3.1. Mild Steel Corrosion Inhibition Properties of p-Hexoxy Coumarate Based Monomeric Ionic Liquids
In order to investigate the effect of the cation on the corrosion inhibition properties of the coumarate ionic liquids, we selected four different common cationic monomers having ammonium, imidazolium, pyridinium and anilinium cations. The protic ionic liquid monomers were easily obtained by acid-base proton exchange reaction between commercially available monomers and p-hexoxy coumaric acid as described in Scheme 2
shown in the experimental section. The chemical structure of the four monomers investigated in this work is shown in Figure 1
, [DMAEM+ HexOCou-] [VIm+ HexOCou-] [VAn+ HexOCou-] and [VPy+ HexCou-]. First, the corrosion inhibition properties of the monomers were evaluated by immersing mild steel AS1020 foils into an aqueous solution of the ionic monomers. By this method, it is expected that the organic ionic compounds may adsorb onto the mild steel surface forming a corrosion inhibition layer, as illustrated in the Figure 1
Potentiodynamic polarization scans of AS1020 mild steel after an exposure of 24 h in 0.01 M NaCl control solution and inhibitors containing solutions (0.01 M NaCl + 8 mM inhibitor monomers) are shown in Figure 2
. Corrosion potentials (Ecorr
), corrosion current density (icorr
), tafel anodic and cathodic slopes (βa
) calculated using Tafel extrapolation are displayed in Figure 2
. It can be observed that, for all the inhibitors, the corrosion potential (Ecorr
) is shifted towards a more positive value compared with the control. The Ecorr
of the control is at −523 mV whereas the inhibitors present potentials ranging from −116 to −543 mV, meaning that all the inhibitors are mainly affecting and suppressing the anodic reaction of the corrosion. The potentiodynamic polarization curves indicate that all compounds acted as anodic inhibitors. The addition of inhibitors to the control solution shifted the corrosion potentials (Ecorr
) to more positive values and significantly reduced the anodic current density [13
]. Anodic inhibitors block the anodic reaction (oxidation of Fe to Fe2+
) by a creation of a barrier coating on anodic sites. They are adsorbed on the metal surface forming a protective film, so reducing the corrosion current and increasing the corrosion potential [13
]. However, differences in both the anodic and cathodic Tafel slopes can be observed, suggesting that, apart from blocking the anodic reaction, the cathodic reaction is also affected by the presence of these compounds. Moreover, the potentiodynamic polarization results also show significant current fluctuations. Those fluctuations refer with metastable pitting, in which there is a constant breakdown and repair of a passive film [1
]. In most cases, the corrosion current decreased considerably compared with the control, yielding very high inhibitor efficiency values. [DMAEM+ HexOCou-]), showed a corrosion current density (icorr
) of 0.008 μA/cm2
and an inhibitor efficiency of 99.1%. Finally, the vinylic compounds, [VIm+ HexOCou-], [VAn+ HexOCou-], and [VPy+ HexOCou-], produced corrosion current densities (icorr
) of 0.020, 0.470, and 0.267 μA/cm2
and acceptable inhibitor efficiencies of 97.8%, 49.2% and 71.1%, respectively. A dramatic effect of the cationic component of the ionic liquid in the corrosion inhibition process is observed. Among the different monomers the inhibition efficiency was [DMAEM+ HexOCou-] > [VIm+ HexOCou-] > [VPy+ HexCou-] >> [VAn+ HexOCou-]. So, the best results were obtained for the compound containing ammonium derivative cation, followed by the one containing imidazolium and pyridinium cations. Finally, the inhibitors containing anilinium cation showed icorr
almost unchanged relative to the control even though Ecorr
was shifted considerably.
Electrochemical Impedance Spectroscopy experiments were carried out in order to further characterize the anticorrosive capacity of different inhibitors. The impedance responses were measured during an immersion in NaCl 0.01 M aqueous solution for 24 h. Figure 3
shows the Nyquist plot at the beginning of the test and after 24 h of immersion in NaCl 0.01 M aqueous solution. The [DMAEM+ HexOCou-] inhibitor presents the largest impedance consistent with the PP data above and reflecting a stronger interaction between the inhibitor and mild steel surface. Moreover, it can be observed that after 24 h of immersion the anticorrosive capacity of [DMAEM+ HexOCou-] inhibitor is not decreased. On the other hand, varying the structure of the inhibitors, a difference in the magnitude of the impedance can be seen depending on the cationic part. [VIm+ HexOCou-] and [VPy+ HexOCou-] inhibitors present a bigger capacitive loop than the control although significantly smaller than for the DMAEM compound. In the case of the [VAn+ HexOCou-] inhibitor, the initial EIS meaurement is similar to the control, but at the end of the 24 h experiment, the impedance is even a little lower than the control suggesting poorer corrosion resistance of the mild steel under these conditions. This data is fully consistent with the PP data above and suggests a. co-dependence of the anion and cation on the adsorption on steel from aqueous solution and hence a significant effect on the corrosion performance.
The Bode plots of the control and the comparison with the different inhibitors as a function of immersion time are shown in the Figure 4
. Due to corrosion reaction occurring on the control sample, in the low frequency range, a decrease in the impedance and the phase angle plateau can be observed, passing from 103.48
and 35° at 2 h of immersion to 103.34
and 30° at 24 h of immersion. On the other hand, steel surfaces immersed in the inhibitors shows different anticorrosive responses. Specifically, the mild steel sample which was immersed in the ammonium coumarate [DMAEM+ HexOCou-] monomer presents an increase on the impedance obtaining 105.64
at the low frequency range. In the corresponding phase angle of the time constant a plateau is observed at 74°, showing that the inhibitor results in predominantly capacitive behavior and correlates with the improved anticorrosive performance. The imidazolium, anilinium, and pyridinium monomers were also compared using the Bode formalism as discussed below.
Mild steel, after an immersion of 2 h in the imidazolium monomer [VIm+ HexOCou-] solution, presented an impedance of 105.42
, although this value decreased to 104.54
after 24 h. The corresponding phase angle at the low frequency range also decreased with time from 61° to 53°. It is worth noting that, upon initial exposure, this inhibitor blocks the corrosion reaction almost as effectively as the DMAEM inhibitor, although the corrosion inhibition effectiveness was lost over time. On the other hand, the vinyl anilinium based monomer shows no effectiveness as an inhibitor which is very surprising given the fact that the anion is common and is expected to be the inhibiting moiety. The Bode plots for the pyridinium monomer show an impedance of 104.50
after the first 2 h whereas after 24 h this value is also decreased to 104.26
with a phase angle of 45°. Optical microscopy images were taken in order to correlate with the EIS results. Figure 4
shows optical images of the samples immersed in control, [DMAEM+ HexOCou-], [VIm+ HexOCou-], [VAn+ HexOCou-], and [VPy+ HexOCou-]. The control sample presents a surface covered by red rust, while the inhibitor immersed samples are not showing evidence of such corroded areas.
In order to corroborate this trend and the corrosion inhibition effect we evaluated all the samples using electron microscopy of the surfaces. Mild steel AS1020 surfaces were analyzed after an immersion of 24 h in 0.01 M NaCl with and without monomeric ionic liquid inhibitors by optical microscopy, scanning electron microscopy and electron diffraction spectroscopy. In Figure 5
, rust deposits can be observed on the surface immersed in the control solution (optical and scanning electron microscopy images). The EDS data (Supplementary Figure S1
) confirms that those precipitates are mainly iron oxide. As it can be seen in the optical images of all samples, the surfaces in contact with solutions containing inhibitors do not present rust deposits although pitting is still evident to different extents in these inhibited samples. EDS analysis confirmed the presence of carbon, oxygen, and nitrogen atoms on these surfaces, indicating the creation of an organic inhibiting layer onto the metallic surface. The surfaces exposed to DMAEM and VIm hexoxy coumarate show the least corrosive attack, once again consistent with the electrochemistry presented above. Thus, it can be concluded that both the p-hexoxy coumarate anion and the cationic monomer have an important effect in the corrosion inhibition phenomena.
3.2. Covalent Incorporation of p-Hexoxy Coumarate Ionic Liquid Monomers into Acrylic Polymer Coatings by UV-Photopolymerization
Photopolymerization was carried out in order to form polymer coatings onto the mild steel AS1020 surface (Scheme 3
). To a typical UV-curable acrylic formulation including a mixture of mono and difunctional acrylic monomers and a photoinitiator, 20 wt% of the different coumarate monomers were added. After UV radiation, the liquid resin became a solid coating and the polymerization of the acrylic double bonds was confirmed by ATR-FTIR. Supplementary Figure S2 in the supplementary material
shows ATR-FTIR spectra of all four coatings comparing the control coating, the monomer mixture and the final coating containing 20 wt% coumarate ionic units. In all the coatings, the disappearance of the band between 1600–1650 cm associated to the double bond can be observed, which confirms the high extent of the photopolymerization (>90% in all the cases). By this method, the cationic moiety of the coumarate ionic liquid monomer is covalently attached to the acrylic network whereas the coumarate anion is interacting ionically with the polymer backbone.
First, we investigated the stability of polymer coatings towards water uptake and leaching of the coumarate counter-anion. It is worth mentioning that, leaching out of the coatings of the corrosion inhibitors when added as additives is one of the main limitations. Whereas in this paper we proposed to attach part of the corrosion inhibitor covalently to the coating by copolymerization and to retain the other active ingredient through strong coulombic interactions. The polymer coatings were immersed in water for 24, 48 and 72 h, patted dry and then weighed so as to study the differences in mass uptake or loss (Supplementary Figure S3
). Interestingly, the acrylic coatings are not absorbing water in spite of the presence of the hydrophilic ionic monomers. Leaching from the film of the unreacted monomer or pendant counter-anions can thus be analyzed for each coating. Interestingly, the Acrylic UV-coatings did not show a significant reduction in mass indicating that the coumarate inhibitors are well integrated into the acrylic polymer coating. Only, the [VPy+ HexOCou-] based coating lost 4% of its mass in the first 24 h and 8% after 72 h which indicates less integration in the polymer coating.
The anticorrosion profile of each polymer coating containing coumarate inhibitors was tested by EIS measurements. Figure 6
shows the Nyquist plot of different polymer coatings after an immersion of 22 h in NaCl 0.005 M aqueous solution (Supplementary Figure S4
). All the coated samples including ionic monomers showed a bigger depressed semicircle than the semicircle of the UV-cured coating control without ionic monomer. The samples coated with acrylic formulations having 20 wt% of [VIm+ HexOCou-] and [VAn+ HexOCou-] showed the biggest capacitive loop that is constant during the length of the experiment. The [VPy+ HexOCou-] containing coating showed less anticorrosive performance amongst the different coatings containing inhibitor which also correlates with its poor performance as a monomer inhibitor.
Pictures of all coatings were taken after EIS measurements (Figure 7
). As aforementioned, the control coating presents some rust deposits on the surface. In contrast, [DMAEM+ HexOCou-], [VIm+ HexOCou-], and [VAn+ HexOCou-] based polymer coatings do not suffer any deterioration. However, as the Nyquist plot showed, the [VPy+ HexOCou-] based coating presents a corroded site which leads to a decrease in the impedance of the coating and thus a decrease in anticorrosion activity (Figure 6
). Figure 7
shows the magnitude of the impedance responses in the form of the Bode plot for [DMAEM+ HexOCou-], [VIm+ HexOCou-], [VAn+ HexOCou-], and [VPy+ HexOCou-] based polymer coatings. The initiation of corrosion can be observed in the control coating due to a decrease in the impedance and in the plateau of the phase angle in the low frequency range (10−1
) with time. The Bode plots for the [DMAEM+ HexOCou-] based coating show an impedance of 104.47
at the low frequency range after 1 h of immersion in NaCl 0.005 M aqueous solution. This impedance is reduced to 104.24
after 9 h and it remains constant until 22 h. That variation corresponds to the initiation of corrosion and the gradual formation of the corrosion products therefore only providing a limited level of protection against mild steel corrosion relative to control. On the other hand, polymer coatings based on [VIm+ HexOCou-] and [VAn+ HexOCou-] present the highest impedance values, which are also constant in time, 105.70
, respectively. Accordingly, they present a high phase angle value, 78° and 85°, respectively (Supplementary Figure S5
). The polymer coating with 20% [VAn+ HexOCou-] showed the best anticorrosive profile, in fact, due to the high impedance and phase angle data and the sensitivity of the experiment, noisy data can be observed. As observed from the Nyquist plot, the [VPy+ HexOCou-] based coating showed lower impedance during immersion. In fact, as the leaching test on [VPy+ HexOCou-] based coating showed, the concentration of the inhibitor in the polymer may be reduced, and as the Bode plot demonstrates, the anticorrosive effect is partly lost.
The scribe test was carried out onto in order to further compare the different coatings. A defect was introduced into a reference UV polymer coating without any inhibitor, and coatings containing 20 wt% [DMAEM+ HexOCou-], [VIm+ HexOCou-], [VAn+ HexOCou-], and [VPy+ HexOCou-] respectively. The coatings were introduced into acidic solutions in order to initiate the corrosion reaction. In Figure 8
, filiform corrosion pictures of all coatings can be observed after 10 days. As can be observed, the control coating without coumarate inhibitors showed a completely rust covered surface. On the other hand, the polymer coatings containing inhibitors passed very well the scribe test showing very little corrosion propagation. As shown in the pictures the polymer coatings including [DMAEM+ HexOCou-], [VIm+ HexOCou-], and [VAn+ HexOCou-] showed excellent performance. On the other hand, the coating including [VPy+ HexOCou-] showed the worst corrosion inhibition. These data are only qualitative, but still it is interesting to note that the order of performance from this test is different to the EIS data discussed above. These differences may be related to the different initiation mechanism for corrosion, (i.e., acid exposure in the scribe test versus chloride in the immersion tests).