3.1Microstructure analysis
In order to understand the structural composition and phase composition of the alloy, the microstructure of high Sn-Pb bronze alloy is analyzed, and its results are shown in Fig. 1. There are different phases from the optical microstructure of high Sn-Pb bronze alloy in Fig. 1a, and its microstructure presents orange-yellow, light blue and large irregular black areas [7, 21–23]. Figure 1c shows the enlarged metallographic structure of area 1 in Fig. 1a. According to the corresponding EPMA surface scanning and X-ray diffraction results (Fig. 1b), it can be seen that the orange yellow region contains a higher amount of Cu element as a pre-eutectoid α (I) Phase (Cu13.7Sn). The light blue area contains higher Sn elements, and the orange-yellow network area contains higher Cu element. The combination of two is called the (δ + α(II)) eutectoid structure, where the light blue area is the δ phase (Cu41Sn11), and the orange-red network is the α(II) phase. The black particles contain more Pb and are randomly distributed. The enlarged metallographic structure of area 2 in Fig. 1a is shown Fig. 1d. The corresponding EPMA surface scanning results show that there is a relatively obvious Pb rich phase on the right side of the photo in Fig. 1e, and the boundary between it and the matrix is mainly α (I) Phase. Figure 1e shows the enlarged metallographic structure of area 3 in Fig. 1a, which exists in the internal region of the rich Pb phase. The reason for the existence of large Pb rich phase in the microstructure is that more Pb is added in the smelting process to increase the fluidity of the alloy. According to the corresponding EPMA surface scanning results, it can be seen that there are Sn and less Cu elements in this position [7, 21, 22].
There were various phases in the structure of high Sn-Pb bronze alloy, which was due to the formation of precipitated phase, eutectoid structure and Pb in the solidification process, so that the alloy phase composition and structure are different.
3.2 Material microstructure potential analysis
SKPFM was used to characterize the micro-area potential of the surface of the structure precipitated during the solidification of the high Sn-Pb bronze alloy. Figure 2 shows the optical microstructure of the high Sn-Pb bronze alloy and its corresponding SKPFM analysis results. Comparing the potential diagram (Fig. 2c and 2d) with the optical microstructure (Fig. 2a and 2b), the region with higher potential in Fig. 2c is (δ + α(II)) eutectoid structure, around which the area with lower potential is α(I), and the bright-colored structure is Pb particles. The potential distribution diagram of the Pb-rich phase and Cu-rich structure in the sample are shown in Fig. 2d. By drawing a straight line with an arrow in Fig. 2c and 2d, Line 1 passed through α(I), (δ + α(II)) eutectoid and a Pb particle. And Line 2 passed through the rich Pb phase, the interface between the rich-Pb phase and the rich-Cu structure, and the rich Cu structure in sequence. The volta potential results of different structures on the line shown in Fig. 2e and 2f showed that the potentials of the α(I) phase and the (δ + α(II)) eutectoid in the alloy matrix were about − 29 ~ -17 mV and − 2 ~ 12 mV, respectively, and the Pb particles were oxidized during the sample preparation process, resulting in a potential increase of up to 47 mV. Although the composition of the α (II) phase and α (I) phase was the same, the magnitude of α (II) phase potential in the eutectoid structure is not clear, because α (II) phase was small and abundant, which makes the potential of the eutectoid structure fluctuate.The potential of the Pb-rich phase increased gradually from the inside to the periphery, while the volta potential at the junction of the two began to drop sharply, and is relatively stable in the Cu-rich structure. Therefore, the potential of the α(I) phase in the high Sn-Pb bronze alloy was significantly lower than that of the (δ + α(II)) eutectoid, and the potential of the Pb-rich phase in the alloy was higher than that of the Cu-rich structure.
The real physical meaning of the surface potential E is actually the relative volt potential difference between the surface of the metal sample and the tip of the test probe by SKPFM. The relationship between it and the free energy of the system is [24, 25]:
$$\varDelta G=-nFE=-nF{(\psi }_{surfance}-{\psi }_{tip})$$
3
where ΔG is the Gibbs free energy of the system, reflecting the properties of the thermodynamic sample, n is the number of electrons transferred, F is Faraday's constant, which is 9.6485 × 104 C/mol, ψsurface is the voltage potential of the sample surface, ψtip is the voltage potential of the SKP probe.
According to Eq. (3), the Gibbs free energy is negatively correlated with the surface potential. This indicates that the higher the surface potential of the sample, the higher its thermodynamic stability, which also reflects a lower corrosion tendency [25–28].
3.3 Weight loss analysis
To study the corrosion weight loss of high Sn-Pb bronze alloys in high humidity and high chlorine environments, the weight loss data were fitted by Eq. (4) [28]. Figure 3 shows the corrosion weight loss curve and the corresponding corrosion rate, and Table 2 shows the fitting results of the weight loss curve.
where D is the weight loss per unit area, g●m− 2, A is the initial corrosion loss weight, g●m− 2, T is the number of corrosion days, day, n is a fitting constant which is related to the protection performance of the rust layer.
By analyzing the fitting results of corrosion kinetics, the two fitting results with R2 > 0.99 showed that the correlation of the power function fitting of the corresponding curve was better, and the results had a higher reliability in predicting the development of corrosion. A is the corrosion weight loss of fitting 1 day, and the value of n reflects the protection of the corrosion product layer. Table 2 showed that the n value of the high Sn-Pb bronze alloy in the environment was 1.0421, indicating that the corrosion products on the surface of the sample did not have a good protective effect on the alloy. By comparing the corrosion rate of different days, the overall trend was to gradually increase and remain relatively stable. The corrosion rates of the samples after tested for 1 day and 30 days were 1.5517 (g/m2 • d) and 1.9425 (g/m2 • d), respectively, indicating that the overall corrosion rate of the samples did not differ significantly. Therefore, the surface corrosion of the sample was relatively serious in the early stage of corrosion, and it had been corroded at a relatively fast speed, indicating that the corrosion product layer on the surface of the sample had no significant protective effect on the substrate.
Table 2
Corrosion kinetics fitting results of high Sn-Pb bronze alloys in salt spray environment
Material | A | n | R2 |
Bronze alloy | 1.6837 | 1.0421 | 0.9997 |
3.4 Corrosion product morphology analysis
3.4.1 Morphology of surface corrosion products
Figure 4 shows the optical corrosion morphology of high Sn-Pb bronze alloy after neutral salt spray test for 0, 5, 10, 20 and 30 days. The surface of the sample before the test in Fig. 4a had a good metallic luster, and the large gray area on the surface of the sample was Pb added during casting. After 5 days of the test (Fig. 5b), the gray-white Pb on the surface of the sample was preferentially corroded and white corrosion products appeared, while reddish-brown corrosion products appeared on the surface of other areas. After 10 days (Fig. 5c), the surface of the sample was also preferentially corroded by Pb, and blue-green corrosion products appeared around the corrosion products. After 20 (Fig. 5d) and 30 days (Fig. 5e), the corrosion was further aggravated, and the blue-green corrosion products on the surface of the sample gradually diffused to the periphery and increased, and there were white corrosion products around it. Under the influence of the fog environment, the corrosion products spread outward in the shape of a river. The reddish-brown corrosion products on the surface of the bronze alloy substrate gradually deepened, and there were less spot-like light green corrosion products on the reddish-brown surface. Raman spectroscopic analysis was performed on the sample tested for 30 days, and the results in Fig. 4f-h showed that the reddish-brown corrosion product (point 1) on the surface of the sample was Cu2O [29–34], and the blue-green corrosion product (Point 2) was the mixed corrosion product of Cu2(OH)3Cl [31–33] and PbCO3 [30, 34], and the white corrosion product (point 3) was PbCO3.
To further observe the corrosion behavior of high Sn-Pb bronze alloy in a high-chlorine and high-humidity environment from a microscopic perspective, the microscopic morphology observation and EDS point scanning of the surface corrosion products after 5 and 30 days were conducted. Figure 5a is a BSE-SEM photo of the corrosion on the alloy surface after 5 days. It could be seen that the Pb-rich phase corrosion on the sample surface was more serious, and the corrosion in the Cu-rich microstructure area was relatively smooth and the processing traces were more obvious. Figures 5b-d correspond to the microscopic corrosion morphology from Pb-rich phase to Cu-rich structure (zones 1–3) in Fig. 6a, where zone 1 was located inside Pb-rich phase and the corrosion product morphology was a relatively compact layered structure. Zone 2 was located at the junction of the Pb-rich and Cu-rich zone, and its corrosion products had a rod-shaped structure with varying lengths. Zone 3 was in the Cu-rich structure of the sample, and there were polygonal particles of different sizes on the surface. Combined with the EDS point scanning analysis results in Table 3, it can be seen that the layered (point 1) and rod-shaped surfaces of the sample mainly contain Pb (content is about 91 wt.%), O (content is about 7 wt.%) ) and Cl (content is about 2 wt.%), the corrosion product formed is Pb(OH)Cl [35]. Zone 3 was in the Cu-rich area, and the polygonal granular corrosion products (point 3) generated on the surface also contain less Cu. Point 4 was on the slightly corroded sample surface, and its Cu and Sn were different from those of the Cu-rich phase. The contents of Cu and Sn were basically the same, but due to being in a neutral salt spray environment, the surface of the sample was easily oxidized.
Table 3
Point scan results of high Sn-Pb bronze alloy after 5 days in neutral salt spray environment
| Cu | Sn | Pb | O | Cl |
1 | — | — | 91.51 | 1.68 | 6.80 |
2 | — | — | 90.72 | 1.94 | 7.35 |
3 | 7.00 | — | 79.99 | 13.01 | — |
4 | 69.29 | 16.20 | 4.29 | 9.75 | 0.47 |
Figure 6a shows the BSE-SEM corrosion morphology photo of the high Sn-Pb bronze alloy after 30 days. It could be seen that more corrosion products appeared on the surface of the sample, and the morphology of the corrosion products in each area was different. Figures 6b-e correspond to the microscopic corrosion morphology from the Pb-rich phase to the Cu-rich structure (zones 1–4) in Fig. 6a. Compared with the corrosion product morphology after 5 days, the corrosion product morphology of the Pb-rich b phase after 30 days was divided into two layers, which was the upper layer being polyhedral and the lower layer being clustered. At the boundary between the Pb-rich phase and the Cu-rich structure (zones 2 and 3), the corrosion products were in the form of relatively fine rods and polygonal layers, and were roughly divided into upper and lower layers. Zone 4 was a Cu-rich phase area, and there were a large number of spherical corrosion products that were not found after 5 days. Combined with the EDS point scanning results in Table 4, it could be seen that the corrosion product elements of the high Sn-Pb bronze alloy after testing in a neutral salt spray environment for 30 days had changed significantly compared with those after 5 days. All corrosion products contain Cu. Moreover, the corrosion product of Pb in the Pb-rich phase contained less Pb (point 1) than at the junction between the two (points 3 and 5). The Cu element content increased, indicating that the Cu inside the Pb-rich phase begins to corrode and formed new corrosion products. Spherical corrosion products (point 9) appeared in the Cu-rich phase and contain higher Sn (containing Sn was 55.01 wt.%) and O (containing O was 26.93 wt.%). By comparing the corrosion products of different corrosion cycles, the initial Pb corrosion products gradually transformed into products containing Cu and Pb. The Cl element content also gradually increased, and Sn oxidation products appeared on the surface.
Table 4
Point scan results of high Sn-Pb bronze alloy after 30 days in neutral salt spray environment
| Cu | Sn | Pb | O | Cl |
1 | 28.42 | — | 57.35 | 3.77 | 10.47 |
2 | 80.41 | — | 4.45 | 2.01 | 13.13 |
3 | 1.70 | — | 87.84 | 10.46 | — |
4 | 58.82 | — | 13.15 | 10.41 | 17.62 |
5 | 6.24 | — | 82.92 | 10.79 | — |
6 | 85.90 | — | 1.12 | 10.89 | 0.41 |
7 | 75.74 | 1.69 | 1.83 | 7.88 | 0.22 |
8 | 7.90 | — | 82.38 | 9.72 | — |
9 | 17.82 | 55.01 | — | 26.93 | 0.24 |
3.4.2 Alloy corrosion sectional analysis
Figure 7 shows the corrosion morphology characteristics and surface scanning distribution diagram of the section of the high Sn-Pb bronze alloy placed in the neutral salt spray chamber for 30 days. The local corrosion micromorphology of the Pb-rich phase after 30 days was shown in Fig. 7a-b, there were a large number of needle-like and rod-like corrosion products in the cross-section of the sample, and they were randomly distributed. According to the surface scanning results (Fig. 7c), the Pb-rich region and its corrosion product layer contain a large amount of O and Cl elements, indicating that the Pb-rich phase had obvious corrosion, and the surrounding Cu-rich structure was not affected by corrosion. Figure 7d-e shows the partial corrosion micromorphology of the Cu-rich structure after 30 days. It can be seen from Fig. 7d that the corrosion products on the cross-section of the sample are irregular needle-shaped and unevenly distributed. Figure 7e shows that the development of corrosion is mainly along the α phase, while the δ phase is hardly corroded. The surface scanning results in Fig. 7f show that the corrosion products on the surface of the sample are mainly Pb, O and Cl elements, while the Cu content in the corroded area on the substrate side decreases significantly, the Sn and Pb contents increase significantly, and the O and Cl. It is mainly distributed in the corroded α phase and coincides with the distribution of Sn.
3.5 Polarization curve analysis
Figure 8 shows the results of polarization curves of bronze alloys in a neutral salt spray environment for different cycles. It could be seen that the polarization curve of the original sample (0 d) is higher than that of other test days. From the fitting results in Table 5, the corrosion current density icorr of other test days was higher than that of the original sample (0 d). And the corrosion current density gradually increases, and the corrosion rate gradually accelerated, which was roughly consistent with the law of corrosion weight loss. Combining the metal systems (Cu, Sn and Pb) contained in the high Sn-Pb bronze alloy, the corrosion potential Ecorr represented the mixed corrosion potential presented by the joint corrosion of the Pb-rich phase and the Cu-rich structure. Secondly, the cathode segment of the polarization curve was mainly oxygen absorption reaction from Fig. 8 [36–38].
Compared with the original sample (0 d) and other test days, the Ecorr in the late corrosion period was lower than that of the original sample (0 d), indicating that the thermodynamic stability of the sample surface was reduced, which might be due to the corrosion equilibrium system of surface oxidation products generated by Pb and Cu was broken, resulting in a decrease in thermodynamic stability, and the hydrolyzed \({CO}_{3}^{2-}\) reacted to generate PbCO3.
In the standard state, the reaction of metallic Pb to PbCO3 [39]:
\(Pb+{CO}_{3}^{2-}=Pb{CO}_{3}+2{e}^{-}\) (E= -0.506 V) (5)
When Cu is transformed into Cu(I), the reaction is [25, 39]
\({2Cu+OH}^{-}={Cu}_{2}O+{H}_{2}O+{2e}^{-}\) (E= -0.361 V) (6)
When Sn is transformed into Sn(II), the reaction is [39]
\(Sn={Sn}^{2+}+2{e}^{-}\) (E= -0.140 V) (7)
As the test time increases, the corrosion product layer cannot protect the substrate, and a high corrosion rate is always maintained. It can be seen from the anode section that the current density of the anode section decreases first and then rapidly increases at about − 0.27~-0.18 V. According to the standard electrode potential of the electrode reaction [36], it may be that the metal Cu and Sn in the sample changed from a low-valence state to a high-valence state substance. Since the electrochemical reaction potential of Pb(II) to Pb(IV) is about 1.69 V [37], the corrosion potential of the reaction is too high, so the conversion of Pb(II) to Pb(IV) is not considered.
When Sn(II) is transformed into Sn(IV), the reaction is [39]
\({Sn}^{2+}={Sn}^{4+}+{2e}^{-}\) (E = 0.15 V) (8)
When Cu(I) is transformed into Cu(II), the reaction is [39]
\({Cu}^{+}={Cu}^{2+}+{e}^{-}\) (E = 0.167 V) (9)
Table 5
Fitting results of polarization curves of high Sn-Pb bronze alloy tested for different times
Time/d | Ecorr/V | Icorr/µA |
0 | -0.5849 | 6.86 |
5 | -0.7301 | 10.205 |
10 | -0.7343 | 11.504 |
20 | -0.7492 | 12.058 |
30 | -0.7619 | 19.972 |