Investigation of tribological, physicomechanical, and morphological properties of resin-based friction materials reinforced with Agave americana waste

In recent years, natural fibers and their composites have attracted the attention of researchers due to environmental awareness and sustainable development. It is crucial to identify new natural fibers as potential reinforcement in polymer composites. This study was aimed to investigate the potential use of Agave americana fibers as a reinforcing component in resin-based friction materials. The tribological, physicomechanical, and morphological characteristics of materials containing different A. americana fiber contents were systematically evaluated. Experimental results indicated that fiber addition effectively improved the fade resistance, recovery behavior, and wear resistance of these materials. From the perspective of overall performance, a friction composite containing 5-wt% fibers possessed the optimal friction stability and wear resistance, exhibiting a fade rate of 13.6%, recovery rate of 97.5%, and sum wear rate of 2.340 × 10–7 cm3·N−1·m−1. Furthermore, sample worn surface morphologies were examined by scanning electron microscope, which revealed that appropriate fiber inclusion helped in the formation of secondary contact plateaus on friction surfaces. In addition, this fiber content significantly reduced abrasive and adhesive wear, which were conducive to good tribological behaviors of friction materials. This research provided a promising method for environment-friendly applications of A. americana waste.


Introduction
Friction materials are widely used in automotive transport systems as brake pads/liners to control the deceleration and immobilization of vehicles quickly and reliably [1]. Friction materials are multi-ingredient composites composed of binders (phenolic resin, nitrile butadiene rubber powder, crumb rubber, calcium oxide, etc), fibrous reinforcements (aramid fibers, Prosopis juliflora fibers, mullite, basalt fibers, mineral fibers, etc), friction modifiers (lubricants and abrasives, like graphite, antimony sulfide, iron sulfide, red mud, iron powder, etc), and fillers (inert and functional, like barium sulfate, vermiculite powder, granite powder, friction dust, china clay, etc) [2][3][4][5]. Generally, more than 10-20 raw ingredients are used in friction materials to achieve a certain set of performance requirements, such as good friction stability, low fade, moderately high recovery, low noise generation, no vibration, and excellent wear resistance, as well as low wear of friction counterparts under different operating conditions [6][7][8]. Among these various ingredients, fibrous reinforcements are often considered essential components for maintaining thermal stability, mechanical strength, and toughness, along with altering the tribological behaviors of friction materials [9]. Asbestos fibers were once used as reinforcement in friction material industry earlier because of their availability and excellent thermal resilience. The usage of asbestos fibers has been banned since they were proved as carcinogenic materials [10]. Thereafter, the humanmade synthetic fibers such as glass, ceramic, and aramid fibers, etc, metallic fibers such as steel, copper, and brass Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. fibers, etc, as well as combinations thereof have been utilized in friction materials as substitutes for traditional asbestos fibers [11]. However, these synthetic fibers showed adverse effects on environment during dispose due to their non-biodegradable nature. The heavy metallic fibers also have a disadvantage that their wear debris formed during braking process can cause immense danger to the aquatic life [12]. Considering the above problems, the need for green friction materials using biodegradable and renewable fibers is gaining wider importance.
With improved environmental protection and sustainable development consciousness, natural fibers, especially plant fibers, have received intense attention as potential reinforcements for composite materials [13,14]. Plant fibers, such as from Leucas Aspera, Prosopis juliflora bark, Areva Javanica, hemp, bamboo, rattan, flax, sisal, and corn stalk, provide several significant advantages over synthetic fibers, including wide availability, biodegradability, renewability, light weight, relatively low cost, no or minor irritation to human eyes, skin, and respiratory system, high-specific mechanical properties, satisfactory acoustic insulation characteristics, and environment-friendly features [15][16][17][18][19][20][21]. Thus, many studies have been performed concerning the effects of the type, length, aspect ratio, relative content, orientation, and surface modification of plant fibers on the tribological, mechanical, thermal, and physical behaviors of friction composite materials.
Xu et al [22] have developed brake friction materials reinforced with sisal fibers and examined the effects of fiber content on tribological properties. They found that composite materials with a sisal fiber/resin ratio at 4/3 (by wt) exhibited optimal friction and wear behaviors. The friction coefficient of sisal fiber-brake materials shows relatively low fluctuation rates under different temperature conditions, compared to asbestos and steel/ mineral fiber reinforced materials. Nirmal et al [23] have fabricated bamboo fiber-reinforced epoxy composites (BMBFRE) and evaluated the influence of fiber orientation (random, parallel, and antiparallel) on adhesive wear and frictional performance under dry contact conditions. They found that BMBFRE composites in antiparallel orientation exhibited superior wear resistance and friction behaviors at different sliding velocities compared to random and parallel orientations. The specific wear rate, friction coefficient, and interface temperature of antiparallel oriented BMBFRE composites were improved by ∼60, 46.4, and 34.7%, respectively, in comparison to neat epoxy composites. El-Tayeb [24] have reported the preparation and tribological assessment of polyester composites based on sugarcane fibers under dry sliding conditions. They concluded that sugarcane fibers have positive effects on enhancing composite friction coefficient and wear resistance properties, yielding them a promising reinforcement material for tribological applications. Similarly, the naturally derived filler materials such as crab, periwinkle, scallop, palm kernel, and cashew nut shells were investigated and evaluated to develop the environment-friendly friction composite materials [10,12,25,26].
Despite the above advantages of natural fibers, the hydrophilic nature and poor interfacial bonding between fibers and matrix limits the usage of natural fibers in composite materials. In order to overcome these issues, various surface modifications known as chemical treatment on natural fibers were proposed by researchers such as alkaline treatment, benzoylation treatment, acetylation treatment, and silanization treatment, etc [27][28][29]. Among these chemical treatments, alkaline treatment has proved to be an efficient and cost-effective method, which can remove the hydrophilic constituents in fibers and improve the compatibility of reinforcement fibers with the hydrophobic matrix, leading to the enhancement of composite properties [29]. Hence, alkaline treatment was chosen as the fiber modification method in the current study.
Agave americana, belonging to the Agavaceae family, is native to Central America and Mexico, and has been successfully introduced and cultivated in Southwest and South China [30]. A. americana is widely known for its application in pulque (an alcoholic beverage) production and the resulting discarded leaves estimated to account for ∼46% of the harvested plant, representing a large quantity of underutilized plant fiber resources [31]. These fibers, as lignocellulosic fibers, are mainly composed of cellulose, lignin, and hemicellulose and have been used in several applications, including in ropes, twines, yarns, textiles, and handicrafts [32,33]. However, to the best of our knowledge, few studies have been conducted regarding the use of these fibers as reinforcing fibers in composite materials, particularly in brake friction composite materials, in spite of the fact that these fibers have interesting physical and mechanical properties, including high strength, low density, high extensibility, and high rupture energy [30,34].
Therefore, this study focused on the preparation and performance evaluation of A. americana fiberreinforced resin-based friction composite materials. For this purpose, five friction materials with varying fiber content were fabricated using a hot-press molding method. Then, the obtained friction materials were characterized in terms of physical, mechanical, and tribological properties. In addition, worn surface morphologies were examined and analyzed to reveal wear mechanisms in these materials. These results will not only help expand the potential application of A. americana waste, but also provide a basis for the design and development of new ecofriendly brake friction materials.

Experimental details 2.1. Fiber preparation
A. americana leaves were obtained from Kunming (Yunnan Province, China). The leaves were cleansed with distilled water and the leaf margin and tip thorns removed. After air-drying, the leaves were ground to pieces and screened with a 40-60 mesh (0.425-0.250 mm) sieve. The real images of A. americana leaves and fibers are shown in figure 1. Subsequently, the prepared fibers were subjected to surface treatment. They were soaked in 6-wt% aqueous NaOH for 60 min, followed by 2-wt% aqueous H 2 SO 4 for 20 min, repeatedly rinsed with distilled water, and finally oven-dried at 70°C to constant mass. The morphologies of raw and treated fibers are shown in figure 2.

Sample preparation
Five friction composite materials were prepared in this study and their detailed compositions are summarized in table 1. Among these raw materials, A. americana fibers and compound mineral fibers were used as reinforcing ingredients. Phenolic resin was used as a binder. Graphite, petroleum coke, antimony sulfide, and zinc stearate were used as lubricants. Alumina and porous iron powder were used as abrasives. Vermiculite powder, friction powder, and barium sulfate were used as particulate fillers. These composite materials were designated FMS-0, FMS-2.5, FMS-5, FMS-7.5, and FMS-10, depending on their A. americana fiber content (wt%). The preparation of friction composite samples was performed by mixing, hot-pressing, and post-curing, and the specific procedure conditions are presented in table 2. The mixing sequence of ingredients involved in this study is given in table 3. The resulting composite samples were finally machined into preset dimensions for physicomechanical and tribological tests ( figure 3).

Physicomechanical characterization
The physicomechanical performances of the prepared composite samples were characterized in terms of density, hardness, and impact strength. Density measurements were carried out based on Archimedes drainage

Tribological characterization
The tribological behaviors of composite samples were characterized in terms of friction coefficient, fade and recovery, and wear rates. Tribological tests were performed using a JF150D-II constant-speed friction tester (Wanda Machinery, Changchun, China, figure 4). The friction tester mainly consisted of control, loading, heating, cooling water, friction testing, and power systems. The samples were press-fitted onto the surface of counterpart disk using a pressurizing device, and the corresponding applied load was adjusted by the loading system. The counterpart disk was driven to rotate by a motor, and the frictional force between the samples and counterpart disk was detected by a tension-compression sensor. The temperature during the test was monitored using a thermocouple sensor and was controlled and regulated by the heating and cooling water systems. A whole test was composed of two parts: fade and recovery procedures. Detailed test procedures are described in table 4, based on Chinese National Standard GB/T 5763-2008. The friction coefficient (μ) was recorded automatically and wear rate (W) was calculated according to the following formula [35]: where A (mm 2 ) is the area of the composite sample, Δh (mm) the thickness change of the sample, R (mm) the horizontal distance between the disk and sample centers (here, R=150 mm), N the disk rotational number, and f (N) the average friction force.  The friction coefficients decrease temporarily at elevated temperatures and should be regained at lower temperatures, which are referred to as fade and recovery, respectively. These characteristics are essential for performance evaluation of friction materials. The fade (F) and recovery (R) rates were calculated according to the following formulas respectively [36]: where μ F100°C and μ F350°C are the sample μ at 100 and 350°C during the fade procedure, respectively; μ R100°C the sample μ at 100°C during the recovery procedure.

Worn surface morphology characterization
After completion of tribological tests, worn surface morphologies of these materials were characterized by means of an EVO18 scanning electron microscope (SEM; Carl Zeiss AG, Oberkochen, Germany). Prior to SEM observation, samples were gold sputtered to enhance surface conductivity.

Physicomechanical properties
The experimental evaluation of the physicomechanical properties of these friction material samples were performed in triplicate to minimize error, and the corresponding test results are given in figures 5(a)-(c). Sample densities exhibited a downward trend with increased fiber content, with FMS-0 samples exhibiting the highest density and FMS-10 the lowest (2.34 and 2.10 g·cm −3 , respectively; figure 5(a)). It can be attributed to the factor that the density of A. americana fibers was relatively low compared with other ingredients, and when the total mass of the friction composite samples remained unchanged, adding A. americana fibers reduced the overall density of the polymer composites. This is in accordance with the previous research of Nishino et al [37] for their study on the kenaf fibers reinforced composites. Hardness refers to the resistance to indentation or penetration. In the current study, hardness values trended similarly to that of density, with hardness maximized in FMS-0 and minimized in FMS-10 (99.4 and 92.3 HRR, respectively; figure 5(b)). The main reason for this change was that the total mass fraction of each component in the friction material formula was 100%, and as the content of A. americana fibers increased, the content of hard particle ingredients such as alumina decreased proportionally, leading to a reduction in the overall hardness of the polymer composite system. In general, brake friction materials with medium hardness can provide better tribological properties, and too high hardness will easily lead to braking noise and stiff pedal feel [38,39].
Impact strength, as a measure of impact resistance of these samples, was used to evaluate the toughness and brittleness of the polymer composites to a limited extent. In this study, there was no specific trend for impact strength with fiber inclusion in these samples ( figure 5(c)). Among all tested composites, sample FMS-7.5 (0.491 J·cm −2 ) exhibited the highest impact strength, followed by sample FMS-5 (0.478 J·cm −2 ), which indicated that appropriate addition of A. americana fibers could improve the impact resistance performances of friction composites. Similar behaviors were seen in the previous findings of Ma et al [40] on their study of the corn stalk fibers and cow dung fibers reinforced friction composites. Usually, the reinforcement fibers have a significant effect on determining the impact strength of the polymer composites by preventing the crack propagation and acting as the load transfer medium to the composite matrix [4]. Composite materials with appropriate content of A. americana fibers, such as 7.5-wt% and 5-wt%, could provide better fiber-matrix interfacial adhesion conditions, thereby improving their impact resistance properties to a certain degree [40,41].

Tribological performances 3.2.1. Friction performance
The friction behavior of friction materials is directly related to the security and stability of automotive braking [42]. Friction testing was conducted to evaluate the effect of fiber content on changes in friction coefficient. The friction coefficients of FMS-0, FMS-2.5, FMS-5, FMS-7.5, and FMS-10 as a function of temperature during the fade and recovery processes showed that sample coefficients increased as the temperature increased from 100 to 150°C and then decreased as the temperature rose to 350°C ( figure 6(a)). The increased friction coefficient might have been due to the glass transition of the binder resin in the composites [43]. The subsequent high temperature decreased friction coefficient was ascribed to the heat fade effect caused by thermal decomposition of organic components, including binder resin, compound mineral fibers, and organic fibers, at elevated temperatures [36,44]. It was interesting to note here that the friction coefficients of the tested samples were in the range of 0.38-0.51, which was in conformity with the Chinese national standard.
During the recovery process, the friction coefficients of all samples initially increased by decreased temperature from 300 to 200°C and then decreased with further temperature decrease to 100°C ( figure 6(b)). The recovery performance of friction materials was found to be affected by both wear debris alteration and surface layer morphology [45]. At high test temperatures, high friction coefficients were mainly attributed to wear debris formation, leading to the counterpart disk scraping. However, lower friction coefficients at lower test temperatures were primarily correlated with rheological changes between wear debris and surface layer morphology [6]. In addition, it was worth pointing out that fluctuations in friction coefficients were in a relatively stable range (0.40-0.48), which was beneficial to braking stability of friction materials [42].
The friction coefficient changes of these samples were further evaluated by determining the fade and recovery rates of composite samples. The order of fade rate was clearly FMS-0>FMS-10>FMS-7.5>FMS- 2.5>FMS-5 and the recovery rate order FMS-5>FMS-2.5>FMS-7.5>FMS-10>FMS-0 (figure 7). Sample FMS-5 presented the lowest fade rate and highest recovery rate (13.6 and 97.5%, respectively), demonstrating excellent fade resistance and recovery properties. However, sample FMS-0 had the worst performance among all samples, with a fade rate of 22.4% and recovery rate of 85.6%. These results suggested that the inclusion of these fibers improved the friction behaviors of resin-based friction materials and, in particular, sample FMS-5 showed optimal friction stability during the entire testing series.

Wear performance
Wear resistance, an important parameter for determining the service life of friction composites, has been found to be primarily dependent on the test temperature as well as material composition [41]. The observed variation in wear rate with test temperature showed that sample wear rates experienced a significant increasing trend with increased test temperature ( figure 8(a)). This might have been due to the fact that thermal deformation and decomposition of the resin binder could cause weakening of interfacial interactions between the resin matrix and ingredients. Thus, the ingredients became loose and sometimes detached from the matrix, consequently resulting in increased wear rate [46,47]. Similar trends have been reported by Lee et al [48] and Ji et al [44]. Notably, the wear rate of sample FMS-0 was clearly higher than other samples, especially at elevated temperatures. In addition, the sum wear rates of these samples were in the order FMS-0>FMS-10>FMS-2.5>FMS-7.5>FMS-5 ( figure 8(b)). Sample FMS-5 exhibited the lowest sum wear rate (2.340×10 -7 cm 3 ·N −1 ·m −1 ) followed by FMS-7.5, while FMS-0 achieved the highest (2.705×10 -7 cm 3 ·N −1 ·m −1 ). These results suggested that fiber incorporation effectively enhanced the wear resistance of these composites, with 5-wt% fiber contents appearing to be the optimal proportion from the wear performance point of view. To  understand the above observations, the wear mechanisms involved needed to be investigated, the details of which were presented in the section below.

Worn surface analysis
The tribological properties of friction materials are closely associated with their corresponding worn surfaces [49]. Here, worn surface characterizations were carried out by SEM observation to understand the effects of A. americana fibers on wear mechanisms and tribological behaviors. SEM micrographs of worn surfaces of samples showed that the worn surface of sample FMS-0 exhibited the roughest surface topography and possessed the most wear debris, scratches, and grooves, as well as spalling pits, which corresponded to the highest wear rate of these composites ( figure 9(a)). The scratches and grooves have been reported to be surface damage phenomena associated with hard particles and debris [50]. Generally, hard particles from resin wear debris behaved as thirdbody abrasives, which nicked and destroyed the worn surface, producing typical abrasive wear characteristics. In addition, cold soldering joints were easily produced between the matrix surface and counterpart disk, which then detached from the surface when subjected to shear forces, thus leading to the formation of spalling pits, indicating adhesive wear behavior [51]. The wear mechanisms for sample FMS-0 were thought to be predominantly abrasive and adhesive wear.
Worn surface morphologies of samples reinforced with A. americana fibers, including samples FMS-2.5, FMS-5, FMS-7.5, and FMS-10, clearly showed relatively smoother worn surfaces, compared to FMS-0 (figures 9(b)-(e)). Specifically, FMS-2.5 surfaces exhibited of plenty of wear debris, microcracks, scratches, and grooves, together with broken fibers, which could have accounted for its high wear rate. The presence of microcracks was thought to be due to an unstable pressure and temperature field generated on the friction surface, as well as discrepancies in the thermal expansion coefficients of different zones, leading to typical fatigue wear characteristics [50]. For sample FMS-5, a high amount of secondary plateaus, some fine wear debris, and a few slight scratches were found on the worn surface, with polished fibers showing only a little breakage, which indicated a preferable interface adhesion between the fibers and composite matrix. The secondary contact plateaus formed on friction surfaces were believed to contribute to enhanced friction stability and wear resistance performances of this sample type [40]. The presence of secondary contact plateaus was mainly due to the accumulation and compaction of wear debris under the combined action of shearing forces, normal pressure, and friction heat [52,53]. The present observations might have correlated with the optimal tribological behaviors observed in sample FMS-5 and, in particular, with its lowest wear rate. Worn surfaces of sample FMS-7.5 were covered with some particles and debris pieces, shallow parallel scratches, bare fibers, fiber-shedding pits, and a few secondary contact plateaus, which agreed with its slightly reduced wear resistance with respect to sample FMS-5. For FMS-10, numerous wear particles and debris, obvious scratches and grooves, microcracks, and fiber-shedding pits were observed on worn surfaces, which were indicative of aggravated wear of the friction surface. The reason for this observation was that too many fibers in the friction material negatively affected the fiber-matrix interface bonding strength, which then easily allowed fiber pullout and shedding under applied frictional forces, resulting in increased wear rate [35].

Conclusions
This study evaluated the possibility of using A. americana fibers as reinforcement fibers in friction composite materials. The effects of fiber content on the tribological, physicomechanical, and morphological characteristics of prepared fiber-reinforced friction materials were examined. The main conclusions drawn were: (1) As A. americana fiber content increased, the density and hardness of the obtained composite materials exhibited a downward trend, while impact strength showed no specific changes and sample FMS-7.5 presented the maximum impact strength, at 0.491 J·cm −2 , followed by sample FMS-5, at 0.478 J·cm −2 .
(2) The incorporation of A. americana fibers apparently improved the fade resistance and recovery properties of the composite samples. Remarkably, the best friction stability was obtained in sample FMS-5, with a fade rate of 13.6% and recovery rate of 97.5%.
(3) A. americana fibers were also effective for enhancing the wear resistance of resin-based friction materials.
(4) The SEM results revealed that appropriate fiber addition (specifically 5-wt%) not only facilitated secondary contact plateau formation on friction surfaces, but also significantly reduced abrasive and adhesive wear, which explained the improved tribological performances of these samples.
Overall, the obtained results of the present study provided evidence that adding A. americana fibers appropriately improved the tribological, physicomechanical, and morphological properties of the polymer composites. Thus A. americana fibers can be effectively used as reinforcing components in the development of resin-based brake friction composite materials, which expanded the environment-friendly application of A. americana waste.