A Novel Environmentally Friendly Synergistic Mixture for Steel Corrosion Inhibition in 0 . 51 M NaCl

The ubiquitous use of plain and low-alloy steels in neutral, chloride bearing environments presents an everyday challenge for corrosion protection professionals. This paper explores the possibility of developing a non-toxic and environmentally friendly synergistic inhibitor mixture made of propolis, tannin, sodium benzoate, PEG400 and starch that could be applied into the solution to induce the formation of a persistent protective layer on steel. Components of the mixture were chosen based on the references giving their characteristics relevant to their possible action in the solution and/or at the surface of the steel. The efficiency of the protective layer formed under the influence of the inhibitor mixture, and then measured by the LPR probe in the inhibitor-free solution, was the lowest under quiescent conditions (75%) and the highest (95%) under flow conditions. Both, LPR and EIS data indicate that the inhibitor modified layer presents a barrier for diffusion of oxygen that acts as a primary corrosion reaction depolarizer in the investigated neutral chloride solution. The demonstrated persistence of the inhibitor modified layer is of primary interest as it gives possibility batch application.


Introduction
Steel is the most widely used metallic material thanks to its good mechanical properties and low price.However, when it comes into contact with aerated neutral electrolyte, especially in the presence of chloride ions, steel is subject to rapid corrosion degradation.
Neutral saline aqueous environments of various origins are frequently simulated in laboratory by the use of 0.35-0.85M NaCl solutions. 1Corrosion rates recorded on plain steel by polarization and weight loss techniques are fairly consistent and range between 0.1 and 0.5 mm per year. 2,3These rates are generally not acceptable and protective measures have to be applied.Due to the inherent characteristics of corroding systems (e.g.complex geometry or need for good thermal conductance) corrosion inhibitors frequently stand out as the most practical or the only solution.Unfortunately, protection of steel in neutral chloride media is challenging from theoretical and practical viewpoint.It can readily be observed that considerably more literature is found on inhibition of steel corrosion in acidic media than in neutral media and that the inhibitor effectiveness is generally higher.From the theoretical aspect, the reason is that in near-neutral, neutral and alkaline media, inhibitor action is complicated by formation of iron oxides and hydroxides at the surface of metal. 4n acidic media, bare iron (or steel) surface is positively charged and behaves as a soft Lewis acid, hence, soft Lewis base compounds with heteroatoms O, N, S, P and/or π electrons in their structure, are good inhibitor candidates due to the possibility of soft acid-soft base adsorption bond formation. 5,6On the other hand, when oxide is present, surface acidity and basicity depends on oxide's isoelectric point, which e.g. for Fe 2 O 3 , equals pH 8.6. 7Below this pH it may be assumed that the surface oxide is protonated, positively charged and acting as a hard Lewis acid.Hard Lewis bases would therefore be good inhibitor candidates capable of hard acid-hard base adsorptive bond formation.
In the present study, tannin, sodium benzoate, propolis, starch and polyethylene glycol (PEG400) were chosen as components of the investigated inhibitor mixture.When Božović et al.: A Novel Environmentally Friendly Synergistic Mixture ... such a mixture rich in OH groups is used for inhibiting corrosion, in near neutral and neutral pH, part of the OH groups will be deprotonated forming anions that have electronegativity practically equal to their ionization potential, and therefore act as almost pure electron donors or hard Lewis bases. 5 Adsorptive hard acid-hard base bond which is of predominantly ionic character would therefore be possible between compounds of the mixture and the oxidized steel surface.Additionally, metal cations in high oxidation state, such as ferric ions dissolved from the steel surface into the electrolyte, are hard Lewis bases, so that formation of protective deposits of hardly soluble complexes and/or chelates of ferric ions with anionic organic ligands is possible. 8omponents of the inhibitor mixture investigated in the present study were chosen after careful consideration of their environmental friendliness and previously published characteristics relevant to corrosion inhibition.E.g., ethanolic propolis extract applied to mild steel forms a very effective anticorrosion coating in neutral and near neutral aqueous 3.5% w/v NaCl solution. 9Propolis was also proven to be effective on copper in neutral chloride solution, 10 but copper is more favourable for adsorption of organic compounds under such conditions. 1 Adsorption of effective propolis layer on steel from the solution is probably hindered by its insufficient concentration due to the limited solubility of propolis in water.Dispersion of propolis in neutral chloride medium produces a yellowish emulsion. 9ased on literature data, PEG is used as an alternative solvent for non-ethanolic propolis extraction, 11 and when used together with propolis, could beneficially influence propolis dispersibility in a water solution. 12Furthermore, it has been shown that a compact layer, with barrier effect for the diffusion of oxygen in 0.1 NaCl solution, has been deposited on mild steel from the solutions containing 1000, 2500 and 5000 g ppm of PEG after 30 min of immersion time, showing that PEG itself can act as a corrosion inhibitor. 13Formation of the protective ferric ''tannate film'' deposited from solution of tannin is also possible, and it was shown to begin at pH = 3, but relevant efficiencies are attained only at concentrations ~1000 ppm. 14Recent investigation of low carbon steel inhibition by 1-5% tannic acid solution in 3.5% NaCl shows that efficiencies up to 51% can be obtained. 15Concentrations of the same order are needed for efficient steel corrosion inhibition in chloride media by starch 16 and sodium benzoate. 17In the present study, minimal quantities of individual components that would be effective, based on the above literature references, were chosen for the inhibitor mixture composition.
In general, extra costs are incurred in neutral chloride media because the enhancement of inhibitor effectiveness is frequently attained through the increase in concentration up to a few thousands of ppm. 189][20][21][22] In the present case, different chemical nature of components of the mixture used was expected to yield a combination of mechanisms of action, which is usually argued to be the basis for synergism. 23ropolis was chosen as a starting component of the mixture.It is generally accepted that propolis is composed of approximately 50% resin (polyphenolic fraction), 30% wax, 10% essential oils, 5% pollen and 5% other organic and inorganic compounds. 9Main propolis phenolic esters and flavonoids are of amphiphilic nature, 24 and are expected not only to inhibit corrosion but also to increase the hydrophobicity of steel or corrosion products due to adsorption with the hydrophilic part of the molecule oriented to the surface and the hydrophobic part oriented towards the liquid phase. 25Chestnut tannin extract 6 was chosen as a second component in order to increase the water-soluble inhibiting fraction of the mixture.Since polyphenols are weak organic acids known to decrease pH of neutral solutions, 14 along with the inhibitive effect of the tannate layer, iron dissolution enhancing effect may be expected. 15Increased release of iron ions form steel, on the other hand, is favourable for ferric tannate formation.Hence, sodium benzoate 17 has been added to the mixture for controlled iron release, until the protective surface layer is fully formed.
The last two components added to the mixture were polymeric substances, PEG and starch.Besides with the intention of increasing propolis solubility, PEG was added alongside with starch, with the intention of strengthening surface layer through interaction with tannin.Tannins have an abundance of hydroxyl groups in close proximity capable of strong and specific interaction with carbohydrate polymers via hydrogen bonding and hydrophobic interactions. 26When phenolic components and PEG and starch are mixed, the hydroxyl groups of phenolic chemicals can interact with PEG oxygens 27 and side chains of starch amylopectin, 28 respectively.
In this paper we extend the synergistic approach by hypothesising that a carefully chosen mixture of non-toxic and environmentally benign inhibitive compounds could modify the rust layer on steel.This layer would form in neutral chloride solutions, as to produce a persistent film which would lower corrosion rate considerably.The inherent advantage of the novel mixture would be environmental friendliness as well as the retention and resiliency of the inhibitor modified rust layer even in the inhibitor free solution.High concentration requirement would be counteracted by a long inhibitor free period, following batch application, during which the protective surface film would ensure acceptable corrosion rates.

Božović et al.:
A Novel Environmentally Friendly Synergistic Mixture ... Natural propolis (Naturwaren-niederrhein, GmbH, Propolis Pulver) was used at a concentration of 100 ppm in 0.51 M NaCl. 100 mg of propolis was first dissolved in 5 ml of 70% ethanol before preparing 1 dm 3 of solution.Chestnut Tannin (Tanin Sevnica) and sodium benzoate (AGZ, food additive) were both used at a concentration of 2000 ppm.
PEG400 (Sigma Aldrich) and corn starch (Sigma Aldrich) were used in concentrations of 200 and 400 ppm, respectively.
The inhibitive mixture of 2000 ppm of tannin (T), 2000 ppm of benzoate (B) and 100 ppm of propolis (P), 200 ppm of PEG400 (PEG) and 400 ppm of starch (S) in 0.51 M NaCl is in further text denoted by abbreviation T+B+P+PEG+S, and the other combinations of single components are denoted accordingly.
All the measurements were done in aerated solutions under ambient conditions.

LPR Probe
The measurements of corrosion rate were performed on the LPR (Linear Polarization Resistance) probe manufactured by RCSL with the data collector MS1500L.The corrosion rate was determined by software of the MS1500L instrument.
LPR electrodes were, prior to immersion, treated mechanically by abrasive paper of grit 240, followed by degreasing with ethanol in an ultrasound bath and rinsing in redistilled water.Measurements were made by immersion of the electrodes in the solution of 0.51 M NaCl, as well as in solutions with the chosen concentrations of inhibitors.A protective layer at the electrode surface was formed for 24 h in quiescent solutions.In one experiment, the layer was deposited from the inhibited solution mixed at 80m RPM, in order to rate the influence of solution mixing on layer formation and its protective properties.In the non-inhibited solutions, the rust layer was formed, while in the inhibited solutions, the rust layer modified by the action of corrosion inhibitors was obtained.The layer formed without inhibitor is in the rest of the text referred to as the "rust layer", and the layer formed under the influence of the inhibitors in 0.5 M NaCl or the layer-forming solution, is in the rest of the text referred to as the" inhibitor-modified rust layer".The corrosion rate was measured after the probe was transferred into a fresh 0.51 M NaCl solution and left to stand for 24 h.The LPR measurement was then performed in a quiescent solution as well as in a solution stirred by magnetic stirrer at 90 to 150 RPM.Long-term measurements with the layer formed in the most effective inhibitor mixture as well as the pure rust layer were performed continuously for 10 days.In this way, the persisten-cy of the protective layer is tested by LPR probe, which was previously found to be particularly suitable technique for that laboratory testing of inhibitor persistence. 29

3. Electrochemical Impedance Spectroscopy (EIS)
For EIS, the PalmSens3 device with PC Trace 5.3 software has been used.A cylindrical steel sample made of API X52 5L steel embedded in epoxy resin and having circular cross section exposed to the electrolyte, was used as a working electrode.
Typical chemical composition of commercial grade API 5L used is C (0.26 max.), MN (1.40 max.), Si (0.4 max.), P (0.03 max.), S (0.03 max), Fe (bal.). 30 saturated calomel electrode was used as reference electrode and a graphite electrode was used as an auxiliary electrode.The working electrode surface was 0.2826 cm 2 .Prior to electrochemical testing, the surface of the working electrode was mechanically treated by abrasive paper of various grits, namely: 240, 600 and 800.Subsequently, the surface was degreased in an ultrasonic bath with ethanol and washed with redistilled water.
EIS measurements were performed in quiescent solution, at the open circuit potential, at frequencies from 10 kHz to 1 mHz, with the amplitude of 10 mV.The electrodes on which a protective layer (of rust or inhibitor-modified rust) has been forming for 24 h, as in the experiment with LPR probe, were left to freely corrode for another 24 h in a fresh 0.51 M NaCl solution, in order to simulate a situation 24 h after batch inhibitor treatment.Additionally, EIS measurement has been done in the inhibited solution prior to transfer into the fresh NaCl in order to compare inhibitive effect in inhibitor bearing and inhibitor free solutions.

4. Fourier Transform Infrared Spectroscopy (FTIR)
FTIR spectra were recorded using a PerkinElmer spectrometer Spectrum One.Spectra were obtained in the range from 400 to 4000 cm −1 , each spectrum being an average of ten scans with a resolution of 4 cm -1 .Coupons used as substrate were made of previously mentioned API X52 5L material and were cut to size 30 × 40 × 3 mm, polished with abrasive paper of grit 240, 600 and 800 and degreased before exposure to inhibitor mixture solution.Inhibitor-modified rust layer has been formed for 24 hours, identically as for the EIS measurements, and then the coupons were left in 0.51 M NaCl for 10 days.350 mg of KBr was rubbed onto the surface of steel coupons after ending the experiment and drying of the coupons.Samples bearing KBr were then hydraulically pressed into a 13 mm stainless steel die and the resulting pellets further subjected to FTIR measurement.Preparation of pellets was in conformance with the standard ASTM E1252:2007.

5. Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray (EDX) Analysis
SEM and EDX analysis was recorded using Tescan Vega III, SBU EasyProbe scanning electron microscope with 15 kV accelerating voltage of the electron beam at various magnifications.Measurements were done on surface of the coupons prior to FTIR experiments.

Results and Discussion
Cathodic reaction takes place on iron or steel through two possible processes.At lower pH values of the solution, hydrogen evolution is controlled by the rate of charge transfer reaction, and at higher pH of the solution, oxygen reduction is controlled by the rate of oxygen diffusion.Lorbeer and Lorentz, 31 state pH of 4.2 as critical, above which the dominant cathodic process is oxygen reduction and anodic iron dissolution is inhibited by time-dependent formation of a three-dimensional porous oxide layer on the electrode surface.In the present study, the pH of uninhibited solution was 6, while the pH of inhibited solutions was between 4.6 and 4.7 so the primary conjugate cathodic reaction of metal dissolution was reduction of oxygen.This reaction is known to be under diffusion control and satisfactory degree of corrosion inhibition requires dense blocking of the surface. 23Oxygen reduction generates OH -ions which react with the dissolved Fe 2+ ions yielding Fe(OH) 2 that is considered to be a precursor of other rust forming compounds. 16Formation of particular rust component depends on pH, temperature and oxygen content.The rust layer partially protects metal surface from dissolution. 2 It has been found that increasing the immersion time of iron and steel in the solution enables thickening of the corrosion product layer resulting in the decrease of the corrosion rate.The decrease of the weight loss has been observed for carbon steel during the first 10 days of immersion in 3.5% NaCl, followed by a period of 20 days in which the weight lost remained low, while in the following 30 days the weight loss became more noticeable and higher. 3This observation indicates that dissolution of rust itself also occurs under the attack of chloride ions.The experiments in the present study were focused on modification of the rust layer by the inhibitors in order to obtain a persistent protective layer that lowers the corrosion rate to acceptable values.

1. Measurements at the LPR Probe
Figure 1 a) shows LPR corrosion rates on electrodes with the rust layer or with the inhibitor-modified rust layer formed under the influence of various inhibitors and inhibitors mixtures.The rates were measured after 24 h in layer-forming solutions, 24 h after transfer to quiescent 0.51 M NaCl solution and in in the same solution at the magnetic stirrer rate of 150 rpm.Except for tannin, single components show <30% inhibitive action both in the inhibitor solutions and quiescent 0.51 M NaCl with a formed layer.As expected, 15 the solution of tannin as a chelating agent significantly increases corrosion rate of steel to 1.6 mmpy, but at the same time forms the most protective modified rust layer suppressing the corrosion rate in 0.51 M NaCl to 0.06 mmpy.Addition of benzoate to tannin solution decreases the corrosion rate to 0.11 mmpy but at the same time hinders the formation of protective tannin modified rust layer, increasing the corrosion rate in 0.51 M NaCl to 0.2 mmpy.Further addition of propolis and PEG decreased the corrosion rates in quiescent 0.51 M NaCl, while the addition of starch decreased the corrosion rate in quiescent and mixed 0.51 M NaCl.For the final inhibitor mixture, all corrosion rates were below 0.1 mmpy.equal to U= (0,051 × corrosion rate + 0.005) mmpy has been calculated from these data. 32he efficiency of the inhibitor containing solution, or the inhibitor-modified rust surface layer can be calculated by the equation: (1)   where CR 0 denotes corrosion rate for the system without inhibitor or with only the rust layer and CR denotes corrosion rate for the system with the inhibitor or the inhibitor-modified rust layer.
In parallel to the corrosion rates, the inhibitor/layer efficiencies are presented in Figure 1 b).Tannin shows a highly negative efficiency of -1200% when present in the solution, while at the same time forming a protective modified rust layer that shows the efficiency of ≈ 60% in quiescent and ≈ 84% in mixed 0.51 M NaCl.Benzoate counteracts the facilitation of iron dissolution by tannin but decreases the efficiency of the inhibitor-modified layer below the efficiency of the pure rust layer.Addition of propolis, PEG and starch, gradually "repairs" the layer, at the same time still successfully counteracting the iron dissolution by tannin.Although interpretation is complicated by the mixed mechanism of action of various components, by comparison of single inhibitor component action with their mixed effect, it is revealed that starch acts in synergy with T+B+P mixture in the layer-forming solution as well as in quiescent and mixed 0.51 M NaCl.Starch also shows synergy with T+B+P+PEG mixture in quiescent and mixed 0.51 M NaCl.
Figure 2. shows LPR inhibitor efficiencies with the inhibitor-modified rust layer formed under the influence of T+B+P+PEG+S inhibitor mixture, measured in quiescent 0.51 M NaCl solution and in the same solution mixed at various magnetic stirrer rotation rates.Identical measurements were repeated for 10 days in order to assess inhibitor-modified rust layer persistence.
The increase in the inhibitor efficiency is observed with the increase in magnetic stirrer rotation rate.The increase is most remarkable between the quiescent system and the system with the lowest mixing rate of 80 RPM.This observation may be explained by the increase of oxygen transport to the electrode by stirring causing higher increase in corrosion rate 33,34 for rust covered electrodes than for the inhibitor-modified rust covered electrodes.However, the inhibitor efficiency levels off with further increase in the electrolyte mixing rate.It is reasonable to assume that as the oxygen supply trough mixing completely counteracts the diffusional influence the efficiency becomes almost constant with stirring rate.
The data also show gradual loss of efficiency with time.After 10 days of immersion, with the inhibitor-modified rust layer, the efficiency at 150 rpm is ≈ 70% and the corrosion rate is still below 0.2 mmpy as opposed to inhibitor free system in which the corrosion rate at 150 rpm roughly equals 0.875 mmpy.The concept of improving corrosion resistance of steel by the formation of a surface layer that efficiently blocks corrosive components from surrounding medium, is self-explanatory, and has also previously been proposed in for steel corrosion in hot tapwater. 35,36he brown-orange colour of rust in the system without inhibitor indicates that its main component is lepidocrocite (γ-FeOOH). 37On the inhibitor protected electrode, a dark violet layer is visible, proving formation of ferric tannate. 14Significantly smaller amount of corrosion products is observed in the solution with electrodes having inhibitor-modified rust layer indicating better protection offered by that layer compared to the protection provided by the pure rust layer.

2. Electrochemical Impedance Spectroscopy (EIS)
Figure 3 shows the EIS results for the electrodes with layers formed without and with T+B+P+PEG+S inhibitor mixture.To compare protective effect of the presence of the inhibitors in solution, the measurement was also done in the layer-forming solution immediately before transfer of the electrode with the formed protective layer into fresh 0.51 M NaCl solution.Figure 4 shows the equivalent cir-  1 shows EIS fitting results.Two parallel nested RC circuits were used to model the EIS spectrum of the electrode with the rust layer and EIS spectrum of the electrode in the inhibitor solution.Three parallel nested RC circuits were used to model the EIS spectrum of the electrode with an inhibitor-modified rust layer.
The efficiency of the rust and inhibitor-modified rust layer can be calculated by the equation: (2)   where R denotes R 2 + R 3 + R 4 from Table 1, calculated for the system without inhibitor (R), and system with the inhibitor or the inhibitor-modified rust layer (R 0 ).High efficiency is obtained irrespective of the inhibitor presence or absence in the solution.High frequency resistance R 1 of the unprotected electrode is probably due to the solution resistance while the R 1 of the protected electrodes may be interpreted as a sum of the solution resistance and resistance due to the partial barrier properties introduced by the surface layers. 38The R 1 resistance is disregarded in the effectiveness calculation due to its value being insignificantly small with respect to the inhibitor-modified rust layer resistance.The inhibitor-modified rust layer protected electrode shows the decrease of n 2 parameter close to   0.5, as well as the typical shape of Nyquist plots 39 that indicate that the inhibitor-modified rust layer presents a barrier for oxygen diffusion, as previously concluded from LPR measurements.

Fourier Infrared Spectroscopy (FTIR)
Figure 5 shows FTIR spectra of propolis and tannin and of the layers formed on steel electrodes in uninhibited, T+B+P and T+B+P+PEG+S solutions.Spectrum of the rust layer from the uninhibited solution indicates a mixture of FeCl 3 (spectrum ID = 8ns2iWUHtq7, 40 ), Fe 3 O 4 (spectrum ID = 8ns2iWUHtq7, 41 ) and Fe 2 O 3 (Spectrum ID = AVGw5xAY59X, 42 ).In particular, the band at 574 cm -1 corresponds to the Fe-O vibrations of magnetite, 43 the bands around 750 and 870 cm -1 show presence of goethite and the bands at 1023 and 1177 cm -1 indicate presence of lepidocrocite.Fingerprint region of FTIR spectra between 400 and 1800 cm -1 of tannin, propolis and T+B+P layer, is similar between the three samples and is also similar to the reference tannin spectrum (Spectrum ID = KPLVhGlArJg, 44 ).Dark purple coloration of the electrode surface layer that fully develops over the period of 24 hours, confirms presence of tannin in the form of ferric tannate. 13In particular, the band at 1326 cm -1 in tannin spectrum is characteristic of bending vibration of O-C in a phenolic hydroxyl group. 45The interaction between ferric ions and the phenolic hydroxyl group shifts the O-C bond stretching vibration to higher wavenumbers, in the present case from to 1343 and 1350 cm -1 in T+B+P and T+B+P+PEG+S layers, respectively.The tannin band at 1744 cm -1 that corresponds to C=O stretching of the polyphenol shifts to shorter wavenumbers, 45 i.e. 1726 cm -1 in the case of T+B+P layer and to 1719 cm -1 in the case of T+B+P+PEG+S layer.The peaks at 1417 cm -1 and 1423 cm -1 , in T+B+P and T+B+P+PEG+S layers, respectively, can be ascribed to the adsorption peak of a polyphenol C=O vibration. 46he most remarkable difference in FTIR spectra of rust and inhibitor-modified rust layers is the loss of intensity and shift of magnetite peak at 574 cm -1 of the rust layer to 506 cm -1 in the case of the inhibitor-modified rust layer, probably due to the conversion process to ferric tannate. 45FTIR spectra show a dominant role of flavonoid species, present both in tannin and propolis, on the formation of the inhibitor-modified rust layer.

4. Scanning Electron Microscopy (SEM) and Energy Dispersive X-ray (EDX) Analysis
Before the electron microscope scanning, the loosely held part of the layer has been blown off in the nitrogen gas stream so that only the firmly held parts of the layer remained.Figures 6 a), c) and e) show the surface of metal with the rust layer and Figures 6 b) and d) and f) with the inhibitor-modified rust layer from T+B+P+PEG+S solution.
EDX results for measurement positions #1 to #9 are shown in Table 2.
Oxygen and carbon in the rust layer, found in excess with respect to the substrate, probably originate from surface oxides and carbonates that may form in nwweutral solutions in contact with the air. 47Fine grained structure observed in both cases is also probably due to surface oxides and carbonates.The inhibitor-modified rust layer is significantly denser and uniform in micrograph shown in Figure 6 b) than the rust layer shown in Figure 6 a).Corrosion defects in the form of shallow pits are visible only in the case of the rust layer.The morphology of the tannate layer is close to the one recently reported for tannic acid on Q235 carbon steel in 3.5% NaCl. 15Similarly, due to inhibitor presence, the carbon content of the layer is much higher in the inhibitor-modified rust layer at locations #2, #4, #5 and #8 (16-25 %wt) than in the rust layer at locations #1, #3 and #7 (<6 %wt).Low carbon content is also observed at the location of the crack in the inhibitor layer protruding to the metal surface at position #6, and in the area of the damaged layer where the substrate is clearly visible at position #9.

Conclusion
Rust layer obtained by exposing steel surface to 0.51 M NaCl solution containing a mixture of non-toxic, environmentally friendly compounds (propolis, tannin, benzoate, PEG400 and starch) has shown better protective characteristics than the rust layer formed in the inhibitor free solution.From LPR probe measurements in quiescent and mixed solution and from the values of EIS parameters in quiescent solution, it may be concluded that the inhibitor-modified rust layer primarily presents a barrier for oxygen diffusion.A synergy effect has been demonstrated for  starch as the mixture component, from LPR data.The inhibitor-modified rust layer shows persistence in a mixed solution.The efficiency of inhibitor-modified rust layer measured by LPR after 24 h of exposure to 0.51 M NaCl is the lowest in quiescent solution 75% and the highest, equalling 95% in mixed solution.Visual observation and FTIR data confirm the formation of ferric tannate in the inhibitor-modified rust layer while the SEM/EDX data show increased carbon content of the inhibitor-modified rust layer having a fine grain structure that is more uniform and denser than the one of the rust layer.
The persistence of the inhibitor-modified rust layer is of great technological interest as it allows batch application, increases flow corrosion resistance and offers prolonged protection at acceptable corrosion rates.The suggested concept of attaining these beneficial effects through rust modification by inhibitors of various action mechanisms is not fully exploited in scientific reports and is worthy of further investigation.

Figure 1 .
Figure 1.a) corrosion rates and b) the appending inhibitor efficiencies obtained from the LPR measurements after 24h in the layer-forming solution and after subsequent 24h in 0.51 NaCl at 0 and 150 rpm.
for each system, has been done in triplicate and the expanded measurement uncertainty Božović et al.: A Novel Environmentally Friendly Synergistic Mixture ...

Figure 2 .
Figure 2. LPR inhibitor efficiencies with the inhibitor-modified rust layer formed under the influence of T+B+P+PEG+S inhibitor mixture, measured in quiescent 0.51 M NaCl solution and in the same solution mixed at various magnetic stirrer rotation rates

Figure 3 .
Figure 3. a) Nyquist and b) and c) Bode EIS results on steel electrode after 24 h of immersion in 0.51 M NaCl, in the layer-forming T+B+P+PEG+S solution after 24 h of immersion and after further 24 h of immersion in 0.51 M NaCl with the inhibitor-modified rust layer at the steel surface.

Figure 4 .
Figure 4. Equivalent circuit used for modelling the EIS data.

Figure 5 .
Figure 5. FTIR spectra of propolis and tannin and of the layers formed on X52 5L steel electrodes in pure 0.51 M NaCl, T+B+P and T+B+P+PEG+S solution.

Table 1 .
Results of EIS data fitting to the equivalent circuit shown in Figure4.

Table 2 .
EDX elemental analysis results in %wt for measurement positions in Figure6.