A method for predicting the corrosion resistance of magnesium

Abstract

The application discloses a method for predicting corrosion resistance of magnesium. The method comprises the following steps: determining the relationship between corrosion potential U and background charge Q based on a corrosion model of pure magnesium in water; determining the relationship between corrosion current density i and background charge Q; coupling the corrosion potential U and the corrosion current density i through the background charge Q to obtain a polarization curve of a corrosion process. The corrosion model established by the application can comprehensively consider the influence of Delta U and work function on corrosion, thereby distinguishing the corrosion resistance difference between different crystal surfaces. The application can analyze the corrosion resistance difference between different crystal surfaces of pure magnesium in the corrosion process in water by fitting the polarization curve, can predict the corrosion resistance of magnesium without any experience parameters, and has high effectiveness and accuracy.

Description

Technical Field

[0001] This invention belongs to the field of metal corrosion performance research and relates to a method for predicting the corrosion resistance of magnesium. Background Technology

[0002] As the lightest structural metal, magnesium boasts advantages such as high strength, high mass ratio, and good shock absorption and noise reduction properties. Furthermore, magnesium alloys exhibit good biocompatibility, making them widely used in aerospace, portable electronic devices, automobiles, and medical devices. However, compared to other structural metals, magnesium's high chemical reactivity and poor corrosion resistance severely limit the further application of magnesium alloys as structural materials.

[0003] Surface coatings are considered one of the most effective methods to prevent corrosion of many materials. However, for pure magnesium, the corrosion resistance of magnesium does not improve significantly after adding these conventional coatings, and may even decrease severely, rendering it unusable. Alloying is another effective solution to improve the corrosion resistance of magnesium. For example, adding a low amount of Ca to magnesium can significantly reduce the cathodic corrosion rate, thus exhibiting better corrosion resistance than pure magnesium. However, it is difficult to elucidate the corrosion mechanism of magnesium alloys through experiments alone, because it is difficult to gain a deep understanding of the complex corrosion process from an atomic or electronic perspective. Currently, elucidating the influence of alloying elements on corrosion behavior experimentally remains a challenging task. Therefore, establishing corrosion models and theoretically studying the atomic and electronic information containing the inherent complex corrosion environment has become an urgent problem to be solved in this field.

[0004] The rapid development of density functional theory (DFT) has enabled the calculation of many key parameters characterizing corrosion resistance, such as corrosion interface structure, work function, and electronegativity. Traditional theory holds that a higher work function of the material surface obtained from the electrostatic potential corresponds to stronger corrosion resistance. However, recent experiments have shown that although the work function of a pure magnesium substrate (0001) is slightly smaller than that of a cylindrical (10⁻¹⁰) substrate (3.66 eV vs. 3.67 eV), the former exhibits 15-20 times higher corrosion resistance. This conclusion directly demonstrates that using only the work function of a material to determine its corrosion resistance is insufficiently accurate.

[0005] Because corrosion is a complex process, it is often necessary to overcome the thermodynamic studies that involve kinetics. The work function obtained solely from thermodynamic calculations cannot accurately express the corrosion resistance of a material. Therefore, establishing a polarization curve that incorporates kinetics is essential. In 2016, Taylor [Taylor CDA First-Principles Surface Reaction Kinetic Model for Hydrogen Evolution under Cathodic and Anodic Conditions on Magnesium[J]. Journal of the Electrochemical Society, 2016, 163: C602-C608.] proposed a polarization curve model that simultaneously incorporates thermodynamics and kinetics. This model is based on a reasonable conjecture of the classical Arrhenius equation: k = Aexp(-Ea / RT), where the pre-exponential factor A is based on the typical vibrational frequencies of atoms and molecules in the condensed phase. Taylor's bold assumption successfully yielded polarization curves for pure magnesium of the same order of magnitude as experimental results. Based on Taylor's work, Ma et al. [Ma H., Wu L., Liu C., Liu M., Ke W. First-principles modeling of the hydrogen evolution reaction and its application inelectrochemical corrosion of Mg[J]. Acta Materialia, 2019, 183: 377-389.] further modified the pre-exponential factor A and activation energy Ea based on this model. The pre-exponential factor A can be approximated as nFkT / h according to the literature, where n, F, k, T, and h are the number of charges transferred during the electrode reaction, Boltzmann constant, absolute temperature, and Planck constant, respectively. The activation energy Ea consists of two parts: one is the chemical activation energy ΔG, which is physicochemically defined as the difference between the average energy of one mole of activated molecules and the average molar energy of all reactant molecules.The chemical activation energy ΔG cannot be directly obtained. Therefore, Ma et al. made bold assumptions about some of the results. The part of the energy that is difficult to obtain is approximated as 30% of the energy required for reactants to break chemical bonds, lose electrons and generate ions [Cao Chunan. Principles of Corrosion Electrochemistry [M]. 3rd ed. Beijing: Chemical Industry Press, 2008.]. Another part of the activation energy Ea depends on the product of the absolute potential Ue at equilibrium proposed by Trasatti [Trasatti S. The absolute electrode potential: an explanatory note [J]. Pure Appl Chem, 1986, 58: 955-966.] and the transfer coefficient α in the electric double layer in the electrochemical reaction. In the process of electrochemical corrosion, due to the difference in internal potential between the metal conductor phase and the solution ionic phase, there is a potential difference U between the electrode surface and the solution. That is to say, there is an electric field phase boundary region between these two phases, which is usually called the electric double layer. α is the relative position of the activated particles in the transition state in the electric double layer [Cao Chunan. Principles of Corrosion Electrochemistry [M]. 3rd ed. Beijing: Chemical Industry Press, 2008.]. Using the experimentally obtained parameter α, Ma et al. not only successfully determined the anisotropy of magnesium crystal plane corrosion but also predicted the influence of alloying elements on the anodic dissolution rate. The addition of Ga, Cr, Cd, Hg, In, and As reduced the anodic dissolution rate, a result very similar to the experimental findings. However, their calculations showed that adding Ge reduced the corrosion resistance of magnesium, contradicting the experimental results that adding Ge increased the corrosion resistance of magnesium [Liu RL, Zeng ZR, Scully JR, Williams G., Birbilis N. Simultaneously improving the corrosion resistance and strength of magnesium via low levels of Zn and Ge additions[J]. Corrosion science, 2018, 140: 18-29.]. This contradiction arises because the parameter α requires experimental data. When no corresponding experimental value is available, 0.5 is often used as an approximation, which can lead to erroneous results. Therefore, designing a theoretical model without experimental parameters is crucial for the design of high corrosion-resistant magnesium alloys. Summary of the Invention

[0006] The purpose of this invention is to provide a method for predicting the corrosion resistance of magnesium with high effectiveness and accuracy.

[0007] The technical solution for achieving the objective of this invention is as follows:

[0008] A method for predicting the corrosion resistance of magnesium includes the following steps:

[0009] (1) A layer of water molecules was adsorbed at the top position of magnesium to construct a corrosion structure model of magnesium;

[0010] (2) Select a crystal plane of magnesium and introduce a series of depleted background charge amounts Q to simulate the cathode polarization process, and determine the electrode potential U under different background charge amounts Q. M and electrolyte potential U S The contact potential difference ΔU is calculated according to formula (1):

[0011] ΔU=U M -U S (1);

[0012] (3) Introduce a series of depleted background charge quantities Q to simulate the anodic polarization process and determine the crystal plane to be analyzed. The work function of the electrode material is used to calculate the total potential difference between the electrode and the electrolyte solution, i.e., the equilibrium potential Ue, according to formula (2):

[0013]

[0014] (4) Combining formulas (1) and (2), the relationship between corrosion potential U and Q is determined as follows:

[0015]

[0016] (5) The anodic corrosion current density i is correlated with the vacancy formation energy using the Arrhenius formula (4):

[0017] i (Q),anode =i0×exp[(-ΔEvac(Q)) / RT] (4),

[0018] In the formula, i0 is the exchange current density of Mg, ΔEvac is the vacancy energy, R is the gas constant, and T is the temperature;

[0019] Where ΔEvac is obtained through formula (5): ΔEvac=E(N)-(N-1)E(N,0) / N (5),

[0020] In the formula, E(N) is the total energy of a supermonomer containing N atoms, where one atom is occupied by a vacancy, and E(N,0) is N times the energy of one atom in the crystal structure.

[0021] By calculating E(N) under different background charge Q, a series of ΔEvac(Q) are obtained. The corresponding ΔEvac(Q) is substituted into formula (4) to establish the relationship between i and Q in anodic polarization. Each Q value corresponds to a current density i.

[0022] (6) The free energy ΔG of hydrogen atoms adsorbed on the metal surface H* The cathode exchange current is obtained:

[0023] i (Q),cathode = -ek0×exp(-ΔG H(Q) / RT) / 1+exp(-ΔG H(Q) / RT) (6),

[0024] Free energy ΔG H* =ΔE H +ΔE ZPE -TΔS H (7),

[0025] In the formula, ΔE H ΔE ZPE and ΔS H These represent the changes in adsorption energy, zero-point vibration energy, and surface adsorption hydrogen atom entropy, respectively, with T being the temperature.

[0026] The adsorption energy ΔE of hydrogen atoms on the electrode surface H Obtained through formula (8):

[0027] ΔE H =E(*H)-E(*)-1 / 2E(H2) (8),

[0028] Where E(*H) is the total energy of the system after the adsorption of hydrogen atoms, and E(*) is the total energy of the system after the adsorption of no atoms.

[0029] Where ΔE ZPE For Mg, it is approximately 0.01 eV / atom. When T is 298 K, TΔS H The value is approximately -0.21 eV / atom;

[0030] (7) The corrosion potential U(Q) is compared with the corrosion current density i (Q),anode i (Q),cathode By coupling with the background charge Q, the anodic polarization curve and the cathodic polarization curve are obtained. The voltage and charge at the intersection of the anodic polarization curve and the cathodic polarization curve are the corrosion voltage and corrosion equilibrium charge, respectively. Thus, the corrosion potential and corrosion equilibrium charge of the selected crystal plane of magnesium are obtained. The lower the corrosion equilibrium charge, the slower the corrosion rate of the corresponding crystal plane. The higher the corrosion voltage, the more difficult it is for corrosion to occur on the corresponding crystal plane.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] (1) The prediction model constructed in this invention only relies on the calculation results of DFT, so it has high effectiveness and accuracy without any empirical parameters;

[0033] (2) By establishing a corrosion model, this invention can comprehensively consider the influence of U and work function on corrosion, and thus distinguish the differences in corrosion resistance between different crystal planes. Attached Figure Description

[0034] Figure 1 These are the four positions where water molecules are adsorbed on the Mg(0001) surface and their corresponding adsorption energies, namely hollow-fcc, hollow-hcp, bridge, and top positions.

[0035] Figure 2 The structures are the initial and optimized structures of monolayer water molecules (a) and two-layer water molecules (b) adsorbed on the Mg(0001) surface. All water molecules are initially positioned at the top position.

[0036] Figure 3 (a) Electron density difference diagram on the xy plane after adsorption of a single layer of water molecules. The inset is a charge density difference isosurface, with blue corresponding to charge accumulation and yellow corresponding to charge transfer; (b) Electron density difference diagram on the xy plane after adsorption of two layers of water molecules.

[0037] Figure 4 The electrostatic potentials are (a) the adsorption of a single layer of water molecules on the Mg(0001) surface system; and (b) the adsorption of two layers of water molecules on the Mg(0001) surface system. The Fermi level is set to 0 as the reference frame.

[0038] Figure 5 This is the most stable state of water molecules adsorbed on 11 crystal planes. The most stable state of a single water molecule on all crystal planes is adsorbed at the top position.

[0039] Figure 6 It is the electrostatic potential of water molecules adsorbed on the Mg(0001) surface system after a series of background charges are introduced.

[0040] Figure 7 The potential U of the (0001) plane changes with the background charge. The upper half of the figure represents the anodic polarization process, and the lower half represents the cathodic polarization process.

[0041] Figure 8 (a) is obtained by fitting U and background charge Q (0001), (10-10) A (10-10) B (a) Quasi-polarization curves of the (11-20) plane. (b) (0001), (10-10) obtained by formula transformation. A (10-10) B , and the polarization curves of the (11-20) surface.

[0042] Figure 9 These are quasi-polarization curves of nine crystal planes in pure Mg. The voltage and charge at the intersection points are the corrosion voltage and corrosion equilibrium charge, respectively. Detailed Implementation

[0043] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0044] Example 1

[0045] (1) To simulate the corrosion process of magnesium, the adsorption sites of water molecules on the Mg basal surface were first determined. When water molecules are adsorbed on the magnesium basal surface, there may be three highly symmetric adsorption sites, such as... Figure 1 As shown. The literature determines the most stable adsorption position of water molecules by calculating the adsorption energy. The adsorption energy parameter can directly determine the strength of the binding between the molecule and the substrate. The calculation results show that the absolute value of the adsorption energy at the hollow-hcp position is the smallest, only 0.21 eV, indicating that the water molecule adsorption at this position is the least stable; while the absolute value of the adsorption energy at the top position is the largest, reaching 0.55 eV, indicating that the water molecule adsorption at the top position is the most stable. This result is consistent with the conclusion in previous literature that the water molecule adsorption at the top position is the most stable [Williams KS, Labukas JP, Rodriguez-Santiago V., Andzelm JW First Principles Modeling of Water Dissociation on Mg(0001) and Development of a MgSurface Pourbaix Diagram[J].Corrosion,2014,71:209-223.].

[0046] (2) Further consideration was given to the adsorption of multiple water molecules on the metal surface. Since multiple water molecules may have multiple configurations, following the method described in the literature, we simultaneously placed multiple water molecules at the top position for structural optimization (e.g., ...). Figure 2 (a) was used to calculate the adsorption energy of the entire water layer. After adsorbing four water molecules, the adsorption energy of the entire water layer was -1.82 eV, which is higher than the adsorption energy of four independent water molecules adsorbed at the top position (-0.55 eV × 4). This result indicates that there are interaction forces between water molecules, resulting in the water molecules adsorbed on the pure Mg basal surface being in a disordered state. Therefore, when considering the configuration relationship between water molecules and the metal interface, it is necessary to consider the influence of the number of water molecule layers on the structure. To determine the influence of the number of water molecule layers on the adsorption structure, we first optimized the structure with two layers of water molecules placed at the top position of the Mg basal surface, as shown in Figure 1. Figure 2As shown in (b), the optimized water molecule layer structure near the metal substrate and Figure 2 In (a), the structures of the two water molecule layers are almost identical, indicating that the interaction between the two water molecule layers is not significant. To verify the interaction between the two water molecules, we calculated the adsorption energy of the system with two water molecule layers (-1.92 eV). The calculation results show that although the number of water molecules doubled, the overall adsorption energy of the water molecules only differed by 0.1 eV, indicating that the second water molecule layer has a small impact on the stability of the system.

[0047] (3) Further, the influence of the structure of one-layer and two-layer H2O on the system was determined at the microscopic electronic level. We calculated the charge density difference between the two systems, i.e., the charge density after adsorption (ρ). met+mono Subtract the charge density of water molecules before adsorption (ρ) mono The charge density Δρ(z) of the matrix metal and the base metal is given by ρ. met+mono (z)-(ρ met (z)+ρ mono (z)), such as Figure 3 As shown in (a) and (b), the charge density change caused by adsorption is most pronounced between the bottom layer of water molecules and the top layer of metal surface, indicating a strong interaction between them. The charge density in the metal layer far from the surface exhibits a gentle oscillation, suggesting minimal charge redistribution. Furthermore, the charge density increases below the water molecules and decreases on the metal surface, indicating that charge is primarily transferred from the metal substrate to the water molecules. By comparing the difference in charge density at the water / magnesium interface between one-layer and two-layer H2O structures, we found that the charge density changes at the interface are essentially the same for both types of H2O structures. Moreover, Figure 3 (b) The amount of charge transfer between the two water molecules is very small, indicating that the interaction between the two water molecules is not obvious.

[0048] (4) Continuing the analysis of the potential at the water / magnesium interface, according to the definition of corrosion, the electrode potential can be expressed as: charge travels from an infinitely distant surface through the surface charge layer and into the interior of the metal. Therefore, the equilibrium potential ΔU can pass through the Mg plane (denoted as U). M ) and electrolyte solution (denoted as U s The total potential difference between the two sides is obtained by the formula ΔU = U M -U s (1) Obtaining. ΔU represents the potential that electrons need to overcome to escape from the metal layer into the solution. A high ΔU indicates that electrons have difficulty escaping from the metal layer, which means that the metal has strong corrosion resistance. For example... Figure 4As shown in (a), with the Fermi level set to zero as the reference level, we can see that the potential of the metal far from the water layer oscillates synchronously within a certain range. This also proves that there is almost no charge transfer between the metal layers. The electrode potential of Us in a single water layer is -6.86 eV, and the potential of U in the metal layer is... M The electrode potential reached -8.33 eV, and ΔU was 1.47 eV. After adding an additional layer of water molecules, as... Figure 4 As shown in (b), Us and U M The change is small, and it increases synchronously by 0.03 eV. Therefore, ΔU (1.47 eV) is the same as the ΔU value of the monolayer water system. This indicates that the addition of the water layer only affects the relative potential of the water and the metal layer, but does not affect the ability of electrons to escape from the metal layer into the solution. Therefore, in the subsequent calculations, all systems only adsorbed a single layer of water molecules.

[0049] (5) After determining the equilibrium potential U of the basal plane, we continued to calculate the structures of water molecule adsorption on 11 other common crystal planes using the same method, such as Figure 5 As shown. We first place individual water molecules at high-symmetry positions on different crystal planes. For all crystal planes, the initial single water molecule adsorbed at the top position is the most stable adsorption site. Therefore, we simultaneously place a layer (4) of water molecules at the top positions on these crystal planes. Where (1-100) A Water molecules on surfaces (10-11), (10-13), (11-20), (11-22), (11-23), and (21-30) remained in their original positions after optimization. However, water molecules on surfaces (10-12), (11-21), and (1-100) remained in their original positions after optimization. B Water molecules on the (21-31) plane will deviate from their initial positions. To determine the reason for this deviation, we further calculated the surface energies of these crystal planes to assess their stability: Among them, E slab It includes the total structural energy of the upper and lower symmetrical surfaces, therefore the pre-coefficient is set to 1 / 2 to represent the energy of a single surface, where S is the surface area and E is the energy of a single surface. bulk Let be the energy of a ground-state atom, and n be the number of atoms in the bulk material. The basal plane (0001), being the most stable crystal plane, has a surface energy of only 552 mJ / m². 2 This value is the lowest among all crystal planes. (1-100) A and (1-100) B The surface structures are very similar, but due to the huge differences in surface structure, the surface energies of the two are very different, among which (1-100) B The surface energy of the surface is as high as 870 mJ / m 2The surface energy of this crystal plane is the highest among all crystal planes, indicating that this crystal plane structure is unstable, while (1-100) A The surface energy of the surface is only 622 mJ / m 2 It is the most stable crystal plane besides the basal plane. (10-12), (11-21), (1-100) B The surface energies of planes (21-31) are relatively high, so water molecules are unstable when adsorbed on these planes, and the structure will change significantly. At the same time, we note that since these 11 crystal planes are almost all non-close-packed, the distance between water molecules is relatively large, so the relative interaction is very weak. As a result, the adsorption structure of each water molecule on the crystal plane is almost the same, and there is no relative interaction between water molecules.

[0050] (6) The corrosion potential U is determined by the following parameters: ΔU as a function of background charge Q and the work function. A series of depleted background charge values ​​Q are introduced to simulate the cathodic polarization process, and the electrode potential U under different background charge values ​​Q is determined. M and electrolyte potential U S The contact potential difference ΔU is calculated according to formula (1): ΔU=U M -U S (1). A series of depleted background charge quantities Q are introduced to simulate the anodic polarization process, and the crystal plane to be analyzed is determined. The work function of the electrode material is used to calculate the total potential difference between the electrode and the electrolyte solution, i.e., the equilibrium potential Ue, according to formula (2):

[0051] When a background charge Q is introduced, the work function will remain unchanged due to the synchronous changes in the Fermi level and the vacuum level, such as... Figure 6 As shown. Unlike the stable and invariant work function, the value of ΔU changes continuously with the applied background charge, and this change is linear. It is noteworthy that the vacuum level tilts after the introduction of the background charge. This is because the symmetry of the system is completely broken after the introduction of the background charge, preventing the introduction of dipoles and causing the vacuum level to not converge. The tilt of the vacuum level does not affect the calculated value of ΔU, because the calculation of ΔU only requires determining the potential difference between the water molecules and the surface of the metal layer.

[0052] (7) After determining the required parameters, we compared the model with experimental results to determine its accuracy. We selected the base plane and cylindrical surface, which could be directly compared with the experimental results, for calculation. Combining formulas (1) and (2), the relationship between the corrosion potential U and Q was determined as follows: The calculated change of U with Q is as follows: Figure 7As shown in the figure, the upper part represents the electron loss process of the metal at the anode, and we find that the U at the anode increases with the increase of the surface charge Q. The lower part represents the electron gain process of hydrogen ions at the cathode, and the U at the cathode decreases with the increase of the applied charge Q. When the external Q is less than 0.5e, the potentials U at the anode and cathode change linearly with Q, which is consistent with the experimental results showing that the corrosion potential changes linearly with the corrosion current density. However, when the external background charge is greater than 0.5e, the electrode potential deviates from the linear result, which means that adding a large background charge will disrupt the stability of the system.

[0053] (8) By fitting U and Q, we determined the (0001) surface and (10-10) A Noodles, (10-10) B Quasi-polarization curves of the (11-20) plane and the (11-20) plane, as shown Figure 8 As shown in (a), we found that the corrosion voltage U on the (0001) surface... corr Greater than the other three planes (U) e(0001) >U e(10-10)A >U e(11-20) >U e(10-10)B This indicates that corrosion is less likely to occur on the substrate. This result is consistent with experimental results, namely that the substrate has the highest corrosion potential. It is noteworthy that the (10-10) plane is composed of (10-10)... A and (10-10) B Structural composition, in which U corr(10-10)A Than U corr(11-20) High 0.25V, while U corr(11-20) Than U corr(10-10)B The high value is 1.29V. Therefore, the corrosion potential U of the (11-20) plane is... corr Above the mixed (10-10) plane. These results are consistent with experimental results [Song GL, Xu Z. Crystallization and electrochemical corrosion of polycrystalline Mg[J]. Corrosionscience, 2012, 63: 100-112.; Ehsan G., Reza A., Terence GLE effect of crystallographic texture and twinning on the corrosion behavior of Mg alloys: A review[J]. Journal of Magnesium and Alloys, 2021: In Press.], i.e. U corr(0001) >U corr(11-20) >Ucorr(10-10) In addition to the experimental results, our computational model can also distinguish (10-10). A and (10-10) B The corrosion resistance between structures is difficult to characterize experimentally due to their similar atomic arrangements.

[0054] (9) Based on the above results, we only need to correlate the background charge Q with the corrosion current i to obtain an accurate polarization curve. According to formula i (Q),anode =i0×exp[(-ΔE vac(Q) ) / RT] and i (Q),cathode = -ek0×exp(-G H(Q) / RT) / 1+exp(-G H(Q) To obtain the conversion current i, we only need to calculate ΔE under different background charge Q conditions. vac and ΔE H To compare the experimental measurements with our current calculations, we set k0 to 200s. -1 , et al. JK, Bligaard T., Logadottir A., ​​Kitchin JR, Chen JG, Pandelov S., Stimming U. Trends in the Exchange Current for Hydrogen Evolution[J]. Journal of The Electrochemical Society, 2005, 152: J23-J26.] The corrosion current density of different metals was estimated using the same parameters. Furthermore, we coupled the voltage U and current i together through the background charge Q to obtain... Figure 8 The polarization curve is shown in (b). Wherein, the (0001) plane and (10-10) are... A Noodles, (10-10) B Corrosion current i of surface (11-20) and surface (11-20) corr They are 10 -20.6 A / cm -2 10 -19.8 A / cm -2 10 -15.1 A / cm -2 and 10 -13.1 A / cm -2 In these crystal planes, the corrosion current density i corr Corrosion voltage U corr The trends are completely opposite, with the etching voltage U of the crystal plane being... corrThe higher the value, the more likely it is to correspond to i. corr The lower the value. Among all crystal planes, the i of the basal plane is... corr It is the lowest, U corr The highest value indicates that corrosion on the basal surface is not only the most difficult to occur, but also has the lowest corrosion rate after it occurs. Experimental results [Hagihara K., Okubo M., Yamasaki M., Nakano T.. Crystal-orientation-dependent corrosion behavior of single crystals of a pure Mg and Mg-Al and Mg-Cu solid solutions[J]. Corrosion Science, 2016, 109: 68-85.] show that in Hanks equilibrium salt solution, the i values ​​of the pure Mg basal surface (0001), prism surface (11-20), and prism surface (10-10) are the highest. corr 8.17 μA / cm -2 10.00μA / cm -2 and 13.4 μA / cm -2 Considering that the experimental samples are usually polycrystalline materials, and that chloride ions in the experimental environment accelerate the corrosion rate [Atrens A., Shi Z., Umehreen S., Johnston S., Song GL, Chen X., Pan F. Review of Mg alloy corrosion rates[J]. Journal of Magnesium and Alloys, 2020, 8: 989-998.], we calculated the i-th... corr The ratio agrees well with the experimental results, i corr Slightly higher than the experimental results. (By comparison) Figure 8 From the results of (a) and (b), we find that, qualitatively speaking, through i corr The result can be consistent with Q, i.e., i(Q). (0001) <i(Q) (10-10)A <i(Q) (11-20) <i(Q) (10-10)B The reason for these two consistent descriptors is ΔE. vac and ΔE H The linear variation with Q implies a linear relationship between Q and ln(i / i0). Based on the above discussion, we find that to save computational costs, Q can be used instead of the corrosion current density i in the polarization curve. corr .

[0055] Importantly, (10-10)A Noodles, (10-10) B The three characteristic structures of the (11-20) plane and the (11-20) plane often appear mixed in experiments, making it difficult to distinguish their corrosion resistance. Theoretically, the work function is usually used to determine the corrosion resistance of different metals. For these three crystal planes, Work... (10-10)A =Work (10-10)B Work (11-20) This demonstrates that using only the work function to determine the corrosion resistance of these three structures will lead to inaccurate conclusions. This is because the corrosion resistance of a metal depends not only on the work function but also on the contact potential ΔU. For different metals, such as Mg and Pt, the work function varies greatly (3.66 eV vs. 5.65 eV). Under these conditions, the contact potential ΔU can be ignored, and the work function can be used as an empirical formula to determine the corrosion resistance of different metals. However, when determining the corrosion resistance of different surfaces of the same material, the difference in work function is limited, while the difference in ΔU is considerable. Considering both the work function and ΔU, (10-10) A The structure exhibits the best corrosion resistance (10-10). B The structure with the worst corrosion resistance (11-20) has corrosion resistance between (10-10). A and (10-10) B Between structures. In our calculations, by establishing a corrosion model, we can comprehensively consider the influence of ΔU and work function on corrosion, thus distinguishing the differences in corrosion resistance between different crystal planes.

[0056] (10) Next, based on our proposed model, we continue to discuss the electrochemical corrosion behavior of the remaining surfaces that were previously calculated. Figure 9 The quasi-polarization curves for the remaining eight different Mg crystal surfaces are shown. Based on the intersection of the anodic and cathodic polarization curves, the corrosion potential and corrosion equilibrium charge of the pure Mg surface can be obtained. In the above discussion, we have demonstrated that the corrosion equilibrium charge can qualitatively replace the equilibrium current in determining the corrosion resistance of the material. The (0001) surface has the lowest corrosion equilibrium charge, indicating the slowest corrosion rate; the (10-12) surface has the highest corrosion voltage, indicating that corrosion is most difficult to occur on this surface; the (21-30) surface has the highest corrosion equilibrium charge and the lowest corrosion voltage, indicating that the (21-30) surface is not only prone to corrosion but also has the fastest corrosion rate after corrosion occurs. These conclusions are completely consistent with the results obtained from existing theoretical work using polarization curves fitted with experimental parameters, demonstrating that the model can obtain reasonable results without using experimental parameters.

Claims

1. A method for predicting the corrosion resistance of magnesium, characterized in that, Includes the following steps: (1) A layer of water molecules is adsorbed at the top position where the absolute value of magnesium adsorption energy is the largest, and a corrosion structure model of magnesium is constructed. (2) Select a crystal plane of magnesium and introduce a series of depleted background charge quantities Q to simulate the cathode polarization process, and determine the electrode potential U under different background charge quantities Q. M and electrolyte potential U S The contact potential difference ΔU is calculated according to formula (1): ΔU = U M -U S (1); (3) Introduce a series of background charge quantities Q to simulate the anodic polarization process and determine the crystal plane to be analyzed. Let Q be the work function of the electrode material. The total potential difference between the electrode and the electrolyte solution, i.e., the equilibrium potential U(Q), is calculated according to formula (2): U(Q) = (2); (4) Combining formulas (1) and (2), the relationship between the equilibrium potential U and Q is determined as follows: U(Q) = (3); (5) The anodic corrosion current density i is correlated with the vacancy energy using the Arrhenius formula (4): i (Q), anode = i0 exp[( ] (4), In the formula, i0 is the exchange current density of Mg, ΔEvac is the vacancy energy, R is the gas constant, and T is the temperature; ΔEvac is obtained through formula (5): ΔEvac = E(N) (N 1) E(N, 0) / N(5) In the formula, E(N) is the total energy of a supermonomer containing N atoms, where one atom is occupied by a vacancy, and E(N,0) is N times the energy of one atom in the crystal structure. By calculating E(N) under different background charge Q, a series of ΔEvac(Q) are obtained. The corresponding ΔEvac(Q) is substituted into formula (4) to establish the relationship between i and Q in anodic polarization. Each Q value corresponds to a current density i. (6) The free energy ΔG of hydrogen atoms adsorbed on the metal surface H* The cathode exchange current is obtained: i (Q), cathode = -ek0× / (1 + )(6), Free energy ΔG H* = ΔE H +ΔE ZPE -TΔS H (7), In the formula, ΔE H ΔE ZPE and ΔS H These represent the changes in adsorption energy, zero-point vibration energy, and surface adsorption hydrogen atom entropy, respectively, with T being the temperature. The adsorption energy ΔE of hydrogen atoms on the electrode surface H Obtained through formula (8): ΔE H = E(*H)-E(*) - 1 / 2E(H2) (8), Where E(*H) is the total energy of the system after the adsorption of hydrogen atoms, E(*) is the total energy of the system after the adsorption of no atoms, and E(H2) is the total energy of gaseous hydrogen molecules. (7) The corrosion potential U(Q) is compared with the corrosion current density i (Q), anode i (Q), cathode By coupling with the background charge Q, the anodic polarization curve and the cathodic polarization curve are obtained. The voltage and charge at the intersection of the anodic polarization curve and the cathodic polarization curve are the corrosion voltage and corrosion equilibrium charge, respectively. Thus, the corrosion potential and corrosion equilibrium charge of the selected crystal plane of magnesium are obtained. The lower the corrosion equilibrium charge, the slower the corrosion rate of the corresponding crystal plane. The higher the corrosion voltage, the more difficult it is for corrosion to occur on the corresponding crystal plane.

Images