Introduction

Fishing forms the major livelihood activity to the sizeable population living in the coastal states of India. Commercial fishing activities are mainly carried out using the traditional wooden fishing canoes and mechanized boats. Recent years, steel is extensively used for construction of fishing boats due to the scarcity of the conventional boat-building timbers. Boats constructed with steel are having low maintenance cost and safer to operate longer fishing days in the sea. The recommended steel for the construction of fishing boat is BIS 2062-grade B steel. The major concerns regarding the steel fishing boats are the corrosion, which occurs in the hull and the welding joints. Micro-sized organic pigments are used for protecting marine structures from corrosion and are highly susceptible to coating defects, poor adhesion, low-impact resistance, optical transparency, abrasion and scratch resistance, lower coating flexibility, etc. The major disadvantage of polymeric coating is the pinholes and pores, through which the corrosive agents will penetrate leading to coating failures (Husain et al. 2013). To override the above problems, nano-sized materials have recently been used in paints and these paints have exhibited excellent efficiency improvements compared to the conventional coating formulations (Fernando 2004). Nano-oxides of Ti, Zn, Fe, Ca, Al, etc. are employed in the coatings recent years (Dhoke and Khanna 2009).

The conventional approaches towards the coating of steel materials with ceramic, polymeric, electrodeposition, etc., are effective to some extent in aggressive environments. Nano-materials provide an efficient corrosion resistance in self-assembled nano-phase particle, due to its high surface area and adhesion to the substrate (Voevodin et al. 2003; Bjerklie 2005). Three layer TiO2–Al2O3 nano-composite coatings (Vaghari et al. 2011) and TiN-based coatings (Ilevbare and Burstein 2001) exhibited excellent corrosion resistance on SS316L. Not much work has been done to introduce a mixture consisting of multifunctional nano-sized inorganic metal oxides as coating material instead of one or two metal-oxide composites over carbon steel.

Rare-earth oxide, such as cerium oxide, has been widely used in solid-oxide fuel cells and as thermal and anti-corrosion barrier during high-temperature catalytic applications. Nano-sized cerium oxide incorporated in nickel, aluminium, and magnesium matrices improved the corrosion resistance and wear resistance (Arurault et al. 2004; Low et al. 2006; Chandrasekar and Pushpavanam 2008; Xue et al. 2010). A recent report showed the development of super-hydrophobic surface of magnesium alloys and aluminium alloys using nano-cerium-oxide thin films (Ishizaki et al. 2011; Liang et al. 2013). The commonly employed method to prevent corrosion in mild steel is to provide a permanent impermeable coating. Nano-cerium oxide is used as inhibitor to reduce the corrosion in H2SO4 medium (Sharmila et al. 2013). Hydroxides and oxides of cerium and titanium composite coatings on 2024 aluminium alloy improved the adhesive strength, micro-hardness, and corrosion resistance. Higher concentration of cerium oxide has enhanced the barrier resistance and hence corrosion resistance (Han et al. 2012). A mixture of cerium-oxide and titanium-oxide-reinforced aluminium showed fouling and corrosion resistances (Ashraf and Shibli 2008).

Increased research interest was shown on TiO2 due to its unique physicochemical properties, excellent chemical stability, good photo-catalyst, and its ability to protect materials from UV radiation (Zheng et al. 2008; Dawidczyk et al. 2008; Woan et al. 2009). TiO2 has ability to switch the surface wettability between hydrophilicity and super-hydrophobic (Wang et al. 1997; Huang et al. 2014; Hoshian et al. 2015). Nano-TiO2 is an excellent cathodic inhibitor, photo-catalyst, an antimicrobial additive, and the antifouling agent in the marine environment (Lovern et al. 2007; Ashraf and Shibli 2008; Etacheri et al. 2015; Zhang et al. 2015). Zhang et al. (2015) studied the removal of humic acid through photodecomposition using Fe2O3/TiO2 nano-wires. The mechanism explained is that Fe2O3 absorbs visible light and transfers the conduction band electron to the electron traps of anatase TiO2.

Nano-Fe2O3 is extensively used for magnetic storage devices, catalysis, targeted drug delivery, magnetic hypothermia, magnetic resonance imaging, and as biosensors (Yang et al. 2008; Laurent et al. 2011; Lee and Hyeon 2012; Cao et al. 2012). Not much work has been carried out using the above three metal oxide together for surface modification of steel. The present study aimed to understand the combined effect of multifunctional nano-oxides, viz., nano-iron oxide, nano-titanium oxide, and nano-cerium oxide, on surface modification of boat-building steel.

Materials and methods

The nano-titanium oxide (4–8 nm, anatase phased and HNO3 stabilized) and iron-oxide Fe2O3 (4–8 nm, 5% aqueous suspension) were purchased from Reinste Nano Ventures India Ltd. Nano-cerium oxide (30–45 nm) was prepared in the laboratory as per the procedure described by Fu et al. (2005). Carbon steel BIS 2062-grade B was procured from Kerala State Inland Navigation Corporation, Cochin, India and the composition are as follows. Carbon 0.22%, Mn 1.5%, Si 0.4%, P 0.045%, and S 0.045% and the rest is iron. Dimethyl formamide is purchased from CDH Mumbai, India. All the chemicals were used as received without any purification.

Preparation of specimen

The steel samples were cut into 3 × 5 cm sizes and pickled using 5% H2SO4 for 3 min, washed with Milli Q type 1 water, and allowed to dry in the air. Then, the specimens were polished up to 1000 grits using a series of silicon carbide papers ranging from 300 to 1000 grits. Degreased the specimen by sonication for 30 min in acetone and then with water. The specimen is kept in the desiccators until the treatment. The nano-Fe2O3, CeO2, and TiO2 mixtures were dispersed in dimethyl formamide as per the details given in Table 1. The suspension was sonicated for 1 h and decanted to remove solid residues. 30 μl of the mixtures of the solutions were pipetted over the specimen and spread using glass slides. Allowed to dry under infrared lamp. The process was repeated for three times on both sides of the specimen and finally kept at 100 °C in an air oven for 1 h. Removed, wrapped with dry tissue paper, and kept in a zipper bag. The specimens were kept in the desiccators until the analysis.

Table 1 Details of treatments of nano-Fe2O3, TiO2, and CeO2 for surface modification boat-building steel BIS 2062
Full size table

Electrochemical evaluation

The linear sweep voltammetric (LSV) analysis of surface modified specimen was done using Metrohm PGstat 302 N with FRA. 3.5% (wt/vol) NaCl is used as an electrolyte, Pt as counter electrode, Ag/AgCl (3 M KCl) as reference electrode, and specimen with exposed area of 1 cm2 as working electrode. The specimen was kept in the electrolyte for 1 h to stabilize the OCP. The LSV was carried out with a scan rate of 0.001 V s−1. Tafel slope analysis was done using the Nova software. The electrochemical impedance spectra of the surface-modified steel specimen were done in 3.5% NaCl with Pt, Ag/AgCl (3 M KCl), and sample, respectively, as a counter, reference, and working electrode. The specimens were scanned from 100 kHz to 0.1 Hz at 60 decades after 30 min immersion in the electrolyte. The generated EIS data were fitted with simple Randle’s equivalent circuit model. All the analyses were repeated at least four times and the average was used for discussion.

Corrosion rates on the surface-modified steel samples were analyzed by a weight loss method as per the ASTM G1. The pre-weighed specimen was immersed in 3.5% NaCl for 40 days under room temperature 28 ± 3 °C. The specimen was removed after the experiment, washed with water, and sonicated to remove adsorbed dirt for 20 min. Dried in an air oven at 100 °C for 30 min and allowed to cool in a desiccator. The final weight was recorded and calculated the corrosion rate using the formula CR = 534W/DAT, where W is the weight loss, D is the density, A is the area in square inch, and time in hours.

Surface characterization

The steel specimen of 0.5 × 0.5 cm was polished using a series of silicon carbide papers up to 2000 grits, degreased using acetone, and washed with water. Surface modification of the specimen was carried out by pipetting 2 μl of nano-material as described earlier. Surface topography of the steel was analyzed under non-contact mode using Park XE 100 AFM. The Si scanning probe with tip radius of less than 10 nm was used. The data were processed using the XEI software. The microstructure of the specimen was analyzed using JEOL Model JSM-6390LV SEM.

Results and discussion

Nature of corrosion products

The nature of corrosion products formed over boat-building steel under atmospheric and 3.5% NaCl environment was compared with the nano-iron oxide using spectrophotometry (Fig. 1). The corrosion product formed in all the three cases was almost similar in the UV region at 192, 280, and 380 nm. In the visible region, a peak at around 480 nm was recorded in all the cases. Fe2O3 exhibits three types of electronic transitions viz. d–d orbital transitions linked with Fe3+ ions, O 2p–Fe 3d ligand to metal charge transfer, and pair excitations coupled with 2 neighboring Fe3+ cations. Ligand-to-metal charge transfer (O 2p—Fe 3d) bands exist in the region of 440–800 nm (Pailhe et al. 2008; Gaudon et al. 2010).

Fig. 1
figure 1

Spectrophotometric evaluation of nano-Fe2O3 and corrosion products formed over the boat-building steel due to air exposure and 3.5% NaCl exposure

Full size image

Microstructure analysis

Clean the surface of the steel specimen was coated individually with nano-iron oxide, nano-titanium oxide, nano-cerium oxide, and the mixture of three metal oxides (3MO) to study the surface characteristics. The surface topography of the specimen is analyzed using AFM and is shown in Fig. 2. The surface topography of the steel surface exhibited that the coatings were uniformly distributed and adhered over the substrate and the grain boundaries of the steel surface were fully covered by the nano-materials. Higher surface area and adhesiveness over the substrate are characteristics of nano-materials (Voevodin et al. 2003; Bjerklie 2005). A line profile drawn on the 1000 nm AFM micrograph and the results of average surface roughness (Ra) and root mean square roughness (Rq) are shown in Fig. 3. Ra highlights the general height variations in the surface and Rq are the root mean square average of the roughness profile ordinates. The order of roughness was nano-cerium-oxide coated > nano-iron oxide > nano-titanium oxide > bare steel > 3MO-coated steel. The nano-metal oxides were occupied in the steel surface uniformly with the lowest surface roughness, indicating that the grain boundaries were sealed completely by the 3MO.

Fig. 2
figure 2

AFM topography of nano-metal oxide and 3MO-coated steel

Full size image
Fig. 3
figure 3

Average roughness (Ra) and root mean square roughness (Rq) of untreated F0, nano-CeO2, nano-Fe2O3, nano-TiO2, and 3MO-coated steel

Full size image

The scanning electron micrographs of untreated steel F0, 3MO-coated steel F4 and F5 are shown in Fig. 4. The nano-3MO particles were uniformly distributed on the steel substrate and the particles were occupied over the grain boundaries. The results were comparable with AFM topographic images. This shows that the 3MO was strongly and uniformly adhered over the surface of steel.

Fig. 4
figure 4

Scanning electron micrographs of bare-steel specimen and 3MO-coated specimens (F4 and F5)

Full size image

Electrochemical evaluation

The electrochemical characteristics of the surface-modified steel specimens were studied using linear sweep voltammetric analysis. Tafel plots of all the 3MO-coated boat-building steel specimens are shown in Supplementary file as Fig S1. The linear sweep voltammetric parameters are shown in Fig. 5. The corrosion potential Ecorr, corrosion current density Icorr, and polarization resistance Rp were ranged, respectively, from − 0.753 ± 0.065 to − 0.613 ± 0.030 V, 3.353 × 10−6 ± 1.47 × 10−6 to 3.59 × 10−5 ± 4.85 × 10−5 A cm−2, and 1961 ± 980 to 6043 ± 2562 Ω cm2. The lowest polarization resistance and higher corrosion current density showed by the specimen coated with 0.005% each of Fe2O3, TiO2, and 0.01% CeO2. The surface coating with 1:1:1 ratio of 3MO showed the highest Icorr and thereby increased corrosion. When Fe2O3 concentration increased from 0.005 to 0.010% in 3MO mixture, the specimen showed decreased corrosion resistance. Whereas nano-CeO2 concentration increased from 0.005 to 0.01% in 3MO mixture, the specimen showed higher corrosion resistance. In the case of F6 the increased Fe2O3 and TiO2 contents in the 3MO, the specimen showed a decrease in the polarization resistance. This indicates the concentration of Fe2O3 and TiO2 has an important role in the corrosion resistance of the 3MO-coated specimen. The optimum concentration CeO2 was 0.01% and its further increase was shown with lower polarization resistance indicating the higher corrosion. Further to check, the impact of higher concentration (double and triple times) by maintaining the same ratios as in F5 showed not much improvement in corrosion resistance (results not shown in the manuscript). Hence, the optimum ratio with maximum efficiency was 0.005:0.01:0.005 of Fe2O3, CeO2, and TiO2. Increased concentration of nano-CeO2 in the surface enhanced the barrier resistance and hence corrosion resistance (Han et al. 2012).

Fig. 5
figure 5

Linear sweep voltammetric data of three metal-oxide mixture-coated boat-building steel

Full size image

There was no significant difference between the Ecorr values between the treatments. The values were about ± 0.075 V than untreated specimen. In general, the corrosion resistance was cathodic in nature (Bethencourt et al. 2004). The cathodic shift was mainly due to the increased concentration of nano-cerium oxide. Here, the Tafel plot showed a longer plateau immediately after the corrosion potential. The plateau of the Tafel curve was due to the influence of CeO2 in the matrix (Hasannejad et al. 2013). The similar behaviour was noted earlier studies on aluminium incorporated with cerium oxide (Ashraf and Shibli 2007). The treatment F5 with the highest polarization resistance highlights the coating formed over the steel matrix smoother with the lowest roughness (Hasannejad et al. 2013). This was further correlated with the results of AFM studies. Based on corrosion current density, the corrosion inhibition efficiency was about 40%.

EIS studies

The Nyquist plots of the 3MO-coated steel after immersing 30 min in 3.5% NaCl are shown in Fig. 6. For estimating the EIS parameters quantitatively, the data are fitted with simple Randle’s equivalent circuit model and the results are shown in Fig. 7. The constant phase element (CPE) was used instead of capacitance in this study. Rp in the high-frequency (HF) and low-frequency (LF) regions were ranged from 9.21 ± 6.54 to 93.08 ± 44.94 Ω cm2 and 1.13 ± 0.09 to 372.25 ± 160.77 Ω cm2, respectively. The CPE in the HF and LF regions was, respectively, ranged from 1.09 × 10−8 ± 5.26 × 10−11 to 2.21 × 10−8 ± 2.00 × 10−8 F and 1.98 × 10−4 ± 7.32 × 10−5 to 5.64 × 10−3 ± 3.86 × 10−4 F, respectively. There are three types of Nyquist plots which were shown by the specimens with different treatments. The untreated control exhibited the Nyquist plot with clear HF and LF domains and both partially overlapped each other (merged) in the middle. The bare steel specimen always has a thin layer of iron oxide and this prevents the interaction of surface with the mild atmospheric corrosives. The merger of HF and LF domains in the Nyquist plots highlights that the thin films of native iron oxide have strong interaction with internal layers and hence higher polarization resistance in the HF domain. The smaller LF domain and lowest polarization resistance indicated the instability of the internal layer in 3.5% NaCl medium. The treatment F1 showed a typical Nyquist plot with independent wider HF and smaller LF domains (Type II). This inferred that the 3MO forms a barrier protection, but the internal layer was highly unstable. Type II plots indicate that Fe2O3–TiO2–CeO2 (1:1:1) form a strong barrier in the outermost region, but it is highly susceptible to degradation under aggressive corrosive marine environment. In the case of F1, Rp in the LF domain was significantly lower, indicating the instability of the internal layer. A similar situation was common in pure aluminium, where aluminium oxide exhibits stronger oxide layer on the surface, but under aggressive environments, it will not resist the stress and, hence, undergo corrosion (Ashraf and Shibli 2007). Here, the equal concentration 3MO form good protective layer probably resists the penetration of electrolyte. This layer was stronger than native Fe2O3 layer formed over the bare metal. The surface coating may not withstand under the aggressive NaCl environment and this was further correlated with the results of LSV studies. The surface has higher amounts of electrons from the oxides of 3MO and the mechanism was discussed in later sections.

Fig. 6
figure 6

Nyquist plots of 3MO-coated steel. a F0–F6 combined plots and b F0 and F1 alone

Full size image
Fig. 7
figure 7

EIS parameters after fitted with simple Randle’s equivalent circuit model

Full size image

These Nyquist plots of F2–F6 were classified as Type III plots. The Nyquist plots showed partial semicircle in the HF domain (Fig. 6). Type III plots showed two independent wider HF and LF domains. The polarization resistance in HF and LF domains was significantly higher than the bare metal, F0, highlighting the improved surface and internal layers due to the nano-3MO coating. This ensures higher corrosion protection of the steel due to the nano-3MO coating. The lowest CPE values in F5 highlight that the coating prevents the uptake of electrolyte to the internal layers, which further prevented corrosion. The coating helps to introduce longer layer capillary paths in the thin film for the approach of electrolytes or corrosive ions to the substrate (Fig. 8). The straight-line slope of the Bode plots (not given in the paper) showed that the coating formed over the steel seldom allowed to penetrate the electrolytes. Probably, occupying the nano-particles in the channels or pores blocked the transport or diffusion of electrolytes from the surface to the internal layers. This highlights the integrity of the coatings and enhanced corrosion protection (Yeh et al. 2006; Lamaka et al. 2008; Domínguez-Crespo et al. 2009). In F1, the 3MO was in 1:1:1 ratio with the lowest concentration of oxides and these oxides occupy the pores of native iron-oxide layer which led to the total closure of the path layer. This led to the instability in the intersection of steel and 3MO layer evidenced by the two separate HF and LF domains with wider gap. Hence, lower polarization resistance increased corrosion.

Fig. 8
figure 8

Pictorial representation of steel and coated surface based on three types of Nyquist plots

Full size image

Corrosion rate

Pre-weighed surface-modified specimens were exposed in 3.5% NaCl for 40 days and corrosion rates were calculated as per the ASTM G1. The results are shown in Fig. 9. The corrosion rates were ranged from 1.7 × 10−3 to 2.5 × 10−3 mils per year. The specimen having 0.005% Fe2O3, 0.005% TiO2, and 0.01% CeO2 exhibited lowest corrosion rate. The results correlated with the findings of LSV, EIS, and AFM studies. The treatment showed 35% more efficient than control.

Fig. 9
figure 9

Corrosion rates of 3MO-coated steel

Full size image

Stress healing

To study the response of the steel if scratched under marine environments, the fishing boats undergo a series of rubbing and scratching during berthing in the shore. The 3MO surface-modified specimen F5 was immersed in the 3.5% NaCl for 24 h to stabilize the OCP in 3.5% NaCl. Next day, a deep scratch was made over the specimen using a sharp stainless steel knife. Changes in open-circuit potential (OCP) before and after the scratch are noted and are shown in Fig. 10. The system has shown increased OCP immediately after the scratch, but regained to the original form within 1 min. The untreated specimen took more than 4 min to reach the stable OCP value. This shows that the healing stress was improved due to the surface treatment of 3MO over the boat-building steel.

Fig. 10
figure 10

Open-circuit potential variation when stress was applied over untreated control and 3MO-treated steel

Full size image

Mechanism of resistance to corrosion

The native Fe2O3 present over the outermost layer of the steel probably behaving like a semiconductor due to the nano-metal-oxide treatment. The bandgap for Fe2O3, TiO2, and CeO2, respectively, was 2.2, 3.2, and 2.76 eV. The conduction band was 0.73, − 0.25, and − 0.32 and valence band was 2.93, 2.91, and 2.44 V, respectively, for Fe2O3, TiO2, and CeO2. The conduction band of the internal layer of the steel was either partially filled or overlapped with valence band; hence, there was no bandgap [http://web.ift.uib.no/AMOS/PHYS208/larsP-N/p-n-slide-lars.pdf (viewed on 30-7-2016)]. The partially filled electron or band in the layer moves the next higher energy levels; hence, the current conduction occurs easily. The outermost layer of steel, native Fe2O3, and conductivity were increased due to the doping of 3MO in the surface. The 3MO treatment over the surface of steel will make the surface behave as a semiconductor.

The bonding in metal-oxide semiconductors is different from the valence electrons of oxygen which are partially or fully transferred to the metal ion. Hence, metal oxides are highly ionic in nature. Fe2O3 and TiO2 have 3d orbital from metal atoms of Fe and Ti and 2p from oxygen. The d orbital contributes more towards conduction band and p orbital towards the valence band (Fujishima and Honda. 1972; van de Krol. 2012). 3d orbital of transition metal is generally shrunken state and exhibits lesser overlap compared to s and p orbitals. The free holes in the valence band form polarons through electrostatic interaction with surrounding oxygen ions. The charge transfer through polarons plays an important role in Fe2O3 and TiO2. The oxygen vacancies of Fe2O3 located in the below the conduction band and it was partly responsible for the intrinsic n-type semi-conductivity. When the 3MO-coated layer interacts with the aqueous environment, the hydroxylation occurs. This mainly depends on the pH of the solution. Specific adsorption of H+ or OH− species affects the charge distribution and potential distribution in the liquid-coating interface. The high electrochemical environment, the electrons, exists initially in the traps and inter-band states are emptied to fill the holes. Hence, a decrease in the electron density in the film and charge separation between the valence and conduction bands was facilitated. This results in a decrease electron–hole transport and competition with hole transport. Finally, long-lived photo-generated holes accumulated near the surface and this promotes oxidation of water (Barroso et al. 2013). Similar mechanism was explained in the case of nano-titanium oxide (Tang et al. 2008; Cowan et al. 2010). This shows that both iron oxide and titanium actively involve the photo-oxidation of water which results in deterrence microorganisms in the vicinity of the surface results a maximum corrosion protection. This might be the reason the black layer formed over the treated specimen.

The conduction band of CeO2 was a Ce 4f band. Cerium ions are known to exhibit variable valency from Ce4+ to Ce3+ ions. The electrons in the cerium-oxide molecules behave as small polarons and oxygen vacancy defect were dominated in the chemical and electronic properties. These oxygen vacancies are mobile and contributed to the oxygen ion transport in the solid solutions (Campbell and Peden 2005; Zhang et al. 2009; Macedo et al. 2010). Oxygen storage capacity of cerium oxide helps to change the oxidation states and this was due to the nearly equal energy of 4f and 3d orbitals. This will help to distribute the electron density between the f and d orbitals (Holgado et al. 2000). The oxygen vacancy defect formation will decrease the oxygen content in the cerium oxide. Cerium oxide also acts as excellent oxygen buffers due to their redox capacity (Chen et al. 2006). The valence and defect structure of CeO2 is dynamic and may change with physical parameters like surface stress (Sheldon and Shenoy 2011). Redox process was prominent in CeO2 under an electrical field (Deshpande et al. 2005). The electrochemical behaviour of CeO2-based space charge models showed that the defect formation energies are different from bulk material. This results in an electrical gradient in the surface to maintain the thermodynamic equilibrium. Thermodynamic equilibrium changes with variation in stress and strain due to point defects and related volume changes (Sheldon and Shenoy 2011). Nano-CeO2 particles exhibit increased oxygen vacancies as the size of the particle decreases (Tsunekawa et al. 1999; Deshpande et al. 2005). The oxygen vacancy formation energy in CeO2 is lower in nano-sized CeO2 particles. It has higher electronic conductivity (Chiang et al. 1996, 1997). The mechanism probably is the large volume of holes created by nano-cerium oxide; this has facilitated the continuous flow of electron transport between the native Fe2O3 and internal layer. This has prevented the electrochemical disturbances to reach the internal layers. This was further reflected the stress recovery experiment. Concluding the increased electronic activity by the three nano-oxides over the surface of steel synergistically prevented electrochemical disturbances and microbial activities. This result increased resistance towards the degradation of the boat-building steel under corrosive environments.

Conclusion

The boat-building steel BIS 2062 surface modified with the mixtures of three multifunctional nano-oxides, Fe2O3, TiO2, and CeO2. The treatment showed uniform coating over the steel, improved the surface roughness than untreated steel, and the grain boundaries were filled with nano-oxides. The electrochemical evaluation showed that the corrosion inhibition was cathodic in nature and the optimum ratio was 0.005% each of Fe2O3, TiO2, and 0.01% CeO2. Electrochemical impedance spectral evaluation exhibited increased polarization resistance on the surface and internal layer of the steel. The resistance to stress was improved due to the nano-3MO treatment over the surface. The surface modification with nano-oxides made the native Fe2O3 layer to behave as a semiconductor, and increased oxygen vacancies of nano-oxides made increased electronic activities. This resulted in deterrence of microbial and electrochemical changes. The surface modification of boat-building steel with three multifunctional nano-oxides is having potential application in marine environments.