Tribological properties of boride based thermal diffusion coatings

Abstract

Engineering components, e.g. tubing systems for the down-hole applications in the oil and gas industry (in particular, sucker rod pumps, progressing cavity pumps and some other components of the artificial lifting systems), as well as numerous valves and seats, bearings, gears and plungers, require protection against friction and sliding abrasion service conditions. The hard boride based coatings on steels and alloys obtained through the thermal diffusion process have a high potential for these severe application conditions over many other types of coatings as they can be obtained on the entire working surfaces of large size and complex shape products. Intensive tribological studies of the iron boride based coatings on carbon steel obtained at Endurance Technologies Inc. have been conducted using the Cameron–Plint testing unit (reciprocating sliding of the metallic rod under the load over a flat surface of the coated samples). The friction wear loss, friction coefficient and structural changes of the coatings have been studied in dry and lubricating (water–oil) friction conditions, which simulate actual application conditions. It was demonstrated that the obtained boride coatings have the friction loss significantly smaller than untreated steel (e.g. ∼10–30 times in the dry conditions and at least 5 times in the lubricating conditions) with no peeling and flaking-off. The friction coefficients of the boride coatings are steady over the test duration. The influence of the thickness on the boride coatings performance is demonstrated. The encouraging results are explained by the specific coating structure of the hard coating obtained through the thermal diffusion process and the thin ‘tribofilm’ formed during a friction mode.

Keywords

Coatings Iron boride Friction Coefficient of friction Thermal diffusion

Introduction

Friction issues and associated wear friction losses take place in numerous industrial processes where engineering components are used.^1–5 For instance, tubing systems for the down-hole applications in the oil and gas industry (in particular, sucker rod pumps, progressing cavity pumps and some other components of the artificial lifting systems), as well as numerous valves and seats, bearings, gear systems and plungers, require protection against the damage associated with friction and sliding abrasion service conditions. In these situations, friction may occur in dry conditions and in lubricating conditions when oil and, more often, oil and water may be between the friction pairs. As an example of the regular failure in oilfield operations, wear occurs in sucker rod pumps applications due to reciprocating, up and down, movement of the rod and couplings against tubing in the pump completion, as well as between tubing and casing. In particular, wear becomes higher when the coupling-to-tubing slap is a result of the motion with some angle. Rotational movement of the rod and harmonic oscillation also cause wear of the rod, couplings and tubing. Produced fluids, which are pretty corrosive due to the presence chlorides, sulphates, H₂S, bacteria and some specially introduced chemical and lubricants, and which often contain abrasive particles, accelerate wear problems significantly. All these problems, alone or in combination, lead to the hole-in-tubing failures. Many different options have been tried to resolve these issues,^6–8 but practically there were no real economical solutions until now.

In industry, for the friction and sliding abrasion conditions, metallic (steels and hard alloys), advanced ceramics and composites and some polymeric materials (elastomers) can be used.^1–5 However, the elastomers, which are used in some cases, cannot withstand elevated temperatures involved in many industrial applications (e.g. in superheated steam assisting processes); in addition, they become brittle after certain service time and also have lower corrosion resistance. Monolithic advanced ceramics and composites are successfully used due to high mechanical properties (first of all, hardness), wear and corrosion resistance;^1–3,9–16 however, the components from these materials can be produced only with rather small sizes due to manufacturing limitations. Due to the requirements and features dealt with the design and dimensions of the industrial components, e.g. long tubular components (which often have a length over 1 m) and/or complex-shape components, tight dimensional tolerance, as well as the necessity to maintain ductility of the material, steels and alloys are mostly used in the industry. However, they do not have adequate wear resistance, including resistance to the friction action, as well as resistance to corrosive environments. In order to minimise all these problems, surface engineering, particularly, the advanced coatings on the components made from steels and alloys have to be used. Different surface engineering, e.g. coating, methods are used in industry; they include painting, dipping, cold and thermal spraying, chemical and electrochemical methods (anodising, electroplating, electroless plating), plasma-assisted technology, physical vapour deposition and chemical vapour deposition (CVD) and some others.^{1–5,17–24} However, the majority of these methods have serious limitations due to the difficulties to apply coating into long size (e.g. inner surface of small diameter tubing with several metres length) and complex shape components. Many other coating methods experience spalling and/or delamination or fast destruction of thin (with micron sizes) coating layers with low hardness, i.e. lower performance, under severe wear, corrosion, temperature conditions or microcracking due to the substrate and coating thermal expansion mismatch, which also results in the fast degradation of products under severe conditions. To the contrary, some surface engineering methods based on the CVD principles may be employed to overcome these serious application/service and manufacturing challenges.²⁴

A thermal diffusion coating method which is based on the CVD principles is successfully used for the processing of the protective coatings for the mentioned applications and for large-size components. This method involves the deposition of the selected element(s) in an active vapour form onto the preheated substrate with consequent chemical modification of the substrate metal surface through the formation of new hard inorganic substances.^18,19,21 It allows effective protection of the entire surface of the metallic components with different shapes and dimensions, including inner surfaces of tubing (over 1 m length) maintaining tight tolerance without machining. Among the options of thermal diffusion methods, boronising of steels and alloys can be used to create hard and chemically inert coatings based on iron borides and some other borides, which can successfully protect metallic components from wear and corrosion at elevated temperatures.^25–29 As known, metal borides have a strong advantage due to their high hardness, chemical inertness and related application properties, e.g. wear and corrosion resistance.³⁰ For example, the iron boride coatings on steels and alloys demonstrated lower abrasion and erosion rates in comparison with carburised and nitrided steels, chromium and electroless nickel coatings and the coatings based on Cr₃C₂, CrB, NiCrBC, TiN, WC and some others obtained by physical vapour deposition, plasma spray, nitriding and CVD methods.^22,29–31 The boride coatings also demonstrated resistance to the action of corrosive environments, such as acids and salts, as well as to high temperature oxidation.^{26,28,32–35} However, despite the needs of the data, which may be important for selection and evaluation of the coatings for the mentioned application conditions, tribological properties of the boride coatings are not studied enough. Although some studies demonstrated the applicability of boronised coatings in friction conditions,^{28,31,32,36–40} they were conducted for different steels and at different testing conditions, therefore their results are hardly comparable. It is not clear, what coating thickness is desirable to reach high performance in the wear, particularly, in friction applications. In some publications,^{31,32,36–40} only rather thin coatings (below 100 μm) are used; however, these thin coatings were tested just because the authors could obtain these thicknesses but they may be not very even and well consolidated.

The thermal diffusion method involves the following processing steps: chemical modification of the substrate metal through the deposition of the gaseous (vapour) phase of the required atoms formed during heat treatment, its diffusion into the structure of the heated substrate material and formation of new hard inorganic phases based on the metal from the substrate and this acting atom, and, finally, diffusion and growth of these new phases, which modify the structure and properties of the base material surface. Diffusion and reaction of the ingredients formed in the gaseous state occur in significantly lower temperatures compared to the ‘conventional’ solid state processes.

In the boronising process, the formed B-contained vapour occurred due to thermal treatment of the specific mix composition used, interacts with a metallic substrate yielding iron borides, and this interaction continues with diffusion of iron borides into the substrate. Formed iron boride Fe₂B reacts with the diffusing B yielding FeB, and FeB diffuses into the substrate and interacts with Fe atoms, which also diffuse from the substrate towards to FeB with the formation of Fe₂B. Thus, diffusion and interaction occurs in two directions The diffusion process greatly contributes to the microstructure development in the reaction zone. The formation of the iron boride thermal diffusion coatings has a parabolic behaviour, which is confirmed by numerous studies.^{25–30,32,37,41,42} Because of this, phase formation and, particularly, case depth (coating thickness), can be managed by the modification of temperature and time.

There are different methods used for the tribological studies of metallic and ceramic materials and coatings. Mostly, they involve the studies of the systems as pin-on-disc, ball-on-disc and block-on-disc configurations,^{1,3,15,36,38,43–45} as well as some others like ring-counter sample.^39,40 However, these methods do not simulate well the application conditions and components configurations required in many oil processing, metallurgical and engineering situations, such as sliding and rolling–sliding conditions, so the data obtained through the mentioned ‘more traditional’ evaluation of friction situations may be not very useful to predict the field behaviour. It is obvious that tribological properties and the features of material destruction occurred at the friction modes are defined, in a high extent, by the testing methods. It is not clear yet, which factors govern wear, particularly, friction resistance of boride based coatings.

The present paper covers the intensive study of tribological properties of boronised steel in comparison with untreated steel widely used in industry for friction applications in the conditions, which simulate actual situations often occurring in oil and gas and similar application conditions. The study has been conducted for the boronised coatings manufactured by Endurance Technologies Inc. (ETI) through the thermal diffusion process. In particular, the behaviour of the coatings with different thicknesses has been compared, which should be important for the appropriate product design and processing. This study is mostly focused on the development and implementation of the products with improved friction resistance for oil processing, e.g. for artificial lift system, different valves and seats and some other complex-shape components of engineering tools and devices, and it has to fill the gap between the laboratory experiments and actual product design and manufacturing.

Experimental

Starting materials and processing

As the substrate material, carbon steel (CS) was used. Although stainless steels and special alloys have higher corrosion resistance than CS, and they are generally considered as more ‘advanced’ materials, their performance in wear applications is on the same level as CS, but they are significantly more expensive. However, the performance of boride coatings in wear and corrosion applications is significantly higher than of bare carbon and stainless steels and alloys.^{25–30,32,34,45–49} Therefore, it is much more beneficial in industry to use boride coatings with high corrosion resistance onto inexpensive CS.

Batch (pack) compositions used for the thermal diffusion boronising process consist of three major ingredients, such as a boron-sourcing material, a process activator (initiator) and inert filler; all ingredients are powders. The ratios of the ingredients of the boronising batch compositions, e.g. B-sourcing material∶activator∶inert filler are basically in the range of (1·5–15)∶(3–10)∶(80–95); however, the types of the ingredients and their particular ratios, are the proprietary of ETI.^41,42 The pack composition is selected in accordance to the required thickness of the coating, demanding properties, as well as in accordance to the substrate material (type of steel or alloy). The selected powder ingredients are mixed, and this homogeneous mix is used to fill a containment vessel or retort with the steel components, which have to be boronised. Among different steels and alloys, which are boronised at ETI, CS with chemical compositions (wt-%) of C 0·15–0·20, Si 0·20–0·23, Mn 0·60–0·75, Cu 0·30–0·35, Ni 0·10–0·15, Cr 0·10–0·16, Mo 0·02–0·05, and Fe as a balance is mainly used in industry as an inexpensive substrate material, and, because of this, it has been selected for boronising and for the actual tribological studies. The surface of the substrate is preliminary cleaned and blasted using the hard powders with certain grits and cleaned to remove the foreign impurities and to create a smooth working surface. The boronising is conducted in the high temperature furnaces at the temperature range of 800–1100°C; the temperature, heating–cooling profile and heat treatment conditions are selected based on the requirements to the product (e.g. coating thickness and properties, etc.). Because the iron boride formation is defined by the diffusion process, i.e. it depends, in a high extent, on temperature and time, these parameters can be varied to obtain the desirable case depth. The boronised products are cleaned up from the remaining spent mix and inspected.

Testing and evaluation

The coating thickness (case depth), structure and morphology were determined using an optical microscope and scanning electron microscope (SEM; JEOL JSM-6610). The metallographic samples were prepared through cutting, grinding and polishing using the Buehler equipment. X-ray diffraction analysis was conducted for the flat test coupons with the coatings prepared by polishing away the surface of the certain thickness of the coupons repeatedly. The thicknesses polished away were selected based on the microscopic studies. For instance, ∼20–25 and ∼75–150 μm of the coatings were removed from the top surface. Special steel holders were used for mounting the samples for X-ray diffraction analysis.

The coatings were also characterised by the Knoop hardness determination at 100 g load (HK0·1) in accordance with ASTM E384-10. The samples for testing were ground and polished using the Buehler equipment. The Knoop hardness was calculated in accordance with formula where P is force (kgf) and d is length of the long diagonal of the replica (mm) made at the indentation.

Surface roughness was determined using a Mitutoyo profilometer SJ301 with pre-programmed Gaussian filter.

Friction testing has been conducted using the Cameron–Plint testing unit TE77 (reciprocating sliding of the metallic rod under the load over a flat surface of the coated samples). This ‘rod-on-flat’ configuration method was selected over some other known methods as simulating rather well the actual friction conditions often occurring in the sucker rods and progressing cavity pumps service situations. The rod specimen was held in the line contact holder, which reciprocates in a sinusoidal manner at 10 Hz. A dead weight load is applied via a loading plate to the roller bearing atop the drive arm. The stationary plate test coupon is supported by flexures, and the friction force is transmitted via the transducer link to the piezoelectric force transducer. During the test, the friction force was monitored. After the designated time, the test stops automatically. The schematic of the Cameron–Plint testing unit and the motion at the testing are demonstrated at Figs. 1 and 2. The wear friction loss was calculated based on the weight difference of the initial and tested samples; the volume loss was calculated using density of the materials involved.

Schematic of Cameron–Plint T77 wear test rig

Diagram of rod-on-flat contact configuration of test samples

The metallic rod was cut from Thompson ‘L’ grade shaft, which is a preheated CS with hardness of HRC65 and which is widely used in oil and gas processing and engineering applications. The specimen rod had 6·35 mm length and 6·35 mm diameter; the surface finish of the rod was better than 0·2 μm, sharp corners of the rod were deburred before testing. The friction wear loss, friction coefficient and structural changes of the coatings have been studied in dry and lubricating (water–oil) friction conditions, which simulate actual application conditions. The ratio between water and oil (SAE30 mineral oil) was selected as ∼75∶25, which also simulated the actual field situations. In order to prevent delamination of water and oil, a small amount (∼1%) of the organic detergent was added to the water–oil mix. During the test, the lubricant was kept agitated by the 10 Hz, 15 mm rod stroke. The amount of lubricant used in each test was controlled at 20 mL to maintain the same fluid dynamic lubrication conditions. The friction testing conditions are presented in Table 1.

Table 1

Friction test parameters

Test parameters	Dry	Lubricating
Load/N	50	200
Friction force conversion factor/N/V	20	50
Stroke peak to peak/mm	15	15
Speed/Hz	10	10
Sliding time/min	30 and 120	180 and 360
Temperature/°C	21	21

Before testing, the rod and test coupons were sonicated in a bath of Stoddard solvent for 5 min and cleaned with a Kimwipe followed by sonication in ethanol for 5 min and cleaning with a Kimwipe. The coupons were then rinsed in ethanol, dried with compressed air and heated at 110°C for 1 h; the cleaned and dried coupons were store in a desiccator until their use. After the testing, the same cleaning procedure was used.

The surface profile traces were taken of each test coupon before and after friction testing using a Taylor Hobson, Form Talysurf Series 2 contacting type profilometer, used with a diamond stylus of radius 2·5 mm, vertical resolution of 16 nm and a spatial resolution of 250 nm. The traces were 10 mm length across the 6·35 mm wear scar in three locations: 2·5 mm from both sides of the ends of the scar and in the centre. Waviness analysis was performed with Least Square Line fit, Gaussian filter, Cutoff (L_c): 0·0025 mm.

The parameters of the friction testing can be found in Table 1. The variation of friction force as a friction of time was monitored, and the remote monitoring system (RMS) filtered signal was recorded by a digital acquisition system. The computer recorded friction force signal (RMS filtered signal, in V) was converted to the friction force F_t according the friction conversion factor (N/V) and then coefficient of friction (COF) was determined by dividing by the normal load F_n applied The morphology of the worn areas of the tested samples was studied using SEM.

Results and discussion

The boronised coating structure has the ‘saw-tooth’ morphology when CSs are used as the substrates (Fig. 3). With regards to case depth determination, it was measured from the top of the coating surface to the middle of the ‘teeth’. Generally, the inner zone (which is closer to the substrate) mostly consists of the Fe₂B phase, and the outer zone mostly consists of the FeB phase. Due to optimisation of the pack composition and process parameters, the Fe₂B phase and the FeB phase have similar ‘saw-tooth’ morphology despite their different crystalline lattice parameters (see Fig. 3). Thus, the FeB phase grows on the Fe₂B repeating the morphology of the Fe₂B phase. Although these phases have different crystallographic structures and different values of coefficients of thermal expansion, the delamination and microcracks between the phases and are not observed, even when the case depth is 250–260 μm, with good bonding between the ‘layers’. The columnar grains of the layers reinforce the coating in general. The mechanical interface between the boride coating and the substrate and the Kirkendall porosity are not observed. The thermal diffusion process provides rather uniform coating for the entire treated surface with no delamination issues. X-ray diffraction analysis of the surfaces obtained by polishing away certain thicknesses confirmed that the top layer mostly consists of the FeB phase with only a few percentage of the Fe₂B phase, while the areas in the middle of the cross-section of the coatings consist of the FeB and Fe₂B (indicated as Fe_0·91B_0·09) phases. The details of the coating formation and coating composition have been described earlier.^41,42

a 6 h; b 12 h; c 15 h

Because the boronising process is based on the thermal diffusion principles, the process can be managed by modification of temperature and time. In this particular case, the process soak time was varied to reach the required coating thickness (Fig. 4). The boronised samples tested had the coating thickness (case depth) of ∼100–115, 200–215 and 250–260 μm denoted as B-100, B-200 and B-250, respectively.

Influence of processing time on the coating thickness (case depth)

Some physical properties of the boronised coatings and the steel are shown in Table 2. The Fe–B coatings of 100–250 μm thickness have hardness in about 10 times higher than uncoated CSs. The obtained results correlate well with other published data related to hardness of the iron boride coatings.^{26–30,32,34,48} It can be specified that the inner zone of the coatings, which mostly consists of the Fe₂B phase has slightly lower hardness values than the outer area of the coatings mostly consisted of the FeB phase (HK0·1 1400–1500 and 1700–1850, respectively). The boronising process promotes surface smoothness; practically, the level of surface roughness (R grade number) is reduced to one grade or more for the boronised steel vs untreated CS.

Table 2

Some parameters of materials tested

Material	Coating thickness/μm	Hardness HK0·1	Roughness/μm – R grade N
CS	…	175–185	7·0–8·0 N9-10
B-100	100–115	1720–1760	3·0–3·5 N8
B-200	200–215	1750–1850	2·5–3·2 N7-8
B-250	250–260	1700–1760	2·8–3·3 N7-8

The samples with worn ‘grooves’ after the friction mode in dry and lubricating conditions are shown in Fig. 5. As expected, the depth of the ‘groove’ after the testing in dry condition is significantly deeper in comparison with testing in lubricating conditions despite shorter testing time and smaller load applied. The comparison of the weight and volume loss of the tested samples (with boronised coatings and uncoated) can be seen in Figs. 6 and 7. It is clear that the uncoated CS has the weight and volume loss significantly higher than the boronised samples, e.g. in 10 times or even greater in dry and more than 5 times in lubricating conditions. This is related to not only consolidated low-porous coating structure and significantly higher hardness of the iron boride (see Table 2), but also to the specific ‘saw-tooth’ morphology of the boronised coatings when the ductile material (steel) supports the coating and that the stress and crack propagation occurring under the friction mode are delayed due to the columnar structure and perpendicular direction of the ‘teeth’ to the substrate. Due to the specific morphology, the adhesion of the iron boride coating is very high without delamination of the coating from the base material. No spalling and flaking-off were observed for the boronised coatings. The high weight loss was also observed for the rods. The most significant weight loss was observed for the rods when the boronised coatings were tested in the dry conditions; this is explained by high hardness of these coatings (that is significantly higher than hardness of the rod material). Comparing the performance of the boronised coatings of different thicknesses, B-200 demonstrated the lowest friction wear loss; B-250 is ranked as next, and B-100 is worse than B-200 and B-250. In the lubricating test conditions, boronised coating B-200 also demonstrated superior performance among the tested samples.

Samples of carbon steel after friction testing in dry (top) and lubricating (bottom) conditions

a weight loss, %; b volume loss, cm³

Because the samples B-200 demonstrated the very good performance in dry and lubricating conditions after relatively short testing time (0·5 and 3 h, respectively), they were also tested for longer time, i.e. for 2 h in dry conditions and for 6 h in lubricating conditions. Considering wear rates for this coating in dry conditions determined at different time periods (Fig. 8), it is higher in the beginning of the testing (during first 30 min), but then, with the test duration, it becomes lower. Thus, wear rate of B-200 is ∼1·4×10⁻³% (weight loss) and ∼0·32×10⁻³ cm³ (volume loss). The rough surface is removed more intensively during the initial testing period, and then the material removal becomes slower. Examination of test data on each specimen revealed that higher wear loss, particularly at the first stage of testing, is associated with rougher specimen surfaces, while less rough specimens of the same boronising treatment resulted in a lower wear rate (compare the wear loss data and the roughness data from Table 2). This is presumably due to the relatively higher local contact stresses in the rough asperities, which lead to easier fracture and faster removal of the high asperities (similarly to the results with other coatings),⁵⁰ although the apparent/macro contact loading conditions were the same.

Friction wear rate for B-200 (weight loss, %), dry test conditions

Considering tested boronised samples, the best performance of the coating with ∼200 μm thickness is probably associated with denser microstructure and a higher level of consistency because all the samples with different case depths have the same level of hardness and similar ‘saw-tooth’ morphology providing good bonding between steel and the coating. The structure of the sample B-100 is not so consistent as B-200 and B-250 with less even case depth at the all surface (is not very ‘mature’). The structures of B-200 and B-250 are well-consolidated with fewer internal pores; this may be dealt with complete diffusion process. However, the structure of B-250, as a thicker one, may have more microcracks due to elevated stress condition between the FeB and Fe₂B phases, and these microdefects may affect negatively the performance at friction modes. The point that coating hardness is not of the highest importance for the friction resistance is confirmed by different values of the friction losses for the boronised samples with different case depths despite the same level of their hardness.

The data of COF for the tested samples (Figs. 9 and 10) showed that boronised samples have the ‘stable’ behaviour with the test duration, i.e. practically no or only very small increase with time; this may be a good indicator of a rather high level of friction resistance. It can be seen that COF at the testing in dry conditions slightly grows at the beginning of the friction test, and may be dealt with the removal of original roughness of the coatings. Then, when the coating became smoother, the behaviour is more stable, and the friction wear rate became lower. This correlates with the data from Fig. 8 indicating different slopes of the graph of the wear loss vs time. As opposed to the stable behaviour of COF of iron boride coatings, COF of CS grows significantly (i.e. about 2 times) during friction, which indicates to a significant wear associated with the damage of the affected surface. The worn material remaining between the surfaces of the test coupon and the reciprocating steel rod promotes further degradation and wear in the case of CS; however, the boronised coating is much more resistant to the action of the worn microparticles. The minimal change of the COF for the boronised samples vs friction time indicates to the ‘tribological equilibrium’ for the used couples attained after the short time, i.e. reasonable adaptation for the existing conditions.

Coefficient of friction change during testing at dry conditions

Coefficient of friction change during testing at lubricating conditions

The studies of the worn surfaces of the tested materials after the friction show that the wear mechanisms for bare steel and coated steel are different. The morphology of the worn surfaces (SEM images) after the friction mode conducted in dry and lubricating conditions can be seen in Figs. 11 and 12. Steel, as a ductile material, is worn due to localised plastic flow originated in the areas with surface defects. This failure occurs mostly due to the removal of the ‘platelets’ formed under the plastic flow and ‘ploughing’ indicating high wear loss of soft material by a slightly harder steel rod. As opposed to bare steel, wear of the hard ceramic boronised coatings is dealt with microcracking and chipping mechanism. Small ‘pits’ can be seen in the Fig. 11b, and the ‘pitting’ does not grow significantly for the boronised samples. For the boronised samples, the debris occurrence may be mostly through the microcrack formation. The morphology of the surfaces of the boronised samples is shown when they were tested longer as it is difficult to see debris for the boronised samples after the short period time testing. Considering wear of boronised samples under the friction mode in dry conditions, the surface oxidation of the ‘fresh’ layer (at the initial stage of the material removal at friction) can be supposed. This thin nanosized oxide layer (‘tribofilm’), which may have a composition (FeB)_xO_y, works as a lubricious buffer between contacting materials and promotes reduction of COF and sliding of the metallic rod over the coating surface and, therefore, reduces wear loss. This formation of the ‘tribofilm’ of the (FeB)_xO_y composition has some similarity with the formation of thin boron oxide film on B₄C, which was proposed by Erdemir.^51,52 However, the oxide scale is brittle, and this brittleness is associated with the microcrack formation and wear. In the case of bare steel, the formation of iron oxide due to the surface oxidation has, probably, only a very small effect on possible reduction of COF, and fast wear is defined by very low hardness of steel and removal of rather ‘large’ platelets under plastic deformation. The positive effect of the formed ‘tribofilm’ on the reduction of the friction wear correlates well with earlier studies conducted by^53–57 for some other coatings.

a CS (0·5 h); b B-200 (2 h)

a CS (3 h); b B-200 (6 h)

In all cases, the worn solid particles (both from steel rod and from the coating material), being between the metallic rod and the surface of the test coupon, create additional damage of the coupon as these particles are inserted into the coupon surface by the rod. In the case of the boronised coating, harder FeB particles may create the wedging action to the coupon surface increasing microcracking (this microcracking can be seen in Figs. 11b and 12b). However, our studies demonstrated that the amount of the worn hard FeB is negligible compared to sufficient wear of the steel rod, i.e. practically mostly worn steel particles (from the rod) are accumulated during the friction test.

In the case of lubricating conditions, the oxidation of the surface is delayed due to the presence of the oily ingredient in the lubricating system; this oil promotes sliding of the rods and reduces the COF significantly. However, the presence of the liquid (water and oily surfactant) penetrating into the occurred surface defects affects the wear mechanism at the friction mode. The high water content in the lubricating system (∼75%) accelerates wear of the materials to compare with only oil lubrication.

The temperature raise during the reciprocating friction may affect the materials friction resistance and change of COF. Although temperature during tribological testing was not measured, it could be assumed that the temperature increase for the boronised samples was not high in comparison with uncoated samples. Lubas,^39,40 who conducted the tribological studies for different materials at the ring-counter sample friction mode, indicated that the boronised samples (with a coating thickness of ∼40 μm) had the smallest temperature raise.

The obtained results of the friction resistance of the boronised coatings with different thicknesses, indicate that hardness of the coating is not the only dominant factor affecting the performance. The tested boronised coatings with different thicknesses have the same level of hardness; however, their performance is different. Although all boronised coatings have similar morphology consisted of FeB (top layer) and Fe₂B (bottom layer), it may be speculated that the coating of 200 μm thickness had the optimal ratio between the contents of these phases, and this coating had more ‘consolidated’ structure with fewer defects. Probably, only the combination of high hardness, uniform consolidated microstructure, minimal porosity and microcracks in the coating provides the optimal level of friction resistance. A high level of hardness should be more important when a combination of friction and sliding abrasion with hard particles with sufficient sizes (e.g. like a ‘sand’ with particles larger than 50 μm) takes place, i.e. when sand is continuously supplied between the pairs being in friction mode, which usually happens in the oil production.

The boronised coatings are successfully applied to protect tubing, casing and couplings for the sucker rods and some other equipment components of artificial lifting systems used in oil production where they have to serve in severe friction or friction/corrosion conditions. The conducted studies allowed the optimised selection of the appropriate coating structure and processing for actual products with required properties.

Conclusion

The tribological studies have been conducted at the roll-on-plate friction mode in dry and lubricating conditions, which simulate actual friction situations occurring in the oil and gas processing and some engineering applications. These studies demonstrated that boronising, i.e. the formation of hard iron borides through the thermal diffusion process, provides significant reduction of the wear loss under the studied conditions over bare steel (greater than 30 times). No delamination and flaking-off were observed for the boronised coatings. The boronised coatings have only minimal growth of the COF during testing. The boronised coating thickness (case depth) has importance in the selection and manufacturing of components with high performance, e.g. the samples with the case depth of ∼200 μm demonstrated the best performance over the samples with thinner (∼100 μm) and thicker (∼250 μm) cased depth. A high level of the friction resistance of the boronised coatings is defined by their hard monolithic structures, ‘saw-tooth’ morphology and strong diffusion related bonding with the substrate materials. The results of the conducted tribological studies are important for the appropriate selection, design and processing of the products with improved friction resistance for oil processing, e.g. for artificial lift system, as well as for different valves and seats, gears and bearings and some other complex-shape components of engineering tools and devices. These coating demonstrated the adequate performance in the actual field conditions, e.g. at the oil and gas down-hole environments where friction is involved.

Footnotes

Acknowledgements

Endurance Technologies Inc. acknowledges the financial support of the NRC Canada for the process development and optimisation (IRAP project 801309).

References

Hutchings

: ‘Tribology, friction and wear of engineering materials’; 1992, London, Butterworth-Heinemann.

Arthur

(ed.): ‘Wear resistant surfaces in engineering’, UK Department of Trade and Industry, HMSO, London, UK, 1985.

Takadoum

: ‘Materials and surface engineering in tribology’; 2008, London, ISTE - John Wiley & Sons.

Brushan

and Gupta

: ‘Handbook of tribology: materials, coatings, and surface treatments’; 1991, New York, McGraw-Hill.

Brushan

: ‘Introduction to tribology’, 2nd edn; 2013, New York, Wiley.

Hedayat

, Yannacopoulos

and Postlethwaite

: ‘Wear and CO₂ corrosion of steel couplings and tubing in heavy oil screw-pump wells’, Wear, 1997, 209, (1–2), 263–273.

, Humphreys

and Matthews

: ‘Reciprocating-sliding wear of sucker rods and production tubing in deviated oil-wells’, Wear, 1989, 134, (1), 13–28.

Mottram

: ‘Sucker rod wear in directionally drilled steamflood wells’, J. Can. Petrol. Sci., 1991, 30, (2), 156–160.

Medvedovski

: ‘Wear-resistant engineering ceramics’, Wear, 2001, 249, (9), 821–828.

10.

Medvedovski

and Llewellyn

: ‘Oxide ceramics for abrasion and erosion resistance applications’, Interceram, 2002, 51, (2), 120–126.

11.

Imbeni

, Hutchings

and Breslin

: ‘Abrasive wear behaviour of an Al₂O₃-Al co-continuous composite’, Wear, 1999, 233, (1–2), 462–467.

12.

Mondal

: ‘Zirconia toughened alumina for wear resistant engineering and machinability of steel application’, Adv. Appl. Ceram., 2005, 104, (5), 256–260.

13.

Chang

, Binner

JGP

and Higginson

: ‘Dry sliding wear behaviour of Al(Mg)/Al₂O₃ interpenetrating composites produced by a pressureless infiltration technique’, Wear, 2010, 268, (1), 166–171.

14.

Guo

, Cai

, Zhang

and Yang

: ‘Sintering, properties and microstructure of low friction SiC ceramic seals containing graphite fluoride’, Adv. Appl. Ceram., 2013, 112, (6), 341–344.

15.

Basu

and Kalin

: ‘Tribology of ceramics and composites. a materials science perspective’; 2011, Hoboken, NJ, Wiley.

16.

Buckley

and Mioshi

: ‘Fundamental tribological properties of ceramics’, Cer. Eng. Sci. Proc., 1985, 6, (7/8), 919–939.

17.

Roberge

: ‘Handbook of corrosion engineering’; 2000, New York, McGraw-Hill.

18.

Budinski

: ‘Overview of surface engineering and wear’, 4–21; 1996, Materials Park, OH, ASTM Int. Publishing.

19.

Budinski

: ‘Surface engineering for wear resistance’; 1988, Englewood Cliffs, NJ, Prentice-Hall Inc.

20.

Davis

J R

(ed.): ‘Surface hardening of steels: understanding the basics’; 2002, Materials Park, OH, ASM International.

21.

Davis

J R

(ed.): ‘Surface engineering for corrosion and wear resistance’; 2001, Materials Park, OH, ASM International and IOM Communications, Maney Publishing.

22.

Levy

: ‘The erosion-corrosion behavior of protective coatings’, Surf. Coat. Technol., 1988, 36, (1–2), 387–406.

23.

Reeber

: ‘Surface engineering of structural ceramics’, J. Am. Ceram. Soc., 1993, 76, (2), 261–268.

24.

Choy

: ‘Chemical vapor deposition of coatings’, Prog. Mater. Sci., 2000, 48, (2), 57–170.

25.

Voroshnin

and Lyakhovich

: ‘Boronizing of steel’; 1978, Moscow, Metallurgy.

26.

Sinha

: ‘Boriding (Boronizing)’, in ‘ASM handbook’, Vol.4, ‘Heat treating’, 437–447; 1991, Materials Park, OH, ASM International.

27.

Chatterjee-Fischer

: ‘Boriding and diffusion metallizing. Surface modification technologies’ (ed. Sudarsan

T S

); 1989, New York, Marcel Dekker Inc.

28.

Dearnley

and Bell

: ‘Engineering the surface with boron based materials’, Surf. Eng., 1985, 1, (3), 203–217.

29.

Hunger

H.-J

and Trute

: ‘Boronizing to produce wear-resistant surface layers’, Heat Treat. Met., 1994, 2, 31–39

30.

Telle

, Sigl

and Takagi

: ‘Boride-based hard materials’, in ‘Handbook of ceramic hard materials’ (ed. Riedel

), 802–945; 2000, Weinheim, Wiley.

31.

Ulutan

, Celik

, Gasan

and Er

: ‘Effect of different surface treatment methods on the friction and wear behavior of AISI 4140 steel’, J. Mater. Sci. Technol., 2010, 26, (3), 251–257.

32.

Petrova

and Suwattananant

: ‘Surface modification of ferrous alloys with boron’, J. Electron. Mater., 2005, 34, (5), 575–582.

33.

Suwattananant

and Petrova

: ‘Oxidation kinetics of boronized low carbon steel AISI 1018’, Oxid. Met., 2008, 70, (5–6), 307–315.

34.

Petrova

, Suwattananant

and Samardzic

: ‘The effect of boronizing on metallic alloys for automotive applications’, J. Mater. Eng. Perfrom., 2008, 17, (3), 340–345.

35.

Krasnev

: ‘Improvement of corrosion resistance of steels by surface modification’, in ‘Corrosion resistance’ (ed. Shih

), 295–316; 2012, Rijena, InTech Europe.

36.

Asthana

, Usta

, Ucisik

and Liang

: ‘Wear and surface characterization of boronized pure iron’, J. Tribol., 2007, 129, (1), 1–10.

37.

Martini

, Palombarini

and Carbucicchio

: ‘Mechanism of thermochemical growth of iron borides on iron’, J. Mater. Sci., 2004, 39, (3), 933–937.

38.

Taktak

: ‘Tribological behaviour of borided bearing steels at elevated temperatures’, Surf. Coat. Technol., 2006, 201, (6), 2230–2239.

39.

Lubas

: ‘Practical application of boron-modified sliding pairs in I. C. engine’, Tribol. Int., 2010, 43, (11), 2046–2050.

40.

Lubas

: ‘Tribological characteristics of the surface layer of steel modified with boron under mixed friction conditions’, J. KONES Powertrain Transp., 2011, 18, (2), 271–278.

41.

Medvedovski

, Chinski

and Stewart

: ‘Thermal diffusion coatings for wear-resistant components for oil and gas industry, in ‘Ceramic Trans. 243, Innovative processing and manufacturing of advanced ceramics and composites II’, A collection of papers presented at the 10th Pacific Rim Conf. on ‘Ceramics and glass technology’, (ed. Ohji

, Colombo

, Naito

and Garay

J E

), Coronado, CA, USA, June 2013, Oak Ridge National Lab, 13–29.

42.

Medvedovski

, Chinski

and Stewart

: ‘Wear- and corrosion-resistant boride-based coatings obtained through thermal diffusion CVD processing’, Adv. Eng. Mater., 2014, 16, (6), 713–728.

43.

Filici

, Kapsiz

and Durat

: ‘Applications of Taguchi design method to study wear behaviour of boronized AISI 1040 steel’, Int. J. Phys. Sci., 2011, 6, (2), 237–243.

44.

Ingole

, Liang

, Usta

, Bindal

and Ucisik

: ‘Multi-scale wear of a boride coating on tungsten’, Wear, 2005, 259, (7–12), 849–860.

45.

Pranay

, Liang

, Usta

and Ucisik

: ‘Wear and surface characterization of boronized pure iron’, J. Tribol., 2007, 129, (1), 1–10.

46.

Martini

, Palombarini

, Poli

and Prandstraller

: ‘Sliding and abrasive wear behavior of boride coatings’, Wear, 2004, 256, (6), 608–613.

47.

Venkataraman

and Sundararajan

: ‘The high speed sliding wear behaviour of boronized medium carbon steel’, Surf. Coat. Technol., 1995, 73, (3), 177–184.

48.

Meric

, Sahin

, Backir

and Koksal

: ‘Investigation of the boriding effect on the abrasive wear behavior in cast irons’, Mater. Des., 2006, 27, (9), 751–757.

49.

Tabur

, Izciler

, Gul

and Karacan

: ‘Abrasive wear behavour of boronized AISI8620 steel’, Wear, 2009, 266, (11–12), 1106–1112.

50.

Jiang

and Arnell

: ‘The effect of substrate surface roughness on the wear of DLC coatings’, Wear, 2000, 239, 1–9.

51.

Erdemir

: ‘Tribological properties of boric acid and boric-acid-forming surfaces. I. Crystal chemistry and mechanism of self-lubrication of boric acid’, Lubr. Eng., 1991, 47, 168–172.

52.

Erdemir

, Bindal

, Zuiker

and Savrun

: ‘Tribology of naturally-occurring boric acid films on boron carbide’, Surf. Coat. Technol., 1996, 86–87, (Part 2), 507–511.

53.

Biswas

: ‘Some mechanisms of tribofilm formation in metal/metal and ceramic/metal sliding interactions, Wear, 2000, 245, (1–2), 178–189.

54.

Aizawa

, Mitsuo

, Yamamoto

, Sumitomo

and Muraihi

: ‘Self-lubrication mechanism via the in situ formed lubricious oxide tribofilm’, Wear, 2005, 259, (1–6), 708–718.

55.

Inman

, Datta

, Du

, Burnell-Gray

and Luo

: ‘Microscopy of glazed layers formed during high temperature sliding wear at 750°C’, Wear, 2003, 254, (5–6), 461–467.

56.

Erdemir

and Donnet

: ‘Tribology of diamond-like carbon films: recent progress and future prospects’, J. Phys. D: Appl. Phys., 2006, 39, (18), R311–R327.

57.

Pastewke

, Moser

, Gumbsch

and Moseler

: ‘Anisotropic mechanical amorphization drives wear in diamond’, Nat. Mater., 2011, 10, (1), 34–38.