Motivation Urbanization and industrialization are leading to increased air quality problems worldwide, with particulate matter emissions playing a significant role. Particles emitted by vehicle braking systems in particular contribute considerably to air pollution and pose a significant health risk. Grigoratos and Martini show that a significant proportion of non-exhaust particulate emissions in urban areas come from braking systems [1]. Furthermore, Kelly and Fussell emphasise that the toxicity of these particles depends on their size, source and chemical composition, which underlines their harmful effects on health [2]. In addition to the effects on human health, fine dust particles also have a negative impact on technical systems. Due to their small size, they penetrate sensitive components and can impair their functionality. This can, for example, lead to increased wear and corrosion in mechanical components such as bearings and gearboxes. A common example is the necessary oil change in vehicles, which is required to extend the service life of combustion engines due to wear particles in the oil [3]. Measures against particle contamination are also becoming increasingly important in the field of electronics, as electrically conductive particles can lead to short circuits or hinder the correct placement of elements on printed circuit boards. Particle resistance must therefore be guaranteed over the entire life cycle of such components [4]. In view of these findings, the European Union (EU) wants to regulate and limit particulate matter emissions from tire and brake wear with the introduction of the Euro 7 standard in 2025 [5]. The physical and chemical processes that take place during the braking process lead to these emissions, the exact mechanisms of which are still to be identified by scientists [1]. The existing methods for measuring particulate matter emissions are not sufficient to adequately reflect realworld emission conditions. Amato et al. point out that the existing methods are not effective enough to precisely identify particulate matter pollution in urban areas, which leads to significant uncertainties in the assessment of environmental impacts [6]. Despite the availability of technologies to reduce particulate emissions, such as drum brakes, brake discs with special surface coatings [7], active filter systems with extraction, and passive filters without extraction [8], existing strategies are often not implemented consistently or effectively. There is a need for specifically adapted measures that Science and Research 19 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 Particulate matter emissions in brake systems - Development and application of an extended measurement methodology for particulate matter emissions from dry-running friction systems Francesco Pio Urbano, Katharina Bause, Arne Bischofberger, Sascha Ott, Albert Albers* Submitted: 20.09.2024 accepted: 15.11.2024 (peer review) Presented at the GfT Conference 2024 With the Euro 7 norm, brake wear emissions are regulated. The authors present a new method to measure wear particles in dry-running friction systems by integrating a sampling system into a test bench. The friction system is enclosed, and particles are transported to a measuring station during braking. Initial results show particle concentration depends on friction material, operating conditions, and correlates with wear. This method improves understanding of emissions, enabling the development of low-emission friction systems, which can help reduce particulate emissions and minimize health risks. Keywords particle emissions in brake systems, particulate matter measurement, dry-running friction systems, development of measurement methods, emission reduction Abstract * Francesco Pio Urbano, M. Eng. (corresponding author) Dipl.-Ing Katharina Bause Arne Bischofberger, M. Sc. Dipl.-Ing Sascha Ott Univ.-Prof. Dr.-Ing. Dr. h.c. Albert Albers IPEK - Institut für Produktentwicklung am KIT - Karlsruher Institut für Technologie Kaiserstraße 10, Karlsruhe, Germany Figure 1) developed in accordance with the IPEK-X-inthe-loop approach [13]. The test bench enables the friction pairing to be subjected to near-application loads and the simulation of typical braking processes in mid-range vehicles. The flywheel mass module, the drive machine and the axial force actuator serve as components for generating realistic test conditions. The flywheel mass module enables the setting of mass inertias of up to 3 kgm 2 , the drive unit allows speeds of up to 6,000 rpm and the axial force actuator generates axial forces of up to 10 kN. These configurations allow surface pressures to be generated in the friction contact that correspond to the actual operating conditions. The test bench enables friction systems to be stressed with specific braking work of 10 J/ mm² and specific braking power of up to 9 W/ mm 2 under realistic conditions. The test bench was equipped with a particle measurement system to record emissions. The friction system was enclosed and a closed pipe system was installed. A fan generates a constant volume flow that transports the particles emitted during the braking process to the measuring and collection point. As the fan could interfere with the formation of the friction layer, baffles were installed to minimize this effect. An aerosol spectrometer measures the particle mass concentrations of various sizes (PM10, PM2.5) in real time, while a cascade impactor collects fine dust particles of the sizes PM1, PM2.5 and PM10 and larger PM10. These particles are then examined under a scanning electron microscope and chemically analyzed. In addition, the filter box was modified so that macroscopic particles fall into a Petri dish, which are then examined under a digital microscope. The tests focus on dry-running friction systems from various application areas such as holding brakes, vehicle clutches and vehicle brakes. Science and Research 20 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 target the sources of emissions in order to achieve a significant reduction [6]. A major research gap lies in the insufficient knowledge of the different formation mechanisms of particles of different size classes. As the mechanisms for particle formation depend on the respective size, specifically adapted reduction strategies are also required. A deeper understanding of these relationships is therefore essential in order to develop precise measurement methods and effective emission reduction measures. This requires a method that makes it possible to analyze the formation mechanisms in a differentiated manner. This article is aimed precisely at developing such a method. Research method In order to specifically reduce emissions from brakes as they occur, it is necessary to fully understand the mechanisms and conditions in the friction contact as well as in the residual system. The approach of system tribology and system tribometers, as presented by Behrendt, Basiewicz and Klotz, among others, is recommended for this purpose [9-11]. The existing models for investigating the formation of particulate matter have not yet been fully researched. This article deals with the presentation of an extended measurement methodology based on the work of Sutschet et al. [12], which enables the detection and analysis of wear particles in dry-running friction systems. The influence of stress on particle formation is investigated and experimental results of a friction pairing are presented. The method presented quantitatively records the correlation between load and particle emissions in dry-running friction systems. In addition, by analyzing the wear particles, conclusions are to be drawn about the processes in the friction contact. These investigations will be carried out on a brake test bench (see Figure 1: Dry friction test bench with extension for measuring and collecting fine dust from dry-running friction systems [12] Experimental design and execution The procedure for particle measurement of different load levels and friction systems is presented below (see Figure 2). The different stresses (load levels) are set by varying the pressure, mass inertia and sliding speed. Some load levels are in the basic stress range (red), while others are partly in the misuse range (purple). The maximum sliding speeds at the start of braking vary between 10 and 20 m/ s, while the surface pressure is in the range of 0.3 to 0.7 MPa. The sliding speeds of load levels 1, 3 and 4 are identical, while the two load levels 2 and 5 exhibit the highest surface pressure at equivalent values. This allows the factors influencing particle formation to be systematically investigated. A new friction disk is used for each load stage, which is run in with 1,000 brake operations. The load stage is then carried out with 200 brake operations. Results The results of the investigations of two friction systems are presented below. Friction system B consists of a counter friction disk made of low-alloy steel (C45) with an organic, mass-pressed friction lining. This metal-free friction lining is used in external shoe brakes and disc brakes. According to the manufacturer, the friction lining can permanently withstand temperatures of up to 250 °C and can briefly withstand temperatures of up to 350 °C. Further friction lining B contains a higher proportion of abrasive materials and resin components, making it harder and stiffer. Friction system A consists of a cast iron counter-friction disk (GGG40) and an organic, mass-pressed, metal-free friction lining. Both friction linings are geometrically identical and are available as friction rings with an inner diameter of 140 mm and an outer diameter of 160 mm. Figure 3 shows the mean friction coefficients of the two friction Science and Research 21 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 Figure 2: Test procedure and specific friction work and friction power values for the different load levels Figure 3: Boxplots of the mean friction coefficients of the two friction pairings A (top) and B (bottom) cle concentrations. Furthermore, the two load levels 4 and 5 are significantly above the level of the other load levels. The exception to this is load level 1 for friction pairing A. The PM2.5 emissions are below the PM10 emissions for both friction pairings and across all load levels. The bar chart in Figure 6 compares the two friction linings of the friction pairings in terms of their weight loss across the different load levels. Despite identical test conditions within a load level, it can be seen that friction pairing A shows a significantly lower weight reduction overall. Figure 7 shows the results of the EDX analysis. The atomic concentrations of PM10 emissions from friction pairing B were determined for load levels 1 and 4. Science and Research 22 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 pairings. There are no substantial differences between the load levels within a given friction pairing, as shown in Figure 3. Although the two load levels 4 and 5 are outside the range of the basic load. No significant difference is found between the friction pairings either. This is due to the fact that friction pairing B has a greater scatter. The course of PM10 emissions over the 200 brake operations is shown in Figure 4. The figure shows that the concentration of PM10 emissions varies greatly and that certain load levels, such as L4 and L5, tend to have higher particle concentrations. Figure 5 shows the results of the integration of PM2.5 and PM10 emissions over the 200 brake operations. It can be seen that the higher the load level, the higher the parti- Figure 5: Sum of the two particle concentrations over the entire brake circuits for friction pairing A (left) and B (right), each shown with different scaling Figure 4: Progression of PM10 emissions for friction pairing A (top) and friction pairing B (bottom), each shown with different scaling Discussion The friction pairings investigated in this paper have so far shown no clear correlation between friction coefficient and the measured particle concentrations. The two misuse load levels 4 and 5 have a similar friction coefficient level to the other levels in the basic load range. However, there are significant differences between the friction pairings in terms of particle concentrations. The concentrations of friction pairing B are between 5 and 49 times higher for PM10 emissions compared to friction pairing A, depending on the load level. For PM2.5 emissions, the difference is in the range of 6 to 35 times. Load level 2 shows the greatest difference for both emission categories. A similar trend is also observed when comparing the weight loss between the two friction pairs. At all load levels examined, friction pairing B shows higher wear compared to friction pairing A. The most significant difference in wear occurs in load level 2, where the wear is 54 times higher. Load level 1 shows the smallest difference with a 5-fold increase in wear. This pattern is also reflected in the particle concentrations, with load level 1 showing the smallest difference and load level 2 the largest. The differences between the two friction pairings can be explained by the fact that friction pairing B is a material with a higher proportion of abrasive materials. This leads to an increase in particles originating from the counter-friction disk. This can be demonstrated by the EDX analyses in Figure 7. Since friction pairing B is a metal-free material, the iron concentrations determined can only originate from particles of the counter-friction disk. As the load level increases, an increased iron concentration can be observed, which indicates increased wear of the counter-friction disk. Science and Research 23 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 Figure 7: Atomic concentration of PM10 emissions of friction pairing B for load levels 1 and 4 Figure 6: Comparison of the weight reduction of the friction linings between the two friction pairings concentrations. For the two friction pairings used, a correlation between particle concentrations and weight reduction can be established across the different load levels. In future, this could make it possible to use particle concentration measurements to draw conclusions about the qualita- Science and Research 24 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020 Furthermore, abrasive materials lead to a smoothing effect of the counter friction partner. This effect is illustrated in Figure 8, where the matt friction surface of friction pairing A can be seen on the left and the shiny, smooth friction surface of friction pairing B on the right. Another possible explanation for the higher particle emissions of friction pairing B could be that the friction lining of friction pairing B has a higher hardness and stiffness than friction pairing A. As a result, the friction ring is not fully loaded during loading (see Figure 9, red circle) and local areas are subjected to higher loads. The power density increases for these more highly stressed surfaces and the surface temperature rises sharply locally. As the wear of organic materials is largely determined by temperature, particle emissions increase as a result. Figure 10 illustrates the relationship between the weight reduction of the friction lining and the measured particle Figure 10: Relationship between weight loss and particle concentrations for friction pairing A (left) and B (right) Figure 9: Section of the friction surface of the friction lining of friction pairing A (left) and B (right) Figure 8: Section of the friction surface of the counter-friction disk of friction pairing A (left) and B (right) tive progression of wear. This would be an important step towards understanding particle concentrations as a sensor and gaining information from the tribological system. Conclusion and outlook In this article, two different friction pairings are examined with regard to their emission and wear behavior using a brake test bench. The method according to Sutschet et al [12] is used to measure particles in the micro and macro range. It is shown that the particle concentration increases with increasing stress. There are also significant differences in the emission and wear behavior of the friction pairings. A correlation between friction coefficient and load level as well as between friction coefficient and wear is not recognized. However, the measured particle concentrations correlate with wear, which opens up the potential to use wear particles as sensors in the future and to obtain information from the tribological system. Future studies will include additional friction pairings to determine whether the observed behavior also occurs with other combinations. The load levels are also expanded and analyzed in order to test the transferability of the results. In addition, the aim is to quantitatively record the previously qualitative correlation between particle concentration and wear in order to allow conclusions to be drawn about the actual wear of the friction lining. In addition, chemical analyses are to be carried out to determine the material composition before and after exposure. These results are to be used in conjunction with thermomechanical stresses and surface changes to develop theoretical models for the underlying wear mechanisms. This knowledge will help to identify optimal brake materials and operating points to minimize particulate emissions. Acknowledgments The investigations presented in the publication were performed as part of the IGF-Project 22080-N. The authors acknowledge the funding of the research project. The IGF-Project 22080-N of the “Forschungsvereinigung Antriebstechnik e.V. (FVA)” is funded by the Federal Ministry for Economic Affairs and Energy through AiF within the program for Industrial Collective Research (IGF) based on a decision of the German Bundestag. Literature [1] T. Grigoratos und G. Martini, “Brake wear particle emissions: a review,” Environmental Science and Pollution Research, Jg. 22, Nr. 4, S. 2491-2504, 2015, doi: 10.1007/ s11356-014-3696-8. [2] F. J. Kelly und J. C. Fussell, “Size, source and chemical composition as determinants of toxicity attributable to ambient particulate matter,” Atmospheric Environment, Jg. 60, S. 504-526, 2012, doi: 10.1016/ j.atmosenv.2012.06.039. [3] Guillermo E Morales-Espejel. “Verschleiß und Oberflächenermüdung bei Wälzlagern - Evolution.” Zugriff am: 14. November 2024. [Online.] Verfügbar: https: / / evolution.skf.com/ de/ verschleiss-und-oberflachenermudung-bei-walzlagern/ # [4] P. Trunz, “ZVEI_Leitfaden: Technische Sauberkeit in der Elektrotechnik, 2. Auflage, Version 2019,” 2018. [5] Europäische Union, “Verordnung (EU) 2024/ 1257 des Europäischen Parlaments und des Rates vom 24. April 2024 über die Typgenehmigung von Kraftfahrzeugen und Motoren sowie von Systemen, Bauteilen und selbstständigen technischen Einheiten für diese Fahrzeuge hinsichtlich ihrer Emissionen und der Dauerhaltbarkeit von Batterien (Euro 7), zur Änderung der Verordnung (EU) 2018/ 858 des Europäischen Parlaments und des Rates und zur Aufhebung der Verordnungen (EG) Nr. 715/ 2007 und (EG) Nr. 595/ 2009 des Europäischen Parlaments und des Rates, der Verordnung (EU) Nr. 582/ 2011 der Kommission, der Verordnung (EU) 2017/ 1151 der Kommission, der Verordnung (EU) 2017/ 2400 der Kommission und der Durchführungsverordnung (EU) 2022/ 1362 der KommissionText von Bedeutung für den EWR.” [6] F. Amato et al., “Urban air quality: The challenge of traffic non-exhaust emissions,” Journal of hazardous materials, Jg. 275, S. 31-36, 2014. doi: 10.1016/ j.jhazmat.2014.04.053. [Online]. Verfügbar unter: https: / / www.sciencedirect.com/ science/ article/ pii/ S030438941400315X [7] LASERLINE. “Die Bremse für weniger Feinstaub - Neue Euro 7 Abgasnorm.” Zugriff am: 14. November 2024. [Online.] Verfügbar: https: / / www.laserline.com/ de-int/ news-detail/ die-bremse-fuer-weniger-feinstaub/ ? creative= 428605098678&keyword=%2Bfeinstaub%20%2 Bbremse&matc [8] MANN+HUMMEL. “Bremsstaubpartikelfilter.” Zugriff am: 14. November 2024. [Online.] Verfügbar: https: / / oem.mann-hummel.com/ de/ oemprodukte/ feinstaubfilter/ bremsstaubpartikelfilter.html [9] M. Behrendt, “Entwicklung eines Systemtribometers mit Abbildung mehrachsiger instationärer Beanspruchungskollektive zur Analyse von Reibung und Verschleiß im Mischreibungsgebiet im Kontext nasslaufender Umschlingungs-CVT,” in Albers (Hg.) 2009 - Forschungsberichte des IPEK - Institut, Bd. 36. [10] M. Basiewicz, “Ein Beitrag zur Validierung nasslaufender Lamellenpakete aus Anfahrelementen von Fahrzeugen im Betriebszustand ‘geregelter Dauerschlupf’,” Dissertation, Institut für Produktentwicklung - IPEK, Karlsruher Institut für Technologie (KIT), Karlsruhe, 2020. [11] T. Klotz, “Ein Beitrag zur experimentellen Untersuchung trockenlaufender Friktionspaarungen im Hinblick auf deren Schädigungs- und Erholungsverhalten während und nach kurzzeitig stark erhöhter Beanspruchung,” in Albers, Matthiesen (Hg.) 2022 - Forschungsberichte des IPEK - Institut, Bd. 149 (Dissertation). [12] A. Sutschet, K. Bause, A. Bischofberger und S. Ott, “Feinstaubemissionen trockenlaufender Friktionssysteme in Fahrzeugen,” Forsch Ingenieurwes, Jg. 87, Nr. 2, S. 521- 528, 2023, doi: 10.1007/ s10010-023-00664-9. [13] A. Albers, M. Behrendt, S. Klingler und K. Matros, “Kapitel 6: Verifikation und Validierung im Produktentstehungsprozess,” in Handbuch Produktentwicklung, U. Lindemann, Hg., München: Hanser, 2016, S. 541-569. Science and Research 25 Tribologie + Schmierungstechnik · volume 71 · issue 4/ 2024 DOI 10.24053/ TuS-2024-0020