Theoretical and experimental studies on the corrosion inhibition potentials of some purines for aluminum in 0.1 M HCl

Abstract

Experimental aspect of the corrosion inhibition potential of adenine (AD), guanine (GU) and, hypoxanthine (HYP) was carried out using weight loss, potentiodynamic polarization and electrochemical impedance spectroscopy (EIS) methods while the theoretical aspect of the work was carried out by calculations of semi-empirical parameters (for AM1, MNDO, CNDO, PM3 and RM1 Hamiltonians), Fukui functions and inhibitor–metal interaction energies. Results obtained from the experimental studies were in good agreement and indicated that adenine (AD), guanine (GU) and hypoxanthine (HYP) are good adsorption inhibitors for the corrosion of aluminum in solutions of HCl. Data obtained from electrochemical experiment revealed that the studied purines functioned by adsorption on the aluminum/HCl interface and inhibited the cathodic half reaction to a greater extent and anodic half reaction to a lesser extent. The adsorption of the purines on the metal surface was found to be exothermic and spontaneous. Deviation of the adsorption characteristics of the studied purines from the Langmuir adsorption model was compensated by the fitness of Flory Huggins and El Awardy et al. adsorption models. Quantum chemical studies revealed that the experimental inhibition efficiencies of the studied purines are functions of some quantum chemical parameters including total energy of the molecules (TE), energy gap (E_L–H), electronic energy of the molecule (EE), dipole moment and core–core repulsion energy (CCR). Fukui functions analysis through DFT and MP2 theories indicated slight complications and unphysical results. However, results obtained from calculated Huckel charges, molecular orbital and interaction energies, the adsorption of the inhibitors proceeded through the imine nitrogen (N5) in GU, emanine nitrogen (N7) in AD and the pyridine nitrogen (N5) in HPY.

Introduction

Industrial revolution that is ever expanding within different parts of the world has several advantages and disadvantages in the quality of environment. Most industries utilize metals or their ores (such as mild steel, aluminum, zinc, and copper) in the fabrication of their installations. In most cases, these metals are exposed to aggressive medium/media and are prone to corrosion [1]. Corrosion is an electrochemical process that gradually returns the metal to its natural state in the environment. Corrosion in industries is often activated by processes such as acid wash, etching, prickling and others. Aluminum owes its widespread use after steel, to its excellent corrosion resistance to the air formed film strongly bonded to its surface. This film is relatively stable in aqueous solutions over a pH range of 4–8.5 [2]. In such solutions the surface film is insoluble but may be locally attacked by aggressive anions, primarily chlorides. The effect of Cl⁻ ions (which can be generated by hydrolysis of HCl) on the corrosion of aluminum and its alloys has been the subject of several studies [3–5]. The cost of replacing metals due to corrosion is often exorbitant and economically unbearable. Therefore, industries have adopted several options to control corrosion of metals including anodic/cathodic protection, painting, electroplating and galvanizing. However, the use of corrosion inhibitors has proven to be one of the most effective methods.

Inhibitors are compounds that retard the rate of corrosion of metals by been absorbed on the surface of the metal either through the transfer of charge from charge inhibitor molecule to charged metal surface (physical adsorption) or by electron transfer from the inhibitor’s molecule to the vacant d-orbital of the metal(chemical adsorption) [6]. Numerous studies have been carried out on the corrosion of metals in different environments and most of the well-known and suitable inhibitors are heterocyclic compounds [7–10]. For these compounds, their adsorption on the metal surface is the initial step of inhibition [11,12]. The adsorption of inhibitor is linked to the presence of heteroatoms (such as N, O, P, and S) and long carbon chain length as well as triple bond or aromatic ring in their molecular structure [13]. Generally, a strong coordination bond leads to higher inhibition efficiency. The corrosion inhibition potentials of some purines and their derivatives have been reported by several researchers [14–19].

Although quantum chemical studies limits the corrosion inhibition efficiency with molecular orbital energy levels of some organic compounds, semi-empirical method emphasizes the approaches that are involved in the selection of inhibitor by correlating the experimental data with quantum chemical properties such as energy of the highest molecular orbital (E_HOMO), the energy of the lowest unoccupied molecular orbital (E_LUMO), total negative charge (TNC), electronic energy (EE), binding energy (E_b), core–core repulsion energy (CCR), dipole moment and other parameters [20,21]. Also, the use of Fukui functions, calculated through Milliken, Lowdin or Hierfield charges have proven to be very useful in predicting the sites for electrophilic and nucleophilic attacks.

The present study is aimed at investigating the inhibitory and adsorption properties of some purines, namely (AD), guanine (GU) and hypoxanthine (HYP) for the corrosion of aluminum in HCl using gravimetric, electrochemical and quantum chemical methods.

Experimental

Material

Aluminum sheet (AA 1060 type) and purity 98.5% was used in this study. Acid solution of 0.1 M HCl was prepared by diluting analytical grade with distilled water. Various concentrations (ranging from 2.0 × 10⁻³ to 10.0 × 10⁻³ M) of the inhibitors were also prepared in the acid media. All reagents were obtained from Zayo-Sigma Chemicals. Fig. 1 shows chemical structures of adenine (AD), guanine (GU) and hypoxanthine (HYP).

Fig. 1.

Molecular structures of (a) adenine (b) guanine and (c) hypoxanthine.

Experimental procedure

Weight loss measurements

Aluminum coupons of dimension 5.0 × 4.0 × 0.15 cm were cut and wet-abraded with silicon carbide abrasive paper (from grade #1000 to #1200), rinsed with distilled water and in acetone before they were dried in the air. The pre-cleaned and weighed coupons were suspended in beakers containing the test solutions using glass hooks and rods. Tests were conducted under total immersion conditions in 150 mL of the aerated and unstirred test solutions. Immersion time was varied from 1 to 5 days (120 h) in 0.1 M HCl. The coupons were retrieved from test solutions after every 24 h, appropriately cleaned, dried and re-weighed. The weight loss was taken to be the difference between the weight of the coupons at a given time and its initial weight. The effect of temperature on Al corrosion and corrosion inhibition was investigated by repeating the experiments at 303 and 333 K respectively. All tests were run in duplicate and the data obtained showed good reproducibility.

Electrochemical measurements

Metal samples for electrochemical experiments were machined into test electrodes of dimension 1.0 × 1.0 cm² and sealed with epoxy resin in such a way that only one square surface area (1 cm²) was left uncovered. The exposed surface was cleaned using the procedure described above. Electrochemical tests were conducted in a Model K0047 corrosion cell using a VERSASTAT 400 complete DC voltammetry and corrosion system, with V3 Studio software. A graphite rod was used as a counter electrode and a saturated calomel electrode (SCE) as a reference electrode. The latter was connected via a Luggin capillary. Measurements were performed in aerated and unstirred solutions at the end of 1 h of immersion at 303 K. Impedance measurements were made at corrosion potentials (E_corr) over a frequency range of 100 kHz–10 mHz, with a signal amplitude perturbation of 5 mV. Potentiodynamic polarization studies were carried out in the potential ranging from −1000 to 2000 mV versus corrosion potential at a scan rate of 0.33 mV/s. Each test was run in triplicate [22].

Quantum chemical calculations

Full geometric optimization of each of the studied purines was carried out using molecular mechanics, ab ignition and DFT level of theories in the HyperChem release 8.0 software. Semi-empirical parameters were calculated using optimized structure of each of the purines as an input to the MOPAC software, while Muliken and Lowdin charges were calculated using GAMMES software. All quantum chemical calculations were carried out on gas phase.

Results and discussion

Weight loss measurements

The corrosion rate, CR g cm⁻² h⁻¹ and inhibition efficiency, IE%, as functions of concentration in the acid media were calculated using the equation [23]:

$$\text{CR}(\text{g}{\text{h}}^{-1}{\text{cm}}^{-2})=\frac{\mathrm{\Delta }W}{\mathit{At}}$$ (1)

$$\text{IE}\%=\left(1-\frac{\mathrm{\Delta }{W}_{\text{inh}}}{\mathrm{\Delta }{W}_{\text{blank}}}\right)\times 100$$ (2)

where ΔW is the weight loss in g, A is the surface area of the coupon and t is the immersion time, ΔW_inh and ΔW_blank are the weight losses (g) of aluminum in the presence and absence of the inhibitor respectively. The results obtained are presented in Table 1.

Table 1.

Corrosion rates of aluminum and inhibition efficiencies of adenine (AD), guanine (GU) and hypoxanthine (HYP) at 303 and 333 K respectively, in 0.1 M HCl.

Inhibitor	C (M)	Corrosion rate × 10−4 (g h−1 cm−2)		Inhibition efficiency (IE%)
		303 K	333 K	303 K	333 K
	Blank	1.26	8.11	–	–
AD	0.002	0.356	4.20	71.73	48.27
	0.004	0.306	3.51	75.70	56.74
	0.006	0.233	3.21	81.49	60.41
	0.008	0.173	2.49	86.28	69.29
	0.01	0.119	2.23	90.58	72.58
GU	0.002	0.231	3.06	81.65	62.28
	0.004	0.177	2.74	85.95	66.26
	0.006	0.127	2.52	89.92	68.91
	0.008	0.090	2.43	92.89	70.06
	0.01	0.044	2.15	96.53	73.50
HYP	0.002	0.216	3.59	82.84	55.79
	0.004	0.17	3.22	86.54	60.36
	0.006	0.136	2.85	89.22	64.88
	0.008	0.117	2.61	90.72	67.83
	0.01	0.098	2.35	92.23	71.01

Fig. 2 shows plots for the variation of IE% with concentration for AD, GU and HPY in 0.1 M HCl and at 303 and 333 K. The plots reveal that the inhibition efficiencies of the studied purines increase with increase in the concentration of the respective purine which suggest that the inhibition efficiency is a function of the amount of the inhibiting species present in the system and that the area of the aluminum surface covered by the adsorbed inhibitors is increased. Again, it is obvious from the plots that all the studied purines had high inhibition efficiencies with GU as the most effective inhibitor suggesting that not only the inhibitory power of the inhibitors increased with concentration but the performance also is a function of the type of purine.

Fig. 2.

Variation of inhibition efficiency with concentration for AD, GU and HPY for the corrosion of aluminum in 0.1 M HCl at 303 and 333 K.

Electrochemical impedance spectroscopy

Nyquist plots displayed in Fig. 3 revealed semicircles for all systems over the studied frequency range. The high frequency intercept with the real axis in the Nyquist plots is assigned to the solution resistance (R_s) and the low frequency intercept with the real axis is ascribed to the charge transfer resistance (R_ct).

Fig. 3.

Electrochemical impedance spectra of aluminum in 0.1 M solutions of HCl in the absence and presence of 0.01 M AD, GU and HYP at 303 K.

The impedance spectra were analyzed by fitting information to the equivalent circuit model R_s(Q_dlR_ct). In this equivalent circuit, the solution resistance was shorted by a constant phase element (CPE) that is placed in parallel to the charge transfer resistance. The CPE is used in place of a capacitor to compensate for deviations from ideal dielectric behavior arising from the inhomogeneous nature of the electrode surfaces. The impedance of the CPE is given by [24];

$$Z_{\text{CPE}}=Q^{-1}{(j\omega )}^{-n}$$ (3)

where Q and n represents the CPE constant and exponent respectively, j = (−1)^1/2 is an imaginary number, and ω is the angular frequency in rad s⁻¹ (ω = 2πf), while f is the frequency in Hz.

The corresponding electrochemical parameters are presented in Table 2 and from the results obtained, it can be stated that the presence of AD, GU and HYP increases the magnitude of R_ct, with corresponding decrease in the double layer capacitance (Q_dl). The increase in R_ct values in inhibited systems, which corresponded to an increase in the diameter of the Nyquist semicircle, confirms the corrosion inhibiting effect of the purines. The observed decrease in C_dl values, which normally results in the double-layer thickness can be attributed to the adsorption of the purines (with lower dielectric constant compared to the displaced adsorbed water molecules) onto the aluminum/acid interface, thereby protecting the metal from corrosion.

Table 2.

Impedance and polarization data for aluminum in 0.1 mol dm⁻³ HCl in the absence and presence of 0.01 mol dm⁻³ adenine (AD), guanine (GU) and hypoxanthine (HYP) at 303 K.

System	Impedance				Polarization
	Rct (Ω cm2)	Cdl (μΩ−1 Sn cm−2)	N	IE%	Ecorr (mV versus SCE)	icorr (μA cm−2)	IE%
Blank	101.00	21.89	0.99	–	−703.00	265.12	–
AD	907.00	3.72	0.92	88.87	−685.93	29.56	88.85
GU	1756.52	2.51	0.91	94.25	−707.72	14.93	94.37
HYP	1175.80	2.68	0.98	91.41	−703.05	23.28	91.22

Inhibition efficiency from the impedance data was estimated by comparing the values of the charge transfer resistance in the absence (R_ct) and presence of inhibitor (R_ct,_inh) as follows [18]:

$$\text{IE}\%=\left(\frac{R_{\text{ct}(\text{inh})}-R_{\text{ct}}}{R_{\text{ct}(\text{inh})}}\right)\times 100$$ (4)

The magnitude and trend of the obtained values presented in Table 3 are in close agreement with those determined from gravimetric measurements.

Table 3.

Langmuir, Flory Huggins and El Awardy et al. parameters for the adsorption of AD, GU and HPY on Al surface at 303 and 333 K.

Isotherm	T (K)	Slope	Intercept	n/1/y	$\mathrm{\Delta }{G}_{\text{ads}}^0(\text{kJ}/\text{mol})/\text{B}$	R2
Langmuir	AD (303 K)	0.8545	−0.2400		−8.73	0.9986
	AD (333 K)	0.7463	−0.3632		−8.80	0.9972
	GU (303 K)	0.8982	−0.183		−9.06	0.9997
	GU (333 K)	0.9032	−0.0541		−10.77	0.9997
	HPY (303 K)	0.9332	−0.0982		−9.55	1.0000
	HPY (333 K)	0.8505	−0.1458		−10.19	0.9996
Flory Huggins	AD (303 K)	1.1246	3.0421	1	−27.77	0.8392
	AD (333 K)	1.7446	2.8094	2	−26.42	0.8930
	GU (303 K)	1.2798	3.4904	1	−30.37	0.9289
	GU (333 K)	4.2163	4.2271	4	−34.64	0.9411
	HPY (303 K)	1.8878	4.0206	2	−33.44	0.9784
	HPY (333 K)	3.1649	3.5061	3	−30.46	0.9520
El Awardy et al.	AD (303 K)	9.0275	25.803	0.1108	−25.55	0.7581
	AD (333 K)	2.3778	7.1637	0.4206	−29.02	0.8748
	GU (303 K)	13.356	39.577	0.0748	−26.13	0.8568
	GU (333 K)	1.4599	5.5222	0.6850	−33.71	0.9208
	HPY (303 K)	9.6585	30.287	0.1035	−27.09	0.9332
	HPY (333 K)	1.6404	5.5832	0.6096	−31.40	0.9334

n is applicable to Flory Huggins while 1/y and B are for El Awardy et al. adsorption isotherms.

Potentiodynamic polarization data

Polarization measurements were undertaken to investigate the behavior of aluminum electrodes in 0.1 M solutions of HCl in the absence and presence of the purines. The current–potential relationship for the aluminum electrode in various test solutions is shown in Fig. 4 while the electrochemical data obtained from the polarization curves are presented in Table 3.

Fig. 4.

Polarization curves of aluminum in 0.1 M solutions of HCl in the absence and presence of 0.01 M AD, GU and HYP at 303 K.

Addition of the purines is seen to affect the cathodic partial reaction mostly, thereby reducing the cathodic current densities and the corresponding corrosion current density (i_corr). This indicates that the purines functioned as cathodic inhibitors for the corrosion of aluminum in 0.1 M HCl solutions. Adenine (AD) however, is also observed to affect the anodic arm of the Tafel plot, slightly, indicating that is functioned as a mixed inhibitor in 0.1 M HCl [11]. It was also seen that the potential range in the Tafel plots is short. This can be explained as follows. A typical Tafel plots will show Tafel region, plateau region and high polarization region. This study revealed the dominance of the Tafel region and thus a short potential range. The values of corrosion current densities in the absence (i_corr) and presence of inhibitor (i_inh) were used to estimate the inhibition efficiency from polarization data (IE_i%) using Eq. (5) and the results are also presented in Table 3 [25]

$${\text{IE}}_i\%=\left(1-\frac{i_{\text{inh}}}{i_{\text{corr}}}\right)\times 100$$ (5)

Adsorption study

The nature of interaction between the corroding surface of the metal during corrosion inhibition can be explained in terms of the adsorption characteristics of the inhibitor. In this study, results obtained for degree of surface coverage at 303 and 333 K were fitted to a series of different adsorption isotherms including Flory–Huggins, Langmuir, Freundlich and Temkin isotherms. The tests revealed that Langmuir adsorption model best described the adsorption characteristics of the studied purines [26]

$$\frac{C}{\theta }=C+\frac{1}{b_{\text{ads}}}$$ (6)

where k is the adsorption equilibrium constant, C is the concentration of inhibitor and θ is the degree of surface coverage of the inhibitor. From the logarithm of both sides of Eqs. (6) and (7) was obtained,

$$\mathit{log}\left(\frac{C}{\theta }\right)=\mathit{log}C-\mathit{log}{b}_{\text{ads}}$$ (7)

By plotting values of log(C/θ) versus values of log C, straight line graphs were obtained as shown in Fig. 5 while adsorption parameters deduced from the isotherms are presented in Table 4. From the results obtained, R² values ranging from 0.9972 to 1.000 were obtained. This indicated a high degree of fitness of the adsorption data to the Langmuir model. However, values obtained for slopes were less than unity indicating the existence of interaction between the adsorbed species and that some components of GU, AD and HPY molecules will occupy more than one adsorption sites on the Al surface [27]. Therefore, Flory Huggins and El Awardy et al. isotherms were also used to explain the existence of interaction.

Fig. 5.

Langmuir isotherms for the adsorption of AD, GU and HPY onto Al surface at 303 and 333 K.

Table 4.

Calculated values of activation energies (E_a) and heats of adsorption (Q_ads) for the corrosion of aluminum in 0.1 M HCl in the absence and presence of various concentrations of adenine (AD), guanine (GU) and hypoxanthine (HYP).

Inhibitor	Concentration mol dm−3	Activation energy, Ea (kJ mol−1)	Heat of adsorption, Qads (kJ mol−1)
	Blank	52.13
AD	0.002	69.10	−10.58
	0.004	68.31	−9.15
	0.006	73.44	−11.20
	0.008	74.66	−10.84
	0.01	78.94	−12.28
GU	0.002	72.34	−10.48
	0.004	76.39	−12.01
	0.006	77.46	−14.72
	0.008	82.39	−18.18
	0.01	84.55	−24.37
HYP	0.002	78.69	−14.19
	0.004	82.35	−15.23
	0.006	85.18	−15.86
	0.008	86.93	−16.21
	0.01	88.98	−16.69

Flory–Huggins adsorption models consider that prior to adsorption, some molecules of water must be replaced by corresponding molecules of the inhibitor such that the following equilibrium (Eq. (8)) is established [28],

$$A_{\text{soln}}+n{({\text{H}}_2\text{O})}_{\text{ads}}=A_{\text{ads}}n{({\text{H}}_2\text{O})}_{\text{soln}}$$ (8)

$$\frac{\theta }{{(1-\theta )}^n}=b_{\text{ads}}C$$ (9)

where n is the number of adsorption site. From Eq. (8), Flory Huggins derived an adsorption model expressed by Eq. (9). The main characteristic of the above isotherm is the appearance of the term, θ/(1 − θ)ⁿ in the expression. From the logarithm and rearrangement of Eqs. (9) and (10) was obtained,

$$\mathit{log}\left(\frac{\theta }{C}\right)=\mathit{log}{b}_{\text{ads}}+\mathit{nlog}(1-\theta )$$ (10)

Fig. 6 shows the Flory–Huggins plots for the adsorption of AD, GU and HPY on Al surface at 303 and 333 K. Adsorption parameters deduced from the plots are also presented in Table 4. From the results obtained, it can be seen that the numerical values of n change from 1 to 2, 1 to 4 and 2 to 3 for AD, GU and HPY at 303 and 333 K, respectively. These changes indicated that the number of water molecules that must be replaced by the respective inhibitor’s molecule increases with increase in temperature supporting the formation of multi-molecular layer of adsorption as the temperature increases from 303 to 333 K.

Fig. 6.

Flory Huggins isotherm for the adsorption of AD, GU and HPY on aluminum surface at 303 and 333 K.

The strength of adsorption of AD, GU and HPY on the surface of Al and the possibility of formation of multi-molecular layer of adsorption were also investigated using the El-Awady et al. kinetic isotherm, which can be written as Eq. (11) [29],

$$\log \left(\frac{\theta }{1-\theta }\right)=\mathit{log}{b}^{\prime }+\mathit{ylog}C$$ (11)

where y is the number of inhibitor molecules occupying one active site and 1/y represents the number of active sites on the surface occupied by one molecule of the inhibitor. ‘y’ is also related to the binding constant, B through B = b^(1/y). Fig. 7 shows El-Awady et al., isotherm for the adsorption of the studied purines while adsorption parameters deduced from the isotherm are also presented in Table 4. The results obtained reveal that values of 1/y are less than unity confirming that a given inhibitor’s molecules will occupy more than one active site (i.e. 1/y < 1). Also, B values were found to increase with temperature. Generally, larger value of the binding constant (B) implies better adsorption arising from stronger electrical interaction between the double layer existing at the phase boundary and the adsorption molecule. On the other hand, small values of the binding constant suggest weaker interaction between the adsorbing molecules and the metal surface. Therefore, the extent of adsorption of AD, GU and HPY on Al surface increases with temperature,

Fig. 7.

El awardy et al. isotherm for the adsorption of AD, GU and HPY on aluminum surface at 303 and 333 K.

The equilibrium constant of adsorption (b_ads) obtained from the adsorption models, is related to the standard free energy of adsorption $\mathrm{\Delta }{G}_{\text{ads}}^0$ according to Eq. (12) [30]:

$$b_{\text{ads}}=\frac{1}{55.5}\mathit{exp}\left(\frac{\mathrm{\Delta }{G}_{\text{ads}}^0}{\mathit{RT}}\right)$$ (12)

where R is the molar gas constant, T is the absolute temperature and 55.5 is the molar concentration of water in the solution. Values of $\mathrm{\Delta }{G}_{\text{ads}}^0$ calculated from Eq. (12) are also presented in Table 4. From the results obtained, the free energies are negatively less than the threshold value (−40 kJ/mol) expected for the mechanism of chemical adsorption hence the adsorption of AD, GU and HPY on Al surface is consistent with electrostatic interactions between the inhibitors’ molecules and charged metal surface, which support physisorption mechanism [31].

Effect of temperature

The adsorption of an organic inhibitor can affect the corrosion rate by either decreasing the available reaction area (geometric blocking effect) or by modifying the activation energy of the anodic or cathodic reactions occurring in the inhibitor-free surface in the course of the inhibited corrosion process. The adsorption mechanism of AD, GU and HYP onto aluminum was investigated by changing the temperature of the systems from 303 to 333 K. The apparent activation energies (E_a) for the corrosion process in the absence and presence of AD, GU and HYP were calculated using a modified form of the Arrhenius equation [32]:

$$\mathit{log}\frac{{\text{CR}}_1}{{\text{CR}}_2}=\frac{E_a}{2.303R}\left(\frac{1}{T_1}-\frac{1}{T_2}\right)$$ (13)

where CR₁ and CR₂ are the corrosion rates at temperatures T₁ and T₂, respectively. Calculated values of E_a are presented in Table 5. The activation energies are higher in inhibited HCl solutions compared to the uninhibited system (blank). This is frequently interpreted as being suggestive of formation of an adsorption film of physical/electrostatic nature [33].

Table 5.

Computed values of semi-empirical parameters for adenine, quinine and hypoxanthine.

	CNDO	MNDO	AM1	RM1	PM3
EHOMO (eV)
Adenine	−11.430	−9.591	−9.511	−9.424	−9.062
Quanine	−10.130	−9.054	−8.993	−8.885	−8.688
Hypoxanthine	−11.577	−9.876	−9.851	−9.811	−9.596
ELUMO (eV)
Adenine	3.520	−0.102	−0.021	0.137	−0.263
Quanine	3.231	−0.164	−0.100	0.084	−0.280
Hypoxanthine	2.776	−0.665	−0.583	−0.509	−0.800
EL–H (eV)
Adenine	14.950	9.489	9.491	9.561	8.799
Quanine	13.361	8.890	8.892	8.968	8.408
Hypoxanthine	14.353	9.211	9.268	9.302	8.796
μ (Debye)
Adenine	6.157	5.803	5.989	6.327	6.266
Quanine	3.747	2.170	2.322	2.482	2.491
Hypoxanthine	6.416	5.728	5.717	6.100	5.986
Eb (eV)
Adenine	−221.16	−69.17	−67.54	−71.12	−69.31
Quanine	−1555546.00	−39442.70	−1057934.00	−38629.40	−39568.00
Hypoxanthine	−1385995.00	−35667.00	−34597.80	−32733.40	−35460.80
EE (eV)
Adenine	−9490.55	−7789.66	−7778.36	−7779.22	−7385.55
Quanine	−6028329.00	−4974777.00	−4967677.00	−4966004.00	−4740755.00
Hypoxanthine	−5158937	−4222164.00	−4215761.00	−4212423.00	−4021481.00
TE (eV)
Adenine	−2650.37	−1741.59	−1738.20	−1741.78	−1476.84
Quanine	−1678629.00	−1099277.00	−1096446.00	−1096585.00	−943142.00
Hypoxanthine	−113224.00	−981435.00	−978842.00	−976977.00	−848255.00
CCR (eV)
Adenine	3,643,243	3,221,244	3,217,150	3,215,699	3,147,134
Quanine	4,349,700	3,875,500	3,871,231	3,869,402	3,797,613
Hypoxanthine	3,659,719	3,240,729	3,236,920	3,234,293	3,173,226

The heat of adsorption (Q_ads) was quantified from the trend of surface coverage with temperature using the following equation [34]:

$$Q_{\text{ads}}=2.303R\left[\mathit{log}\left(\frac{{\theta }_2}{1-{\theta }_2}\right)-\mathit{log}\left(\frac{{\theta }_1}{1-{\theta }_1}\right)\right]\times \frac{T_1{T}_2}{T_2-T_1}$$ (14)

where θ₁ and θ₂ are the degrees of surface coverage at temperatures T₁ and T₂, and R is the gas constant. Negative Q_ads values were obtained for the inhibition behavior of AD, GU and HYP (Table 5). This implies that the inhibition of Al corrosion by the studied purines is exothermic and that their inhibition efficiencies decreased with increase in temperature (see Table 1) which is a good indication of a physisorptive kind of interaction between these purines and the metal surfaces.

Quantum chemical study

Global reactivity

Quantum chemical principles have been widely used to study corrosion inhibition including structure optimization calculations, semi-empirical, ab initio and DFT calculations. In this study, calculated values of semi-empirical parameters for different Hamiltonians (namely, CNDO, MNDO, AM1, RM1 and PM3) were correlated with experimental inhibition efficiencies, while Fukui functions were used to study electrophilic substitution within the inhibitors.

Table 5 presents values of the frontier molecular orbital energies (i.e. energy of the highest occupied molecular orbital (E_HOMO), energy of the lowest unoccupied molecular orbital (E_LUMO) and the energy gap (E_L–H)), total energy (TE), electronic energy (EE), core core interaction energy (CCR) and dipole moment (μ). Figs. 8–11 present plots for the variation of E_L–H, TE, EE and E_b with experimental inhibition efficiencies of the studied inhibitors. E_HOMO indicates the tendency of an inhibitor to donate electron while E_LUMO is an index that indicate the tendency of a molecular specie to accept electron. The difference between E_LUMO and E_HOMO is the energy gap (i.e. E_L–H). In view of this, corrosion inhibition efficiency is expected to increase with increasing values of E_HOMO and with increase in the value of E_LUMO and that of the energy gap. Correlations between calculated values of E_HOMO and experimental inhibition efficiencies were very poor and R² were lower than 0.45 for all the Hamiltonian considered. Similarly, calculated values of E_LUMO did not correlate significantly with values of experimental inhibition efficiencies. This observation suggests that the inhibition efficiencies of the studied purines are not affected by electron transfer process, a mechanism that favors physical adsorption as proposed earlier. On the other hand, better correlation was obtained between E_L–H and experimental inhibition efficiencies of the studied purines. Generally, the energy gap of a molecule is a quantum chemical parameter that indicates hardness or softness of molecular specie. Hard molecules are characterized with larger value of energy gap and are less reactive than soft molecules, which are characterize by small energy gap [35]. Therefore, corrosion inhibition potential of a molecule is expected to increase with decreasing value of E_L–H as observed in the present study. Although PM3 Hamiltonian did not give excellent correlation between experimental inhibition efficiency with E_L–H, calculated values of R² for CNDO, MNDO, AM1 and RM1 Hamiltonians were within the range of 0.7297 and 0.8155 indicating better relationship between E_L–H and the measured inhibition efficiency (Fig. 8). Excellent correlations were also found between experimental inhibition efficiency and TE and also for EE and E_b (Figs. 9–11). Correlations between IE_exp and TE were excellent for MNDO, AMI, RM1 and PM3 Hamiltonians as indicated in the plots (Fig. 9). Similarly, excellent correlations were found for MNDO, AM1, RM1 and PM3 Hamiltonians with respect to the variation of IE_exp and EE of the molecules. However, AM1 Hamiltonian did not give excellent correlation between IE_exp and E_b. Since each Hamiltonians is based on specific assumption, it can be stated that the failure of some of these assumptions for some molecules can lead to poor correlation.

Fig. 8.

Variation of E_L–H with experimental inhibition efficiencies of ADN, GUN and HYP for CNDO, MNDO, RM1 and PM3 Hamiltonians.

Fig. 9.

Variation of TE with experimental inhibition efficiencies of ADN, GUN and HYP for MNDO, AM1, RM1 and PM3 Hamiltonians.

Fig. 10.

Variation of EE with experimental inhibition efficiencies of ADN, GUN and HYP for MNDO, AM1, RM1 and PM3 Hamiltonians.

Fig. 11.

Variation of E_b with experimental inhibition efficiencies of ADN, GUN and HYP for CNDO, MNDO, RM1 and PM3 Hamiltonians.

Ionization energy and electron affinity of the inhibitors were calculated using the method of finite difference approximation as follows [36],

$$\text{IE}=E_{(N-1)}-E_{(N)}$$ (15)

$$\text{EA}=E_{(N)}-E_{(N+1)}$$ (16)

where IE and EA are ionization energy and electron affinity respectively, E_(N−1), E_(N) and E_(N+1) are the ground state energies of the system with N − 1 and N + 1 electrons respectively. Calculated values of IE and EA are presented in Table 6. Correlation between IE_exp and IE and between IE_exp and EA were excellent (R² ranged from 0.79 to 0.89). Also, strong correlations were obtained between IE and E_HOMO and between EA and E_LUMO indicating that IE is associated with the tendency of the inhibitor to donate electron while EA is associated with the tendency of the inhibitors to accept electron. Similar findings have been reported by other researchers [37]. The global hardness which is the inverse of the global softness (i.e. η = 1/S) can be evaluated using Eq. (17),

$$S=\frac{1}{[(E_{(N-1)}-E_{(N)}-(E_{(N)}-E_{(N+1)})]}$$ (17)

where S is the global softness and η is the global harness. Calculated values of S and η are also presented in Table 6. Although calculated values of S and η did not show strong correlation with the experimental inhibition efficiencies (R² = 0.589 and 0.657 respectively), correlations between these parameters and E_L–H were strong indicating that S and η are related to E_L–H which is an index for measuring the softness of a molecule.

Table 6.

Some quantum chemical descriptors for AD, GU and HPY.

Inhibitor	Hamiltonian	IE (eV)	EA (eV)	S	η (eV)	χ (eV)	δ
HYP	CNDO	10.9662	−2.0136	0.0770	12.9798	6.4899	−0.0128
	MNDO	9.4434	−0.1960	0.1037	9.6394	4.8197	0.0694
	AM1	9.4234	−0.2467	0.1034	9.6701	4.8351	0.0684
	RM1	9.3801	0.9197	0.1182	8.4604	4.2302	0.1140
	PM3	9.1975	−0.4223	0.1040	9.6198	4.8099	0.0701
QU	CNDO	9.1143	−2.4394	0.0866	11.5537	5.7769	0.0165
	MNDO	8.5389	−0.2467	0.1138	8.7856	4.3928	0.1005
	AM1	8.4968	0.0251	0.1180	8.4717	4.2358	0.1135
	RM1	8.3759	0.0642	0.1203	8.3117	4.1558	0.1205
	PM3	−4.8242	13.2261	−0.0554	−18.0503	−9.0252	−0.4206
AD	CNDO	10.1571	−2.7802	0.0773	12.9373	6.4687	−0.0120
	MNDO	9.1715	−1.1005	0.0974	10.2720	5.1360	0.0498
	AM1	8.9903	−0.7983	0.1022	9.7885	4.8943	0.0646
	RM1	8.1140	−0.6235	0.1144	8.7375	4.3687	0.1024
	PM3	8.4825	−0.3308	0.1135	8.8134	4.4067	0.0994

The fraction of electron transferred, δ can be calculated using Eq. (18) [38],

$$\delta =\frac{({\chi }_{\text{Al}}-{\chi }_{\text{inh}})}{({\eta }_{\text{Al}}+{\eta }_{\text{inh}})}$$ (18)

where χ_Al and χ_inh are the electronegativities of Al and the inhibitor respectively and can be evaluated as χ = (IP + EA)/2. η_Al and η_inh are the global hardness of Al and the inhibitor respectively. Calculated values of δ are also presented in Table 6. δ values did not correlate strongly with the experimental inhibition efficiencies of the inhibitors and were relatively low, indicating that fewer electrons were transferred from the inhibitor to the metal surface. Therefore, the inhibition of the corrosion of aluminum by AD, GU and HYP supports the transfer of charge or electron from the inhibitor to the metal surface indicating the occurrence of physical and chemical adsorption mechanism. However, physical adsorption mechanism precedes chemisorption mechanism.

Local reactivity

In this study, local reactivity was investigated using the Fukui functions deduced through DFT and MP2 calculations. The Fukui function has been formally defined as

$$f(r)={\left[\frac{\delta \mu }{\delta v(r)}\right]}_N$$ (19)

where v(r) is the external potential and the functional derivatives must be taken at constant number of electrons. Assuming that the total energy E as a function of N and functional of v(r) is an exact differential, the Maxwell relations between derivatives may be applied to write

$$f(r)={\left(\frac{\delta \rho (r)}{\delta N}\right)}_v$$ (20)

Eq. (20) is the most standard presentation of the Fukui function. Owing to the discontinuity of the chemical potential at integer N, the derivative will be different if taken from the right or the left side. Therefore, there are three different functions, f⁺(r) (corresponding to the situation when the derivative is taken as N increases from N to N + δ), f⁻(r) corresponding to a situation when N decreases to N − δ) and f⁰(r) the average of the two. In practice, condensed Fukui function is often used and are defined as follows,

$$f_k^{-}=q_k(N)-q_k(N-1)$$ (21)

$$f_k^{+}=q_k(N+1)-q_k(N)$$ (22)

In this work, condensed Fukui functions were calculated using Huckel and Muliken charges at DFT and MP2 level of theories. Values of Mulliken charges and Fukui functions calculated for DFT level of theories using Muliken charges are presented in Tables 7 and 8 respectively. Fukui functions computed for MP2 level of theory using Muliken and Lowdin charges are also presented in Tables 9–11. From the results obtained, values of condensed Fukui functions were negative indicating unphysical results [39]. Therefore, calculation of the binding energies between Al and the inhibitors, for the various positions of hetero atoms was carried out.

Table 7.

Milliken charges of GU, AD and HPY radical (N), cation (N − 1) and anion (N + 1) calculated from DFT.

Atom no.	N	N − 1	N + 1	N	N − 1	N + 1	N	N − 1	N + 1
1	−0.7902	−0.7589	−0.8082	−0.7660	−0.7341	−0.7915	−0.5819	−0.5061	−0.6562
2	0.2771	0.3325	0.1718	0.2733	0.3244	0.1761	0.2015	0.2458	0.1719
3	−0.5039	−0.4366	−0.5719	−0.5251	−0.4609	−0.5798	−0.5401	−0.4290	−0.6267
4	0.4495	0.5163	0.3833	0.4391	0.4784	0.3539	0.4080	0.4416	0.3438
5	−0.5341	−0.4344	−0.6306	−0.6336	−0.5295	−0.6736	0.2737	0.3327	0.2470
6	0.1986	0.2895	0.1636	0.8830	0.9417	0.8213	0.6021	0.6408	0.5002
7	−0.5637	−0.4845	−0.6381	−0.7965	−0.7318	−0.8196	−0.7857	−0.6776	−0.8196
8	0.6477	0.7381	0.5186	−0.8149	−0.8037	−0.8267	−0.4961	−0.4242	−0.5629
9	−0.5898	−0.5137	−0.6474	0.6750	0.7290	0.5698	0.2766	0.3166	0.1733
10	0.2463	0.3025	0.2252	−0.5217	−0.3956	−0.6328	−0.7631	−0.7213	−0.7845
11				0.0.2485	0.3283	0.2080

Table 8.

Condensed Fukui functions for AD, GU and HPY calculated from Milliken charges using DFT.

Atom no.	HPY		GU		AD
	$f_k^{+}$	$f_k^{+}$	$f_k^{+}$	$f_k^{+}$	$f_k^{+}$	$f_k^{+-}$
1	−0.0180	−0.0313	−0.0255	−0.0319	−0.0742	−0.0759
2	−0.1053	−0.0554	−0.0972	−0.0512	−0.0296	−0.0443
3	−0.0680	−0.0673	−0.0547	−0.0642	−0.0867	−0.1111
4	−0.0662	−0.0668	−0.0852	−0.0394	−0.0643	−0.0336
5	−0.0965	−0.0997	−0.0401	−0.1041	−0.0267	−0.0590
6	−0.0350	−0.0909	−0.0617	−0.0588	−0.1019	−0.0387
7	−0.0744	−0.0792	−0.0231	−0.0647	−0.0339	−0.1081
8	−0.1291	−0.0904	−0.0118	−0.0112	−0.0669	−0.0719
9	−0.0575	−0.0762	−0.1053	−0.0540	−0.1033	−0.0400
10	−0.0212	−0.0562	−0.1111	−0.1261	−0.0214	−0.0418
11			−0.0405	−0.0798

Table 9.

Huckel charges and condensed Fukui functions for GU calculated from Milliken and Lowdin charges using MP2 level of theory.

Atom and atom number	Huckel charge	Milliken		Lowdin
		$f_k^{+}$	$f_k^{-}$	$f_k^{+}$	$f_k^{-}$
N(1)	0.2591	−0.0370	0.0048	−0.0613	0.0219
C(2)	0.1757	−0.0674	0.0180	−0.1075	0.0323
N(3)	−0.5442	0.0079	−0.1262	0.0372	−0.1346
C(4)	0.2082	−0.1063	0.0433	−0.1192	0.0434
N(5)	−0.5898	−0.0672	−0.1977	−0.0705	−0.2581
C(6)	0.3965	−0.0376	−0.0576	−0.0235	−0.0526
N(7)	0.0237	−0.0208	−0.0533	−0.0250	−0.0719
N(8)	0.2467	−0.0004	−0.0422	−0.0292	−0.0515
C(9)	0.4078	−0.2194	0.0627	−0.2545	0.1051
O(10)	−0.8803	−0.1562	−0.2689	−0.1693	−0.3117
C(11)	−0.0993	0.0386	−0.1307	0.0569	−0.1385

Table 10.

Huckel charges and condensed Fukui functions for AD calculated from Milliken and Lowdin charges using MP2 level of theory.

Atom and atom number	Huckel charge	Milliken		Lowdin
		$f_k^{+}$	$f_k^{-}$	$f_k^{+}$	$f_k^{-}$
N(1)	−0.5113	−0.0769	−0.0716	−0.0841	−0.0577
C(2)	0.2450	−0.0286	−0.1250	−0.0402	−0.1479
N(3)	−0.3972	−0.0658	−0.1378	−0.0584	−0.2491
C(4)	0.1554	−0.0116	−0.0552	0.0161	−0.0348
C(5)	−0.0362	−0.0552	−0.0397	−0.0587	−0.0249
C(6)	0.3228	−0.0495	−0.0680	−0.0433	−0.0767
N(7)	−0.0216	−0.0168	−0.0369	−0.0211	−0.0498
N(8)	−0.4966	−0.1472	−0.0482	−0.1892	−0.0315
C(9)	0.1837	−0.1299	−0.0509	−0.1957	−0.0621
N(10)	0.2447	−0.0163	−0.0241	−0.0567	−0.0303

Table 11.

Huckel charges and condensed Fukui functions for HPY calculated from Milliken and Lowdin charges using MP2 level of theory.

Atom and atom number	Huckel charge	Milliken		Lowdin
		$f_k^{+}$	$f_k^{-}$	$f_k^{+}$	$f_k^{-}$
N(1)	0.2471	−0.0166	−0.0401	−0.0447	−0.0594
C(2)	0.2316	−0.1389	−0.0396	−0.2119	−0.0499
N(3)	−0.5196	−0.1297	−0.0810	−0.1636	−0.0671
C(4)	0.2621	−0.0249	−0.0320	−0.0016	−0.0235
N(5)	−0.4258	−0.1048	−0.0867	−0.1223	−0.0911
C(6)	0.2277	−0.0254	−0.0904	−0.0318	−0.1241
N(7)	−0.3770	−0.0738	−0.0389	−0.0788	−0.0177
C(8)	0.3372	−0.0945	−0.1278	−0.0996	−0.1486
O(9)	−0.2551	−0.0426	−0.0776	−0.0410	−0.0812
C(10)	−0.0292	−0.0053	−0.0815	0.0206	−0.1285

The interaction energy between the inhibitor and the aluminum atom can be calculated using Eq. (23) [40],

$$E_{\text{int}}=E_{(\text{Al}-X)}-(E_x+E_{\text{Al}})$$ (23)

where E_Al is the total energy of the iron atom, E_x is the total energy of inhibitory compound and E_(Al−X) is the energy of interaction between aluminum and the inhibitor. When absorption occurs between the compound and the aluminum atom, the energy of the new system is expressed as E_x + E_Al. Table 12 shows the binding energies for various positions of the hetero atoms in AD, GU and HPY respectively. Since every systems prefer to remain in states of lowest energy, it can be stated that in GU the adsorption of the inhibitor (hence inhibition mechanism) occurred in the imine nitrogen (i.e. N5), in AD, the adsorption site is emanine nitrogen (i.e. N7) and in HPY, the adsorption site is in the pyridine nitrogen (i.e. N5). Considerations of Huckel charges for these sites (Tables 9–11) also revealed that these sites possess reasonable negative charges that can interact with the charged metal (Al) surface. Since inhibition efficiency of an inhibitor is closely related to the strength of adsorption, it can be stated that the listed adsorption sites enhances the efficiencies of the studied purines due to the high concentration of electrons. Hence the HOMO and LUMO molecular orbital of AD, GU and HPY are produced in Fig. 12. It is significant to note that the LUMO corresponds to f⁺(r) (i.e. tendency to accept electron) while the HOMO corresponds to f⁻(r) (i.e. tendency to donate electron). Therefore, the nature and distribution of the lopes in the HOMO and LUMO diagrams (Fig. 12) also support the findings deduced from the interaction energies and from Huckel charges (Table 12).

Table 12.

Binding energies for various positions of hetero atoms in AD, GU and HPY.

Inhibitor	Atom	E(Al−X) (J/mol)	Einh + EAl (J/mol)	Eint (J/mol)
GU	N1	18.433	18.376	0.057
	N3	13.156	18.376	−5.22
	N5	17.266	18.376	−1.11
	N7	18.531	18.376	0.155
	O10	16.714	18.376	−1.662
AD	N1	17.202	19.061	−1.859
	N3	17.786	19.061	−1.275
	N7	18.992	19.061	−0.069
	N8	13.056	19.061	−6.005
	N10	20.081	19.061	1.02
HPY	N1	21.895	20.294	1.601
	N3	15.206	20.294	−5.088
	N5	18.204	20.294	−2.09
	N7	19.122	20.294	−1.172
	O9	22.539	20.294	2.245

Fig. 12.

HOMO and LUMO molecular orbitals of AD, GU and HPY.

Conclusions

From the results and findings of this study, the following conclusions were made,

• i.AD, GU and HPY are excellent corrosion inhibitors for the corrosion of Al in acidic medium.
• ii.The inhibitors are adsorption inhibitors and their adsorptions on Al surface were spontaneous and were consistent with physiosorption mechanism. The adsorption behavior of the inhibitors supported the models of Flory Huggins, Langmuir and El Awardy et al.
• iii.Inhibition efficiencies of AD, GU and HPY strong correlations with some quantum chemical parameters. Fukui functions at DFT and MP2 levels of theory are unphysical in predicting the sites for electrophilic and nucleophilic attacks but considerations of interaction energies between Al and various sites of the hetero atoms revealed likely adsorption sites.

Conflict of interest

The authors have declared no conflict of interest.

Compliance with Ethics Requirements

This article does not contain any studies with human or animal subjects.