of 1: 1 and 2: 1 in mass took place as follows: first the quantities of each component were weighed on a digital scale 25 accurately 0.001 g; then the components were mixed (in a mixer), for three minutes, with low rotation speed (30 rpm), in order to avoid bubbles. This mixture was finally spread on the ubuçu fiber sheets to be pressed.

1. Data analysis: In this phase, the 30 information regarding the type of installation for wooden and wooden blocks were taken into account.

9/63 these were applied to the universe of development of the taco floor of the composite of ubuçu fiber and castor resin.

2. Design Project: Design the floor in view of the information previously mentioned. In this way, attention was focused on the 5 best use of fiber and resin, its manufacturing process, the development of the male and female grooves and the future installation of the floor.

Bearing in mind the dimensions of the wooden blocks, we chose to design composite blocks with dimensions of 6 cm wide, 1.2 cm thick and 28 cm long. This dimensioning was thought taking into account the best use of the fiber, since the ubuçu leaf will have its ends discarded and will have an average useful area of 64 cm in length and 26 cm in width.

To obtain a thickness of 1.2 cm, 12 sheets of 15 ubuçu will be necessary, which will be pressed according to the procedure previously described for the manufacture of the composites plates. With the sheets open, a rectangle 64 cm long by 26 wide is obtained, these dimensions being taken into account when manufacturing the metal mold to be pressed in the hydraulic press. It is possible to observe that with 20 this dimensioning it is possible to manufacture six clubs at a time, thus directing the production to obtain the largest number of pieces per pressed composite block.

After the pressing of this composite block, the equipment to be used to manufacture the clubs will be those commonly used in 25 joineries, that is: band saw, disc saw, router, bench sander, among others that are perfectly suited to the work with the composite.

1.1 Experimental Procedure

First, the experimental procedures for the characterization of the ubuçu fiber will be exposed. Having done this, 30 the experimental procedures applied to the composites now manufactured will be explained later.

10/63

1.1.1 Tests for the Characterization of the Ubuçu Fiber

-1.1.1 Scanning Electron Microscopy (SEM / EDS)

The morphological characterization of the fiber was made by scanning electron microscopy (SEM), coupled to the EDS. The two faces of the 5 sample (internal and external) were observed and the images were obtained through secondary electrons, since of all the SEM signals, this is the most used because it provides greater resolution, great depth of field, three-dimensional and easy printing. interpretation. In this analysis, an electron beam acceleration voltage of 15 - 20kV was used. Specific regions of the fibers were analyzed by EDS to determine the elements present. For this analysis, samples of ubuçu fiber were cut in dimensions 1 cm ² .

1.1.1.2 X-Ray Diffraction

The determination of the crystallinity index and the structural characterization 15 of the fiber were carried out by X-ray diffraction (powder method). It used a Siemens D5000 diffractometer, one scan being performed between 80- 5 ^and with 0.02 step ^9, voltage 40kV, current 30 mA , and copper tube (λ = 1,5406Â). The fiber crystallinity index (% CI) was calculated using the following equation (LIMA, 2005):

- ^Ac% Ãc + Ja (1) where, Aa is the area below the line of the amorphous region and Ac is the area remaining below the crystalline peaks. In order to obtain the result, the calculation was performed by dividing the remaining area below the crystalline peaks by the total area of the diffractogram.

11/63

To perform X-ray diffraction analysis, the fiber was previously ground to a fine powder. The determination of the granulometry of this powder has not been determined.

1.1.1.3 Thermogravimetric Analysis (TGA)

Thermogravimetric analyzes (TGA) were performed using a Perkin Elmer equipment, TGA-7. The analyzes were performed using different heating rates, namely, 5, 10, 15, 20, 30, 40 and 50 ° C min ¹ , with the aim of evaluating the activation energy of the fiber degradation process. This assessment was made using the Kissinger method (KISSINGER, 1957), as described below.

COSTA et al., (1999) show that every kinetic study is described by the basic equation that relates the reaction conversion rate of / dt, as a function of the concentration of one of the reactants g (a) through the velocity constant k, at constant temperature, as described below:

gives *

Since the driving force here is temperature, it is introduced using the Arrhenius equation. The Arrhenius equation determines the velocity constant k - A exp (E / RT), where E is the activation energy, R is the gas constant, T is the absolute temperature, and A is the pre-exponential factor. Thus the reaction speed increases exponentially with temperature. Substituting k in (2) according to the Arrhenius equation, we get:

(3)

12/63

For thermo-rigid systems that obey the cure reaction kinetics of order n, the conversion rate is a function of the reagent consumption. Assuming that g (a) = (1-a) .n, where α is the fractional concentration of the reagents consumed at time t, also called the degree of conversion:

& a λ / E \ f. x rHwH · ₍₄₎

Considering the degree of conversion, a, proportional to the heat involved, and assuming the temperature dependence with the Arrhenius equation, equation (3) becomes the basis for dynamic kinetic methods that use multiple scans, such as the method de Kissinger (KISSINGER, 1957), which relates the maximum peak temperature of the system's curing reaction with the heating speed.

Kissinger's method (MIRANDA et al., 1997) uses thermodynamic parameters obtained from DSC and TGA thermograms of samples evaluated at different heating speeds. The Kissinger equation is based on the second derivative of the / dt = A.exp (-E / RT). (1-a) .n equation with respect to temperature. Supposing that the relation of / dt = (da / dT) is true. (DT / dt) = (da / dT) .çp, where φ = dT / dt is the heating speed, we have:

Deriving (4) in relation to temperature, we obtain:

3 ² <z Λα-, Γ E / da \ n (6)

13/63

Considering that the derivative is equal to zero at the peak point of the cure reaction in the TGA thermogram, or at the maximum temperature of the cure reaction, Tmax, is the a _c in this T, we have:

£

RTmax ²

max.

n - ac (7)

If Tmax = (da / dT) _max and a _c are measured and n is known, then the activation energy can be calculated. Combining (4) and (5) and rearranging has:

(3)

The product n (1 -a _c ) ⁿ ' ¹ is dependent on <p, but may be close to the unit. Thus (G) can be simplified:

Φ 'RTmax * (9)

Obtaining the logarithm of this equation and differentiating it as a function of T, we have:

(10)

14/63

Thus, the Kissinger method provides the activation energy value, E, obtained from the graph of ln (qp / Tmax ² ) versus 1 / Tmax for a series of scans at different heating speeds, φ, where -E / R 5 corresponds to the slope of the line.

TGA analyzes were carried out in an atmosphere of N ₂ , from room temperature to 800 ⁵ C, in a platinum sample port and under a gas flow of 20 mL / min. The noise rate was 0.4 pts and the mass of each sample was approximately 5.102mg.

io To perform the thermogravimetric analysis, fiber was used, as already reported in item 3.2.1.2.

This test was carried out at the Composites Laboratory of the School of Chemistry at the Federal University of Rio de Janeiro - UFRJ.

1.1.1.4 Apparent Density in Relation to Water

The calculation of the apparent density in relation to distilled water was performed using a pycnometer and a digital scale with an accuracy of ± 0.0001 g. First, the digital scale was tare with the pycnometer 20 empty, then the fiber mass was weighed in the pycnometer and then the water mass in the pycnometer. Then, the mass of the pycnometer / solid / distilled water was measured, thus obtaining the apparent fiber density in relation to water.

1.1.1.5 Thermal Analysis - Differential Scanning Calorimetry (DSC)

According to MOTHÉ & AZEVEDO (2009), the DSC can provide information on characterization and specific measures, such as: glass transition, temperature and crystallization time, melting point, specific heat,

15/63 oxidation, purity, thermal stability, boiling point, degree of cure speed, reaction kinetics and others.

The differential exploratory calorimetry analyzes for the ubuçu fiber samples were performed on a Perkin-Elmer device, model PYRIS 5 Diamond DSC, under the following conditions:

Sample mass: 3.601 mg;

Temperature Range: -18 to 400 ° C;

Atmosphere: Nitrogen;

Heating Ratio: 10 ° C min '¹;

Gas Flow: 20 mL min '¹;

Sample holder: standard aluminum pan.

As the DSC allows quantitative determinations, the peak area is related to the energy involved in the process. To this end, standards are used to calibrate the equipment. These patterns show a variation of 15 known enthalpy, usually of fusion, and the peak area of this process is compared with the process area presented by the sample. In this test, Indian metal was used as a reference standard in the calibration of temperature, energy and physical constants related to these parameters.

²⁰ The samples were cut into 0.5 cm ² squares. The procedures adopted during the DSC test were as follows: First, the temperature was maintained for 1 min at -18 ° C, then a run was performed from -18 to 400 <C at 10 ° C per minute, the temperature for 1min at 400 ° C, then the sample was cooled from 400 Ό 25 to -18 ° C to 300 ° C per minute, maintaining this temperature for 15 minutes and finally the temperature was raised from -18 ° C to 400 ° C to 10 ° C / min.

The analysis of Differential Exploratory Calorimetry - DSC was performed at the Composites Laboratory of the School of Chemistry at the Federal University of Rio de Janeiro - UFRJ.

16/63

1.1.1.6 Infrared Spectroscopy

The ubuçu fiber can be considered a natural polymer consisting basically of cellulose, hemicellulose and lignin. According to WINANDY & ROWELL (1984) these constituents contribute differently to the mechanical resistance. Cellulose is primarily responsible for fiber resistance, for its high degree of polymerization and linear orientation. Hemicellulose acts as a matrix for cellulose and increases the packing density of the cell wall. Lignin not only holds the fibers together, but also acts as a stiffening agent for cellulose molecules within the fiber wall. Knowledge of the internal structure of lignocellulosic fibers and their chemical composition is important for understanding the influence of fiber on the characteristics of a composite and also in the choice of chemical and physical treatments that can be applied to the fibers. The FTIR technique makes it possible to identify the components and structure of plant fibers.

According to PIRES (2009), the spectroscopy technique in the infrared region allows to evaluate vibrations between atoms, characterizing the functional groups of the monomeric units. The technique is based on the fact that the chemical bonds of substances have specific vibration frequencies, which correspond to the energy levels of the molecule (in this case called vibrational levels). If a given molecule receives electromagnetic radiation with the same energy as one of its vibrational modes, then the radiation will be absorbed.

A Perkin-Elmer spectrometer, model 1720, ranging from 4000 cm ' ¹ to 500 cnT ¹ , was used at room temperature, using a KBr cell. As this method requires small particle size so that there is an even distribution in the tablet, the fibers were ground, kiln dried and pressed in KBr tablets (1mg sample / 100 mg KBr). The fiber was analyzed as received, that is, no surface treatment was carried out to improve its adhesion with the resin.

17/63

FTIR analyzes for ubuçu fiber were carried out at the Federal University of Rio de Janeiro, Laboratory of Composites at the School of Chemistry.

-1.1.7 Weight test

Before the tensile tests the ubuçu tissue weight test was carried out, which consists of determining the mass per unit length and can be represented in terms of grams per square meter (g / m ² ) or grams / linear meter (g / m).

To calculate the grammage of the ubuçu fiber (fabric), Standard NBR 12.984 - Determination of the grammage of non-fabrics was used. The apparatus used was a calibrated ruler (caliper) with a precision of 1 mm and a digital scale with a precision of 0.0001 g. The grammage test consists of cutting five specimens with a one-square-meter area (surface area of 0.01 m ² ) that are weighed on a digital scale with an accuracy of 0.0001 g (maximum capacity of 4000 g and minimum of 0 , 5g).

From the values of mass (m) and surface area (A) of each sample, weights (G) were determined using equation (1):

G = m / Y ₍₁₁₎

The average weighing is multiplied by 100, to determine the mass per unit area in grams per square meter (g / m ² ).

1.1.1.8 Fiber Tensile Test

The tensile test of the ubuçu fibers was carried out in an INSTRON Universal Testing Machine, model - 5569. The standard adopted was ASTM D1682 (Determination of the tensile strength and elongation in fabrics Burst load and elongation). The test speed adopted was 5 r

18/63 mm / min and loading was continuous until the specimens ruptured. Pneumatic grips were used for the tensile test of films and fibers.

The dimensions of the ubuçu fiber samples were 195 x 100 mm, 5 the longest length being fixed in each claw (lower and upper). The useful length to be analyzed in the test was approximately 140 mm. In all, 10 samples of the ubuçu fabric were tested. The claws used are 25.4 mm wide and 50.8 mm long.

The fiber tensile test was carried out at the Macromolecule Institute 10 Professor Eloísa Mano - IMA belonging to the Federal University of Rio de Janeiro.

1.1.2 Composite Fabrication and Tests Performed

1.1.2.1 Composite Manufacturing Methodology

The objective of this work is to analyze the feasibility of manufacturing the ubuçu fiber composite with castor resin, using an accessible technology to do so. Thus, it was decided to analyze the fiber in its natural state, that is, without going through a greenhouse in order to improve the resin / fiber interface 20 (D ALMEIDA, 2007). The composite plates were manufactured using the compression molding technique and involved the steps described below: Manufacture of a metallic mold in which ubuçu leaves embedded in castor or epoxy resin would be pressed. The general dimensions of the mold are: 200 x 150 x 5 mm (length, width and internal thickness). In order to facilitate demoulding, the metal mold was coated with plastic sheets, thus avoiding the use of waxes and other release products.

Aiming at a more isotropic behavior of the composite, it was decided to work with fiber sheets organized in two directions, that is, 0 and 90 °. Thus, respecting the preferential direction of the fibers, the ubuçu leaves

I

19/63 were opened and cut according to the dimensions of the metal mold in which they would be pressed (200 x 150 mm).

After cutting the leaves, the fibers and the amount of resin (epoxy or polyurethane) were weighed on a digital scale with an accuracy of 5 0.001g.

a) After weighing the fibers and the resin, the components of the respective resins used in this work were mixed, that is, castor resin (mixture of the polyol with the prepolymer) and epoxy resin (mixture of the resin with the hardener ). Thus, in the ¹⁰ manufacture of epoxy resin and castor oil, the components were weighed on the balance and then mixed (in the mixer) for three minutes. Soon after, the resin was poured into the ubuçu fibers previously organized. After the fibers were embedded in the resin, their leaves were placed in the mold respecting the position (0 ^o and ^{15 90} °) · The resins were spread on the fibers with the help of a hard bristle brush.

With the composite placed in the mold, it was subsequently closed and pressed with a force of 6 ton. The entire process was carried out at room temperature and the demoulding of the composite was done after a minimum of 20 hours.

The specimens for the tests of static bending, normal and parallel compression, abrasion, impact and DMTA were machined from the molded composite plates as described above.

For the purpose of analyzing the improvement of the fiber / resin interface, 25 a composite plate was also made with kiln-dried fibers at 80 ° C for 20 min and using castor resin in the proportion 1: 1 of its components. This procedure was adopted aiming at the reduction of moisture absorbed by the fiber and it was used the same methodology of manufacture of plates described previously. It must be emphasized that the total elimination of water is difficult, due to the hydrophilic character of the fibers, which present

20/63 structurally bonds with water molecules. Thus, in this experiment, as soon as the fibers were removed from the greenhouse, the castor resin was passed quickly to minimize new moisture absorption. Then, the composite was placed in the mold. After the manufacture of this plate, specimens were made for the flexion test at three points.

1.1.2.2 Normal and Parallel Compression

To meet the dimensioning required by NBR 7190 (ABNT, 1997) for the specimens, 04 composite plates were glued, each with an average thickness of 5mm, which in total had a thickness of 20mm. After the glued plates, the block was cut in a band saw in order to obtain the samples to be tested.

The normal and parallel compression tests were performed on an EMIC Universal Testing Machine, coupled to a 5 ton load cell. Continuous loading was applied until the specimens ruptured, and the head displacement was adopted at a rate of 100 kgf / min, according to the recommendations of NBR 7190 (ABNT, 1997). A minimum of 4 specimens were tested for each type of test, with dimensions of 20 x 30 x 50 mm, the largest being in the direction of the fibers (BERALDO et al., 2003).

These tests were carried out in the Materials Characterization Laboratory of CEPEL - Eletrobrás Electric Energy Research Center, located in Ilha do Fundão.

1.1.2.3 Mechanical Dynamic Thermal Analysis (DMTA)

For the DMTA tests, rectangular specimens were manufactured in the dimensions 30 x 8 x 3 mm, the length being greater in the direction of the fibers. A total of 6 units were tested, 2 for each type of composite (2 for castor resin composite in a 1: 1 ratio, 2

21/63 for the castor »composite in the proportion of 2: 1 and 2 for the composite of epoxy resin and ubuçu fibers). The DMTA equipment used for the analysis was a Perkin Elmer DMA, and the test was conducted in flexion mode at three points, with a fixed frequency of 1Hz and heating rate of 3 ° C / min. The temperature range used was -20 ° C to 200 ° C. This test was carried out at the Composites Laboratory of the School of Chemistry at UFRJ.

1.1.2.4 Static Flexion

The flexion test of the composites was carried out according to io ASTM D790 (Flexural Strength and Flexural Elasticity Module), in an INSTRON Universal Testing Machine, model - 5569. The test speed adopted was 2.5 mm / min with continuous loading until the specimens rupture.

In this test the ratio between the distance between the supports and the thickness of the specimen was 16, in order to minimize the component of the shear stress. A minimum of 6 specimens were tested for each type of composite analyzed. The test was carried out at room temperature (= 25 ° C), and the loading of three points was used. The test was carried out at the Institute of Macromolecules Professor Eloísa Mano - IMA belonging to the 20 Federal University of Rio de Janeiro.

1.1.2.5 Abrasion test

The abrasion resistance test was carried out in accordance with the recommendations of DIN 53 516 (1987), and consists of determining the wear resistance of the material, in an abrasimeter, for an estimated distance of 40 meters. Thus, in the scope of this work, this test simulates the dragging of a given object on a finished floor.

The abrasion test was carried out at the Composites Laboratory of PUC Rio, using a Parabor abrasometer, with sandpaper (granules

22/63 sand) number 180, with an attack index of 219.73 mg (corresponding to the abrasiveness of the sandpaper). In this test, 3 repetitions were used for each material to be tested.

To make the specimens, the composite plates were glued with their respective adhesives, cut and pressed in a 4 x 4x4 cm block, which was later turned (lathe) until reaching a cylindrical shape, with a diameter of 16 ± 0, 2 mm and an average height of 1.5 cm (dimensions compatible with the sample holder of the abrasimeter used). The abrasion resistance was calculated as a function of the mass loss suffered by the material for ίο the 40 meter course.

1.1.2.6 Scanning Electron Microscopy (SEM)

The SEM analysis aimed to observe the surface of the composites specimens after the abrasion test, comparing them with 15 other woods used for making floors. Thus, a specimen for the castor bean composite (1: 1) with ubuçu fiber was analyzed, a specimen for the castor bean composite (2: 1) with ubuçu fiber and a specimen for the composite of epoxy resin and ubuçu fiber. The woods used to compare the abrasion resistance properties with the 20 composites in this study were maçaranduba and coconut.

1.1.2.7 Dart Drop Impact Resistance Test

The falling weight dart impact tests were performed on a Fractovis Gravity Drop CEAST 6789 machine. This machine is composed of two chambers, an upper one in which a steel dart is positioned which, falling freely along two guides, it reaches the sample placed at the bottom where the adjustable supports that support it are. Knowing the height of the drop h, the theoretical impact speed is obtained from the relation v = while the energy available at the moment of

23/63 impact that depends on the falling mass m, and neglecting the energy lost by friction, is given by, E = mgh.

Different steel weights can be placed on the upper part of the dart, allowing the mass to fall freely from 2.97 kg to 25.97 kg. Likewise, the speed of free fall can vary continuously from 0.5 m / sec to 4.42 m / sec. As a consequence, the energy available for the impact varies from 0.74J to 407J.

To achieve the desired impact speed, the dart is automatically positioned at the required height using a motor connected to the machine.

io The base on which the sample is located is adjustable in order to accurately correct the height of the shot and thus ensure the impact speed. To determine the exact impact speed of the test, a fixed photocell, located parallel to the dart, measures the speed of the dart an instant before impact.

The impact resistance test was performed according to the ISO Standard

6603-2 (Puncture Impact Behavior of Rigid Plastics) and consists of determining material strength, ductility and energy absorption. For this test, a 40 mm diameter disk, a semi-spherical impactor (tup) with a radius of 12.7 mm and an impact speed of 1 m / s was used, associated with the equipment mentioned above. The test was performed at room temperature (± 20 ° C) and the impact mass used was 28.6 kg.

5 specimens with dimensions of 5 x 5 cm ² and thickness of 0.4 cm were manufactured for each type of composite, that is, 5 specimens for the composite of fiber and castor resin in the proportion of 1: 1, 5 specimens 25 for the castor fiber and resin composite in the ratio of 2: 1 and 5 specimens for the castor fiber and resin composite.

The dart tip is equipped with extensometers capable of measuring the force acting on the precursor. The force exerted on the precursor is measured indirectly: an applied load causes deformation and, therefore, a variation in electrical resistance proportional to the length of the conductor. The extensometers are placed inside a Wheststone bridge to

24/63 increase the sensitivity of the measured potential difference (signal) in relation to the variation of the extensometer resistance (indicated as “shunt”). The Wheatstone bridge is connected to fast acquisition and control equipment (Ceast DAS 4000). Through the calibration curve provided by the manufacturer, the DAS 4000, converts the potential difference in load applied to the striker, thus obtaining a force x time point curve.

This test was carried out at the Materials Science and Technology Research Institute (INTEMA), Faculty of Engineering, National University of Mar Del Plata.

1.1.2.8 Determination of the Volumetric Fraction

The volumetric fraction of fibers and matrix was estimated from the volume of the components and the composite. The equations used are presented below:

_T Fm Mm

Vfm. = otherwise vm ~ --Ve pm where Vf _f and Vf _m are the volumetric fractions of fibers and matrix, respectively, Mf is the mass of the fibers, Mm is the mass of the matrix, and pf are the density of the matrix and fiber, respectively . The Mm will be determined by weighing the composite sink, if the fiber mass has already been weighed before, the remainder is the Mm.

25/63

1.1.2.9 Design Project - Turun Mamona Floor

Subsequent to the analysis of the tests performed, a composite floor of ubuçu fiber and castor resin was made using the Industrial Design methodology, which, in summary, can be outlined as follows:

1. Project theme: Design a “taco” type floor using the composite molded by compression and formed by ubuçu fiber and castor resin;

2. Data collection: Search the market for the types of existing floors and adapt the composite material to the most suitable format, in view of its production process;

3. Data analysis: Filter the most relevant information and that will serve as guiding points for the development of the “flooring” product;

4. Design Project: Design the floor in view of the information previously collected;

5. Construction of models, mock-ups and prototypes to validate the ideas previously conceived;

6. Construction of the prototype for presentation;

7. Technical drawings and details of the production chart;

8. Development of the Visual Identity Manual that will contain the logo or product logo, use of colors in RGB and CMYK systems, use of the logo in different printed, electronic, gifts and packaging media;

9. Final presentation.

Thus, before carrying out the project, the various styles of floors existing on the market, the various types of fitting of the pieces, the application modalities and the problems presented in the application and use of each one were observed. After observing the situations mentioned,

26/63 a “taco” type floor was designed, which is easy to manufacture, assemble and maintain.

After the projected floor, a logo was created that identifies the product. These steps were carried out in order to demonstrate the importance of the Industrial Design activity in the production process and in the distribution of a given product on the market.

1.2 Characterization of the Ubuçu Fiber

1.2.1 SEM / EDS

It is possible to observe that the fibers have a preferential direction, which corresponds to the direction of the length of the bag containing the ubuçu fruits. It can also be seen that on the external face (Figure 1 A), the fibers are more visible, while on the inside (Figure 1B), there is an organic material covering the fibers.

In both regions, however, it is observed that the fibers are joined together and have a preferential direction. This difference in aspect may be due to the washing process carried out by riverside dwellers, which apparently is not efficient enough to remove the organic residue from the inner part of the fiber bag.

In the external region, it is possible to observe the presence of parenchymatic cells, characteristics on the surface of plant fibers, a fact also observed by MARTIN et al., (2009) in sisal fibers. In the inner region it is apparent that the fibers are covered with organic material.

In regions that were cleaner of organic material it was possible to observe the presence of protrusions on the surface of the fibers.

Through EDS, protrusions were identified as being rich in Si, which agrees with morphological studies of other plant fibers, such as coconut, carried out by D'ALMEIDA et al (2007) and also by GEETHAMMA et al (1998), where such protrusions are called silicated marks. A f

27/63 presence of Si-rich compounds, notably silica, is due to the fact that the fibers naturally have a defense against attack by predators, as this substance makes the fibers more abrasive.

It was also observed in the ubuçu fiber, that its surface is coated with a thin layer of wax, possibly of aliphatic and non-polar origin, called cuticle, this data being supported by studies carried out by SATYANARAYANA et al (1990).

1.2.2 X-Ray Diffraction

The result obtained in the X-ray diffraction spectrum of the fibers is very similar to those found by D'ALMEIDA et al (2007) for other lignocellulosic fibers, such as piassava and sisal, and reflects the crystalline structure of cellulose. The main peak observed at 22.76 ⁹ corresponds to the plan (002) of diffraction of native cellulose. The widest peak around 16.3 ⁹ corresponds to the signal of the planes (101) and (ioT). The peak at 33.52 corresponds to the plane (040), and is also characteristic of native cellulose, as described by RONG et al (2001) and GEORGET et al (1999).

The crystallinity index found for the ubuçu fiber, according to equation 1 (page 39) was 60.6 ⁹ . This value is consistent with the values found by RONG et al (2001) for lignocellulosic fibers such as sisal (62.8 °) and for wood fibers from wild pine (65.1 °) found by TSERKI et al (2005) .

1.2.3 Thermogravimetric Analysis - TGA

At 100 ⁹ C, the fiber showed a mass loss of about 10.45%, relative to the loss of moisture. This data agrees with values found for other lignocellulosic materials, such as curauá, found by SILVA (2010). It is observed that up to 264 ^S C, the ubuçu fiber can be considered thermally

28/63 stable, with small mass losses, basically attributed to moisture loss. This behavior is comparable to that found for the curauá fibers (D'ALMEIDA, 2007) and for the sisal fibers (JÚNIOR, 2006), where the onset of degradation occurs between 260 and 300 ⁹ C.

The second degradation stage starts at around 265 ⁵ C and is linked to the decomposition of hemicellulose. This value is comparable to that found by JÚNIOR (2006) for the decomposition of sisal hemicellulose (around 260 ° C). The temperature at which the maximum rate of mass loss occurs in this region was approximately 299-C. Although this value is ίο slightly higher than the temperature range normally associated with the decomposition of hemicellulose into iignocellulosic fibers (from 260 to 281 ^Q C), this is not unexpected, as the degradation temperature can be altered by the intrinsic differences between different lignocellulosic materials species, of composition, which ends up affecting the entire set of 15 properties of the fibers, including the peaks of degradation temperature (SCHNIEWIND, 1979).

The third stage of degradation is associated with the thermal decomposition of cellulose. The temperature value at which the maximum rate of mass loss occurs in this region was 379 ⁹ C, which directly agrees with data from 20 other lignocellulosic materials, such as that found by OLIVEIRA (2008) for sisal fiber (360 = 0) and by D'ALMEIDA et al (2007) for the fiber of caroá (350 ° C).

The activation energy value found for the ubuçu fiber was 180.2 kJ / mol, being consistent with that found for the piassava (185 kJ / mol) and 25 for the curacao (135 kJ / mol), (D ' ALMEIDA et al., 2007).

1.2.4 Apparent Density in Relation to Water

According to the pycnometer method adopted to calculate the apparent fiber density in relation to water, a value of 0.97 was obtained

29/63 g / cm ³ , a value close to that found by TANOBE (2003) for the vegetable loofah (0.92 g / cm ³ ).

1.2.5 Fourier Transform Infrared Spectroscopy (FTIR)

Table 1 shows the main movements for the absorption bands in the FTIR of cellulose, hemicellulose and lignin according to XIAO (2001); FANG (2002); SUN et al., (2004); MARTIN et al., (2006) and MEGIATTO Jr. et ak, (2007).

Initially, the general aspect of the infrared spectra of ubuçu fibers, showing bands typical of the functional groups present in lignocellulosic fibers [FANG, 2002; SUN 2004; XIAO 2001].

For a better understanding of the results obtained in the spectroscopy test, the bands of the spectrum are described separately. Absorption bands around 3288.6 cm * ¹ are observed, which are related to the stretching of OH groups, which correspond to the characteristic signal of the axial vibration of cellulose hydroxyls (hydroxyls of carbon 2,3 and 6 of glucose), [ D'ALMEIDA & CALADO (2005)]. The vibration bands around 2918.4 cm ' ¹ can be related to the stretching of aliphatic CH groups of the primary and secondary carbons, (FANG, 2004).

The bands that go from 1996 - 2112 cm ' ^{1 are related} , which correspond to the stretching of CH bonds of aliphatics and aromatics (XIAO, 2001). Related to lignin are the bands at 1607.2 cm ' ¹ , assigned to the acetal group related to carbonyl aldoses (d'ALMElDA & CALADO, 2005). At the peaks of 1513.3 cm ' ¹ , 1313.3 cm * ¹ and 1370.9 cm ¹ are the bands referring to hemicellulose and the stretching of the CO bond bond of carboxylic acids, JÚNIOR (2006).

Concluding the analysis of the infrared spectrum of the ubuçu fiber, the Figure illustrates the last absorption bands found, that is, between 600 and 1400 cm * ¹ .

30/63

The absorptions present in the region between 1239 - 1029 cm 'are observed, which are typical of xylans, (JÚNIOR, 2006). The absorption band observed at 769.5cm ' ¹ may be associated with β-glycosidic bonds between the sugar units, XIAO (2001).

Thus, it is possible to affirm that the ubuçu fiber presents a typical spectrum of lignocellulosic materials, in which bands referring to the absorption of groups present in cellulose, tignin and hemicellulose are observed.

Changes observed in the magnitude of these signals will allow to verify the efficiency of any modification on the surface of the fibers, such as, for example, an acetylation treatment as was done in coconut fiber [D'ALMEIDA & CALADO, 2000].

1.2.6 Differential Scanning Calorimetry - DSC

It is possible to observe that the DSC curve obtained for the ubuçu fiber without treatment shows typical behavior of lignocellulosic fibers, JÚNIOR (2006).

Observing the DSC, the peaks referring to the three major constituents of the fiber, cellulose, hemicellulose and lignin, can be identified, which will soon be exposed and discussed in detail. Up to 98 ° C there is what can be attributed to the loss of moisture present in the samples, a fact also verified by MARTIN et al., (2009) for sisal fibers (75 ° C). Initially, a crystallization peak around 140 ° C is observed.

Between 288 to 292 ° C, the degradation of the fraction of hemicellulose occurs, given this comparable with that found by MARTIN et al., (2009) for sisal fiber (297 ° C).

There was a decomposition of the cellulose chains of the ubuçu fiber at 332.82 ° C, this result being comparable to that found by OLIVEIRA (2008) for the sisal fiber (around 300 ° C).

According to BHADURI et al., (1994), this endotherm is a consequence of dehydration and depolymerization of the cellulose component of the fiber,

31/63 leading to the formation of flammable volatile products. CHAND et al. (1987), on the other hand, attribute this endotherm to dehydration and to the breakdown of hydroxyl groups in the cellulose molecule, resulting in the evolution of water. In their work, FAIRBRIDGE et al., (1978), carried out studies of cellulose decomposition in nitrogen and oxygen atmospheres and observed that cellulose degradation occurs via two competitive reactions of the first order, which produce a condensed fraction, a fraction gaseous and ash. Still speaking of works carried out with lignocellulosic fibers, CARASCHI & LEÃO (1999), characterized the curauá fiber in a thermal and chemical way, finding an endotherm of 270 ° C related to cellulose degradation, a value relatively lower than that found for fiber ubuçu. These small variations are due to the different compositions of the various species of plant fibers.

Thus, the DSC curve corresponding to the ubuçu fiber presents practically all events, including lignin decomposition, which occurred from 332 ° C and presented its maximum at approximately 371 ° C, a data comparable to that found by OLIVEIRA (2008) for the sisal fiber (around 400 ° C).

1.2.7 Weight

The weight of the ubuçu fiber samples was compared with three other materials, jute fiber fabric and glass fiber fabric. The objective is to analyze which of these materials would be lighter or more dense for future use as reinforcement in polymeric matrix.

The sample of the jute fiber fabric is flat, Basket type, bidirectional, approximately 2 mm thick, composed of jute threads approximately 1 mm thick. The samples of the fiberglass fabrics are also flat, Basket type, differing in the more closed and more open weave / warp, they are bidirectional, with approximately 2 mm of thickness and glass threads with approximately 1 mm of thickness.

32/63

In order to better clarify the structure of the tissues now analyzed, a brief description follows.

According to SILVA et al., (2009), the flat fabric is characterized by the weft cables intertwined in the warp cables, where the weft cables 5 intertwine one above and one below the warp cables, and so on, being that in the next column, what was above becomes below and what was below passing over and so on until the end of the fabric, as can be seen.

The basket flat fabric is a variation of the flat fabric, and for the purpose of comparing grammage, it was the type of fabric used for fiberglass and jute fiber.

The ten samples analyzed had an area of 195 x 100 mm each. According to the weight test standard, each sample was weighed on a digital scale with an accuracy of 0.0001 g. Table 12 illustrates the 15 weight variability computed for the ten samples of ubuçu fibers.

Samples Weight (g) Weight (g / m ² ) 01 3.0427 156 02 5.2871 271 03 3.3545 172 04 3.7185 190 05 4.1063 210 06 2.9746 152 07 4.3977 225 08 3.5139 180 09 3.1527 161 10 6.4513 330

Table 1 - Weights and weights computed for ubuçu fiber samples

33/63

The average weight was 3.9999g and the average weight was 204.7 ± 57.2. The weight found for the ubuçu fiber was 205.12 g / m ² . The weight of the ubuçu fiber (205.12 g / m ² ) was lower than the values found by SILVA et al., (2009), that is, 440 g / m ² to 344 g / m ² for the sisal tissue in the basket plane.

For the grammage test of fiberglass fabrics, the same methodology adopted for the ubuçu fiber was followed, that is, 10 samples with dimensions of 195 x 100 mm. Table 13 illustrates the values of weights and weights obtained for fiberglass fabrics.

Samples Weight (g) Weight (g / m ² ) 01 4.4251 226 02 4.9004 251 03 4.0045 205 04 3.7896 194 05 3.9765 203 06 4.2056 215 07 4.3789 224 08 3.9878 204 09 3.9345 201 10 3.7899 194

Table 2 - Weights and weights computed for glass fiber fabric samples

The average weight for the samples of the fiberglass fabrics was 4.6793g and the average weight was 239.4 + 12.3. The weight found 15 for the fiberglass fabric was 239.96 g / m ² . Table 14 illustrates the results of weights and weights for the second type of fiberglass fabric.

34/63

Samples Weight (g) Weight (g / m ² ) 01 4.8911 250 02 4.7689 244 03 4.2808 219 04 4.7671 244 05 4.4050 225 06 5.0548 259 07 4.5086 231 08 4.6125 236 09 4.8892 250 10 4.6150 236

Table 3 - Weights and weights computed for samples of fiberglass fabrics

The average weight for the second samples of the glass fiber fabrics was 4.1392g and the average weight was 212.2 + 17.7. The weight found for the second type of fiberglass fabric was 212.27 g / m ² . Thus, it is possible to conclude that the ubuçu fiber fabric is lighter (205.12 g / m ² ) than the two types of fiberglass fabrics analyzed, that is, 239.96 g / m ² and 212.27 g / m ² , respectively.

io In the grammage test for jute fabric, only two samples were used and their weights and weights were computed according to the procedure for ubuçu and glass fiber. Table 15 presents the results obtained.

Samples Weight (g) Grammage (g / m *) 01 4.1821 214 02 3,6951 189

35/63

Table 4 - Weights and weights computed for samples of jute fiber fabrics

The average of the computed weights for the jute tissue samples was

3.93 + 0.34. An average weight of 201.97 g / m ^{2 was found} , demonstrating that the jute fiber is slightly lighter than the ubuçu fiber (205.12 g / m ² ). It is possible to conclude, at this stage, that the vegetable fibers analyzed here are lighter than the glass fiber.

According to the NBR - 13370 standard, nonwoven is a flat, flexible and porous ίο structure, consisting of a veil or blanket of fibers or filaments, directionally or at random, consolidated by mechanical (friction) and / or chemical (adhesion) processes ) and / or thermal (cohesion) and combination thereof. Nonwovens can be classified by the manufacturing process, raw materials, fiber / filament characteristics, consolidation process, grammage, transformation and / or conversion process, or association of these elements. As for weight (weight per unit area), we have: Light: less than 25 g / m ² ; Medium: between 26 and 70 g / m ² ; Heavy: between 71 and 150 g / m ² and Very Heavy: above 150 g / m ² .

1.2.8 Tensile Strength

It can be seen that the 10 fiber samples tested showed varied behaviors, both with regard to quantitative parameters and the aspect of the force versus displacement curves. The details of the curve of each specimen are shown in Table 2.

The behavior variability observed in the force versus displacement graph can be directly related to the fact that the samples have different weights, as shown in Table 12. Another factor that can explain the behavior variability is precisely the formal interlacing aspect of each sample ( some were more intertwined than others, with differences in tissue structures).

36/63

Another fact that may still be directly responsible for the variability of mechanical behavior may be the chemical and physical heterogeneity of plant fibers (for this purpose, a chemical test must be carried out in the future to prove this hypothesis). According to CARVALHO (2005), fabrics are sets of textile fibers and, therefore, discontinuous behavior can also be understood as a result of localized ruptures, of one or some fibers, and the subsequent redistribution of efforts between the nearby fibers that start to operate under new tension levels. It can be seen that a sample of ubuçu showed a knurled curve, with a constant load drop (body 10). The rest of the bodies, on the other hand, showed maximum tensile strength and subsequently a sharp drop in the curve.

Two rupture modes were identified: the rupture of the ubuçu fiber elements, and the slipping of the fibers, both causing reasonable displacements, a fact also verified by CARVALHO (2005) when studying the sisal fibers in tension.

The maximum load supported in the tensile test was 596N for the specimen 6, and the minimum load supported was observed for the specimen 2, that is, 263N. Table 16 lists all the maximum load values obtained. The average of the values was 432N and the standard deviation was 112. The coefficient of variation was 0.26, which indicates the great variability of the data.

Samples Maximum load (N) Elongation at Break (%) at 1 368.88 4.9 2 263.18 9.4 3 565.68 6.7 4 388.33 5.0 5 360.25 6.5 6 596.53 8.8

37/63

7 498.69 8.7 8 399.09 10.9 9 333.24 7.0 10 551.16 27.4

Table 5 - Maximum load supported and elongation at break

Table 16 also illustrates the percentage elongation at break. The highest value found was for the specimen 10 (27.4%) and reflects the progressive rupture of this sample. The lowest percentage elongation was obtained by the specimen 1 (4.9%), which shows a sudden failure.

Two types of preferential failure were observed. Most samples presented a preferential fracture plane, in the longitudinal direction of the specimens and, therefore, parallel to the direction of the fibers. Some samples (2,6 and 7) ίο showed an elective slide between the region of the claws and the rest of the specimen, giving rise to a yieldor at the edge of the specimen.

The mean value obtained for the breaking load was 432N and comparing it with data obtained in the literature (in the case of tensile testing for nonwovens) it is possible to conclude that this value was higher than that found by 15 D'ALMEIDA & D'ALMEIDA (2010) for jute fabric (on average 360 N) and very close to that found for fiberglass (645 N).

For the purpose of better visualizing the appearance of the rupture in the specimens after the tensile test, Table 2 shows the samples tested and the appearance of the force versus displacement curve. It is possible to observe that there was 20 for most samples a preferential fracture plan, in the direction of fiber growth.

According to CHAND & HASHMI (1993) and GREENBERG (1989), the variation of the tensile strength and modulus of elasticity are directly related to the variation of the crystallinity index (Ic). That is, with the amount of cellulose present in the fibers, since cellulose is the only component present in the fiber that crystallizes (ROWELL et al, 1997). In this i

38/63 sense, the ubuçu fiber fits, since its crystallinity index was 60.6 °, which is quite compatible with sisal (62.8 °). However, studies have been found that demonstrate that the mechanical properties of natural fibers are also related to structural parameters such as the angle or orientation of the microfibrils and cellulose crystals that make up the fiber, (REDDY, 2005). In relation to this last item (structural parameters), it would be necessary to conduct a study to verify the angle of the microfibrils of the ubuçu fibers and try to relate them to the tensile strength data.

The modulus of elasticity obtained for the tensile test on the fabric is shown in Table 17. The average of the values of the modulus of elasticity was 297.3 + 134.6 MPa, the maximum value being 509.2 MPa (body 3 ) and the minimum modulus of elasticity attributed to the specimen 8 (167.5MPa).

Samples Module of Elasticity (MPa) 1 315.1 2 215.4 3 509.2 4 337.8 5 361.4 6 268.9 7 148.9 8 167.5 9 145.1 10 503.8

Table 6 - Tensile elasticity module for ubuçu fibers

39/63

1.3 Characterization of the ubuçu fiber composite and castor resin

In the tests performed, ubuçu fiber was evaluated as a reinforcement for polymeric matrices based on castor resin and epoxy resin. In other words, this research 5 analyzed the behavior of three types of composite materials, preferably Materials 1,2 and 3 (as specified in Table 18). To complement some more specific studies, two more composites were made, Materials 4 and 5. Thus, from this point forward, for a more objective and quick reading, when information about ίο composites is exposed, reference will be made to their numbers, that is, composite material 1,2, 3, 4 or 5, as Table 18 for material designation.

Table 7 - Designation of the researched composite materials

Material Composite Designation Material 1 Ubuçu fiber and castor resin in the 1: 1 ratio of its components. Material 2 Ubuçu fiber and castor resin in a 2: 1 ratio of its components. Material 3 Composite formed by ubuçu fiber and epoxy resin with a formulation recommended by the manufacturer. Material 4 Composite formed by ubuçu fiber (oven dried for 20 minutes at 80 ° C) and castor resin in a 1: 1 ratio.

40/63

Material 5

Composite formed by ubuçu fiber (short fiber) and castor resin in a 1: 1 ratio.

1.3.1 Static Bending Test

It is observed that Material 3 presented an average flexural strength (53.1 ± 14.9 MPa) greater than the average of Material 1 (36.7 ± 9.6MPa) and Material 2 (23.2 + 3.1 MPa).

In relation to composites with polyurethane matrix, the average flexural strength found for Material 1 (36.7 MPa) was higher than that found for Material 2 (23.2 MPa), which proves that for the formulation of resin from castor oil for its application in a composite more resistant to bending, it is necessary to produce it taking into account the 1: 1 proportion of its components. The mean flexural strength value found for Material 1 can be compared to that found by VILARIM (2003) for the composite of sisal (fabric) and castor resin (in the order of 40 MPa) and for the composite with short fibers of sisal and castor resin (22.9 MPa). In both materials researched by VILARIM (2003), these data were for untreated sisal fibers, similarly to the one used in this work. It should be noted here that sisal fiber is considered to be one of the most resistant of the most commonly used lignocellulosic fibers.

Another material that can be compared with the data obtained for Material 2 is the composite formed by coconut (treated and long fibers) in a PU matrix, where an average flexural strength of 42.5 MPa was obtained (VILARIM, 2003) . The average value of flexural strength obtained for Material 1 is comparable to that found in the literature for vegetable fiber composites with PU. Taking into account that the composite developed here was manufactured without any previous treatment in the fiber, aiming to improve its fiber / matrix interface and consequently its resistance. It is possible to affirm that the composite formed by ubuçu fiber and castor resin in a 1: 1 ratio is

41/63 technically feasible and, even more, that a future treatment on the surface of the fibers to improve the fiber / matrix interface may make this composite more resistant.

The flexural strength value found for Material 3 (53.1 MPa), 5 can be compared to that found by PAULA (1996), who obtained an average value for the sisaí / epoxy composite of 67 MPa. In addition, the mean flexural strength for Material 3 was higher than that found by ROZMAN (2000) for the coconut / PP composite (36.7 MPa) and that found by SATYANARAYMA et al., (1990) for the composite formed per coconut / polyester io (38.5 MPa). Other examples of composites that can be compared with the average flexural strength value found in this work for Material 3, can also be presented here: POTHAN et al., (1999) researching the banana / polyester composite, found an average value of flexural strength of 10 - 40 MPa; OWOLABI (1998), when researching the coconut / PF composite (PF 15 phenol formaldehyde resin), found an average flexural strength value of 30.5 - 98.0 MPa. Therefore, the average flexural strength result found for Material 3, is quite compatible with that of other composite materials found in the literature, which reinforces the technical feasibility of using ubuçu fiber as reinforcement in matrices of 20 origin. oil (epoxy, polyester, PP, PF).

It is possible to observe that Material 3 presented a greater modulus of elasticity (6.5GPa ± 0.4) than Material 1 (2.2 ± 0.2GPa) and that Material 2 (0.9 ± 0.1GPa). The values obtained are compatible with the fact that the epoxy resin has a modulus of elasticity (stiffness) greater than that of castor resin, since the 25 volumetric fraction used was the same in the three materials.

The mean values of the modulus of elasticity of Material 1 (2.2 GPa) was higher than that found by VILARIM (2003) for the sisal composite in the form of untreated fabric and castor resin (2.0 GPa) and is similar at the value of 2.3 GPa for the sisal composite in the form of fabric with treatment 30 and castor resin. Another comparison of values can still be performed between the elasticity modulus of Material 1 (2.2 GPa) and that obtained by VILARIM

42/63 (2003) for the composite coconut (without treatment) / castor resin (1.6 GPa) and coconut (with treatment) / castor resin (2.19 GPa). Material 2 presented a much lower stiffness than that of Material 1, because the proportion of components in 2: 1 is not the most suitable when a behavior with greater rigidity of the composite in question is desired.

Material 3 showed an average elastic modulus value of 6.5 GPa. This value can be compared to that found by PAULA (1996) for the sisal (untreated) / epoxy (5.2 GPa) composite and for the sisal composite (with treatment) / epoxy (6.7 GPa). The mean value of stiffness obtained by ίο Material 3 was higher than that found by ROZMAN (2000) for the coconut (without treatment) / PP (0.95 GPa) and for the coconut (with treatment) / PP (3.9) GPa). POTHAN et al., (1999) when researching the flexural modulus of elasticity of the banana / polyester composite, found a value between 2.3-5.3 GPa. Thus, it is possible to conclude that Material 3 15 presented rigidity compatible to the composites presented in the literature, when it comes to the use of vegetable fibers as reinforcement in polymeric matrices of petroleum origin.

According to VILARIM (2003), in the flexion test, the flexural strength and the modulus of elasticity are controlled by the strength of the material 20 located on the surface of the composite. The crack always starts in the specimen's tractive region and propagates to the compressive side. In this way, the fibers that effectively act in the reinforcement of the composite are those located on the tractive side of the specimen and oriented in the direction of the applied tension. The fibers with perpendicular orientation to it, act in order to decrease the mechanical strength of the composite and, in this case, the flexural strength is dominated by the flexural strength of the matrix. The good mechanical results obtained in comparison with the data referring to other composites may be due to the fact that the ubuçu fiber can be seen as being a unidirectional fabric.

Aiming to improve the fiber / matrix interface of the composite corresponding to Material 1 (since the main objective of the research is the technical viability of a ί

43/63 ecological material), it was proposed to manufacture a composite where the fiber was previously dried in an oven for 20 minutes, at a temperature of 80 ° C. This procedure was adopted aiming at the reduction of moisture absorbed by the fiber and the same methodology for the manufacture of plates of other materials was used.

In a comparison between the average flexural strength values of composites manufactured with castor resin in a 1: 1 ratio, it is observed that the drying of the ubuçu fiber in an oven to remove moisture and consequently an improvement in the fiber / Fo matrix matrix efficient with an increase in the flexural strength of composite 4 (40.1 ± 4.9MPa) in 10 compared to that found for Material 1 (36.7 MPa).

According to d'ALMEIDA (2006), drying the fiber in an oven aims at reducing the moisture present in it, and, by reducing this moisture, the fiber / matrix interface is considerably improved, reaching higher mechanical strength values than those of the fiber used without drying. The 15 results obtained corroborate this fact, and it is clearly seen that the drying process of the ubuçu fiber was satisfactory, improving the fiber / matrix interface and consequently leading to an improvement in the material's resistance.

It is observed that the modulus of elasticity in flexion of Material 4 (1.8 ± 0.3 GPa) is slightly lower than that of Material 1 (2.2). However, the stiffness of Material 20 presents a scale of values comparable to those found in similar composites in the literature (VILARIM, 2003).

In order to analyze the effect of the reinforcement geometry on the flexural strength, a composite plate was also manufactured with chopped and short ubuçu fibers. For the manufacture of this composite (Material 5), the same 25 percentage of fibers used in the previous composites manufactured with the ubuçu non-chopped fibers was used.

It is observed that the average flexural strength value found for Material 5 (15.8 GPa) was lower than the average values found for all other composites with resin in the proportion 1: 1. This result corroborates the assumption that the good bending behavior obtained can be associated with the direction observed in the ubuçu fibers. In this way, the geometry

Λ I; 44/63 natural fiber must be taken into account when manufacturing composites. However, the value found is comparable to that found by VILARIM (2003) for the composite of sisal (untreated short fiber) and castor resin (22.9 MPa).

It is known that in short fiber composites, the ends of the fibers act as points of stress concentration and thus function as crack initiation regions (AGARWAL; BROUTMAN, 1990). Under flexion, the fibers that effectively act in the reinforcement of the composite are those located on the tractive side of the specimen and oriented in the direction of applied ίο stress. Thus, in the case of short fibers, a smaller fraction of fibers is effectively contributing to the behavior of the composite, although it has the same volumetric fraction of fibers as the composite 1.

It is observed that the value of the modulus of elasticity found for Material 5 (1.7 GPa) is very close to that of Material 4 (1.8 GPa) and lower than that of Material 15 (2.2 GPa). In order to better visualize the difference in flexural behavior of the various composites, the profile of the flexion stress (MPa) vs flexion extension (mm) curves is shown in Table 3.

According to VILARIM (2003), in the flexion test, the flexural strength and the modulus of elasticity are controlled by the strength of the material 20 located on the composite surface. The crack always starts in the specimen's tractive region and propagates to the compressive side. In this way, the fibers that effectively act in the reinforcement of the composite are those located on the tractive side of the specimen and oriented in the direction of the applied tension.

²⁵ To better visualize the types of failure that occurred in the flexion test, Chart 4 illustrates the appearance of the specimens after the test, as well as some comments regarding the type of failure that occurred. From each group analyzed, a sample was taken to make comments.

I

45/63

1.3.2 Parallel Compression Test

The compression test parallel to the fibers was performed only with the three initial materials, that is, Material 1,2 and 3.

It is observed that the greatest resistance was presented by the samples of the

Material 3, that is, the composite formed by ubuçu fiber and epoxy resin, with an average of 48.8 ± 2.46 MPa. The lowest average (8.1 + 1.83 MPa) was obtained by Material 2, that is, ubuçu fiber and castor resin in a 2: 1 ratio, again, probably due to the amount of polyol and prepolymer in the 2: 1 ratio is more flexible than 1: 1 ratio. Material 1 ίο had an average tensile strength of 29.1 ± 0, 99 MPa.

Even though the objective of this research is not the application of the material in the vertical direction for the manufacture of floors, the parallel compression test is necessary because it provides the opportunity to observe the mechanical behavior of the fibers in the vertical direction.

The tensile strength results for the parallel compression test obtained were slightly lower than those found by OLIVEIRA (2005) for the bamboo and resorcinol composite (40.74 MPa) and those found by GARBINO et al. (2002), from 55 MPa.

From the results found for Material 2 (8.1 MPa) it is evident that this proportion did not present good resistance to compression, since the specimens flambelled in an elastic way at their bases. The value obtained of parallel compressive strength for Material 1 (29.1 MPa) is in the comparison range of the values found by FREIRE & BERALDO (2003) for the composite of bamboo and resorcinol formaldehyde (30 MPa).

It can be seen that, with the exception of Material 2, the 4 specimens of each material showed uniform behavior, both with regard to quantitative parameters and in the aspect of the curves.

In some projects, the major requirement is compressive strength. Thus, the designer must specify a material that presents good resistance to compression, that does not deform easily and that ensures

46/63 good dimensional accuracy when requested for compression efforts. The ratios that apply to the tensile test are also valid for the compression test. This means that when a body is subjected to compression, it will initially undergo an elastic deformation and then a plastic deformation. Hooke's Law is valid for the elastic deformation phase in the compression test (TELECURSO MECÂNICA, 2000).

The factors that condition the compression rupture mode occur at the structural level (specimen geometry), at the macro-structural level (at the blade level) and at the micro structural level (fiber, matrix and interface level). According to DIETENBERGER et al., (1999), compression failures usually appear on the surface of the part as small white lines or through areas formed by the kneading or warping of the cells.

According to FLECK (1996), the axial compression strength of composite 15 is usually relatively low, since most of the individual compression failure mechanisms are dictated by the matrix properties.

A problem that always occurs in the compression test is the friction between the specimen and the plates of the test machine. Another problem is the possible occurrence of buckling, that is, curving of the specimen 20 due to the effect of instability in compression. Buckling occurs mainly in specimens that are longer in relation to the diameter.

In his studies with fiberglass composites and polymeric matrices, FLECK (1996) identified that the main mechanisms of compression failure 25 are:

1. Elastic micro buckling: it is a deformation due to instability and deforms the matrix by simple shear;

2. Plastic micro buckling: deformation due to instability that occurs at stresses large enough for ^the matrix to deform by shear in a non-linear manner;

47/63

3. Fiber crush: the failure occurs at the fiber level due to shear instability, such as buckling inside the fiber;

4. Fissures / cracks: fissures in the matrix parallel to the fibers, being associated with a low tenacity of the matrix;

5. Curve delamination: it is associated with low matrix toughness;

6. Shear planes: fracture in a band oriented around 45 ° in relation to the loading axis.

Some failures under compression for fiberglass composites and polymeric matrices present in the study by FLECK (1996), as they are a fibrous composite, will be considered a parameter to perform a correlation with the failures that occurred in ubuçu fiber composites and resins castor / epoxy.

Studies of compression failures in composites reinforced with vegetable fibers are quite rare in the literature. Thus, this research aims to contribute, albeit briefly, with a study on the types of failures that occurred in each composite material researched.

According to JUMAHAT et al., (2009), among all failure modes in fibrous composites, buckling is recognized as the dominant failure mechanism in axial compression. The compression failure is probably caused by the local instability of the fibers incorporated in the matrix. The instability may be due to the waviness of the fiber (since it is not a rectilinear body), a free edge region, a region rich in resin and a poor interfacial bond (fiber / matrix).

In fact, the buckling failure in this research was identified more frequently in the specimens submitted to the compression test, followed by delamination and shear planes around 45 ° in relation to the loading axis.

48/63

In order to better visualize the appearance of the rupture in the specimens after the parallel compression test, the samples tested and the appearance of the stress curve (MPa) versus elongation (mm) are shown here, side by side (Table 5).

In order to have a clearer idea of the aspect of the failures that occurred in each type of material, the most significant samples of each group were exposed. Sample 3 of Material 2 showed a small elastic buckling and much less severe than what happened, for example, with Sample 2 of Material 3, in which there was buckling followed by delamination. Still comparing, sample 4 of Material 1 showed failures much less severe than those occurred in Material 3.

1.3.3 Normal Compression Test

The normal compression test for fibers was also performed only with the three initial materials, that is, Material 1,2 and 3.

It is observed that the greatest resistance was presented by the samples of the

Material 3, that is, the composite formed by ubuçu fiber and epoxy resin, with an average of 65.5 ± 1.44 MPa. The lowest average (27.3 ± 2.57 MPa) was obtained by Material 2, that is, ubuçu fiber and castor resin in the proportion of 2: 1, probably due to the adhesive being more flexible in this proportion. Material 20 1 showed very close values (59.8 ± 5.9 MPa) to those of Material 3.

In this research, the sense of making a normal compression test is precisely to try to analyze the technical feasibility of applying the composites now presented in the manufacture of floors. In the case of alternative materials for floors, bamboo is widely used today, and, for this reason, it was decided to compare the properties of this material with the data obtained for the composites of this research.

The results obtained were superior to those found by OLIVEIRA (2005) for the composite formed by sheets of bamboo and phenolic resin (17.3MPa) and those found by GONÇALVES et al, (2002), who obtained 18 30 MPa for the bamboo composite and phenolic resin. In this way, the value of

49/63 compressive strength obtained for Material 1 (59.8 MPa) is considered acceptable for application of the material in the manufacture of residential floors.

For comparison purposes with woods commonly used for the manufacture of residential floors, we have the values of compressive strength 5 found by CALIL JÚNIOR et al., (2000) for jatoba wood (91 MPa) and maçaranduba (83 MPa).

According to PICANÇO (2005), the compressive strength of the composite is influenced by the type, length and volumetric fraction of the fibers. During a compression test of ίο mortar composites reinforced with vegetable fiber, it is common for the specimen, already broken, to keep its parts joined by the fibers, thus not losing its continuity and avoiding its catastrophic fracture. This reaffirms the role of the fibers as elements capable of connecting the faces of the cracks, giving the material a certain capacity, even if minimal, of loading, after the cracking of the matrix.

In fact, in the normal compression test carried out in this research, some specimens, even having already been ruptured, presented resistance to avoid a catastrophic fracture. Some specimens of Materials 1 and 3 did not break the maximum capacity of the universal testing machine (approximately 68 MPa), thus, the test ended as soon as the sample reached maximum strength without breaking.

In order to better visualize the appearance of the rupture in the specimens after the compression test, side by side (Table 6), the tested samples and the appearance of the stress curve (MPa) versus 25 elongation (mm) are illustrated here.

1.3.4 Impact Test

According to VILARIM (2003), characterizing the mechanical properties of composite materials is not a simple task. The first question that arises is how to apply the standard methods, developed for materials

50/63 homogeneous, to composite materials. This is especially the case for local properties, such as impact resistance and fracture toughness parameters (BURZIC et al., 2001).

The behavior regarding the impact of fiber-reinforced composites depends on many factors including the nature of their constituents, fiber / matrix interface, the construction and geometry of the composites, as well as the test conditions. The specimen fracture occurs due to factors such as matrix fracture, fiber / matrix detachment and fiber pullout, called “pull-out”. Of these, the “pull-out” mechanism can be highlighted for being an important energy concentration / dissipation mechanism in the fiber reinforced composite (RAZERA, 2006).

In the mechanism, known as "pull-out", the fiber is pulled out from inside the matrix, usually after the crack propagation, that is, shortly after the occurrence of the mechanism that is known as "fiber bridging" and, in most cases , is found in polymeric composites reinforced with fragile fibers, ductile matrices and an interface involving intense interactions. In this way, the crack propagates through the matrix and the fiber forms a bridge interconnecting the two surfaces of the fractured matrix. On the other hand, the crack tip gives rise to regions with stress concentration even outside the propagation plane of the same. These stresses can cause “debonding” (3) which is characterized by the detachment between the fiber and the matrix, due to the presence of a weak interface. This mechanism leads to the disruption of the existing interaction between the fiber and the matrix. This region in which tensions are concentrated, may also lead to fiber failure (4) due to its high fragility, or even matrix failure (5) for the same reason (ANDERSON, 1995); (TITA, 2002).

NAGLIS (1998) tells us that whatever the intended application, structure or components are subject to static and / or dynamic efforts that can result in damages that lead the material to premature failure. Dynamic or impact loading is particularly important to be

51/63 analyzed, because, under high deformation rates, the fracture of the material can occur preferably in a fragile way.

In the analysis of the results of the impact tests of the “dart fall” type, the impact resistance was quantified in terms of the impact energy absorbed per unit of thickness of the specimen in the rupture region.

It is possible to observe that the samples of Material 1 showed a uniform behavior, with an average maximum impact resistance of 1800N. This value was higher than that found by JÚNIOR (2006) in an ίο impact test of the Izod type for phenolic matrix composites reinforced with sisal fibers (400 N). RAZERA (2006) in an Izod impact test for the hybrid composites of phenolic matrix, cotton fibers and jute (fabric) found a value of 92.7 N. For the lignin and coconut fibers composite, an impact resistance was obtained of 643.8 N and for the composite of lignin 15 and banana fibers an impact resistance of 159 N was obtained (RAZERA, 2006).

VILARIM (2003) carried out a fracture toughness test with two loading speeds (0.5 and 1 mm / min) for castor oil-based polyurethane resin and obtained the values of 275 N and 250 N, 20 respectively. From these data it can be concluded that the ubuçu fiber worked as a reinforcement in the impact resistance, since it obtained a value of 1800N.

It is possible to observe that the samples of Material 2 showed a uniform behavior, with a maximum impact resistance of 3250N.

This value was higher than that found for Material 1 (1800N), showing that Material 2 is ductile in nature, since the mixing ratio of the resin in 2: 1 makes it more flexible than the proportion of resin in 1: 1 of the Material 1. The maximum impact resistance values computed for Material 2 were from samples 5 (3200 N) and sample 6 (3180 N). The minimum impact resistance value was from sample 2 (1750N). Note that the lowest impact resistance value for Material 2 (1850 N from sample 2) is still i

52/63 higher than the maximum value found by Material 1 (fragile in nature, due to the proportion of castor resin used, 1: 1).

Table 8 illustrates the images of the fractures of the specimens after the impact test. The photos were taken from below. Firstly, the type of fracture that occurred in the samples can be considered as semi-ductile. The failure mechanisms were fracture in the preferential direction of the fiber, fracture of the fiber / m actress.

Comparing the images in Tables 7 and 8, it is possible to observe that the fractures that occurred in Material 1 were much more severe than those that occurred in ίο Material 2, the latter being more resistant to impact, presenting a more ductile behavior (ductile type fracture) when compared to Material 1. The metal dart managed to cross (although less severely than Material 1), three samples, that is, 2, 3 and 5. This ductile behavior of Material 2 must be directly related to the question of castor resin (2: 1 ratio) which makes it exhibit a more flexible, malleable behavior than the proportion of resin 2: 1 (rigid), since the geometry and amount of fibers used in the two composites was the same .

It is possible to observe that the samples of Material 3 showed a uniform behavior, with a maximum impact resistance of 2980N. This value was higher than that found for Material 1 (1800N), and lower than that found for Material C (3250N), showing that Material 3 is more fragile in nature than Material 2. The maximum impact resistance values computed for Material 3 were from samples 3 (2980 N) and 25 sample 2 (2700 N). The minimum impact resistance value was from sample 1 (2250N). Note that the lower impact resistance value for Material 3 (2500N) is still higher than that found for Material 1 (fragile in nature, due to the proportion of castor resin used, 1: 1).

Comparing the images in Tables 7, 8 and 9, it is possible to observe that 30 fractures occurred in Material 3 were much more severe than those occurred in Material 2, the latter being more resistant to impact, presenting a λ

J 53/63 more ductile behavior when compared to Material 3. The metallic dart managed to pass through all the samples of Material 3, proving the more fragile nature of this material compared to Material 2. In samples 3 and 6 of Material 3 it is possible observe that the fibers that make up the 5 threads are partially rotated in a single direction, due to the natural twist of the threads that form the ubuçu fiber fabrics.

The force values were divided by the thickness of the sample, in order to take into account its geometry. It is possible to observe that the lowest average of the maximum relative force (N / mm) was computed for Material 1 io (425 + 61.1 N / mm), followed by Material 2 (545 ± 101.1N / mm) and finally , by Material 3 (555 ± 53.5N / mm).

According to VILARIM (2003), the two basic mechanisms of energy absorption of materials are the creation of new fracture surfaces and the deformation of the material.

^{15 In the} impact test, the first mechanism to operate is the deformation of the material and, if the incident energy is high enough, crack initiation and propagation occur, acting as the second energy absorption mechanism. In the case of composites, the mechanisms of energy absorption that most contribute to their toughness occur during the propagation of crack and are the extraction of fibers and the detachment of the interface (GASSAN, 1999). These mechanisms consume energy by creating a larger fracture surface area within the composite and by the friction work at the interface.

The average values of fracture energy for each type of material are presented in two different plots. First we have the initiation and then the propagation of the fracture. For the initiation stage, Material 1 had a value of 1256.6 + 401.8 Nmm / mm, Material 2 had a value of 2256 ± 519.9 Nmm / mm and Material 3 had a value of 1750.7 ± 530, 6 Nmm / mm. For the propagation stage, Material 1 presented a value of

1674.4 + 542.6 Nmm / mm, Material 2 presented a value of 3511.7 + 544.8

Nmm / mm and Material 3 showed a value of 2728.4 ± 238.7 Nmm / mm.

54/63

It is possible to observe that Material 2 needed more energy to start the fracture (2256 Nmm / mm), differently from Material 1, which started its fracture with 1256 Nmm / mm and Material 3, with the fracture start energy of 1750 Nmm / mm. The fracture propagation energy in Material 2 was also higher (3511 Nmm / mm) than that of Material 1 (1674 Nmm / mm) and Material 3 (2728 Nmm / mm). It is concluded, once again, that Material 2 showed a greater capacity to absorb impact energy to initiate and propagate the fracture than the other two materials. The energy required for the start and the propagation of fracture of Material 2 was 5767 Nmm / mm, that is, well above that of Material 1 (2930 Nmm / mm) and Material 3 (4478 Nmm / mm).

Thus, according to the data presented, the material that presented the best behavior, that is, it absorbed a greater amount of energy and consequently presented a greater impact resistance was Material 2. It presents a semi-ductile behavior, with plastic deformation before of the fracture. At maximum force, Material 2 showed a greater displacement (8.1 ± 0.8 mm), when compared with Material 1 (4.4 ± 0.61 mm) and with Material 3 (4.5 ± 1.1 mm).

1.3.5 Analysis of impact fracture surfaces using scanning electron microscopy (SEM)

The following micrographs illustrate some important factors for fracture analysis of Material 1.

In short, the main failure mechanism for Material 2 was a deformation of the matrix followed by fracture. Both fracture patterns are observed, that is, ductile and fragile fracture, which suggests that the matrix is of a semi-ductile nature.

The following micrographs illustrate some important factors for fracture analysis of Material 3. There was massive detachment of the fibers, followed by the fracture and subsequent “puli-out” or pullout of the fibers on the surface of

I

55/63 fracture (white arrows). This fact was also observed by RAZER A (2006), for composites of phenolic resin reinforced with banana fibers and phenolic resin reinforced with long coconut fibers.

The “pull-out” mechanism dissipates energy during mechanical friction, 5 increasing the impact resistance, as the fibers dissipate energy during the debonding process, tending to stop the crack propagation (RAZERA, 2006).

Another observed factor is a hole left by the fiber that was extracted from the matrix, although it was observed that there was, in the matrix, a type of fracture ranging from a fragile type to a semi-ductile (black arrow). The occurrence of fiber extraction from the matrix was also observed by VILARIM (2003), for the composite of sisal and castor resin.

In short, the failure mechanisms in Material 3 were the massive detachment of fibers, fracture and extraction of fibers from the matrix. As well as, a deformation of fragile to semi-ductile of the matrix which led to the fracture of both the matrix and the fibers.

It was observed that in Material 3 the phenomenon of “pull-out” or extraction of fibers from the matrix occurred, that is, there was energy dissipation by mechanical friction, and simultaneously the “debonding” took place. The voids and bubbles affect the material's properties, as they act as stress concentrators, reducing the absorption of impact energy (LIN, 1990). Thus, it is possible to conclude that materials 1 and 2 presented a better interface than material 3.

VILARIM (2003) tells us that the high energy absorption of composites 25 with fabric is due to the friction between the fibers that form the fabric threads, resulting from the shearing action. Some authors consider the high fracture toughness and impact resistance the main advantage of composites with natural fibers. However, the mechanical strength of the composite is very dependent on the interfacial strength and, as seen, a high strength on the interfacial can decrease the energy absorption capacity of the composite.

In short, greater impact resistance is not always associated

J

56/63 with high mechanical resistance. The balance between these properties will depend on the intended application.

1.3.6 Abrasion test

One of the main objectives of this work is to, in a first analysis, prove the technical and ecological feasibility of using alternative materials in the manufacture of floors, specifically, the polyurethane composite of castor resin and ubuçu fiber. Therefore, it is known that to manufacture this type of product, research is needed, focusing on other parameters related to the final product such as: acceptance, price, material resistance (resistance to abrasion, compression and flexion), aesthetics and functionality.

SILVA & BITTENCOURT (2002) affirm that in Brazil much research is done on abrasive wear on metals, but in relation to wood (a natural composite), there is still a long way to go in establishing procedures, standards and characterization of materials for purposes specific, such as the manufacture of floors, for example.

This part of the research is dedicated to the study of the castor resin composite and ubuçu fiber for the production of components (specifically floors and coverings) to be used in homes. Based on the knowledge of the properties of raw materials hitherto presented and discussed, regional availability and the ecological problem in relation to some woods, it is proposed to use an alternative material that generates income for the population that produces it, specifically, the riverside population. of the State of Amapá. Thus, the abrasion test strengthens this idea, providing technical data on the use of this composite in the manufacture of hardwood floors. After presenting the results obtained for the composites in the abrasion test, the requirements that a hardwood floor needs to have to be applied in a residence will be presented.

*

57/63

Each sample had an approximate diameter of 16 mm and a thickness of at least 15 mm. As the first stage of the abrasion test consisted of weighing each specimen, the computed values for the mass mean of the specimens divided by 5 groups of materials. Material 1 presented an average weight of 2.9673 ± 0.10, Material 2 an average weight of 3.2768 ± 0.01 and finally Material 3, an average weight of 3.2456 ± 0.02 . It is possible to observe that Material 2 presented a higher density for its specimens, followed by Material 3 and finally, Material 1.

ίο Material 1 had a final mass of 2.23 + 0.08, Material 2 had 2.44 ± 0.05 and Material 3 a final mass of 2.59 ± 0.07.

Material 1 showed slightly less abrasive wear (24.52 ± 0.08%) than Material 2 (25.49 ± 0.05%), but statistically, as previously presented, these results do not differ. What can be concluded in relation to the abrasive wear of these two materials is that both presented a final mass of more than 60% after the 40m course, which technically makes them viable to be used as flooring materials.

In relation to this last statement, KOGA & BITTENCOURT (2002) tell us that in the abrasion test, for a material to be considered of 20 commercially acceptable resistance for making floors, it is necessary that it reaches the 40m, with a final mass greater than 60% of its initial mass.

Of the analyzed materials, Material 3 showed the lowest abrasive wear (20.13 + 0.07%). This result can be directly related 25 to the fact that epoxy resin is more rigid than castor resin in the two proportions analyzed.

As the composite materials analyzed here are considered alternative, we searched the literature for results of abrasion tests carried out with materials in this category. In this context, OLIVEIRA (2005) 30 researched the use of glued laminated bamboo in the manufacture of floors and found an abrasive wear for this grass of 20.34%, comparable with the

58/63 results found here. Regarding so-called alternative woods, BARCELLAR & D’ALMEIDA (2010) carried out an abrasion test for the pupunha wood and determined an abrasive wear of 9.8%, whereas for the coconut palm an abrasive wear of 19.8%.

OLIVEIRA (2005) found abrasive wear values for ipe (12.63%), sucupira (15.68%) and jatobá (23.67%). Comparing the values of abrasive wear for hardwoods with those found in this research, it is possible to conclude that the composites now analyzed, present technical viability of use in the manufacture of floors, since the abrasive wear ίο presented was less than 26% for all the samples. It is important to emphasize that the comparison between the composites and these woods was due to the fact that the woods are considered natural composites.

1.3.6.1 Scanning Electron Microscopy - Analysis of the

Surface of samples used in the abrasion test

After the abrasion test was carried out, the tested samples were conditioned and taken to the electron microscopy laboratory in order to analyze the surfaces that suffered abrasion. In this way, a sample was collected from each group of material analyzed and images of 20 four magnifications were captured under microscopy, that is, 50; 100; 150 and 200 times.

It is possible to observe that the surface of Material 1 presents a risk caused by the abrasive sand granules (white arrow), in general, the existence of fibers exposed on the surface is not perceptible.

It is possible to observe that the surface of Material 2 is quite irregular and has an excessive number of exposed fibers. Comparing this block of micrographs with those of Material 1, it is possible to conclude that greater amounts of fibers were exposed on the surface of Material 2, which makes it less resistant to abrasion.

In general, the surface of Material 3 is more even in terms of damage than that of Materials 1 and 2. The surface of Material 3 is

I

59/63 quite regular and has some exposed fibers. Comparing this block of micrographs with those of Materials 1 and 2, it is possible to conclude that lesser amounts of fibers were exposed on the surface of Material 3, as well as the surface was more regular or less damaged than the others, which should be Indeed, the rigid nature of the epoxy matrix. In general, the three composite materials analyzed here showed no cracks in their surface, which demonstrates the effectiveness of reinforcing the ubuçu fibers in the polymeric matrices.

Pupunha and coconut woods are considered alternatives and, according to BARCELLAR (2010), they can be applied in the manufacture of floors and coverings, since both showed wear of 9.8% and 19.8%, respectively. In this way, the pupunha and coconut palm samples were tested in abrasion and their surfaces analyzed in SEM to be compared with those of the composite materials studied here.

In general, the surface is more regular in terms of damage than that of Materials 1 and 2, resembling the surface presented by Material 3.

In general, the surface is more irregular in terms of damage than that of Materials 3 and pupunha, being quite similar to Materials 1 and 2.

Comparing the coconut and peach palm woods with the composites manufactured in this research, it is possible to conclude that the cracks found in these are harmful, as they cause it to expand more than normal, compromising its structure, even having less surfaces damage than that of composites (specifically pupunha). The composites in general presented a more resistant surface to the formation of grooves from the sandpaper granules, this problem was identified in greater quantity in the alternative woods, which proves the technical feasibility of using the composites now analyzed in the manufacture of floors and coatings.

I

60/63

1.3.7 Dynamic Mechanical Analysis - DMA

According to GEETHAMMA (2005), DMA is an efficient tool for determining the glass transition temperature, Tg of polymeric materials, among other parameters that can be evaluated. The dynamic Tg can be indicated as the temperature at which the maximum tan, the maximum E ”occurs or at the midpoint of the E 'curve versus temperature (MURAYAMA, 1982).

PAIVA (2006) mentions that the mechanical dynamic properties of composites are determined by the properties of their components, system morphology and the nature of the interface between the phases.

TRINDADE (2005) explains that in the analysis of results of composites reinforced with lignocellulosic fiber, one must consider that there is a difference between the coefficients of thermal expansion of the fibers and the matrices, which can influence the results.

Table 19 shows the average results of the glass transition temperature (Tg), peak tan 5 and storage module (E ¹ ), at room temperature, for the analyzed composites.

Table 8 - Glass transition temperature (Tg), tan $ peak and storage module for the three composite materials analyzed

Composite Material (maximum Tan δ) Peak d _and tan δ AND' (Pan) (25 ° Ç) Mater 87 0.24 5.30 ial 1 E + 0.12 Mater 25 0.22 3.05 was going! 2 E + 0.12 Mater 60 0.33 6.50

J

61/63 ial 3

E + 0.12

According to Table 19, the Tg value (maximum tan 8) was 87 ° C for Material 1, slightly higher than that found by VILARIM (2003) for the untreated sisal fiber composite with castor polyurethane (83.1 ° C) and for the coconut fiber and castor resin composite (82.1 ° C). For Material 2, the lowest Tg (25 ° C) was found, this value being due to the flexible nature of the castor resin matrix in the proportion of 2: 1 of its components. The values of tan delta peaks found for Material 1 (0.24), for Material 2 (0.22) and for Material 3 (0.33) are comparable to those found by VILARIM (2003) for composite of sisal fiber and castor resin (0.25) and for the composite of coconut fiber and castor resin (0.30).

In general, the composites showed a decrease in the storage module (E ') with increasing temperature. As can be seen, for Materials 1 and 3, from 60 ° C these composites show a sharp drop in the storage module, E ', which reflects the loss of elastic properties in relation to the plastic properties of the materials. This temperature was also found by VILARIM (2003) for composites of sisal fiber and castor resin and coconut fiber with castor resin.

Once the glass transition is over, above 100 ° C, there is a matrix in the rubbery state while the ubuçu fibers maintain their rigidity, as they only show some change in behavior with the temperature from approximately 260 ° C, as seen in the TGA test (page 116).

It is also possible to verify that Material 2, which has a matrix of a more flexible nature than that of Material 1, presented a low storage modulus, when compared to Materials 1 and 2. This fact is credited to the really more flexible and less resistant matrix as it was formulated. Thus, according to the results of the

62/63 storage module (E ^! ), Materials 1 and 3 are considered more rigid than material 2, because the material matrix 2 is more flexible than the material matrix 1 and 3.

RAY et al., (2002) affirm that the value of E 'decreases due to the dissipation of energy in the transfer of effort in the fiber / matrix interface and in composite with weak interface there is a greater dissipation of this energy. Thus, it is possible to study the possibility of a possible weak interface for Material 2, since it had a smaller storage module than other materials.

io GEETHAMMA et al., (2005) state that the incorporation of fiber causes an increase in the loss module, E ”, indicating an increase in heat dissipation. The introduction of fiber into the system restricts the mobility of polymeric chains immediately close to the fibers, and as a result, the glass transition of the matrix can occur over a wide temperature range. This factor may be leading to the behavior of Materials 1 and 3 observed from 60 ° C, in which wide "bands" for the composites (E ") are observed, which is also reflected in the tan £ curves.

It is possible to observe that Material 1 presented a tan 5 of 0.24, a value very close to that found by VILARIM (2003) for the composite of 20 sisal fibers and castor resin (0.25). Material 2 had a tan £ of 0.22 and Material 3 a tan 5 of 0.33. These values can once again be compared with those found by VILARIM (2003) for the composite of coconut fiber and castor resin (0.30).

According to VILARIM (2003), the value of tan £ indicates the degree of mechanical damping and the more ductile the material, the greater the degree of damping. The more rigid composites showed a lower degree of damping, that is, Materials 1 and 2.

According to DUTRA et al., (2000), with a greater interaction between the fiber and the matrix, there is a restriction on the movement of the segments of the macromolecular chains located on the surface of the fiber. This restriction causes a

63/63 increase in glass transition temperature and decrease in damping, with a consequent increase in material stiffness. The fact announced by this author happened with Material 1, since it had a tg (maximum tan delta) of 87 ° C and its degree of damping (peak tan delta) 5 was 0.24. It can be concluded that this material had a good fiber / matrix interaction, being more rigid than Materials 2 and 3.

Those skilled in the art will value the knowledge presented here and will be able to reproduce the invention in the modalities presented and in other variants, covered by the scope of the attached claims.

Claims (13)

Claims Ecological Composite, Ecological Composite Production Process and Use of Ecological Composite in Coatings. 1. Ecological composite characterized by being highly resistant and comprising ubuçu. 2. Ecological composite characterized by comprising parts of ubuçu and resins derived from vegetable oil. 3. Ecological composite according to claim 1 or 2, characterized in that the composite is a mixture of ubuçu fibers and castor resin. Ecological composite according to claim 1 or 2, characterized in that resins derived from vegetable oil comprise castor oil resins. 5. Ecological composite according to claim 2, characterized in that the resins derived from vegetable oil comprise vegetable-based polyurethanes. 6. Ecological composite, according to claim 2 or 5, characterized in that the ubuçu parts and the resins derived from vegetable oil are in a 1: 1 ratio. 7. Ecological composite, according to claim 2 or 5, characterized in that the ubuçu parts and the resins derived from vegetable oil are in a 2: 1 ratio. 8. Production process for ecological composites characterized by understanding the steps of: a) obtaining parts of ubuçu; b) mixing the parts of a) with resins derived from vegetable oil. Process according to claim 8, characterized in that resins derived from vegetable oil comprise castor oil resins. 10. Use of ecological composites characterized by composites comprising parts of ubuçu and being used for the preparation of 2/2 coating surfaces. 11. Use according to claim 10, characterized in that the ubuçu parts are mixed with resins derived from vegetable oil. 12. Use according to claim 10, characterized in that resins 5 are castor oil resins. 13. Use, according to claim 10, characterized in that the coating surfaces comprise floors, walls, among other coating surfaces, as well as the combination thereof.