Shuduan Deng, Xianghong Li, Guanben Du

1 Yunnan key laboratory of wood adhesives and glue products, Southwest Forestry University, Kunming 650224, China

2 College of Chemical Engineering, Southwest Forestry University, Kunming 650224, China

Keywords:Rice (Oryza sativa L.)Saline-alkaline stress Abscisic acid (ABA)OsABA8ox1-kd Endogenous ABA levels

ABSTRACT Starch is one of the richest natural polymers with low-cost, non-toxic and biodegradable, but is seldom directly used as corrosion inhibitor due to its poor inhibitive ability and low water solubility.To solve this problem, cassava starch-acryl amide graft copolymer (CS-AAGC) was prepared through grafting acryl amide (AA) with cassava starch (CS), and it was firstly examined as an efficient inhibitor for 1060 aluminum in 1.0 mol·L-1 H3PO4 media.The adsorption behavior of CS-AAGC and its electrochemical mechanism were investigated by weight loss and electrochemical methods.Additionally, the inhibited aluminum surface was fully characterized by a series of SEM, AFM, contact angle measurements and XPS.Results confirm that CS-AAGC performs better inhibitive ability than CS, AA or CS/AA mixture,and the maximum inhibition efficiency of 1.0 g·L-1 CS-AAGC is 90.6% at 20°C.CS-AAGC acts as a mixed-type inhibitor while mainly retards the anodic reaction.EIS has three time constants, and the polarization resistance is significantly increased in the presence of CS-AAGC.The micrograph of inhibited aluminum surface is of hydrophobic nature with low surface roughness and little corrosion degree.

1.Introduction

Aluminum and its alloys are the most widely used nonferrous metals in our daily life and industrial applications, and they exhibits highly corrosive resistance owing to Al2O3film formed on the surface.However,this oxide film can be substantially dissolved in acid solution,and then the exposed aluminum substrate is continuously corroded.To efficiently retard the corrosion of aluminum in acid solution,adding inhibitor to the corrosive medium is one of the most widely used methods [1].Polymer inhibitor is likely to adsorb more metal surface area comparing with low-molecular weight compound,and then exhibits better inhibitive performance[2].Some petroleum polymers of ethoxylated fatty acids[3,4],poly 3-(decyloxy sulfonic acid)aniline[5],polyvinyl pyrrolidone[6]and poly 3-(dodecyloxy sulfonic acid)aninline[7]were studied as efficient inhibitors for aluminum in HCl solution.Noticeably, these polymers are high cost, and most important of all, along with the shortage of oil resources, the industrial applications are limited.Entering into the 21st century, with the urgent need for sustainable development, natural polymer, as an eco-friendly renewable resource with low cost, and its application as corrosion inhibitor has been received more and wider attentions [8].Several natural polymers like pectin [9], iota-carrageenan [10], hydroxypropyl methylcellulose [11]and exudate gum [12]were reported as efficient inhibitors for aluminum in HCl solution.However, both petroleum and natural polymers were little studied as the corrosion inhibitors for aluminum in H3PO4solution.

Phosphoric acid (H3PO4) is widely used for acid cleaning and electropolishing of aluminum.Also, it is used in pickling delicate,costly components and precision items where rerusting after pickling has to be avoided [13].Although H3PO4is a medium-strong acid, it can severely corrode aluminum surface [13].Inorganic compounds of potassium chromate (K2CrO4) [14]and sodium molybdate (Na2MoO4) [15,16]had been tested as efficient inhibitors for aluminum corrosion in H3PO4.As early as in 1976,organic inhibitor of aniline (C6H5—NH2) was reported as a good inhibitor for aluminum in H3PO4solution [17].Since then, however, there was almost no published paper as organic inhibitor for aluminum in H3PO4media.Until 2009, N-heterocyclic compound of purine was reported to be a moderate inhibitor for aluminum in 1.0 mol·L-1H3PO4, but the mixture of purine/I-exhibited a better inhibitive performance [18].Recently, some plant inhibitors of Dendrocalamus brandisii leaves extract (DBLE) [19], Mentha pulegium leaves extract (MPLE) [20]and Coriandrum sativum L.seeds extract (CLE) [21]were tested, but still moderately inhibited aluminum corrosion in H3PO4solution.Obviously,it is rather difficult to seek the eco-friendly and efficient inhibitor for aluminum corrosion in H3PO4.Through many attempts in our laboratory,it is found that most low-molecular-weight organics and petroleum polymers as well as plant extracts could not be qualified for efficient inhibitors for aluminum corrosion in H3PO4.Thus, there may be a unconventional way to explore eco-friendly and efficient inhibitor for aluminum in H3PO4solution using the natural polymers.

Starch, one of the richest natural polymers, is non-toxic and biodegradable with low-cost.However,natural starch is easy to be bonded together,has the aging problems and microbial corruption.It is seldom directly used as corrosion inhibitor due to its poor inhibitive ability and low water solubility.In view of these limitations,our work team develop an useful chemical modification method of graft copolymerization on starch to prepare the starch graft copolymer that exhibits high efficiency in inhibiting aluminum corrosion in HCl and HNO3[22,23]solution.However, the inhibitive mechanism of starch graft copolymer is still uncertain,and the inhibition is also be decided by acid nature.H3PO4solution is more complex than HCl owing to a series of acid anions ofand.Most of efficient inhibitors for aluminum in HCl solution cannot still be qualified as the effective inhibitors in H3PO4media.

Herein,we firstly reported the inhibition effect of cassava starchacryl amide graft copolymer (CS-AAGC) on the corrosion of aluminum in H3PO4solution.The adsorption behavior of CS-AAGC on aluminum surface and its electrochemical mechanism were discussed according to weight loss and electrochemical methods,respectively.Moreover, the inhibited aluminum surface was fully characterized by scanning electron microscope(SEM),atomic force microscope (AFM), contact angle measurement and X-ray photoelectron spectroscopy (XPS).Lastly, the inhibitive mechanism of CS-AAGC is proposed.It is expected to provide a novel type of green efficient inhibitor for aluminum in H3PO4solution.

2.Experimental

2.1.Materials and reagents

1060 aluminum specimens were used for all experiments, and the main elements mass compositions (%) are 0.25% Si, 0.35% Fe,0.05% Cu, 0.03% Mn, 0.03% Mg, 0.05% Zn, 0.03% Ti, 0.05% V and 99.16%Al.Analytical reagents H3PO4(85%)and AA(C3H5NO)were purchased from Shanghai Chemical Reagent Company of China.CS is of industrial-grade that is composed of about 17% amylose and 83% amylopectin [24].

2.2.Preparation of CS-AAGC

CS-AAGC was prepared using the redox initiation system with the initiator of (NH4)2S2O8and NaHSO3[22,23], and the synthesis route is shown in Fig.1.5.0 g CS was mixed with 150 ml H2O was heated at 80°C for 0.5 h with continuous stirring and bubbling N2, and cooled to 55°C, then 0.3 g (NH4)2S2O8and 0.35 g NaHSO3were added and continually stirred for 10 min, finally 7.5 g AA were added.The reactant mixture was stirred for 3.5 h,then cooled to room temperature and poured to 350 ml absolute C2H5OH to obtain the white precipitate.Through filter, the crude product was obtained and then dried in a vacuum oven at 50°C for 12 h.It was refluxed in acetone for 12 h,then filtered to obtain the solid product of CS-AAGC, which was finally dried in vacuum oven at 50°C for 12 h, and stored in a desiccator.Raman spectra of CS and CS-AAAGC were fully characterized and discussed in our recent published paper [22].According to literature [24], the molecular weight range of CS-AAGC is 1.0×105-2.0×105g·mol-1for the amylose with n=990,and about 2.0×107g·mol-1for amylopectin with n=7200.

SEM micrographs of CS and CS-AAGC were characterized, and also shown in Fig.1.CS appears spherical or dumbbell-shaped particles, while CS-AAGC is like a massive floccule.

2.3.Weight loss and electrochemical measurements

Detailed procedures of weight loss and electrochemical measurements have been described several times in our earlier papers[25,26].The aluminum coupons of 5.0 cm×2.0 cm×0.20 cm were used in weight loss measurement,both corrosion rate(v)and inhibition efficiency (ηw) were calculated through the mean weight loss value of three parallel aluminum specimens [25,26].Electrochemical tests were carried out on PARSTAT 2273 advanced electrochemical system (Princeton Applied Research) and working electrode with exposed surface area was 1.0 cm×1.0 cm.The polarization curves were measured in the potential range from-250 to+250 mV versus OCP at a scan rate 0.5 mV·s-1,and inhibition efficiency(ηp)was calculated from corrosion current densities[25,26].Electrochemical impedance spectroscopy (EIS) was measured at stable OCP in the frequency range from 100 kHz to 10 mHz, and the signal amplitude was 10 mV root mean square.Each electrochemical experiment was done at least in triplicate to check the reproducibility.

Fig.1.Synthesis route of CS-AAGC including SEM of CS (a) and CS-AAGC (b).

2.4.Surface analysis

SEM, AFM, contact angle and XPS measurements were examined by FEI QUANTA 200 scanning electron microscope (America),SPA-400 SPM Unit atomic force microscope(Japan),OCA20 optical contact angle (Dataphysics company, Germany) and X-ray photoelectron spectrometer (PHI-5500 ESCA, USA), respectively.

3.Results and Discussion

3.1.Weight loss measurements of the corrosion inhibition of CS-AAGC

In uninhibited H3PO4solution (blank), v is as high as 2.07 g·m-2·h-1.In the presence of chosen compounds, v values are dropped to 1.40, 1.13 and 0.19 g·m-2·h-1at 1.0 g·L-1CS, AA and CS-AAGC, respectively.Fig.2 shows the inhibition efficiency(ηw) with different concentrations of CS-AAGC (c) in 1.0 mol·L-1H3PO4solution at 20°C using weight loss method with immersion time of 6 h.Clearly, ηwof either CS, AA or CS-AAGC prominently increases with the additive concentration from 0.1 to 0.5 g·L-1,and then slightly changes with increasing inhibitor concentration till to 1.0 g·L-1.Thus,the adsorption amount on aluminum surface gradually reaches stable state from 0.5 to 1.0 g·L-1.Also,ηwfollows the order: CS-AAGC ＞CS/AA mixture ＞AA ＞CS.With the additive concentration of 1.0 g·L-1, ηwis 32.5% for CS; 45.6% for AA; and 90.6% for CS-AAGC.Accordingly, CS behaves as a poor inhibitor,and AA a moderate inhibitor, while CS-AAGC acts as an efficient inhibitor for aluminum in H3PO4solution.When CS is mixed with 1.0 g·L-1AA,ηwincreases to about 15%,but reaches only 57.6%for the mixture of 1.0 g·L-1CS/1.0 g·L-1AA,which is much lower than that of CS-AAGC.Thus, it can be concluded that the graft copolymerization on CS is an effective chemical modification method to significantly enhance the inhibitive performance.By contrast, the inhibition efficiency cannot reach a pleasant value by simply mixing CS with AA.The reason can be presumed as follows:there is no synergistic inhibition between CS and AA, and so the inhibition performance is not so good for CS/AA mixture.But when AA is chemically grafted with CS to form CS-AAGC that easily combines the merits of CS and AA.

Fig.2.Inhibition efficiency(ηw)curves with different inhibitor concentration(c)in 1.0 mol·L-1 H3PO4 at 20°C using weight loss method with immersion time of 6 h.

Table 1 lists the comparison of inhibition efficiency of CS-AAGC with the literature data of other previous reported inhibitors for aluminum in H3PO4solution.As can be seen from Table 1, CSAAGC performs better inhibition than reported inhibitors except for K2CrO4and Na2MoO4.It should be noted that the toxic of K2CrO4limits its applications in industrial engineering.Though Na2MoO4is non-toxic, it is of high-cost, which also restricts its wide application.Clearly, other organics and plant leaves extracts cannot qualify for efficient inhibitors for aluminum corrosion in H3PO4solution.Through overall comprehensive consideration, it is only CS-AAGC that is qualified for the eco-friendly and efficient inhibitor with low-cost for aluminum corrosion in H3PO4solution.In our previous studies, CS-AAGC is a good corrosion inhibitor for aluminum in HCl [22]and HNO3[23].In summary, CS-AAGC can be deemed as a universal and efficient inhibitor for aluminum in various inorganic acids.

The adsorption of acid inhibitor on metal surface in aqueous water solution can be commonly discussed by adsorption isotherm.It should be noted the adsorption species in acid inhibition system involve inhibitor, water, H3O+, acid ion and etc.Thus, it is very difficult to determine the accurate surface coverage of inhibitor molecules on metal surface.Commonly, the surface coverage(θ) is approximately equal to inhibition efficiency (η).As recently discussed by Walczak et al.[27],the assumption of θ=η is not necessarily valid, and then accurate value of standard Gibbs adsorption energy is difficult to be determined.

In the present study, CS-AAGC exerts inhibitive ability via its adsorption on aluminum surface in H3PO4solution, and herein,the surface coverage is proportional to inhibition efficiency.Accordingly, θ can be determined from inhibition efficiency as following relationship:

where f is the proportional factor.

Langmuir adsorption isotherm is applied for simulating the adsorption process, which has the following equation [28]:

where c is the concentration of CS-AAGC (g·L-1), and K represents the adsorptive equilibrium constant (L·g-1).

Substituting Eq.(1) into Eq.(2) obtains the following equation:

The experimental data of c/ηw- c and the corresponding fitted straight line is shown in Fig.3, and its the linear correlation coefficient (R2) of 0.9989.Thus, the adsorption of CS-AAGC on aluminum surface in H3PO4follows Langmuir adsorption isotherm.In addition, f is 1.02 from the slope.With obtained f and intercept of c/ηw- c line, the calculated K is 12.05 L·g-1.

Fig.3.Langmuir isotherm adsorption mode of CS-AAGC on aluminum surface in 1.0 mol·L-1 H3PO4 solution at 20°C using weight loss method with immersion time of 6 h.

Fig.4.Potentiodynamic polarization curves for the corrosion of aluminum in 1.0 mol·L-1 H3PO4 solutions in the absence and presence of CS-AAGC at 20°C in the potential range from -250 to +250 mV versus EOCP at 0.5 mV·s-1.

3.2.Potentiodynamic polarization curves for aluminum in H3PO4 solution

Potentiodynamic polarization curves of aluminum in 1.0 mol·L-1H3PO4solutions without and with CS-AAGC at 20°C are depicted in Fig.4.After adding CS-AAGC to H3PO4medium,both cathodic and anodic branches move to lower current densities,and the corrosion potential moves to positive comparing with blank solution,and this trend is enhanced with the increase of CSAAGC concentration.This result clearly indicates that CS-AAGC can efficiently retard both cathodic and anodic reactions of aluminum in H3PO4solution, while mainly the anodic reaction.

Electrochemical corrosion parameters are summarized in Table 2.As compared with blank solution, corrosion potential(Ecorr)in inhibited H3PO4solution by CS-AAGC moves positive position,and the positive changing Ecorr(ΔEcorr)value is more enlarged with the increase of inhibitor concentration,which implies that CSAAGC mainly inhibits the anodic reaction.When the additive concentration of CS-AAGC is 1.0 g·L-1, ΔEcorrreaches the maximum value of 32 mV.Owing to that the maximum ΔEcorris much lower than 85 mV,CS-AAGC can still be arranged as a mixed-type inhibitor for aluminum in H3PO4solution [29-31].The addition of CSAAGC causes icorrdecrease significantly from 119.0 to 9.6 μA·cm-2,and so CS-AAGC efficiently retards aluminum corrosion in H3PO4solution.Upon changing from icorrto obtain ηp, ηpincreases with the additive CS-AAGC concentration, and attains as high as 91.9%with the addition of 1.0 g·L-1.Thus,CS-AAGC acts as a good inhibitor for aluminum in H3PO4solution.

Table 2 Potentiodynamic polarization parameters for the corrosion of aluminum in 1.0 mol·L-1 H3PO4 solution containing different concentrations of CS-AAGC at 20°C(immersion time is 2 h)

The presence of CS-AAGC does not change cathodic Tafel slope(bc), which means the cathodic hydrogen reaction mechanism remains unchangeable.In all cases,bcvalues of the cathodic polarization curves with are within 137-145 mV·dec-1,which indicates the cathodic reaction of aluminum in H3PO4solution(2H++2e-→H2↑)is actively controlled by charge transfer.On the other hand,bavalues of anthodic polarization curves are as high as 612-741 mV·dec-1, which may be partly attributed to concentration polarization for the diffusion of the formed corrosion products on anodic reaction active sites.With the addition of CS-AAGC, babecomes large to more extent, which can be explained on the adsorption inhibitor molecules on anodic reaction sites, namely,extra concentration polarization of diffusion of inhibitive film.

3.3.EIS of aluminum in uninhibited and inhibited 1.0 mol·L-1 H3PO4 solution

EIS of Nyquist curves,Bode modulus and Bode phase angle plots of aluminum electrode in 1.0 mol·L-1H3PO4solutions without and with CS-AAGC at 20°C are shown in Fig.5(a), (b) and (c), respectively.Irrespective of adding any concentrations of CS-AAGC to H3PO4, the shapes of Nyquist and Bode plots almost remain unchangeable throughout all tested inhibitor concentrations.Thus,there is almost no alter in the corrosion mechanism whether the inhibitor is added to the corrosive media.

As can be seen Fig.5(a), each Nyquist diagram can be divided into three sections: (i) the first capacitive loop at high frequencies(HF), (ii) a small inductive loop at medium frequencies (MF), and(iii) the second capacitive loop at low frequencies (LF).The first large capacitive loop at HF can be related to formed Al2O3film on aluminum substrate [32].Owing to ex situ examination of aluminum electrode,it is reasonable to deduce that aluminum surface was easily oxidized by O2in the air or water solution.Burstein and Cinderey [33]confirmed that it is rather difficult to produce an oxide free Al surface, even if the freshly produced surface is obtained and it is easily oxidized very fast by O2immediately.It should be noted that Al2O3can be dissolved substantially in H3PO4solution.Accordingly, the first capacitive loop at HF would be related to the dissolution process of Al/Al2O3/electrolyte interface [34].During this process, Al+was firstly formed at the inner Al/Al2O3interface, and it can further be oxidized to Al3+when it was migrated through Al2O3/electrolyte interface.With the addition of inhibitor, CS-AAGC molecules can adsorb on Al/Al2O3/solution interface, and then efficiently slow down the dissolution rate of Al/Al2O3interface in H3PO4solution,which enlarges the diameter of capacitive loop at HF.It should be noted that these capacitiveloops are not perfect semicircles, which confirms that there is frequency dispersion that originates from the roughness and inhomogeneous of electrode surface [35,36].The small inductive loop at MF may be related to the adsorption-desorption process of intermediates formed on aluminum surface [15].For the present studied inhibition system, it is assumed the intermediates on aluminum electrode surface contain[35], acid anions (,), CS-AAGC inhibitor molecules, or [CS-AAGC-Al+]complex.The second capacitive loop at LF may be associated with Al-dissolution [37].

Fig.5.EIS of the corrosion of aluminum in 1.0 mol·L-1 H3PO4 solutions without and with CS-AAGC at 20°C (solid lines show fitted results) and the equivalent circuit: (a)Nyquist plots; (b) Bode modulus; (c) Bode phase angle plots; (d) Equivalent circuit used to fit the EIS.

In Fig.5(b), the Bode modulus at lgf from -2 to 1.0 increases with the concentration of CS-AAGC, which confirms that the inhibitive action is enhanced at higher inhibitor concentration.Further inspection Bode phase curves in Fig.5(b)reveals that there are one phase peak at high frequency (lg f=2-3), one phase valley at MF(lg f=-0.6 - 0.2) and another small peak at LF (lg f=-2 - -1),which are assigned to respective capacitive loop at HF, inductive loop at MF and the second capacitive loop at LF.The phase peak angle is lower than 90o, which again confirms that there is frequency dispersion for electrode surface.In a word, there are three time constants in EIS for aluminum corrosion in H3PO4solution.The unique characteristic EIS of three time constants for aluminum in H3PO4solution had been published for several times [18-21].

The equivalent circuit of Rs(Q(Rt(C1(R1(LRL)(C2R2)))))as shown in Fig.5(d) is used to deal with EIS result [19].Rsis solution resistance, and Rtis charge transfer resistance.RLand L are inductive elements associated with the inductive loop at MF, and they are in parallel.Q is constant phase element, and is parallel with series resistors of Rt,R1,RL,R2and two double layer capacitances of C1and C2.R1, RLand R2are resistances that correspond to the first capacitive loop at HF,the inductive loop at MF and the second capacitive loop at LF,respectively[19].Accordingly,R1corresponds to the dissolution resistance of Al2O3, and R2indicates the dissolution resistance of Al.C1is the double layer capacitance of Al2O3/electrolyte,and C2the double layer capacitance of Al/corrosion product.

With obtained resistance values of Rt,R1,R2and R3,the polarization resistance (Rp) can be calculated through the following equation [19]:

Inhibition efficiency from EIS(ηR)is calculated by the following equation [38]:

where Rp(0)and Rp(inh)are polarization resistances in the absence and presence of inhibitor, respectively.

The constant phase element of Q has following definition [39]:

where Y0is a proportional factor,ω is the angular frequency.n is a deviation parameter (-1 ≤n ≤+1), which has the meaning of a phase shift.If n=0, Q represents a pure resistor, for n=-1 an inductor; and for n=+1 a pure capacitor [39].

Another important parameter of double electric capacitance(Cdl) with the following equation [39]:

where fmaxis the maximum frequency related with maximum impedance in Nyqusit gram.

EIS fitted parameters for the corrosion of aluminum in H3PO4media are listed in Table 3.The chi-squared (χ2) are rather lower,which indicates that the fitted data are acceptable with high accuracy [40,41].All Rsvalues are in the range between 0.9 to 1.5 Ω·cm2,and thus IR drop for the present electrolyte can be negligible.Generally, a higher resistance is always associated with a slower corrosion system.With the addition of CS-AAGC, all resistors of Rt,R1,R2,R3and Rpare increased,which indicates the corrosion of aluminum is dropped to more extent in inhibited H3PO4solution.The electrochemical parameters of Q and Cdlis dropped to more extent for the inhibited system, which can be attributed to CS-AAGC inhibitor molecules adsorb on aluminum/solution interface,and then leads to an increase in the thickness of the electrical double layer and the decrease in local dielectric constant[42].It is mentioned that n is lower than 1 irrespective of adding CS-AAGC, and so the frequency dispersion effect is existed on aluminum electrode surface.Another inductive parameter of L becomes larger with the increase of CS-AAGC concentration.In uninhibited H3PO4solution, the adsorbed intermediates were mainlyand acid ions of.When CSAAGC was added to H3PO4solution, the adsorbed intermediates were transformed to CS-AAGC inhibitor molecules.Consequently,it is reasonable to assume that large L value in inhibited solution can be related to the adsorption/desorption process of CS-AAGC molecules on aluminum surface.It should be noted that C2is significantly higher than C1, which indicates that the dissolution of Al in H3PO4solution is easier to be corroded than that of Al2O3.C1almost remains unchangeable, while decreases to more extent with 1.0 g·L-1CS-AAGC, which implies that CS-AAGC may adsorb on Al2O3surface.On the other hand,C2increases with the concentration of CS-AAGC,which may be caused by the amount of corrosion product formed on Al decreases with the addition of CS-AAGC.

Inspection of Table 3 reveals that ηRincreases with the CS-AAGC concentration from 0.1 to 1.0 g·L-1, and the maximum ηRis up to 88.9%at 1.0 g·L-1,which confirms that CS-AAGC is an efficient inhibitor for aluminum corrosion in H3PO4solution.All experimental results of weight loss, polarization curves and EIS show that CSAAGC can efficiently retard the corrosion of aluminum in H3PO4.

3.4.Surface analysis of aluminum surface

Aluminum surfaces are characterized by SEM,AFM and contact angle examinations, and the results are presented in Fig.6.As shown in Figs.6(a)-(c), the freshly abraded aluminum surface before immersion appeared rather smooth,but still some abrading scratches can be seen.For 3D-AFM high resolution graph of Fig.6(c), the average roughness (Ra), root-mean-square roughness(RMS) and maximum peak-to-valley (P-V) are 29.64 nm,236.34 nm and 30.0 nm.By careful inspection,small particles covered on the aluminum surface,which may be attributed to the oxidized particles by O2in air.As shown in Fig.6(d),the water contact angle of pre-treated aluminum was in the range of 83.0°-86.9°, so the whole aluminum surface can be soaked with aqueous water solution.As aluminum was totally immersed in 1.0 mol·L-1H3PO4solution for 6 h,its surface was severely corroded and some black holes were immerged as shown in Figs.6(e)-(g), and turned to be rather rougher with Ra=83.61 nm, RMS=749.9 nm and PV=106.6 nm for 3D-AFM graph as shown in Fig.6(g).The contact angle of corroded aluminum surface was decreased to 67.8°-68.9°as shown in Fig.6(f),and thus the wettability was enhanced for the corroded surface.After adding 1.0 g·L-1CS-AAGC to 1.0 mol·L-1H3PO4media, as shown in Fig.6(i), aluminum surface had almost no corrosion products, and its original abraded scratches can be observed.For more high resolution SEM graph of Fig.6(j), there were still no evident corrosive holes.In 3D-AFM of Fig.6(k), the inhibited aluminum surface displayed smooth except for a little small round shallow pit, and Raof 27.56 nm, RMS of 188.2 nm and P-V value of 32.81 nm.As can be seen from Fig.6(l), the contact angle of the inhibited aluminum surface by CS-AAGC was turned into the hydrophobic nature with an obtuse angle of 115.6o,which can efficiently shield aluminum surface corrosion from H3PO4solution.

Fig.6 (continued)

ηR/%-69.3 83.7 88.9 2 Rp/Ω·cm 81.4 264.9 498.8 731.4 Cdl/μF·cm-2 109 14 108 χ2 5.6×10-4 2.5×10-3 1.5×10-2 1.2×10-2 2 R2/Ω·cm 23.4±0.5 66.5±0.6 146.7±0.8 155.8±1.0 EIS parameters for the corrosion of aluminum in 1.0 mol·L-1 H3PO4 solution containing different concentrations of CS-AAGC at 20°C (immersion time is 2 h)C2/mF·cm-2 495±4 511±5 519±7 588±7 2 L/H·cm2RL/Ω·cm 5.7±0.4 15.5±1.3 30.1±1.5 53.4±2.2 2±1 2±1 15±2 14±2 2 R1/Ω·cm 22.6±0.4 46.2±0.6 112.4±0.8 131.1±0.9 C1/μF·cm-2 16±2 25±4 15±6 9±5 2 Rt/Ω·cm 29.7±0.8 136.7±1.2 200.6±1.3 391.1±1.3 n 0.9634±0.02 0.9484±0.03 0.9998±0.02 0.7647±0.02 Q/μF·cm-2 123±8 19±5 11±5 11±4 2 3.9±0.2 3.4±0.3 4.5±0.3 4.9±0.3 Table 3 c/g·L-1Rs/Ω·cm 00.1 0.5 1.0

Fig.6.SEM,3D-AFM and contact angle images of aluminum surfaces:(a,b,c,d)before immersion;(e,f,g,h)after 6 h of immersion at 20°C in 1.0 mol·L-1 H3PO4 solution;(i,j,k, l) after 6 h of immersion at 20°C in 1.0 g·L-1 CS- AAGC +1.0 mol·L-1 H3PO4 solution.

The inhibited aluminum surface after immersing in 1.0 mol·L-1H3PO4solution containing 1.0 g·L-1CS-AAGC for 6 h at 20°C was composed of corrosion product and adsorbed inhibitor, and its chemical composition was analyzed by XPS that is shown in Fig.7.XPS scanning survey result of Fig.7(a) indicates that Al, P, C, N and O elements exist in inhibited aluminum surface.To further analyze the corresponding compounds and functional groups, and the high resolution fitted curves of C1s,N1s, P2p, O1s and Al2p elements are also summarized in Figs.7(b)-(f).As shown in Fig.7(b), C1s peak is decomposed of five fitted peaks.Binding energy (BE) at 284.8 eV is attributed to C—H or C—C group [43]; 285.3 eV to C—N; 286.5 eV to C—O; 288.3 eV to N—C=O; and 289.2 eV to C=O [44].Two independent fitted peaks of N1s in Fig.7(c) with BE at 400.3 eV and 402.2 eV are related to C—N and N—H groups, respectively.Thus, these fitted functional groups provide the direct evidence of the adsorption of CS-AAGC onto aluminum surface.

Fig.7.XPS analysis for aluminum surface exposed to 1.0 mol·L-1 H3PO4 solution with 1.0 g·L-1 CS-AAGC solution for 6 h at 20°C:(a)survey XPS;(b)C1s;(c)N1s;(d)P2p;(e)O1s; (f) Al2p.

Considering the acid medium of H3PO4, two independent BE peaks at 134.4 and 135.2 eV for P2p as shown in Fig.7(d)correspond toand H2PO4-,respectively.Accordingly,the main corrosion products of aluminum in H3PO4would be AlPO4and Al(H2PO4)3.Further analyzing the XPS fitted result of O1s presented in Fig.7(e), the main corrosion products are composed of Al2O3(530.6 eV), AlPO4(531.8 eV), Al(OH)3/AlOOH (532.6 eV) and aluminum water salts (533.7 eV).XPS curve of Al2p is fitted with two peaks in Fig.7(f), and BE at 75.4 eV is associated with Al2O3[45].It should be noted that BE at 75.7 eV may also be assigned to Al—O—CO— or Al—O—N— [46], which inflects that the coordinated bonds between CS-AAGC and aluminum surface.

3.5.Corrosion and inhibition mechanism

When aluminum surface was immersed in H3PO4solution, it can be dissolved owing to the aggressive attack of H+:

If the covered Al2O3layer is dissolved to the solution, and then the naked Al surface is exposed to H3PO4solution.In the experiment,continuous gas bubbles(H2)can be observed were liberated from aluminum surface, and so the aluminum corrosion in H3PO4solution can be arranged as the hydrogen evolution corrosion (Al+3H+→Al3++3/2 H2).For one hand, the cathodic reaction is the discharge of proton (H++e-→1/2H2↑), and may contain the following steps [22]:

On the other hand, the anodic dissolution taking place on aluminum surface is the reaction of Al →Al3++3e-, and the reaction mechanism is through following reactions [13]:

When CS is added to H3PO3solution,the corrosion of aluminum surface in H3PO3solution is slightly retarded.On the other hand,the graft polymer of CS-AAGC exhibits efficient inhibitive performance.Accordingly, the vinyl monomer of AA can be deemed as the active site, and it can strongly adsorb on metal surface, and then the molecular skeleton of CS in CS-AAGC acts as a barrier which shields metal surface from aggressive acid solution.

The adsorption of CS-AAGC on aluminum surface can be further discussed as follows:CS-AAGC contains many N and O atoms, and so CS-AAGC can be protonated by H+in the acid solution:

In a word, CS-AAGC exists in the forms of CS-AAGC and CSAAGCHxx+in acid media,and three main adsorption modes are presented as follows:

(i) During the corrosion process of aluminum in H3PO4solution,the acid anion ions ofcan adsorb on aluminum surface.Owing to the electrostatic attraction force,interact with these anion ions, and then physical adsorb on aluminum surface.

According to the CS-AAGC corrosion inhibition mechanism, it can be deduced that the graft copolymer through natural polymer(pectin, Gum Arabic, chitosan, cellulose, etc.) chemically modified by grafting monomers (sodium allylsulfonate, acryl amide, acrylic acid, etc.) can be deemed as potential efficient inhibitors for aluminum in H3PO4media.

4.Conclusions

(1) The grafting polymer of CS-AAGC acts as an eco-friendly effective inhibitor for 1060 aluminum corrosion in 1.0 mol·L-1H3PO4,and the maximumηwis higher than 90%.Inhibition efficiency follows the order:CS-AAGC ＞CS/AA mixture ≈AA ＞CS,and increases with the additive concentration.The adsorption ofCS-AAGConaluminumsurfacein1.0 mol·L-1H3PO4solution obeys Langmuir adsorption isotherm.

(2) The presence of CS-AAGC shifts the corrosion potential to positive, and can be arranged as a mixed-type inhibitor while mainly retards the anodic reaction.Nyquist consists of a large capacitive loop at HF followed by a small inductive one at MF and the second capacitive loop at LF.EIS has three time constants, and the polarization resistance strengthens with the concentration of CS-AAGC.

(3) SEM and AFM confirm the corrosion degree and surface roughness of aluminum surface are drastically decreased after adding CS-AAGC to the media.Contact angle image suggests that the inhibited aluminum surface is of hydrophobic nature.XPS proves that inhibited aluminum surface not only contains the corrosion products of but also the adsorptive inhibitor of CS-AAGC.

Declaration of Competing Interest

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.

Acknowledgements

Funding support from the National Natural Science Foundation of China(51561027),Training Programs of Young and Middle Aged Academic and Technological Leaders in Yunnan Province (2015HB049,2017HB030) and Special Project of ‘‘Top Young Talents” of Yunnan Ten Thousand Talents Plan(51900109)are acknowledged.