Spontaneous formation of mono-n-butyl phosphate and mono-n-hexyl phosphate thin films on the iron surface in aqueous solution and their corrosion protection property

Abstract

Mono-n-butyl phosphate (BP) and mono-n-hexyl phosphate (HP) thin films were directly formed on an iron surface by immersing pure iron samples in aqueous solutions containing BP or HP. FTIR analyses, XPS characterizations and water contact angle data show that the iron surfaces were covered with the BP or HP thin films. The spontaneous formation of alkyl phosphate thin films on the iron surface is largely attributed to the electrostatic interaction between Fe²⁺ ions on the surface of iron and RO–PO₃H⁻ (R represents n-butyl or n-hexyl) ions and the specific adsorption of phosphate groups on the iron substrate. Moreover, electrochemical results clearly demonstrate that the as-formed thin films can effectively protect the iron substrate from corrosion in NaCl corrosive solutions.

---

1. Introduction

Iron is currently the most widely used metal and iron-based alloys are used in countless applications as construction materials, medical devices, tools, wires and appliances, so corrosion control of iron and its alloys has always been an old and new research subject.¹ At the present time, there are various ways of protecting iron and iron alloys from corrosion (oxidation) including coating,^2–6 plating,⁷ hot dip galvanizing,⁸ cathodic protection,⁹ and combinations of different protection methods. In particular, self-assembly of organic molecules on the metal surfaces provides a promising film formation method, by which a barrier of corrosion-resistant material between the material substrate and the damaging environment can be established in an easy form.^10–12 In recent years, much work in this field has focused on self-assembled monolayers (SAMs) of alkanethiols adsorbed onto the surfaces of coinage metals (gold, silver and copper)^13–15 and SAMs of silanes on silicon oxide surface,¹⁶ especially thiol-based SAMs on gold,^17–20 which are extensively used as model systems in a wide range of studies. In contrast, less attention has been paid to the spontaneous formation of ordered molecular layers on the iron surface. Because iron is easily oxidized,²¹ it is very difficult to prepare the alkanethiol SAMs on the iron surface with oxides by following the similar film-forming procedures. Meanwhile, due to the lack of special interaction between iron substrate and common organic molecules, so far few typical self-assembly systems have been created on the surfaces of iron and its alloys. On the other hand, the existing self-assembly methods can not meet the requirements for preparing practical corrosion protection layers directly on the unmodified iron surface. The self-assembling process is usually carried out in organic solvents;¹⁷ however, on an industrial scale, organic solvents are increasingly falling into disfavor because of both air-pollution and water-pollution issues.¹⁷ From the perspective of corrosion protection, it is very necessary to develop a practical self-assembly system based on spontaneous adsorption of non-toxic inhibitor molecules onto the iron surface in water phase.

Key technical problems that probably occur in creating the practical self-assembly systems include: (i) what type of inhibitor is selected as the film-forming substance? (ii) How big should the solubility of the film-forming substance in water be? (iii) In what manner do the inhibitor molecules dissolved in an aqueous solution spontaneously adsorb onto the iron surface? According to theory of organic corrosion inhibitors, the film-forming substance should have heteroatoms (S, O, N, and P atoms) that act as the active centers for the process of adsorption on the metal surfaces.^22,23 The existing data have shown that the inhibition efficiency follows the sequence O < N < S < P. Thus, alkyl phosphate (or phosphonate) esters are ideal candidates among various film-forming substances.²⁴ It is worth noting that, phosphonic acid compounds have been reported to form SAMs on the oxide surfaces, including Al₂O₃, TiO₂, ZrO₂, SiO₂.^19,25–33 At the same time, the spreading, adsorption, and self-assembly of soluble amphiphilic molecules at the air–water interface attract our great attention.

Seeing that short-chain alkyl phosphate esters (less than 8 carbon atoms) are environmentally-friendly amphiphilic substances and can be synthesized easily, we chose mono-n-butyl phosphate (BP) and mono-n-hexyl phosphate (HP) as the film-forming substances with the aim to explore whether this type of amphiphilic molecules could assemble ordered thin films on the iron surface in aqueous solutions. Excitingly, it was found that, when an iron electrode was immersed in an aqueous solution containing BP or HP, both alkyl phosphate esters were able to form self-assembled thin films on the iron substrate via the spontaneous adsorption onto the iron surface at the solid–water interface. More importantly, electrochemical measurement results indicate that the as-formed thin films possessed good corrosion resistance ability in NaCl corrosive medium. The present study opens a facile and effective route for the design and assembly of corrosion protection layers on the surfaces of iron and its alloys.

2. Experimental

2.1. Materials

Iron electrode was made from a 99.98% pure iron rod with 0.2 cm diameter. The iron rod specimen was embedded in epoxy resin mould, leaving only its cross-section exposed to BP or HP aqueous solutions and corrosive electrolyte solutions. Before each test, the electrode was ground with emery papers of decreasing particle size to #2000 finish, and then rinsed with deionized water, degreased with 95% ethanol, and finally dried in a stream of N₂.

BP and HP were synthesized by a direct esterification reaction of phosphorus pentoxide (P₂O₅) with 1-butanol or 1-hexanol, followed by purification of products. The detailed procedure for the synthesis and purification was given in Section S1 of ESI.† Structural formulae of BP and HP molecules are shown in Fig. 1.

Fig. 1 Chemical structural formulae of BP (a) and HP (b).

The as-synthesized BP and HP products were dissolved in ultrapure water (∼18 MΩ cm) according to the designated weight ratios in order to prepare three groups of aqueous solutions with different weight percentage concentrations (0.1 wt%, 0.01 wt%, and 0.001 wt%). The BP and HP aqueous solutions were used as the film-forming solutions in which BP or HP could spontaneously adsorb onto the iron substrate and form self-assembled thin films. The polished iron electrodes and small iron plates (0.5 cm × 1.0 cm) were immersed in the BP or HP aqueous solutions for 10 min, 30 min and 1 h respectively depending on experimental requirements. A 3.5 wt% NaCl solution was prepared with ultrapure water and analytically pure NaCl.

2.2. Surface analyses

Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray spectrometry (EDS) and contact angle method were used to characterize the characteristic functional groups, compositions and hydrophobicity of the as-formed BP and HP thin films on the iron substrate, respectively. Before each test, the iron plate samples used were firstly ground and cleaned by using the same procedure as for the treatment of the iron electrodes, and then immersed in 0.1 wt% BP (or HP) aqueous solution for 1 h to allow the BP or HP to fully adsorb on the iron substrate, and finally blown dry with ultrapure N₂. As for FTIR measurements, the spectra of absorbance versus wavenumber were scanned from 4000 to 600 cm⁻¹ and recorded on a Bruker TENSOR27 infrared spectrophotometer (BRUKER, Ettlingen, Germany). For XPS measurements, iron plate samples were placed on a copper block and the spectra were collected on PHI5300 spectrometer. The energy scale of spectrometer was calibrated by using an argon-ion etched copper plane and all binding energies were reported with reference to C 1s transition at 284.6 eV. For contact angles tests, each test was performed three times.

2.3. Electrochemical measurements

Electrochemical measurements were carried out in a conventional three-electrode glass cell with bare iron electrode or BP and HP thin film coated electrodes as working electrode, while a saturated calomel electrode (SCE) and a bright platinum plate served as counter and reference electrodes respectively. A 3.5 wt% NaCl aqueous solution was used as the test solution. The reference electrode was led to the surface of the working electrode through a Luggin capillary. The potentials in this paper were referred to the SCE. All electrochemical measurements were carried out at room temperature (22 °C).

Before each electrochemical test, the working electrode was immersed in the test solution for 1 h to attain stable open circuit potential (OCP) and the changes were recorded as a function of exposure time.³⁴ The polarization curves were measured with a CHI 604C electrochemical workstation by scanning the potential with 0.2 mV s⁻¹ going from the cathodic to the anodic side. Electrochemical impedance spectroscopic (EIS) measurements were performed with an ACM electrochemical workstation at respective OCPs with the AC voltage amplitude of ±5 mV in the frequency range from 100 kHz to 50 mHz with eight points per decade.

3. Results and discussion

3.1. Surface tension tests

Surface tension shows a contractive tendency of the surface of a liquid, which allows the liquid to resist an external force. Alkyl phosphate ester is a kind of amphiphilic substance; when it is added to pure water, the hydrophobic group (tail) tends to move upwards from bulk water phase to water surface but the hydrophilic group (head) remains in the water phase, enabling this substance to have the higher concentration near the surface than in the bulk solution. This phenomenon may be reflected well by comparing the surface tension data of pure water and alkyl phosphate-containing aqueous solutions. Surface tension tests for different BP and HP aqueous solutions were done at 22 °C after the solutions were stored for 1 day. As can be seen from Table S1 in ESI,† for the same phosphate ester, the higher the concentration, the lower the surface tension value; but for different phosphate esters, the longer the alkyl chain, the lower the surface tension. A comparison of the surface tension values shows that the surface excess (i.e. the difference between the concentration at air–water interface and that in the bulk water phase) of both HP and BP is positive. And moreover, surface excess concentration of HP is higher than that of BP at the air–water interface under otherwise identical conditions. This means that it was easier to pull BP or HP molecules from the bulk phase and add them to the new surface. As a result, the surface tension was naturally decreased in the presence of both alkyl phosphate esters.

In theory, an amphiphilic substance may be treated as a surfactant when it contains at least 8 carbon atoms in the hydrophobic group. However, on the basis of the conductivity versus concentration curves shown in Fig. S1 (see ESI†), BP did not show any characteristic behavior of a surfactant, while HP behaved like a surfactant to a certain extent although the alkyl chain was a little short. For the HP aqueous solution, an apparent turning point (166 ppm) should correspond to the critical micelle concentration (CMC) of HP, which implies that HP may act as a surfactant despite of the relatively short carbon chain.

The reduction of surface tension of water in the presence of BP and HP mainly depends on surface excess concentration of both alkyl phosphate esters at the air–water interface. Because HP behaved like a surfactant, it is obvious that HP was more inclined to accumulate at the interface as compared to BP. Analyzing the change of surface tension caused by addition of alkyl phosphate esters is very helpful for having a good understanding of how BP and HP spontaneously self-organize into the ordered adsorbed layers on the iron surface in water phase.

3.2. FTIR, EDS and XPS characterizations

As far as whether alkyl phosphates can spontaneously assemble onto the iron surface in water phase, the key problems include (i) adsorption capacity of an alkyl phosphate on the iron surface in aqueous phase and (ii) surface excess concentration of alkyl phosphates near the iron–water interface. When an iron electrode was immersed in an aqueous solution of BP (or HP), the competitive adsorption between water molecules and alkyl phosphate molecules would occur on the iron surface. As stated above, BP can be considered to be an organic inhibitor containing O and P heteroatoms, and its phosphate group may act as the reaction center for the adsorption process of BP on the iron surface. The strong chemical binding interaction between the phosphate group and the iron substrate can help BP molecules dominate over H₂O molecules in the competitive adsorption. Meanwhile, the hydrophobic alkyl chains of adsorbed BP molecules can exclude water molecules from the iron surface. As for HP, its enrichment at solid–water interface has to be considered in addition to the afore-mentioned reasons. The strong tendency of escaping from bulk water phase to the iron–water interface is very helpful for the preferential adsorption of HP molecules on the iron surface.

It is reported that OH-terminated surfaces have an affinity for phosphonic acids (PAs), whose head groups have the similar molecular structure with those of alkyl phosphate esters. In our study, the iron samples were directly ground and polished in atmosphere before they were used, so the iron surface was sure to be oxidized to a certain extent in air. Such iron surface was hydrophilic and would favor the specific adsorption of head groups of alkyl phosphates at the solid–water interface. On the other hand, both BP and HP are acidic substance and can ionize out H⁺ ions even if HP is slightly soluble in water. The electrostatic interaction between Fe²⁺ ions on the surface of iron and RO–PO₃H⁻ (R represents n-butyl or n-hexyl) ions is also advantageous to the binding of alkyl phosphate to the iron surface.

Except for the film-forming problem, what we are more concerned about is whether the as-formed film can exist independently as a physical barrier once the iron samples are taken out of the aqueous solution. Accordingly, after an iron plate sample was immersed in the pure water only containing an alkyl phosphate ester (BP or HP), whether or not the iron sample was covered with the BP or HP thin films was examined by using FTIR and EDS methods.

Fig. 2 shows FTIR spectra of the pure iron samples uncoated and coated with the BP and HP thin films. No characteristic absorption peak was seen from the FTIR spectrum of the unmodified iron sample (a). By contrast, the FTIR spectra of both BP (b) and HP (c) thin film modified iron samples were very similar in outline and clearly showed the characteristic peaks of main functional groups of BP and HP molecules. The methylene stretching frequencies can be considered to be an important diagnostic for the completeness and packing of the alkyl phosphate thin films. Their intensity and position reflect the coverage and ordering of as-formed films.³⁵ Herein for BP and HP thin films, asymmetrical stretching vibrations of methylene groups occurred 2966.80 cm⁻¹ and 2957.91 cm⁻¹ respectively,³⁶ whereas symmetric stretching vibrations happened at 2872.55 cm⁻¹ and 2871.45 cm⁻¹ respectively.³⁶ The scissoring mode of methylene appeared at 1485.71 cm⁻¹ or 1480.19 cm⁻¹.³⁶ Because the wavenumber of asymmetric methylene stretching vibrations is higher than 2922 cm⁻¹, it is inferred that the ordering of the as-formed alkyl phosphate thin films is not high enough.³⁵ At the same time, the coverage of BP film is expected to be higher than that of HP film due to the higher absorption intensity of the characteristic peaks. For phosphate groups, the stretching of H-bond produces the broad shallow bands around 2791.97 cm⁻¹ or 2788.87 cm⁻¹. The characteristic absorption peaks of P=O bonds are localized in the range of 1740–1600 cm⁻¹, 1350–1150 cm⁻¹ and 1090–910 cm⁻¹, so the peaks at 1752.42 cm⁻¹, 1730.08 cm⁻¹, 1672.22 cm⁻¹, 1576.28 cm⁻¹, 1578.73 cm⁻¹, 1310.31 cm⁻¹, 1153.33 cm⁻¹, 1163.26 cm⁻¹, 1061.09 cm⁻¹, 1052.19 cm⁻¹, 943.64 cm⁻¹ and 952.19 cm⁻¹ are caused by stretching of P=O bonds.³⁶

Fig. 2 FTIR spectra of a bare iron plate sample (a) and the BP (b) and HP (c) films modified iron plate samples.

In order to exam the stability of as-formed alkyl phosphate thin films, the iron plates modified with BP and HP layers were used to carry out electrochemical tests, including EIS and polarization curve experiments, followed by FTIR measurements. It is clearly observed form Fig. S2 and S3 (see ESI†) that the former characteristic peaks corresponding to main functional groups of BP and HP still remained after the electrochemical testing although the peak positions shifted to different degrees. This confirms that the adsorbed BP and HP molecules were strongly bound to the iron surface and did not leave the surface even under excitation of a sinusoidal perturbation signal or at different applied DC potentials.

EDS characterization may help us to identify the elemental compositions of the as-formed thin films on the iron substrates. Herein the EDS data was listed in Table 1 for comparison. The EDS spectra for the iron samples covered with BP and HP respectively gave the signals of P, O and C elements in addition to the Fe signals from the substrate. Although the contents of P, O and C elements are not accurate enough, the EDS analyses, together with FTIR spectra and XPS spectra, show the existence of alkyl phosphate esters on the iron substrates, without question.

Table 1 Elemental compositions (wt%) of the iron samples in the absence and presence of BP thin films and HP thin films

Samples	Fe	C	P	O
Iron plate	100
Iron plate with BP films	77.06	15.23	7.41	0.30
Iron plate with HP films	78.12	16.68	4.87	0.33

---

XPS method is employed to further investigate the surface compositions of the iron samples before and after being covered with adsorbed layers and determine the binding mode between alkyl phosphate esters and iron substrate. Fig. 3 shows the high resolution spectra of major elements, C, O and Fe, for the iron samples before modification by the alkyl phosphate esters. The spectra were analyzed through the deconvolution fitting procedure by using XPS Peak-Fit 4.1 software. As shown in Fig. 3, The O 1s core level spectrum can be fitted into three main peaks. The first peak located at 529.8 eV is related to oxygen atoms bonded to ferric oxides,^37,38 and the second peak centered at 531.7 eV is attributed to OH⁻ of hydrous iron oxides such as FeOOH or O–P in the phosphate group such as PO₄³⁻.^37,38 Here, because the iron samples were not coated with alkyl phosphate esters, this peak is mainly attributed to FeOOH. And the last peak located at 533.2 eV is associated with the chemisorbed H₂O. The high resolution Fe 2p_3/2 spectrum shows two peaks. The small peak located at 706.5 eV is attributed to metallic iron.³⁷ The strong peak centered at 711.1 eV is associated with ferric compounds such as Fe₂O₃, FeOOH.^37,39,40 As a comparison, Fig. 4 and 5 show the high resolution spectra of the major elements (C, P, O and Fe) for the iron samples modified by the BP and HP layers respectively. Compare to plain iron samples, the XPS spectra of the iron samples modified with BP and HP show the characteristic peaks of P, which indicates that a certain amount of alkyl phosphate ester (BP or HP) molecules adsorbed onto the iron substrate when the iron plate samples were immersed in the pure water containing BP or HP.³⁷ The C 1s region can be fitted into two peaks. The main peak centered at 284.4 eV is related to C–H or C–C bonds and the other peak at 275.5 eV indicates the presence of carbon impurities in the iron sample. The P 2p core level spectra shown in Fig. 4(b) and 5(b) are fitted into a single peak at 132.8 eV or 132.9 eV corresponding to P–O in phosphate groups of BP and HP,⁴¹ respectively, which suggests the presence of adsorbed layers of BP or HP on the iron samples. The O 1s region is fitted into three main peaks, as shown in Fig. 4(c) and 5(c). The first peak located at 529.7 eV corresponds to O²⁻, which is related to oxygen atoms bonded to ferric oxides.^37,38 The second peak located at 531.7 eV is attributed to OH⁻ of hydrous iron oxides such as FeOOH and O–P in the phosphate group such as PO₄³⁻.^37,38 Because of the formation of the alkyl phosphate esters adsorbed layers, the phosphate group appeared on the iron surface compared to the plain iron sample. The third peak centered at 532.6 eV may be associated with the presence of H₂O.^37,38 The Fe 2p_3/2 peak is fitted into two peaks, as shown in Fig. 4(d) and 5(d). The large broad peak at lower binding energy 711.1 eV is associated with ferric compounds such as Fe₂O₃, FeOOH^37,39,40 and iron phosphates,³⁷ and the small peak located at 714.4 eV may be attributed to the satellite of Fe(III).³⁷ The peak located at 706.5 eV disappeared in Fig. 4(d) and 5(d), which implies that there is no metallic iron on the surface of iron sample after modification with the BP and HP layers.

Fig. 3 High resolution XPS spectra of major elements on the surface of plain iron plate sample: (a) C 1s; (b) O 1s; (c) Fe 2p.

Fig. 4 High resolution XPS spectra of major elements on the iron surface modified with BP layers: (a) C 1s; (b) P 2p; (c) O 1s; (d) Fe 2p_3/2.

Fig. 5 High resolution XPS spectra of major elements on the iron surface modified with HP layers: (a) C 1s; (b) P 2p; (c) O 1s; (d) Fe 2p_3/2.

On the basis of fitted XPS spectra, the peak positions and the corresponding atomic ratios of the most intense core peaks of BP and HP are listed in Tables 2 and 3 respectively. The peak area ratios were used to calculate atomic ratios through Scofield cross sections³⁸ and the instrument transmission factor by assuming a homogeneous sample. The similar outline of XPS spectra and the peak positions for the iron samples modified by BP and HP layers indicate that the formation types of the two alkyl phosphate ester layers on the iron surface are similar. They both assemble onto the iron substrate via electrostatic interaction. But as shown in Tables 2 and 3, the different P atomic ratios for the iron samples modified with BP and HP indicates that the compositions of the two adsorbed layers are different. The higher P atomic ratio of BP layer indicates that the adsorption ability of BP is stronger than that of HP. This may be an important reason why the protection efficiency of BP layers is higher than that of HP layers although its carbon chain is shorter.

Table 2 The peak positions and atomic ratios obtained by fitting XPS spectra shown in Fig. 4

Element	C 1s		P 2p		O 1s		Fe 2p
Position (eV)	275.5	284.4	132.8	529.7	531.7	532.6	711.1	714.4
Ratio	0.0454	0.624	0.0223	0.122	0.0644	0.0913	0.0267	0.00342

---

Table 3 The peak positions and atomic ratios obtained by fitting XPS spectra shown in Fig. 5

Element	C 1s		P 2p		O 1s		Fe 2p
Position (eV)	275.5	284.4	132.9	529.7	531.7	532.6	711.1	714.4
Ratio	0.0659	0.718	0.0179	0.0719	0.0431	0.0758	0.00638	0.000479

---

3.3. Contact angle tests

Hydrophobicity is an important index to evaluate the potential application of the as-formed alkyl phosphate thin films as corrosion protection layers. In this study, the contact angles of water droplets on the iron plates unmodified and modified with alkyl phosphate thin films were measured. Prior to the measurements, the iron plate samples were immersed in 0.1 wt% BP and HP solutions for 1 h at the room temperature respectively, allowing both phosphates to form adsorbed layers on the iron plate surfaces. Each test was repeated three times, and the data averaged. Fig. 6 shows the representative photographs describing water droplets on the untreated and treated iron surfaces. The results measured by contact angle method were collected in Table 4.

Fig. 6 Photographs describing water droplets on the bare iron surface (a) and the iron surfaces modified with BP (b) and HP (c) thin films.

Table 4 Water contact angles on the iron surfaces unmodified and modified with BP and HP thin films

	1	2	3	Average value
Pure iron	44.5°	40.2°	38.0°	40.9°
Iron plate with BP films	90.6°	90.0°	90.7°	90.4°
Iron plate with HP films	90.4°	90.0°	92.5°	91.0°

---

According to the results indicated in Fig. 6 and Table 4, the bare iron surface is hydrophilic since the water contact angle is about 40.9°, whereas the iron surfaces modified with two phosphate thin films show a tendency changing from hydrophilic to hydrophobic since the water contact angles are over 90°. The hydrophilicity is beneficial to attachment of polar functional groups to the iron surface. The change of the water contact angles not only confirms the formation of alkyl phosphate thin films on the iron substrates but also provides favorable evidence that the hydrophobic parts (tail) of phosphates are pointed outward. Moreover, the phosphate thin films may behave as a barrier to block the corrosive media from contacting the metal substrates. It is worthy to note that the water contact angle is not entirely dependent on the chain length of alkyl phosphate molecules. Theoretically, the carbon chain of an amphiphilic molecule is longer, the hydrophobicity is stronger, and correspondingly the water contact angle should be larger. But in fact, the water contact angles on the iron surfaces modified with both BP and HP thin films almost have the same values, which in turn demonstrates that the hydrophobicity depends not only on the assembled molecules but also on coverage, ordering, number and size of defects, and other properties of the as-formed thin films.

3.4. Electrochemical characterizations

3.4.1. EIS measurements. The BP thin films were prepared on the iron surfaces by directly immersing iron electrodes in 0.1 wt% BP solutions for 10 min and 30 min respectively. The same was true of the preparation of HP thin films. In order to evaluate the corrosion resistance of the phosphate thin films in common corrosive media, the 3.5 wt% NaCl aqueous solution was selected as the corrosive medium. As a contrast test, the corrosion behavior of the unmodified iron electrode in the NaCl solution was also investigated. EIS measurements were performed in potentiostatic mode at the open circuit potentials (OCPs). The changing trends of OCP values of unmodified and modified iron electrodes with immersion time in 3.5 wt% NaCl solutions were indicated in Fig. S4 (see Section S2 of ESI†).

Nyquist spectrum of the bare iron electrode shows a capacitive loop in high frequency and a Warburg line in low frequency, as seen in Fig. 7(a). The capacitive loop is attributed to the relaxation time constant caused by the coupling of the charge-transfer resistance (R_ct) and the double-layer capacitance (C_dl). It seems to be a depressed semi-circle rather than a regular one because of the common “dispersing effect”.^42,43 In this case, the electric double layer does not function as a pure capacitor but a constant phase element (CPE), whose admittance and impedance are, respectively, defined as

Y_CPE = Y₀(jω)ⁿ (1)

andwhere Y₀ is the magnitude and n the exponential term.^42–44 CPE has been widely used to account for the deviations brought about by surface roughness of electrodes.^42,43,45 The appearance of Warburg impedance indicates that the corrosion reaction of iron was strongly influenced by the transport of dissolved oxygen from the bulk solution to the electrode surface since the oxygen was depolarization agent in the corrosive medium.⁴⁶ In contrast, the Nyquist spectra for the iron electrodes covered with BP and HP thin films only gave a segment of circular arc rather than a perfect semi-circle. The disappearance of Warburg impedance and the significant increase in the diameter of capacitive loop fully reflect that the as-formed phosphate thin films were able to act as a protective barrier against corrosion attack from the aggressive substances, including dissolved oxygen, Cl⁻ ions and water molecules. However, a closer examination reveals an unusual phenomenon that the anticorrosion property of the phosphate films did not increase monotonously with increasing the chain length of self-assembled molecules or prolonging the film-forming time. For the BP films, the immersion time was longer, the corrosion resistance effectiveness was better. But for the HP films, increasing the immersing time was not conductive to the enhancement of corrosion resistance. A typical example is that the HP films prepared in the case of 30 min immersion showed the lower corrosion resistance ability than the same films prepared in the case of 10 min immersion. It is important to note that, the corrosion resistance of the HP films obtained after 30 min of immersion was even almost the same with that of the BP films under the identical conditions, although HP had the longer carbon chain than BP. This means that the corrosion resistance ability of the as-formed alkyl phosphate thin films is more dependent on their effective coverage on the whole iron surface. Because the adsorption mode of BP is different from that of HP on the iron surface, the structure of the BP thin films will be distinct from that of HP thin films, which has in turn produced influence on the corrosion resistance properties of two alkyl phosphate films. The detailed reasons will be analyzed further in the sections that follow.

Fig. 7 Nyquist plots (a) and Bode phase plots (b) of the iron electrodes uncovered and covered with BP and HP thin films in 3.5 wt% NaCl aqueous solution at the respective OCPs.

In order to determine how many time constants the above-mentioned capacitive loop contains more accurately, the obtained impedance data are also presented in the form of Bode plots, as shown in Fig. 7(b). Considering that each Bode phase plot displays a relatively regular peak, it is confirmed that each capacitive loop shown in complex plane contains one time constant. For the bare iron electrode, it is generally believed that the corrosion reaction takes place uniformly on the whole electrode surface. Thus, the impedance spectrum for the bare iron electrode can be analyzed by the equivalent circuit shown in Fig. 8(a),⁴⁶ where R_s stands for the solution resistance, CPE_dl represents a CPE used to model the double-layer capacitance, W the Warburg impedance, and R_ct the charge-transfer resistance. But for the BP or HP thin film modified iron electrodes, it is reasonable to assume that the corrosion reaction only occurred at those regions that were not effectively covered by the alkyl phosphate ester molecules. Based on the assumption, a reaction model which describes the corrosion reaction taking place on the electrode surface partially covered with a corrosion protective layer is established and shown in Fig. 8(b). In fact, this circuit is essentially equivalent to the equivalent circuit given in Fig. 8(c).^47,48 Here R_s and R_ct elements have the same physical meaning as what they represent in the circuit shown by Fig. 8(a), but the CPE′_dl element contains the contributions from the double-layer capacitance (C_dl) and the capacitance (C_f) of BP (or HP) thin films. The values of the above elements were obtained by fitting the impedance data and were listed in Table 5.

Fig. 8 Equivalent circuits used to fit the EIS spectra for the bare iron electrode (a) and the iron electrodes covered with alkyl phosphate protective layers (c). A model describing the corrosion reaction on the iron surface partially covered with corrosion protective layers (b).

Table 5 Values of elements of equivalent circuit obtained by fitting the impedance spectra shown in Fig. 7 and protection efficiency (η) values under different conditions

Electrodes	Immersion time	R s (Ω cm^2)	CPE′dl		R ct (Ω cm^2)	W (Ω cm^2)	η (%)
			Y 0 (Ω^−1 cm^−2 s^n)	n
Pure Fe	—	1.500	0.002336	0.7257	211.8	0.01064	—
Modified with BP	10 min	1.768	0.00158	0.7359	558.2	—	62.1
Modified with BP	30 min	2.033	0.001571	0.7803	821.8	—	75.1
Modified with HP	10 min	1.908	0.001150	0.8036	888.4	—	76.2
Modified with HP	30 min	1.905	0.001670	0.7869	676.0	—	68.7

---

The protection efficiency of BP and HP thin films can be determined using the following formula

where η represents protection efficiency, R_ct and R′_ct are the values of charge-transfer resistance for the iron electrodes uncovered and covered with alkyl phosphate thin films, respectively. The η values of BP and HP layers in all case are given in Table 5. It is seen that, the longer film-forming time helped to improve the protection efficiency of the BP layers but made that of the HP layers decrease obviously. On the basis of previous assumption that the corrosion reaction only occurs on uncovered regions, the coverage of the BP or HP layers on the iron substrate should be approximately equal to the η value. The obtained η values reflect such a fact that BP and HP molecules were able to form the self-assembled layers on the iron substrate in water phase; moreover, the as-formed the layers worked as the stable isolation layers that prevented the aggressive species from the iron substrate. At the same time, the changes of Y₀ values of CPE′_dl before and after modification with BP or HP films also reveal that the partial coverage of the alkyl phosphate thin films on the iron surface reduced the double-layer capacitance of the bare iron surface.

3.4.2. Polarization curves. The corrosion protection ability of the two phosphate thin films prepared under different conditions was further evaluated by means of polarization curves. At first, the influence of the film-forming time on the corrosion resistance of the films was investigated. The BP and HP thin films were formed by immersing iron electrodes in the aqueous solutions containing 0.1 wt% BP or HP for the different time (10 min, 30 min and 1 h). Fig. 9(a) shows the polarization curves of the iron electrodes untreated and treated with BP thin films in 3.5 wt% NaCl solutions. All cathodic curves show a current plateau in a wide potential region, especially when the iron electrode was coated with the BP and HP thin films. The appearance of the current plateau should be related to the mass transport of oxygen species, because the electrochemical reduction of oxygen (i.e. depolarization reaction) in the corrosive medium is usually diffusion controlled. The result also supports the previous inference about the origin of the Warburg impedance at the bare iron electrode. The BP thin films affected the corrosion behavior of iron substrate in the following ways: (i) making corrosion potential shift to the positive direction, (ii) changing the Tafel slope of anodic polarization processes to varying degrees, and (iii) decreasing the corrosion current densities to different extents depending on the film-forming time. The impact trend of BP thin films on the corrosion resistance is that, the longer the film-forming time, the better the thin films protect the iron substrate from being corroded.

Fig. 9 Polarization curves for the bare iron electrode and the iron electrodes modified with BP and HP thin films in 3.5 wt% NaCl solution. The BP or HP thin films were formed on the iron surfaces by immersing iron electrodes in the aqueous solutions that contained 0.1 wt% BP or HP for 10 min, 30 min and 1 h respectively.

In strong contrast to the BP thin films, the HP thin films display the lower protection performance to the corrosion of the iron substrate (see Fig. 9(b)), although the presence of the thin films also shifted the corrosion potential to the positive direction. It is interesting that, the longer the film-filming time, the worse the corrosion resistance of the as-formed thin films.

Next, how the concentration of the film-forming substances (i.e. BP and HP) affected the corrosion protection property was investigated. The BP and HP thin films were assembled on the iron surfaces by immersing iron electrodes in aqueous solutions containing 0.001 wt%, 0.01 wt% and 0.1 wt% respectively for a fixed period of time (1 h), and then the corrosion behavior of the bare electrodes and modified electrodes was compared under the identical conditions. For the BP thin films, the corrosion protection ability was gradually enhanced with increasing the concentration of BP in the assembling system, as indicated in Fig. 10(a). However, for the HP thin films, increasing the concentration of HP in the assembling system caused an adverse impact on the corresponding corrosion protection ability, as seen in Fig. 10(b).

Fig. 10 Polarization curves of the iron electrodes uncovered and covered with BP and HP thin films in 3.5 wt% NaCl solution. The BP or HP thin films were formed on the iron surfaces by immersing iron electrodes in the aqueous solutions that contained 0.001 wt%, 0.01 wt% and 0.1 wt% BP or HP for 1 h respectively.

The η values of two alkyl phosphate thin films can also be calculated as follows⁴⁹

where I_corr and I′_corr are the corrosion current densities for the iron electrodes unmodified and modified with the protection layers, respectively. Due to the strong impact of diffusion on the cathodic processes, the values of I_corr and I′_corr can not be simply determined by following the conventional Tafel extrapolation method. When a cathodic reaction was under control of diffusion, the extrapolation of anodic Tafel straight line and cathodic current plateau line produced an intersection. The current value corresponding to the intersection point was considered to be I_corr or I′_corr. In this way, the electrochemical parameters calculated from the polarization curves shown in Fig. 9 and 10, including I_corr and I′_corr, anodic Tafel slopes (b_a) and cathodic Tafel slope (b_c), together with η values in each case, were collected in Tables 6 and 7 respectively. It should be pointed out that the η values obtained using eqn (4) are basically in agreement with those obtained by eqn (3) in the range of experimental errors.

Table 6 Electrochemical parameters for the corrosion process obtained from the polarization curves shown in Fig. 9

Electrodes	Immersion time	b a (mV dec^−1)	−bc (mV dec^−1)	I corr (μA cm^−2)	η (%)
Pure Fe	—	81.08	∞	144.01	0
Modified with BP	10 min	76.16	∞	59.07	59.0
Modified with BP	30 min	45.37	∞	43.18	70.0
Modified with BP	1 h	44.34	217.53	10.68	92.6
Modified with HP	10 min	46.22	∞	25.42	82.3
Modified with HP	30 min	50.00	∞	48.69	66.2
Modified with HP	1 h	40.03	∞	60.10	58.3

---

Table 7 Electrochemical parameters for the corrosion process obtained from the polarization curves shown in Fig. 10

Film-forming solutions	Immersion time	b a (mV dec^−1)	−bc (mV dec^−1)	I corr (μA cm^−2)	η (%)
0.1 wt% BP	1 h	44.34	217.53	10.68	92.6
0.01 wt% BP	1 h	46.07	∞	37.90	73.7
0.001 wt% BP	1 h	38.86	∞	55.73	61.3
0.1 wt% HP	1 h	40.03	∞	60.10	58.3
0.01 wt% HP	1 h	52.89	∞	43.60	69.7
0.001 wt% HP	1 h	35.81	∞	41.94	70.9

---

3.4.3. Interpretation of film formation processes. Theoretically, the structure of alkyl phosphate thin films should be similar to that of common SAMs such as alkanethiol SAMs. The head groups bind to the substrate surface, functional tail groups are exposed at the film surfaces, and spacer chains separate the head and tail groups and, in some cases, drive the self-assembly by lateral interactions between adjacent molecules.

Herein the driving force for the film formation is that the difference in the concentration of alkyl phosphate between near the solid (iron) surface and in the bulk solution is positive. At the same time, the iron substrate's affinity for the head groups of alkyl phosphate esters plays an important role in the film formation process. Otherwise, alkyl phosphate molecules are unable to displace the water molecules adsorbed on the iron surface, let alone spontaneous formation of the organized adsorbed layers. Based on the EIS and polarization curve results, especially the changing trends of the corrosion protection ability with the self-assembling time and the concentration of film-forming substances, we speculate that the formation processes of two alkyl phosphate thin films are similar but not identical.

Due to the shorter carbon chain, BP has the better water-solubility and hydrophilicity than HP. According to the surface tension results (see Table S1 in ESI†), HP molecules have the stronger trend to escape from the bulk water phase and accumulate at the air–water interface than BP molecules. This implies that, as compared to HP molecules, the transfer of the same number of BP molecules from the bulk solution to the air–water interface needs the longer time under otherwise equal conditions. In other words, the formation speed of BP films at solid–water interface should be slower than that of HP films. Besides, the short carbon chain is disadvantageous to excluding water molecules from the solid–water interface. Thus, the longer immersion time and the higher concentration of film-forming substance in assembling systems enabled more BP molecules to adsorb on the iron surface, thereby forming the adsorbed layers with fewer defects, as illustrated by the schematic diagram shown in Fig. 11(a).

Fig. 11 Schematic diagrams for the formation process of BP (a) and HP (b and c) thin films in water phase.

HP molecules tend to accumulate at air–water interface because of the longer carbon chain. When the iron electrode was immersed in the HP-containing aqueous solution, a number of HP molecules would accumulate at the electrode–water interface in a short time, which brings a favorable factor to prepare the HP layers on the iron surface. However, it is possible that some HP molecules would condense into droplets like liquid lens at the solid–water interface (a common phenomenon taking placing at liquid–liquid interface) with the prolonging of time, due to the relatively low solubility in water. This situation is more likely to occur when the HP concentration is relatively high. Besides, the CMC of HP is about 166 ppm (see Fig. S1 in ESI†). This means that there are a certain amount of micelles in the 0.01 wt% or 0.1 wt% HP aqueous solutions. The appearance of the micelles, especially at the solid–water interface, has an adverse impact on the ordering of the HP self-assembled films on the iron substrate. The above two adverse factors hampered the ordered assembly of HP layers at the electrode–water interface, which can be reflected from the corresponding electrochemical results that the corrosion protection ability of HP thin film decreased when the immersion time became longer or the concentration of film-forming substance became higher. The formation process of HP layers on the iron surface under this condition may be explained by another schematic diagram that is shown in Fig. 11(b). Interestingly, it is relatively easy to obtain high quality HP thin films with excellent corrosion protection ability in the aqueous solution containing small amount of HP (e.g. 0.001 wt%). An important reason is that HP molecules exist in the form of monomers when the concentration of HP is much smaller than its CMC. Because HP behaves as both surfactant and inhibitor, HP molecules are able to adsorb onto the iron surface through specific adsorption, forming a relatively compact and ordered layer in a shorter time as compared to BP molecules. The formation process of HP layers in such a situation can be described using a schematic diagram given in Fig. 11(c), which is similar to that shown in Fig. 11(a). It should also be noted that there still exists the possibility of aggregation between HP molecules caused by the rearrangement of the adsorbed HP molecules when the self-assembling time is long. Thus, choosing the appropriate film formation time and using the aqueous solutions containing a low concentration of HP are very necessary to reduce the amount of defects in the HP thin films and ensure that as-formed thin films possess high corrosion protection ability.

In short, we have demonstrated that BP and HP may spontaneously form the relatively ordered adsorbed layers on the iron surface in water phase. Moreover, the as-formed BP and HP thin films can effectively protect the iron substrates from corrosion in NaCl solutions. The film formation process is driven by (i) the concentration difference of BP or HP between at the solid–water interface and in the bulk solution and (ii) the specific adsorption of BP or HP via binding of head groups to iron surface. Main advantages of the film formation method include spontaneous assembly of amphiphilic inhibitors in water phase, convenient surface pretreatment of iron samples in air, and especially future applications on an industrial scale.

4. Conclusion

A simple but effective film-forming method was developed to assemble the alkyl phosphate protection layers directly on the iron substrate in water phase by taking full advantage of the surface excess of BP and HP at solid–water interface and specific adsorption of head groups of BP or HP molecules on the iron substrate. The formation of BP and HP thin films was fully demonstrated by FTIR, EDS, XPS and contact angle results. The as-formed alkyl phosphate layers show the strong corrosion protection ability to the iron substrate. For BP with short carbon chain, the long film-forming time and high concentrations are conducive to the enhancement of the protection efficiency of BP films. In contrast, for HP with long carbon chain, due to the lower solubility in water and the stronger trend to accumulate at the solid–water interface, it is easier to assemble HP layers with high corrosion resistance on the iron substrate in aqueous solution with low HP content. The practical film-forming method is expected to find wide applications in the corrosion protection of metal and alloys.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21373129) and the Fundamental Research Funds of Shandong University (2014YQ004).

References

1. J. L. Yao, B. Ren, Z. F. Huang, P. G. Cao, R. A. Gu and Z. Q. Tian, Electrochim. Acta, 2003, 48, 1263–1271.
2. S. Pour-Alia, C. Dehghanian and A. Kosari, Corros. Sci., 2014, 85, 204–214.
3. M. Kohl, A. Kalendová and J. Stejskal, Prog. Org. Coat., 2014, 77, 512–517.
4. B. Nikravesh, B. Ramezanzadeh, A. A. Sarabi and S. M. Kasiriha, Corros. Sci., 2011, 53, 1592–1603.
5. L. K. Wu, X. F. Zhang and J. M. Hu, Corros. Sci., 2014, 85, 482–487.
6. G. A. Zhang, N. Yu, L. Y. Yang and X. P. Guo, Corros. Sci., 2014, 86, 202–212.
7. H. N. T. Le, B. Garcia, C. Deslouis and Q. L. Xuan, Electrochim. Acta, 2001, 46, 4259–4272.
8. A. Amadeh, B. Pahlevani and S. Heshmati-Manesh, Corros. Sci., 2002, 44, 2321–2331.
9. J. G. Kim and Y. W. Kim, Corros. Sci., 2001, 43, 2011–2021.
10. Y. Ein-Eli, M. Auinat and D. Stearosvetsky, J. Power Sources, 2003, 114, 330–337.
11. T. Shimura and K. Aramki, Corros. Sci., 2008, 50, 2407–2414.
12. A. R. El-Sayed, U. Harm, K. M. Mangold and W. Fürbeth, Corros. Sci., 2012, 55, 339–350.
13. H. Y. Ma, C. Yang, B. S. Yin, G. Y. Li, S. H. Chen and J. L. Luo, Appl. Surf. Sci., 2003, 218, 143–153.
14. J. C. Love, L. A. Estroff, J. K. Kriebel, R. G. Nuzzo and G. M. Whitesides, Chem. Rev., 2005, 105, 1103–1169.
15. H. Y. Ma, C. Yang, S. H. Chen, Y. L. Jiao, S. X. Huang, D. G. Li and J. L. Luo, Electrochim. Acta, 2003, 48, 4277–4289.
16. S. Onclin, B. J. Ravoo and D. N. Reinhoudt, Angew. Chem., Int. Ed., 2005, 44, 6282–6304.
17. R. Hofer, M. Textor and N. D. Spencer, Langmuir, 2001, 17, 4014–4020.
18. H. T. Rong, S. Frey, Y. J. Yang, M. Zharnikov, M. Buck, M. Wühn, C. Wöll and G. Helmchen, Langmuir, 2001, 17, 1582–1593.
19. S. Q. Sun and G. J. Leggett, Nano Lett., 2007, 7, 3753–3758.
20. A. M. Mahmoud, T. Tang, D. Jed Harrisona, W. E. Lee and A. B. Jemere, Biosens. Bioelectron., 2014, 56, 328–333.
21. E. S. M. Sherif, R. M. Erasmus and J. D. Comins, Electrochim. Acta, 2010, 55, 3657–3663.
22. S. Ramesha, S. Rajeswari and S. Maruthamuthu, Appl. Surf. Sci., 2004, 229, 214–225.
23. Z. Zhang, S. H. Chen, Y. H. Li, S. H. Li and L. Wang, Corros. Sci., 2009, 51, 291–300.
24. T. Ishizaki, R. Kudo, T. Omi, K. Teshim, T. Sonod, I. Shigematsu and M. Sakamoto, Mater. Lett., 2012, 68, 122–125.
25. T. Hauffman, O. Blajiev, J. Snauwaert, C. van Haesendonck, A. Hubin and H. Terryn, Langmuir, 2008, 24, 13450–13456.
26. B. M. Silverman, K. A. Wieghaus and J. Schwartz, Langmuir, 2005, 21, 225–228.
27. G. Zorn, I. Gotman, E. Y. Gutmanas, R. Adadi, G. Salitra and C. N. Sukenik, Chem. Mater., 2005, 17, 4218–4226.
28. E. S. Gawalt, G. Lu, S. L. Bernasek and J. Schwartz, Langmuir, 1999, 15, 8929–8933.
29. O. Yildirim, M. D. Yilmaz, D. N. Reinhoudt, D. H. A. Blank, G. Rijnders and J. Huskens, Langmuir, 2011, 27, 9890–9894.
30. B. Branch, M. Dubey, A. S. Anderson, K. Artyushkova, J. K. Baldwin, D. Petsev and A. M. Dattelbaum, Appl. Surf. Sci., 2014, 288, 98–108.
31. M. Dubey, I. Gouzman, S. L. Bernasek and J. Schwartz, Langmuir, 2006, 22, 4649–4653.
32. I. Gouzman, M. Dubey, M. D. Carolus, J. Schwartz and S. L. Bernasek, Surf. Sci., 2006, 600, 773–781.
33. R. Bhure, Mater. Sci. Eng., C, 2013, 33, 2050–2058.
34. C. X. Li, L. Li, C. Wang, Y. Y. Zhu and W. C. Zhang, Corros. Sci., 2014, 80, 511–516.
35. G. Zorn, I. Gotman, E. Y. Gutmanas, R. Adadi, G. Salitra and C. N. Sukenik, Chem. Mater., 2005, 17, 4218–4226.
36. J. G. Speight, Lange's Handbook of Chemistry, McGraw-Hill Co. Inc, New York, NY, 16th edn, 2005.
37. H. Zarrok, A. Zarrouk, B. Hammouti, R. Salghi, C. Jama and F. Bentiss, Corros. Sci., 2012, 64, 243–252.
38. K. Babić-Samardžija, C. Lupu, N. Hackerman, A. R. Barron and A. Luttge, Langmuir, 2005, 21, 12187–12196.
39. G. Gunasekaran and L. R. Chauhan, Electrochim. Acta, 2004, 49, 4387–4395.
40. A. Galtayries, R. Warocquier-Clérout, M. D. Nagel and P. Marcus, Surf. Interface Anal., 2006, 38, 186–190.
41. X. F. Cui, Q. F. Li, Y. Li, F. H. Wang, G. Jin and M. H. Ding, Appl. Surf. Sci., 2008, 255, 2098–2103.
42. X. J. Wu, H. Y. Ma, S. H. Chen, Z. Y. Xu and A. F. Sui, J. Electrochem. Soc., 1999, 146, 1847–1853.
43. H. Y. Ma, X. L. Cheng, G. G. Li, S. H. Chen, Z. L. Quan, S. Y. Zhao and L. Niu, Corros. Sci., 2000, 42, 1669–1683.
44. H. Y. Ma, C. Yang, S. H. Chen, Y. L. Jiao, S. X. Huang, D. G. Li and J. L. Luo, Electrochim. Acta, 2003, 48, 4277–4289.
45. M. Ehteshamzade, T. Shahrabi and M. G. Hosseini, Appl. Surf. Sci., 2006, 252, 2949–2959.
46. W. J. Guo, S. H. Chen, B. D. Huang, H. Y. Ma and X. G. Yang, Electrochim. Acta, 2006, 52, 108–113.
47. C. L. Zhou, X. Lu, Z. Xin, J. Liu and Y. F. Zhang, Corros. Sci., 2014, 80, 269–275.
48. T. C. Huang, Y. A. Su, T. C. Yeh, H. Y. Huang, C. P. Wu, K. Y. Huang, Y. C. Chou, J. M. Yeh and Y. Wei, Electrochim. Acta, 2011, 56, 6142–6149.
49. H. Zarrok, A. Zarrouk, B. Hammouti, R. Salghi, C. Jama and F. Bentiss, Corros. Sci., 2012, 64, 243–252.

---

Footnote

† Electronic supplementary information (ESI) available: Additional experimental data. See DOI: 10.1039/c5ra03899e

---