1. Introduction

Mineral acids are widely used in various industries for pickling of metals at temperatures up to 60^°C. This technique besides being used to remove corrosion scales from steel surfaces without causing acid attack of the bulk metal is also effectively applied in cleaning of industrial equipment and acidization of oil wells. Corrosion of oil equipment is of significant cost, which is additional to mechanical and other operational problems that require work-over and repairs. It is therefore imperative to add corrosion inhibitors to the solution during pickling in other to reduce the degree of metal attack and the rate of acid consumption (Umoren et al., 2008; Loto et al., 2014).

Corrosion can be viewed as the process of returning a metal to their natural state, i.e. the ores from which they were originally obtained. Approximately one-fifth of the iron and steel produced annually is used to replace rusted metal (Zumdahl, 1993). Mild steel and aluminium are used in fabricating various reaction vessels, reaction tanks, and pipes etc, for industrial uses due to their availability and low cost. However when used in aggressive media they tend to corrode which causes severe loss and malfunctioning of industrial equipment. Organic and inorganic compounds such as indigo dyes (Oguzie et al., 2004), methyl red (Ebenso, 2004a,b), have been used in recent past as corrosion inhibitors. However, discharge of such materials has become unacceptable due to environmental hazards (Sharmila et al., 2010). Plants extracts known for their biodegradability, eco-friendliness, cost effectiveness, low toxicity and easy availability are considered green inhibitors (Sethuraman et al., 2005). The recent trend is towards environmentally friendly and organic inhibitors. Accordingly, great research efforts have been directed towards formulating these inhibitors.

Some of the plant extracts that have been studied include Gum acacia (Mobin and Khan, 2013), Azadirachta indica and Hibiscus Sabdariffa (Oguzie, 2008), ethanol extract of Vernonia amygdalina (Odiongenyi et al., 2009), Telfaria occidentalis (Oguzie, 2005), Allium sativum (Okafor et al., 2005), and ethanol extract of Musa acuminata peel (Eddy et al., 2008), Phyllanthus amarus leave (Okafor et al., 2008), aqueous extract of Fenugreek leaves (Noor, 2007), Calendula Officinalis flower (Subha et al., 1999), aqueous extract of olive leaves (El-Etre, 2007), Berberine (Yan Li et al., 2005) and Damsissa (Abdel-Gaber, 2008) have been studied. Also, Sansevieria trifasciata (Oguzie, 2007), Aloe plant (Al-Turkustani et al., 2010), Bauhinia purpurea leaves (Patel et al., 2009), bamboo leaf extract (Li et al., 2014), Camellia Sinensis (Loto et al., 2014) and red onion skin acetone extract (James and Akaranta, 2009) investigation as corrosion inhibitors have been carried out. Despite the availability and variety of plant materials only few have been thoroughly investigated and data obtained for various corrodents.

African breadfruit (Treculia Africana Decne) is a tropical forest tree that belongs to the plant family Moraceae. The tree plant has beautiful shape and makes a good shade arising from its broad leaves. It bears big heavily seeded oval fruit heads of mass 2.5–13.62 kg and attains 45 cm in diameter, yielding 5–15 kg edible cooking oil seeds after processing. The tree is evergreen, grows up to 40 m high, and produces 30–50 fruits annually (Akubor, 1997; Nwokocha and Ogunmola 2005). The fruits and leaves are good fodder for goats, sheep and pigs (Etukudoh, 2003).

To the best of knowledge of the authors, the only literature on the use of African breadfruit leave extract as corrosion inhibitor is our earlier study (Ejikeme et al., 2012), in which we examined the inhibitive effect of the plant extract on the corrosion of aluminium in HCl solution. The results of the said study encouraged us to investigate further the corrosion inhibition of the plant extract in a more aggressive 1M H₂SO₄ solution on both aluminium and mild steel. This comparative study we speculate will help to access the applicability or otherwise of the plant extract as corrosion inhibitors in the two widely used construction materials.

2. Materials and Methods

2.1. Materials

The composition of the mild steel is as follows: 0.21% C, 0.38% Si, 0.09% p, 0.01% Al, 0.05% Mn, 0.05% S and the remainder iron. The aluminium was purchased at System Metals Industries Limited, Calabar, Nigeria. It was of 98% purity and of the type AA 1060. Other materials and apparatus used include: rotary evaporator, digital high precision weighing balance (Metler), UV-visible spectrophotometer (Sperctron 21D) PEC Medicals, U.S.A. All reagents used were of analytical grade unless otherwise stated.

2.2. Preparation of African Breadfruit Extract

The leaves of African breadfruit were obtained from Nsukka, Southern Nigeria. The leaves were sun-dried and crushed to powder using a manual grinding machine. About 300g of the ground leaves was weighed and soaked in 96% ethanol. After 48h they were filtered and the filtrate concentrated to paste using rotary evaporator. Various weights of the inhibitor ranging from 0.1g - 0.5g were weighed and dissolved in 1M H₂SO₄.

2.3. Phytochemical Screening

The phytochemical analysis of the extract was by standard methods as described in Evans (2000) and Harbone (1998). Specifically the extract was screened for phytochemicals that included tannins, flavonoids, alkaloid, terpenes, saponins, phlobatannins, deoxy sugar, and cardiac glycoside.

2.4. Gravimetric Corrosion Study Technique

For the corrosion study, previously weighed metal coupons were placed in 100 ml of 1M H₂SO₄ solution (blank) and in 100 ml of 1M H₂SO₄ solution with the various concentrations of BLE ranging from 0.1 – 0.5g in open beakers labeled A, B, C, D, E. The beakers were placed in a thermostated paraffin oil bath maintained at 30℃ or 60℃ as the case may be. In each experiment, the clean metal coupons were suspended with the aid of glass rod and hook. These coupons were retrieved progressively at 2h intervals for 8h, washed, dried in acetone and re-weighed. The weight loss was taken as the difference in weight of the specimen before and after immersion determined by weighing with the digital balance. The tests were conducted in triplicate to guarantee the reliability of the results and the mean value of the weight loss was reported.

3. Results and Discussion

3.1. Phytochemical Screening

The result of the phytochemical screening of the plant extract has been reported earlier (Ejikeme et al., 2012). Briefly, the plant extract contain a preponderance of alkaloids, terpenes, flavonoids, tannins, anthraquinones etc. Heteroatoms (P, O, N, S, Se) regarded as centers for adsorption (Obot et al., 2009) which abound in the phytochemicals of BLE are obviously involved in the adsorption of the extracts to the surface of the metals, inhibiting corrosion at such parts of the metal; the net effect of this process being the protection of the metal surface from the attack of the aggressive ions (SO₄^2-) of the acid.

3.2. Corrosion Data

Assessment of corrosion rate, inhibition efficiency and surface coverage for aluminium and mild steel with different inhibitor concentrations was carried out using Eqns. 1 to 3, respectively.

3.2.1. Corrosion Rate (χ)

The corrosion rate (χ) was computed using Eqn. 1 as follows;

(1)

where, M₀ = weight (g) before immersion; M₁= weight (g) after immersion A= area of specimen (cm²), t = exposure time (h)

3.2.2. Inhibition Efficiency (%I)

The inhibition efficiency (%I) was evaluated using Eqn. 2.

(2)

where; χ₀ =corrosion rates in absence of inhibitor; χ₁= corrosion rates in presence of inhibitor

3.2.3. Surface Coverage (θ)

The surface coverage (θ) of the inhibitor was obtained from the experimental data using Eqn. 3.

(3)

where; W₀ =corrosion rates in absence of inhibitor; W₁= corrosion rates in presence of inhibitor

The weight loss of mild steel and Al in 1M H₂SO₄ in the presence and absence of BLE is given in Figs. 1 - 4.

It could be seen from the figures that weight loss of mild steel and aluminium in H₂SO₄ varies with time and was reduced in the presence of the extract compared to the blank (Loto et al., 2014). Furthermore, the weight loss increased with temperature, thus higher at 60℃ for each metal. Aluminium was found to show a lower weight loss than mild steel at the same experimental conditions, indicating that aluminium corroded less than mild steel in 1M H₂SO₄ solution.

Generally, inhibition efficiency (Fig. 5) increased with the concentration of the extract; with the highest efficiency achieved with 0.5g/l of the extract. Though the inhibition efficiency increased with increasing inhibitor concentrations, it decreased with increase in temperature. Decreased inhibitor efficiency with increase temperature is suggestive of physisorption (Ebenso, 2004a). The implication of this observation is that corrosion rate was higher in mild steel at all temperatures in nearly all concentrations in 1M H₂SO₄.

Figure 1. Plot of weight loss against time for mild steel corrosion in 1M H2SO4 at 30℃

Figure 2. Plot of weight loss against time for mild steel corrosion in 1M H2SO4 at 60℃

Figure 3. Plot of weight loss against time for aluminium corrosion in 1M H2SO4 at 30℃

Figure 4. Plot of weight loss against time for aluminium corrosion in 1M H2SO4 at 60℃

Figure 5. Plot of inhibition efficiency of BLE for mild steel and Al corrosion in 1M H2SO4

This result corroborates the weight loss data. It may be argued that aluminium can form an oxide on its surface which alongside with the adsorbed extract made it less prone to attack by the acid corrodent. However, it should be noted that since BLE contain alkaloids, terpenes, flavonoids, tannins, anthraquinones, etc, which have fused benzene rings and heteroatoms in the ring, the chemical complexity of the extract makes it is difficult to assign the inhibiting action to a particular constituent or group of constituents (Li et al., 2014; Nnanna et al., 2014).

The increase in corrosion rate with increase in temperature is due to the fact that chemical reactions are nearly doubled with increase in temperature.

3.3. Isotherm Study

The mode and extent of the interaction between the BLE inhibitor and the metal surfaces were studied by applying adsorption isotherms. In isotherm studies, the linearity of the plot and good correlation coefficient may be interpreted to suggest that the experimental data for the studied inhibitor obey a particular adsorption isotherm but considerable deviation of the slope from unity shows that the isotherm may not be strictly applied. Temkin adsorption isotherm which assumes a uniform distribution of adsorption energy which increases with increase of the surface coverage is given by Eqn. 4 (Mu et al., 2005).

(4)

Equation 4 can be transformed into Eqns. 4a to 4c.

(4a)

(4b)

(4c)

where: a is the lateral interaction term describing the molecular interactions in the adsorption layer and the heterogeneity of the metal surface; θ is the degree of surface coverage; K_ads is the equilibrium constant of adsorption process; C is the concentration of the inhibitor.

Plots of surface coverage (θ) against logarithm of the concentration of the extract (ln C) at 30℃ and 60℃ are shown in Fig. 6. Correlation coefficients of the linear regression (R²) plots of > 0.76 were obtained (Table 1) implying some degree of agreement of the data to the Temkin adsorption isotherm model, especially for the adsorption of BLE to aluminium. The values of the lateral interaction term ‘a’ are all greater than zero (a > 0) indicating lateral attraction between adsorbing molecules of BLE and the surfaces of both mild steel and Al.

Table 1. Some parameters of the linear regression between θ and ln C Temp (℃) Mild Steel Aluminium Kads a R2 Kads a R2 30 0.847 0.926 0.7653 0.642 0.962 0.8802 60 0.923 1.000 0.9076 0.679 1.020 0.9497

Freundlich adsorption isotherm is given by Eqn. 5, and the linearized version by Eqn. 6. Values of lnθ were plotted against lnC (Fig. 7) using corrosion data. It is seen from the figure that the data obtained within the temperature range followed the Freundlich adsorption isotherm model better than Temkin model.

(5)

(6)

where: k and C, represents the equilibrium constant of adsorption and extract concentrations respectively.

Figure 6. Temkin adsorption isotherm for Al and mild steel corrosion in 1M H2SO4 at 30 and 60℃

Figure 7. Freundlich adsorption isotherm for Al and mild steel corrosion in 1M H2SO4 at 30 and 60℃

In both tests with aluminium and mild steel, the fitting at 60℃ was better than that at 30℃ considering the R² of the various graphs of the Freundlich adsorption isotherm plots.

Figure 8 shows the linear dependence of C/θ as a function of C which is obtained using the Langmuir adsorption isotherm is represented by Eqn. 7.

(7)

where: k is the equilibrium constant of the adsorption process, θ is the degree of surface coverage, C the molar inhibitor concentration in the bulk solution and k is the equilibrium constant of the process of adsorption.

African breadfruit leave extract adsorption conformed to Langmuir adsorption isotherm for aluminium more than mild steel at the temperatures studied as can be seen from the R² values from the figures. Also, it fitted the adsorption data worse than Freundlich isotherm for aluminium and mild steel.

3.4. Thermodynamic and Kinetic Studies

The values of ∆G°_ads for the extract are given in Table 2. These values were found to be between -10.438 kJmol^-1 and -10.645kJmol^-1 for mild steel and -10.078 kJmol^-1 and -10.795 kJmol^-1 for aluminium which confirms that BLE adsorbed through physisorption on the aluminium and mild steel surfaces. The negative value of ∆G°_ads indicates that the adsorptions of BLE unto the metal surfaces were spontaneous and that the layer adsorbed were stable (Bouklah, 2006). Values of ∆G°_ads up to -20kJmol^-1 like the ones obtained with BLE in this study are consistent with electrostatic interaction between the charged metal (which indicates physical adsorption) while those more negative than -40kJmol^-1 involved charged sharing or transfer from the inhibitor molecule to the metal surface to form a co-ordinate type of bond (which indicates chemical adsorption) (Umoren et al., 2008). Overall, first order kinetic was better followed at lower temperatures for aluminium and vice–versa for mild steel. The values of the rate constant k, were evaluated from the plots of log ∆w against time.

The value of activation energy (Ea) for the corrosion process in the absence and presence of the inhibitor were calculated using Arrhenius equation represented by Eqn. 8.

(8)

where; P₁=corrosion rates at temperature T_1, and P₂ = corrosion rates at temperature T₂.

The values of K_ads are relatively small indicating that low interaction between the adsorbed molecules and the metal surfaces obtained and that the extracts molecule are physically adsorbed on the metal surfaces through adsorption.

The values were observed to be lower at higher temperatures for both mild steel and Al (Oguzie 2007; Al-Turkustani et al., 2010). Figures 8 to 11 show plots of the logarithm to base 10 of ∆w (final weight loss) against time at 30℃ and 60℃ in H₂SO₄ for mild steel and aluminium. The linear plots reveal a first order kinetic. The least correlation coefficient for the adsorption process for both metals using the BLE for the first order plot was 0.932 (Table 3). This high correlation shows that the process can be modeled to first order kinetics. From Table 3, it is evident that the adsorption process for aluminium at 30°C conforms best to first order kinetics while comparison of the process at 60°C show that it conformed more to first order kinetic for mild steel.

The heats of adsorption (Q_ads) were calculated using Eqn. 9.

(9)

where: θ₁=degree of surface coverage at temperature T₁ and θ_{2 =} degree of surface coverage at temperature T₂.

Figure 8. Langmuir adsorption isotherm for Al corrosion in 1M H2SO4 at 30℃and 60℃

Figure 9. Plot of log ∆W against time for mild steel in 1M H2SO4 in the absence and presence of BLE at 30℃

Table 2. Calculated parameters from Freundlich adsorption isotherm plot for mild steel and aluminium in 1MH2SO4 of the extract Temp (℃) Mild Steel Aluminium Kads ∆Gads n R2 Kads ∆Gads n R2 30 1.136 -10.438 0.582 0.893 0.985 -10.078 0.436 0.936 60 0.843 -10.645 0.768 0.973 0.890 -10.795 0.471 0.988

Figure 10. Plot of log ∆W against time for mild steel in 1M H2SO4 in the absence and presence of BLE at 60℃

Figure 11. Plot of log ∆W against time for Al in 1M H2SO4 in the absence and presence of BLE at 30℃

Figure 12. Plot of log ∆W against time for Al in 1M H2SO4 in the absence and presence of BLE at 60℃

Table 3. Comparison of linear regression (R2) from first order plot Extract Conc. in 1M H2SO4 (g/l) Mild Steel Aluminium Mild Steel Aluminium R2 Rate constant k(sec-1) x 10-3 Rate constant k(sec-1) x 10-3 30°C 60°C 30°C 60°C 30°C 60°C 30°C 60°C 0.5 0.977 0.981 0.986 0.939 0.023 0.020 0.028 0.027 0.4 0.966 0.974 0.996 0.953 0.023 0.024 0.031 0.028 0.3 0.954 0.971 0.999 0.932 0.017 0.020 0.028 0.028 0.2 0.961 0.973 0.998 0.945 0.015 0.019 0.028 0.027 0.1 0.997 0.975 0.998 0.941 0.017 0.025 0.031 0.032 Blank 0.882 0.970 0.999 0.929 0.015 0.025 0.037 0.030

From Table 4, it is evident that in all cases the values of Q_ads are negative indicating that the adsorption and hence the inhibition efficiency, decreases with rise in temperature (Ebenso, 2004b). The values of Q_ads for mild steel were more negative than those obtain for aluminium. Similar result had been reported by Ebenso (2004a). Activation energy values were higher for mild steel (40.28 - 52.46kJ/mol) than aluminium (7.46 - 19.75kJ/mol). This general increase in the activation energy of the inhibition process with increase in the concentration of the inhibitors shows the process followed a physical adsorption mechanism.

Table 4. Calculated values of activation energy (Ea) and heat of adsorption (Qads) for mild steel and Al corrosion in 1M H2SO4 Concentration(g/l) Mild Steel Aluminium Ea (kJmol-1) Qads (kJmol-1) Ea (kJmol-1) Qads (kJmol-1) Blank 40.28 - 7.46 - 0.1 43.94 -18.53 11.27 -8.52 0.2 44.92 -10.50 10.68 -6.85 0.3 43.17 -8.68 8.99 -4.47 0.4 43.56 -8.06 14.92 -10.81 0.5 52.46 -20.33 19.75 -16.60

4. Conclusions

African breadfruit leave extract acted as an inhibitor in the corrosion of mild steel and aluminium in H₂SO₄ solution. Corrosion rate was found to be greater in mild steel than aluminium and increased with increase in temperature. Inhibition efficiency increased with increase in the concentration of the extract but reduces with increase in temperature indicating that the adsorption of African breadfruit leave extract on the metals surfaces was unfavourable at high temperature. The adsorption of the extract on the metals surfaces obeyed Freundlich, Temkin and Langmuir adsorption isotherms at 30 and 60℃. Heat of adsorption was negative in all cases which show that adsorption and hence inhibition efficiency decreases with rise in temperature. Activation energies were higher in mild steel (40.28–52.46kJ/mol) than aluminium (7.46–19.75kJ/mol). Mechanism of physical adsorption is proposed for the adsorption process.