Journal Information
Vol. 8. Issue 6.
Pages 5536-5547 (November - December 2019)
Share
Share
Download PDF
More article options
Visits
288
Vol. 8. Issue 6.
Pages 5536-5547 (November - December 2019)
Original Article
DOI: 10.1016/j.jmrt.2019.09.022
Open Access
Novel application of a thermoplastic composite with improved matrix-fiber interface
Visits
288
S. Lopez de Armentia
Corresponding author
slopezar@pa.uc3m.es

Corresponding author.
, B. Enciso, G. Mokry, J. Abenojar, M.A. Martinez
Dept. of Material Science and Engineering, Univ. Carlos III de Madrid, IAAB Av. Universidad, 30, 28911 Leganes, Spain
Article information
Abstract
Full Text
Bibliography
Download PDF
Statistics
Figures (12)
Show moreShow less
Tables (8)
Table 1. Marble composition (%wt.).
Table 2. Glycerol contact angle (º) on fibers with APPT.
Table 3. Contact angle (º) on as-received, treated and calcinated fibers.
Table 4. Surface energy and its components of glass fiber as-received.
Table 5. Asignation of peaks of untreated, APPT treated and calcinated fibers spectra.
Table 6. Surface energy and its components.
Table 7. Crystallinity and melting temperature for LDPE and composites.
Table 8. Absorbed energy and fall speed in drop tower tests.
Show moreShow less
Abstract

Ornamental materials are widely used in many fields such as construction. However, they have a main issue: up to 60% of processed marble pieces break due to their low toughness. In this work, an externally bonded reinforcement which can cope with this problem is presented. It must absorb the shock energy when marble is subjected to an impact without increasing severely the production cost. A thermoplastic polymer matrix composite is proposed due to its high toughness compared with conventional thermoset polymer matrix composites. However, thermoplastics present an issue related to the adhesion with the reinforcement, therefore fibers must be treated to solve this matter. Atmospheric pressure plasma torch offers a time saving and environmental friendly solution for improving adhesion between the fibers and the matrix. Finally, when marble is reinforced with the composite an important increase in absorbed energy is found in both, Charpy and drop tower test.

Keywords:
Glass fiber reinforced thermoplastic
Atmospheric plasma treatment
Fiber/matrix adhesion
Impact behavior
Externally bonded reinforcement
Full Text
1Introduction

In the last years, the use of thermoplastics as matrix in composite materials has been growing steadily due to their recyclability and ability to be processed rapidly. They also offer improved fracture toughness and they can be joined using welding techniques. Besides, they can be easily stored for long periods of time without loss of performance or ability to be processed and it is easy to repair or remold them [1–3]. These thermoplastic composites can be used in many different industries, such as automotive, aerospace or sports [4]. In this work, they are proposed to be used as reinforcement of some brittle materials in the construction field, such as natural stone, where brittleness is one of the most important issues.

When studying composite materials, it is known that their final properties do not depend only on the properties of the separate constituents (matrix and reinforcement), but in how they interact with each other. Interaction between them occurs on the interface and can be studied from the adhesion between matrix and reinforcement. Adhesion plays an important role to transfer the stress from the matrix to the fibers, and hence it changes the final properties [5].

In order to increase the adhesion, surface properties of fibers can be modified by physical or chemical treatments. Chemical modification methods include the use of surface modifiers on fibers, as silane coupling agents or graphene oxide [6,7], compatibilizers [8], electrochemical treatment [9] or oxidation [10].

Physical modification methods change structural and surface properties of fibers and thereby influence the mechanical bonding with the matrix. It can be divided into macroscopic modification, such as fiber fibrillation [11] and polymer rivet [12], and microscopic methods: etching [13], plasma treatment [14] or grit blasting with micro particles [15].

Plasma treatment has been widely used to improve the adhesion of coatings and paintings on polymer surfaces [16]. It increases surface energy, wettability and adhesion capability by three different mechanisms, without affecting bulk properties. Ablation consists on the removal of organic residues as well as surface layers at a molecular level; cross-linking, meaning the formation of covalent links by the interaction between two or more radicals and surface activation due to the generation of polar groups (−COOH, −OH), which increases surface energy [17]. Besides, previous works have proved the effectiveness of this treatment on glass surfaces [18]. For this reason, atmospheric pressure plasma torch (APPT from now on) was used to treat glass fibers and improve adhesion between them and polyethylene.

One of the most important novelties of this work is the application of the composite with improved adhesion between fibers and matrix. It is related to natural stone, which has been the most commonly used material in construction, ornamentation, funerary art and a variety of artistic objects through the ages. The natural brittleness of stone results on a need to reinforce it. Specifically, consolidants are commonly used with this aim [19,20]. Glass fiber with epoxy resin is also employed in order to strengthen the marble before sawing the blocks to get slabs [21]. Thermoset composite materials are now being used worldwide for building new structures as well as for in-service structures rehabilitation. They are used as load-bearing structural components for infrastructure applications, as a corrosion-resistant alternative to external steel reinforcements [22]. However, an important drawback is found with this type of composite materials. They cannot be reused or recycled due to its irreversible cure. The thermoplastic composite developed in this work, acting as reinforcement of marble in order to reduce its fragility can cope with this issue. On polymers, brittleness depends, among others, on their crystallinity: the higher the level of crystallinity, the lower the toughness of the polymer. It is known that fibers added to a thermoplastic polymer can play a dual role: on one hand, they can act as nucleating agent, promoting higher crystallinity and on the other hand, they can restrict mobility of chains, promoting lower crystallinity [23]. When the fibers are treated, the high density of nuclei could lead to a transcrystallization process, resulting on high or low crystallinity depending on the amount and the nature of the fibers [24,25].

In this work, the effect of plasma treatment on glass fiber is studied; on one hand, from the point of view of the adhesion with the matrix and, on the other hand, the effect of this change in adhesion on the properties of the polymer.

As previously said, marble is naturally brittle. This brittleness is determined by the energy that is capable to absorb during an impact. With the externally bonded reinforcement this energy could be increased, making the toughness of the marble higher with the consequent reduction of broken pieces. In consequence this study is focused on the improvement of the adhesion between glass fiber and LDPE by plasma treatment in order to use this composite material as externally bonded reinforcement of marble pieces. Adhesion is a key factor, due to the necessity of having a good transmission of the energy through the composite.

2Experimental part2.1Materials

The reinforcement was a glass fiber type E in form of bidirectional woven tape with 5cm width supplied by Vetrotex (Alcalá de Henares, Spain). Its main components are SiO2 (53–57%wt.) and Al2O3 (12–15%wt.). Usually, commercial glass fiber is presented with a sizing. To remove it, they were subjected to a desizing calcination at 500°C for 1h [26]. Calcinated fibers are used just to compare, since it is not recommended to use unsized fibers.

As matrix, a low density polyethylene (LDPE) DOW 780 E in pellets, supply by Dow Chemical (Madrid, Spain) was used.

Marble was supplied by Asociación Española del Mármol de Andalucía (AEMA). Its composition was obtained by X-Ray Fluorescence (Table 1).

Table 1.

Marble composition (%wt.).

CaO  MgO  SiO2  Al2O3  BaSO4  P2O5  SO3 
60.79±0.26  0.75±0.17  0.37±0.16  0.07±0.04  0.08±0.01  0.12±0.01  0.04±0.01 
2.2Glass fiber reinforced polyethylene2.2.1Plasma treatment

An APPT device from Plasma Treat GmbH (Steinhagen, Germany) was utilized to treat the as-received glass fibers. It presented a torch ending nozzle that rotated at 1900rpm, through which plasma was expelled through a circular hole onto the samples. The setup operated at a frequency of 17kHz and a high tension discharge of 20kV. The air plasma was generated at a working pressure of 2bar.

In order to determine the optimum parameters of APPT (speed and nozzle-sample distance), glycerol contact angle was measured for different combinations of these parameters. Plasma treatment changes the surface energy and, consequently, wettability of surface. At least 5 drops of 4μL of glycerol were measured for each sample.

2.2.2Surface characterization

Dataphysics OCA15 plus goniometer and SCA20 software (DataPhysics Instruments GmbH, Filderstandt, Germany) were used in order to determine the surface energy. Deionized water, glycerol and diiodomethane contact angle measurements were used.

Owens–Wendt–Rable–Kaelble method [27] enables the determination of both dispersive and polar components of surface energy.

Van Oss, Chaudhury, Good (vOCG) [28] method is also used to determine Lifshiftz-van der Waals (LW) and acid-base (AB) contributions of surface energy. For its part, acid–base comprises two components: electron donor or basic component (γ) and electron acceptor (γ+).

To study the topography Atomic Force Microscope and Scanning Electron Microscope were used. Nanoscope IV AFM (Digital Instruments, Santa Barbara, CA) with tapping mode in air at room temperature was used to take images. The probe used was NSG30 supplied by NT-MDT. With NanoScope Analysis Software, roughness of the surfaces were compared. Images of scanning electron microscopy were taken by TENEO-LoVac (Eindhoven, Nederlands).

2.2.3Chemical characterization

APPT and calcination effect on the chemical bonds were studied by means of attenuated total reflectance spectroscopy (ATR). The spectra were obtained using Tensor 27 Fourier Transform Infrared (FTIR) spectrometer from Bruker (BrukerOptikGmbH, Madrid, Spain), equipped with a DuraSample Diamond accessory formed by a 0.5mm diameter diamond embedded in a ZnSe crystal. The ratio signal-to-noise is better than 8000:1 (5.4×10−5 noise absorbance). Spectra were recorded with a resolution of 4cm−1 from 4000 to 400cm−1 by taking 32 scans.

2.2.4Adhesion tests

Peel tests were carried out following the ASTM D1876 – 08(2015)e1 Standard. It was modified, having the thickness (0.5mm) and width (50mm) of glass fiber fabric. Peel test samples were prepared by using hot plates press Fontijne Presses TPB 374 (Barendrecht, Netherland) with 140°C, 0.5MPa and 10min.

T-peel test was carried out with the universal testing machine Microtest DT005 FR (Madrid, Spain) with a load cell of 1kN. Samples consisted on two glass fiber fabrics with polyethylene between them (Fig. 1a).

Fig. 1.

Scheme of (a) T-peel, (b) Pull-off and (c) Pull-out tests.

(0.05MB).

For pull-off tests, aluminum dolls with a diameter of 20mm were used. This test provides the traction resistance of the joint. They were adhered to the glass fiber by melting polyethylene between them as is shown in Fig. 1b. Then, samples were tested with the universal testing machine, according to ASTM D4541 - 17 Standard. Resistance (σ) is calculated with Eq. (1):

where F is the force, A is the area and Rdoll is the doll radius.

Finally, pull-out tests (ASTM D7913 Standard) were carried out by embedding a glass fiber roving in a LDPE pellet (Fig. 1c). It was melting in the oven at 140°C and then it was tested to obtain the shear resistance of the fiber-matrix joint. The universal testing machine with a load cell of 50N is used.

At least 15 specimens were prepared. With the embedded fiber length (lef), the diameter of the fiber (df) and the maximum force (Fmax), interfacial shear strength (IFSS) τ can be calculated from Eq. (2)[29]:

2.2.5Composite manufacturing

Monolayer composite is manufactured by using hot plates press. Firstly, two layers of polyethylene were processed from melted pellets at 130°C, 0.22MPa and 10min. Glass fiber fabric was positioned between them and the composite was manufactured at 140°C, 0.22MPa and 10min. Surface energy of composite with untreated and treated fibers was calculated.

In order to study the influence of fibers and their treatment, melting point and crystallinity of LDPE in different conditions (without fibers, with untreated and with plasma treated fibers) were obtained by Differential Scanning Calorimetry. DSC 883 Mettler Toledo (Grafensee, Switzerland) equipment was used. Aluminum crucible with sample of 5–10mg were firstly heat from −40°C to 200°C at 10°C/min to remove the thermal history. Then it was cooled down to −40°C at 20°C/min and heated again in the same conditions. Results are analyzed in the second scan. Crystallinity (χc) is determined from Eq. (3):

where ΔHc is the enthalpy obtained by the melting peak area, ΔHmº the melting enthalpy of LDPE 100% crystalline (288J/g) and wf the weight fraction of fibers [30].

2.2.6Mechanical tests

Three points bending tests (ASTM D790 – 17 Standard) were carried out with the universal testing facility, with a load cell of 150N. The three points bending device was composed of a punch and two support steel rods. The distance between supports was 25mm. Flexural strength (σf) and deformation (εf) were determined from Eqs. (4) and (5).

F represents the applied force, L the distance between supports, b and h width and thickness of the sample, respectively and s the deflection produced.

LDPE samples and monolayer samples of untreated and treated glass fiber embedded on polyethylene matrix were tested. The flexural modulus was calculated from the slope of the elastic zone on the strain–stress curve. Besides, the stress needed to produce a 3.5% deformation on the sample was calculated. Displacement rate of the punch was set on 2mm/min.

2.3Thermoplastic composite reinforced marble2.3.1Adhesion test

Adhesion between reinforcement and marble has an important role to play in transmitting the waves formed during the impact.

In order to study adhesion between the composite and the marble, peel tests were carried out. Samples were prepared with the hot plates press (130°C, 0.2MPa, 10min), after the marble surface was cleaned with acetone. Universal testing machine with 1kN load cell equipped with a rail was used in order to have a perpendicular load throughout the test (Fig. 2).

Fig. 2.

Adhesion test.

(0.23MB).
2.3.2Impact tests

When a sample is subjected to an impact, compressive waves pass through the material. They are reflected in the opposite surface as tensile waves, which are the main cause of failure in ceramic materials, such as marble. The function of the reinforcement is to absorb this energy in order to reduce the intensity of the tensile reflection, leading to a less brittle material.

In order to determine absorbed energy of the reinforced and non-reinforced marble, two different tests were done.

Drop tower Neurtek (Éibar, Spain) was used to determine the maximum falling height of a body of 1kg on the material without breaking it. From this height, absorbed energy and fall speed were calculated.

On the other hand, Charpy pendulum test was carried out. Absorbed energy needed to break marble, with and without reinforcement, was measured with this test. Initial energy was fixed in 295J. Nestor pendulum for Charpy–Izod tests was used in this work.

3Results3.1Glass fiber reinforced polyethylene3.1.1Plasma treatment optimization

First, the parameters of the plasma are optimized: nozzle-sample distance and speed. Plasma treatments improves the wettability of a surface, hence it is possible to optimize the parameters by measuring the contact angle. In this case, it is done with glycerol. Nozzle-sample distance of 8mm and 5m/min gives the maximum wettability, as can be seen in Table 2.

Table 2.

Glycerol contact angle (º) on fibers with APPT.

    Speed (m/min)
    10 
Nozzle-sample distance (mm)4  53.8  58.4  68.7 
8  91.0  51.7  69.7 

Fibers without treatment show a contact angle of 104°. The improvement of wettability with plasma treatment is observed in Fig. 3. It is clearly observed a lower contact angle when the fiber is treated with APPT. The best condition combination is a nozzle-sample distance of 8mm and speed of 5m/min. With these parameters the lowest angle is obtained.

Fig. 3.

Glycerol drop on (a) untreated and (b) plasma treated fibers.

(0.06MB).
3.1.2Fiber surface3.1.2.1Scanning electron microscopy

In order to study the plasma and the calcination effect on fiber surfaces, micrographs with SEM were taken. In these images, cleaning effect of plasma (Fig. 4b) is observed. Furthermore, when fibers are calcinated (Fig. 4c), their surface becomes cleaner and softer due to the elimination of the organic sizing, which is irregular and inhomogeneous (Fig. 4a). However, there are some parts of sizing that remains after 1h of calcination.

Fig. 4.

SEM images of (a) as-received, (b) treated by APPT and (c) calcinated glass fiber.

(0.37MB).

It seems that plasma treatment does not remove sizing from the surface, but a more exhaustive analysis (by FTIR-ATR, for instance) is required in order to prove this.

3.1.2.2Atomic force microscopy

By atomic force microscopy (AFM), fiber surfaces can be studied in nanometric scale, in this way, it is possible to verify what has been appreciated by SEM. Height images taken with AFM are shown in Fig. 5. Second order “planefit” was done in order to remove the effect of the fiber curvature. At least six profiles, with a length of 3μm, were chosen on each sample in order to determine its roughness.

Fig. 5.

AFM micrographs of (a) as-received, (b) plasma treated and (c) calcinated glass fiber.

(0.26MB).

It is proved that the roughness of the surface changes when plasma treatment is applied. Surface becomes more homogeneous and the roughness decreases (Fig. 5a and b). This nanoroughness can lead to an improvement of adhesion when fibers are treated. It can be seen how the desizing process softens the surface, although remains of the sizing can still be observed (Fig. 5c).

When roughness values are compared, it can be concluded that plasma treatment, as well as the removal of the sizing, produces in both cases very flat surfaces. However, the calcinated surfaces have lower roughness than the one observed for the plasma treated fiber.

This was previously observed by other authors [27]. APPT produced etching on the sizing due to the flow of plasma. Reduction of roughness and smoothing effect of this treatment were observed on epoxy and polyester composites surfaces [31].

3.1.2.3Surface energy

Wettability is a key factor on adhesion between two elements. It can be studied by contact angle measurements and surface energy calculation. These contact angle values are presented in Table 3.

Table 3.

Contact angle (º) on as-received, treated and calcinated fibers.

  As-received  APPT  Calcinated 
Water  112.4±1.2  42.9±7.2 
Glycerol  104.5±2.1  51.7±4.2  32.9±1.8 
Diiodomethane  28.1±3.5  52.2±11  90.3±5.8 

An increase in wettability is produced when the treatment is applied, except for diiodomethane. This is due to the dispersive character of this liquid, while plasma affects the polar component of surface energy.

When sizing is removed and the glass surface is exposed, contact angles change completely. The wettability values observed on the sizing polymer as compared to the bare glass surface are very different, due to low surface energy of polymers, in contrast with high energy of ceramic and glass.

From these data, surface energy is calculated. OWRK and acid-base methods are used to calculate the most significant components. These components can be seen in Table 4.

Table 4.

Surface energy and its components of glass fiber as-received.

  As-received  APPT  Calcinated 
γTotal, OWRK  54.1±1.4  53.4±7.0  73.3±5.1 
γPolar, OWRK  5.4±0.4  27.9±5.0  62.1±4.8 
γBase, vOCG  0.1±0.2  41.5±11.6  68.1±2.1 

It is observed how plasma treatment increases the polar component of glass fiber due to the incorporation of polar groups (CO, C–O, etc.) on the surface, which presents dispersive nature when untreated. This increase in polarity was reported previously [31]. Furthermore, basic component rises with APPT, which leads to think that the introduced groups are those that present electron-donor character: amine, imine, ketone, ester, ether or alcohol functionalities [32].

On the other hand when the sizing is removed from the glass surface, the polar and basic character corresponds with glass [18]. This polar and electron donor character is related to the chemical structure of glass (SiO).

3.1.2.4Infrared spectroscopy

Spectra and the identification of the peaks are presented in Fig. 6 and Table 5, respectively. Peaks corresponding to the sizing (light rectangles in Fig. 6) appear around 2800–2900cm−1 (−CH2 and −CH3 bonds) [33] and at 1730cm−1 (carbonyl groups). This peak at 1730cm−1 corresponds to the polar component of the wax present in the sizing [36].

Fig. 6.

FTIR-ATR spectra of as-received, treated and calcinated fiber.

(0.17MB).
Table 5.

Asignation of peaks of untreated, APPT treated and calcinated fibers spectra.

ν (cm−1) UT  ν (cm−1) APPT  ν (cm−1) calcinated  Asignation 
700  700  700  Si–O [6] 
800–1200  800–1200  800–1200  SiOCH3, SiOC, SiOSi [30,34] 
1730  1730  –  CO [31] 
–  2100–2160  –  CO–CHN2[35] 
2850  2850  –  −CH2[30] 
2920  2920  –  −CH3[30] 

Untreated and treated fibers spectra show these peaks, which disappear when fibers are calcinated due to the removal of the sizing. It has been observed previously by other authors when fibers were heat-cleaned [37]. With these results, and together with SEM and AFM characterization, it can be concluded that sizing is almost completely removed from the surface.

When fibers are treated with APPT bands at 2800–2900cm−1 do not disappear, allowing to conclude that sizing is not removed with this treatment. Effect of plasma appears at 2100–2160cm−1 (dark rectangle in Fig. 6). These bands are due to diazoketone (−CO−CHN2) formed when introducing polar groups (nitrogen and oxygen) with atmospheric plasma. The introduction of these elements with the treatment was previously observed by X-ray photoelectron spectroscopy (XPS) [32,37]. Ketone groups explain the basic character of the surface energy when fibers are treated with atmospheric plasma. Plasma effect is also observed in the band at 1730cm−1, but this peak appears overlapped with the wax of the sizing. In spite of the double contribution in this peak (wax and plasma), the area of this peak is reduced when plasma is applied, so it seems that the treatment reduces the amount of wax present on the fiber surface.

Finally, groups due to fibers (SiOSi, SiO) appear between 800 and 1200cm−1 and 700cm−1, respectively. Overlapped with them, between 800 and 1200cm−1, the fiber-sizing bonds, and peaks of coupling agent silane, appear.

3.1.3Adhesion tests

Peel tests offer an advantage compared with other tests. They allow to evaluate quantitative and qualitatively the joint, as these results show the peel resistance and the type of failure.

Fig. 7 shows peel resistance when the fiber is both treated and untreated. An increase of more than 45% is produced when the joint with treated fiber is subjected to peel stress. This is due to the cleaning and elimination of low cohesion superficial layers when plasma is applied. By AFM it has also been observed a more homogeneous surface, which promotes adhesion (Fig. 5b).

Fig. 7.

(up) Peel strength and (down) failure of polyethylene with (a) as-received and (b) APPT treated fibers.

(0.15MB).

In Fig. 7 down it is shown the type of failure produced when fibers are untreated (Fig. 7a) and treated with APPT (Fig. 7b). Failure mode is cohesive with plasma treatment, being adhesive without treatment. It indicates the good adhesion between matrix and reinforcement when fibers are treated.

Pull-off tests were carried out by using an aluminum doll. With doll radius, pull-off resistance was calculated, as can be seen in Fig. 8a. Pull-off adhesion improves significantly, increasing the resistance eight times when fibers are treated. However, in all cases, failure is produced in the interface between the aluminum doll and the polyethylene, therefore suggesting that adhesion between polyethylene and fiber would present higher values.

Fig. 8.

(a) Pull-off and (b) pull-out adhesion.

(0.07MB).

Finally, pull-out tests reveal an increase in adhesion of 35% when treatment is applied. Deviations of these results are very important, but using t student it can be concluded that, with 90% confidence, adhesion with the treatment is higher than without it. Results are shown in Fig. 8b. With these tests, it is shown how plasma treatment affects the interface by improving the adhesion from chemical and mechanical anchoring.

3.1.4Composite manufacturing

Monolayer composite (GF-LDPE) is manufactured by using hot plates press. When surface energy of composite is calculated, it is observed that adding fibers, as well as treated fibers, results in drastic changes (Table 6).

Table 6.

Surface energy and its components.

  LDPE  LDPE+untreated GF  LDPE+APPT GF 
γTotal, OWRK  28.3±0.8  36.1±2.9  44.9±3.6 
γPolar, OWRK  3.8±0.4  23.5±1.6  3.1±3.2 
γBase, vOCG  5.6±1.7  42.4±11.7  0.0±0.0 

LDPE shows dispersive character, which corresponds to the non-polar molecules of the polymer attracted by London forces. These forces appear in non-polar polymers, such as elastomers and soft thermoplastic polymers [38].

When the fibers are added without treatment, the surface energy of the composite increases due to an increment in the polar component; base component also increases. However if the fibers are treated, polar component of the composite presents the same value that LDPE, besides they do not present any base component. However, total surface energy increases.

In the case of untreated fiber, with the application of heat, wax of the sizing migrates to the surface, being the responsible of the high polar component. When fibers are treated, the wax is removed from their surface and the polar component of the final composite is reduced.

Other changes on the polyethylene when fibers are untreated or treated can be seen by DSC analysis. Table 7 shows crystallinity of polyethylene and melting temperature for LDPE and composites. Untreated glass fibers act as nucleating agent whilst long fibers present a preferential direction, which facilitates crystals growth. APPT fibers reinforced polyethylene shows the lowest value of crystallinity, which is due to the transcrystallization process [24,25]. Treated fibers have an activated surface providing a large number of nucleation points. They block spherulites growth except in perpendicular direction where other fibers and spherulites are found, which blocks perpendicular growth [23].

Table 7.

Crystallinity and melting temperature for LDPE and composites.

  χc (%)±Tm (°C)±
LDPE  15  113 
LDPE+untreated GF  27  114 
LDPE+APPT GF  14  112 
3.1.5Flexural tests

Results of three points bending tests on GF-LDPE were presented in Fig. 9. All samples subjected to the bending test are more flexible than the total travel of the machine therefore exceeding its measuring limits.

Fig. 9.

Flexural modulus (a) and stress at 3.5% strain (b).

(0.09MB).

It can be seen how the presence of untreated fibers in the matrix does not present an impact in the flexural modulus (Fig. 9a), but the stress at 3.5% strain is higher (Fig. 9b). This can be explained because the interface is weak but it needs a certain force to be broken. Once broken, forces are supported only by the matrix and the modulus is given by polyethylene. Furthermore, when fibers are treated and the interface is improved, an increase is observed in both, modulus and stress. It can be seen that an interphase with good properties is crucial for improving the mechanical properties of the composite.

Previous studies with basalt fibers and PBT did not find any differences in flexural modulus, between untreated fibers and the ones that were treated with silane coupling agent [39]. However, in this work, a significant improvement on flexural modulus when fibers are treated is observed.

3.2Marble reinforced with thermoplastic composite3.2.1Marble-composite adhesion

Peel test allows to determine adhesion qualitatively, studying the kind of failure, and quantitatively. When the composite with untreated fibers is adhered to the marble, it is found that the joint does not present any resistance due to the presence of the wax on the composite surface. However, with treated fibers, the resistance of the joint is 0.64±0.11N/mm, and the failure can be seen in Fig. 10.

Fig. 10.

Mixed failure obtained in adhesion test.

(0.29MB).

The joint between marble and composite with treated fibers shows a mixed failure. It can be observed how a part of polyethylene is on the marble surface after the peel test. Adhesion is therefore adequate for the final purpose.

3.2.2Impact tests

From the maximum height in drop tower test, absorbed energy and fall speed are calculated, resulting in the values shown in Table 8. When the reinforcement is adhered to marble, absorbed energy increases by a factor of five, and fall speed doubles. This is due to the good interface between them, and the ability to absorb energy without the breakage of the flexible composite material.

Table 8.

Absorbed energy and fall speed in drop tower tests.

  Non-reinforced marble  Reinforced marble 
Energy (J)  0.59  2.45 
Fall speed (m/s)  1.08  2.21 

Absorbed energy when the sample breaks can be measured by Charpy test. Energy values with and without reinforcement are shown in Fig. 11. As in drop tower tests, the reinforcement of marble increases drastically the absorbed energy. Furthermore, in Charpy test, the breakage of the sample also changes when marble is reinforced, as can be seen in Fig. 12. The scheme, shown below, presents the efforts generated during the impact in both cases [40]. Marble, like every ceramics, has high compressive strength, but it presents low tensile strength. When the compressive waves travel across, they are reflected like tensile waves, which produces breakage by tensile strength when marble is not reinforced (Fig. 12 down). When the marble is reinforced, the intensity of the tensile waves is lower thanks to the good adhesion between the components of the composite, as well as with the marble piece.

Fig. 11.

Absorbed energy in Charpy test.

(0.06MB).
Fig. 12.

Breakage of marble without (a) and with reinforcement (b).

(0.19MB).
4Conclusions

  • APPT effect on glass fiber surface:

    • Cleaning and etching: these effects are observed by SEM and AFM, where a smoother and more homogeneous surface is found with APPT. Furthermore, by FTIR, a reduction in the amount of wax on the surface is found.

    • Activation: this effect is observed by surface energy measurements, DSC and FTIR. Polar and electron donor groups are introduced in the surface, which act as nucleation points resulting on transcrystallization process.

  • GF-LDPE adhesion: by peel, pull-out and pull-off tests an important improvement of the adhesion is observed with plasma treatment.

  • Composite flexural properties: the quality of the interface affects the flexural modulus and the strength of composites: the better the interface, the higher these parameters.

  • Reinforcement of marble: when marble is reinforced with the composite both, drop tower and Charpy tests, show an important increase in absorbed energy when the material is subjected to an impact. It means that the toughness of the marble increase drastically. It is important to highlight that this increase is achieved thanks to the good adhesion of every interface (matrix-fiber and composite-marble).

Conflicts of interest

The authors declare no conflicts of interest.

Funding

This work was supported by Interreg SUDOE project SOE1/P1/E0307.

References
[1]
R. Stewart.
Thermoplastic composites – recyclable and fast to process.
Reinf Plast, 55 (2011), pp. 22-28
[2]
R. Kumar, R. Singh, I.P.S. Ahuja, R. Penna, L. Feo.
Weldability of thermoplastic materials for friction stir welding—a state of art review and future applications.
Compos B: Eng, 137 (2018), pp. 1-15
[3]
S.S. Yao, F.L. Jin, K.Y. Rhee, D. Hui, S.J. Park.
Recent advances in carbon-fiber-reinforced thermoplastic composites: a review.
Compos B: Eng, 142 (2018), pp. 241-250
[4]
R. Sepe, F. Bollino, Boccarusso, F. Caputo.
Influence of chemical treatments on mechanical properties of hemp fiber reinforced composites.
Compos B: Eng, 133 (2018), pp. 210-217
[5]
B. Enciso, J. Abenojar, M.A. Martinez.
Influence of plasma treatment on the adhesion between a polymeric matrix and natural fibres.
Cellulose, 24 (2017), pp. 1791-1801
[6]
M. Iorio, M.L. Santarelli, G. González-Gaitano, J. González-Benito.
Surface modification and characterization of basalt fibers as potential reinforcement of concretes.
Appl Surf Sci, 427 (2018), pp. 1248-1256
[7]
M. Jing, J. Che, S. Xu, X. Liu, Q. Fu.
The effect of surface modification of glass fiber on the performance of poly(lactic acid) composites: graphene oxide vs. silane coupling agents.
Appl Surf Sci, 435 (2018), pp. 1046-1056
[8]
L.A. Savas, U. Tayfun, M. Dogan.
The use of polyethylene copolymers as compatibilizers in carbon fiber reinforced high density polyethylene composites.
Compos B: Eng, 99 (2016), pp. 188-195
[9]
X. Liu, C. Yang, Y. Lu.
Contrastive study of anodic oxidation on carbon fibers and graphite fibers.
Appl Surf Sci, 258 (2012), pp. 4268-4275
[10]
X. Zhao, D. Xiong, X. Wu.
Effects of surface oxidation treatment of carbon fibers on biotribological properties of CF/PEEK materials.
J Bionic Eng, 14 (2017), pp. 640-647
[11]
V.C. Li, Y.W. Chan, H.C. Wu.
Interface strengthening mechanisms in polymeric fiber reinforced cementitious composites. Special session honoring Prof. H. Krenchel. BMC4 Proceedings, (1994), pp. 7-16
[12]
M. Grujicic, V. Sellappan, M.A. Omar, N. Seyr, A. Obieglo, M. Erdmann, et al.
An overview of the polymer-to-metal direct-adhesion hybrid technologies for load-bearing automotive components.
J Mater Process Technol, 197 (2008), pp. 363-373
[13]
D. Long, H. Guo, J. Cui, X. Chen, M. Lu.
Rapid etching of carbon fiber induced by noble metal nanoparticles.
Mater Lett, 197 (2017), pp. 45-47
[14]
Z. Liu, P. Chen, D. Han, F. Lu, Q. Yu, Z. Ding.
Atmospheric air plasma treated PBO fibers: wettability, adhesion and aging behaviors.
Vacuum, 92 (2013), pp. 13-19
[15]
K. Schulze, J. Hausmann, B. Wielage.
The stability of different titanium-PEEK interfaces against water.
Procedia Mater Sci, 2 (2013), pp. 92-102
[16]
M.A. Martinez, J. Abenojar, S. Lopez de Armentia.
Environmentally friendly plasma activation of acrylonitrile–butadiene–styrene and polydimethylsiloxane surfaces to improve paint adhesion.
Coatings, 8 (2018), pp. 428
[17]
T. Scalici, V. Fiore, A. Valenza.
Effect of plasma treatment on the properties of Arundo Donax L. Leaf fibres and its bio-based epoxy composites: a preliminary study.
Compos B: Eng, 94 (2016), pp. 167-175
[18]
J. Abenojar, M.A. Martínez, N. Encinas, F. Velasco.
Modification of glass surfaces adhesion properties by atmospheric pressure plasma torch.
Int J Adhes Adhes, 44 (2013), pp. 1-8
[19]
E. Sassoni, G. Graziani, G. Ridolfi, M.C. Bignozzi, E. Franzoni.
Thermal behavior of Carrara marble after consolidation by ammonium phosphate, ammonium oxalate and ethyl silicate.
Mater Des, 120 (2017), pp. 345-353
[20]
E. Tesser, L. Lazzarini, S. Bracci.
Investigation on the chemical structure and ageing transformations of the cycloapliphatic epoxy resin EP2101 used as stone consolidant.
J Cult Herit, 31 (2018), pp. 72-82
[21]
Processing of Marble Slabs and Equipment Requirements, Al Qamer Company for Energy and Infrastructure. [cited 2019 May 20]. Available from: http://www.alqamer.com/index.php/2016-02-01-19-25-41/processing-of-marble.
[22]
C. Bellini, W. Polini, L. Sorrentino, S. Turchetta.
Mechanical performances increasing of natural stone by GFRP sandwich structures.
Procedia Struct Integr, 9 (2018), pp. 179-185
[23]
J.P. Mofokeng, A.S. Luyt, T. Tábi, J. Kovács.
Comparison of injection moulded, natural fibre-reinforced composites with PP and PLA as matrices.
J Thermoplast Compos, 25 (2011), pp. 927-948
[24]
L. Ludueña, A. Vázquez, V. Alvarez.
Effect of lignocellulosic filler type and content on the behavior of polycaprolactone based eco-composites for packaging applications.
Carbohydr Polym, 87 (2012), pp. 411-421
[25]
R.J. Young, M. Heppenstall-Butler.
Polymer transcrystallinity in composites. Encyclopedia of materials: science and technology.
2nd ed., (2001), pp. 7535-7540
[26]
D. Olmos, K. Bagdi, J. Mózco, B. Pukánszky, J. González-Benito.
Morphology and interphase formation in epoxy/PMMA/glass fiber composites: effect of molecular weight of the PMMA.
J Colloid Interface Sci, 360 (2011), pp. 289-299
[27]
N. Encinas, M. Pantoja, J. Abenojar, M.A. Martínez.
Control of wettability of polymers by surface roughness modification.
J Adhes Sci Technol, 24 (2010), pp. 1869-1883
[28]
C.J. Van Oss, R.J. Good, M.K. Chaudhury.
Additive and nonadditive surface tension components and the interpretation of contact angles.
Langmuir, 4 (1988), pp. 884-891
[29]
A. Godara, L. Gorbatikh, G. Kalinka, A. Warrier, O. Rochez, L. Mezzo, et al.
Interfacial shear strength of a glass fiber/epoxy bonding in composites modified with carbon nanotubes.
Compos Sci Technol, 70 (2010), pp. 1346-1352
[30]
N.L. Batista, M.C. Rezende, E.C. Botelho.
Effect of crystallinity on CF/PPS performance under weather exposure: moisture, salt fog and UV radiation.
Polym Degrad Stab, 153 (2018), pp. 255-261
[31]
N. Encinas, M. Lavat-Gil, R.G. Dillingham, J. Abenojar, M.A. Martínez.
Cold plasma effect on short glass fiber reinforced composites adhesion properties.
Int J Adhes Adhes, 48 (2014), pp. 85-91
[32]
N. Encinas, J. Abenojar, M.A. Martínez.
Development of improved polypropylene adhesive bonding by abrasion and atmospheric plasma surface modifications.
Int J Adhes Adhes, 33 (2012), pp. 1-6
[33]
M. Pantoja, J. Abenojar, M.A. Martínez.
Influence of the type of solvent on the development of superhydrophobicity from silane-based solution containing nanoparticles.
Appl Surf Sci, 397 (2017), pp. 87-94
[34]
M. Pantoja, J. Abenojar, M.A. Martínez, F. Velasco.
Silane pretreatment of electrogalvanized steels: effect on adhesive properties.
Int J Adhes Adhes, 65 (2016), pp. 54-62
[35]
T. Groβ, T. Hermann, B. Shi, A. Jäger, P. Chiu, P. Metz.
Further studies on sultones derived from carbine cyclization cycloaddition cascades.
Tetrahedron, 71 (2015), pp. 5925-5931
[36]
D. Muscat, M.J. Tobin, Q. Guo, B. Adhikari.
Understanding the distribution of natural wax in starch–wax films using synchrotron-based FTIR (S-FTIR).
Carbohydr Polym, 102 (2014), pp. 125-135
[37]
J. Abenojar, M.A. Martínez, F. Velasco, M.A. Rodríguez-Pérez.
Atmospheric plasma torch treatment of polyethylene/boron composites: effect on thermal stability.
Surf Coat Technol, 239 (2014), pp. 70-77
[38]
R.B. Seymour, C.E. Carraher Jr.
Introducción a la química de los polímeros, 3rd. ed.
Reverté, (2002),
[39]
C. Arslan, M. Dogan.
The effect of silane coupling agents on the mechanical properties of basalt fiber reinforced poly(butylene terephtalate) composites.
Compos B: Eng, 146 (2018), pp. 145-154
[40]
D. Mohotti, T. Ngo, P. Mendis, S.N. Raman.
Polyurea coated composite aluminum plates subjected to high velocity projectile impact.
Mater Des, 52 (2013), pp. 1-16
Copyright © 2019. The Authors
Journal of Materials Research and Technology

Subscribe to our newsletter

Article options
Tools
Cookies policy
To improve our services and products, we use cookies (own or third parties authorized) to show advertising related to client preferences through the analyses of navigation customer behavior. Continuing navigation will be considered as acceptance of this use. You can change the settings or obtain more information by clicking here.