Effects of Phyllanthus muellerianus Leaf-Extract on Steel-Reinforcement Corrosion in 3.5% NaCl-Immersed Concrete

Abstract

This paper investigates Phyllanthus muellerianus leaf-extract effects on steel-reinforcement corrosion in concrete immersed in 3.5% NaCl, simulating saline/marine environment. Different concentrations of the leaf-extract were admixed in steel-reinforced concrete samples, which were immersed, with normal control, in the test-environment, while positive control samples were immersed in distilled water. Electrochemical measurements of corrosion-rate (by linear-polarization-resistance instrument), corrosion-current (by zero-resistance-ammeter) and corrosion-potential (by high impedance multimeter) were obtained for assessing the reinforcing-steel corrosion. Analyzed results showed that the corrosion-rate exhibited excellent correlation (R = 98.82%, Nash-Sutcliffe Efficiency = 97.66%, ANOVA p-value = 0.0006) with function of the admixture concentration and of the corrosion noise-resistance (ratio of corrosion-potential and corrosion-current standard deviations). The 0.3333% Phyllanthus muellerianus (per weight of cement) exhibited optimal efficiency, η = 97.58% ± 1.28% (experimental) or 95.33% ± 4.25% (predicted), at inhibiting concrete steel-reinforcement corrosion in the test-environment, which compares well with the positive control performance model, η = 97.96% ± 0.03%. The experimental and predicted models followed the Langmuir adsorption isotherm, which indicated physisorption as the Phyllanthus muellerianus leaf-extract adsorption mechanism on the reinforcing-steel. These support suitability of the N-, S-, and O-containing and π-electron rich Phyllanthus muellerianus leaf-extract as an environmentally-friendly inhibitor for effective corrosion-protection of steel-reinforcement in concrete designed for the saline/marine environment.

1. Introduction

Steel-reinforced concrete is a global material of choice for building structures and infrastructure, especially due to its relatively lower cost and the inherent protection of the reinforcing-steel embedment from environmental degradation within the concrete [1,2]. The reinforcing-steel (steel-rebar), in turn, improves the load-bearing strength properties of the concrete for the steel-reinforced concrete application. The inherent protection of steel-reinforcement in concrete takes the form of high alkalinity of concrete pore environment, usually of pH > 12, which promotes the development of a thin passive film of oxide layer on the reinforcing-steel surface [1,3,4,5,6]. However, destruction of the thin passive oxide layer, by penetrations of aggressive agents through the concrete, from its service-environment, unto the steel-rebar, renders the rebar susceptible to corrosion degradation. Dominant, out of the many, environmental agents capable of promoting steel-rebar corrosion include ingress of chloride ions, in artificial saline from de-icing salts in temperate region, or in natural marine from seawater in coastal areas, into concrete unto the steel-rebar [2,7,8,9,10,11].

Steel-reinforced structures, usually in these chloride-contaminated environments, include: steel-reinforced energy structures, commercial or domiciliary buildings, tunnels, roads/pavements, bridges, storm barriers, wharfs, harbors, platforms, and man-made concrete islands [2,12,13,14,15,16,17,18,19]. The rusts produced by the ensuing corrosion degradation of the steel-reinforcement are expansive within the concrete such that they generate hoop stress that leads to cracks, spalling, delamination and loss of structural integrity of the steel-reinforced structure. Maintenance, rehabilitations and repairs, for averting catastrophic collapse of corrosion deteriorated steel-reinforced structures, constitute huge spending and substantial parts of fiscal budgetary allocations in many countries [17,18,20]. Otherwise, the insidious nature of chloride-induced corrosion failure of steel-reinforced concrete structures and infrastructures in saline/marine environment, if unchecked, could culminate in safety risks to life and/or loss of properties [21,22].

Despite many other corrosion-protection methods, the use of green, environmentally-friendly corrosion inhibiting admixtures, especially from natural plant materials, is attracting preference among researchers and construction stakeholders due to many advantages. Corrosion inhibiting substances from natural plant materials are of relatively lower cost due to their ready availability in the environment. They combine this desirable quality with being biodegradable, non-toxic and non-hazardous to the environmental ecosystem. These properties are unlike traditional synthetic chemical inhibitors, e.g., compounds of chromates and nitrites [23], which are known to be both toxic to the environment and capable of causing temporary or permanent damage to organs (e.g., kidneys and/or liver) of living beings [24,25]. Rather, corrosion inhibitors from natural plant sources are rich in biocompatible organic compounds that are able to exhibit corrosion-protection by adsorption on the metal-solution interface [26,27]. These naturally occurring substances can be safely extracted, synthesized, characterized and applied for the corrosion inhibition process using simple and standard procedures in contrasts to injurious chemicals and processes required for synthesizing traditional chemical inhibitors [26,28]. However, the fact that corrosion inhibitors are medium and/or material specific necessitates test on the corrosion inhibition property of the extract from specified plant on given metallic material, and in a given medium, of interest [29]. In a closely related consideration, substances that had exhibited positive inhibition effects on reinforcing-steel corrosion but in concrete pore solution are still recommended for further tests in physically cast concrete, in which they may require additional compatibility criteria [30,31,32].

Studies on biochemical characterization of Phyllanthus muellerianus (P. muellerianus) Euphorbiaceae leaf-extract showed that the plant is rich in lone-pairs and π-electron containing heteroatoms and biocompatible phytochemical constituents [33,34]. From those studies, phytochemical constituents found in P. muellerianus include tannins, flavonoids, phlobatannins, alkaloids, saponins, and terpernoids. According to further reports [34,35], P. muellerianus leaf-extract is non-toxic, rather, the leaf of this plant has been found suitable for medicinal purposes including potencies of deep wound healing and antiplasmodial (i.e., against malaria causing parasite) activities. Corrosion research works have also identified P. muellerianus leaf-extract as an effective inhibitor of steel-reinforcement corrosion in concrete for the industrial/microbial service-environment [33,36].

Special motivation for this work had been drawn from studies that had successfully employed leaf-extract from Phyllanthus amarus (a different plant but from the same Euphorbiaceae family [37] as P. muellerianus) for inhibiting steel corrosion in acidic sulfate and in chloride contaminated media [29,38]. However, there is a dearth of studies on the anticorrosion potential of P. muellerianus leaf-extract on reinforcing-steel in concrete for chloride contaminated medium. Therefore, the objective of this study was to investigate the effects of P. muellerianus leaf-extract on steel-reinforcement corrosion in concrete immersed in 3.5% NaCl, simulating saline/marine environment.

2. Materials and Methods

2.1. Experimental Materials

Steel-reinforced concrete slabs for the study were cast in duplicates of size 100 mm × 100 mm × 200 mm that contained similar concentrations of P. muellerianus leaf-extract admixture, and which had been obtained following details from studies [33,36]. Six variations of the P. muellerianus leaf-extract were employed as admixtures in the duplicated design of steel-reinforced concrete specimens for the saline/marine test-environment. For these, the leaf-extract concentrations ranged from 0% P. muellerianus for the control or normal samples (Ctrl) in increment of 0.0833% (per weight of cement for the concrete mixing) up to 0.4167% admixtures. In addition to these, another duplicate of steel-reinforced concrete specimen with 0% P. muellerianus admixture was employed as set of positive controls for immersion in distilled water environment (i.e., Ctrl in Water). This usage of positive control without admixture was to ascertain that the corrosion effects in the experimental samples for the saline/marine environment followed from the immersion of the samples in their test-environment rather than from any other environmental effects.

In each sample of steel-reinforced concrete slab, was embedded, during casting, 150 mm part of a 190 mm length by 12 mm diameter deformed steel-rebar, such that the remaining 40 mm protrusion could be used for electrochemical test connections. Prior to their embedment in the concrete samples, each of the steel rods had been subjected to surface preparations as per ASTM G109-99a [39]. The reinforcing-steel is composed of elements by weight percent (wt. %) that include: 0.273 C, 0.780 Mn, 0.403 Si, 0.240 Cu, 0.109 Ni, 0.142 Cr, 0.039 P, 0.037 S, 0.0083 Nb, 0.0086 Co, 0.0063 Sn, 0.0037 Ce, 0.0032 V, 0.016 Mo and the balance Fe.

2.2. Experimental Setup

The saline/marine test-environment into which the twelve samples of steel-reinforced concrete slabs, for corrosion testing in this environment, were longitudinally and partially immersed was constituted of 3.5% NaCl test-solutions in plastic bowls. This usage of 3.5% NaCl medium for the saline/marine environment followed practice in studies [40,41,42] as well as the identification from text of the 3.5% by weight as the typical content of salt in most seawater [19]. The remaining two samples for corrosion test in this study were the duplicate of Ctrl in Water samples that were also longitudinally immersed in plastic bowls but which contained distilled water. The diagrammatic and pictorial representation of this longitudinal immersion model of samples followed that which has been presented in [43,44].

From each of these fourteen samples, three different, non-destructive tests (NDT) of electrochemical measurements were obtained in five days for the first 40 days and in seven days interval for the next seven weeks. These totaled 89 days of electrochemical test-experiments. The NDT electrochemical tests to which the steel-reinforced concrete samples were subjected include:

• Corrosion-potential (CP) measurements versus Cu/CuSO₄ electrode (CSE), Model 8-A, obtained from Tinker & Rasor^® (San Bernardino, CA, USA), using a high impedance digital multimeter (MASTECH^® instrument, Guangdong, China) [44], conforming to ASTM C876-91 R99 [45].
• Corrosion-current (CC) measurements, versus CSE using zero resistance ammeter (ZRA), Model ZM3P obtained from Corrosion Service^® (Markham, ON, Canada) [43,46,47,48].
• Corrosion-rate (CR) measurements from linear polarization resistance using the three-electrode LPR Data Logger, Model MS1500L, obtained from Metal Samples^® (Munford, AL, USA) [46,48], and that gave direct readout of CR in mpy unit. The three-electrode system by the instrument include a brass plate auxiliary electrode, a Ag/AgCl SCE reference electrode (EDT direct-ION^®, Dover, UK) and the steel-reinforcement working electrode [40,49,50]. The LPR Data Logger instrument was connected to the steel-reinforced concrete specimen using typical electrochemical cell setup detailed in [26,51].

2.3. Initiation of Corrosion Test-Data Analyses—Statistical Distribution Fitting

As prescribed in ASTM G16-95 R04 [52], for avoiding grossly erroneous conclusion, analyses of the non-destructive electrochemical measurements of steel-reinforcement corrosion were initiated through fittings of the test-data to the Normal and Weibull distributions [51]. Compatibility of the dataset from each test-variable per sample was also investigated using the Kolmogorov-Smirnov goodness-of-fit test-techniques [53,54]. Measurements of central tendencies, μ, and measurements of dispersions, σ, for each of the distribution fittings of datasets were obtained by maximum likelihood estimation (MLE) procedures [55] which for the Normal distribution employs the well-known formula [51,56] for μ_Normal in Equation (1) and σ_Normal in Equation (2):

$${\mathsf{\mu }}_{\mathrm{Normal}}=\frac{1}{n}{\displaystyle \sum _{i=1}^n{x}_i}$$

(1)

$${\mathsf{\sigma }}_{\mathrm{Normal}}=\sqrt{\frac{1}{n-1}{\displaystyle \sum _{i=1}^n{\left(x_i-\mathsf{\mu }\right)}^2}}$$

(2)

whereas for the Weibull distribution, it was required that the MLE procedures be applied for estimating the Weibull shape, k, and scale, c, parameters through the solution of the simultaneous Equations (3) and (4) given by [51,56,57]:

$$\frac{n}{\widehat{k}}-n\ln \left(\widehat{c}\right)+{\displaystyle \sum _{i=1}^n\ln x_i}-{{\displaystyle \sum _{i=1}^n\left(\frac{x_i}{c}\right)}}^{\widehat{k}}\ln \left(\frac{x_i}{c}\right)=0$$

(3)

$$\widehat{c}-{\left\{\frac{1}{n}{\displaystyle \sum _{i=1}^n{x}_i^{\widehat{k}}}\right\}}^{\frac{1}{\widehat{k}}}=0$$

(4)

From the unbiased estimated values of k and c, the Weibull mean (μ_Weibull) and standard deviation (σ_Weibull) were then evaluated by the respective expression in Equations (5) and (6):

$${\mathsf{\mu }}_{\mathrm{Weibull}}=c\mathrm{\Gamma }\left(1+\frac{1}{k}\right)$$

(5)

$${\mathsf{\sigma }}_{\mathrm{Weibull}}=\sqrt{c^2\left\{\mathrm{\Gamma }\left(1+\frac{2}{k}\right)-\left[\mathrm{\Gamma }{\left(1+\frac{1}{k}\right)}^2\right]\right\}}$$

(6)

2.4. Corrosion Noise-Resistance (R_n) Analyses

Measurements of dispersion from the descriptive statistics that exhibited better compatibility with the scatter of the corrosion-potential and the corrosion-current test-data were used for modeling the corrosion noise-resistance [56,58,59]. This noise-resistance was modeled as the ratio of standard deviation of corrosion-potential (CP) to the standard deviation of corrosion-current (CC), using the formula in Equation (7):

$$R_{\mathrm{n}}=\frac{{\mathsf{\sigma }}_{\mathrm{CP}}}{{\mathsf{\sigma }}_{\mathrm{CC}}}$$

(7)

2.5. Surface Coverage and Inhibition Efficiency Analyses

Measurements of central tendencies from the descriptive statistics that exhibited better compatibility with the scatter of CR test-data were used for modeling the surface coverage (θ) and inhibition efficiency (η) of P. muellerianus leaf-extract on the steel-reinforcement. These facilitated estimations of θ and η, for each admixed P. muellerianus concentration in concrete relative to the Ctrl sample, through the respective relationship in Equations (8) and (9) [60,61]:

$$\mathsf{\theta }=\frac{{\mathrm{CR}}_{\mathrm{Ctrl}\mathrm{sample}}-{\mathrm{CR}}_{\mathrm{Admixed}\mathrm{sample}}}{{\mathrm{CR}}_{\mathrm{Ctrl}\mathrm{sample}}}$$

(8)

$$\mathsf{\eta }=\frac{{\mathrm{CR}}_{\mathrm{Ctrl}\mathrm{sample}}-{\mathrm{CR}}_{\mathrm{Admixed}\mathrm{sample}}}{{\mathrm{CR}}_{\mathrm{Ctrl}\mathrm{sample}}}\times 100$$

(9)

2.6. Adsorption Isotherm Modeling

The surface coverage θ was subjected to the fitting of Langmuir adsorption isotherm through Equation (10) [56,60,62,63,64]:

$$\frac{\rho }{\mathsf{\theta }}=\frac{1}{K_{\mathrm{ads}}}+\rho$$

(10)

In Equation (10), ρ is the concentration of P. muellerianus leaf-extract admixed in concrete, and K_ads is the Langmuir equilibrium constant for the adsorption-desorption process. The K_ads estimation facilitated the modeling of the nature of P. muellerianus leaf-extract adsorption through use of the K_ads for the separation factor, R_L, computation in Equation (11) [56,60,65,66]:

$$R_{\mathrm{L}}=\frac{1}{1+K_{\mathrm{ads}}C{R}_0}$$

(11)

The R_L computation, therefore, finds usefulness for indicating P. muellerianus leaf-extract adsorption on reinforcing-steel as irreversible if R_L = 0, or favorable if 0 < R_L < 1, or linear if R_L = 1, or unfavorable if R_L > 1 [60,66]. In addition to these, the K_ads of the Langmuir isotherm model was used for estimating the Gibbs free energy of adsorption ΔG_ads using Equation (12) [56,60,62,63,64]:

$$\Delta G_{\mathrm{ads}}=-2.303RT\log \left(55.5K_{\mathrm{ads}}\right)$$

(12)

In Equation (12), R is the molar gas constant ≡ 8.314 J/mol∙K, and T is the absolute temperature ≡ 300 K, while 55.5 represents the concentration of water in solution expressed in molar. Estimated value of Gibbs free energy of adsorption that is around or more positive than −20 kJ/mol suggests prevalent physisorption (or physical adsorption) mechanism, while estimated value that is around or more negative than −40 kJ/mol suggests prevalent chemisorption (or chemical adsorption) mechanism.

3. Results and Discussion

3.1. Statistical Distribution Fitting and Analyses of Corrosion Test-Variables

Figure 1 shows the mean plots, obtained from the statistical analyses of corrosion test-variables from each sample by the Normal and the Weibull distributions, for the corrosion potential Figure 1a, corrosion current Figure 1b and corrosion rate Figure 1c. The mean plots of corrosion potential and corrosion current, Figure 1a,b, also include the range of standard deviations for each of the datasets of these corrosion test-variables for each steel-reinforced concrete samples. In addition, and for aiding direct interpretations from the plotting, the plots of corrosion potential (Figure 1a) include linear plots of corrosion risk probability as per ASTM C876-91 R99 [45] while the plots of corrosion rate (Figure 1c) include a linear plot of typical corrosion rate criteria from literature [18,67].

From the plots in Figure 1, the general trend whereby corrosion test-variables of the normal control samples (Ctrl) were high-valued compared to the corrosion test-variables of samples admixed with P. muellerianus leaf-extract could be observed from each of the electrochemical tests. This general trend was such that values of the corrosion test-variables tend to range from the high-values for the Ctrl samples through reduced values for the samples with P. muellerianus down towards the low-values for the positive control samples that were immersed in water (Ctrl in Water). According to [68], the high values of CP test-variable, which classified to the severe corrosion risk region of [45], encountered in the Ctrl samples indicate prevalence of anodic (i.e., corroding) areas in the steel-rebar embedment in the concentration cell of the 3.5% NaCl-immersed normal control concretes. By these, it could also be understood from [68] that the lower CP values obtained from the 3.5% NaCl-immersed concrete samples that were admixed with P. muellerianus suggest region whereby the plant extract promotes passive layer that enhances resistance of the steel-rebar to dissolution. In similar manner, and according to [69], the high CC values in the Ctrl samples gives the measure of the corrosion activity between the corroding anode, the steel-rebar in the samples, and the passive cathode of the Cu in the Cu/CuSO₄ reference electrode [68,69]. Therefore, the low values of CC obtained from the P. muellerianus admixed concretes also corroborate the low values of CP by the low CC values indicating low measure of corrosion activity between the steel-rebar in the concrete and the Cu in the reference electrode employed for this electrochemical test.

In addition to these, and with just some few exceptions, especially in the corrosion potential plots of some samples, there were general agreements in the pattern of corrosion test-variable models by the Normal and by the Weibull distributions. However, the few over-predictions and discrepancies in model of corrosion test-variables necessitate investigation of the distribution that describes the scatter of the test-data from each corrosion test-variable better among the Normal and the Weibull distribution models.

Figure 2 therefore shows results of the Kolmogorov-Smirnov goodness-of-fit (K-S GoF) analyses of the scatter of the dataset of corrosion test-variables, from each sample of steel-reinforced concrete employed in this study, like the Normal and like the Weibull distribution. The plots in Figure 2 also include linear plot of α = 0.05 level of significance for ascertaining analyzed dataset scattering like, or otherwise, each of the distribution fitting models. These showed that all the datasets of corrosion potential and of corrosion current scattered like both the Normal and the Weibull probability distributions according to the K-S GoF criteria at α = 0.05 level of significance. However, datasets of corrosion rate from six steel-reinforced concrete samples, out of the 14 samples studied and which include datasets from the duplicates of normal control (Ctrl and Ctrl Dup), were not scattered like the Normal distribution. In contrast, all datasets of corrosion rate scattered like the Weibull distribution. These results indicate that while both the Normal and the Weibull distributions could be employed for describing the corrosion potential and the corrosion current, only the Weibull distribution, but not the Normal, could be used for describing the corrosion rate test-data.

3.2. Corrosion Rate and Corrosion Noise Resistance for Correlation Modeling

Based on the results from the K-S GoF analyses, the Weibull distribution model was employed in the study as the descriptive statistics for the corrosion test-variables. By these, the corrosion noise resistance, R_n, was also evaluated for each steel-reinforced concrete samples using ratios of Weibull model of standard deviations of corrosion potential to the standard deviations of corrosion current as per Equation (7). Figure 3, therefore, shows the results of the corrosion noise resistance for the steel-reinforced concrete samples, superimposed on the plots of corrosion rate, in increasing order of the corrosion rate effect on the embedded reinforcing-steel in the samples.

From this figure, it could be noted that the duplicate samples of normal Ctrl exhibited the lowest corrosion noise resistance as well as the highest corrosion rate in the experimental study. In addition, the 0.3333% P. muellerianus sample exhibited the highest corrosion noise resistance as well as the least corrosion rate in the study. This 0.3333% P. muellerianus sample and the 0.4167 P. muellerianus_Dup sample both exhibited higher corrosion noise resistance and lower corrosion rate than the values of these variables for the duplicate samples of “Ctrl in Water”. Despite this, it is still worth noting that the duplicate samples of positive control, “Ctrl in Water” samples exhibited the kinds or agreements in corrosion rate effects that was exhibited by the duplicate samples of normal control, the Ctrl immersed in NaCl medium.

As could be observed in Figure 3, studies have shown that the corrosion noise resistance is generally high-valued at low corrosion rate and low-valued at high corrosion rate due to the fact that the corrosion noise resistance tracks the linear polarization resistance [36,56,59,60]. This relationship of corrosion noise resistance with corrosion rate, and vice versa, had been employed for detailing correlation expressions between these corrosion variables in reported works [36,56,60]. In a similar manner, it could be shown that the corrosion rate, CR, obtained from this study, exhibited relationship with the corrosion noise resistance, R_n, and the P. muellerianus leaf-extract admixture concentration, ρ, which can be expressed in compact form as Equation (13):

$$\mathrm{CR}=0.5057\left[\rho +{\displaystyle \sum _{\mathsf{\lambda }=0}^5{\left(-1\right)}^{\mathsf{\lambda }+1}\cdot a_{\mathsf{\lambda }}\cdot {10}^{3\mathsf{\lambda }}\cdot {\left(1/R_{\mathrm{n}}\right)}^{\mathsf{\lambda }}}\right]$$

(13)

In Equation (13), values of the coefficients a_λ {λ = 0, 1,…, 5} are as presented in

Table 1. Values of the coefficients a_λ in Equation (13).

λ	aλ
0	0.6396
1	4.6268
2	29.3848
3	49.3821
4	31.1072
5	6.2572

.

For the correlation fitting expression in Equation (13), the correlation coefficient, R = 98.82% and the Nash-Sutcliffe efficiency, NSE = 97.66%, which interpret to excellent model fitting efficiency [70,71]. Analysis of variance for the correlation fitting expression (see

Table 2. Analysis of variance for the correlation fitting expression in Equation (13).

Source of Variations	df	SS	MS	F	p-Value
Treatment	6	33.3438	5.5573	34.7807	0.0006
Residual	5	0.7989	0.1598
Total	11	34.1428

) indicated that p-value = 0.0006, which implies that the correlated relationship between the dependent variable, CR, and the independent variables ρ and R_n is statistically significant within 95% confidence interval.

3.3. Corrosion Inhibition Effects by Experimental and Correlated Model

Plots of inhibition efficiency are presented in Figure 4, in ranking order of the P. muellerianus leaf-extract performance on concrete steel-reinforcement corrosion in the tested medium, using the experimental and the correlation predicted CR data applications to Equation (9). For further comparisons of corrosion effects, the figure also includes a linear plot of the analyzed corrosion rate reduction effect from the positive control (“Ctrl in Water”), idealized as inhibition efficiency relative to the normal control (Ctrl), through use of Equation (9). By these, therefore, it could be deduced from the figure that the P. muellerianus leaf-extract concentrations effectively inhibited steel-reinforcement corrosion in concrete immersed in the chloride contaminated environment. However, in addition to these, the inhibition efficiency performance of the P. muellerianus leaf-extract concentrations tended towards that of the “Ctrl in Water” by both the experimental and the predicted models of corrosion. Thus, while the 0.3333% P. muellerianus leaf-extract exhibited optimal inhibition efficiency, η = 97.58% ± 1.28% (experimental) or 95.33% ± 4.25% (predicted), this inhibition effect ranged towards that of the idealized inhibition efficiency, η = 97.96% ± 0.03%, by the “Ctrl in Water”.

The high range of idealized inhibition effects by the positive control samples laid credence to the fact that the corrosion effects from the normal controls followed from the immersion of the normal control samples in the saline/marine environment and not from any other effects. It is also worth noting that the optimal inhibition efficiency in this study by the 0.3333% P. muellerianus leaf-extract concentration finds similarity with the performance from previous work [36] by this same concentration on steel-reinforcement corrosion in H₂SO₄-immersed concrete. This performance was despite the longer experimental period of steel-reinforced concrete immersion in the chloride induced corrosive test-medium in this study. Rather, the inhibition performance by the 0.3333% P. muellerianus leaf-extract in the chloride contaminated medium in this study still surpassed that of the same admixture concentration in that acidic sulfate immersed medium. In addition to these, almost all concentrations of P. muellerianus leaf-extract inhibitors in this study exhibited excellent model, η > 90% of inhibition efficiencies both by the experimental and the correlation prediction models. The 0.16667% P. muellerianus leaf-extract admixture concentration modeled with the least inhibition efficiency of η = 84.65% ± 12.88% (experimental) or 88.22% ± 0.26% (predicted) in this study indicates positive as well as “very good” efficiency model [70,71]. These exhibited further contrasts with results from that previous work whereby insufficient amount of P. muellerianus leaf-extract admixture exhibited negative inhibition effect (i.e., promotes corrosion aggravation) on steel-reinforcement corrosion in the H₂SO₄-immersed concrete.

3.4. Adsorption Isotherm Model of Experimental and Correlated Data

Figure 5 shows plots of the results obtained from fitting the experimental and the correlated predicted data to the Langmuir adsorption isotherm, through requisite applications of Equations (8) and (10), while

Table 3. Parameters from Langmuir adsorption isotherm fitting of corrosion data.

Isotherm Parameter	Experimental Model	Predicted Model
Kads	62.9422	71.1493
R, correlation coefficient, (%)	99.51	99.90
RL, separation factor	3.2995 × 10−3	2.9192 × 10–3
ΔGads Gibbs free energy (kJ/mol)	–20.3492	–20.6549

presents the adsorption fitting parameters. In the figure, the ratio of P. muellerianus concentration to the surface coverage model (ρ/θ) was plotted, as the ordinate, against the concentration (ρ) of the plant extract, as the abscissa, for obtaining the intercept 1/K_ads, the reciprocal of the Langmuir equilibrium constant, as per Equation (10). Figure 5 shows that both the experimental and correlation predicted data linearly fit the Langmuir adsorption isotherm model, which by their R > 99% correlation coefficients, as tabulated, interpret also to excellent isotherm modeling efficiency. The values of the separation factor that were in the range 0 < R_L < 1 indicate favorable adsorption [60,66] by the P. muellerianus leaf-extract inhibitor on the concrete steel-reinforcement. In addition, the negative values of Gibbs free energy of adsorption indicate spontaneity of the P. muellerianus leaf-extract adsorption while the values of the parameter around –20 kJ/mol imply prevalent physisorption as the mechanism of P. muellerianus leaf-extract adsorption on the reinforcing-steel surface.

The results from the study showed P. muellerianus leaf-extract as an effective environmentally-friendly inhibitor of steel-reinforcement corrosion in steel-reinforced concrete immersed in 3.5% NaCl test-environment, for representing saline/marine environment. This inhibition effectiveness of the natural plant extract on steel-rebar corrosion in the chloride contaminated medium could be due to its constituents of lone pair and π-electrons rich organic hetero-atoms that had shown positive inhibition prospects in acidic environment in previous studies.

3.5. Organic Bio-Constituent Model from P. muellerianus Leaf-Extract

The effectiveness of P. muellerianus leaf-extract on the corrosion protection of reinforcing steel in the chloride contaminated medium engender interest in the bio-constituent model that could be available in the leaf-extract of the natural plant [56]. For this, the spectrum obtained from the Fourier Transform Infrared (FT-IR) spectroscopy instrument, the Perkin-Elmer^® FT-IR System (Spectrum BX, Waltman, MA, USA), application to the P. muellerianus leaf-extract is presented in Figure 6.

As had been previously detailed, the functional classes of adsorbed vibrations, for the identified frequencies in the P. muellerianus leaf-extract FT-IR spectrum (Figure 6), had been reported, along with the phytochemical characterization, by [33]. However, the rendering of this FT-IR spectrum to the Euclidean Search of the Fluka library, by the Perkin-Elmer^® instrument [72], suggested hit list of 10 organic compounds that are presented, in this study, as 3-D optimized structures of the compounds in Figure 7.

Figure 7 shows that, out of the 10 compounds identified in the P. muellerianus leaf-extract spectrum, eight contain aromatic rings, seven are N- (nitrogen), two are S- (sulfur), one is Br- (bromine) and seven are O- (oxygen) containing lone-pair and/or π-electrons rich organic compounds. In addition, it is worth noting that five of the identified compounds for P. muellerianus find similarities to those that were identified for Morinda lucida in [72] while the remaining five were not found in Morinda lucida but in P. muellerianus leaf-extract. In this study, however, P. muellerianus leaf-extract concentrations exhibited high effectiveness performance, which compares well to that obtained from positive (distilled water immersed) control samples, on the inhibition of steel-reinforcement corrosion in concrete for the saline/marine simulating environment. This performance was in similar manner with that by the P. muellerianus for the industrial/microbial environment in [36], and with that obtained by the also N-, S-, and O-containing Morinda lucida for the saline/marine environment in [72]. The corrosion inhibition effects by these natural plant-extracts find agreements with what obtained in reported works [73,74,75,76], where effective corrosion inhibition of steel material in acidic environments had been observed with uses of N-, S-, and O-containing and π-electron rich chemical derivatives. These considerations, therefore, also garner further support for the usage of the N-, S-, and O-containing and π-electron rich P. muellerianus leaf-extract as an environmentally-friendly inhibitor of steel-reinforcement corrosion in concrete designed for the saline/marine environment.

4. Conclusions

Anticorrosion effects of P. muellerianus leaf-extract on steel-reinforcement in concrete immersed in 3.5% NaCl, for simulating saline/marine environment, was investigated, in an experimental design employing normal control (in tested medium) and positive control (in distilled water). Statistical analyses of the experimental data identified the Weibull distribution as the descriptive statistics of better-fit for the corrosion test-data, according to the Kolmogorov–Smirnov goodness-of-fit criteria. Analyzed results showed that the corrosion rate from LPR instrument exhibited excellent correlation (R = 98.82%, Nash–Sutcliffe Efficiency = 97.66%, ANOVA p-value = 0.0006) with function of the P. muellerianus leaf-extract concentration and of the corrosion noise-resistance.

The 0.3333% P. muellerianus leaf-extract admixture (per weight of cement) exhibited optimal inhibition efficiency, η = 97.58% ± 1.28% (experimental) or 95.33% ± 4.25% (predicted), on steel-reinforcement corrosion in concrete samples immersed in the saline/marine test-environment. These results of corrosion inhibition effectiveness from the P. muellerianus leaf-extract admixtures compare well with the performance model, η = 97.96% ± 0.03%, of corrosion test-results obtained from the positive (i.e., the distilled water immersed) control samples. Subjection of the experimental and correlation predicted data to adsorption isotherm modeling indicated they both followed the Langmuir adsorption isotherm. This isotherm model identified spontaneous/favorable adsorption and prevalent physisorption as the mechanism of P. muellerianus leaf-extract on the corrosion-protection of the reinforcing-steel in the concrete for the saline/marine environment.

These performances support suitability of the N-, S-, O-containing and π-electron rich P. muellerianus leaf-extract, as indicated from Euclidean Search Hit List application to the FT-IR spectrum of the plant extract, as an environmentally-friendly corrosion inhibitor. Usage of the extract from this natural plant is therefore recommended for effective corrosion-protection of steel-reinforcement in concrete designed for the saline/marine environment.