Abstract
An automotive friction brake pad is a complex system consisting of several components with unique and balanced properties related to operation conditions. There are efforts to develop brake pads with longer lifetime and better friction performance and wear properties. Those properties are related to composition of the pads, and therefore, new materials are being evolved. Tuning the friction and wear properties can be achieved with the selection of a functional filler and optimizing its amount in a formulation of friction brake pad. Laboratory-developed and laboratory-prepared nanocomposite material kaolin/TiO2 (KATI) has been introduced to formulation of the commercially available automotive low-steel brake pad. Kaolin was utilized as a matrix for anchoring TiO2 nanoparticles. New unused pads and pads after AK master, a standard dynamometer testing procedure of friction performance, were investigated using light and scanning electron microscopy providing information on the structure and its changes after the friction processes. Moreover, MTK wear test was used to compare wear rate of the newly developed pad with the reference low-steel pad. Improved durability of the brake pad formulation has been observed together with sufficient friction performance. Microscopic analysis shown homogenous distribution of the KATI nanocomposite in the friction layer. From the obtained results, it can be assumed that the new formulation is promising regarding to the life cycle of the pads and reduction of wear rate and thus potential production of wear particulate emissions.
1. Introduction
Typically, a brake pad is a multicomponent composite material characteristically consisting of more than 10 constituents. Phenolic resin usually serves as a matrix, and several forms of metals, ceramics, minerals, polymers, and carbonaceous ingredients are present in a typical brake pad. Due to significantly different material properties of single components, the resulting friction layer established on the surface of friction composites is strongly heterogeneous and also its surface is rough with many asperities across it [1, 2]. Friction materials have to be durable, mechanically and thermally stable, provide a consistent friction force, be tribologically compatible with the other part of friction pair, have balanced vibration and noise, be environmentally acceptable, and last but not the least have to be cost-effective in design, manufacturing, and usage [3].
Significant phase transformations of brake pad material occur during friction processes. These transformations lead to origination of newly formed friction layer and have a great effect on the friction efficiency of brake pads as well as an effect on wear debris formation and its properties. While braking, both parts of braking couple in the disc brake system (i.e., pads and rotating disc) are worn and thus wear debris is generated. It is well known that the wear debris may have a negative impact on the environment and human health [4]. Heat created during friction is dissipated through exchange mainly with the environment and rise tremendously in the contact regions.
Nowadays, environmental issues are being discussed due to the potential influence of wear products on the air quality in urban areas. It has been proved that the nonexhaust emission comes also from the wear of the pads and rotors [5]. Therefore, there are enormous efforts to develop new brake pad formulations with lower particulate emissions and stable friction performance. In connection with the development of new formulations for automotive brake pads, new raw materials are being investigated. The influence of nanopotassium titanate filler on the performance of nonasbestos organic brake pads has been already studied. It has been shown that nanopotassium titanate has improved friction performance and wear resistance compared to bulk potassium titanate [6]. Moreover, potential new abrasives in nanometer size have been studied. It has been revealed that all the friction and wear parameters were significantly and beneficially influenced due to the nanosize of the tested abrasives. Nanosilica increased wear resistance as well as nano-SiC and alumina [7]. Recent work [8] has also shown the significant influence of the particle size of SiO2 on the transformation of friction mechanisms.
Therefore, nanostructured composite material kaolin/TiO2 could be one of the suitable candidates and thus the aim of the study was the introduction of this nanomaterial into the brake pad formulation together with subsequent detailed characterization of friction surfaces of tested brake pads and their chemical and phase composition as well as evaluation of their changes during the friction process.
2. Materials and Methods
2.1. Brake Pad Samples
Federal-Mogul Motorparts (Germany) provided reference commercial brake pads and manufactured the brake pads with modified composition used in this study. The composition of the reference commercially available low-steel brake pad was adjusted, and nanocomposite kaolin/TiO2 (KATI) (Figure 1) was introduced to the formulation as a new possible functional filler. Detailed description of laboratory procedure for preparation of KATI is presented in the study [9]. The prepared KATI contained about 60 wt% of TiO2 nanoparticles with crystallite size of 18 nm.
Hence, the phase composition of the pads is confidential; therefore, Table 1 shows only content of the groups of ingredients in volume% without detailed specification.
AK Master and MTK wear test with both samples were performed using full-scale dynamometer at Federal-Mogul Motorparts, Germany. These procedures simulate various braking scenarios with various brake pressures and speed, and thus, these studied materials were exposed to different conditions, which may occur during usage of automotive friction brakes. The dynamometer program AK Master describes the friction value behavior of a friction material with regard to the influences of pressure, temperature, and speed. Its main purpose is to compare friction materials under the most equal conditions possible. To take account of the different cooling behavior of the different test stands, the fading series are temperature-controlled. Project-related brakes and brake discs must be used. Test conditions are defined by inertia, press rate increase, sampling frequencies, temperature measurement, and cooling conditions. The friction performance is tested under various braking conditions, such as speed, pressure, and disc temperature representing different situations during vehicle driving. Firstly, pads and rotor go through a friction check in new conditions (30 brake applications), followed by a bedding section to get pad and rotor surface adapted (62 Ba). After that, a characterization of the friction value at standard pressure, speed, and initial temperature is measured (6 Ba). Next point is speed/pressure sensitivity of the friction coefficient (5 × 8 Ba). Again, a characterization as before is added (6 Ba). A cold temperature check followed by a highway simulation with elevated speed is next (3 Ba). This section is closed by a newly conducted characterization (18 Ba). The fade characteristic is tested by continuously increasing disc temperature up to 550°C initial temperature by executing 15 brake applications. The next characterization (18 Ba) is followed by pressure lines. This means 8 Ba with increasing pressure at 100°C initial temperature, an increase of disc temperature up to 500°C by 9 Ba, and again a pressure line at 500°C starting temperature (8 Ba). Next sections are again a characterization (18 Ba), followed by a second fade (15 Ba), and finally a characterization (18 Ba ≥ total 274 Ba).
Friction values are determined by the following formula. Efficiency factor is determined as 100%. The brake torque is measured during the brake application, and an arithmetic average is calculated to each of them. The value is calculated as follows:
Evaluation of wear behavior of pad material in Main-Taunus endurance course based on vehicle data was done. The course is a driving cycle containing 240 brake applications each, depending on the driving conditions tracked from the vehicle driving. 20 test cycles are run during the MTK test, which means 4800 brake applications in total. The wear of friction pads is determined by thickness measurement before and after the test at four defined positions.
The inner brake pad (placed in the inner side of the disc) was selected for further experiments due to higher pressures and temperatures achieved during braking. For better manipulation with the studied samples, back plates were removed from the pads. Since friction composites could be considered as homogenous material on the macro level, randomly selected area of 25 mm2 of the initial sample surface was labelled and the rest of the sample was covered with aluminum foil to ensure that the same one will be analyzed by all of the techniques used. Since the friction layer formation depends on many factors and varies at the sample surface, two spots (called site 1 and site 2) of 25 mm2 from different areas of the tested sample surface were labelled. Rest of the sample was covered with aluminum foil. The samples prepared this way were analyzed using a combination of selected analytical methods.
2.2. Material Characterization Techniques
Light digital microscope VHX-500 (Keyence Corporation, Japan) was used for macroscale characterization of the friction surface of the studied samples together with surface roughness.
The EDS mapping of the selected (400 × 400 μm) areas from both new and tested samples was performed using TESCAN S8252G microscope equipped with field emission electron gun and Oxford Peltier-cooled SDD detector with 150 mm2 active area. The following analytical conditions were used: analytical working distance 6 mm, acc. voltage 20 kV, probe current 3 nA, mapping resolution of 1024 × 1024 pixels, and 10 accumulations. The experimental data were collected, and maps were obtained using the Aztec software.
The samples of the tested brake pads were also measured using TESCAN TIMA (TESCAN Integrated Mineral Analyzer) equipped with tungsten emitter at the following working conditions: acc. voltage 25 kV, working distance 15 mm, and probe current 7 nA. The EDAX Element 30 detectors were used for collecting the characteristic X-rays. The detectors are Peltier-cooled SDD (silicon drift detectors) with 30 mm2 active area. Two areas of CCA 5 × 5 mm were measured from each sample. Liberation analysis with high-resolution mapping mode was used. Pixel spacing was 1 μm/pixel, and 1000 X-ray counts and BSE intensity were collected from each pixel. Combination of BSE intensity and X-ray spectra was used for distinguishing the phase boundaries. The data were interpreted, and reports were processed by the TIMA 1.5.49 software.
3. Results and Discussion
Table 2 lists parameters of friction performance evaluation including the average friction coefficient dimensionless values (ranging between 0 and 1) in the individual sections of the AK Master test. The average and minimum values of the entire AK Master test are also presented. For friction composites with organic polymer matrix, the optimum friction coefficient is around 0.4. Both composites achieved very similar friction efficiency in the standard AK Master sections. In sections with high heat load (fade 1 and fade 2), the composite with KATI addition achieves higher average friction coefficient values which is a desirable feature, since other similar materials have the opposite tendency in this AK Master section. Both composites had a comparable weight loss of the disc after the test, while pad wear was reduced by 44.6% after KATI addition.
Table 3 presents results from MTK wear test and shows significant reduction of pad wear which leads to increased lifetime of the pad with KATI nanocomposite compared to the reference pad. It is worth mentioning that mean wear of inner and outer pad is the same. Commercial cast iron disc was used for both dynamometer tests and its wear rate is comparable for all tested pads.
Light digital microscopy is a useful technique for studying sample surface in the macroscale. It enables to image the surface in real colors, which may help to identify some of the present structures. For instance, in the case of the reference pad, it can be presumed that shiny gold-like features could be brass fibers (Figure 2(a)). Nevertheless, it might not be helpful in the cases without high color contrast samples or without having knowledge about the studied sample. 3D reconstruction of the surface shows the roughness and profile of the surface; however, the scale should not be considered as absolute values due to the influence of the sample placement in the microscope. Therefore, the 3D reconstruction of the surface should be used only for theoretical comparison of each point or area at one image instead of comparing among images.
(a)
(b)
(c)
The surface of the reference pad is quite rough with remarkable fibrous structures. The surface of two selected areas after the dynamometer testing is more flat, but still exhibits notable roughness and contains cavities. The first area (Figure 2(b)) contains some fibers, but the second area (Figure 2(c)) contains no visible fibers. Contact plateaus (fractions of the pad surface that mediate direct contact with a disc) are clearly visible at both areas.
The surface of the initial pad containing KATI nanocomposite is less rough. Some fibers can be seen, but they are not made of brass as in the reference pad. Some cavities can be seen at the surface of the tested pad (Figures 3(b) and 3(c)). Contact plateaus are not clearly visible in both cases of the tested samples.
(a)
(b)
(c)
Combination of electron image with EDS mapping enables visualization of elemental composition of each component of pad surface. Due to the lack of colors in the software, several elements, which were presented at the same areas of the sample (most probably create together a compound), have the same color (e.g., Cu and Zn—brass or K, Si, Al, and Mg—clay minerals). Oxygen is not included in these maps because it was present almost in every point, and thus, such map including oxygen would be useless. Figure 4 confirms the presumption that some fibers are made of brass and some are made of steel. Layered structure of clay minerals is also clearly visible.
(a)
(b)
(c)
Contact plateaus based on iron visible in Figure 5, together with high amount of fine wear particles based on chromium, tin, or barium, and some residues of layered structure of clay minerals, are also visible. Brass fibers create contact plateaus together with steel fibers but are in minority. Phase contrast in the contact plateaus together with different intensity of yellow color in the EDS map indicates that part of the surface of the plateaus consists of elemental iron and other parts are created by iron oxide, which is consistent with finding of Österle et al. [10].
(a)
(b)
(c)
The larger part of the surface of site 2 of the reference pad (Figure 6) is created by contact plateaus than in site 1 of the same pad (Figure 5). Therefore, the content of iron and also copper and zinc in the analyzed area is significantly higher. Other difference is in the presence of wear particles based on chromium and titanium, which are not clearly visible in Figure 5. Finally, this friction surface contains larger carbon-based structures than site 1; nevertheless, weight percentage of carbon is slightly lower.
(a)
(b)
(c)
The surface of the selected area of the pad with KATI nanocomposite (Figure 7) contains steel fibers, layered structure of clay minerals quite high number of fine particles based on titanium and chromium, where titanium is in significantly higher amount according to EDS, because of the addition of kaolinite/TiO2 filler. There are also some areas rich in carbon particles composed of sulfur as well. Iron layer evidently covers a large particle of brass.
(a)
(b)
(c)
The surface of site 1 pad with KATI nanocomposite (Figure 8) is partly created by contact plateaus and partly by wear particles. Some of the contact plateaus are composed of iron or iron oxides and some of brass. There are only few clay minerals structures visible in the image. Contrary to the reference pad, it is evident that there is significantly higher number of very fine wear particles on the surface composed of carbon, titanium, chromium, iron, and tin. As in the previous sample, there are bright blue-green particles with unknown elemental composition.
(a)
(b)
(c)
The surface of site 2 pad with KATI nanocomposite (Figure 9) is different from the surface of site 1 of the same pad. The contact plateaus are mainly created by iron and iron oxides, and only minority of them are created by brass. Layered structure of clay minerals is more visible in this site, and fine wear particles do not cover the top of contact plateaus, but they are trapped around them. Carbon-rich areas can be distinguished.
(a)
(b)
(c)
EDS mapping is a useful technique, which provides information about the distribution of single elements; however, multicomponent samples such as brake pads are hitting limits of this technique and even knowledge and experience with this kind of samples cannot guarantee a good result. Thus, there are tendencies to find other complementary techniques to obtain results with limited misinterpretations. Therefore, TESCAN TIMA microscope was used in this pilot study, and based on the published research paper available, it was the first time when it was used for this kind of sample.
Since initial pads have some defined composition (even industrially protected) and significant chemical and morphological changes could not be expected, only pads after dynamometer testing were further examined. Additionally, phase mapping provided by TIMA was also very time-consuming (approx. 12–15 hours). Signal from backscattered electrons and X-ray was collected to create a map of phase composition as can be seen in the following images.
The phase composition of site 1 (25 mm2) of the reference pad after dynamometer testing is shown in Figure 10. Most of the presented phases were identified via software. Due to difficulties with the detection limit and for certain level of simplification, carbon-S-rich was stated. Fe-Cu oxide was revealed to be the major phase at the surface. Since there are no well-defined phases with corresponding composition, it can be assumed that it can have intermetallic character and originated due to the local high temperatures, when melting of metals or their oxides can occur and subsequently those melted metals are mixed due to the kinetic force. The blue frame is a signal from the background (Al foil). Some pixels have not been classified (below 0.5% of the surface).
The phase composition of site 2 of the reference pad is given in Figure 11. Fe-Cu oxide is dominant at the surface of this area as well; however, significant part is created also by fine grained particles based on corundum, Ca oxide, Sn sulfide, and others. The surface of this sample appears to be more fragmented into contact plateaus and agglomerated particles, as well as in site 1 of the reference pad, intermetallic compounds can be found. Contrary to site 1, there is a higher incidence of iron pure and Fe oxide phases.
The surface of site 1 of the pad with KATI nanocomposite (Figure 12) contains Al-Si titanates, carbon-S-rich, and Fe-Cu oxide as the dominant phases. It can be assumed that Al-Si titanates represent the added KATI nanocomposite. Interestingly, Al-Si titanates are accompanied with clays as another phase. Noticeable amount of Fe oxide can be also seen. Generally, distribution of KATI nanocomposite among the surface is quite homogenous, which is desired for better friction performance of the pad.
The last examined surface is site 2 of the pad with addition of KATI nanocomposite (Figure 13). It can be stated that the surface of sites 1 and 2 is nearly similar with Al-Si titanates homogenously dispersed and accompanied with clays. Site 2 contains approximately more Fe oxide and carbon phases. The incidence of other phases is nearly similar.
Exported elemental maps and cumulative spectra of individual grains (the cumulative spectra are summed from all analyzed pixels comprising the grain Figure 14) support the idea of glassy-like nature of the mixed intermetallic phases (the Fe-Cu oxide and Fe-Cu-Zn oxide). The elemental maps of a representative grain (Figure 15) show that both Fe (Figure 15(c)) and Cu (Figure 15(a)) are homogeneously distributed within the grain (Figure 15(b)). The same applies for other elements present in these phases in smaller amounts (Si, S, Sn, and others (see Table 4)). Since the material appears to be chemically homogeneous at least on the micronscale level (spatial resolution of EDS), it would suggest involvement of at least some degree of melting and not only mechanical mixing caused by friction process. The presence of Sn and S in the spectra of the intermetallic phases may point to melting of the grains of Sn sulfides which can occur around 880°C [11].
(a)
(b)
(c)
(d)
It can be said that addition of TiO2 to the brake pad formulations is still not frequently studied, and this study with KATI nanocomposite is one of the first in this field; however, there are some studies with titanates. Mahale et al. [6] studied the influence of nanopotassium particles on the friction performance of nonasbestos organic brake pad. Addition of nanopotassium titanates improved performance and wear resistance of the tested pad formulation compared to the one with potassium titanates with micronsized particles due to its lubricating character that led to the reduction of friction coefficient fluctuations. This finding is in accordance with the study of Kim et al. [12] describing the positive impact of potassium titanate on fluctuation of friction coefficient. It has been also reported that potassium titanate has a positive impact on friction coefficient at higher temperatures [13, 14]. Based on these findings, we assume the KATI nanocomposite can stabilize friction coefficient at higher temperatures and then has a positive impact on the wear of brake pads. An original idea was to introduce KATI to the brake pad formula as a new functional filler; nevertheless, this material has probably lubricating properties, which contribute to reduction of wear and thus could increase of the lifetime of the pad.
4. Conclusions
Nanostructured composite material kaolin/TiO2 (KATI) has been successfully introduced to the existing automotive brake pad formulation. This pad with modified formulation has been tested together with commercially available low-steel pad on full-scale brake dynamometer to determine its friction performance and wear properties. It has been found that friction performance of the pad with KATI nanocomposite is comparable with the commercial low-steel pad; however, the average pad durability was significantly higher expressed by pad thickness loss which was about 50% lower compared to the reference pad formulation. Microscopic techniques revealed structures of the friction surface where some intermetallic phases have been found, what indicates occurrence of local high temperatures. The nanocomposite KATI was homogenously distributed on the surface of the pad before and also after the dynamometer testing which indicates good thermal stability of the modified composite. This modified brake pad formulation may be promising regarding the durability of pads and the needs to control wear and reduce particulate emissions. Most expected phases of the initial compounds were determined; however, some of the detected phases do not correspond to possible initial components. It can be interpreted as a result of phase transformation. When selecting a suitable nanomaterial for application in automotive friction composites, nanocomposite materials appear to be suitable candidates due to decrease of potential environmental risk, because of anchoring of nanoparticles in a matrix. New useful properties of the friction composite can also be achieved by using a lower amount of suitable nanomaterial than conventional bulk material.
Data Availability
No data were used to support this study.
Conflicts of Interest
The authors of the paper certify that they have no affiliations with or involvement in any organization or entity with any financial interest.
Acknowledgments
This study was supported by the project no. LO1203 “Regional Materials Science and Technology Centre - Feasibility Program” funded by the Ministry of Education, Youth and Sports of the Czech Republic and by the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 636592.