In order to achieve a sustainable global society, the United Nations adopted the Sustainable Development Goals (SDGs) in 2015. This initiative has led to increased awareness regarding the reduction of friction loss, machine maintenance, parts replacement frequency, and the use of lubricants and additives in various regions and companies (UN, 2015). “Sustainable consumption and production” is one of the key global objectives, highlighting the importance of extending the lifespan of machines and their components in the field of tribology. When considering the life expectancy of machines, it becomes evident that the wear of mating parts has a direct impact, with wear debris believed to significantly reduce their operational lifespan. An investigation using an atomic force microscope revealed that the size of wear debris is quite small, typically in the range of several nanometers (approximately 15–40 nm). This was observed in scenarios involving dry friction between identical material pairs, such as an almost pure iron (99.5%) pin against an iron disk, pure capper (99.99%), pure zinc (99.99%), and pure nickel (99.7%). These tiny debris particles eventually aggregate, forming larger particles exceeding 10 μm in size (Mishina et al., 2003). The presence of wear debris or abrasive particles in the surrounding environment contributes to abrasive wear, emphasizing the importance of employing filtration to remove these particles and thereby extend the service life of machinery. Utilizing the ferrography method to determine the size, shape, color and other characteristics of wear debris, coupled with vibration analysis, represents powerful tools for predicting wear patterns at contact points within a tribosystem. This field has been extensively researched over the past 4 decades (Roylance et al., 2000). Recently, real-time measurement of wear debris in lubricating oil has been made possible through a laser–photo diode detector system, which measures particle size distributions and counts along the oil tank–rubbing surface path (Iwai et al., 2010; Hong et al., 2018). This technology enables the detection of wear debris with a diameter larger than 1 μm. As the capability to detect smaller diameters continues to improve, the accuracy of wear amount estimation becomes increasingly precise. However, the challenge remains that wear debris with a diameter less than 1 μm is often overlooked due to limitations in measurement systems. Consequently, we still lack a comprehensive understanding of whether these small particles can enter contact points and function as abrasives.

Over the past few decades, numerous authors have explored the entrainment of particles between sliding surfaces and have sought to determine how these particles can or cannot infiltrate. The inception of this phenomenon, often referred to as “particle entrainment between contacts,” has been discussed in several literature sources in the context of lubrication contaminants (Tallian, 1976; Loewenthal and Moyer, 1979; Cusano and Sliney, 1982a; Cusano and Sliney, 1982b). Additionally, it gained prominence with the rise or ceramics bearing technology (Dwyer-Joyce, 1999; Mitchell et al., 2000; Wang et al., 2000; Wang et al., 2007). As a result, there has been a significant focus on carbonaceous hard coatings as an appealing material for reducing friction and wear. However, this approach comes with its share of challenges, including issue related to peeling caused by cyclic impacts (Abdollah et al., 2010; Abdollah et al., 2011; Abdollah et al., 2012a; Abdollah et al., 2012b), droplets emanating from the surface (Lee et al., 2018; Lee et al., 2019; Lee et al., 2020), and oil additives affecting mechano-chemical wear (Kassim et al., 2020; Kassim et al., 2021). It’s noteworthy that these particles have been observed to be smaller than 1 μm in diameter (Tokoroyama et al., 2018), underscoring the growing importance of detecting such small wear debris and accurately predicting wear levels. The prediction of wear amounts on surfaces, such as ball or roller bearings, has been accomplished by considering the role of particles acting as abrasives passing through a contact point. This concept involves the removal volume of the surface and its relationship with the hardness between the surface and the particles, a phenomenon known as three-body abrasive wear. Dwyer-Joyce, for instance, reported a bearing mass loss that correlated with particle size, demonstrating an increase in mass loss as the size of diamond particles ranged from 1 to 6 μm in diameter. Notably, when using diamond particles with a diameter less than 1 μm, a mass loss was still observed, implying that diameters less than 1 μm can indeed enter a contact point.

Modeling abrasive wear through wear debris has been a subject of study for many years. Third-body wear caused by such debris is influenced by factors such as debris hardness, normal load between mating surfaces, sliding speed, and so on. However, determining the ratio of particles entrained in the system can be quite challenging, primarily because small-diameter particles result in shallow, narrow, and often undetectable scars. Consequently, a comprehensive understanding of how abrasive wear progresses due to small-diameter particles can only be achieved through in-situ observation. In-situ direct observation of these particles at a contact point was successfully carried out by Strubel and colleagues (Strubel et al., 2016; Strubel et al., 2017a; Strubel et al., 2017b). Their observations were conducted using particles with diameters of 10 μm or larger, revealing that larger and heavier particles were more likely to be entrapped than smaller and lighter ones. It was observed that smaller-diameter particles did not enter the system because a reverse flow occurred at a midpoint between mating surfaces, preventing their entrainment in a fully flooded condition. Another critical aspect to consider is the thickness of the oil film outside the Hertzian contact area. Cann et al. reported that a decrease in the oil film thickness (h _oil∞) in the vicinity of the sliding track led to a reduction in the oil film thickness at the contact point. This reduction was attributed to inadequate oil replenishment from the vicinity of the sliding track. Both of these viewpoints—particle entrainment and oil film thickness—intersect and may explain the potential for particle entrainment, even if the particle diameter is around the micro to sub-micro range.

One significant limitation in the observation of micro to sub-micro diameter particles lies in optical constraints. Using white light and a standard CCD camera, particles with diameters less than 0.3 μm cannot be reliably detected. To overcome this optical limitation, particle image velocimetry (PIV) has emerged as a widely-used technique for observing fluid motion by employing fluorescent particles (Kinoshita et al., 2007). In this study, a novel approach was employed, utilizing fluorescent-stained solid particles for direct observation under boundary lubrication conditions. Bright fluorescent particles were obtained through a staining technique involving glass particles (silica) and Rhodamine B (a fluorescent agent) combined with a silane coupling agent to achieve sufficient intensity. With this staining method, it has become possible to directly observe the behavior at the contact surface of particles within 1 μm or less, which sets it apart from other studies the most.

In this study, we examined the quantity of entrainment particles by specifically examining the oil film thickness in the surrounding area beyond the point of contact, which we refer to as the “outside oil film thickness.” Subsequently, we utilized silica particles with average diameters of 0.8, 1.0, and 3.0 μm to elucidate the impact of the outside oil film thickness on particle entrainment. Additionally, we conducted dry friction tests to draw comparisons with the lubricated conditions. Furthermore, the establishment of the staining method in this study opens up the possibility of observing the behavior of nanoparticles (Wu et al., 2018; Gao et al., 2021; Nomede-Martyr et al., 2021; Wu et al., 2021; Bukvić et al., 2023), allowing for the investigation of their tribological properties, which have been garnering attention in recent years.

The relationship between the number of entrained particles and the non-dimensional value “r,” which represents the outside oil film thickness divided by the particle diameter, is summarized in Figure 12. In dry conditions, a substantial number of particles could enter the contact point regardless of their diameter. However, the entrainment number decreased when the “r” value approached approximately 1. Consequently, the maximum number of entrained particles increased within the range of 1 ≤ r ≤ 2. Beyond this range, the number of entrained particles gradually decreased. The reason behind this change in the number of entrained particles due to the “r” value is believed to be related to reverse flow in front of the contact point. A cross-sectional schematic image illustrating this phenomenon is presented in Figure 13. When “r” is less than or equal to 1 as shown in Figure 13i, the reverse flow creates a vortex with a radius smaller than that of the particles because the oil film thickness is less than the particle diameter. As a result, the vortex flow does not significantly affect the motion of particles. In the case where “r” falls within the range of 1 ≤ r ≤ 2 (Figure 13ii), a vortex flow is generated along the direction from the disk to the oil surface, with a vortex radius equal to the particle radius. Consequently, particles can move from the disk surface to the oil surface and back to the disk surface again. This enhances the maximum number of entrained particles. Finally, for values of “r” greater than or equal to 2 (Figure 13iii), the vortex flow radius becomes larger than the particle radius. Thus, once particles attach to the disk surface and follow the vortex flow, they are unable to return to the contact point. Based on the mechanisms described above, specific outside oil film thickness values that enhance the number of entrained particles is suggested.

• We conducted investigations into the entrainment of glass beads with different diameters, such as 3.0 μm, 1.0 μm, and 0.8 μm. Utilizing a fluorescent staining technique facilitated by silane coupling, we were able to observe the contact point in-situ. The high-speed camera effectively detected the number of entrained particles.