High Temperature Tribological Properties of Polybenzimidazole (PBI)

All data and results are available upon requests by the Abstract: Polybenzimidazole (PBI) is a high performance polymer that can potentially replace metal components 9 in some high temperature conditions where lubrication is challenging or impossible. Yet most characterisations 10 so far have been conducted at relatively low temperatures. In this work, the tribological properties of PBI were 11 examined with a steel ball-PBI disc contact at 280  C under high load and high sliding speed conditions. The dry 12 friction coefficient is relatively low and decreases modestly with increasing applied load. Surface analysis shows 13 that PBI transfer layers are responsible for the low friction observed. In-situ contact temperature measurements 14 were performed to provide for the first time direct links between the morphology and distribution of the transfer 15 layer, and the temperature distribution in the contact. The results show that high pressure and high 16 temperature in heavily loaded contacts promote the removal and the subsequent regeneration of a transfer 17 layer, resulting in a very thin transfer layer on the steel counterface. FeOOH is formed in the contact at high 18 loads, instead of Fe 2 O 3 . This may affect the adhesion between PBI and the counterface and thus influence the 19 transfer layer formation process. To control PBI wear, contact temperature management will be crucial.

Polybenzimidazole (PBI) [1], on the other hand, has a glass transition of 427 o C, which makes it a promising 30 candidate for high temperature conditions. It has recently regained potential for tribological applications due 31 to a reduction in manufacturing cost.

32
Apart from an increased glass transition temperature, PBI also has a higher hardness and modulus than other 33 HPPs. As a results, it has better scratch resistance against a stainless steel indenter at room temperature than 34 polyetheretherketone (PEEK), PI and poly-para-phenylene (PPP) [2]. Under room conditions, PBI also exhibits 35 better tribological behaviour than PEEK and PAI when it is rubbed against a steel counter [3]. Sharma et al. [4] 36 investigated the behaviour of PBI under a range of loads and temperatures (100 -200 o C) and found the 37 coefficient of friction to be fairly independent of load until a limiting pressure-velocity (PV) value is reached, 38 after which the coefficient of friction drops. The PV value is the product of average contact pressure (P) and 39 sliding speed (V) and is often used as a measure of the amount of frictional heat generated during rubbing [5-

50
and Unal et al. [13,14] observed a decrease in coefficient of friction and wear rate with increasing load at room 51 temperature. The effect of bulk temperatures up to 300 o C was investigated for PI [13,15]. When the bulk 52 temperature of PI reaches 200 o C, a significant increase in wear rate and a reduction of the friction coefficient 53 are observed. Lancaster [16] observed that the wear rate for amorphous polymers is at its lowest just before 54 the glass transition and then sharply increases. The severe wear produced above a critical temperature is 55 associated with adhesive wear at the interface [17]. Good adhesion of the polymer to the countersurface 56 favours the formation of a transfer layer [18,19]. However, it is not always an assurance of low wear of the 57 polymer [18]. These results highlight that both load and temperature strongly impact on the tribological 58 properties of polymers, which is likely due to their effects on the process of material transfer and the properties 59 of the transferred material.

60
The self-lubricating ability of polymers is frequently attributed to the formation of a so-called transfer layer when 61 polymers are rubbed against steel countersurfaces [20]. However, the concept of a 'transfer layer' is ill-defined 62 and is usually characterised in term of thickness, coverage and homogeneity. An efficient transfer layer is 63 commonly described as a "thin," "uniform," and "stable" layer of material that has been transferred from one 64 surface to another by adhesive wear [21]. Transfer layer formation is often related to thermal softening of the 65 polymer and polymer molecular structure (alignment) [22]. The connection between the occurrence of a 66 transfer layer and wear of the polymer in a polymer-steel system has been widely studied [ Table 1. PBI discs and balls were wiped with isopropanol and dried in an oven at 150°C for at least 2 101 days [37]. Steel balls were cleaned using toluene in a sonication bath for fifteen minutes followed by sonication 102 in isopropanol (IPA) for another five minutes and were dried using a dry cloth.

103
Bare sapphire discs and aluminium coated sapphire discs were used for the contact temperature measurements.

104
Sapphire discs were cleaned using isopropanol and sonicated for 15 min. Aluminium coated sapphire discs were 105 wiped with isopropanol.

110
A sphere-on-flat geometry was used where a stationary ball was loaded against a rotating disc using a dead 111 weight. Frictional force was obtained by the amount of deflection of a flexure arm attached to the ball measured 112 by a displacement transducer. A heating element located around the disc heated the sample to the pre-set 113 temperature. The test chamber was thermally insulated to minimise temperature fluctuations and heat losses.

114
The surface temperature of the disc was 280°C as monitored by thermocouples located below and above the 115 disc. In this study, rotating PBI discs were rubbed against stationary steel balls under dry sliding conditions, i.e. 116 no lubricant was used. The applied normal loads ranged from 3 to 12 N and the sliding speed was fixed at 2 m·s -117 1 . While the usefulness of PV values may be questioned, and PV values are ill-defined for a sphere-on-flat 118 geometry used in our tests, it was calculated based on the initial average pressure and ranged between 75 and 119 119 MPams -1 . As rubbing progressed, a steady state was reached where the coefficient of friction plateaued.

120
The steady state PV values (40 -80 MPams -1 ) was lower than the initial PV values as contact area increased 121 due to wear of the disc and the ball. Note that these values exceed the PV values commonly studied, which 122 typically are about 3 -7 MPams -1 [4]. This means that in this work, the PBI specimens were exposed to very 123 severe operating conditions. Each test lasted for a total of 30 minutes, or 3600 metres sliding distance. The 124 coefficient of friction changed during the initial running-in period and the duration of this running-in period 125 varied among samples. This is due to variations in surface finishes among different samples, but it has no effect 126 on the steady state coefficient of friction. After the initial running in period, the coefficient of friction stabilised 127 when the steady state regime was reached. The reported coefficients of friction in this work are those measured 128 during this steady state and were similar among repeated experiments. Friction tests were also conducted with 129 PBI disc-PBI ball contacts and results compared to those obtained with PBI disc-steel ball contacts. Flattened regions (see Figure 1(h) -1(j)), called wear scars, with a diameter of approximately a few hundreds 142 m, developed on the steel balls where they rubbed against the PBI disc. It was observed that the wear scar 143 was covered by a thin polymeric layer (< 1 m) transferred from the PBI disc. This transfer layer is seen in Figure   144 1(h) -1(j) as dark grey material. The wear volume of each steel ball was obtained from the surface topographic 145 image of the wear scar and its surroundings. The raw images were background corrected with the nominal 146 profiles (6 mm diameter) of the steel ball, as such the wear volume of the ball appears as a negative volume.

147
The morphology of the transfer layer on the steel ball wear scars was obtained by flattening raw height images 148 of wear scars with straight line background correction.

149
Worn steel ball and PBI disc surfaces were also examined using the backscattered electron (BSE) mode of a 150 scanning electron microscope (SEM) at 10 kV energy (Hitachi S-3400N, Hitachi High-Technologies, Japan). The 151 energy dispersive X-ray (EDX) mode of the SEM and X-ray Photoelectron Spectroscopy (XPS) (Thermo Scientific 152 K-Alpha+ X-ray photoelectron spectrometer) were used to obtain chemical information of pristine PBI discs, The contact temperature is defined as the temperature at the rubbing interface, which can be substantially 160 different from the ambient temperature of the test due to frictional heating. To qualitatively show the general 161 trend of temperature rise with test conditions and possibly heat distributions in tribological contacts involving 162 PBI, in-situ contact temperature measurements were performed with Infrared (IR) thermography. To implement 163 IR thermography, the disc needs to be IR transparent. As a result, the steel ball-PBI disc contact for friction 164 measurements cannot be employed. Taking advantage that sapphire fully transmits IR radiation in the 165 wavelength range of 3 -5 µm, the test configuration for these temperature measurements was changed to a 166 stationary PBI ball against a rotating sapphire disc. It needs to be noted that, due to differences in thermal 167 conductivity, the actual heat profiles in this stationary PBI ball-rotating sapphire disc contact is likely different 168 from that of the stationary steel ball-rotating PBI disc contact used in the friction tests.

169
To measure the temperature increase in the contact, a stationary PBI ball was loaded against a rotating sapphire 170 disc from the bottom with an EHL rig (manufactured by PCS instrument). An infrared camera (X6540SC, FLIR) 171 was placed above the contact to capture IR irradiation emitted during rubbing. The camera has a 320 × 256 172 focal plane array. With a 5× lens, it has a lateral resolution of 6.3 µm. During rubbing, IR radiation came from

187
With a quicker drop in contact pressure (due to wear) and higher contact temperature as compared to a PBI-188 steel system, it is not surprising that the friction of a PBI-PBI system is lower.  Table 1). This means that 40 MPa is a reasonable value 207 for the failure stress as the temperature increase in the contact due to frictional heating may be substantial 208 in our tests. The contact temperature rise is further discussed in section 3.3.2.

209
The wear rate of PBI discs (solid squares) and steel balls (open circles) are presented in Figure 2  procedure. The isopropanol layer will have no effect on the friction results as it will be remove during rubbing.

Figure 3 (a) and (b) Morphology and (c) and (d) thickness profiles of transfer layers on wear scars of steel balls. The arrow shows the sliding direction. Area around the scars appear lower in (a) and (b) due to the curvature of the ball. The dashed lines show where the thickness profiles are obtained. The insert in (c) shows the profile of polymeric accumulations at the outlets of the contacts.
Based on the SEM micrographs of steel ball wear scars, two regions can be identified (see Figure 4). This is best suggest that the layer is not uniform and is thin (an average thickness of about 200 nm). In Figure 2, it was 285 shown that the friction is independent of the applied load, whilst the thickness of the developed transfer layer 286 is highly dependent on load. The results here thus suggest that the existence of the transfer layer, rather than 287 its thickness, governs the observed friction.

288
The iron oxide found on the cleaned steel ball is -Fe2O3 (530 eV, Figure 5(c)). This is not observed in the EDX 289 analysis (Figure 4) probably because the oxide layer is very thin (a few nm, confirmed by ellipsometry 290 measurements, results not shown). The -Fe2O3 peak is also clearly identified in the spectra of the wear scars 291 produced at = 3 and 5 N ( Figure 5(c)). The fact that oxides are detected in the wear scars by both XPS and 292 EDX suggests that firstly the conditions in the rubbing steel ball-PBI disc contacts promote the formation of 293 oxides. Secondly, the transfer layer is locally patchy, i.e. oxides underneath the layer are exposed and interact 294 with the X-ray beam directly; or iron oxides are dislodged from the steel ball and debris was embedded into the 295 transfer layer. . While PBI discs used in this study have been dried, it is likely that only water 301 locked in PBI at and near the disc surface was removed. As the load increases, the PBI disc is increasingly worn, 302 such that PBI originally located deep below the surface (see Figure 1 where the depth of the centre of disc wear 303 track is about 7 m), which might still hold a relatively large amount of water, is exposed. This newly exposed 304 PBI can interact with the freshly exposed steel of the wear scar and as a result -FeOOH may be formed. The 305 results also suggest that rubbing may have promoted such a reaction. However, more work is required to 306 understand how the rubbing process gives rise to the formation of various oxides. and heterogeneous polymer transfer layer exists on the wear scar (see Figure 7(e)). Additionally, the polymer 375 experiences a very high wear rate and a large amount of debris is produced.

376
A consequence of rubbing, likely due to the elevated contact temperature, is the increased growth of oxides on 377 the steel surface (see Figure 4). The high stress, high temperature region in the contact has the thinnest oxide 378 as they are rapidly removed. Note the oxides of the steel ball change from Fe2O3 in low load conditions to 379 FeOOH in high load conditions which may impact on the potential formation of a transfer layer.

405
There is a direct link between the temperature distribution inside the contact and the local formation of a 406 polymer transfer layer. The morphology of the transfer layer is heterogeneous within the contact and is load 407 dependent. At low load and low contact temperature conditions, a relatively thick PBI transfer layer rapidly 408 develops on the steel surface. At high load and high contact temperature conditions, the majority of the contact 409 is covered by a very thin polymeric transfer layer due to the constant removal and regeneration of the transfer 410 layer, giving rise to a high wear rate of the PBI. XPS results suggest that at low and high load conditions different 411 iron oxides develop, -Fe2O3 and -FeOOH respectively. The change in surface chemistry may also contribute 412 to differences in transfer film morphology and wear rate. The insensitivity of the measured friction in the steel-413 PBI contact to the applied load suggests that it is the existence rather than the morphology of the transfer layer 414 that controls the friction.

415
For practical applications, one should be aware that, while PBI-steel contacts show low friction at high 416 temperature conditions, under more severe loading (pressure and/or sliding velocity) conditions PBI can 417 experience substantial wear. This is due to both high applied shear stress as well as elevated contact 418 temperature from frictional heating. While the temperature rise does not necessary reach the glass transition 419 or melting point of the polymer, a relatively small temperature rise may sufficiently lower the shear strength of 420 the polymer strength. To mitigate such severe wear rates and the related failure of mechanical components, 421 the implementation of an effective heat management strategy will be crucial to ensure that polymeric 422 transferred materials on steel are not removed from the steel-PBI contact.