Many mechanical erosion models have been established to correlate the erosion resistance of target materials to their mechanical properties and hydrodynamic parameters[1-2]. A detailed literature review on this aspect is out of scope of this article. Meng and Ludema[3] provide an exhaustive overview up to 1995, found 182 equations and selected 28 for special study. Lyczkowski and Bouillard[3] gave one up to 2002. As pointed by Tsai et al[4], over a fairly wide range of variables, at least in flowing slurries, the overall dependence of the particle and target hardness (HP and H respectively), on erosion is approximately given by ė0 ∝ HP1/2/H. Generally, the erosion resistance of target materials increases with their hardness if no substantial change takes place in the erosion mechanisms and, as illustrated in Fig. 1, the power law erosion rate of materials can give a fairly good fit to the correlation between the mechanical erosion rate and surface hardness[5-7].
where κH and nH(n＞0) are experimental constants depending heavily on the erosion mechanisms.
At least two irreversible processes are involved in the erosion-corrosion process, namely, the electrochemical corrosion at surface and plastic deformation in surface layer. The fluxes of these two irreversible processes can be represented by the corrosion rate ċ (or the anodic current density iA=ċ/zF, where z is the number of electrons involved in the corrosion reaction of the electrode material and F is the Faraday constant) and the plastic strain rate (=Ṅλb, where Ṅ, λ and b are the flux, the mean free path and Burger’s victors of dislocations, respectively). When the anodic dissolution on surface and the plastic deformation in surface layer occurred simultaneously, they will enhance each other leading to the synergistic effect[6, 8, 18],
where is plastic deformation rate in an inert environment, iA, 0 is the corrosion current density of material free of dynamic plastic deformation, FC and FP are the general driving forces for the plastic deformation (such as the force produced by particle impingement) and anodic dissolution (the potential), respectively, LC→P and LP→C are the coefficients representing the cross effects. The second term in Eq.(3) stands for the mechanical impact enhanced corrosion[6-7] and it indicates that the anodic dissolution rate increases linearly with the plastic deformation rate[18-19].The second term in Eq.(2) implies that the plastic deformation in the surface layer would be promoted by the corrosion occurring on surface. The reduced resistance to the plastic deformation can be characterized by the degradation of surface strength or hardness[6, 8, 9, 18].The degradation surface hardness due to the presence of anodic dissolution ΔH can be formulated as follows[6]:
where H is surface hardness measured in an inert environment, ΔH is defined as the difference between the hardness values measured in corrosive solution while anodic current is present on surface and in the inert environment. B is a constant related to the active volume of dislocations and test conditions, ith the threshold current to cause the surface strength degradation. The phenomenon of corrosion-induced surface hardness degradation has been experimentally observed in carbon steels and commercial pure iron using the micro-hardness and nano-indentation techniques[6, 8-9, 18].An example of corrosion-induced micro-hardness degradation is demonstrated in Fig. 2.
If the increasing erosion rate caused by the anodic dissolution-induced hardness degradation is the only mechanism of corrosion-enhanced erosion, the increment of erosion rate Δė due to the presence of anodic dissolution can be defined as the corrosion-enhanced erosion, namely, ėc=Δė0=ė-ė0. By combining Eqs.(1) and (4), the normalized corrosion-enhanced erosion wastage, i.e., the wastage ratio of ėc/ė0, can be correlated to the anodic current density iA as follows[6, 8, 18],
where Z is an experimental constant, ith is the threshold anodic current density to cause the corrosion-enhanced erosion. According to Eq.(5), the erosion will be enhanced by the chemo-mechanical effect when corrosion occurs simultaneously and the erosion augmentation defined by ASTM G119(=1+ėc/ė0) will be approximately a linear function of the logarithm of anodic current density. It has been shown the prediction of Eq.(5) agrees well with the experimental results obtained of carbon steels, as shown in Fig. 3[6, 8]. The practical engineering, the slurry pipe is normally operated under open circuit potential (OCP). The experiments in Ref.8 indicated that the corrosion-enhanced erosion at the OCP was predictable using the curve obtained under galvanostatic control as the corrosion current density at the OCP is known.
As shown in Fig. 8, the corrosion-enhanced erosion is also affected by the concentration of solid particles in slurry when the anodic current density is held unchanged. The impact on the sand concentration can be predicted when the normalized wastage ratio ċ/ė0 is employed to replace anodic current density, as shown in Fig. 4.
Until now, it is still to build a universal model for the corrosion-enhanced erosion because of complex mechanisms. A recent research[10-12], indicated that when the hydrodynamic condition and anodic current density were held unchanged, the erosion rates of carbon steel in acidic slurries were significantly higher than those in alkaline or near-neutral ones. The erosion rates in corroding slurries with high or near-neutral pH were not affected by the slurry chemistry but the slurry chemistry impact were pronounced in acidic slurries. The high-to-low order of erosion rates was the same as that of in-situ nano-indentation hardness measured in same corroding environments, indicating the anodic dissolution-induced surface hardness degradation as likely the mechanism for the high erosion wastage in acidic slurries. However, it is still unclear why the exposure to acidic electrolytes results in larger surface hardness loss when anodic dissolution rate is same.
Flowing the concept of ‘wear map’ developed by Ashby, a group led by Stark[13] built up the erosion-corrosion map to demonstrate that correlation between the erosion-corrosion mechanisms and/or performance of materials and process parameters, such as temperature, potential, flowing velocity, particle concentration, impact angle etc. The erosion-corrosion maps offer a directly perceived illustration about the effects of various parameters on the erosion-corrosion mechanisms and performance of materials. However, there are four different wastage components involved in erosion-corrosion process and so many factors relating to mechanical, chemical and material aspects that can affect the erosion-corrosion behavior of material. Besides, the interactions of these factors are very complicated. It is difficult to demonstrate these complex relationships with a few maps. Stark et al[13, 14], made efforts to group various parameters into dimensionless ones and to build the erosion-corrosion maps using these dimensionless parameters. However, more work need to be done to understand the physical meanings of these non-dimensional parameters. It is unknown whether or not the erosion-corrosion experimental data from difference sources can be correlated using these parameters.
Although a lot of efforts have been made to establish a theoretical model that allows us to predict the performance of an engineering components based on the erosion-corrosion experimental data obtained in laboratory, limited progress has been achieved[2-3, 5].In practical engineering, the erosion-corrosion rates (ERC) are normally to be correlated to the operating parameters using empirical equations determined by experiments. An example is as follows[15-16],
where ECR is the penetration rate caused by erosion-corrosion, Fi is inhibitor factor, FM and FS, are empirical constants that account for the material hardness and sharpness of sand particles, respectively, FP is the penetration factor for steel (based on 1” pipe diameter); Fr/D is the penetration factor of elbow; CP is weight fraction of sand; QF is the production rate of fluid, r is ratio of pipe diameter in inches to in. pipe; Uimp is the characteristic particle impact velocity. The parameters in Eq.(5) can be determined by experiments or experience. Generally the dependence of erosion-corrosion rate on flowing velocity is formulated as follows[6, 17]:
The β value depends on the relative contributions of corrosion and erosion to total loss. When the solid erodent is present, the corrosion current density increases with increasing sand concentration[18] and the erosion may dominate the total material wastage. The value of β is often used as a diagnostic tool for the erosion-corrosion mechanism, as summarized in Table 2[19].
If the solid erodent is present in the electrolyte, the impingement of solid particles can induce the plastic deformation in the surface layer. The anodic dissolution would be promoted by the dynamic plastic deformation. Both theoretical analysis[6, 20] and experimental results[18, 21], indicated that the anodic dissolution current density iA would increase linearly with the dynamic plastic deformation rate . When the metallic components are exposed to flowing slurry, plastic deformation is induced by the solid particle impingement and the overall plastic deformation rate in surface layer increases with increasing impingement velocity and sand concentration, so that the
In flowing slurries, it is believed that the corrosion is rate-control process if β is close to 1 and the erosion will dominate the material loss when β is close to 3[17, 22]. However, a recent study indicated that when the corrosion is controlled by the repeated breakdown of passive films due to particle impingement and repassivation, the β value is close to 3, as shown in Fig. 5[22].
A careful design of flow geometry can effectively to minimize erosion-corrosion caused by disturbed flow, such as limiting weld root protrusion and steps of flanges[19, 23], utilizing long radius elbows[24] and gradual changes in the flow cross-section[19], replacing the elbow with plugged tee[25]. Utilizing helically-formed pipes to enhance the swirl flow in pipe can reduce the critical flow velocity to suspending solid particles and the erosion rate significantly. The helically-formed pipes can also reduce the pressure drop and improve the particle distribution across the elbow. It leads to erosion uniformly distributed over entire inner pipe surface and hence reduces the potential erosion instead of localized wall penetration[26].
Increasing the thickness of materials in critical areas, using impingement plates to shield the critical areas, and sometimes, rotating pipes can extend the life of tailing lines. In addition, acceptance of a high erosion rate with regular inspection and replacement may be less costly than using more expensive materials is a practice used extensively in minerals processing and oil/gas industries[19].
A proper design also includes optimizing the particle size by grinding and the flow velocity[19], as well as slurry pH and sand control[27]. The erosion rate is reduced significantly when the particle size is less than 100 μm[28-29]. For some geometries where throw sufficient power is possible, cathodic protection is a good option. If the corrosion following the removal of protective film is liquid-phase mass transport controlled, a decrease in flow velocity can retard corrosion process. However, pitting corrosion in flowing slurry has been found in the both field[17] and laboratory tests[30]. The initiation of pitting requires corrosion product layers with local defects[31]. These defects may be the non-uniform growth of the layers and/or to the local mechanical destruction by various fluid-induced mechanical forces. The rupture of protective film results in the rapid anodic dissolution of bare metal in a localized way and, if the repassivation is in any way hindered pits are likely to commence at that sites. The surface roughness generated by erosion may thus be responsible for the enhanced pitting corrosion during erosion-corrosion because the rough surface is likely to enhance the localized micro-turbulence[23]. A lower flowing velocity can generally reduce the erosion-corrosion damage and it will reduce the economic output as well[23].Actually, the flowing velocity cannot be lower than certain limit value to keep particle in suspension[32]. Lower velocities may result in the sliding abrasion of horizontal pipe bottom[19]. Since the localized corrosion will lead a more serious problem than the uniform corrosion, Postlethwaite[33] suggested that slurry pipelines should operate under the conditions that the pipe wall is free from rust and scale to prevent pitting corrosion.
In the two-phase liquid/solid flow, the erosion-corrosion performance relies on both the mechanical properties and electrochemical characteristics. Generally, an increase of Cr-content in steels will improve the erosion-corrosion resistance[23-34]. It has been widely recognized that the erosion resistance of metallic materials increases with increasing relative hardness (the difference between the hardness of target material and particles)[7, 29, 35-37]. This conclusion is correct only when no substantial change occurs in the erosion mechanism[38-39]. Finnie[22] reported the erosion rates of annealed metals were inversely proportional to their Vickers hardness but the erosion rates of some heat-treated steels were almost unaffected by their hardness. Wentzel et al found that the slurry erosion resistance of white cast irons containing tungsten increases linearly with hardness in the low hardness but this relationship does not exist in the high hardness range[40]. Finnie attributed this phenomenon to the low strain hardening rates of metallic materials having high yield strength[22]. In the corrosive slurries, the total weight loss caused by slurry erosion, sometimes, does not decrease with increasing hardness of target materials[29, 41]. It has been known the total material loss in corrosive slurry is a sum those caused by erosion and corrosion respectively while the latter does not relate to the hardness of materials. Wang and Stack[42] isolated the corrosion contribution to the total weight loss and found that only the erosion resistance of mild and stainless steels increases with increasing hardness. Sundararajan[43] found that the resistance of metallic, including metallic matrix composites (MMCs), against the erosion caused by solid/gas mixture was often not well correlated to the mechanical properties measured under quasi-static loading conditions. He suggested using the dynamic hardness in evaluating the erosion resistance[33]. Although, attempts have been made, it is still hard to generalize the effects of hardness[22, 29, 44]. As pointed by Kato[34], the wear resistance is not only dependent on hardness, but also on the ductility of material, as well as surface roughness.
Actually, the correlation between the erosion resistance and hardness of base metal depends on the erosion mechanism. Heitz[30] pointed out that, if the mechanical damage is restricted in the surface layer, especially in corrosion product scale or passive film, it normally exists in the single-phase flow, the adherence, cohesion and hardness of surface layer determines mechanical stability. In this case, the hardness of base metal is not relevant to the erosion-corrosion process but certain chemical changes in these layers may be the cause of a breakdown with subsequent onset of erosion-corrosion. In many corrosion systems, a protective film is likely to form on a metal surface when it exposes to its environment and the film plays an important role in the erosion-corrosion mechanism of materials. The passive film has an ability to inhibit erosion-corrosion damage to a certain extent through inhibiting corrosion as long as it is chemically stable in the environment[45]. The protective function of film relates to its formation kinetics, mechanical properties and hydrodynamic conditions of fluid[46-49]. The kinetics of film formation depends on the composition of materials and conditions of environment[35, 50]. For a ferrous alloy, the protective ability of the film increases with increasing chromium concentration in a matrix[40].
As is summarized by Finner et al. in 1967[51], for example, erosion rate tends to increase with reduced hardness for pure metals but not for steels. Hutchings[52] proposed to correlate the erosion rate to the microhardness of steels measured on the eroded surface. An alternative explanation is the high strain rate created by solid particle impingement may play a role, so that the erosion models based on the dynamic hardness were proposed[53, 54]. The impingement velocity of particles during slurry erosion is relatively low. Lu et al[6, 8], found the erosion rate was reduced with increasing hardness of carbon steels. Wood[26] reported slurry erosion rates measured from metallic and ceramic materials and found that the erosion rate was reduced with increasing hardness and erosion, no matter the materials were brittle or ductile. The corrosion-enhanced erosion is also affected by the corrosion mechanism. The experimental evidences have indicated that the erosion resistance of carbon steels increases with increasing carbon content in composition[28, 55]. Steels with low-bainitic structure are generally more resistant to erosion-corrosion than those have ferrite + pearlite structure[56].
High chrome cast steels are more resistant to erosion-corrosion than plain cast iron and this attributes to its higher Cr concentration in matrix and martensitic structure[57]. Generally, the erosion-corrosion resistance of steels increases with increasing chromium concentration in matrix. Therefore, stainless steels are more resistant to erosion-corrosion than carbon and low alloy steels, while the erosion-corrosion resistance of austenitic stainless steels is better than ferritic stainless steels[32, 58]. Experimental data indicate that addition of alloy elements Cr, Mo, Mn, N would improve the erosion-corrosion resistance of stainless steels[59-61].
Lindsley et al.[62] investigated the erosion resistance and morphology of spheroidized Fe-C alloys with various carbon content and microstructure, and they found that the erosion resistance increased as the mean free path between both the grain boundaries and the carbides decreases. Same phenomenon has been also reported in ferrous alloys containing 0.4% to 1.4% C[52]. These variables control dislocation motion in the ferrite and, in turn, affect the plastic deformation and the erosion resistance of materials. A Hall-Petch-type relationship was found between the mean free path of microstructure and both erosion rate and hardness[52].
Chrome white irons (CWIs) are specifically developed for abrasion resistant applications[63], because of their excellent abrasive resistance and moderate ability against impact, which necessary for crushing, grinding and slurry erosion applications[64-66]. It is often applied as weld hardfacing alloy deposited on the surface of low carbon steel pipelines to improve the erosion-corrosion resistance steel pipes[67]. Generally, high carbon content is required for the formation of carbides to provide erosion resistance[55, 68], but the chromium content in matrix is critical to the corrosion resistance of material[29, 69, 70]. The optimum C content appeared to depend on the Si level[59]. Dodd pointed out the alloys contained 2%~2.5% C, 20%~28% Cr with 2% Mo have good resistance to erosion-corrosion at pH values down to 4. A part of chromium is consumed in the formation of carbides. The experimental evidence has indicated that the minimum Cr content in matrix is 12%. Based on the distribution of Cr between the matrix and carbides determined by electronic probe[71] and the experimental results of dry sand wear and corrosion, Lu et al[29] establish the wear- and corrosion performance map of chrome white iron, as shown in Fig. 6.
However, the erosion-corrosion resistance of chrome white irons depends heavily on the morphology, distribution and size of the secondary phase, as well as erosion mechanism[29, 72-73]. The matrix structure is adjusted by heat treatment and alloy content to balance wear properties and toughness[53, 74]. The effect of microstructure on corrosion resistance is still unclear[75-76]. The slurries in the oil sand production are often corrosive. CWIs become less resistant to wear when corrosion is present[77-78]. When chrome white iron was eroded in slurry with low pH, the contribution of synergism to total material loss was reported being as high as 86.3%[78]. Therefore, a comprehensive understanding of mechanisms of synergistic effect is critical to improve the erosion resistance in corrosive media.
The metal matrix composites (MMCs) comprise metallic binder and hard particles phases, and they are normally used as hard coating. The binder materials include Ni, Co, Al and their alloys[79-81]. Sometimes, austenitic stainless steel was also adopted[82]. The hard particles commonly used include WC, B4C, and SiC. Scanning electron microscopic (SEM) study confirmed that wear of composite is mainly governed by the synergistic effect of the two simultaneous processes: (1) corrosion, erosion and abrasion of the matrix by the slurry; and (2) fracture and removal of the hard particles due to erodent impingement at a high speed[67-68, 83]. The erosion-corrosion resistance of MMC depends heavily on the corrosion resistance of binder, since the binder material tends to dissolve preferably. The hard particles will be readily removed by erosion if they loss support of binder[41, 67, 73]. The differences in binder composition will influence the MMCs’ hardness and corrosion behavior, which in turn affects the synergistic action of erosion-corrosion. The erosion-corrosion mechanism depends heavily upon the corrosion kinetics of binder[67, 69] and mechanical properties of hard particles[84]. The inherent corrosion resistance of pure nickel and cobalt binder did not increase the erosion-corrosion resistance of the MMCs in slurry, but both the nickel-chromium-cobalt grades and the nickel-chromium grades were found to improve the erosion-corrosion behavior compared to the pure cobalt grade[69, 85]. This fact indicates that the erosion-corrosion resistance of MMC can be further improved by optimizing the binder composition.
The performance of MMCs is closely related to the corrosion kinetics of binder[67, 69], as well as the volume fraction and mechanical properties of hard particles[86]. Pugsley et al[87] investigated the cavitation erosion performance of a range of tungsten carbide-cobalt (WC-Co) composites of various grain sizes (0.5 ~ 5 μm) and cobalt contents (6%~15%wt). They found that the correlation of the erosion-corrosion resistance and the binder content depended on the erosion-corrosion mechanism, while the WC gain size is of strong influence on the erosion-corrosion mechanism. The interface structure between the hard particles and matrix can affect the erosion-corrosion resistance significantly[70]. The surface analysis has indicated that the performance of WC-Co-Cr system can suffer localized corrosion in area adjacent to the interface of particle/matrix[88].
Coating techniques play an important role in minimizing the loss caused by erosion-corrosion. The CWIs and MMCs are most commonly used coating materials. WC and W2C are almost as hard as diamond. They are particularly suitable to be used in hardfacing. Tungsten carbide overlay coatings can be applied by high-velocity oxygen fuel (HVOF) spraying or welding. The coating thickness obtained by HVOF technique is between 10 to 1000 mm, which is suitable for the valves, pistons and pump impellers. The extremely abrasive conditions in oil drilling demand thicker coatings and better bonding. The plasma-transferred arc (PTA) process using multilayer technique can produce facings up to 120 mm. It is widely used in oil sand industry in Canada[89]. Other coatings processes, involving chemical vapor deposition, have been used to produce ultra-hard coatings like diamond for pump components and mechanical seals. As a low cost approach to improve the surface hardness of steel pipe, induction-quenching technique was utilized to improve the internal pipe surface. However, the process parameters are still needed to be optimized to prevent pitting corrosion caused by the heterogeneity of microstructure[90].
In many cases, it is technically difficult to change the nature of slurry to be transported. According to the understanding of damage mechanisms resulting from the synergism of mechanical and (electro)chemical factors, as those described in the previous sections, the material performance can be improved by reducing corrosion rate.
The addition of inhibitor could efficiently reduce the corrosion and corrosion-enhanced erosion wastages of carbon steel in slurry containing 1% silica sand and saturated with CO2[91]. Chromates and nitrites with high concentrations act as passivating inhibitors, whereas chromates at low concentration act as a cathodic inhibitor, and were used in the first long distance coal-slurry pipelines[92, 93]. However, chromates are highly toxic, so that effort have been made to find non-chromate inhibitors used in the cooling water systems, zinc, sodium tripolyphorphate (Na5P3O10) and nitrilotris (methylene) triphosphonic acid, N[CH2PO(OH)2]3, showed little benefit in erosive slurries when used alone or along with chromates[19].
Solution conditioning involves raising the pH and/or deaeration, as illustrated in Fig. 9. Both have been applied to long-distance slurry pipelines. An increase in pH can promote pitting because thicker scale is likely to form[94-95]. Deaerationis achieved by adding oxygen scavengers, such as bissulphite or hydrazine, or by non-chemical steam stripping or vacuum deaeration. The latter two methods of deaeration are not suitable for slurry pipelines. They are used extensively with oil-well water injection systems[96].
Erosion-corrosion is big concern in oil and gas production. Although significant efforts have been made, it is still difficult to evaluate the erosion-corrosion wastage of engineering materials and/or components in a quantitative way, because of the complexity resulting form the synergisms among the factors relating to mechanics, chemistry and materials. The economic loss due to erosion-corrosion can be minimized by utilizing proper design, material selection and chemical control.
2013, Vol. 42 Issue (2): 95-105
2013, Vol. 42 Issue (2): 95-105