Report

Mechanisms of antiwear tribofilm growth revealed in situ by single-asperity sliding contacts

See allHide authors and affiliations

Science  03 Apr 2015:
Vol. 348, Issue 6230, pp. 102-106
DOI: 10.1126/science.1258788

Additive explanation for anti-wear

Additives in oil are vital for protecting engines from wear by forming films at sliding interfaces. Zinc dialkydithiophosphate (ZDDP) has been used for decades to reduce engine wear. Now there is a strong incentive for finding a replacement for ZDDP: Its breakdown products shorten catalytic converter lifetime. Gosvami et al. examined exactly how ZDDP produces an anti-wear film under high stress or elevated temperature (see the Perspective by Schwarz). Understanding these mechanisms will help in the development of higher-performance and more effective additives.

Science, this issue p. 102; see also p. 40

Abstract

Zinc dialkyldithiophosphates (ZDDPs) form antiwear tribofilms at sliding interfaces and are widely used as additives in automotive lubricants. The mechanisms governing the tribofilm growth are not well understood, which limits the development of replacements that offer better performance and are less likely to degrade automobile catalytic converters over time. Using atomic force microscopy in ZDDP-containing lubricant base stock at elevated temperatures, we monitored the growth and properties of the tribofilms in situ in well-defined single-asperity sliding nanocontacts. Surface-based nucleation, growth, and thickness saturation of patchy tribofilms were observed. The growth rate increased exponentially with either applied compressive stress or temperature, consistent with a thermally activated, stress-assisted reaction rate model. Although some models rely on the presence of iron to catalyze tribofilm growth, the films grew regardless of the presence of iron on either the tip or substrate, highlighting the critical role of stress and thermal activation.

Additives are crucial components of lubricants used in a wide range of tribological applications, including vehicles, turbines, and manufacturing equipment (1). Antiwear additives and friction modifiers can extend many industrial and automotive application lifetimes by orders of magnitude, resulting in considerable energy and material savings. One of the most crucial modern antiwear additives is zinc dialkyldithiophosphate (ZDDP), chemical formula Zn[S2P(OR)2]2, where R is an alkyl or aryl group (2, 3) (fig. S2). Extensive macroscopic studies have shown that ZDDP molecules decompose at rubbing interfaces (4, 5) and form protective surface-bonded tribofilms that reduce wear by minimizing metal-to-metal contact of steel and iron (3) and other material pairs (6, 7). ZDDP-derived tribofilms consist of rough, patchy, pad-like features that are composed of pyro- or orthophosphate glasses in the bulk, with an outer nanoscale layer of zinc polyphosphates and a sulfur-rich layer near the metal surface (3). However, the tribochemical film growth pathways are not established, and the factors that determine the film morphology and thickness (generally 50 to 150 nm) are unknown (3). Furthermore, ZDDP’s effectiveness as an antiwear additive for advanced engine materials is not yet clear. For low-weight materials such as Al- and Mg-based alloys, ZDDP forms robust tribofilms primarily on load-bearing inclusions but not on surrounding softer matrices (6, 8). Although ZDDP tribofilms can be formed between other nonferrous material pairs [e.g., low-friction diamond-like carbon (DLC) films], they are often less durable than those formed when steel or iron is present, for reasons not yet understood (9, 10). ZDDP often increases frictional losses (3) and produces Zn-, P-, and S-containing compounds in automobile engine exhaust, thus reducing catalytic converter efficiency and lifetime (1, 3, 11, 12). However, despite decades of research, no suitable substitute for ZDDP has yet been found (12), and research efforts have sought to identify the beneficial mechanisms underlying the growth and antiwear properties of ZDDP-derived tribofilms.

Various macroscopic methods have been developed to produce ZDDP tribofilms (13, 14), and the resulting films have been studied by many ex situ mechanical and chemical approaches (3, 15) and atomistic simulations (8). It is widely assumed that the tribofilm acts as a protective layer that is continually replenished, reducing metal-to-metal contact (3). Some studies indicate that antiwear properties result from ZDDP’s ability to reduce peroxides in the base stock, thereby preventing corrosion (16, 17). One model explaining ZDDP tribofilm formation on steel is based on hard and soft acid-base (HSAB) reactions (18), which require the exchange of Zn2+ and Fe3+ cations between the ZDDP and iron oxide wear particles, respectively, where the latter are digested within the tribofilm (19). Direct experimental evidence for this model is lacking (3), nor does the model explain tribofilm formation on nonferrous surfaces (6, 7). In contrast, Mosey et al.’s first-principles atomistic simulations proposed that tribofilm formation results from contact pressure–induced cross-linking of zinc phosphate molecules (8), which are a thermal or catalytic decomposition product of ZDDP (11, 15).

There is no general consensus on the growth mechanism, and no models conclusively explain either the tribofilm patchiness or why the film thickness is limited. All prior experiments have been conducted for macroscopic, multi-asperity contacts (specific asperity contact areas and pressures thus being unknown) that are then analyzed post mortem and ex situ, often after extracting the sample from base stock, which may alter the tribofilm (20). Although macroscopic in situ studies of zinc polyphosphates under static compression (21, 22) have shown irreversible loss of crystallinity and little increase in polymerization with increased pressure, these studies do not involve dynamic sliding. In situ single-asperity sliding studies have several advantages: Contact loads and geometries can be controlled and quantified; local tribofilm properties such as morphological evolution with nanometer resolution, tribofilm volume, friction, adhesion, and wear can be measured concurrently; and results can be compared with atomistic simulations (8).

We conducted in situ single-asperity studies with an atomic force microscope (AFM). The AFM tip was slid against an Fe-coated or uncoated Si substrate at temperatures up to 140°C while immersed in ZDDP-containing base stock (23) to dynamically generate the tribofilm (fig. S3). Low-load (10 to 20 nN) contact-mode imaging revealed a soft, weakly bound thermal film, formed without prior sliding, that was easily removed by sliding with a load of 100 nN (fig. S4). This well-known “thermal film” is formed from adsorbed decomposition products of ZDDP (15, 24). Typical thermal film thicknesses of ~10 nm were obtained after ~1 hour of heating the base stock bath, but the thickness can vary with the age of the oil and heating time (24). After removing the thermal film with the tip, sliding was continued within the same region with a higher normal load to induce the growth of the tribofilm.

The morphological evolution of the tribochemical products with increasing sliding cycles (one sliding cycle = one 1 μm × 1 μm image) revealed randomly located nucleation sites and subsequent growth of the sliding-induced tribofilm (Fig. 1). The tribofilm grew vertically and laterally (only within the region scanned at higher normal load) with further sliding, leading to a rough surface (movie S1). The total film volume increased linearly with sliding time during the first ~1200 cycles (Fig. 1, inset), indicating a zero-order reaction (23). The growth rate then increased rapidly, fitting well to a power-law function corresponding to an nth-order reaction with n = 0.22 (fig. S7), which indicates a complex reaction pathway. The observed growth of a patchy film matches well with macroscopic results (25, 26). Such macroscopic studies cannot make clear whether the patchiness resulted from multiple asperities applying a range of pressures at different contact points, or by other means. Because the loads and contact geometry are well controlled in our single-asperity experiments, the heterogeneity is apparently intrinsic to the growth mechanism. This may indicate that nucleation is sensitively dependent on surface heterogeneities such as defects or roughness and/or that there are instabilities in the growth (perhaps affected by atomic-level stress variations within the single-asperity contact) immediately after randomly occurring nucleation events.

Fig. 1 Morphology and volumetric growth of tribofilm.

Tribofilm volume (mean ± SD) versus sliding cycles, with linear and power-law fits to the initial and subsequent growth regimes, respectively. Inset shows a zoom-in of the initial growth period. Around the perimeter, clockwise from upper left: periodically acquired 2 μm × 2 μm AFM images of an iron oxide surface using a DLC-coated silicon AFM tip immersed in ZDDP-containing base stock, acquired at a nonperturbative load of 20.0 ± 0.1 nN. Below each image is the number of previously acquired 1 μm × 1 μm scans (“sliding cycles”) at a load of 340 ± 2 nN (4.2 ± 0.5 GPa). The images demonstrate progressive tribofilm growth where the higher load was applied.

At these stresses (~4 GPa), the tribofilm growth rate was low, and the volume rarely reached a limiting value within the time frame of our experiments (~10 hours), whereas growth typically saturates within a few hours in macroscopic experiments (27). This discrepancy may be due to differences in sliding speeds (~80 μm/s for these AFM experiments versus millimeters to meters per second for macroscopic tests) or contact areas (on the order of 10 to 100 nm2 in AFM versus ~109 nm2 for macroscopic tests), both of which reduce the area per unit time covered by AFM. The far larger amount of fluid exchange and the multi-asperity nature of the macroscopic contacts will also affect growth. Fortunately, AFM experiments performed at higher normal stresses (~6.5 GPa) enhanced the growth rate, and films reached a limiting height after prolonged sliding. We observed tribofilm wear once it reached a thickness of ~30 to 40 nm, preventing further growth (fig. S5). At this thickness, there was no observable contrast in friction between the tribofilm and the surrounding substrate. However, before the tribofilm growth had saturated, a transient increase in friction was observed (fig. S10). Further study is required to determine whether this effect is due to changes in tribofilm adhesion, modulus, roughness, or interfacial shear strength. However, the increase seen is consistent with macroscopic studies that report transient increases in friction for ZDDP-infused base stocks (28).

Within the subnanometer vertical resolution limits of our instrument, the tribofilms formed without any observable wear of the iron oxide substrate. The proposed HSAB mechanism requires substantial plastic deformation and wear of the substrate (18). Considering the nanoscale dimensions of the nucleation centers observed in our experiments, the possibilities of cation exchange and digestion of atomic-scale debris via molecular-level mechanical mixing cannot be excluded. However, such a mechanism does not explain observations of similar macroscopic ZDDP tribofilms on other substrates such as DLC and silicon (6, 7, 10, 29). We also observed formation of tribofilms in AFM experiments using Si substrates with no Fe present (fig. S6); the growth rate and morphology of these films were indistinguishable from those we formed on Fe substrates.

Our results also provide direct evidence that the tribofilm is not a product of sliding-induced transformation of the adsorbed thermal film, because growth occurred in regions where the thermal film was completely removed (fig. S4). Rather, these results indicate that tribofilm growth is fed by molecular species from solution into the contact zone, where tribochemical reactions occur.

Tribofilm growth rate and morphology were investigated as a function of normal load, which is directly related to the initial contact pressure (contact pressure at a fixed load will decrease as the compliant tribofilm’s thickness increases). Multiple tribofilms were generated by sliding the AFM probe for 2000 sliding cycles at 100°C for a range of fixed loads (i.e., different initial contact pressures) (Fig. 2). Tribofilm morphologies and volumes clearly reveal that growth is strongly affected by contact pressure. Beyond 5.2 ± 0.6 GPa, tribofilm deformation and pile-up was observed and the growth rate was stabilized, indicating concurrent tribofilm generation and removal. This agrees with macroscopic observations and directly demonstrates the sacrificial property of ZDDP tribofilms beyond a critical thickness and contact pressure at the nanoscale (30).

Fig. 2 Tribofilm volumetric growth rate dependence on contact pressure.

Tribofilm growth rate is exponential at low contact pressures (data are means ± SD). Further growth is inhibited above ~5 GPa as the tip wears away newly deposited material. The 2 μm × 2 μm topographic contact-mode AFM images shown were acquired at a nonperturbative load after generating tribofilms in the central 1.0 μm × 0.5 μm regions at various contact pressures.

The stress-dependent growth rate Γgrowth rate (nm3/s) fits well to a stress-activated Arrhenius model (Fig. 2):Embedded Image (1)where the prefactor Γ0 depends on the effective attempt frequency and the molar volume of the growth species (23), ΔGact is the free activation energy of the rate-limiting reaction in the growth process, kB is Boltzmann’s constant, and T is absolute temperature. The fit assumes that ΔGact is influenced by stress according to Embedded Image (2)where ΔUact is the internal activation energy (i.e., the energy barrier in the absence of stress), σ is the mean value of the stress component affecting the activation barrier (assumed to be the compressive contact pressure), and ΔVact is the activation volume (31). The good fit suggests that tribofilm formation is an activated process (31). We find ΔUact = 0.8 ± 0.2 eV and ΔVact = 3.8 ± 1.2 Å3, consistent with parameters for single atomic bond breaking or formation processes. The stress dependence suggests that the observed heterogeneous nucleation (Fig. 1) could result from atomic-scale surface roughness, which would lead to varying contact areas and stresses for a given normal load; thus, the energy barrier for the relevant tribochemical reaction would be lower where the local stress is higher.

Experiments performed as a function of temperature provide further support for an activated tribochemical reaction mechanism (Fig. 3). The volumetric growth rate of tribofilms generated by 5000 sliding cycles at ~4.4 GPa depended exponentially upon temperature. From fitting Eq. 1, we obtain ΔGact = 0.62 ± 0.10 eV. Using the initial contact pressure determined from AFM force-distance data and using ΔVact from data in Fig. 2, we obtain ΔUact = 0.74 ± 0.10 eV using Eq. 2, in excellent agreement with the value obtained from the stress-dependent data. This confirms the applicability of reaction rate theory by using independent stress- and temperature-dependent measurements. Our results provide a robust basis to support the idea that tribofilm growth occurs via stress-activated and thermally activated tribochemical reactions, in contrast to previous empirical approaches (25). Our data do not provide any direct support for the HSAB model (18), which asserts that tribofilms can form even at contact pressures as low as 1 MPa, where the entropy of mixing, not stress and temperature, drives the reaction (19). The data are consistent with molecular dynamics simulations showing that tribofilm formation can be driven by contact pressure (8). However, we note that the simulation studies were performed on simpler zinc phosphate systems (with no sulfur), and effects of sliding were not investigated. Here, we used sliding experiments to show directly the role of pressure and temperature in forming tribofilms from ZDDP itself.

Fig. 3 Tribofilm volumetric growth rate dependence on temperature.

Growth rate (mean ± SD) versus temperature data fitted with an exponential function (Eq. 1). The 2 μm × 2 μm topographic contact-mode AFM images shown were acquired at a nonperturbative load after generating tribofilms in the central 1.0 μm × 0.5 μm regions at 80°, 100°, 120°, and 140°C at an initial contact pressure of ~4.4 GPa.

Ex situ chemical analysis using energy-dispersive spectroscopy (EDS) and Auger electron spectroscopy (AES) identified the tribofilms’ elemental composition. Point spectroscopy and elemental mapping by EDS (Fig. 4A) revealed clear signatures of Zn, S, and P inside the tribofilm, as expected from ZDDP-derived products (32). Much smaller peaks corresponding to P, S, and Zn were observed outside the tribofilm region; these are attributable to the thin (~10 nm), weakly bound thermal film (a large fraction of which is likely dissolved during solvent rinsing before the EDS measurements). Elemental maps (Fig. 4A) reveal uniform distributions of P, S, and Zn inside the tribofilm. The distribution of Fe was uniform and indistinguishable between regions inside and outside the tribofilm, further showing that no measurable wear or displacement of Fe was involved in tribofilm formation. AES, more surface-sensitive than EDS, revealed Zn, S, and P in the tribofilm region only (Fig. 4B). Far more Fe was seen outside the tribofilm, indicating that little or no Fe is mixed into the outermost regions of the tribofilm.

Fig. 4 Ex situ chemical characterization.

(A) EDS point spectra (estimated sampling depth of ~1 μm) acquired for regions (a) inside and (b) outside the tribofilm (i.e., for the portion of the substrate covered with the thermal film). (c) Secondary electron image of the 10 μm × 5.0 μm tribofilm. Corresponding elemental maps are shown for (d) Fe, (e) Zn, (f) P, and (g) S. (B) (h) Optical and (i) secondary electron image of a 10 μm × 5.0 μm tribofilm obtained by scanning AES. (j) AES spectra for the tribofilm and the substrate (estimated sampling depth ~3 nm).

The observed reduction of tribofilm robustness with increased thickness is consistent with reports that the modulus and hardness of macroscopic ZDDP tribofilms decrease with thickness (20). Furthermore, the contact pressure dependence of tribofilm formation reported here (Fig. 2) can explain the reported gradient in composition, structure, and mechanical properties of ZDDP tribofilms. Specifically, because the tribofilm has a lower modulus than the substrate, the contact stress at constant load decreases as the tribofilm thickens. This in turn reduces the amount of stress-induced polymerization and other reactions that produce the tribofilm, resulting in a weaker, more compliant, graded structure and a further reduction in contact pressure. This feedback-driven self-limiting growth mechanism hinges on the stress dependence of the thermally activated growth that we have uncovered (Fig. 2).

Our results show that the ZDDP tribofilm growth rate increases exponentially with applied pressure and temperature under single-asperity contact, in very good agreement with stress-assisted reaction rate theory; the kinetic parameters are consistent with a covalent bond reaction pathway. Repeated sliding at sufficiently high loads leads to abundant tribochemical reactions and the associated nucleation and growth of robust tribofilms with a pad-like structure similar to macroscopically generated films. The tribofilm is not a product of the weakly adsorbed thermal film, but instead is generated from molecular species fed continuously into the contact zone. We confirmed the sacrificial nature of the tribofilm beyond a threshold thickness, indicating that layers grown at lower applied pressures are weaker. The observations imply that ZDDP’s antiwear behavior derives from mechanical protection provided by the tribofilm, as opposed to corrosion inhibition. We suggest that this in situ approach can be directly applied to understand further molecular-level tribochemical phenomena and functionality, such as the behavior of other important lubricant additives (e.g., friction modifiers) or films formed in vapor-phase lubrication (33).

Supplementary Materials

www.sciencemag.org/content/348/6230/102/suppl/DC1

Materials and Methods

Supplementary Text

Figs. S1 to S10

Movie S1

References (3442)

References and Notes

  1. See supplementary materials on Science Online.
  2. Acknowledgments: Supported by the Marie Curie International Outgoing Fellowship for Career Development within the 7th European Community Framework Programme under contract PIOF-GA-2012-328776 (F.M.), the Nanotechnology Institute through the Ben Franklin Technology Development Authority, the University of Pennsylvania School of Engineering and Applied Sciences, and NSF grant CMMI-1200019. We thank the University of Pennsylvania Nano/Bio Interface Center Facilities and the Nanoscale Characterization Facility in the Singh Center for Nanotechnology for use of facilities, Evans Analytical Group (East Windsor, NJ) for AES measurements, ExxonMobil’s Corporate Strategic Research laboratory for materials and financial support, Q. Tam for Matlab analysis, T. D. B. Jacobs for transmission electron microscopy analysis of AFM probes, and A. Jackson for helpful discussions.
View Abstract

Stay Connected to Science

Navigate This Article