New evaluation method for the characterization of coatings by electroerosive alloying

The running-in coatings were formed on the surface of tin bronze QSn10-1 by electroerosive alloying (EEA) with soft antifriction materials such as silver, copper, Babbitt B83 and graphene oxide (GO). The mass transfer, surface roughness, coating thickness and dry friction coefficient of the running-in coatings were measured and analyzed by a precision electronic balance, three-dimensional optical profiler, metallographic microscope, scanning electron microscope (SEM), energy dispersive spectrometer (EDS) and friction and wear tester. An evaluation indicator system for the characterization was constructed based on six factors, including material price, time, mass transfer, roughness, thickness and friction coefficient of the coatings by electroerosive alloying. The Shannon entropy method was used to calculate the weight of different indices, and a comprehensive evaluation method for running-in coatings performance was proposed by combining the technique for order preference by similarity to ideal solution (TOPSIS) and a multicriteria decision-making technique. The TOPSIS model was employed for the comprehensive evaluation ranking of the characterization of the running-in coatings by electroerosive alloying.


Introduction
The wear and failure of rotating parts of key mechanical equipment usually start from the surface; therefore, to improve the reliability and stability of rotating parts of key mechanical equipment, improving the surface quality of components is necessary (Gençer et al 2019, Kobernik et al 2020. One of the most effective methods is to use functional coatings to improve the surface quality of parts, which can improve part reliability and stability (Mertgenc et al 2019, Cao et al 2020. Tin bronze plain Bush bearings have good thermal conductivity and mechanical properties, but the running friction is slightly higher, sometimes affecting the reliability and durability of plain bearings (Dinesh andMegalingam 2021, Prasad 2012). Therefore, it is necessary to build functional coatings for tin bronze plain bearings to improve reliability and durability.
Electroerosive alloying (EEA), also known as electro-spark deposition (ESD) or electro-spark alloying (ESA), the technology is the advanced method for repairing and strengthening the surface of metal materials. This method has the advantages of simple equipment, convenient operation and wide application range. The alloyed coating has higher wear resistance, good corrosion resistance, excellent friction performance, high fatigue strength, high temperature resistance and other special properties, so it has better practical value and wide application prospect. Widely used in molds, automotive, aerospace, nuclear industry, marine vessels, turbines, electrical transmission and other industries of mechanical parts of the surface strengthening and local material addition manufacturing. Silver, copper, Babbitt B83 and graphene oxide soft antifriction materials were alternately alloyed by electroerosive alloying to form tin bronze bearing surface functional coatings. The mass transfer, surface roughness, coating thickness and dry friction coefficient of the running-in coatings were measured and analyzed. The results of different comprehensive evaluation methods are quite different (Prasad et al 2020). Therefore, it is necessary to find an effective comprehensive evaluation method with strong objectivity, convenient calculation and small external disturbance to evaluate the performance of electroerosive alloying coatings.
Entropy-weighted is an objective and applicable method for the determination of weight value, was introduced into the comprehensive assessment. The information entropy can clearly reveal the utility of each indicator and avoid the interference of subjective factors, which ensures that it is more objective and credible than the subjective methods for comprehensive evaluation of multivariate index. The Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is a comprehensive evaluation based on distance and is widely used for multiple attribute decision making. The TOPSIS model can objectively and comprehensively reflect the characterization of the coatings by calculating the closeness degree between an evaluation value and its ideal solution.
Here, the Shannon entropy method and the TOPSIS model were applied to comprehensively evaluate the characterization of the coatings by electroerosive alloying. Shannon information entropy was introduced into the comprehensive evaluation of the properties of electroerosive alloying coatings as an objective and applicable method to determine the weight value. The TOPSIS model is more effective and accurate for calculating the comprehensive order of the objects in the comprehensive evaluation of the properties of the tin bronze sliding Bush bearing coatings of electroerosive alloying.

Materials and experimental methods
The experimental research methods consisted of adapting existing technology for producing coatings based on the electroerosive alloying method to the working conditions, design and technological features of Bush bearings.

The substrate material and the coating materials
The tin bronze sliding Bush bearing material QSn10-1 (Cu 89.10%, Sn 9.38%, P 0.72%, Others 0.80%) was cut into sizes of 25 mm by 29 mm by 4 mm as the substrate. Silver (Ag 99.99%), copper (Cu 99.99%) and Babbitt B83 (Sn 83.10%, Sb 11.02%, Cu 5.83%, others 0.05%) materials were used to make 3 mm diameter electroerosive alloying electrodes. Graphene oxide was a dispersible solution of 4 mg ml −1 with water as the solvent. The materials for experiments were provided by Zhejiang Shenfa Bearing Co., Ltd.
The surfaces of the tin bronze Bush substrate of the sliding bearing and the electrodes of electroerosive alloying were ground by metallographic grinding paper with different grain sizes (400, 600, 800 and 1,000 grit). The roughness of the ground surface is not more than 2 μm. The substrate and electrodes were cleaned for 20 min in absolute ethyl alcohol using an ultrasonic cleaning device before electroerosive alloying.
2.2. Electroerosive alloying process parameters DZ-4000III electroerosive alloying equipment can change the electrical parameters of the electroerosive discharge voltage (20 V-250 V), energy storage capacitor (30 μF-420 μF), and discharge frequency (1300 Hz -6000 Hz). The rotating speed of the electroerosive alloying equipment electrode was 2,600 r min −1 . The DZ-4000III electroerosive alloying surfacing machine was produced by the Institute of Surface Engineering Technology of China Academy of Agricultural Mechanization Science and Technology.
The moving speed of the electrode is approximately 2 mm s −1 during electroerosive alloying. The discharge frequency of the first layer was 3 kHz, the discharge frequency of the second layer was 4 kHz, and the discharge frequency of the third and fifth layers was 5 kHz. At room temperature, argon gas (Ar 99%) with a flow rate of 10 L min −1 was used as a protective gas to create the external environment atmosphere for the electroerosive alloying so that the alloying area was protected from being polluted by oxygen or nitrogen in the air.
The graphene oxide solution was applied to the surface of the specimens by a manual precoating method, and then the specimens were dried by ambient air, followed by the fifth layer using a B83 electrode. After the first electroerosive alloying of Babbitt B83, there will be many small pores on the surface, and the graphene oxide solution will enter Babbitt B83 along the pores. After drying, the second electroerosive alloying of Babbitt B83 will be carried out. The graphene oxide thus enters Babbitt B83 as an antifriction additive.
The technological parameters of electroerosive alloying coatings on the surface of the tin bronze sliding Bush bearing, discharge voltage, energy storage capacitor and working efficiency are shown in table 1.

Property investigation
The mass transfer data of the running-in coatings by electroerosive alloying were measured by a Mettler-Toledo AL204 precision electronic balance with a weighing accuracy of 0.1 mg.
The surface roughness and surface profile of the functional coatings were measured by Bruker Contour GT-k1 three-dimensional optical profilometers.
The electroerosive alloying running-in coatings specimen was cut from the cross section, and a small piece was embedded in the Bakelite for the analysis of microstructure and element distribution. After polishing and cleaning, to reveal elements of the granular structure, the surfaces of the samples were subjected to etching with 4% nitric acid alcohol with an exposure time of 10 s (Zhang et al 2022). The electroerosive alloy-treated crosssection morphology of the running-in coatings was analyzed using scanning electron microscopy (SEM) (FEI Quanta 200) and metallographic microscopy (LECIA DMi8 M). The distribution of chemical elements on the cross-section of the electroerosive alloying functional coatings was detected by using energy dispersive spectrometry (EDS).
An assessment of tribological properties was performed in the ball-on-plate reciprocating rig on the MWF-500 tribometer. The study investigated the tribological properties of electroerosive alloying layers under dry friction conditions. The test temperature was 25°C. The low sliding velocity of 20 mm s −1 was chosen to ensure boundary lubricating conditions and was maintained constant for all the stripes (Zhang et al 2022). The track length was 6 mm. An 8 mm diameter bearing steel (GCr15) ball was used as the counterface. The applied loads were 5 N, 10 N and 15 N. For the initial 600 seconds, the applied load was 5 N; the following 600 seconds, the load was 10 N; and then the load increased to 15 N for the final 600 seconds.

Evaluation methods
In this study, multiattribute decision-making methods of Shannon's entropy and TOPSIS have been used for the quantitative assessment of the characterization of the running-in coatings by electroerosive alloying.

Construction of the characterization indicator system of the coatings
Based on analyzing and sorting the characterization indicator system of relevant scholars and institutions, according to the connotation requirements of the coatings by electroerosive alloying, combined with data availability, according to the four scientific, systematic, practical and operational principles, six indicators are selected. The influencing factor indicator system of the characterization of the running-in coatings by electroerosive alloying is constructed in table 2. Table 2 shows that the factors affecting the characterization of the running-in coatings by electroerosive alloying include material price, electroerosive alloying time, mass transfer, roughness, thickness and friction coefficient, which constitute the six decision-making aspects of the indicator system. The selection basis of each aspect is described as follows.
(1) Electroerosive alloying material price Almost any conductive material can be used as an electrode for electroerosive alloying, but the price of different materials varies widely. The price of the deposited material has a significant influence on the cost of the deposited coating. Researchers need to find cheaper materials to achieve better performance.
(2) Electroerosive alloying time A single discharge formed a very small discharge micro area, the material melting amount was less, and the electroerosive alloying process was accompanied by the erosion of workpiece material, resulting in a long deposition time and low deposition efficiency. In addition, current electroerosive alloying equipment is mostly hand-held, and manual operation under low efficiency increases labor intensity. Therefore, the deposition time is a very important factor in electroerosive alloying.
(3) Mass transfer The electroerosive alloying running-in coatings are the result of gradual deposition and accumulation through multiple electro alloying discharges (Aghajani et al 2020). The amount of mass transfer on the substrate surface is an important index to evaluate electroerosive alloying. At the beginning of alloying, the weight of the running-in coating on the substrate surface increases most obviously. With increasing electroerosive alloying time, the mass transfer from the electrode to the substrate decreases gradually. After a period of time, with increasing alloying time, the weight of the electro alloyed running-in coating does not increase. This is because with increasing alloying time, the oxide content of the coating surface increases gradually, the surface thermal residual stress increases gradually, and the bonding force between the coatings decreases gradually. The surface material is more likely to splash during electroerosive alloying, which hinders mass transfer during the alloying process (Kiryukhantsev-Korneev et al 2020).
(4) Roughness Surface roughness has a major influence on the use of parts. Generally, a small surface roughness value will improve the fit quality, reduce wear, and prolong the service life of parts (Wang et al 2021). The rougher the surface is, the smaller the effective contact area between the mating surfaces, the greater the pressure, the greater the friction resistance, and the faster the wear . The running-in coating surface roughness of the tin bronze plain Bush bearing will directly affect the initial running-in effect and running-in time.
The surface roughness of electroerosive alloy running-in coating is affected not only by the alloying process parameters but also by the production, operation process, mechanical accuracy of the welding gun and properties of the alloying materials.

(5) Thickness
The thickness of the electroerosive alloying running-in coating is one of the most important indicators of electroerosive alloying. If the coating thickness is too small, the coating will wear off easily and cannot exhibit a special running-in coating performance (Vereschaka et al 2020). If the coating thickness is small, in strict working conditions (high speed and high specific pressure) running-in, the bearing face may exhibit scratches.
(6) Friction coefficient Electroerosive alloying technology can change the physical and chemical properties of the substrate surface, such as roughness, element composition, phase composition, hardness, elasticity and so on, thus affecting the tribological properties of the substrate surface .
The friction coefficient of the electroerosive alloying layer has an important influence on the practical application prospect of the workpiece, so a small friction coefficient should be selected. The material friction resistance should be relatively small; otherwise, it will consume a large amount of power.

Entropy method
Entropy is used very often in physics. In 1948, Shannon introduced the concept of entropy into information technology. Shannon, a researcher, first explained the concept of information entropy in academic circles, which is used to evaluate the degree of chaos or disorder in information systems (Mukhametzyanov 2021). In this paper, Shannon information entropy is applied to the comprehensive evaluation of a multi-index system as an objective weight calculation method. In the calculation and generalization of the Shannon information entropy method, the greater the entropy weight is, the greater the change degree of the corresponding indicator, the more information that can be obtained, and the more accurate the comprehensive evaluation . The information entropy value can accurately reflect the utility of different indicators and avoid the influence of subjective factors, so it is more objective than other subjective methods in the comprehensive evaluation system of various indicator systems (Minhas et al 2020).
The principle of using the entropy method to calculate the indicator weight is as follows.
(1) Construct the original decision matrix According to the experimental test data, the values of each index form an m × n matrix where m is the number of evaluation objects, namely, the number of electroerosive alloying running-in coatings. n is the number of indices, namely, the number of running-in coating performance evaluation indices.
The performance matrix can be presented as follows (Oluah et al 2020): (2) Data normalization There are different test schemes and different evaluation criteria in the electroerosive alloying running-in coating research program. Considering the difference in numerical units and orders of magnitude, the experimental test data cannot be directly used for comprehensive evaluation. It is necessary to normalize the test data obtained by detection or calculation. Based on the characteristics of the evaluation object, the evaluation index of test data is divided into two categories: the positive type index and the negative type index. The positive type index refers to the index that the larger the experimental test value is, the better the effect is. The negative type index is the index that the smaller the experimental index value, the better the effect.
When x ij is the positive type index, the normalized value S ij can be calculated by equation (2) (Mukhametzyanov 2021): When x ij is the negative type index, the normalized value S ij can be calculated by equation (3)  (1) Construct the weighted standardized matrix The weighted standardized matrix is established as equation (8)  (2) Calculate the positive and negative ideal solutions The positive ideal solution and the negative ideal solution are two basic concepts in the TOPSIS model method. If the positive ideal solution is assumed to be the optimal solution, all the attribute values reach the optimal values in all the test schemes. The negative ideal solution is considered to be the worst solution, and all its attribute values reach the worst value.
In this step, the positive ideal solution A + and the negative ideal solution A − were obtained (Yang et al 2022).  (4) Calculate the relative closeness To compare the evaluation value of the test data with the distance between the two ideal solutions, the relative closeness of object i is derived using Formula (13) (Rana et al 2021): The value of relative closeness reflects the relative superiority of the alternatives. A larger C i indicates that alternative i is relatively better, whereas a smaller C i indicates that this alternative is relatively poorer.

Results and analysis
4.1. The experimental results and analysis of the running-in coatings In this study, the analysis of the experimental results of the running-in coating was investigated, as shown in table 3.
The material prices of the running-in coating are shown as C 1 . With an increase in the number of layers, the kinds of coating materials increase, and the comprehensive price of coating materials increases. The lowest price of silver coating is 60 yuan, and the highest price of Ag + Cu + B83 + GO + B83 composite coating is 165 yuan. The coating deposition time is shown in table 3 as C 2 . The minimum deposition time is 5 min for silver coating, and the maximum deposition time is 50 min for Ag + Cu + B83 coating. The weight of the sample was measured by a precise electronic balance with an accuracy of 0.1 mg, and the mass transfer C 3 of the matrix after electroerosive alloying was calculated. As shown in table 3, the minimum value of mass transfer is 25.0 mg of Ag coating, and the maximum value of mass transfer is 394.2 mg of Ag + Cu + B83 coating.
A Bruker Contour GT-K1 three-dimensional optical profiler was used to observe the surface of the electroerosive alloy running-in coating, and the surface roughness of the coating was detected and analyzed, as shown in figure 1. The surface roughness of the running-in coating is shown as C 4 . As shown in table 3, the minimum value of the surface roughness is 15.46 μm for the Ag coating, and the maximum value of the surface roughness is 32.30 μm for the Ag + Cu + B83 coating.
The thicknesses of the running-in coatings are shown as C 5 in table 3. As indicated in figure 2, the thickness of the electroerosive alloying running-in coating increases gradually with the increase in the number of layers. The minimum thickness is 15 μm, and the maximum thickness is 160 μm.
The friction coefficient of the coatings is denoted by C 6 in table 3. The silver coating has a maximum coefficient of friction of 0.31. There is little difference in the coefficient of friction of the other three coatings as the surface coating is Babbitt B83, as illustrated in figure 3. The minimum friction coefficient of the Ag + B83 coating is 0.177.

The entropy weight of the running-in coating indicators
To facilitate comprehensive evaluation and comparative analysis of the characteristics of electroerosive alloying running-in coatings, normalized and standardized original data were processed, and the entropy method was adopted to calculate the entropy value, information utility value and weighting value corresponding to the six indicators of electroerosive alloying coatings, as shown in table 4.
The smaller the entropy value is, the larger the information utility value is and the larger the entropy weight is, indicating that the indicator is more important and contains more information, which should be given more attention.
According to the weight of indicators in table 4, in order of importance, first, all the indicators are mass transfer (C 3 ); second, thicknesses (C 5 ); third, time (C 2 ); fourth, material prices (C 1 ); fifth, friction coefficient (C 6 ); and finally, roughness (C 4 ). The weight of the indicator mass transfer (C 3 ) is as high as 20.69%, which cannot be underestimated in the performance evaluation of electroerosive alloying coating. This is mainly because the special properties of the coating can be realized only when the effective mass transfer of the deposited material from the electrode to the substrate is completed.
Among all the indicators, the average weight of six indicators is 16.67%. The three indicators higher than the average weight are mass transfer (C 3 ), thicknesses (C 5 ), and time (C 2 ). This shows that these three indicators are more important.
Although the weight values of material prices (C 1 ), friction coefficient (C 6 ) and roughness (C 4 ) are lower than the average weight of 16.67%, the material price directly affects the economic benefit, the friction coefficient directly reflects the friction reduction performance, and the surface roughness of the coatings directly affects the postprocessing process. Therefore, the effects of material prices (C 1 ), friction coefficient (C 6 ) and roughness (C 4 ) should still be considered.

The comprehensive evaluation ranking of the running-in coatings
To compare the characteristics of all the electroerosive alloying running-in coatings, the Euclidean distance, relative closeness and comprehensive evaluation ranking of the electroerosive alloying coatings were obtained using experimental data, as shown in table 5.
Based on the calculated Euclidean distance, the relative closeness of different evaluation objects is calculated. The schemes are ranked according to the relative closeness. The larger the relative closeness is, the better the scheme is. In order of relative closeness, they are Ag + Cu + B83 + GO + B83 coating, Ag + Cu + B83 coating, Ag + B83 coating and Ag coating. Table 5 indicates that with the increase in coating material species, the comprehensive evaluation of the coating improved.
The thermal conductivity of tin bronze Bush bearings is better; as the basic material, it can efficiently conduct the energy generated by friction, which is conducive to reducing the risk of Bush bearing failure caused by local overheating (Yuan et al 2020, Tarel'nik et al 2017. The soft metal material silver is often used in the design of high load, medium and low speed sliding Bush bearings and has excellent mechanical engineering performance, chemical corrosion resistance and tribological performance. The good wettability between silver and copper on soft metal materials is beneficial to improving the metallurgical bonding performance of metals in the process of electroerosive alloying. However, the detection and analysis of its operation show that the friction coefficient of silver as an anti-friction metal coating is high, which makes shafts wear easily. Its tribological characteristics become an obstacle to its further application. The ε-phase (Cu 6 Sn 5 ) is formed between the copper in running-in coating and the tin in Babbitt alloy, which is beneficial to strengthening the metal bonding of running-in coating and refining the grain . Therefore, based on the comprehensive analysis of the above factors, the soft metal materials silver and copper are very suitable as intermediate transition coatings. Tin-based Babbitt B83, a classic bearing material for plain bearings, has good embeddability and compliance and plays a role in reducing friction and preventing bite injury in industrial applications (Ni et al 2019). It is especially beneficial to running-in at the initial stage of operation and is very suitable for use as surface coating. Graphene has been used to improve tribological properties due to its extraordinary properties (Mudra et al 2021, Uysal et al 2016. Graphene oxide is considered a promising material for reducing friction and wear, owing to its structural features (Wu et al 2017, Song et al 2020. Therefore, adding graphene oxide to the running-in coating is beneficial for improving the friction and wear properties of the surface.
Facilitation of running-in conditions and improvement of the friction surface operating regime in the post running-in period are provided by improving the Bush bearing by applying the running-in coating by electroerosive alloying. The running-in coating deformation under the action of high specific loads provides parts with automatic adjustment and compensation for manufacturing errors.

Industrial validation of evaluation method
The running-in coatings of the tin bronze bearing bush that were formed by alternately electro-spark deposition applying the antifriction material of silver, copper, Babbitt B83 and graphene oxide. Operate the control panel of the electro-spark deposition equipment, and adjust the electrical parameters such as discharge voltage, energy storage capacitor and discharge frequency according to the parameters in table 1.
Clean it up during installation, and there should be no Sundries, so as not to affect the surface of bearing bush or lubrication system. The installation standard shall be strictly implemented in the installation process of the bearing bush, so that the installation measurement dimensions shall meet the specification requirements.   Special attention shall be paid to the assembly clearance between the journal and the bearing bush shall meet the requirements.
After filling the lubricating oil into the machined bearing Bush, the commissioning test with light load of 10 N and rotating speed of 700 r min −1 is carried out for 3 h in total, and the external temperature of the bearing Bush is detected at any time. After the light load commissioning test, carry out the commissioning test for 8 h with the load of 30 N and the rotating speed of 700 r min −1 , and check the external temperature of the bearing Bush at any time.
According to the analysis of the test data, the tin bronze bearing bush with the Ag + Cu + B83 + GO + B83 coating runs stably, and the external temperature of the bearing bush only increases by 8°C. The tin bronze bearing bush with other running-in coatings runs stably, and the external temperature of the bearing bush increases by 10°C-12°C. After the commissioning test, there is no burning phenomenon on the surface of the bearing Bush with the Ag + Cu + B83 + GO + B83 coating, the scratch is not obvious, and the wear is slight. It was found that the surface of the bearing Bush with other running-in coatings was burnt and blackened, and the surface was seriously worn. As can be seen in figure 4, the bearing shells with the Ag + Cu + B83 + GO + B83 coating showed no burning on the journal part, while the bearing shells with other running-in coatings showed burning on the journal part. From the overall situation, that temperature rise of the slide bearing Bush with the Ag + Cu + B83 + GO + B83 coatings is small, the operation is stable and reliable, and the durability is good.
The operating characteristics of the Ag + Cu + B83 + GO + B83 running-in coatings are such that it can withstand short term damage under lubricating oil dynamic conditions without seizure and at relatively high temperatures, and it has very good fatigue properties under rated load conditions. The industrial application of scientific results verifies the availability and effectiveness of the evaluation method.

Conclusions
On the basis of the obtained results and their interpretation, the following conclusions can be drawn.
(1) A characterization indicator system for the running-in coatings was constructed by electroerosive alloying.
(2) The entropy method was adopted to calculate the entropy weight corresponding to the six electroerosive alloying coating indicators. The weight value of the mass transfer indicator is 20.69%, which is the highest among the indicators of all the running-in coatings. (3) The entropy method and the TOPSIS model were employed for the comprehensive evaluation ranking of the characterization of the running-in coatings. The Ag + Cu + B83 + GO + B83 coating has the highest comprehensive score and the best comprehensive characteristics.
(4) In the industrial application, the availability and effectiveness of the evaluation method are verified.