3.1. Microstructure evolution
Figures 1–4 are showing microstructures of heat-treated plain carbon steels i.e. AISI 1020, 1030, 1040, 1045, and 1050 having carbon content in the range of 0.19–0.54 wt.%. Heat-treatment processes; annealing, normalizing, quenching, and austempering of plain carbon steels resulted in the formation of coarse ferrite-pearlite, fine ferrite-pearlite, martensite, and bainite phases in the microstructures respectively. Microstructures of all the annealed plain carbon steels (Fig. 1) comprised of coarse lamellar pearlite phase in the matrix of ferrite. The annealing process involves slow furnace cooling which provides sufficient time for recrystallization and grain growth, resulting in the coarse-grained microstructure. Due to slow and extended cooling within the furnace, the growth of ferrite is increased within the cementite plates resulting in the formation of coarse pearlite. The lower carbon content (0.19 wt.%) of AISI 1020 steel, resulted in a smaller volume fraction of pearlite in the matrix of ferrite (Fig. 1a). Since the carbon content (0.34 wt.%) of AISI 1030 steel is greater than AISI 1020 steel, it has a large volume fraction of pearlite (Fig. 1b). Similarly, AISI 1040, 1045, and 1050 steels are comprised of increasing pearlite volume fractions in the microstructures (Figs. 1 (c-e)) due to an increase in carbon contents ranging from 0.37–0.54 wt.%.
Relatively fine pearlite in the fine ferritic matrix is observed in the microstructures of normalized plain carbon steels (Fig. 2).
After quenching, packets and blocks of lath martensite (Fig. 3) are visible in all plain carbon steels. It is observed that with an increase in carbon content from 0.37–0.54 wt.%, the martensite laths become finer as reported previously [27].
Austempering heat-treatment resulted in the formation of lower bainite (Fig. 4). Due to the higher carbide formation, volume fractions of bainitic ferrite (BF) decreased as an increase in carbon content. The microstructure (Fig. 4b) shows bainitic ferrite, which has Widmanstätten side-plate morphology.
3.2. Micro Vickers hardness
Figure 5 (a) and (b) shows the micro-Vickers hardness results of plain carbon steels obtained after various heat treatments i.e. annealing, normalizing, quenching, and austempering. Microstructures of annealed plain carbon steels comprised of coarse pearlite phase in the matrix of ferrite. The presence of ferrite in the microstructure of annealed samples caused low hardness values. But with an increase in carbon contents from 0.19 to 0.54 wt.% the amount of cementite phase increased, resulting in a gradual increase in hardness.
A gradual increase in hardness values was observed with an increase in carbon contents. Bainite formed by austempering has been reported to be a comparatively hard phase than ferrite and pearlite. Therefore, the hardness of austempered plain carbon steel was higher than normalized and annealed samples. Supersaturated lath martensite resulted from quenching and possessed the highest hardness among all.
3.3. Electrochemical properties
Figure 6 is showing Tafel polarization scans of heat-treated plain carbon steels in 3.5% NaCl solution. All the samples are polarized in the ± 50 mV potential range with respect to their open circuit potential.
The kinetic parameters like anodic (βa) and cathodic (βc) slopes, corrosion current densities (icorr), corrosion potentials (Ecorr), and corrosion rates were calculated by Tafel fit, with the help of Echem Analyst software (version 5.62).
Table 3 is showing the calculated values of polarization curves. Localized galvanic corrosion cells were formed in seawater due to various concentrations and morphologies of ferrite, pearlite, and cementite. The relatively active potential of the ferrite phase compared with the pearlite phase may have promoted its preferential dissolution. The corrosion of plain carbon steel starts with the oxidation of the ferrite phase due to the occurrence of reaction mentioned in equation 2 [32].
(2)
On the cathodic site, the reaction (Eq. 3) is
(3)
So, the overall reaction (Eq. 4) will be
(4)
Hence the ferrite phase acts as anode and cementite as cathode which will further enhance corrosion of plain carbon steels [33]. As shown in Fig. 7 (a) with an increase in carbon concentration from 0.19 to 0.54 wt. % in the plain carbon steel (annealed condition), the corrosion rate also increased from 2.378 to 9.666 mpy due to an increase in the pearlite phase providing more sites for active cell formation.
Table 3
Kinetic parameters calculated from Tafel scan of all heat-treated plain carbon steels
AISI Steel Grades
|
βa
(mV-decade−1)
|
βc
(mV-decade−1)
|
icorr
(µA-cm−2)
|
Ecorr
(mV)
|
Corrosion Rate
(mpy)
|
Annealing
|
1020
|
36.30
|
42.60
|
5.200
|
− 594.0
|
2.378
|
1030
|
33.20
|
51.90
|
6.230
|
− 566.0
|
2.845
|
1040
|
85.10
|
142.0
|
12.00
|
− 648.0
|
5.500
|
1045
|
84.20
|
113.1
|
16.30
|
− 545.0
|
7.352
|
1050
|
99.70
|
1331
|
21.20
|
− 585.0
|
9.666
|
Normalizing
|
1020
|
34.80
|
36.60
|
3.440
|
− 660.0
|
1.571
|
1030
|
41.60
|
73.60
|
6.330
|
− 579.0
|
2.892
|
1040
|
58.50
|
109.9
|
8.960
|
− 625.0
|
4.095
|
1045
|
86.40
|
138.5
|
13.20
|
− 546.0
|
5.940
|
1050
|
78.30
|
162.3
|
13.90
|
− 583.0
|
6.362
|
Quenching
|
1020
|
42.00
|
77.00
|
5.640
|
− 683.0
|
2.576
|
1030
|
67.40
|
147.0
|
8.290
|
− 640.0
|
3.786
|
1040
|
67.70
|
173.1
|
12.50
|
− 638.0
|
5.725
|
1045
|
79.20
|
154.5
|
14.90
|
− 572.0
|
6.719
|
1050
|
66.80
|
148.4
|
17.30
|
− 586.0
|
7.889
|
Austempering
|
1020
|
21.80
|
25.10
|
5.870
|
− 627.0
|
2.681
|
1030
|
69.10
|
136.6
|
13.70
|
− 600.0
|
6.275
|
1040
|
71.90
|
128.2
|
8.200
|
− 626.0
|
7.747
|
1045
|
81.60
|
549.1
|
18.60
|
− 610.0
|
8.381
|
1050
|
80.00
|
296.5
|
26.30
|
− 582.0
|
12.12
|
Decreased corrosion rate after normalizing might be due to the fine grain-sized microstructure compared to annealed samples (Fig. 7 (b)). After normalizing, the highest corrosion rate (6.362 mpy) was exhibited by AISI 1050 steel (0.54 wt. % C) in 3.5% NaCl solution. It has been reported that in 0.5 M NaCl solution, quenched samples exhibited better corrosion resistance than the annealed samples due to less localized galvanic cells [34] formation.
In quenched microstructure carbon is entrapped in BCT crystal structure, resulting in the uniform distribution of carbon in the matrix. Hence, the martensite phase behaves as noble and acts as the cathodic phase while the Widmanstatten ferrite at grain boundaries acts as an anode. Increasing the carbon concentrations from 0.19 to 0.54 wt %, the corrosion rate also increased from 2.576 to 7.889 mpy due to the formation of high carbon martensite. The martensite has more corrosion resistance in comparison to ferrite-pearlite phases. This infers that uniform distribution of carbon in the matrix enhances corrosion resistance.
In the case of austempering, bainitic microstructure showed that increment in the carbon concentration resulted in enhancement of corrosion rate. The corrosion rate is increased from 2.681 to 12.12 mpy with the increase of carbon concentration [25] from 0.19 to 0.54 wt %. Among all microstructures, the highest corrosion rate is observed in the bainitic microstructure due to more active corrosion cells.
3.4. Corrosion morphology
SEM micrographs and EDS spectra of heat-treated AISI 1030 and 1050 steel obtained after corrosion testing in 3.5% NaCl solution are shown in Figs. 8 and 9. The SEM-EDS elemental maps of the corroded surface was used to determine the elemental distribution of the corrosion product (Figs. 10 and 11).
The ball shape morphology (Fig. 8) of corrosion products was observed on the surfaces of annealed, quenched, and austempered samples of AISI 1030 steel after corrosion testing which might be goethite, as reported by C. Yong et al.[30]. While the normalized steel sample revealed flake-like morphology (Fig. 8c) of corrosion product.
The relatively large ball shape morphology (Fig. 8g) of corrosion products was visible on the austempered AISI 1030 sample resulting in a large corrosion rate as revealed in Table 3.
In the case of AISI 1050 steel, again ball shape morphology (Fig. 9) of corrosion products is observed on surfaces of annealed and normalized samples. While on the surfaces of quenched and austempered samples more uniform porous corrosion deposits (Figs. 9e & 9g) are observed. The elemental maps (Fig. 10) of annealed, normalized, and quenched AISI 1030 steel samples exhibited the localized corrosion attack, whereas, in austempered AISI 1030 steel, uniform corrosion occurred.
When iron ions react with chlorine or oxygen or any other such anions, they tend to form metallic oxides, hydro-oxides, or even chlorides and then deposit as particles. These oxides or hydro-oxides accumulate and may deposit in the areas called crevices and then further trigger corrosion. High percentages of oxygen in the elemental maps indicates high corrosion rates similar to reported works earlier [35–39].
In the case of AISI 1050, annealed and normalized samples (Figs. 11a & 11b) underwent localized corrosion attack resulting in the ball-like morphology of corrosion products. In Figs. 10 and 11, elemental maps show the occurrence of Fe, O, and C in almost every map and these three elements are the dominant part of corrosion scales [40–46].
Magnetite (Fe3O4) [47] along with goethite (α-FeOOH), and lepidocrocite (γ-FeOOH) [48] were observed in the corrosion deposits mainly however small fractions of hematite (Fe2O3) [49, 50] calcite (CaCO3) and green rusts (hydrated ferrous-ferric compounds containing CO32-, Cl- or SO4-2) were also observed [51–54].
Figure 12 shows higher magnification morphologies of corrosion deposits after corrosion testing of AISI 1050 steel of annealed and austempered microstructures.
The porous/spongy corrosion deposits observed in austempered AISI 1050 steel (Fig. 12b ) seemed to be the main cause of the high corrosion rate. Corrosion causes porous morphology which increases the ingress of electrolyte and results in the accelerated dissolution process. However, it seems that the solid morphology of the corrosion deposit leads to the lowest corrosion rate of the normalized AISI 1050 steel (Fig. 12a). Since the solid corrosion product with hairline cracks has a very low ability to retain the electrolyte, results in the lower corrosion rate.