Corrosion inhibition effect of a benzimidazole derivative on heat exchanger tubing materials during acid cleaning of multistage flash desalination plants

Desalination xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Desalination journal homepage: www.elsevier.com/locate/desal

Corrosion inhibition effect of a benzimidazole derivative on heat exchanger tubing materials during acid cleaning of multistage flash desalination plants Ikenna B. Onyeachu, Moses M. Solomon , Saviour A. Umoren, Ime B. Obot, Ahmad A. Sorour ⁎

Center of Research Excellence in Corrosion, King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia

ARTICLE INFO

ABSTRACT

Keywords: Desalination plant Cu-Ni alloy Acid cleaning Corrosion inhibition Benzimidazole

A benzimidazole derivative, 2-(2-bromophenyl)-1-methyl-1H-benzimidazole (2BPB) has been studied as a corrosion inhibitor for Cu-Ni 70/30 and 90/10 alloys in 1 mol/dm3 HCl solution at low and high temperatures using the weight loss, electrochemical (potentiodynamic polarization (PDP), electrochemical impedance spectroscopy (EIS), & cyclic voltammetry (CV)), and surface characterization (scanning electron microscopy (SEM) & Fouriertransform infrared spectroscopy (FTIR)) techniques. The effect of immersion time (up to 72 h) and addition of iodide ions on the inhibition efficiency of 2BPB have also been investigated. At low temperature, 1.0 g/L 2BPB inhibits Cu-Ni 70/30 and 90/10 alloys by 88.9 ± 4.8% and 57.5 ± 1.3%, respectively. The performance of 2BPB improves with increase in immersion time and addition of iodide ions but slightly depreciates with rise in temperature. 2BPB acts as a mixed type corrosion inhibitor and adsorbs on the alloys surfaces through physical adsorption mechanism. SEM and FTIR results confirm the adsorption of 2BPB on the alloys surfaces. 2BPB is a potential low toxic candidate for the formulation of acid corrosion inhibitor for Cu-Ni alloys.

1. Introduction

and evaporator condensers made of Cu-Ni 70/30 and 90/10 alloys experienced corrosion rate between 53 and 66 mpy during 6 h acid cleaning with 2% HCl solution containing no corrosion inhibitor [6]. Khadom et al. [7] also observed that Cu-Ni 90/10 alloy exposed to 5% HCl solution for 3 days at 55 °C experienced a corrosion rate of 20.83 g/ m2 in the absence of a corrosion inhibitor. The severe corrosion experienced by these Cu-Ni alloys could be due to the impediment to form a stable Cu2O passive layer during acid cleaning in the absence of a corrosion inhibitor. It therefore becomes imperative for an effective corrosion inhibitor to be added to acid cleaning solution before deployment. Few research studies have been devoted towards finding effective corrosion inhibitors for the acid cleaning of Cu-Ni alloys (Table 1). These reports reveal that nitrogen–containing organic compounds are, by far, the most investigated corrosion inhibitors for the alloys. Related studies on pure metallic copper have also shown sulfur- and nitrogen–containing compounds to be highly effective corrosion inhibitors [8,9]. The sulfur and nitrogen moieties in the corrosion inhibitor compounds provide lone pair electrons to form coordinate bonds with Cu0, Cu+, or Cu++ species leading to the formation of protective complexes on the metal surface [8,9]. Obviously, the efficacy of the compounds in Table 1, based on the inhibition efficiencies

Copper-nickel alloys exhibit good corrosion resistance, excellent thermal and electrical conductivity, good mechanical ductility, and excellent antifouling properties [1,2]. The good corrosion resistance property is due to the formation of a protective passive film, Cu2O on the surface [1,2]. These unique properties allow copper-nickel alloys to be used as brine heaters and evaporator condensers in the multistage flash (MSF) desalination plants [3]. One of the operational challenges of MSF desalination plants is the formation of inorganic scales on the surface of the alloy components; a common phenomenon capable of decreasing plant production efficiency and enhancing under deposit corrosion [4,5]. The frequently adopted descaling technique in the MSF desalination plants is acid cleaning [4,5] using hydrochloric or sulfuric acids in the concentration range of 2–5% [6]. The use of hydrochloric acid is usually preferred because of the formation of soluble chloride products in lieu of insoluble sulfates. During acid cleaning in the absence of a corrosion inhibitor, brine heaters and evaporator condensers in MSF desalination plants usually suffer accelerated corrosion. For instance, the technical report on one of the largest desalination plants in the world revealed that brine heaters

⁎

Corresponding author. E-mail address: [email protected] (M.M. Solomon).

https://doi.org/10.1016/j.desal.2019.114283 Received 25 November 2019; Accepted 11 December 2019 0011-9164/ © 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Ikenna B. Onyeachu, et al., Desalination, https://doi.org/10.1016/j.desal.2019.114283

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Table 1 Some reported organic-based corrosion inhibitors for copper-nickel alloys in acid cleaning solution. S/N

Inhibitor

1

Chemical structure

Cu-Ni alloy

Acid studied

Maximum η

Conc. with max. η

Ref.

Benzotriazole

90/10

1.5 M HCl

99.8% @ 35 °C; 86.8% @ 55 °C

0.1 M @ 35 °C; 0.02 M @ 55 °C

[11]

2

Benzotriazole

90/10

0.5 M H2SO4

98.0%

1.82 mM

[10]

3

Naphthylamine

90/10

5% HCl

85.6% @ 35 °C

8.5 mM

[7]

4

Phenylenediamine

90/10

5% HCl

45.0% @ 45 °C

8.5 mM

[7]

η = inhibition efficiency.

[7,10,11], provide strong motivation for the development of acid cleaning corrosion inhibitors for Cu-Ni alloys using nitrogen-containing compounds. Consequently, this work was designed to evaluate the corrosion inhibition performance of a benzimidazole derivative, 2-(2-bromophenyl)-1-methyl-1H-benzimidazole (2BPB), against Cu-Ni 70/30 and 90/10 alloys corrosion in a simulated acid cleaning solution. Despite being a nitrogen–containing compound, 2BPB exhibits low bioaccumulation (log Po/w = 4.005 [12]) and has no long alkyl chains which usually increase the toxicity of corrosion inhibitors [13]. This ensures environmental safety when the acid cleaning solution containing 2BPB is eventually discharged into water bodies, as is common in desalination plants. Previous works have shown that 2BPB exhibited excellent inhibition for carbon steel in acidic solutions [13,14]. The weight loss and electrochemical techniques were used to acquire information about the corrosion kinetics and the inhibition mechanism of the Cu-Ni alloys in 1 mol/dm3 HCl solution without and with 2BPB. Fourier-transform infrared spectroscopy (FTIR) technique was applied to characterize the products formed on the surfaces of the alloys after corrosion in 1 mol/dm3 HCl solution without and with 2BPB. Finally, the effect of 2BPB on the surface degradation of the CuNi alloys was studied with scanning electron microscopy (SEM). The effects of temperature, immersion time, and iodide ions addition on the inhibitive performance of 2BPB were also explored.

0.0003% P, 0.5% Sb, 0.0583% Pb, 0.0202% Si, 0.017% S, 0.0056% As, 10% Ni, and the remainder is Cu [11]. For weight loss experiments, the dimension of the Cu-Ni 70/30 coupons was 0.250 × 3.280 × 3.114 cm (surface area = 23.63 cm2) while that of the Cu-Ni 90/10 coupons was 0.178 × 2.810 × 3.406 cm (surface area = 21.36 cm2). Samples for electrochemical experiments were mounted in epoxy resin adopting a disc-like shape with an exposed surface area of 0.63 cm2. Prior to each test, the Cu-Ni alloys samples were abraded with 400, 600, and 800 grit emery papers, ultrasonically cleaned in 1:1 (v/v) ratio of ethanol-water solution for 15 s, and then dried with warm air [15]. 2.3. Weight loss experiments The standard immersion procedure as described in the NACE/ASTM G31 [16] was followed. Experiments were conducted at 25 and 60 °C under static and normal atmospheric conditions for up to 72 h. Sample treatment after each immersion test followed the ASTM-G 01–90 procedure [17]. The weight loss (WL) was calculated using Eq. (1).

WL (g) = W0

W1

(1)

where W0 and W1 are the initial and final weight (g) of the samples. The corrosion rate (v) was computed using the following equation [16]:

v (mm/yr) = 2. Experimental section

87600 × WL AT

(2)

where WL = average weight loss (g), ρ = density of the sample (g cm−3), T = immersion time (h), and A = surface area of the sample (cm2). The inhibition efficiency (η) of 2BPB and 2BPB + KI was calculated using Eq. (3):

2.1. Chemicals and solutions The benzimidazole derivative, 2-(2-bromophenyl)-1-methyl-1Hbenzimidazole (97% purity), was purchased from Merck, USA. The stock solution of 2BPB was prepared in isopropanol according to the method previously reported [13]. The stock solution was then diluted with 1 mol/dm3 HCl solution to obtain desired working concentrations of 0.1–1.0 g/L. The synergist, potassium iodide was also procured from Merck, USA and was used at a concentration of 1 mM.

(%) =

v(blank) v(inh.) × 100 v(blank)

(3)

where v(blank) = corrosion rate in uninhibited solution and v(inh.) = corrosion rate in inhibited solution.

2.2. Materials and preparation

2.4. Electrochemical experiments

Two grades of Cu-Ni alloys (70/30 and 90/10) were used in the study. The chemical composition of the Cu-Ni 70/30 alloy is as follows: 67.84% Cu, 30.6% Ni, 0.63% Fe, 0.6% Mn, 0.02% Zn, 0.018% C, 0.002% S and other trace elements [1]. The chemical composition of the Cu-Ni 90/10 alloy is: 0.148% Sn, 0.2% Fe, 0.134% Zn, 0.015% Al,

A three-electrode system consisting of a Cu-Ni alloy (70/30 or 90/ 10) working electrode, a KCl-saturated Ag/AgCl reference electrode, and a graphite rod counter electrode was used for all electrochemical experiments. The electrochemical experiments performed include electrochemical impedance spectroscopy (EIS), potentiodynamic

2

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Fig. 1. Representative plots showing the variation of open circuit potential (Eocp) with time for (a) Cu-Ni 70/30 and (b) Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution in the absence and presence of 2BPB at 25 °C.

polarization (PDP), and cyclic voltammetry (CV). Prior to the electrochemical experiments, the corrosion system was monitored for 1 h, which was sufficient for a steady state open circuit potential (Eocp) condition to be achieved (Fig. 1). At the Eocp, the EIS experimental parameters were as follow: initial frequency = 100,000 Hz, final frequency = 0.01 Hz, point acquired = 10 points per decade, signal amplitude = 10 mV at corrosion potential (Ecorr). The PDP tests were carried out by polarizing the working electrode away from the Eocp beginning from an initial potential of – 0.25 V vs. Eocp to a final potential of 0.7 V vs. Eocp at a scan rate of 0.2 mV/s. The CV studies were carried out in the potential range of −0.6 V vs. Ag/AgCl to +0.6 V vs. Ag/AgCl with a sweep rate of 10 mV/s and 4 repeat cycles. All the electrochemical experiments were performed at room temperature (approximately 25 °C).

2.5. FTIR and SEM experiments After 72 h of immersion in 1 mol/dm3 HCl solution without and with 1.0 g/L 2BPB, the corrosion product films on the surfaces of the alloys were carefully scrapped and subjected to FTIR analysis. This was conducted in the attenuated total reflectance (ATR) mode within the wavenumber range of 4000 cm−1 to 500 cm−1 at a resolution of 4 cm−1 with the aid of a PerkinElmer Version 10.03.05 FTIR spectrometer. The surface morphologies of the corroded alloys after 72 h immersion in the 1 mol/dm3 HCl solution without and with 1.0 g/L 2BPB was observed with a scanning electron microscope (SEM), JEOL JSM-6610 LV model.

3

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

3. Results and discussion

frequency. This is indicative of resistive behavior and relates to the solution resistance [8,19]. Secondly, a linear relationship exists between log /f/ and log /Z/ at the intermediate frequency region (i.e. log /f/ = 0 Hz to log /f/ = 3 Hz). This relates to the charge transfer at the metal/electrical double layer interface [8,19]. Thirdly, in the low frequency region (i.e. log /f/ = −2 Hz to log /f/ = 0 Hz), the total impedance increases with a decrease in the frequency. This corresponds to the Warburg impedance and signifies diffusion-controlled corrosion process [8,19]. Oxygen and chloride ions from the bulk solution can diffuse into the alloy-solution interface at the instant of sample immersion and species like CuCl2− can diffuse from the interface to the bulk solution at longer immersion time [8]. In the presence of 2BPB, the diameter of the capacitive loop (Fig. 2(a & b)) increases and the impedance modulus is shifted to a higher value (Fig. 2(b & d)). This observation, which is concentration dependent, indicates higher resistance to charge transfer in the inhibited systems than in the uninhibited acid solution. This implies that, the corrosion process was retarded in the inhibited medium than in the

3.1. Electrochemical studies 3.1.1. EIS measurements The electrochemical reactions at the metal/electrolyte interface can be studied using a non-destructive EIS technique. Herein, we applied this technique to study the corrosion of Cu-Ni alloys exposed to 1 mol/ dm3 HCl solution in the absence and presence of 2BPB at room temperature. The impedance spectra, in Nyquist (Fig. 2(a & c)) and Bode modulus (Fig. 2(b & d)) formats recorded for Cu-Ni 70/30 and Cu-Ni 90/10 alloys in the studied acid medium are shown in Fig. 2. The impedance spectra for the uninhibited systems (Fig. 2(a & c)) exhibit two capacitive loops at high and low frequency regions, and are related to charge transfer [1] and diffusion processes [18], respectively. Three distinguishable segments characterize the corresponding Bode graphs (Fig. 2(b & d)). Firstly, in the high frequency region (i.e. log /f/ = 3 Hz to log /f/ = 5 Hz), the total impedance decreases with increasing

(b)

(a)

(d) (c)

Fig. 2. Electrochemical impedance spectra for (a,b) Cu-Ni 70/30 and (c,d) Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution in the absence and presence of 2BPB at 25 °C in Nyquist and Bode modulus representations. 4

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

CE

(a)

(b)

Fig. 3. The physical model, equivalent circuits used for modelling, and the corresponding fitted graphs. (a) Equivalent circuit was used for modelling impedance of Cu-Ni 70/30 in 1 mol/dm3 HCl solution without and with 0.1–0.3 g/L 2BPB; impedance of Cu-Ni 90/10 in 1 mol/dm3 HCl solution in the absence and presence of the studied concentrations of 2BPB. (b) Equivalent circuit used for modelling the impedance of Cu-Ni 70/30 in the HCl solution containing 0.5–1.0 g/L 2BPB. Rs = solution resistance, ϕ1 = constant phase element (CPE) of the passive film layer or the outer oxide film formed on the metal surface, Rf = resistance of the passive film, ϕ2 = CPE of the electrical double layer (EDL), Rct = charge transfer resistance of the EDL, W = Warburg diffusion impedance. CE denotes the counter electrode.

where Y0 = magnitude of CPE, j = imaginary unit (j2 = −1), and w = angular frequency. The n parameter, which usually lies between 0 and 1, is adjustable and defines the CPE power [1,2]. The CPE behaves as an ideal capacitor when n = 1. For the value of n = 0 or 0.5, the CPE represents an ideal resistor or a Warburg impedance, respectively [19]. The frequency-dependent impedance response of the ECs in Fig. 3 can be described by a simple circuit analysis as given in Eqs. (5) & (6). This allows the calculation of the polarization resistance (Rp) as the summation of Rf and Rct (Eq. (7)). The computed Rp values were used to calculate the inhibition efficiency (ηEIS) following Eq. (8) [10].

uninhibited and could be attributed to the adsorption of inhibitor molecules on the alloys surfaces [20]. By closely inspecting Fig. 2(a), it is observed that, the diffusion phenomenon diminishes in the presence of inhibitor and is near absent in the systems inhibited with higher concentrations of 2BPB (0.5 g/L to 1.0 g/L). This suggests that the corrosion mechanism becomes more charge transfer – controlled than diffusion controlled in the presence of higher 2BPB concentrations [19,20]. In order to account for the above observation, the impedance data for Cu-Ni 70/30 in HCl solution containing 0.5–1.0 g/L 2BPB were modelled using equivalent circuits (EC) devoid of Warburg diffusion impedance (Fig. 3(b)). The other impedance data were modelled using a similar EC but with inclusion of Warburg diffusion impedance (Fig. 3(a)). In the ECs, Rs = solution resistance, ϕ1 = constant phase element (CPE) of a passive film layer or an outer oxide film formed on the alloys surfaces, Rf = resistance of the passive oxide film, ϕ2 = CPE of the electrical double layer (EDL), Rct = charge transfer resistance of the EDL, W = Warburg diffusion impedance. The impedance of the CPE is defined by the equation below [1]:

ZCPE

1 = Y0 (jw )n

1

/ Z / =Z R s +

1 + ZCPE1

Z Rf +

(

1 1 ZCPE2 + Z Rct + ZW

)

1

(5) 1

/ Z / =Z R s +

1 ZCPE1

Rp = Rf + Rct

(4) 5

+ Z Rf +

(

1 1 ZCPE2 + Z Rct

)

1

(6) (7)

Desalination xxx (xxxx) xxxx ± 2.0 ± 1.3

– 63.0 66.0 81.7 86.1 88.9 – 25.9 57.5

EIS

318.8 ± 6.1 862.6 ± 7.3 936.7 ± 8.4 1742.7 ± 6.0 2294.9 ± 4.8 2884.5 ± 7.1 23.4 ± 2.0 31.6 ± 1.3 55.0 ± 3.7 0.71 1.03 0.70 0.70 1.00 0.90 5.43 4.60 0.59 308.4 ± 7.0 147.0 ± 12.3 877.0 ± 10.2 1701.0 ± 8.8 2204.5 ± 6.3 2759.0 ± 9.5 9.2 ± 3.6 17.9 ± 0.6 25.8 ± 3.6 ± ± ± ± ± ± ± ± ± 0.5 0.4 0.5 0.5 0.5 0.6 0.5 0.6 0.5 271.8 ± 0.0 551.0 ± 0.0 552.0 ± 0.0 882.0 ± 0.0 580.5 ± 0.0 511.1 ± 0.0 1.3 ± 0.0 585.2 ± 0.0 583.5 ± 0.0 34.1 ± 0.0 19.4 ± 0.0 13.4 ± 0.0 – – – 9.0 ± 0.0 7.0 ± 0.0 11.8 ± 0.0 10.4 ± 5.1 15.6 ± 2.2 59.7 ± 6.5 41.7 ± 3.2 90.4 ± 3.2 125.5 ± 4.6 14.2 ± 0.3 13.7 ± 2.0 29.2 ± 3.7 214.0 ± 0.0 5.2 ± 0.0 57.1 ± 0.0 56.4 ± 0.0 24.1 ± 0.0 22.1 ± 0.0 5.5 ± 0.0 3.4 ± 0.0 3.4 ± 0.0

± ± ± ± ± ± ± ± ±

0.1 0.1 0.1 0.1 0.0 0.1 0.1 0.1 0.1 0.8 1.0 0.8 0.8 0.8 0.9 0.9 0.9 1.0

Rp(0)

Rp(1)

× 100

(8)

Cu

Cu+ + e

(9)

Cu+

Cu2 +

(10)

Ni

Ni2+

+e

+ 2e

(11)

0.0 0.1 0.1 0.1 0.0 0.0 0.1 0.1 0.1

In the second region (from A–B), there is a sharp decrease in current density with potential increase. This transition can be associated with the formation of CuCl precipitates on the copper surface, according to Eq. (12) [8,23]. Based on reports by other authors [19,24,25], this chemistry is consistent with the primary oxidation of the more abundant element (Cu) in the alloy from the atomic Cu0 to the univalent ion Cu+ which eventually combines with the Cl− ions in solution to form the insoluble CuCl. Nevertheless, this protection is not sustained as the formation of soluble Cu/Ni chloride products [8] triggers the sharp increase in current density with potential increase, as indicated by the region between B and C. This chemistry is illustrated by the Eqs. (13) and (14).

± ± ± ± ± ± ± ± ± Cu-Ni 90/10 2BPB 2BPB

2BPB 2BPB 2BPB 2BPB 2BPB

Rp(1)

3.1.2. PDP measurements The polarization curves of Cu-Ni 70/30 and 90/10 alloys in 1 mol/ dm3 HCl solution in the absence and presence of 2BPB at room temperature are given in Fig. 4. The anodic branch of the Cu-Ni 70/30 curve in the blank acid solution (inserted graph in Fig. 4(a)) exhibits phenomena characteristic of active dissolution and a series of active – passive transitions. The region from O to A corresponds to the active dissolution of the alloy in the acid solution, which was proposed to proceed according to the following reactions [1,22]:

1.3 1.4 1.5 1.8 1.8 1.8 1.4 1.5 1.4 Cu-Ni 70/30

Blank 0.1 g/L 0.3 g/L 0.5 g/L 0.7 g/L 1.0 g/L Blank 0.1 g/L 1.0 g/L

(%) =

where Rp(0) and Rp(1) are the polarization resistances in the absence and presence of 2BPB, respectively. The values of impedance parameters are summarized in Table 2. In the table, it is found that the Rp of Cu-Ni 70/30 in the uninhibited acid solution (318.8 ± 6.1 Ω cm2) is significantly higher than that of Cu-Ni 90/10 (23.4 ± 2.0 Ω cm2). This indicates a higher corrosion resistance exhibited by Cu-Ni 70/30 alloy in the studied acid solution than Cu-Ni 90/10. Such can be attributed to the greater tendency for the Cu-Ni 70/ 30 alloy to form a layer of protective oxide film in the chloride solution, based on previous reports [1,21]. Nevertheless, the addition of 2BPB to the corrosive medium enhances the corrosion resistance property of both Cu-Ni alloys. For instance, the Rp of Cu-Ni 70/30 and Cu-Ni 90/10 increased to 862.6 ± 7.3 Ω cm2 and 31.6 ± 1.3 Ω cm2, respectively upon addition of 0.1 g/L 2BPB. This corresponds to a corrosion protection efficiency of 63.0 ± 6.1% and 25.9 ± 2.0%, respectively. Increase in the concentration of the inhibitor further boosted the corrosion resistance of the alloys. The highest concentration of 2BPB studied (1.0 g/L) inhibited the Cu-Ni 70/30 and Cu-Ni 90/10 alloys corrosion by 88.9 ± 4.8% and 57.5 ± 1.3%, respectively. This shows that, 2BPB possesses inhibiting effect but is a better corrosion inhibitor for Cu-Ni 70/30 alloy than Cu-Ni 90/10. It is valuable to point out the increase in the Rf value with increasing 2BPB concentration (Table 2). For instance, the Rf value increased from 15.6 ± 2.2 Ω cm2 in the case of Cu-Ni 70/30 to 125.5 ± 4.6 Ω cm2 as the inhibitor concentration was raised from 0.1 g/L to 1.0 g/L. Similar behavior is observed in the case of Cu-Ni 90/10. This can be explained in terms of the gradual growth and increase in the thickness of the adsorbed inhibitor layer. It can also be deduced from Table 2 that the surfaces of the alloys immersed in both unprotected and protected acid solutions were rough (n1 ranges between 0.8 ± 0.0 and 1.0 ± 0.1) and this could be due to the presence of oxide and/or adsorbed inhibitor films. Furthermore, the contribution of diffusion to the corrosion process is confirmed by the value of n2 (Table 2).

0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0

Rct (Ω cm2) n2 Y2 (μF cm−2 s−(1−αc)) W (mΩ−1 s−0.5) Rf (Ω cm2) n1 Y1 (μF cm−2 s−(1−αc)) Rs (Ω cm2) Alloy System/conc.

Table 2 Electrochemical impedance parameters for Cu-Ni 70/30 and Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution in the absence and presence of 2BPB at 25 °C.

x2 × 10−3

Rp (Ω cm2)

ηEIS (%)

± ± ± ± ±

6.1 7.3 8.4 6.0 4.8

I.B. Onyeachu, et al.

Cu+ + Cl CuCl + Cl 6

CuCl CuCl2

(12) (13)

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

(a) E

O

(b)

Fig. 4. Potentiodynamic polarization curves for (a) Cu-Ni 70/30 and (b) Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution in the absence and presence of 2BPB at 25 °C.

7

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

icorr the corrosion current density in the presence of 2BPB. The presence of 0.1 g/L 2BPB decreases the icorr from 12.86 μA/cm2 and 23.36 μA/ cm2 in the case of Cu-Ni 70/30 and 90/10, respectively to 7.71 μA/cm2 and 17.82 μA/cm2 corresponding to inhibition efficiency of 40.05% and 23.72%. Increase in the concentration of the inhibitor further decreased the corrosion current density. For instance, by increasing the concentration to 1.0 g/L, the icorr further decreased to 2.88 μA/cm2 and 10.89 μA/cm2 in the case of Cu-Ni 70/30 and 90/10, respectively and the protection efficiency increased to 77.60% and 53.38%. This confirmed the earlier assertion that the inhibiting property of 2BPB towards Cu-Ni alloys is concentration dependent and that 2BPB is a better inhibitor for Cu-Ni 70/30 than Cu-Ni 90/10. As earlier mentioned, the polarization resistance of Cu-Ni 70/30 (318.8 ± 6.1 Ω cm2) in the studied corrosive environment is higher compare to that of Cu-Ni 90/10 (23.4 ± 2.0 Ω cm2) (Table 2). This behavior is also noted in Table 3, as the corrosion current density of CuNi 70/30 alloy in 1 mol/dm3 HCl solution (12.86 μA/cm2) is lower than that of Cu-Ni 90/10 alloy (23.36 μA/cm2). This again reflects the higher corrosion resistance of Cu-Ni 70/30 alloy than Cu-Ni 90/10 and is in agreement with the report of other authors [26,27]. The authors [26,27] had reported that, the corrosion resistance of Cu-Ni alloy increased with increase in Ni content up to approx. 30%, it remained constant or slightly decreased at Ni content above 30%. Accordingly, the corrosion resistance is due to increased passivity occasioned by the formation of a compact film consisting of NiO or Ni(OH)2 and Cu2O.

Table 3 Electrochemical polarization parameters for Cu-Ni 70/30 and Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution in the absence and presence of 2BPB at 25 °C. System/conc.

Alloy

−Ecorr (mV vs. Ag/AgCl)

icorr (μA/cm2)

βa (mV/dec)

βc (mV/dec)

ηPDP (%)

Blank 0.1 g/L 0.3 g/L 0.5 g/L 0.7 g/L 1.0 g/L Blank 0.1 g/L 1.0 g/L

Cu-Ni 70/30

215.14 221.99 174.31 198.61 229.91 187.01 487.65 507.59 500.37

12.86 7.71 6.33 5.73 3.46 2.88 23.36 17.82 10.89

57.60 83.60 56.40 69.60 70.40 117.60 32.10 40.00 30.40

31.70 38.80 53.60 71.20 34.20 104.40 103.60 48.70 12.20

– 40.05 50.78 55.44 73.09 77.60 – 23.72 53.38

2BPB 2BPB 2BPB 2BPB 2BPB

Cu-Ni 70/30

2BPB 2BPB

Ni2 + + 2Cl

NiCl2

(14)

Generally, it is accepted that the corrosion resistance property of copper in chloride-rich aqueous environment is due to the formation of cuprous oxide (Cu2O) [19,23–25] (Eq. (15)), which imparts passivity on the metal surface. Ismail et al. [26] reported that, in the case of Cu-Ni alloys, the formation of NiO, according to Eq. (16), provides additional passivity and the extent of this passivity is directly related to the content of Ni in the Cu-Ni alloy. Therefore, the secondary transition to passivation noticed between regions C and D can be attributed to the formation of the protective Cu2O and NiO products on the alloy surface. The relatively steady current density observed between regions D and E, therefore, indicates that the passive layer on the Cu-Ni alloy has become enriched and thickened with the Cu2O and NiO products.

2CuCl2 + 2OH NiCl2 + 2OH

Cu2 O + H2 O + 4Cl NiO + H2 O + 2Cl

3.1.3. CV measurements The cyclic voltammograms of Cu-Ni 70/30 and 90/10 alloys immersed in 1 mol/dm3 HCl solution without and with 2BPB are presented in Fig. 5. In the voltammogram for Cu-Ni 70/30 alloy in the blank acid solution (Fig. 5(a)), the forward sweep of the 2nd, 3rd, and 4th cycles exhibits three peaks designated as Ia, IIa, and IIIa, at potential of 15.96 ± 2.01 mV vs. Ag/AgCl, 212.0 ± 4.23 mV vs. Ag/AgCl, and 457.8 ± 3.52 mV vs. Ag/AgCl, respectively. The forward sweep denotes the anodic reactions while the reverse sweep represents the corresponding cathodic reactions. In previous reports [28,29], the anodic oxidation of Ni and the formation of NiO and/or Ni(OH)2 (Eqs. (11), (14), & (16)) occur at potential < 100 mV vs Ref. It would be reasonable to assign the peak at lower potential (i.e. 15.96 ± 2.01 mV vs. Ag/AgCl) to the anodic oxidation of Ni and the formation of NiO and/or Ni(OH)2. The IIa and IIIa peaks at 212.0 ± 4.23 mV vs. Ag/AgCl and 457.8 ± 3.52 mV vs. Ag/AgCl, respectively were identified to be due to Cu−Cu+ and Cu+−Cu2+ couples [8]. Specifically, Appa Rao and Kumar [8] reported the Cu−Cu+ and Cu+−Cu2+ couples at +250 mV and +450 mV, respectively. The IVc, Vc, and VIc peaks in the reverse sweep correspond to the cathodic reduction reactions [8,28,29]. By comparing the cyclic voltammogram of Cu-Ni 70/30 alloy immersed in the uninhibited acid solution (Fig. 5(a)) with that of the CuNi 90/10 alloy (Fig. 5(b)), it is observed that, the intensity of peak Ia in Fig. 5(b) significantly decreased in Fig. 5(a). This clearly shows the influence of Ni content on the corrosion process. The cyclic voltammograms of Cu-Ni 70/30 alloy in 0.1 and 1.0 g/L 2BPB inhibited solution are shown in Fig. 5(c & e). Similar graphs for Cu-Ni 90/10 are given in Fig. 5(d & f). A comparison of Fig. 5(c–f) to (a & b) reveals the following: (i) the current density of the peaks in the 2BPB-inhibited voltammograms are shifted to lesser values. For instance, the current density of peak IIa in Fig. 5(a) is 74.12 ± 4.23 mA/ cm2. It decreased to 52.93 ± 3.13 mA/cm2 and 36.12 ± 0.23 mA/ cm2 in the systems inhibited with 0.1 g/L and 1.0 g/L 2BPB, respectively. (ii) The intensity of peak Ia in Fig. 5(a) decreased significantly in Fig. 5(c & e) and the extent of decrease is concentration dependent. (iii) The peak IIIa in Fig. 5(a & b) is near absent in Fig. 5(c–f). All these observations are indicative of corrosion inhibition effect of 2BPB [8].

(15) (16)

In the presence of the inhibitor (2BPB), it is observed from Fig. 4(a) that both the anodic and cathodic current densities are reduced and the corrosion potential slightly shifts towards the anodic direction. Under the studied conditions, the main cathodic reaction would be the reduction of oxygen in the solution (4H+ + O2 + 4e− → 2H2O) [1,2,19]. This infers that, 2BPB acted as a mixed type inhibitor in the process of inhibiting the alloy corrosion in the studied corrosive medium. An interesting observation is made when comparing the anodic branches of the PDP curves acquired in the inhibited systems in Fig. 4(a) with that in the uninhibited solution. The extent of transition observed in the regions A to B and C to D becomes diminished in the 2BPB inhibited curve. This suggests that, the inhibitor interferes with the oxidation reactions of the alloy and the level of interference depends on the concentration of the inhibitor in the system. In other words, the adsorption of 2BPB on the alloy surface decreases the dissolution rate of the alloy. In the Cu-Ni 90/10 curves (Fig. 4(b)), the presence of 2BPB is found to have minimal effect on the corrosion process, which is in agreement with the EIS results and portrays 2BPB as a better inhibitor for Cu-Ni 70/30 than 90/10 alloy. The polarization parameters derived from the active region of the curves are summarized in Table 3. They include the corrosion potential (Ecorr), corrosion current density (icorr), and the anodic and cathodic Tafel slopes (βa & βc). The inhibition efficiency also given in the table was calculated using the equation: PDP

i corr × 100 0 icorr

= 1

(17)

0

where icorr is the corrosion current density in the absence of 2BPB and

8

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

IIa

(b)

(a)

Ia Ia II a

IIIa

IIIa

IVc

VIc

IVc

VIc

Vc

Vc

(d)

(c)

(f)

(e)

Fig. 5. Cyclic voltammograms of (a, c, e) Cu-Ni 70/30 and (b, d, f) Cu-Ni 90/10 alloys in 1 mol/dm3 HCl solution (a,b) without and containing (c,d) 0.1 g/L and (e,f) 1.0 g/L 2BPB at 25 °C.

9

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Fig. 8 shows the effect of temperature rise (25 °C and 60 °C) on the corrosion rate and inhibition efficiency of 2BPB for (a) Cu-Ni 70/30 and (b) Cu-Ni 90/10 alloys. Expectedly, the corrosion rate increases with the rise in temperature because of the increase in the average kinetic energy of the corrosive species. The corrosion inhibition performance of 2BPB is found to slightly decline with the rise in temperature, particularly for Cu-Ni 70/30 alloy where it decreases from 64.81% at 25 °C to 59.23% at 60 °C. For Cu-Ni 90/10 alloy, inhibition efficiency depreciates from 52.50% at 25 °C to 38.66% at 60 °C. This observation can be linked to the detachment of some of the adsorbed inhibitor film from the alloys surfaces [34,35]. The slight decrease in the inhibition efficiency with rise in temperature may infer mixed adsorption mode (i.e. both chemical and physical adsorption) but with physical adsorption being the more prevalent mechanism. To further probe the role of temperature on the corrosion behavior of the alloys in the studied corrosive systems, the apparent activation energy (Ea) for the corrosion process and the heat of adsorption (Qads) of 2BPB on the alloys surfaces were calculated using Eqs. (18) & (19) [34,36], respectively.

Fig. 6. Variation of corrosion rate and inhibition efficiency with the concentration of 2BPB at 25 °C.

log

3.2. Weight loss measurements The weight loss technique was also used to study the corrosion and corrosion inhibition of the Cu-Ni alloys in 1 mol/dm3 HCl solution by 2BPB. Fig. 6 shows the variation of corrosion rate (v) and inhibition efficiency (η) with the concentration of 2BPB at 25 °C after 24 h of immersion. While v varies inversely with the concentration of 2BPB, η is in direct proportion. The least v and the highest η are achieved with the highest studied concentration of 2BPB (1.0 g/L). The presence of this concentration in the corrodent decreased v from 0.099 mm/yr (Cu-Ni 70/30) and 0.117 mm/yr (Cu-Ni 90/10) to 0.033 mm/yr and 0.055 mm/yr, respectively. This corresponds to η of 64.8% and 52.5%, respectively. The implication is that, more 2BPB molecules were available for adsorption when 1.0 g/L was added than when the lower concentrations were added. This resulted in larger surface coverage by the adsorbed inhibitor films and better protection against corrosive attack. It is a fact that the corrosion of a metal and the performance of an inhibitor can be affected by immersion duration and temperature [13,30]. Weight loss experiments were undertaken to investigate the influence of these factors on the corrosion behavior of the alloys and the corrosion inhibition performance of 2BPB. For these sets of experiments, 1.0 g/L 2BPB was used. Fig. 7 illustrates the effect of immersion time on the corrosion rate and inhibition efficiency of 2BPB for (a) CuNi 70/30 and (b) Cu-Ni 90/10 alloys. The immersion duration of 72 h was chosen to mimic a typical acid cleaning duration [6]. From the graphs, it is clear that the corrosion rate of the alloys in both the uninhibited and inhibited systems and the inhibition efficiency of 2BPB increase with increase in immersion time. The inhibition efficiency of 2BPB increased from 52.50% to 59.49% in the case of Cu-Ni 90/10 and from 64.81% to 71.58% in the case of Cu-Ni 70/30. The increase in the inhibition efficiency with increasing immersion time is attributed to an increase in the number of inhibitor molecules adsorbed on the alloys surfaces and the stability of the adsorbed protective film [31]. It is important that the corrosion rate of the MSF metal components fall within the acceptable limit during acid cleaning process. For Cu-Ni alloys, the maximum acceptable corrosion rate in HCl solution under normal atmospheric and aerated conditions is ≥0.1 mm/yr [32,33]. After 72 h immersion, the corrosion rate of Cu-Ni 70/30 and 90/10 alloys in the studied system is 0.06 ± 0.000 mm/yr and 0.16 ± 0.003 mm/yr, respectively. This portrays 2BPB as a promising active for the formulation of acid cleaning inhibitor cocktail.

v2 Ea 1 = v1 2.303R T1

Qads = 2.303R log

1 T2 2

1

(18)

log 2

1

1

× 1

T1 T2 kJ mol T2 T1

1

(19) where v1 and v2 are the corrosion rates (derived from the weight loss technique) at T1 (25 °C) and T2 (60 °C), respectively; θ1 and θ2 are the surface coverage (θ = η/100) at T1 and T2, respectively and R is the molar gas constant. The calculated value of Ea for the corrosion of Cu-Ni 70/30 alloy in the uninhibited and inhibited HCl solution is 18.32 kJ/ mol and 14.09 kJ/mol, respectively. Similarly, the calculated value of Ea for Cu-Ni 90/10 alloy in the uninhibited and inhibited acid solution is 52.03 kJ/mol and 58.30 kJ/mol, respectively. The values are positive and indicate endothermic corrosion process [30,34,36]. For Cu-Ni 70/ 30 alloy, the Ea value for the 2BPB-inhibited system is slightly smaller than that of the uninhibited acid solution whereas the reverse is observed in the case of Cu-Ni 90/10 alloy. In the literature, Ea (blank) < Ea (inhibited) and Ea (blank) > Ea (inhibited) are interpreted to indicate chemical and physical adsorption mechanisms, respectively [34–36]. However, it is argued that the two mechanisms (i.e. physisorption and chemisorption) are not completely separate and that one is a preceding step to another [37,38]. Because the calculated value of Qads for the corrosion inhibition process (i.e. −6.01 kJ/mol and −13.38 kJ/mol for Cu-Ni 70/30 and 90/10 alloys, respectively) is negative, we conclude that the dominant adsorption mechanism is physical adsorption [34,36]. 3.3. Effect of iodide ions on the performance of 2BPB During corrosion inhibitor formulation, intensifiers are added to enhance the corrosion inhibition performance of the active organic molecule [39]. This is because, a single organic molecule cannot provide the needed corrosion protection [39]. One of the frequently used intensifiers is potassium iodide in the concentration range of 0.1 to 2.0 wt% [40–43]. Experiments were conducted to study the effect of addition of 1 mM KI on the corrosion inhibition performance of 2BPB and the results obtained are summarized in Table 4. The results in the table reveal that the addition of KI enhanced the corrosion inhibition performance of 2BPB. For instance, the corrosion rate of Cu-Ni 70/30 and 90/10 alloys in 1.0 g/L 2BPB-inhibited HCl solution after 72 h of immersion at 25 °C is 0.06 + 0.00 mm/yr and 0.16 + 0.00 mm/yr, and

10

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Fig. 7. Effect of immersion time on the corrosion rate and inhibition efficiency of 2BPB for (a) Cu-Ni 70/30 and (b) Cu-Ni 90/10 alloys.

corresponded to inhibition efficiency of 71.58% and 59.49%, respectively. Addition of 1 mM KI to the system further decreased the corrosion rate to 0.01 + 0.00 mm/yr and 0.07 + 0.00 mm/yr for Cu-Ni 70/30 and 90/10 alloys and upgraded the inhibition efficiency to 89.39% and 70.22%, respectively. Similar beneficial effect is observed at 60 °C. This enhanced inhibition is due to the co-adsorption of iodide ions and 2BPB on the alloys surfaces [41–43]. Meanwhile, co-adsorption could be competitive (antagonistic) or cooperative (synergistic) and can be differentiated by computing the synergism parameter (Sθ) using the following equation [44]:

S =

1

(

1

1

+

2 1 1+2

1 2)

(20)

where θ1 is degree of surface coverage of 2BPB, θ2 is the degree of surface coverage of iodide ions and 11+ 2 is the degree of surface coverage of 2BPB + KI mixture. In general, Sθ value greater than unity is indicative of cooperative adsorption while Sθ value less than unity is reflective of competitive adsorption [40–42,44]. In this study (Table 4), the Sθ values are mostly greater than or equals to unity and point to a synergistic effect where the 2BPB molecules adsorb on a layer of iodide ions on the alloys surfaces.

11

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Fig. 8. Effect of temperature on the corrosion rate and inhibition efficiency of 2BPB for (a) Cu-Ni 70/30 and (b) Cu-Ni 90/10 alloys.

3.4. Surface characterization

inferring that, the adsorption of the inhibitor molecules protected the alloys surfaces against corrosive attack [13]. By comparing Fig. 9(c) with Fig. 9(d), it is seen that, the surface in Fig. 9(c) is smoother than the one in Fig. 9(d). In addition, the film on the surface of Fig. 9(c) is more rigid than the one on the surface of Fig. 9(d). This observation confirms the experimental findings (Tables 2 & 3, Fig. 6) that 2BPB is a better inhibitor for Cu-Ni 70/30 alloy than Cu-Ni 90/10 alloy. Obviously, the 2BPB + KI mixture offered superior corrosion protection to the alloys surfaces than 2BPB alone. Clearly, the surfaces in Fig. 9(e & f) are smoother and well-protected compared to the surfaces in Fig. 9(c & d). The co-adsorption of 2BPB and KI ions on the alloys surfaces provided formidable barrier against corrosive ions present in the aggressive medium [13].

3.4.1. SEM studies Fig. 9 presents the SEM images of Cu-Ni alloy samples after exposure to 1 mol/dm3 HCl solution for 72 h at room temperature. After 72 h of immersion in the uninhibited acid solution, the Cu-Ni 70/30 and 90/10 alloys samples exhibit the rough morphology shown in Fig. 9(a) and (b), respectively, which is typical for a corroded surface. Some porous and loosely adherent films are observed on the surfaces in Fig. 9(a & b), which have been identified as an unprotective mixture of Cu/Ni oxides and hydroxides [1]. The SEM micrographs of the 2BPB-inhibited samples (Fig. 9(c & d)) show a less coarse morphology and a more compact surface film,

12

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Table 4 Calculated values of corrosion rate (v), inhibition efficiency (η), and synergism parameter (Sθ) for Cu-Ni 70/30 and Cu-Ni 90/10 alloys immersed in 1 mol/dm3 HCl solution containing 1.0 g/L 2BPB alone and in combination with 1 mM KI. System/conc.

Tempt. (°C)

Cu-Ni 70/30

Cu-Ni 90/10

24 h

Blank KI 2BPB 2BPB + KI Blank KI 2BPB 2BPB + KI

25

60

72 h

24 h

72 h

v (mm/yr)

ηWL (%)

Sθ

v (mm/yr)

ηWL (%)

Sθ

v (mm/yr)

ηWL (%)

Sθ

v (mm/yr)

ηWL (%)

Sθ

0.10 0.06 0.03 0.02 0.21 0.59 0.37 0.18

– 35.71 64.81 75.00 – 34.81 59.23 69.62

– – – 1.0 – – – 1.3

0.20 0.10 0.06 0.01 – – – –

51.09 71.58 89.39 – – – –

– – – 0.9 – – – –

0.12 0.11 0.06 0.03 1.06 0.85 0.65 0.44

– 8.20 52.50 69.64 – 19.96 38.66 48.54

– – – 1.4 – – – 0.8

0.40 0.23 0.16 0.07 – – – –

– 43.22 59.49 70.22 – – – –

– – – 1.0 – – – –

± ± ± ± ± ± ± ±

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

± ± ± ±

0.00 0.00 0.00 0.00

3.4.2. FTIR studies Interaction of organic molecule with a metal surface can result in a shift in the infrared absorption wavenumber and a decrease in the absorption intensity [13,30]. To confirm the adsorption of 2BPB molecules on the alloys surfaces after exposure to 2BPB-inhibited acid solution for 72 h and to identify the interaction centers between the inhibitor and the alloys surfaces, the FTIR spectra of the adsorbed films extracted from the alloys surfaces were compared with the FTIR spectrum of the pure 2BPB (Fig. 10). The FTIR spectrum of the pure 2BPB, which is in conformity with previously reported spectra [13,14] exhibits multiple absorption bands, particularly at the low wavenumber region. For clarity, the low wavenumber region was zoomed (see Fig. 10). In the zoomed spectrum, the weak peaks at 1559.6 cm−1 and 1125.7 cm−1 are assigned to N−H bending and C−N stretching, respectively [13,45]. The peak at 1463.2 cm−1 is linked to the vibration of the unsaturated C]C bonds in the rings [13,45]. The sharp bands at 1019.7 cm−1 and 747.8 cm−1 emanated from the C−H stretching of the imidazoline ring and the ring puckering, respectively [13,45]. The C−Br stretching band can as well be observed at 692.8 cm−1 [13,45]. The weak absorption band observed at 3048.4 cm−1 in the pure 2BPB spectrum is typical of N−H stretching [13,45]. The absorption bands in the spectra of the extracted films are identical to those in the pure 2BPB spectrum. This confirms the claim that, the observed corrosion protection of the alloys surfaces (Fig. 9) is due to the adsorption of 2BPB molecules. However, the position of the bands and the absorption intensity are altered in the extracted film spectra relative to that of the pure 2BPB spectrum. For instance, the N−H stretching and bending bands are shifted to around 2910 cm−1 and 1610 cm−1, respectively in the extracted film spectra. The intensity of N−H, C−N, C]C, and the ring puckering bands diminish considerably in the extracted films spectra. This shows that, the inhibitor molecules interacted with the alloys surfaces and that the N-heteroatom, the imidazoline ring, and the benzyl ring serve as the interaction centers. By comparing the spectrum of the film from Cu-Ni 70/30 alloy surface to the one from the Cu-Ni 90/10 alloy surface, it is observed that, the N−H bending peak at 1614.6 cm−1 in the Cu-Ni 90/10 alloy film spectrum is near absent in that of Cu-Ni 70/30 alloy surface film. This suggests a stronger interaction between 2BPB molecules with Cu-Ni 70/30 alloy surface than with Cu-Ni 90/10 alloy surface, which is in agreement with the other experimental results. The weak absorption band also observed at around 3300 cm−1 in the extracted films spectra is associated with the adsorbed water molecules held within the matrix of the corrosion layer [13].

± ± ± ± ± ± ± ±

0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

± ± ± ±

0.00 0.00 0.00 0.00

solution. These are (i) the N-heteroatom, (ii) the imidazoline ring, and (iii) the benzyl ring. There are therefore three possibilities by which 2BPB molecule can interact with the alloys surfaces. Firstly, the Nheteroatoms in 2BPB can be protonated in the acid solution and adsorb on the alloys surfaces already saturated with halide ions (chloride and/ or iodide ions) from the acid [14] through columbic attraction (physical adsorption). Secondly, the lone electron pairs on un-protonated N atoms can form coordinate bond with the empty 3d orbitals of Cu and Ni (chemical interaction) [8,14]. Thirdly, the inhibitor molecules can adsorb on the alloys surfaces by virtue of donor-acceptor interaction between the π-electrons from the benzyl and imidazoline rings and the empty d-orbital of Cu and Ni atoms [8,14]. Our results (the calculated values of Ea and Qads, the variation of inhibition efficiency with temperature) suggest that, although these three forms of interactions may have been in place, the physical interaction mechanism is the dominant mechanism. 4. Summary and conclusions The Cu-Ni alloys, particularly the Cu-Ni 70/30 and 90/10 alloys are used as brine heater and evaporator condenser because of their outstanding properties [1,2]. Specifically, they have good corrosion resistance property due to the formation of cuprous oxide layer [1,2]. However, the layer is broken and corrosion accelerated during acid cleaning process [3,6]. The corrosion inhibitor technology is a reliable and less expensive corrosion control technique [3,6,8]. In this investigation, the anticorrosion property of 2-(2-bromophenyl)-1-methyl1H-benzimidazole (2BPB) for Cu-Ni 70/30 and 90/10 alloys in 1 mol/ dm3 HCl solution is investigated at low and high temperatures using the weight loss, electrochemical (EIS, PDP, & CV), and surface characterization (SEM & FTIR) techniques. The effect of immersion time (up to 72 h) and addition of iodide ions on the corrosion inhibition performance of 2BPB have also been investigated. It is found that the performance of 2BPB as inhibitor depends on the concentration, immersion time, and temperature. At 25 °C, 1.0 g/L 2BPB is capable of protecting Cu-Ni 70/30 and 90/10 alloys by 88.9 ± 4.8% and 57.5 ± 1.3%, respectively. The performance of 2BPB improves with increase in immersion time and addition of iodide ions but slightly depreciates with rise in temperature. 2BPB behaves as a typical mixed type corrosion inhibitor inhibiting both the anodic and cathodic corrosion reactions and adsorbs on the alloys surfaces mostly through physical adsorption mechanism according to the calculated heat of adsorption parameter. FTIR results reveal that, the N-heteroatom, the imidazoline ring, and the benzyl ring in 2BPB molecule are the adsorption sites. The results obtained in this investigation portray 2BPB as a low toxic promising candidate for the formulation of acid corrosion inhibitor for Cu-Ni alloys.

3.5. Proposed mechanism of corrosion inhibition by 2BPB The FTIR studies (Fig. 10) reveal three obvious interaction centers for the adsorption of 2BPB molecule on Cu-Ni alloys in 1 mol/dm3 HCl 13

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

Fig. 9. SEM images for (a, c, & e) Cu-Ni 70/30 and (b, d, & f) Cu-Ni 90/20 alloys after immersion in 1 mol/dm3 HCl solution (a, b) without inhibitor, (c, d) containing 1.0 g/L 2BPB, (e, f) containing 1.0 g/L 2BPB + 1 mM KI for 72 h at 25 °C. 14

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al.

2BPB

Fig. 10. Comparative FTIR spectra for pure 2BPB and corrosion products formed on Cu-Ni alloys surfaces after 72 h immersion in 1 mol/dm3 HCl solution containing 1.0 g/L 2BPB.

CRediT authorship contribution statement

Declaration of competing interest

Ikenna B. Onyeachu: Conceptualization, Investigation, Methodology, Writing - review & editing. Moses M. Solomon: Conceptualization, Investigation, Methodology, Writing - original draft. Saviour A. Umoren: Conceptualization, Resources, Writing review & editing. Ime B. Obot: Conceptualization, Resources, Writing - review & editing. Ahmad A. Sorour: Conceptualization, Resources, Writing - review & editing.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. References [1] S.J. Yuan, S.O. Pehkonen, Surface characterization and corrosion behavior of 70/30 Cu-Ni alloy in pristine and sulfide-containing simulated seawater, Corros. Sci. 49 (2007) 1276–1304, https://doi.org/10.1016/j.corsci.2006.07.003.

15

Desalination xxx (xxxx) xxxx

I.B. Onyeachu, et al. [2] B.V. Appa Rao, K.C. Kumar, Effect of hydrodynamic conditions on corrosion inhibition of Cu-Ni (90/10) alloy in seawater and sulphide containing seawater using 1,2,3-benzotriazole, J. Mater. Sci. Technol. 30 (2014) 65–76, https://doi.org/10. 1016/j.jmst.2013.08.019. [3] M.S. Wiener, A. Eliezer, Materials and corrosion control in desalination plants, Mater. Perform. 51 (2012) 56–60. [4] J.S. Gill, A novel inhibitor for scale control in water desalination, Desalination 124 (1999) 43–50, https://doi.org/10.1016/S0011-9164(99)00087-9. [5] J. Zhao, M. Wang, H.M.S. Lababidi, H. Al-Adwani, K.K. Gleason, A review of heterogeneous nucleation of calcium carbonate and control strategies for scale formation in multi-stage flash (MSF) desalination plants, Desalination 442 (2018) 75–88, https://doi.org/10.1016/j.desal.2018.05.008. [6] A.U. Malik, S. Ahmad, I. Andijani, N. Asrar, Acid Cleaning of Some Desal Units at Al-Jubail Plant, Technical Report No. TR3804/APP95007, (February 1997). [7] A.A. Khadom, A.S. Yaro, A.A.H. Kadum, Corrosion inhibition by naphthylamine and phenylenediamine for the corrosion of copper-nickel alloy in hydrochloric acid, J. Taiwan Inst. Chem. Eng. 41 (2010) 122–125, https://doi.org/10.1016/j.jtice.2009. 08.001. [8] B.V. Appa Rao, K. Chaitanya Kumar, 5-(3-Aminophenyl)tetrazole – a new corrosion inhibitor for Cu–Ni (90/10) alloy in seawater and sulphide containing seawater, Arab. J. Chem. 10 (2017) S2245–S2259, https://doi.org/10.1016/j.arabjc.2013.07. 060. [9] L. Feng, S. Zhang, Y. Lu, B. Tan, S. Chen, L. Guo, Synergistic corrosion inhibition effect of thiazolyl-based ionic liquids between anions and cations for copper in HCl solution, Appl. Surf. Sci. 483 (2019) 901–911, https://doi.org/10.1016/j.apsusc. 2019.03.299. [10] J.M. Maciel, R.F.V.V. Jaimes, P. Corio, J.C. Rubim, P.L. Volpe, A.A. Neto, S.M.L. Agostinho, The characterisation of the protective film formed by benzotriazole on the 90/10 copper-nickel alloy surface in H2SO4 media, Corros. Sci. 50 (2008) 879–886, https://doi.org/10.1016/j.corsci.2007.10.011. [11] A.A. Khadom, A.S. Yaro, Mass transfer effect on corrosion inhibition process of copper-nickel alloy in hydrochloric acid by benzotriazole, J. Saudi Chem. Soc. 18 (2014) 214–219, https://doi.org/10.1016/j.jscs.2011.06.013. [12] 2-(2-Bromophenyl)-1-methyl-1H-benzimidazole, Material Safety Data Sheet, Sigma Aldrich, www.sigmaaldrich.com. (Retrieved 16/10/2019). [13] I.B. Obot, I.B. Onyeachu, S.A. Umoren, Alternative corrosion inhibitor formulation for carbon steel in CO2-saturated brine solution under high turbulent flow condition for use in oil and gas transportation pipelines, Corros. Sci. 159 (2019) 108140, , https://doi.org/10.1016/j.corsci.2019.108140. [14] I.B. Obot, A. Madhankumar, S.A. Umoren, Z.M. Gasem, Surface protection of mild steel using benzimidazole derivatives: experimental and theoretical approach, J. Adhes. Sci. Technol. 29 (2015) 2130–2152, https://doi.org/10.1080/01694243. 2015.1058544. [15] Y.H. Wang, J.B. He, Corrosion inhibition of copper by sodium phytate in NaOH solution: cyclic voltabsorptometry for in situ monitoring of soluble corrosion products, Electrochim. Acta 66 (2012) 45–51, https://doi.org/10.1016/j.electacta. 2012.01.063. [16] NACE TM0169/G31 − 12a, Standard Guide for Laboratory Immersion Corrosion Testing of Metals, NACE International/ASTM International, 2017, https://doi.org/ 10.1520/G0031-12A. [17] ASTM-G 01–03, ASTM Book of Standards, vol. 3.02, ASTM, West Conshohocken, 2003, https://doi.org/10.1520/G0001-03R11.2. [18] A. Frignani, V. Grassi, F. Zanotto, F. Zucchi, Inhibition of AZ31 Mg alloy corrosion by anionic surfactants, Corros. Sci. 63 (2012) 29–39, https://doi.org/10.1016/j. corsci.2012.05.012. [19] H. Gerengi, M. Mielniczek, G. Gece, M.M. Solomon, Experimental and quantum chemical evaluation of 8-hydroxyquinoline as a corrosion inhibitor for copper in 0.1 M HCl, Ind. Eng. Chem. Res. 55 (2016), https://doi.org/10.1021/acs.iecr. 6b02414. [20] C. Monticelli, A. Balbo, J. Esvan, C. Chiavari, C. Martini, F. Zanotto, L. Marvelli, L. Robbiola, Evaluation of 2-(salicylideneimino) thiophenol and other Schiff bases as bronze corrosion inhibitors by electrochemical techniques and surface analysis, Corros. Sci. 148 (2019) 144–158, https://doi.org/10.1016/j.corsci.2018.12.017. [21] S.A. Radford, G.J.W. Walsh, F.C. Smith, J.R. Tuck, C.D.S. Campbell, Developments in Marine Corrosion, Royal Society of Chemistry, Cambridge, 1998. [22] N.S. Meck, P. Crook, Electrochemical behavior of nickel alloys in HCl solutions containing oxidizing impurities, Paper No. 05334, NACE International, Houston, 2005. [23] R. Babić, M. Metikoš-Huković, M. Lončar, Impedance and photoelectrochemical study of surface layers on Cu and Cu-10Ni in acetate solution containing benzotriazole, Electrochim. Acta 44 (1999) 2413–2421, https://doi.org/10.1016/S00134686(98)00367-3.

[24] E.M. Sherif, S.M. Park, Inhibition of copper corrosion in acidic pickling solutions by N-phenyl-1,4-phenylenediamine, Electrochim. Acta 51 (2006) 4665–4673, https:// doi.org/10.1016/j.electacta.2006.01.007. [25] K.M. Ismail, Evaluation of cysteine as environmentally friendly corrosion inhibitor for copper in neutral and acidic chloride solutions, Electrochim. Acta 52 (2007) 7811–7819, https://doi.org/10.1016/j.electacta.2007.02.053. [26] K.M. Ismail, A.M. Fathi, W.A. Badawy, Effect of nickel content on the corrosion and passivation of copper-nickel alloys in sodium sulfate solutions, Corrosion 60 (2004) 795–803, https://doi.org/10.5006/1.3287859. [27] M. Carvalho, Corrosion of copper alloys in natural seawater – effects of hydrodynamics and pH, Anal. Chem. (2014) 258, https://doi.org/10.1109/TIM.2015. 2477160. [28] S.M. Abd El Haleem, E.E. Abd El Aal, Electrochemical behavior of nickel in HNO3 and the effect of chloride ions, J. Mater. Eng. Perform. 13 (2004) 784–792, https:// doi.org/10.1361/10599490421538. [29] S.L. Medway, C.A. Lucas, A. Kowal, R.J. Nichols, D. Johnson, In situ studies of the oxidation of nickel electrodes in alkaline solution, J. Electroanal. Chem. 587 (2006) 172–181, https://doi.org/10.1016/j.jelechem.2005.11.013. [30] M.M. Solomon, H. Gerengi, S.A. Umoren, Carboxymethyl cellulose/silver nanoparticles composite: synthesis, characterization and application as a benign corrosion inhibitor for St37 steel in 15% H2SO4 medium, ACS Appl. Mater. Interfaces 9 (2017), https://doi.org/10.1021/acsami.6b14153. [31] R. Aslam, M. Mobin, J. Aslam, H. Lgaz, I.-M. Chung, Inhibitory effect of sodium carboxymethylcellulose and synergistic biodegradable gemini surfactants as effective inhibitors for MS corrosion in 1 M HCl, J. Mater. Res. Technol. 8 (2019) 4521–4533, https://doi.org/10.1016/j.jmrt.2019.07.065. [32] M.G. Fontana, Corrosion Engineering, 3rd ed., Mc Graw Hill International, 1987. [33] H. Uhlig, R. Revie, Corrosion and Control, 3rd ed., John Wiley and Sons, 1985. [34] E.E. Oguzie, Corrosion inhibition of aluminium in acidic and alkaline media by Sansevieria trifasciata extract, Corros. Sci. 49 (2007) 1527–1539, https://doi.org/ 10.1016/j.corsci.2006.08.009. [35] M.M. Solomon, S.A. Umoren, I.I. Udosoro, A.P. Udoh, Inhibitive and adsorption behaviour of carboxymethyl cellulose on mild steel corrosion in sulphuric acid solution, Corros. Sci. 52 (2010) 1317–1325, https://doi.org/10.1016/j.corsci.2009. 11.041. [36] M.M. Solomon, H. Gerengi, T. Kaya, S.A. Umoren, Performance evaluation of a chitosan/silver nanoparticles composite on St37 steel corrosion in a 15% HCl solution, ACS Sustain. Chem. Eng. 5 (2017) 809–820, https://doi.org/10.1021/ acssuschemeng.6b02141. [37] M.M. Solomon, S.A. Umoren, Performance evaluation of poly (methacrylic acid) as corrosion inhibitor in the presence of iodide ions for mild steel in H2SO4 solution, J. Adhes. Sci. Technol. 29 (2015) 1060–1080, https://doi.org/10.1080/01694243. 2015.1017436. [38] K. Ramya, R. Mohan, K.K. Anupama, A. Joseph, Electrochemical and theoretical studies on the synergistic interaction and corrosion inhibition of alkyl benzimidazoles and thiosemicarbazide pair on mild steel in hydrochloric acid, Mater. Chem. Phys. 149 (2015) 632–647, https://doi.org/10.1016/j.matchemphys.2014.11.020. [39] M. Finšgar, J. Jackson, Application of corrosion inhibitors for steels in acidic media for the oil and gas industry: a review, Corros. Sci. 86 (2014) 17–41, https://doi.org/ 10.1016/j.corsci.2014.04.044. [40] S.A. Umoren, M.M. Solomon, Effect of halide ions on the corrosion inhibition efficiency of different organic species - a review, J. Ind. Eng. Chem. 21 (2015) 81–100, https://doi.org/10.1016/j.jiec.2014.09.033. [41] A.A. Farag, T.A. Ali, The enhancing of 2-pyrazinecarboxamide inhibition effect on the acid corrosion of carbon steel in presence of iodide ions, J. Ind. Eng. Chem. 21 (2015) 627–634, https://doi.org/10.1016/j.jiec.2014.03.030. [42] I.B. Obot, A. Madhankumar, Synergistic effect of iodide ion addition on the inhibition of mild steel corrosion in 1 M HCl by 3-amino-2-methylbenzylalcohol, Mater. Chem. Phys. 177 (2016) 266–275, https://doi.org/10.1016/j.matchemphys. 2016.04.027. [43] R. Solmaz, E. Altunbaş Şahin, A. Döner, G. Kardaş, The investigation of synergistic inhibition effect of rhodanine and iodide ion on the corrosion of copper in sulphuric acid solution, Corros. Sci. 53 (2011) 3231–3240, https://doi.org/10.1016/j.corsci. 2011.05.067. [44] M.M. Solomon, S.A. Umoren, I.B. Obot, A.A. Sorour, H. Gerengi, Exploration of dextran for application as corrosion inhibitor for steel in strong acid environment: effect of molecular weight, modification, and temperature on efficiency, ACS Appl. Mater. Interfaces (2018), https://doi.org/10.1021/acsami.8b09487. [45] T.S. Xavier, N. Rashid, I. Hubert Joe, Vibrational spectra and DFT study of anticancer active molecule 2-(4-Bromophenyl)-1H-benzimidazole by normal coordinate analysis, Spectrochim. Acta A Mol. Biomol. Spectrosc. 78 (2011) 319–326, https:// doi.org/10.1016/j.saa.2010.10.013.

16