Parameters of NiCrCoAlY powder for HVOF coating.
In petroleum and petrochemical industries, offshore and onshore systems have to function in an aggressive environment that exposes the production equipment components to thermal cycling and wear and corrosion. Although maintenance of material degradation in oil and gas is costly, internal and external parts of the equipment and pipelines must be well inspected and continually maintained. For this reason, highly advanced corrosion and wear–monitoring systems must be installed in the critical areas of the plant to protect pipes and equipment from seawater and crude oil. Therefore, researchers are in search of advanced materials and methods that could be applied in oil and gas pipelines and accessories for increasing their working time. The common manufacturing processing method for improving the surface of piping and accessories is overlay welding or cladding. This method has some limitations, such as its limitation for choosing materials. In addition, the high temperature of welding causes some defects on the final surface, such as thermal residual stress, cracking, and distortion in the substrate. The method is also time consuming and costly. However, the coating method provides a blend of unique properties with low cost. Thermal spray methods are cold spraying techniques that have a considerably less thermal stress, residual stress, and other defects. Among different thermal spray coating techniques, high-velocity oxygen fuel (HVOF) and plasma are the most commonly used thermal spraying coating processes to produce anti-wear and corrosion coatings with different types of materials such as metal, alloys, and ceramic composite . Furthermore, HVOF and plasma thermally sprayed coating processes induce microstructure heterogeneities, which increase the corrosion and wear resistance. In this research, one type of alloy with chemical composition NiCrCoAlY was chosen for increasing corrosion resistivity of carbon steel piping. A corrosion behavior of coated samples in seawater was investigated for 30 days. Potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) results indicated that these types of alloys protected the surface of carbon steel piping from harsh environments. However, the corrosion protection of NiCoCrAlY deposited by HVOF technique is higher than a plasma coating technique.
- carbon steel piping surface
- thermal spray coating techniques
1. Literature review
Corrosion is generally defined as the destructive disintegration process through which a metal is gradually destroyed by electrochemical reactions with the environment . There is no hard and fast rule to classify corrosion, but it can normally be divided into two main groups, uniform and localized corrosion . In uniform corrosion, damage occurs on the entire surface at a uniform rate. In localized corrosion, damage occurs on localized spots due to heterogeneities in the material microstructure (galvanic, intergranular, and dealloyed) and the environment (crevices and hydrogen damage). Localized stress (stress corrosion cracking, corrosion fatigue, and fretting) and the geometry of closed chemical fluid flow system parts (erosion and cavitation) also cause localized corrosion damage [3, 4].
Corrosion damage depends on the extent of particular factors, e.g. aggressive ions or stress amplitude.
Corrosion has huge economic and environmental impact on infrastructures worldwide such as highways, bridges, oil and gas industries, chemical processing, and water and wastewater systems. Corrosion cannot be fully eliminated but its effects can be minimized using different protection and prevention methods . The direct cost attributed to corrosion damage has been estimated in the order of 3–4% of industrialized countries’ gross national product (GNP) .
According to the National Association of Corrosion Engineers (NACE), the global corrosion cost, through direct and indirect losses, was $US 552 billion in 2001, which increased to $US 1.3 trillion in 2009 . Direct and indirect damage to the environment is massive if corrosion-related problems are overlooked in sensitive oil/gas industries and nuclear power plants. For instance, radiation and poisonous gas leakage due to severe pitting in nuclear plants can put workers’ health at risk .
Corrosion measurement and prevention cover a large field of technical activities such as measuring corrosion rate; controlling physical parameters like temperature, pH, and pressure stress; protecting against corrosion like cathodic and anodic protection ; chemical dosing; and prevention by material selection or organic/inorganic coatings. Measuring corrosion rate by weight loss is one of the corrosion testing methods commonly used since the past till now . A material is exposed to an environment in which it needs to be in service for a prolonged time (90 days). The corrosion rate is measured from the net weight loss per unit time, i.e. difference in weight before and after the exposure divided by exposure time. In this way, corrosion damage can be assessed for the future by extrapolating the weight loss results. With advancements in electrochemical sensor technology, different sorts of resistance probes, linear polarization resistance probes, and H2 evolution probes are used for monitoring corrosion systems [11, 12].
Metallic and nonmetallic coatings along with cladding or surface modifications are also common corrosion prevention methods. Their effective use depends on carefully selecting and regular service monitoring of corrosion. Corrosion prevention, monitoring, and testing can save billions of dollars besides minimize hazards [5, 6].
In oil and gas, many components, such as valve, wellhead, Xmass tree, exchanger, and so on, exposed to the harsh environments and cause the long life of equipment significantly decreased. For this reason, choosing a suitable alloy that has a specific properties depend on environment is essential for increasing corrosion resistivity of equipment .
The selection of corrosion resistant alloys is an important step that is directly affected on the long life system. Because, any mistakes for choosing the materials it make some issues for system [5, 13]. Companies with high research facilities first simulate a certain part of the field environment under study. Then, a group of alloys selected some different alloys based on the available information. Then, test all the candidate alloys at the same time and finally choose the best alloys for a certain condition. This method can easily take 1–3 years to accomplish at significant cost [6, 14].
Another selection method is literature review on corrosion data that generally applied to the expected fields. It is quickest and cheapest technique because it neglected some group of alloys that is not suitable for specific condition. Then, the selected alloys are tested under specific condition [13, 15]. However, the chance of error in this method is greater than the previous method, e.g., introducing potential for Corrosion Resistance Alloys (CRA) failure or using a more expensive alloy than required .
Other resources for material selection are using available standards, such as the 2003 ISO 15156 publications derived from the previous NACE 0175 publication for “sour service.” ANSI/NACE MR0175/ISO 15156 gives general information for service in oil and gas productions and in natural gas sweetening plants in hydrogen sulfide (H2S)-containing environments, and it also recommended some materials in the appropriate design codes, standards, and regulations. Generally, it can be applied to help avoid costly corrosion damage to the equipment itself [13, 17].
Temperature, chloride ion concentration, CO2 and H2S, environment pH, and the presence or absence of sulfur (S) are some parameters that directly affect the corrosion properties of alloys .
These parameters that cause the risk of stress corrosion cracking (SCC), rate of metal solution from pits, and other things are considerably increased .
Generally, three items directly affect the corrosion rate of sample, temperature, pressure, and the chloride content in sodium chloride. In order to make the scale for these parameters is universal, centigrade (°C), pounds per square inch, psi and gram/liter, respectively .
Furthermore, the absence of oxygen is critical to the application of these alloys under the conditions shown. If the environment has an oxygen (typically greater than 10 parts per billion), the other alloys should be consider [20, 21].
According to the information presented above, nickel alloys have a long life for oil and gas process fluids. Also it has a high performance and relatively low cost, and at the end of the structure’s lifecycle, nickel alloys are completely recyclable. For this reason, this alloy is offered for both the environments [9, 22].
The Nickel Cobalt Chromium Aluminum Yttrium family of gas-atomized powders is designed to produce thermal sprayed coatings with excellent high-temperature oxidation and high-temperature corrosion resistance. The presence of cobalt improves coating ductility and high-temperature corrosion resistance. The presence of chromium and yttrium improves oxidation resistance by increasing the activity of aluminum and improving the spallation resistance of the oxide scale. The function of chromium and aluminum is to provide a reservoir that continually replenishes the oxide scale. Maintaining the chromium and aluminum ratio is critical for avoiding coating embrittlement [23, 24].
Adding tantalum to the chemical composition exhibits superior oxidation resistance at high temperatures. This chemical composition can serve as overlay coating on moving and rotating instruments to improve their performance and service life, even under harsh environmental conditions. Gas atomization ensures excellent chemical homogeneity and high purity, which results in consistent coating results. Figure 1 represents a SEM photomicrograph of Nickel Cobalt Chromium Aluminum Yttrium powder. The SEM photomicrograph shows gas-atomized morphology that is typical of these materials .
This chemical composition is a premium grade NiCoCrAlY gas-atomized powder that produces thermal spray coatings with excellent surface resistance against oxidation and corrosion at high temperatures .
High-temperature corrosion resistant bond coats act as thermal barrier and oxide-based abradable coatings for hot section components. They also produce superior low oxide coatings that machine well and closely resemble wrought alloys in terms of characteristics .
A noteworthy point of preference of thermal spray procedures is to a great degree wide assortment of materials that can be utilized to make coatings. For all intents and purposes, any material that melts without breaking down can be utilized. Another real point of interest is the capacity in most warm splash procedures to apply a covering to a substrate without significantly warming it. In this manner, materials with high melting points can be Joint to the substrate without changing the characteristics or thermal distortions of the parts. A third advantage is the ability, in most cases, to strip and recoat worn or damaged coatings without changing the part’s properties or dimensions. Also, applying thin coatings on low-cost substrates results in increased efficiency and cost savings [28–30].
There are five different types of thermal spray coatings: flame spraying, wire arc splashing, detonation gun deposition, plasma spray, and high speed oxy-fuel.
Coating produced by HVOF and plasma processes is characterized by lamellar structure embedded with solid particles, oxide, and inclusions (surface residue from shot blasting or surface cleaning) . Molten or semi-molten particles deform upon hitting the substrate. Deformed particles in the coating are called splats and are approximately 1–20 μm thick. The high striking speed of particles produces compact coating, but some voids/pores still form at the surface and interlamellar particle boundaries . The impaction of solid particles at high speed also adds strength due to the peening effect (relieving tensile stress or adding compression by low level mechanical stresses). The high temperature and presence of oxygen in the environment also cause the formation of oxides. The adhesion between substrate and coating is predominantly by mechanical interlocking .
The utilizations of thermal spray coatings are to a great degree changed, yet the vital use classes include upgrading surface wear and/or erosion resistance. Different applications incorporate their utilization for dimensional reclamation, as thermal obstructions, thermal conduits, and electrical channels or resistance; for electromagnetic protection; and improving or retarding radiation. Thermal spray coatings are utilized for all intents and purposes in each industry, including aviation, farming, car, essential metals, mining, paper, oil and gas creation, chemicals and plastics, and biomedicine .
The substrate material used is carbon steel (S45) because it is one of the most popular materials used in oil piping production in both upstream and downstream domains. The substrate material was supplied by Kelvin Steel, Glasgow, with the chemical composition in percentage as follows: C, 0.42–0.50; Mn, 0.5–0.80; Si, 0.17–0.37; Cr (max%), 0.25; Cu (max%), 0.25; Ni (max%), 0.25; S (max%), 0.035; P (max%), 0.035; and Fe balance.
For the layer deposited onto the carbon steel surface, gas-atomized and spheroidal NiCoCrAlTaY powders with nominal size ranges of 45+11 and 37+15 μm, respectively, were used for plasma and HVOF methods.
For increasing the adhesion of deposited layer to substrate, the surface preparation is a very important step. Roughness of sample directly affects the adhesion of deposited layer, which can be controlled by different parameters such as blasting pressure, angel, and so on. .
Grit blasting was carried out with a high efficiency sand blaster with Alumina grit (size 10–20 mesh), 8/10 mm nozzle, and operating at a blasting pressure of >0.5 MPa. The distance between substrate and nozzle was 150 mm with a 30° angle. The grit blasting time was dependent on obtaining the required surface roughness. Upon grit blasting completion, powders with different chemical compositions were sprayed using HVOF and/or plasma gun systems.
The coating thickness selected for all samples in this research after trial and error was approximately 400 μm. Generally, increasing some initial parameters, such as arc current, power feed rate, and so on, helps increase the thermal spray coating thickness.
For the deposition of this powder by HVOF machine, a Model 9MP machine was employed. Table 1 shows the parameters for HVOF coating. For plasma coating, the same specific setup was used as for HVOF coating in this material powder group. The setup of powder coated by plasma machine is tabulated in Table 2.
|Model: 9MP||DJ2700 hybrid|
|Suction and spreader||L/L|
|Gases||Pressure (psi)||Flow (FMR)||SCFH|
|Natural gas (CH4)||110||140.2||320|
|Carrier gas (N2)1||105||350||799|
|Spray distance||254 mm (10”)|
|Spray rate||38 g/min (5 lb./h)|
|Type||#2 (1.8 mm/0.071″)|
|Suction and spreader||L/L|
|Spray distance||114–140 mm (4.5–5.5″)|
|Spray rate||40–57 g/min (5.3–7.5 lb./h)|
The coating morphology was observed through a high resolution FEI Quanta 200F field emission scanning electron microscope (FESEM). Using an EDX-System (Hitachi SU8000) instrument attached to a FESEM instrument, energy dispersive X-ray (EDAX) analysis was performed to investigate the elemental composition of the samples.
The surface of the substrate and coated samples before and after corrosion was also studied by a relatively destructive method. The X-ray diffraction (XRD) equipment is an Empyrean X-ray diffractometer with Cu Ka radiation (k = 1.54178 A operating at 45 kV, 30 mA, 0.026° step size, and scanning rate of 0.1 s 1 in a 2 h range from 10° to 90°).
For measuring the corrosion rate of coated sample, electrochemical techniques, such as DC (polarization) and AC (impedance), were used and applied using a potentiostat/galvanostat model AutoLab PGSTAT30 from Ecochemie (Netherlands) . Polarization experiments were performed using a three-electrode cell, where the samples were the working electrode (WE) while a platinum wire and saturated calomel electrode (SCE) was the counter and reference electrodes, respectively. The electrolyte was 3.5% NaCl. In this test, the linear scan voltammetry, corrosion current (Icorr), and corrosion potential (Ecorr) were calculated from the Tafel plots with a scan rate of 0.001 Vs−1.
A frequency range of 105–10−2 Hz with amplitude potential of 5 mV vs SCE was used to conduct electrochemical impedance spectroscopy (EIS). To obtain the charge transfer resistance values across the electrode-electrolyte interface that could relate to parameters from the polarization output, simulations with analog circuits were performed with the experimental data. Having been installed in a computer interfaced with a potentiostat, general purpose electrochemical software (GPES) and frequency response analyzer (FRA) were used to run the AC and DC techniques, respectively.
3. Results, analysis and discussion
3.1. Microstructural analysis of surface
Figure 2 displays FESEM micrographs of plasma and HVOF-coated samples before corrosion testing. The top view image indicates a relatively dense, uniform, and crack-free coating deposited on the substrate. However, the HVOF-coated samples have a semi-uniform surface compared to the plasma-coated samples.
A small amount of semi-molten and nonmolten particles is visible in the fully melted region (Ni) area. The presence of semi-molten and nonmolten particles provides the coating with high bonding strength and good wear resistance [34, 35]. A few pores appear in black in the micrographs of both types of coating. The porosity size and Average number of porosity with the HVOF method are greater than with plasma.
The coatings deposited by plasma spray (Figure 2a) exhibit the most visible defects. Shrinkage of the molten droplets, as they splat onto the substrate and rapidly solidify, is one of the main causes of porosity and visible defects in plasma spray coatings. Coatings deposited by HVOF (Figure 2b) feature the most desirable structure with minimal porosity, absence of cracks, and a clean interface with the substrate. The finer pores within HVOF coating demonstrate the superior compaction and deposition of particles during spraying. Also, HVOF coating has a rough surface due to the overlap of different splats and the roughness of each single splat [36, 37].
X-ray diffraction patterns and EDX for NiCoCrAlY plasma and HVOF-coated samples are shown in Figure 3. According to the coated samples in Figure 3a and b, Only three phases were detected γ (Co, Ni, Cr), β(Co, Ni, Y)Al, and Al2O3. The morphology of the coatings consists of single-phase splats separated by Al2O3veins as a result of the oxidation occurring in the deposition process.
Previous studied showed the samples that coated with NiCoCrAlY by vacuum plasma coating(VPS) have high performance. Because it can provide good adhesion to the metallic substrate with high density, while this method is more expensive rather than other techniques of thermal spray coatings. For this reason, high-velocity oxygen fuel (HVOF) has a good option for disposition of NiCoCrAlY alloys instead of VPS. However, existing the free oxygen in the combustion gas causes increasing homogeneity in the melted powders and remove Oxidation particles, because they need high temperature. The aluminum and yttrium elements in the metallic powder have high affinity with oxygen and are thus easily oxidized during thermal spraying; consequently, the coatings have high oxide content [38–42].
3.2. Electrochemical corrosion analysis
The open circuit potential (OCP) of coated samples for day 3 and day 30 is shown in Figure 4. The plasma-coated sample (blue) shows the potential moved from −0.421 on day 3 toward noble regions to −0.341 at the end of that period, and the OCP of the HVOF-coated sample (red) was more negative compared to the plasma-coated sample, with a change from −0.305 on day 3 to −0.413 on day 30.
The charge transfer resistance Rct was consistent with the OCP values from polarization measurements and Icorr illustrated in Table 3. It can be seen that the Rct for plasma-coated samples changed from 11.735 to 10.098 kΩ from day 3 to day 30, and for the HVOF-coated samples, it changed from 12.193 to 8.242 kΩ. The higher Rct (EIS data) is due to the greater resistance to charge transfer across the electrode-electrolyte interface, which is consistent with the positive OCP result values belonging to the plasma-coated sample. From the computer simulations for the plasma-coated sample, the Rs (QR) circuit diagram accurately matches the experimental data. Only one semicircle is observed in the Nyquist plot (Figure 5a and c).
|Day||Plasma-coated sample||HVOF-coated sample|
|OCP||ICorr/A (×10−5)||RCT/kOhm||OCP||ICorr/A (×10−5)||RCT/kOhm|
The Bode phase diagrams of the plasma-coated samples (Figure 5b and d) show one-phase maxima, consistent with the presence of one semicircle in the Nyquist plots from Figure 5a and c with Rs (Q1R1) circuit. The Rs (QR) circuit diagram accurately matches the start day of the plasma-coated samples. Only one semicircle and one maximum phase were observed in the Nyquist and Bode plots (Figure 5a and b). Figure 6a and b shows that the Rs (QR) circuit diagram accurately matches the experimental data at the beginning of measurement. Whereas for the HVOF-coated samples, at the end of the period, the equivalent circuit Rs (Q1 [R1(Q2R2)] accurately fits the experimental data for HVOF coating, because two semicircles and maxima phases were observed in the Nyquist and Bode plots (Figure 6c and d). The resistance between RE and WE is the solution resistance Rs, which is in series with two time constants (parallel arrangement of R and Q) that are also in series.
It can be seen that the Nyquist plot (Figure 5c) is a “depressed semi-circle” with the center of the circle below the x-axis [43–45]. Q1 is parallel to R1. R1is the polarization resistance of the area at the substrate/coating interface where corrosion occurs. Also, Q2 is in parallel with a charge transfer resistor R2, corresponding to the pores on the coated layer surface. The main reason for the higher Rct of the plasma-coated sample than the HVOF-coated sample at the end of the period is the presence of a bigger oxide layer on top of the plasma-coated samples that protects the deposited layers from electrolyte penetration to the interface.
In these simulations, the plasma coating had higher charge transfer resistances, Rct, than the HVOF coating due to the higher resistance against corrosion. The charge transfer resistance, Rct, values for plasma and HVOF-coated samples are presented in Table 3.
According to Table 3, the corrosion current density of plasma-coated samples increased from 5.683 × 10−5 to 6.030 × 10−5 from day 3 to day 18, and then suddenly dropped to 5.436 × 10−5A on day 21. After that, it rose again up to 6.231 × 10−5A on day 30. The corrosion current density for HVOF-coated samples moved from 5.267 × 10−5 to 5.635 ×10−5 on day 3 to day 9, but suddenly dropped to 4.897 × 10−5A on day 12. From the middle of the period until the end, the current density increased from 5.593 × 10−5 to 6.289 × 10−5A. The decreasing amount of corrosion current density related to the oxide layer created on top of the deposited layer protected the samples from the corrosive environment. Therefore, the amount of Rct over this period increased.
3.3. Microstructural analysis of corrosion
FESEM micrographs of both coated sample types after 30 days of immersion in crude oil and seawater are shown in Figures 7 and 8. According to Figures 7a and 8a, when samples were exposed to the oil environment, the corrosion rate on the coated samples’ surface was lower than in NaCl electrolyte (seawater) solution. According to Figure 7, the amount of pitting corrosion on the plasma samples was less than the HVOF samples, because there were fewer fine holes produced than on the HVOF samples.
In Figure 8, the HVOF-coated samples corroded significantly after 30 days and large holes appeared on their surface. This coating displayed evident crevices and pitting corrosion in the microstructure containing intersplat porosity, that was detected on the top surface of HVOF-coated samples by previous studies [46, 47]. Also, the size and amount of corroded area inside NaCl solution (seawater) is obviously greater than in the crude oil.
Regarding Figures 7 and 8, there is a big difference between the corroded and uncorroded samples with the HVOF coating method in both environments, whereas in plasma-coated samples, the difference between corroded and uncorroded samples is considerably less than the HVOF method. However, both methods protected the substrate from corrosion.
Generally, there are several factors affecting corrosion resistance, including adhesion force between coating layer and substrate, cooling rate, generation of oxides and compounds, and porosity [48, 49]. The NiCoCrAlY alloy has an excellent corrosion resistance and oxides generated affect chemical stability, bond strength, and durability. As it has high adhesion to the base metal and creates a dense oxide layer, it yields high durability .
Morphological, microstructural, and compositional analyses were performed for both coating methods in order to characterize the samples before and after the corrosion tests. The characterization for seawater (3.5% NaCl) and crude oil after 30 days showed that both methods can protect the substrate from corrosion, but the corrosion rate with HVOF is higher than with the plasma coating method. This result suggests that the potential impact of the plasma thermal spraying process against a corrosive environment is better than that of HVOF spraying systems. Therefore, from this chemical composition group, the plasma samples were chosen for corrosion resistivity.