1 Introduction In rolling element bearings, lubricants such as oils and greases are used to reduce friction and wear. In addition, these lubricants also fulfil other tasks, such as preventing corrosion on the metal surfaces of the bearings. As these requirements cannot be adequately fulfilled by base oils alone, additives are added, which specifically improve the lubricant properties. In order to achieve the desired properties, various additives are used simultaneously in one lubricant. The simultaneous use of various surfaceactive additives might cause interactions between those additives, while forming a tribological boundary layer. Synergistic as well as antagonistic effects of interactions can occur that potentially influence the wear protection performance of the additives. For example, the performance of zinc dialkyl dithiophosphate (ZDDP) can be enhanced by friction modifier additives [1,2]. In contrast, by adding other additives such as dispersants the tribological boundary layer formation rate of ZDDP can be reduced [3,4]. The reduction or at least knowledge of harmful additive interactions is required to ensure optimum use of the additives' potential. Likewise, an increase in the positive additive interactions can be used to reduce the needed quantity of additives. Therefore, the mechanisms of the surfaceactive additives need to be investigated in combination with each other. Previous studies with the sole use of the extreme pressure / anti wear (EP/ AW) additive ZDDP revealed that a homogeneous layer on the rolling elements is desired to assure good wear protection [5]. However, commercial oils commonly contain extreme pressure / anti wear (EP/ AW) and corrosion inhibitor (CI) additives, which led to antagonistic interactions [6,7]. Based on the mechanisms Science and Research 5 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 Influence of Corrosion Inhibitors on the Wear Protection of Extreme Pressure / Anti Wear Additives in Oil-lubricated Rolling Bearings Merle Reimers, Silvia Richter, Georg Jacobs, Joachim Mayer, Florian König* submitted: 19.09.2024 accepted: 08.08.2025 (peer review) Presented at GfT Conference 2024 Surface-active additives have a central influence on the wear behavior of rolling bearings. The simultaneous use of different surface-active additives can cause interactions between these additives and influence the formation of a tribological boundary layer required for wear protection. Until now, these interactions between corrosion inhibitors (CI) and extreme pressure / anti wear additives (EP/ AW) have been insufficiently investigated. This article therefore presents a method for evaluating the influence of CI on the effect of EP/ AW additives in oil-lubricated rolling bearings. In standard FE8 tests according to DIN 51819-3, the mineral oil used with the EP/ AW additive zinc dialkyl dithiophosphate (ZDDP) shows insufficient wear protection. By adding the corrosion inhibitor zinc carboxylate, the wear mass can be significantly reduced but not completely eliminated under the same conditions. To evaluate the formation of the tribological boundary layers, further tests are therefore carried out at a higher relative lubricant film height. The influence on the composition of the boundary layer is analyzed using electron probe microanalysis (EPMA). By adding zinc carboxylate, a higher mass coverage can be achieved in the tribological boundary layer than when using ZDDP alone. Keywords Wear, Additives, Lubricant, Additive Interactions, Extreme Pressure / Anti Wear Additive, Corrosions Inhibitor, Rolling Bearings Abstract * Merle Reimers a (corresponding author) Orcid-ID: https: / / orcid.org/ 0000-0002-8880-5561 Dr. rer. nat. Silvia Richter b , Dr.-Ing. Georg Jacobs a , Dr. rer. nat. Joachim Mayer b , Dr.-Ing. Florian König a a Institute for Machine Elements and Systems Engineering RWTH Aachen University Schinkelstr. 10, 52062 Aachen, Germany b Central Facility for Electron Microscopy RWTH Aachen University Ahornstr. 55, 52074 Aachen, Germany neral oil and 1.0 wt.% ZDDP is defined as the reference. Therefore, three oils were investigated in this study, which are shown in Table 1. The ZDDP used in this study is a primary ZDDP with a C8 chain. It contains 8 wt.% phosphorus, 16 wt.% sulfur and 9.5 wt.% zinc. The calcium sulfonate is composed of 2 wt.% calcium and the zinc carboxylate holds 15 wt.% zinc. 2.2 FE8 Test Method To evaluate the influences of CI on the wear protection of EP/ AW under high loads axial thrust bearing tests (FE8 tests) based on DIN 518193 [8] are carried out. A FE8 test rig with axial cylindrical roller bearings (81212) was used for this purpose. The bearings each consist of 15 rolling elements, two washers and a cage. All rolling elements and washers are made of 100Cr6, while a polyamide cage (PA66) was selected in this study to avoid the chemical effects associated with other cage materials such as brass. Additionally, only rolling elements and washers with an arithmetic mean roughness value Ra = 0.04 - 0.05 µm were used, to eliminate potential influences of varying roughness on the wear test results. Two of the test bearings described are tested simultaneously and loaded by a hydraulic load and a disc spring assembly. Each bearing is individually supplied with 0.1 liters of oil per minute. The shaft with the two bearings is driven at the desired rotational speed with continuous measurement of the frictional torque. A heating ja- Science and Research 6 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 of EP/ AW, it is assumed that the joint use of EP/ AW and CI hinders the forming of homogeneous, tribological boundary layers. Consequently, to ensure a safe operation while using both EP/ AW and CI the interactions need to be better understood. The aim of this contribution is to introduce a method to evaluate the influences of the CI on the wear protection of EP/ AW when simultaneously using EP/ AW and CI in oillubricated rolling bearings. Therefore, axial thrust bearing tests (FE8 tests) accordingly to DIN 518193 [8] are carried out to provide a practical assessment for highly loaded rolling contacts. The base oil with only the EP/ AW is set as reference. For the simultaneous use, the EP/ AW is combined with either one of the two CI of this study. The influences of CI on the wear protection of EP/ AW are evaluated by analyzing the wear mass of the rolling elements sets. Furthermore, electron microscopic methods such as electron probe microanalysis (EPMA) are used to investigate the chemical compositions of tribological boundary layers. This allows the influence of the CI on the tribological boundary layers to be examined. Specifically, the mass thickness and chemical composition of tribological boundary layers formed in the FE8 tests are measured. 2 Materials and Methods The oils used, including the base oil and additives, are presented below. The test method and microanalysis methods are also described. 2.1 Base Oil and Additives The oils consist of mineral oil (API group I) as base oil, 1.0 wt.% zinc dialkyl dithiophosphate (ZDDP) as EP/ AW and either 0.5 wt.% calcium sulfonate or 0.5 wt.% zinc carboxylate as CI. The oil with only mibase oil 1.0 wt.% EP/ AW 0.5 wt.% CI M-ZDDP mineral oil (API group I) ZDDP - M1 calcium sulfonate M2 zinc carboxylate Table 1: Oils used in this study a) Axial thrust bearing test rig (FE8 test rig) oil supply axial load heating and cooling drive 2 test bearings torque measurement 0 -5 +5 0 +5 -5 0 -12.05 +13.70 length [mm] SRR [%] b) Test bearing Figure 1: FE8 test rig and test bearing cket is used to maintain the required test temperature of the oil. The test rig and the bearing type are shown in Figure 1 a. Due to the different linear speeds at the inner and outer end of the cylindrical rolling elements, different slide-roll-ratios (SRR) are present along its length. This leads to different products of relative velocity and pressure along the rolling element, which affects the formation of the tribological layers [9]. This is therefore also included in this study and shown in Figure 1 b. All FE8 tests were conducted under the standardized conditions of 80 kN (1.89 GPa), 80 °C and 7.5 rpm for 80 hours. Each oil was tested once due to the good reproducibility as shown in [5,10]. Furthermore, all FE8 tests were conducted with two test bearings and each bearing is analyzed separately. The FE8 test is analyzed on the basis of the wear mass of rolling element sets consisting 15 rolling elements. The wear mass is determined by weighing the rolling element sets of each test bearing before and after the test with an accuracy of 1~mg. The mean value from the rolling element sets of both test bearings is compared in this work in order to evaluate the influences of CI on the wear protection of EP/ AW. In addition, the total wear masses of the rolling element sets per bearing are given as error bars. Based on [11] a wear mass less than 10 mg is rated as very good wear protection of the oil, when tested under 80 kN, 80 °C and 7.5 rpm for 80 hours. A test result with a wear mass less between 10 mg and 30 mg is still considered good, while a wear mass over 100 mg is rated as very high wear, i.e. very poor wear protection of the oil. 2.3 Microanalysis Methods After the FE8 tests the rolling elements were subjected to a cleaning procedure to remove residues of the lubricant film. The cleaning procedure was established by checking the reproducibility of the thin film analysis. Thus, the cleaning recipe can be described as follows: • Clean all sides of using absorbent cotton and white spirit • Immerse in white spirit for approx. 10 seconds • Clean again with absorbent cotton and white spirit • Dry in the air • Clean with ethanol and acetone in an ultrasonic bath ◦ 5 min ethanol ◦ 5 min acetone ◦ 5 min ethanol ◦ 5 min acetone ◦ 5 min fresh ethanol • Dry with a hair dryer EPMA measurements were carried out with a JEOL JXA-8530F microprobe analyzer equipped with a Schottky Field Emitter and 5 wavelength dispersive spectrometer. The operating conditions applied throughout the present work were: Primary electron energy: 10 keV; beam current: 100 nA; counting time: 10 s/ measuring point. The intensities of O Kα, Zn Lα, P Kα, S Kα and Ca Kα were detected. A synthetic multilayer for the analysis of O was chosen. For the measurements the electron beam was defocused to 10 µm. Thus, local inhomogeneities in the elemental distribution and effects from surface roughness were compensated. The X-ray intensities were calibrated by measurement on the following standards under constant conditions: Fe 2 O 3 for O Kα, Zinc for Zn Lα, GaP for P Kα, FeS 2 for S Kα and Andradite for Ca Kα. The k-ratios for all elements, i.e. the calibrated net intensities from the sample to those of bulk standards, were measured across the sample surface by conventional step scans with a step size of 140 µm. Science and Research 7 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 0 -5 +5 +4 0 Mass coverage [μg/ cm 2 ] -4 0 30 15 exemplary EPMA result of rolling element 0 -9.8 +10.8 length [mm] SRR [%] Oxygen Phosphorus Sulfur Calcium Zinc Total Elements Figure 2: Representation of the EPMA results very high (> 1300 mg). Nevertheless, the wear masses of the oil M1 were in the same range of the oil with ZDDP only (MZDDP). The high wear masses of the oil M-ZZDP align with the results of [13]. In contrast, the wear mass of the rolling elements sets was significantly reduced to 88 mg when ZDDP was used in combination with zinc carboxylate (M2), which means a reduction by more than 90 %. For a deeper understanding of the wear protection depending on the type of oil and used additives, the EPMA results performed along the surface of the rolling elements are shown in Figure 4. There is no homogeneous tribological boundary layer formed in FE8 tests for both oils with ZDDP only (M-ZDDP) as well as with ZDDP and calcium sulfonate (M1). Especially, in the first half of the measurement scan (-5 to 0 mm) of the oil with CI (M1) the partial mass coverages of all elements vary strongly. In general, the concentration of additive elements in the total mass coverage is small. The second half of the scan (0 to +5 mm) reveals almost no layer formation with both oils. Most of the surface area is covered by a total mass of about 1 µg/ cm 2 which corresponds to about 2 nm thickness assuming a density of 5 g/ cm 3 . This is the thickness of a natural oxide layer. If zinc carboxylate is added to ZDDP (M2), the EPMA profiles are more homogeneous. Based on previous results, a homogeneous boundary layer is necessary for a properly working wear protection [5]. This correlates with the results of the FE8 test (Figure 3). There are small variations of the mass coverage depending on the SRR condi- Science and Research 8 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 The electron beam energy was adjusted to 10 keV to get a surface-sensitive excitation of the interesting tribolayer. The information depth ranges from about 250 to 450 nm depending on the chosen X-ray line. Since the information depth is larger than the expected thickness of the tribological boundary layer a special algorithm (inhouse script) was applied to determine elementspecific mass coverages from the measured data, i.e. k-ratios. As physical model the work of Pouchou and Pichoir [12] was used to convert experimental k ratios into elementspecific or partial mass coverages. The EPMA measurements were carried out along the rolling element length as shown in Figure 2. 3 Results The wear masses of the rolling element sets from the FE8 tests are used to evaluate the influences of the CI on the wear protection of EP/ AW. Therefore, the results of the oil including EP/ AW and CI must be compared with the oil containing only EP/ AW and no CI. The wear masses in mg for the three different oils are shown in Figure 3. The wear mass of the reference test M-ZDDP, which contains solely EP/ AW ZDDP is shown twice for better comparison. None of the three oils lead to a wear protection under the standard FE8 test conditions (7.5 rpm, 80 kN, 80 °C, 80 h, λ = ~ 0.05), since all wear masses were higher than 30 mg. Especially, the wear masses of the oil containing ZDDP and calcium sulfonate (M1) and the oil with ZDDP only (MZDDP) were EP/ AW EP/ AW + CI T = 80 ˚C n = 7.5 rpm F = 80 kN t = 80 h (1.89 GPa) ZDDP zinc dialkyl dithiophosphate Ca-Sulf. calcium sulfonate Zn-Carb. zinc carboxylate standardized FE8 test conditions Oil name M-ZDDP M1 M-ZDDP M2 EP/ AW ZDDP ZDDP CI - Ca-Sulf. - Zn-Carb. Figure 3: Wear masses of FE8 tests with standardized test conditions Science and Research 9 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 T = 80 ˚C n = 7.5 rpm F = 80 kN t = 80 h (1.89 GPa) standardized FE8 test conditions M-ZDDP M1 M2 Oxygen Phosphorus Sulfur Calcium Zinc Total Elements Figure 4: EPMA results of rolling elements after FE8 test with standardized test conditions EP/ AW EP/ AW + CI ZDDP zinc dialkyldithiophosphate Ca-Sulf. calcium sulfonate Zn-Carb. zinc carboxylate T = 80 ˚C n = 75 rpm F = 80 kN t = 80 h (1.89 GPa) adapted FE8 test conditions EP/ AW ZDDP ZDDP CI - Ca-Sulf. - Zn-Carb. Figure 5: Wear masses of FE8 tests with adapted test conditions tribological boundary layer enriched with additive-elements has been formed. Due to the adapted FE8 test conditions the influence of the CI on the formation of the layer can be now investigated. In case of the calcium sulfonate as CI an additional Ca mass coverage is visible (M1) compared to the tribological boundary layer of the oil with ZDDP only (M-ZDDP). By adding zinc carboxylate, a higher mass coverage can be found in the tribological boundary layer (M2) compared to the only use of ZDDP (M-ZDDP). In addition, although there is no P portion in both CI used, the P mass coverage increases as well by adding either one of the CI. The CI might have a catalytic effect on the enrichment of P in the boundary layer. 4 Discussion No wear protection was achieved with none of the three oils used in this study at the standardized FE8 test conditions with 7.5 rpm. However, an influence of the CI on the wear protection performance of the EP/ AW can be observed. By adding calcium sulfonate as CI (M1) the wear mass was the same as with ZDDP only (M-ZDDP). In contrast, by using ZDDP with zinc carboxylate simultaneously (M2) over 90 % less wear mass was detected Science and Research 10 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 tions along the surface of the rolling element. It points to the formation of a tribological boundary layer, probably iron oxide layer, since the concentration of the additive elements is low. The thickness varies between 20 and 40 nm assuming a density of 5 g/ cm 2 . To investigate the tribological boundary layer closer, the FE8 test conditions were adapted accordingly to [13]. Except for the rotational speed, which is raised to 75 rpm, the test conditions stayed the same. Even with the adapted FE8 test conditions, testing continues in mixed friction (λ = ~0.2). All three oils were investigated again under the described adapted FE8 test conditions with higher rotational speed. The wear masses are shown in Figure 5. For all three oils the wear masses of the rolling elements sets were much lower compared to the standard FE8 test conditions. This can be explained by the less critical FE8 test conditions. Also, the wear masses do not differ significantly between the oils. A slight increase of the wear mass can be seen by adding calcium sulfonate to the oil with ZDDP (M1). The objective to investigate the tribological boundary layer closer by adapting the FE8 test conditions accordingly to [13] was successful. A homogenous tribological boundary layer was formed and the concentration of all additive elements is also higher for all three oils. A T = 80 ˚C n = 75 rpm F = 80 kN t = 80 h (1.89 GPa) adapted FE8 test conditions M-ZDDP M1 M2 Oxygen Phosphorus Sulfur Calcium Zinc Total Elements Figure 6: EPMA results of rolling elements after FE8 test with adapted test conditions compared to using ZDDP only (M-ZDDP). This correlates with the EPMA measurement. No homogenous tribological boundary layer was formed in FE8 tests for both oils with ZDDP only (M-ZDDP) as well as with ZDDP and calcium sulfonate (M1). In contrast, by adding zinc carboxylate to the oil with ZDDP (M2) the EPMA results show a more homogeneous boundary layer, which is necessary for a properly working wear protection [5]. Since the concentration of the additive elements is low, the tribological boundary layer is probably an iron oxide layer. The more homogeneous boundary layer and the lower wear masses indicates a positive influence of the CI zinc carboxylate on the wear protection of ZDDP. Nevertheless, to study the influence of CI on the wear protection of EP/ AW analyzing the boundary layer is needed. By adapting the FE8 test conditions to higher rotational speed (75 rpm) accordingly to [13] the boundary layer can be analyzed successfully. In comparison to the results of the adapted FE8 test with the oil containing of ZDDP only (MZDDP) the wear mass of the oil also consisting of calcium sulfonate (M1) slightly increases. This indicates a negative influence of the CI calcium sulfonate on the wear protection of ZDDP. The EPMA results of M1 show a calcium coverage in the tribological boundary layer as well as a higher phosphorus coverage compare to the EPMA results of MZDDP. Overall, this study both showed standardized as well as adapted FE8 tests are necessary to analyze the influence of CI on the wear protection of EP/ AW in oillubricated rolling bearings. The FE8 tests with the standardized test conditions (80 kN, 80 °C, 7.5 rpm, 80 hours) are needed to examine the wear protection by means of wear masses. Since no homogeneous tribological boundary layer was formed under the standardized FE8 test conditions, a closer investigation of the influences of the CI on the wear protection of EP/ AW was not possible. Therefore, the adapted test conditions (80 kN, 80 °C, 75 rpm, 80 hours) are used. Homogeneous tribological boundary layer were formed and can be further analyzed with microanalysis methods such as EPMA. Even if the wear masses are lower in the adapted FE8 test due to the less critical operating conditions, a slight influence of the CI on wear protection of EP/ AW can be seen. Nevertheless, the results from the standardized FE8 tests are necessary for a conclusive study. 5 Conclusion While forming a tribological boundary layer, synergistic as well as antagonistic effects of interactions between surfaceactive additives might occur and potentially influence the performance of those additives. The surfaceactive additives extreme pressure / anti wear (EP/ AW) and corrosion inhibitor (CI) additives are assessed as antagonistic [6,7]. However, EP/ AW and CI are used simultaneously in oillubricated rolling bearings. To ensure a safe operation of the rolling bearings, the influence of the CI on the performance which have been insufficiently investigated. A method to evaluate the influences of the CI on the wear protection of EP/ AW when using EP/ AW and CI simultaneously in oillubricated rolling bearings was introduced. Therefore, standard FE8 tests were carried out. Under standard FE8 test conditions, none of the three oils lead to a sufficient wear protection. Nevertheless, the wear mass was significantly reduced by adding zinc carboxylate compared to the sole use of ZDDP. This can be explained by the more homogeneous tribological boundary layer formed by this additive combination. To investigate the tribological boundary layer closer, the FE8 test conditions were adapted by raising the speed accordingly to [13]. Due to the less critical conditions the wear masses reduced significantly and are in close range to each other. Therefore, the FE8 test with the standard test conditions are needed to evaluate the influence of CI on the wear protection of EP/ AW. Nevertheless, the adapted FE8 test conditions are required as well to investigate the influence on the formed tribological boundary layer using electron probe microanalysis. By combining the standard and the adapted FE8 test with accelerated speed, a method was developed to investigate the influence of the CI on the wear protection of EP/ AW when using EP/ AW and CI simultaneously in oillubricated rolling bearings. The method includes an evaluation of the wear masses and EPMA results of both tests. With the introduced method a first step is shown to evaluate the influences of the CI on the performance of EP/ AW when using EP/ AW and CI simultaneously in oillubricated rolling bearings. To study the synergistic and antagonistic effects of the simultaneously use of EP/ AW and CI further investigations such as variation in different additives, base oils, additive concentration and ratio as well as runningin processes are needed. Acknowledgements The project (01IF22309N) is funded by the Federal Ministry of Economic Affairs and Climate Action BMWK (Bundesministerium für Wirtschaft und Klimaschutz) on the basis of a resolution of the German Bundestag. References [1] Eickworth J, Aydin E, Dienwiebel M, Rühle T, Wilke P, Umbach TR. Synergistic effects of antiwear and friction modifier additives. ILT 2020; 72(8): 1019-25. https: / / doi.org/ 10.1108/ ILT-07-2019-0293. [2] Lu R, Shiode S, Tani H, Tagawa N, Koganezawa S. A Study on the Tribofilm Growth and Tribological Properties Science and Research 11 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 schmierstoff-Prüfgerät FE8 - Teil 3: Verfahren für Schmieröl - einzusetzende Prüflager: Axialzylinderrollenlager. Berlin: Beuth Verlag GmbH. [9] Stratmann A. Einflüsse auf die tribologische Grenzschichtbildung beim Betrieb von Wälzlagern in der Mischreibung. Dissertation, Institut für Maschinenelemente und Systementwicklung, Verlag Mainz; RWTH Aachen University. [10] Burghardt G, Wächter F, Jacobs G, Hentschke C. Influence of run-in procedures and thermal surface treatment on the anti-wear performance of additive-free lubricant oils in rolling bearings. Wear 2015; 328-329: 309-17. https: / / doi.org/ 10.1016/ j.wear.2015.02.008. [11] van de Sandt N. Gebrauchsdauer von axial belasteten Wälzlagern bei starker Mischreibung. 1. Aufl. Aachen: Mainz 2004. [12] Pouchou JL, Pichoir F. Electron probe X-ray microanalysis applied to thin surface films and stratified specimens. Scanning Microscopy 1993; 1993(7): 167-89. [13] Burghardt G. Wirkung tribologischer Grenzschichten in Wälzlagern unter Mischreibung. Dissertation, Institut für Maschinenelemente und Systementwicklung, RWTH Aachen University 2017. Science and Research 12 Tribologie + Schmierungstechnik · volume 72 · issue 3-4/ 2025 DOI 10.24053/ TuS-2025-0013 of Tribofilms Formed from Zinc Dialkyl Dithiophosphate (ZDDP) and Molybdenum Dialkyl Dithiocarbamate (MoDTC). Tribology Online 2018; 13(3): 157-65. https: / / doi.org/ 10.2474/ trol.13.157. [3] Papay AG. Antiwear and extreme-pressure additives in lubricants. Lubrication Science 1998; 10(3): 209-24. https: / / doi.org/ 10.1002/ ls.3010100304. [4] Barcroft FT, Park D. Interactions on heated metal surfaces between zinc dialkyldithiophosphates and other lubricating oil additives. Wear 1986; 108(3): 213-34. [5] Rosenkranz L, Richter S, Jacobs G, et al. Influence of temperature on wear performance of greases in rolling bearings. ILT 2021; 73(6): 862-71. https: / / doi.org/ 10.1108/ ILT-03-2021-0076. [6] Mortier RM, Fox MF, Orszulik ST. Chemistry and Technology of Lubricants. Dordrecht: Springer Netherlands 2010. [7] Spikes HA. Additive-additive and additive-surface interactions in lubrication. Lubrication Science 1989; 2(1): 3- 23. https: / / doi.org/ 10.1002/ ls.3010020102. [8] DIN 51819-3: 2016-12, Prüfung von Schmierstoffen - Mechanisch-dynamische Prüfung auf dem Wälzlager-