Study on the corrosion behaviour of a three-metal coupling system under temperature variation

Abstract

This study focuses on a typical three-metal pair (20# carbon steel/H62 brass/6063 aluminum alloy) commonly used in marine engineering. The weight loss method, electrochemical detection and wire beam electrode (WBE) technique were employed to investigate temperature-induced corrosion effects on this three-metal pair and establish a corrosion model. From the polarisation curves of single metals, it was found that under the dual effects of temperature on the corrosion process, the corrosion rate of 6063 aluminum alloy showed an inflection point at 20 °C. Results from polarisation curve superposition and galvanic current tests of the three-metal system indicated that the system followed the mixed potential theory: E(6063 aluminum alloy) < E(20# carbon steel) < E(mixed potential) < E(H62 brass) and |Ig (H62 brass)| = |Ig(20# carbon steel)| + |Ig(6063 aluminum alloy)|. However, a ‘galvanic effect amplification’ phenomenon was also observed, which does not exist in bimetallic systems. The weight loss method revealed that as temperature increased, the corrosion weight loss of both 6063 aluminum alloy and 20# carbon steel in the coupling system increased. Notably, the corrosion rate of the aluminum alloy was more significantly affected by temperature. The WBE technique showed that with rising temperature, the electrode potentials of the three metals decreased to varying degrees. This indicated a reduction in the activation energy of the corrosion process and a decline in the corrosion resistance of the coupling system. Among the three metals, the 6063 aluminum alloy had the lowest electrode potential (lower than that of 20# carbon steel and H62 brass), so it acted as the main anode in the coupling system and suffered corrosion. At 30 °C, 20# carbon steel underwent anodic/cathodic polarity reversal. By integrating the above analysis results, a corrosion model for the three-metal coupling system was established. This model fills the gap of the classical mixed potential theory in multimetal–seawater temperature coupling scenarios.

Keywords

electrode arrays corrosion in marine environments galvanic corrosion aluminum alloys brass carbon steel

Experimental materials and methods

Introduction

The marine environment, as a highly corrosive natural environment, poses a significant corrosion threat to offshore platforms,^1,2 submarine pipelines,^3,4 ships,^5,6 ports^7,8 and other marine engineering facilities.^9,10 Long-term exposure to the marine environment will lead to a decline in the structural strength of marine engineering facilities, affecting their safety and stability and even causing safety accidents, resulting in significant loss of personnel and property.^11,12 With the continuous progress of marine engineering technology, a variety of metals and alloy materials are widely used to adapt to the corrosive marine environment.^13–16 This trend inevitably results in the formation of a galvanic coupling corrosion effect between two or even more different metals, which leads to severe galvanic coupling corrosion.^17,18

The corrosion weight loss method^19,20 is a classical corrosion rate measurement method that calculates the corrosion rate by weighing the difference in mass of a metal specimen before and after corrosion. In galvanic corrosion experiments, the loss-in-weight method provides direct data on the corrosion rate, providing a quantitative basis for assessing the degree of corrosion of different metal combinations in a given environment. Electrochemical testing methods^21,22 can monitor the electrochemical behaviour of metals in the corrosion process in real time, providing an important basis for in-depth analysis of the corrosion mechanism. In galvanic corrosion experiments, polarisation curve tests can determine the self-corrosion potential, corrosion current density and other parameters of the metal and analyse the difference in electrochemical activity of the metal, thus revealing the driving force of galvanic corrosion, while the electrochemical impedance spectroscopy (EIS) test can study the formation and destruction process of the metal surface film, further revealing the corrosion mechanism. Wire beam electrode (WBE)^14–16,23 technology can provide electrochemical parameters such as current density/potential at each point on the electrode surface, and then the corrosion rate, corrosion depth and other parameters of local corrosion can be calculated, and the corrosion inhomogeneity at the electrode/solution demarcation can be effectively characterised, which provides a new way for the study of local corrosion.²⁴ In recent years, WBE technology has been widely used in many local corrosion research fields such as galvanic corrosion and has become an important means of local corrosion monitoring. For the galvanic coupling corrosion problems faced by various types of metal equipment in marine environments, Yang studied the galvanic coupling corrosion behaviour between stainless steel and carbon steel in seawater immersion by WBE technology and found that the marine environment can exacerbate the galvanic coupling corrosion effect that exists between the two and promote the dissolution reaction of the anode. Deng studied the galvanic coupling corrosion behaviour of X70 pipeline steel in seawater thermocline by using vertically aligned WBE. In the early stage, the lower part of the X70 steel was corroded as a major anode, and then polarity reversal took place after the 10th day and the galvanic current increased continuously during the corrosion process. The galvanic coupling corrosion current increased during the process, and the galvanic coupling corrosion effect of steel increased in the vertical direction.

Galvanic corrosion in the ocean is usually characterised by a fast corrosion rate and is not easy to detect, so it has been highly valued in recent years. Traditional studies on galvanic coupling corrosion is mostly focused on two-metal coupling systems. However, in actual marine engineering structures, there are often a variety of metal material combinations, making the three-metal coupling system of galvanic coupling corrosion more complex and representative. Marine galvanic corrosion usually has a fast corrosion rate and is not easy to be found, so in recent years has been highly valued. Traditional galvanic coupling corrosion more focused on two-metal coupling system research, but the actual marine engineering structure, there are often a variety of metal material combinations, three-metal coupling system of galvanic coupling corrosion is more complex and more representative. The aim of this study was to investigate the effect of temperature on galvanic coupling corrosion in a three-metal coupling system and to establish a corrosion model for the three-metal coupling system using the weight loss method, electrochemical tests, and WBE technique.

Materials and media

The experimental materials are selected from three common metal materials used in marine oil and gas gathering systems: 20# carbon steel, H62 brass and 6063 aluminum alloy. 20# carbon steel is widely used in offshore pipelines and platform supports due to its high strength and low cost.^25,26 H62 brass is used for valves, flange gaskets and instrument interfaces because it resists seawater erosion and seals well.^14–16,27 6063 aluminum alloy, which has one-third the density of steel, is used for railings, deck plates and light supports thanks to its resistance to marine atmospheric corrosion.^28,29 In service, the flange joint between a 20# carbon-steel pipe and H62 brass valve, together with the weld between a 6063 aluminum panel and carbon-steel bracket, creates a carbon-steel/brass/aluminum galvanic zone.

The test medium was prepared by manually formulating a seawater solution with reference to the standard,^30,31 and its chemical composition is shown in Table 1.

Table 1.

Chemical components of artificially prepared seawater.

Composition	g/L	Composition	g/L
NaCl	24.53	NaHCO₃	0.201
MgCl₂	5.200	KBr	0.101
Na₂SO₄	4.090	NaBO₃	0.027
CaCl₂	1.160	SrCl₂	0.025
KCl	0.695	NaF	0.003

Table 2 shows the measurement results of dissolved oxygen content in the simulated seawater solution from 0 to 50 °C. It can be seen that the oxygen content decreases rapidly with increasing temperature, and the oxygen content at 0 °C is 13.96 mg/L, which is about twice as much as that at 50 °C, which indicates that the dissolved oxygen capacity of seawater is inversely proportional to the temperature; i.e., the higher the temperature is, the lower the level of oxygen content is.

Table 2.

Variation of dissolved oxygen content of seawater with temperature.

Temperature (°C)	Dissolved oxygen content (mg/L)	Temperature (°C)	Dissolved oxygen content (mg/L)
0	13.96	30	7.61
10	11.20	40	6.44
20	9.01	50	5.73

Experimental set-up

In this experiment, a homemade three-metal WBE device was used to investigate the non-uniformity of galvanic corrosion under different temperature conditions. The experimental set-up is shown in Figure 1. The current density changes on each electrode wire were monitored by connecting the wire bundle electrodes to the CST520 wire bundle electrode current potentiometer so that the data could be displayed in real time on the computer screen.

WBE preparation

Figure 2 shows the fabrication of the WBEs. First, 20# carbon steel, H62 brass and 6063 aluminum alloy rods were machined to Φ2 mm × 25 mm. One end of each rod was bevelled and welded to a data collection lead. Heat-shrink tubing insulated the joint and separated the wires. The rods were then inserted into holes in a custom resin plate, with adjacent holes spaced 1 mm apart. After fixing, the epoxy resin adhesives A and B were mixed at a 5:1 ratio, and the resin plate with metal rods was placed in a prefabricated container and sealed by pouring in the mixed epoxy resin adhesive.^24,32

Corrosion weight loss test specimen preparation

With reference to the standard,³³ the weight loss corrosion specimen placement method shown in Figure 3 was designed by ourselves, and the size of the specimens used for the test was 60 mm × 50 mm × 10 mm, and the surface of the specimens was polished by sandpaper, degreased by ethyl acetate and wiped by anhydrous ethanol. The specimens were then placed in a desiccator and left to stand for 7 days. The specimens were connected in series with bolts and nuts, and metal spacers were used between the specimens with a spacing of 5 mm between neighbouring specimens.

Dynamic polarisation curve

The dynamic potential polarisation curve can reflect the relationship between the polarisation potential or overpotential in the electrode reaction and the passing current density, which is one of the most basic methods to study the law of electrode reaction. The kinetic potential polarisation curve test experiments were conducted using the classical three-electrode system. An 800 mL four-hole flask was used as the electrolyte solution container, with a saturated calomel electrode as the reference electrode and a metal platinum sheet as the auxiliary electrode. The working electrodes of 20 # carbon steel, H62 brass and 6063 aluminum alloy were sealed in epoxy resin, leaving a bare area of 100 mm². Firstly, during the test process, the amplitude of the open-circuit potential change of the working electrode was less than 5 mV/s, and the potential range of scanning was determined according to the open-circuit potential of ±500 mV, and the scanning rate was 1 mV/s. The results of the test were fitted with the curve using the CS Studio software accompanying the electrochemical workstation.

EIS

EIS, also known as AC impedance, is based on the principle of applying a small-amplitude sinusoidal electrical signal to the system to obtain the sinusoidal response of the electrode system to the perturbation and then applying a series of angular frequencies to obtain the frequency response function values of the system to make up the electrochemical impedance spectrum. Kou studied the effect of CO₂ content on the corrosion non-uniformity of 20# steel in oil–water flow using the EIS. The frequency range was 10⁻²–10⁵ Hz, and ZView software was used for data fitting.³⁴

Results and discussion

Effect of temperature on the corrosion behaviour of single metals

The polarisation curves of 6063 aluminum alloy, 20# carbon steel and H62 brass are shown in Figure 4. For 20# carbon steel, under increased seawater temperature, the polarisation curves as a whole move to the lower right, the self-corrosion potential decreases, the self-corrosion current density (Figure 5(b)) increases significantly, and the corrosion process is accelerated, which is obviously affected by temperature. And the anodic polarisation curve at higher temperatures shows passivation characteristics. The reason is the rapid corrosion reaction on the surface of carbon steel to form a large number of metal oxides, hindering the expansion of corrosion to the interior. The internal metal matrix has a certain protective effect, but the overall corrosion rate is still maintained at a very high level. The 6063 aluminum alloy self-corrosion potential compared to 20 # carbon steel is more negative, and there is a more pronounced passivation area. The self-corrosion potential of H62 brass is always kept near −300 mV, which is much higher than the other two metals. In addition, compared with that of 20# carbon steel and 6063 aluminum alloy, the polarisation curve of H62 brass has obvious oxygen limit diffusion current density, which is manifested in the cathodic polarisation curve. There is a section parallel to the longitudinal coordinate of the region, and the higher the temperature, the wider the influence of the limit current density. This indicates that when H62 brass is used as the dip pair cathode, the diffusion and reaction rate of dissolved oxygen determine the electrochemical reaction rate on its surface. The absence of an obvious oxygen-limited diffusion current density at 0 °C is due to the fact that at 0 °C, the diffusion coefficient of dissolved oxygen is smaller, the activation energy of the electrode reaction is higher, the kinetic process of the electrode reaction is inhibited, and the reaction rate is slower. This results in the current density remaining low at higher potentials, making it difficult to observe a significant oxygen-limited diffusion current density. With the increase of seawater temperature, the polarisation curve moves slightly to the lower right, proving that the self-corrosion current density of H62 brass is positively correlated with the temperature, and with brass being a corrosion-resistant class of metal, the cathode and anode electrochemical reaction rates are slower, and the accelerating effect of the temperature increase on the corrosion process is greater than the inhibition caused by the decrease of dissolved oxygen concentration.

Figure 5(a) shows the self-corrosion potentials of 6063 aluminum alloy, 20# carbon steel and H62 brass after stabilisation by immersion in seawater at a temperature of 0–50 °C. The potentials of 20# carbon steel and H62 brass all shift negatively with increasing temperature, whereas that of 6063 aluminum alloy shows a tendency to first increase and then decrease, with an inflection point temperature of 20 °C. The significant differences in self-corrosion potentials of the three metals at different temperatures indicate the theoretical existence of a strong electric coupling driving force. The H62 brass potential (−0.26 to −0.32 V) is significantly higher than that of the other two metals, and therefore, H62 is selected to be the cathode of the coupling system. In contrast, 20# carbon steel and 6063 aluminum alloy self-corrosion potentials are more negative, and the difference is not large. Based on the potential difference alone, it is not possible to completely determine the electrode polarity between the three and the strength of the coupling effect.

Effect of temperature on the corrosion behaviour of trimetallic coupling systems

Superposition of polarisation curves

Superimposing the dynamic potential polarisation curves of the three metallic materials at a seawater temperature of 50 °C (Figure 6) reveals that the polarisation characteristics of the three metals are very different. According to the above analysis, in the three-metal coupling pair where brass acts as the cathode, its stronger cathodic polarisation rate means that even a smaller polarisation current can cause a significant potential shift. This suggests that overall coupling potential of the three-metal system tends to approach that of 20 # carbon steel and 6063 aluminum alloy, thereby weakening the electric coupling corrosion driving force. At the same time, it is also found that there are obvious anodic passivation areas after the superposition of the polarisation curves of 20# carbon steel and 6063 aluminum alloy, which are protected by the oxide film to different degrees, and there is a certain degree of resistance to the anodic dissolution reaction.

Coupling potential and current tests

The experiment assumes that 20# carbon steel and 6063 aluminum alloy are the anode (the galvanic couple current is positive) and H62 brass is the cathode (the galvanic couple current is negative), and the electrochemical workstation is used for the determination of the galvanic couple potential and current of the complex dipole system of the three couplings. Using the electrochemical workstation, the three-metal coupling system was monitored at a temperature of 30 °C. The changes in the electric coupling potential Eg and the electric coupling current Ig over 240 h are shown in Figure 7. It can be seen that the electric coupling potential and current of the three-metal dipole system tend to be stable after being immersed for about 50 h. The electrodes of 6063 aluminum alloy and 20# carbon steel showed anodic corrosion current, indicating that both were corroded when acting as the anode of the dipole system. Their anodic currents stabilised at about 0.071 and 0.036 mA·cm⁻², respectively, and the cathodic current was stabilised at −0.107 mA·cm⁻² when H62 brass served as the cathode, and the potential Eg of the three-metal dipole system remained stable at around −0.88 V. The cathodic and anodic polarity characteristics of the three metals also verified the prediction in the polarisation curve superposition method.

From the graph analysis, we can see the following:

E(6063 aluminum alloy)<E(20# carbon steel)<E(mixed potential)<E(H62 brass)

$\sum Ig = 0$ . Ig| (H62 brass)|=Ig|(20# carbon steel)|+Ig|(6063 aluminum alloy)|.

It is shown that this complex coupling system obeys the mixed potential theory; i.e., the current of the coupling system does not flow to the outside, the anodic and cathodic reaction rates are equal, and the mixed potential is located between the cathodic and anodic potentials.

Corrosion weight loss method

The loss-of-weight method involves placing a corrosion specimen in a specific corrosive environment for a set period of time and then measuring its weight before and after the exposure to find the weight change, which is used to calculate the average corrosion rate of the metal in a particular environment. The loss-of-weight results of the 6063 aluminum alloy/20 # carbon steel/H62 brass coupling system after 15 days of immersion at different temperatures, along with the corresponding self-corrosion results for comparison, are shown in Figure 8. It can be seen that as the temperature increases, the corrosion weight loss of 6063 aluminum alloy and 20# carbon steel in the coupling system increases, and the corrosion rate of aluminum alloy is more obviously affected by temperature. At seawater temperatures of 0 to 50 °C, the 6063 aluminum alloy corrosion rate increased from 0.0792 to 0.5754 mm/year, while the 20# carbon steel corrosion rate increased from 0.0835 to 0.3135 mm/year, indicating an obvious anodic corrosion acceleration effect. Meanwhile, H62 brass as the cathode is protected to different degrees, but there is still the phenomenon of its own dissolution.

WBE technology

This technique uses the principle of calculus to divide the surface of the object to be measured into a number of independent tiny electrodes, with each microelectrode being insulated from each other. The entire metal specimen surface is simulated by arranging and encapsulating these microelectrodes in an orderly and close arrangement, so as to obtain the current potential information of the individual electrode wires for the study of the non-uniform corrosion behaviour of the metal surface.^35,36

Figures 9 to 11 show the electrode potential distributions of the 6063 aluminum alloy/20# carbon steel/H62 brass three-metal coupling system in seawater at different temperatures of 10, 30 and 50 °C, where the 6063 aluminum alloy is located in columns 1–4 of the WBE, the 20# carbon steel is located in columns 5–8 of the WBE, and the H62 brass is located in columns 9–12 of the WBE. As can be seen from Figure 9, when the seawater temperature is low (10 °C), the electrode potentials of the metals after immersion for 1 h are generally very high, among which the potential of 6063 aluminum alloy is relatively negative (−0.74 to −0.82 V), and at this time, the difference between the potentials of the aluminum alloy and the other two metals is the largest, but with the continuous prolongation of the immersion time, the electrodynamic coupling effect continues to diminish, which is manifested in the increase in the potentials of the aluminum alloy and the electrode potentials of the 20 # carbon steel and the H62 brass. The electrode potential of 20# carbon steel and H62 brass shifted negatively. After 96 h of immersion, the potential difference between the three metals became very small, but some parts of the aluminum alloy WBE still show a low potential, acting as the main anode of the dipole system corrosion. At this time, the electric coupling effect is very weak, and the corrosion rate is very small.

When the temperature increased to 30 °C (Figure 10), the potential difference and the electric coupling effect of the three metals at the beginning of corrosion were very obvious. The aluminum alloy showed more negative electrode potential near the side of carbon steel, and the main anodic area was concentrated in columns 3 and 4 within 24 h. Different from the case of 10 °C, the aluminum alloy electrode potential was continuously decreasing, and the corrosion resistance was reduced in the experiment within 96 h, which indicated that the protective ability of the oxide film was obviously weakened. During polarisation, carbon steel and brass electrode potentials continue to shift negatively, but the brass potential change is greater, decreasing from −0.63 V at 1 h to −0.80 V after 72 h or so. As corrosion progressed, the potentials of the two metals gradually converge, and after 72 h, the potential difference between the two became very small. In some areas, the potential of the brass electrode was even lower than that of the carbon steel.

When the seawater temperature is 50 °C (Figure 11), the electrode potential of aluminum alloy is initially very high (−0.68 to −0.78 V) after being immersed for 1 h. This is because, at high temperatures, the oxide film on the aluminum alloy forms rapidly, which enables it to protect the aluminum alloy substrate well. At the same time, the carbon steel shows the most negative potential, and as a main anodic area, the comprehensive corrosion occurs in a more uniform way. However, with the rapid depletion of oxygen in seawater, the oxide film generation rate on the aluminum alloy surface becomes much slower than its destruction rate, so the potential of the aluminum alloy decreases sharply, reaching its most negative value after 48 h of immersion, with some WBE becoming a corrosion concentration area due to complete exposure.

With the continuous reduction of oxygen at the cathode, the concentration of dissolved oxygen is further reduced, the cathodic oxygen reduction limit current density decreases, and the corrosion rate is obviously controlled by the cathode, which leads to the potential of the aluminum alloy increasing again at the late stage of corrosion. As an intermediate metal, carbon steel shows potential change, and aluminum alloy basically remains the same; it first decreases and then increases. In addition, the electrode potential of brass part was negatively correlated with the oxygen concentration, and during the whole immersion process, its potential was rapidly polarised to −0.78 to −0.81 V at 1 h and then gradually increased to −0.71 to −0.74 V after 96 h. The corrosion rate was controlled by the cathodic current density, resulting in the corrosion of aluminum alloy in the late stage.

Figure 12 shows the lateral changes of the column average electrode potentials of 6063 aluminum alloy/20# carbon steel/H62 brass three-metal coupling system at different temperatures after 12 h of experiment. It can be clearly seen that with the increase of temperature, the electrode potentials of the three metals have decreased to different degrees, indicating that the activation energy of the corrosion process decreases and the corrosion resistance of the dipole system decreases. The electrode potential of 6063 aluminum alloy is lower than that of 20# carbon steel and H62 brass, which is corroded as the main anode of the dipole system.

In addition, the fourth column electrode of 6063 aluminum alloy is particularly affected by the dipole effect, the average electrode potential is the most negative, and the corrosion localisation is obvious. 20# carbon steel, as an intermediate metal, has a potential that is generally between the potentials of aluminum alloy and brass, but at 30 °C, the case is different because brass has a high cathodic polarisation rate at this temperature, and the rapid polarisation process makes its potential even lower than that of carbon steel. Polarity has a cathodic shift during the corrosion process. In addition, the decrease of the electrode potential of the metal during the experiment also leads to the decrease of the mixed theoretical potential of the whole dipole system.

EIS

Figure 13 shows the results of the impedance spectra of 6063 aluminum alloy self-corrosion, coupling with H62 brass and the three-metal coupling system when the seawater temperature is 50 °C. The results show that the Nyquist curve of the impedance spectra of 6063 aluminum alloy self-corrosion has the largest radius of the capacitance arc, and the capacitance arc of a bimetallic coupling system has a slightly smaller radius than that of a single metal, while the capacitance arc of a trimetallic coupling system has the smallest radius, which shows that the aluminum alloy corrodes most seriously in the trimetallic coupling system, which proves the accuracy of the WBE results, indicating that 20# carbon steel, as an intermediate potential metal, did not slow down the corrosion of aluminum alloy. On the contrary, because of its electrochemical properties and its large difference to aluminum alloy, as the anode, it exacerbates the corrosion rate of aluminum alloy.

Corrosion morphology and physical phase analysis

Figure 14 shows a comparison of the microscopic corrosion morphology of the specimen surface of 6063 aluminum alloy in trimetallic and bimetallic (6063 aluminum alloy/H62 brass) coupling systems at 50 °C. Compared with those on the bimetallic coupling pair, the corrosion products on the surface of the aluminum alloy in the trimetallic coupling system are granular with a more obvious longitudinal shape and larger lumpy products (Figure 14(a)), whereas the corrosion products of the bimetallic one are more compact and spherical with more hairy flocculent products (Figure 14(b)). This indicates that the corrosion of the anode in the trimetallic dipole system is more inhomogeneous, the product structure is sparser, and the different regions cause more serious corrosion non-uniformity due to the different media environments.

The results of the physical phase analysis (Figure 15 and Table 3) show that the elemental composition of corrosion products and self-corrosion of aluminum alloy in three-metal couples are very different. According to the previous conclusions, after the composition of the couples with other low-potential metals, on the one hand, the driving force of galvanic coupling corrosion increases, and the corrosion reaction is accelerated, and on the other hand, oxygen-absorbing corrosion occurs in the cathode and consumes a large amount of dissolved oxygen, which results in the decrease of the passivation film generation rate on the surface of the aluminum alloy. Expressed in the elemental composition of corrosion products as the main component of passivation film, O content is greatly reduced from a self-corrosion case of 72.33% to 30.81%, while causing more serious corrosion of the aluminum alloy surface, the content of elemental Al is as high as 63.71%, reflecting that the corrosion products in Al(OH)₃ become the main component. Al₂O₃ content decreases, indicating that the oxide film has been unable to form an effective protection. Chen et al. found that loose and porous outer products let Cl⁻ penetrate easily. This matches our energy-dispersive X-ray spectroscopy (EDS) data: oxygen drops sharply and Al(OH)₃ rises. Both results show that poor product density speeds up anodic dissolution.³⁷

Table 3.

Phase analysis results (atomic number normalisation%).

Elemental composition	Self-corrosion	H62/20#/6063
Al	26.35	63.71
O	72.33	30.87
Si	0.15	2.84
Mg	1.12	1.41
Mn	0.04	1.06
Zn	0.01	0.10

Corrosion modelling of three-metal coupled systems

Table 4 shows the corrosion current density of 6063 aluminum alloy in the H62 brass/20# carbon steel/6063 aluminum alloy three-metal coupling system at different temperatures. The two remain clearly positively correlated, indicating that temperature has a dominant accelerating effect on galvanic coupling. This occurs because increasing temperature reduces the overpotential of the oxygen reduction reaction while accelerating both cathodic and anodic reaction rates. As a result, the anodic dissolution rate increases, and the promoting effect of temperature outweighs the corrosion mitigation effect caused by the reduced dissolved oxygen content.

Table 4.

Corrosion current density of 6063 aluminum alloy in three-metal couples under different temperature conditions.

Experimental methods	0/°C	10/°C	20/°C	30/°C	40/°C	50/°C
Ig(weight loss method)/μA·cm⁻²	21.36	27.54	39.41	60.10	83.21	155.12
Ig(polarisation curve superposition)/μA·cm⁻²	23.41	30.57	42.36	63.74	95.12	177.41
Ig(couple current test)/μA·cm⁻²	21.82	30.89	37.17	62.69	86.37	164.48
Ig(wire beam electrode)/μA·cm⁻²	20.34	25.68	36.14	50.14	72.36	129.65

The equations for corrosion current density versus temperature in Table 4 were fitted and listed in Table 5, and again an exponential relationship was found; i.e., the higher the temperature, the more pronounced the accelerating effect of elevated temperature on the rate of corrosion of the dipole system anode.

Table 5.

Fitting results of the curve of anodic corrosion current density versus seawater temperature for three metal couples.

Experimental methods	Fitting result	Goodness of fit
6194 weight loss method	Ig = 15.018exp(0.0197T)	R² = 0.983
Polarisation curve superposition	Ig = 15.416exp(0.048T)	R² = 0.985
Electrochemical workstation testing	Ig = 14.798exp(0.048T)	R² = 0.979
Wire beam electrode	Ig = 14.3369exp(0.043T)	R² = 0.981

The experiment will get different dipole systems in 6063 aluminum alloy anodic corrosion current density to the natural logarithm as the bottom of the logarithm, and draw with the temperature change curve shown in Figure 16, can be seen in the three-metal coupling system of anodic corrosion current density is much higher than the two combinations of bimetallic dipole composition of the anodic current density, and is slightly higher than the system of two anodes respectively with the cathode after the dipole of the anodic corrosion current density and the sum of the two anodes. When the three metals are coupled, they create multiple galvanic circuits in parallel. This parallel arrangment lowers the overall electrical resistical resistance of the system and thus accelerates the corrosion reaction.

Therefore, compared with the traditional corrosion process of bimetallic couples, there is an amplification of the galvanic coupling corrosion effect on the anode reaction after trimetallic or even multimetallic coupling. Due to the requirements of high safety and high concentration of marine equipment, the influence of multiple metal laps and couplings on the anode metal should be considered in a comprehensive manner, and at the same time, the multimetallic coupling system cannot be regarded as the sum of multiple bimetallic couplings in a simple way. Wang et al. used a multi-electrode array to study top-of-line corrosion in pipelines. They found that multimetallic coupling greatly increases the local anodic current density. This matches our WBE results.^14–16

Conclusions and recommendations

In this paper, the self-corrosion and galvanic coupling corrosion behaviours of 6063 aluminum alloy, 20# carbon steel and H62 brass, which are typical metal materials in offshore oil and gas engineering, have been investigated in seawater at different temperatures (0–50 °C) by using electrochemical methods, corrosion loss of gravity and WBE technique. The main conclusions are as follows:

Single-metal polarization curves showed that H62 brass and 20# carbon steel lose corrosion resistance as seawater warms from 0°C to 50°C. For 6063 aluminum alloy, the corrosion rate rises until 20°C and then falls quickly; 20°C is the turning point. The reason is related to the dual role of temperature on the corrosion process; the polarisation curve superposition and the three-metal coupling system of the coupled electric current test results show that E(6063 aluminum alloy)<E(20# carbon steel)<E(mixed potential)<E(H62 brass) and Ig|(H62 brass)|=Ig|(20# carbon steel)|+Ig|(6063 aluminum alloy)|, which indicates that the three-metal coupling system follows the theory of mixed potential; i.e., the current of the coupling system does not flow outward, and the rate of the anodic electric reaction is equal to the rate of the cathodic reaction, and the mixed potential is located between the cathodic and anodic potentials.

Through the weight loss method, it is found that with the increase of temperature, the corrosion weight loss of 6063 aluminum alloy and 20# carbon steel in the dipole system increases, and the corrosion rate of aluminum alloy is more obviously affected by temperature. The seawater temperature increases from 0 to 50 °C, the corrosion rate of 6063 aluminum alloy increases from 0.0792 to 0.5754 mm/year, and the corrosion rate of 20# carbon steel increases from 0.0835 to 0.3135 mm/year, making the acceleration effect of anodic corrosion obvious.

Through the use of the WBE technique to study the lateral change of the column average electrode potential of the three-metal dipole system after 12 h of experiment at different temperatures, it can be found that with the increase of temperature, the electrode potentials of the three metals have different degrees of decline, indicating that the activation energy of the corrosion process decreases, and the corrosion resistance of the dipole system decreases. The 6063 aluminum alloy electrode potential is lower than that of the 20# carbon steel and the H62 brass, and as a dipole system, the main anode was corroded. The anodic corrosion process has obvious localisation characteristics. The corrosion hazard is very great. The dipole in the larger potential difference driven by the first destroyed. 20# carbon steel sits between aluminum alloy and brass in potential. At 30°C, however, brass gains a strong cathodic polarization rate. Its potential drops below that of carbon steel. As a result, the carbon steel surface switches from cathode to anode during the polarization test. There is a cathodic–anodic shift in the process of corrosion of carbon steel.

When three or more mentals are coupled, the anodic response is amplified. This electric coupling effect must be judged by looking at all the mental overlaps and their joint action on the anode. A multi-metal system cannot be treated as just the sum of several single pairs.

Footnotes

ORCID iDs

Wei Li

LinJun Wu

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

References

Melchers

Tan

. Long-term corrosion of abandoned offshore steel infrastructure. Corros Eng, Sci Technol 2023; 58: 712–722.

Lin

Deng

, et al. Galvanic corrosion of E690 offshore platform steel in a simulated marine thermocline. Metals (Basel) 2024; 14: 287.

Meng

. Evaluation of the susceptibility of a submarine pipeline material to galvanic corrosion in the Bohai Sea. Coat Prot/Tuceng yu Fanghu 2023; 44:1-4.

Hou

Zhang

Wang

. Current status and future of marine corrosion protection. Proc Chin Acad Sci 2016; 31: 1326–1331.

Refait

Paul

. New frontiers in marine corrosion research. npj Mater Degrad 2025; 9: 11.

Zhang

. Study on the influence of galvanic corrosion on ship structural materials. IOP Conf Ser Earth Environ Sci 2019; 252: 022015. (6pp).

Peng

Liu

Guo

. Corrosion behaviour and law study of two stainless steels in port seawater environment. Equip Environ Eng 2020; 17: 76–83.

. Corrosion and anti-corrosion of oil pipelines in harbours. Chem Manage 2021;14: 142–143.

Xin

. Research progress on corrosion and protection technology of seawater pipelines in ships. Equip Environ Eng 2018; 15: 98–101.

10.

Yang

Khan

Thodi

et al. Corrosion induced failure analysis of subsea pipelines. Reliab Eng Syst Saf 2017; 159: 214–222.

11.

Chen

Zhang

Soares

. The effect of general and localized corrosions on the collapse pressure of subsea pipelines. Ocean Eng 2022; 247: 110719.

12.

Zhou

Gong

Yuan

, et al. Theoretical analysis of the collapse behaviour of subsea pipelines with corrosion defects under external pressure. Ocean Eng 2024; 307: 118069.

13.

Wang

Zhang

Qian

, et al. Galvanic corrosion induced localized defects and resulting strength reduction of circular steel tubes under axial compression: an experimental study. Thin-Walled Struct 2020; 154: 106881.

14.

Wang

Xia

. Simulation study of coupled corrosion of dissimilar metal pipelines in seawater pipeline systems. Equip Environ Eng 2023; 20: 64–71.

15.

Wang

Tan

Zhu

. Probing top-of-the-line corrosion using coupled multi-electrode array in conjunction with local electrochemical measurement. Corros Eng, Sci Technol 2023; 16:7.

16.

Wang

Tan

Zhu

. Probing top-of-the-line corrosion using coupled multi-electrode array in conjunction with local electrochemical measurement. Corros Eng, Sci Technol 2023; 58: 131–140.

17.

Cheng

Pan

Wang

et al. Micro-galvanic corrosion of Cu/Ru couple in potassium periodate (KIO₄) solution. Corros Sci 2018; 137: 184–193.

18.

Dong

Song

Cai

, et al. Study on galvanic coupling corrosion of titanium–aluminium connectors under simulated marine atmospheric environment. Surf Technol 2024; 53: 11–21.+8.

19.

Zhao

Liao

. Experimental teaching design of to determine the corrosion rate of metal. Appl Math Nonlinear Sci 2024; 9:1.

20.

Ali

NURDIN

Mustapa

Vega

, et al. A comparative study on corrosion rate of carbon steel in NaCl solution with continuous and discontinuous weight loss methods. Key Eng Mater 2022; 930: 35–41.

21.

Neshati

Adib

Sardashti

. Stress corrosion cracking detection of sensitized stainless steel 304 in chloride media by using electrochemical impedance spectroscopy (EIS). J Pet Sci Technol 2012; 2: 55–60.

22.

Hampel

Schenderlein

Schary

. Efficient detection of localized corrosion processes on stainless steel by means of scanning electrochemical microscopy (SECM) using a multi-electrode approach. Electrochem Commun 2019; 101: 52–55.

23.

Pan

Zhang

Xing

. Localized corrosion behavior of T2 copper and H62 brass in NaCl solution for semiconductor applications: an electrochemical study using wire beam electrode. J Phys: Conf Ser 2025; 2941: 012052–012052.

24.

Kou

Zhang

. Progress of galvanic coupling corrosion research based on wire beam electrode technology. Mater Guide 2023; 37: 160–168.

25.

Praveen

Arijit

Jagmohan

. Evaluation of galvanic corrosion in mild steel, 309S and 409M alloy combinations for marine environment applications. Eng Res Express 2024; 6: 045008–045008.

26.

Philippe

Anne

Marc

, et al. Corrosion of carbon steel in marine environments: role of the corrosion product layer. Corros Mater Degrad 2020; 1: 198. -.

27.

Gabriela

Vieira

Aline

. Methods of corrosion prevention for steel in marine environments: a review. Results Surf Interfaces 2025; 18: 100430–100430.

28.

Prabhu

Rao

. Corrosion behaviour of 6063 aluminium alloy in acidic and in alkaline media. Arabian J Chem 2017; 10: S2234–S2244.

29.

Sureshkumar

Jagadeesha

Natrayan

, et al. Electrochemical corrosion and tribological behaviour of AA6063/Si₃N₄/Cu(NO₃)₂ composite processed using single-pass ECAPA route with 120° die angle. J Mater Res Technol 2022; 16: 715–733.

30.

Gao

Zou

Zhao

, et al. Sand impingements on cathodic protection of marine carbon steel in natural sea water. Corros Eng, Sci Technol 2023; 58: 12–22.

31.

GB/T 15748-2013 . Test method for galvanic coupling corrosion of marine metallic materials [S].

32.

Tan

. An overview of progresses and challenges of electrochemically integrated multi-electrode arrays for probing localised corrosion in complex environmental conditions. Corros Eng, Sci Technol 2023; 58: 819–837.

33.

HB 5374-1987 . Methods for determining the current of different metal galvanic couplings [S].

34.

Kou

Yang

et al. Effect of CO₂ content on corrosion non-uniformity of 20# steel in oil–water two-phase flow based on wire beam electrode technique. Corros Eng, Sci Technol 2022; 58: 1–13.

35.

Chen

Liu

. Investigation on localized corrosion behavior of pipeline steel under AC interference in alkaline earth metal environment using wire beam electrode. J Electroanal Chem 2024; 973: 118668.

36.

Wang

Zhong

. Studies of the corrosion behaviours of high-strength alloy steel in an environment containing Cl-using a wire beam electrode (WBE) and EIS techniques. Int J Electrochem Sci 2022; 17: 221153.

37.

Chen

Liu

Yang

, et al. Corrosion behaviours of Ni–Mo–Cr alloy and 316 stainless steel in molten NaCl–KCl–AlCl₃ salts at high temperature. Corros Eng, Sci Technol 2023; 58: 103–107.