SOJ Materials Science & Engineering

Adsorption of vinyl triethoxysilane and γ-aminopropyltriethoxysilane on the surface of aluminiumfrom aqueous solutions was studied by quartz nanobalance technique. Adsorption isotherms were obtained. It was shown that during adsorption silanes displace adsorbed water from the surface. To interpret the adsorption data, we used the well-known approaches of adsorption data: Langmuir, BET, Flory-Higgins, Langmuir multicenter,Temkin,Frumkinand Freundlichadsorption isotherms. The surface orientation of adsorbed molecules was determined. The silanes adsorption heats were calculated by different ways. It was shown that silanes are chemisorbed on the aluminium surface. By means of quartz nanobalance, atomic force microscopy, scanning electron microscope, Fourier transform IR spectroscopy it has been shown that an uniform, selfassembled, anti-corrosionvinylsiloxanenanolayer is formed on the aluminium surface during adsorption of vinyl silane from aqueous solution. This layer covalently bonded to the hydroxylsurface groups of the metal. The thickness of the surface nanolayer can be controlled by changing the deposition conditions. The effect of the vinylsiloxanenanolayer on the dissolution of aluminium in chloride-containing solutions has been studied. It was established that an ordered vinylsiloxanenanolayer, up to five molecular layers thickness, effectively inhibits the uniform and localized aluminium corrosion. It was shown that after 10 days of corrosion tests vinylsiloxanenanolayer is preserved on the aluminium surface, indicating its resistance to water and corrosive components.vinylsiloxane

Keywords: Aluminium; Adsorption; Alkoxysilane;Organosilicon Self-Assembled Nanolayers; Heterogeneous Processes; Corrosion; Electrode Reactions; Corrosion Test

It is known that [1] aluminum and alloys based on it have found application in many areas, such as the aviation industry, engineering, shipbuilding, energy, etc., as they combine lightness and strength, high thermal conductivity and electrical conductivity, reflect light and heat well and in addition, are hygienic and non-toxic. A wide variety of forms of these materials are produced in contributes to their various use. At the same time, being stable under conditions ofuniform corrosion, they have low corrosion resistance in chloride-containing media, becausethey are prone to pitting corrosion [2]. Pitting corrosion is one of the mostdangerous types of metal structures destruction and occurs on metals covered with an oxide filmin the places of the passive film damage. Due to the implementation of high metal dissolutionrates in local areas, the result of pitting corrosion may be failure of the entire structure. Despitethe increased interest of researchers in the aluminium and its alloys pitting formation, observedfor several decades [2-6], the problem of developing effective methods of aluminium and its alloys protectionfrom pitting corrosion is relevant

Usually, during the operation, aluminium and its alloys products are protected from corrosion by polymer and paint coatings, which are well established as one of the most effective ways to reduce corrosion risk in various conditions [7]. Often, when they are used, the metal surface is subjected to preliminary’ chemical treatment (finishing) to improve corrosion resistance and increase the paint (polymer) coating adhesion to the metal. Until recently, the metal surface was chemically treated with hexavalent chromium compounds. Chromatic layers effectively protect the metal from corrosion while providing high coating adhesion. However, hexavalent chromium, which is part of the components, is an environmentally harmful element, the use of which is currently highly undesirable and in the future is subject to a complete ban[8]. In recent years, the efforts of researchers have been aimed at replacing chromatic technologies, in particular in aluminium and its alloys processing [9,10]. However, the problem has not been solved, and the development of more efficient and environmentally friendly methods of aluminium surfaces pre-treatment is an urgent scientific and technical task. Toxic chromium compounds were replaced by organosilanes, whose general formula is as follows - RnSi(R’)4-n. Organosilanes are non-toxic and can be adsorbed on the aluminium surface, forming surface self-assembled siloxanenanolayers[11-13]. At the same time, on the one hand, strong and hydrolytically stable bonds of Al-O-Si (Flanuni et al. 2012) with the hydroxylated metal surface are formed and. on the other, it is possible to provide a high affinity of the nanolayer to a wide range of polymeric andpaint materials due to the variation of the organic radical R chemical nature.

Despite the great long-term interest in the studyof organosiliconnanolayers on inorganic surfaces [11,14]. the mechanism of organosilanes adsorption on aluminium and their influence on the corrosion properties of this metal have not been sufficiently investigated. In this regard, the aim of this work is to study the adsorption of vinyl and amino -containing trialkoxysilans on the aluminium surface and to study the influence of surface siloxane nanolayers on the corrosion and electrochemical behavior of aluminium in chloridecontainineelectrolvtes.

Experimental
А995 aluminium foil [15] with a thickness of 100 μ and thermally deposited(from vacuum) A99: aluminum were used in the work. The aluminum sheathing was placed in a tungsten evaporator and heated to an evaporation temperature by passing 70 A current in 10-6 mm of mercury vacuum, achivable by using vacuum post VUP-4 (made in Russia). Two types of substrates were used for the aluminum layer application: 1) quartz resonator brand QC-10-AuBU gold plated, with AT-cut, and basic frequency of 10 MHz (Elchema, USA) and 2) glass plates with 3 mm thickness and a size 15x30 mm were used. The substrates surface was degreased with alcohol, washed with water and dried in air at room temperature.

The thickness of the metal layer was determined by electrochemical piesoquartznanobalance[13,16] on the EQCN 700 unit (Elchema. USA), measuring the change in the frequency of the quartz resonator during the spraying process and calculating the change in mass during metal deposition and the value of the true metal surface by the method described in [13]. All calculations were carried out taking into account the true value of the metal surface.

Organosiliconnanolayers based on vinyltriethoxysilaneCH₂=C HSi(OC₂H₅)3 (VS) ultrapure

(Witco Co, Switzerland) and aminopropyltriethoxysilane NH₂-(CH₂)3-Si(OC₂H₅)3 (AS) ultrapure (Witco Co, Switzerland) ere formed on the aluminium surface. Organosilanes were applied on the surface by dipping the samples into aqueous solution, which was prepared by adding organosilane to bidistilled water. The aqueous solutions with an interval of 10-5 – 10-1 M concentration were used.

Before using, the solution was kept for 60 minutes for silane hydrolysis. The samples were dipped in the solution and kept for 10 minutes, after which the excess organosilane was washed away by holding in water for 1 minute and dried in air at room temperature. After applying an organosilane layer, the surface was characterized using a piezoelectricquartz nanobalance, Atomic Force Microscopy (АFМ), Scanning Electron Microscopy (SEM), Fourier transformed Infrared Specroscopy (FT-IR).

Solver P47 (NT MDT, Russia)scanning probe microscope was used for AFM studies. Infrared reflection spectra were obtained using Perkin Elmer 2000 Fourier spectrometer in the range of 400-4000 cm-1, with a resolution of 4 cm-1, the number of scans 400. A mirror reflection attachment with an angle of incidence of 80° was used. Scanning electron microscopy of the samples surface before and after corrosion tests andX-ray Spectral Microanalysis (XSMA) were carried out at the Camebax SX50 unit (Samesa, France).

Corrosion tests of aluminium foil samples were carried out in sodium chloride solutions (ultrapure NaCl) with a concentration of 0.001 - 0.1 M and in a climatic chamber (Taiwan) RH 90%, t=60°C for 30 days. Corrosion of aluminium was studied using “in situ’ methods: piesoquartznanobalance[13], resistometry[17] and scanning reflectometry[18].

During resistometric studies of aluminium corrosion, both Al foil(with 100 μm thickness)or thermally-deposited Al samples were used. The working element of the resistonietric thermally deposited indicator was an aluminium layer 1 mkm thick, depositeded to the glass surface. Such indicators are used for rapid assessment (within 1-2 days.) of corrosive aggressiveness of the medium in the study of the metal behavior in natural conditions[17]. The resistonietric indicator is characterized by high sensitivity and response speed (about 30 minutes). The reliability of the corrosion rates obtained using indicators with a working element made of thermally deposited aluminium was confirmed by the gravimetric method on foil samples [17].

The change in the resistance of the corroding metal was measured and the change in the layer thickness was calculated by the formula

wheret is the time of testing, do is the initial sample thickness (μm). dt is the thickness of the sample at time t.ρ is the aluminum resistivity, 1 is the sample length, h is the sample width. R_o is the initial resistance of the sample. R_t- is the sample resistance at time t.

The sample resistance during the corrosion test was measured continuously with a frequency of one measurement in 10 min with an accuracy of 0.01 Om by APPA109N multimeter (APPA, Taiwan), allowing automatic recording of the measured values. The test duration was 100 and 300 hours for sprayed and foil samples respectively. The amount of corrosion was determined by gravimetrically with quartz nanobalance[19] technique on the EQCN700 unit.(Elchema, USA) by measuring the frequency change of the quartz resonator in the spraying process and calculating the mass change at the metal corrosion by the formula:

f0 is the basic base frequency of 10000 kHz; Δ_m is the change in the silane mass (g); Δ_f is the change in the frequency of the piesoquartz resonator (kHz). N is the frequency constant (for ATcut crystals, N = 1670 kHz-mm); is the quartz density equal to 2.65 g/cm³; S is the working quartz area, equal to 0.72 cm².
Studies of corrosion using scanning reflectometry were performed by processing the electrode surface diffuse reflection reflectogram obtained during corrosion tests. They were recorded on a standard computer scanner Epson Perfection 3200 Photo (Epson, Japan) with optical resolution of 3200 dpi and at the same time recorded the change in the corrosion potential of samples continuously for 10 days. The metal corrosion was evaluated by changing the area of the corroded surface during the test. A glass cylindrical cuvette with a smooth even, optically transparent bottom was used for testing. The sample was placed on the bottom of the cell, working surface down, providing a gap between the metal surface and the cuvette bottom with a glass stop. The angle between the metal surface and the scanner glass was selected in such a way as to obtain a white image of the sample [18]. The potential was measuredrelative to the chloridesilver reference electrode using APPA 109N multimeter(Appa Co, Taiwan).

The calculation of the degree of the surface coverage with defects and the average size of a single defect was carried out using the original software for digital image processing, written in the Ruby 1.9.0 programming language and the RMagick 2.12.0 program (IniageMagick 6.5.6-8). Three-dimensional visualization of surface defects was performed using software developed on the basis of Surfer 9.0 program.

The study of the organosilicon nanolayers effect on the electrochemical behavior of aluminium was carried out by polarization method (Kelly 2002). A three-electrode cell with a chloride-silver reference electrode was used. The measurements were carried out using the PI-50-1 potentiostat (Gomel plant of Measuring Devices, Belorus) at fixed potentials or potentiodynamically with a potential sweep rate of 0.1 mV/s. The experiments were carried out on disk (0.8 cm²) and cylindrical (12.14 cm²) aluminum electrodes.

The pitting formation critical potential (Epit), i.e. the potential above which the metal pitting dissolution occurs and stablepittings spread [20], was determined by anodic polarization curves from the fracture on the curve, as the potential upon reaching which a sharp increase in the current is observed (Figure 1) [20,21].

The study of the aluminum surface morphology with surface siloxane nanolayers wascarried out using the “Solver-Pro” atomic force microscope («Micro and Nanoholographicsystems» Company. Ltd, China) “ex situ” (in the air) in a contact mode. Data processing for image androughness data was performed using the following software: NOVA Solver Pro (NT-MDT, Russia).

Ethoxysilanes adsorption on the aluminium surface was determined by measuring the change in the frequency of the quartz resonator and recalculating it in the mass change by formula (2) when ethoxysilane was introduced into the solution. The addition of a small amount of VS corresponding to the solution concentration of 10^-5 M into the water first leads to a lightdecrease in the mass of the sample followed by its increase (Figure 2).

This may indicate the processes of competing adsorption at the metal-solution interface, as a result of which the adsorbed ethoxysilane displaces the solvent molecules, i.e. water, from the surface. Similar effects have been observed previously on copper [19] and they have been observed during the adsorption of corrosion inhibitors on the surface of iron and gold [23]. The sample weight reduction is insignificant in the case of VS (10^-5 M) and is 16.4 ng/cm² which may correspond to the displacement of 13-14 H₂O molecules.

At the same time, about 6 VS molecules are adsorbed on the surface, i.e. during adsorption, each silane molecule displaces 2 or more water molecules. Increasing the amount of silane to concentrations of 10^-3 also leads to displacement of water from the surface. In the case of aminosilane the same effect was observed, but the introduction of AS into the solution leads to the displacement of more solvent from the surface. The reduction in mass was 25.7 ng/cm² at a solution concentration of 10^-3 M which corresponds to the displacement of 20-22 H₂O molecules. Silane adsorption was calculated taking into account the mass of the displaced water. Increasing the concentration of alkoxysilane solution in water leads to an increase in its adsorption (Figure 3 and 4). Besides, it was found that the sorbed phase VS and AS consists of reversible and irreversible parts. The

reversibly sorbed layer is easily removed from the surface when the sample is kept in clean water for 10 minutes and is about 10% of the total amount of adsorbate. The irreversibly sorbed part of silane is water resistant for at least 8 hours, i.e. the adsorbed molecules of silane are dissolved in clean water during at least 8 hours, while maintaining the metal sample with irreversible sorbed VS and AS in the pure (without silane) water.

Figures 3 and 4 show the adsorption isotherms of vinyl - and the amino-silanes (Figure4). Figure 3 presents the isotherm of vinylsilane irreversible adsorption on the surface of freshly sprayed aluminium in a wide range of VS concentrations. The type of adsorption isotherm corresponds to the Langmuir isotherm which may indicate that VS molecules areadsorbed monolayerly. The conversion of the mass into the number of molecules showed that at VS solution concentrations less than 0.001 M per 1 nm2 of the surface, about 21 silane molecules were adsorbed, and the thickness of the monolayer can be calculated from the bond lengths and is about 0.9 nm. The increase in the VS solution concentration above 0.001 M leads to a polymolecular coverage of the surface and during adsorption from the 0.1 M solution on the surface forms a vinylsiloxane layer with a thickness of 10 molecular layers.

The shape of the aminosilane adsorption isotherm corresponds to a multi-molecularadsorption (Figure 4). Moreover, the conversion of the adsorbed AS mass into the number ofmolecules shows that even at low concentrations of 10^-4 M per 1 nm² of the surface 158 aminosilane molecules are adsorbed. The calculation using bond lengths shows that with densemonolayer coverage, there can be no more than 49 molecules nnr on the surface, i.e. at ASsolution concentration of 10^-4 M the metal surface is covered with more than 3 tightly packed,vertically oriented molecular layers of aminosilane. To study the adsorption mechanism of silanes adsorption data were processed in terms of known adsorption approaches: Langmuir (eq. 3), BET (eq. 4),Frumkin, Temkin,Flory-Higgens:

Where С is the VS (M) concentration, mis the mass of adsorbed VS (ng), mm is the mass of a tightly filled monolayer ng,b and k are constants.

It was found that the VS adsorption at concentrations up to 10-3 M and the AS adsorption on the surface of freshly sprayed aluminium is well described by Langmuir (3) and BET (4) isotherms (Table 1, 2).

Representation of the VS and AS adsorption isotherm in the coordinates of equations (3) and (4) gives a linear relationship with the correlation coefficients close to one (Table 1, 2), indicating compliance of the VS adsorption with the conditions of Langmuir and BET isotherms. Apparently, at the degrees of surface coverage with silane molecules of less than one monolayer, there is no interaction between neighboring adsorbed molecules, which allows the use of these isotherms in the description of adsorption.

The solution of equations (3) and (4) makes it possible to determine the area occupied by an individual molecule (“landing site”) and the heat of VS molecules adsorption at adsorption of the first layer. The values of “landing sites” were 0.05 and 1.23 nm² molecule in the calculation by the Langmuir and BET isotherm, respectively, i.e. from 19 to 25 molecules are adsorbed per 1 nm² at a monolayer coverage. This coverage may indicate the horizontal position of silane molecules on the surface (Figure 5).

In case of monolayer adsorption, neighboring molecules most likely do not interact. Apparently, for condensation of a neighboring molecules and formation of self-assembled siloxanenanolayer, polymolecular adsorption is required. Using equation (5):

where R is the universal gas constant,T is the temperature, 55.5 is the mol water concentration in 1 liter of the solution, calculated the К coefficient, from which, in turn, the VS and AS adsorption heats were calculated (Table 1,2) whose values lie in the range from 20 to 46 kJ/mol which exceeds the values recorded for specific adsorption of organic compounds on metals. However, it is believed that the chemisorption corresponds to the value of 100 kJ/mol. Which was determined using ethiols adsorption on gold [24]. At the same time, it is considered [25] that the heats values of the substimtiveadsorption determined by quartz balance are strongly underestimated, since the interaction energy of the water displaced from the surface on metals with high hydrogen overvoltage is much higher than on precious metals. Therefore, we can say that VS is chemisorbed on the aluminium surface. In addition, the obtained values of adsorption energies correspond to the hydrogen bond energies [26].

In the aminosilane adsorption the analysis of adsorption data using the Langmuir and BET approaches showed (Table 2) not 158 (as we mentioned above), but 19.7 and 25 molecules are located per 1 nm2. This means that the aminosilane molecules are also located horizontally on the aluminium surface. The value of x in the Flory-Higgens equation was 3.6, which indicates the substitutive AS adsorption with the displacement of more than 3 water molecules from the surface. The silane molecule occupies more than 53 adsorption places (Table 2) and. apparently. AS molecules are connected by hydrogen bonds with surface hydroxyl groups of neighboring molecules (Figure 6).

Since alkoxysilanes were adsorbed from the solution, and Langmuir and BET isotherms are usually used for adsorption from gas media, adsorption data were processed using isotherms used to describe adsorption in liquid media. As noted above, during adsorption ethoxysilanes replace water on the surface. So if the interaction between the adsorbate molecules and water occurs on the surface, the reaction is realized:

x H₂O molecules are displaced from the surface. To describe the so-called “substitute” adsorption, a number of authors [9,27] use the Flory-Higgens isotherm:

where С is the concentration. θ is the degree of surface coverage, x is the substitutive factorthe number of substituted water molecules, К is the constant of adsorption equilibrium.

The transformation of equation (7) leads to (8) allowing to present the adsoiption data a linear form and calculatex and K:

From Tables 1 and 2 it can be seen that the adsorption data are straightened in the coordinates of the Flory-Higgens isotherm with high fin correlation coefficients. Processing of equation (8) allowed to find for VS the value of x equal to 2.54 cm (Table 1) and the adsorption equilibrium constant. The data obtained confirm the results described above that VS displaces more than 2.5 water molecules from the surface during adsorption In this case, the silane molecule can occupy more than one adsorption place on the surface. So-called multicenter adsorption isotherms are used to describe such situations [9]. In our case, to determine the orientation of the adsorbed molecules on the surface, the multicenter Langmuir adsorption isotherm was used, where the assumption is that there are n adsorption centers on the surface. Then the classical Langmuir isotherm is converted to (9), and into a linear form (10):

Construction of adsorption data in the linearization coordinates of the equation (10) showed a rectilinear dependence with a correlation coefficient close to one (Table 1), indicating the applicability of the Langmuir multicenter adsorption isotherm to VS. The value of n was 2.54, i.e. on average, each vinylsilane molecule occupies 2.54 adsorption centers, which confirms the assumption of the horizontal orientation of VS molecules on the surface (Figure 5).

Thus, the analysis of VS adsorption isotherms showed that at the initial stages VS adsorption occurs with the displacement of the adsorbed water, while the silane molecule occupies more than two adsorption places on the surface and is located horizontally. However, the question about the transformation of a silane molecule, the interaction of neighboring molecules and the formation of connections with the surface remains open. Temkin,Frumkin and Freundlich isotherms take into account the heterogeneity of the surface, the interaction of adsorbate molecules with each other and dissociative adsorption.

The representation of adsorption isotherm in the coordinates of the surface coverage degree-logarithm concentration showed a linear relationship in a wide range of VS concentrations (Table 1). The obtained data are confirmed by AFM studies of the modified surface. Figure 7 shows atomic force images of the thermally deposited aluminium surface after modification from VS water solutions with concentration of 10^-5 and 10^-1 M. At low concentrations, under conditions of pre-bedding, an uneven distribution of the adsorbate on the metal surface was observed. VS is adsorbed in the form of separate islands with a height of about 2 nm. (Figure 7. a).

Increasing the concentration of the solution to 10^-3 M leads to a more even coverage of the surface with a layer of VS with the thickness of about 1-2 nm which corresponds to 1-2molecular layers (Figure 7. b). Further increase in concentration leads to a polymolecular coverageof the surface and during adsorption

from the 0.1 M of the solution, close-packed vinylsiloxanelayer with the thickness of 10 molecular layers is already being formed (Figure 7. c).

Electron microscopy of aluminium samples with VS adsorbed layer of about 2 nm thickness, deposited from a solution with the concentration of 0.1 M. showed (Figure 8) mostly uniform surface coverage, but also heterogeneities associated, apparently, with the formation of polycondensedsiloxane layers in excess of the monolayer coverage are also observed.

To study the chemical processes occurring in the formation of organosilicon nanolayers, the FT-IR spectroscopic study of thealuminium surface, with a vinylsiloxanenanolayer, of about 3 monolayers thick, deposited from the solution with the concentration of 10^-2 M (Figure 9).

FT-IR spectrum of the aluminium surface treated with VS solution (Figure 9) contains a number of bands related to the vinylsiloxane layer formed during hydrolysis and polycondensation of silane molecules on the surface. Thus, the intense band of 1030 cm^-1 lies in the region of the Si-O-Si group oscillations and the band of 905 cm^-1 corresponds to oscillations in the bonds of the bridge oxygen atom in the Si-O-Si fragment. In addition, a band lying in the region of 1000 cm^-1 was found in the spectrum which can be attributed to the oscillations of -Al-O-Sisurface groups. The bands at 1411 and 1600 cm^-1 lie in regions close to the oscillations of the -CH=CH: double bond and at 2950 cm^-1 - to the oscillations of the CH: bonds of the vinyl group. The band at 770 cm^-1 corresponds to the oscillations of silicon-carbon bonds. A wide but low-intensity band of about 3370 cm’1 lies in tlie region of -OH oscillations of the fragment group Si-ON. The spectrum also showed a low-intensity band at 2900 cm^-1 corresponding to the oscillations of the -OC₂H₅ group.

Spectrum analysis allows us to determine the nature of chemical processes occurring on the surface during the selfassembled nanolayer formation. Thus, the first stage of the process is the hydrolysis of vinylsilane molecules with the silanol formation. It should be noted that no bands corresponding to the Si-O-C groups oscillations were found in the spectrum which indicates the absence of nonhydrolyzed silane molecules on the surface. The low-intensity oscillation band of the -OC₂H₅ fragment indicates trace amounts of ethyl alcohol remaining on the surface after washing. The conducted research allows to propose the following scheme for the formation of self-assembled vinylsiloxanenano layers on the aluminium surface (Figure 10, a-с). At the first stage, when vinylsilane is introduced into the solution, its hydrolysis occurs with the formation of silanol (Figure 10. a). Further, the silanol molecules diffuse to the metal oxide-hydroxide surface,displace the adsorbed solvent molecules (H2Oads) from the surface and interact with the hydroxylgroups of the surface, first forming hydrogen bonds (Figure 10, a). Next, the silanol moleculesenter into the condensation reaction with surface hydroxyl groups. In this case, the formation ofmetal-siloxane Al- O-Si bonds between the adsorbed molecules and the surface layer of the metaloccurs (Figure 10. b). At the last stage, neighboring adsorbed molecules enter the polycondensationreaction, forming hydrolysis-resistant bridging Si-O-Si bonds (Figure 10, c), and a vinylsiloxanenanolayer is formed on the surface, firmly bonded to the surface. Thus, during VS adsorption onthe aluminium surface, self-asssembling of the adsorbate molecules is observed. It is caused byhigh affinity of the vinylsilane molecules to the surface reactive groups of the oxide-hydroxidelayer of the metal and to the neighboring adsorbate molecules. Due to this affinity, silanemolecules without external specific action are self-assembled on the surface so that as a result,an oligomericpolymeric vinyilsiloxanenanolayer associated with surface metal atoms is formed.

However, the thickness of the surface nanolayer is determined by the VS concentration in the solution. Table 3 presents data on

the thickness of siloxane layers formed in VS adsorption from water solutions of different concentrations. At low concentrations the formation of a monolayer was observed on the surface and with its increase above 0.001 M multilayer adsorption begins. Thus, during the VS adsorption from the water solution, a uniform, self-assembled covalent-bound to the surface groups of the metal vinylsiloxanenanolayer is formed on the aluminium surface, the thickness of which can be controlled by changing the application conditions, for example, the vinylsilane solution concentration. As a result, the change in the VS concentration in the solution allows the intended formation of a siloxane nanolayer with any given thickness in the range from 1 to 19.3 molecular layers (Table 3).

The study of the influence of surface organosilicon nanolayers on the corrosion and electrochemical behavior of aluminium was carried out. It is known that a necessary condition for the corrosion processes on the metal is the presence of adsorbed or phase water layer on its surface. In this regard, the interaction between the surface siloxane layer and water was studied in detail. For this purpose, the influence of the surface nanolayer on the water adsorption from the vapor phase was studied (Figure 11). It was shown that one siloxane monolayer reduces both reversible and irreversible water adsorption. A greater effect was observed in the case of a more hydrophobic vinylsiloxane layer (Figure 11, curve 3) when the amount of (both reversibly and irreversibly) adsorbed water was reduced by almost half. The presence of a more hydrophilic amine-containing layer on the aluminium surface led to a 17% decrease in reversibly and 39% irreversibly adsorbed water compared to the aluminium surface without an organosilicon layer (Figure 11. curves 1 and 3).

The data obtained indicate that the polymolecularorganosilicon layer formed as a result of the monolayer chemisorption of organosilane is able to reduce the corrosion rate of aluminium

in water. Therefore, in addition to the water adsorption on the aluminium surface from the vapor phase, the influence of surface siloxane layers on the hydration rate of the Al₂O₃ oxide film on aluminium with the formation of Аl (III) and Al corrosion in distilled water was studied. For this purpose, a single layer of VS and AS (from a diluted water solution) was applied to the quartz resonator covered with a freshly deposited layer of aluminium. The sample was placed in water and the mass change was measured over time, the rate of hydrate formation and aluminium corrosion was estimated by weight gain (Figure 12). It was established that the aluminium surface without the organosilicon layer in the water quickly hydrated (Figure 12, curve 1). So, after 10 minutes of testing, the entire surface of aluminium is covered with a hydroxide (А1(ОН)₃) layer. The presence of a more hydrophobic VS layer (two molecular layers thickness, applied from 0.03 M VS solution) on the surface results in a more than two-fold reduction in the weight gain of the aluminium sample after 12 minutes of testing (Figurel2 curves 1, 2).

The resistometric study of the corrosion behavior of aluminium in chloride-containing electrolytes showed that instant metal dissolution rate was 0.315 and 2.68 mm per year in 0.01 M and 0.1 M sodium chloride solution, respectively (Table 4). The formation of a vinylsiloxane layer on the metal surface, with thickness 1-3 molecular layers (at concentration of deposition solution of VS equal 0.001 M) leads to a significant reduction in the corrosion rate of the aluminium foil (Table 4. Figure 13 a) and inhibition of thermally deposited aluminium dissolution (Figure 13, b). Increasing the layer thickness to 10 molecular layers reduces the inhibitory capacity of the nanolayer. Increasing the number of adsorbed molecules provides an increase in the layer thickness, but reduces its ordering and protective properties.

Thus, it was found that the presence of vinylsiloxane layer on the aluminium surface reduces the metal corrosion. Corrosion tests of aluminium foil samples carried out in a climatic chamber with periodic dipping in a 0.1 M NaCl solution and in a chloridecontaining solution showed the inhibition of metal corrosion both in the atmosphere and in the solution (Figure 14.Table 4).

Scanner-refleetoinetic study carried out on aluminium foil showed that the first corrosion products appeal on the surface after 1 hour of exposure in a 0.1 M sodium chloride solution (Figure15, curve 1). After 20 hours of testing, almost the entire surface is covered with corrosion products. The formation of a self-assembled vinylsiloxanenanolayer with a thickness of 3 molecular layers on the aluminium surface reduces the rate of uniform corrosion (Figure 15, curves 1, 2). Thus, the quantitative processing of reflectograms showed that in the presence of vinylsiloxanenanolayer for 50 hours there were practically no traces of uniform corrosion on the aluminium surface (Figure 15, curve 2) and only about 5% of the surface was occupied by corrosion products.

Thus, it was shown that surface organosilicon nanolayers reduce the rate of uniform aluminum corrosion. However, the presence of the organosilicon layer on the surface can reduce the rate of not only uniform, but also local aluminum dissolution.

Using resistonietricmethod allows us to estimate the change in the thickness of the metal conductor in the process of testing and makes it possible to divide the general and localized metal corrosion, although it is known that in chloride-containing media aluminum is susceptible to pitting corrosion. This is confirmed by visual inspection of samples after testing and corrosion rates (Table 4). In order to separate the contribution of uniform and localized aluminum corrosion and the influence of vinylsiloxanenanolayer on them, the initial stages of corrosion dissolution of thermally deposited aluminum in chloridecontaining solution were studied using the method of quartz nano balance. Figure 16 shows the curves of changes in the mass of the freshly thermally deposited aluminium when the samples

are kept in 0.001 NaCl solution. From Figure 16 it can be seen that in the interaction with water, i.e. when uniform corrosion occurs, the mass of the sample increases during the holding time in water, which may be due to the growth of an insoluble in water oxidehydroxide layer on the metal. Above it is shown that the presence of vinylsiloxanenanolayer on the surface, one monolayer thick, leads to slowing uniform aluminum corrosion.

Introduction of chloride ions into water (Figure 16) causes local dissolution of the metal, which is expressed in the loss of mass due to the release of metal ions into the solution. During the first 5 minutes of testing on pure aluminum, the growth of the oxide-hydroxide film was observed. Further exposure to the solution led to a decrease in the mass of the sample, which was caused by a violation of passivity and local metal dissolution. It was found that for the first, at least. 150 minutes of testing, the loss of the sample mass is not recorded (Figure 16. curve 2).Apparently, vinylsiloxanenanolayer inhibits local corrosion at the initial stage. Thus, it wasshown that the surface vinylsiloxanenanolayer is able to inhibit both uniform and local aluminum corrosion.

The study of the electrochemical behavior of aluminum also showed the inhibition of aluminum pitting by the surface vinylsiloxane layer (Figure 17). So, the critical potential of “purer” (without the organosilicon surface nanolayer) aluminum, determined by the fracture of the anodic polarization curve is -0.4 V. The presence of surface vinylsiloxane monolayer leads to Epitincrease, which indicates the inhibition of the pitting formation process. The reason for this may be a positive charge of the surface due to the amino group protonation at pH< 9.5 (eq. 11) facilitating adsorption of chloride anions, responsible for the violation of passivity, to the metalsurface. As a result, on the one hand, the organosilicon nanolayer must inhibit the localized aluminum dissolution and. on the other hand, a positive charge of the surface must activate it As a result, there was no noticeable influence on the potential of pitting formation (Figure 17 curves 1, 2) [28].

Scanner-reflectometric study conducted on aluminium foil showed (Figure 18) that the first corrosion defects (pittings) appear on the surface after 1 hour of exposure in 0.1 M sodium chloride solution. After 25 hours of testing the density of defects is about 40 units per cm² (Figure18). Further exposure to aggressive electrolyte leads to an increase in the number and to the development of defects formed in the first 25 hours of testing. After 150 hours of testing pitting density was 253 pitting per cm on the surface of “pure” aluminium. It was noted that in the presence of the surface vinylsiloxane layer, pittings occur in the first 50 hours: further exposure of the sample in a chloridecontaining solution (over 50 hours of testing) does not lead to the emergence of “new” pittings. The formation of self-assembled vinylsiloxanenanolayer 3 molecular layers thick on the aluminium surface reduces the rate of localized aluminium corrosion (Figure 18). Thus, after 150 hours of testing, the density of pitting on the silicicated surface is more than 4 times less than on the “pure” aluminium and is 53 pittings per 1 cm² (Figure 18. curve 2). Quantitative processing of reflectograms showed that on the aluminiumsurface in the presence of vinylsiloxanenanolayer (Figure 18. curve 2), corrosion products occupyonly 7% of the surface, while on “pure” aluminium after 50 hours of testing, corrosion products occupy the entire (100%) surface of the sample[30,31].

Electron microscopy of the samples of pure aluminum and aluminum with vinylsiloxanenanolayer of about 2 nm thick, applied from a solution with a concentration of 0.01 M VS, showed that after 10 days of preserving in a chloride-containing solution on the aluminum surface (both sprayed and foil), traces of corrosion products, occupying 18% and 29% of the surface for thermally deposited metal and foil, respectively were found. Corrosion products were also found on samples with siloxane nanolayer, but the proportion of the affected surface was about 2%. Localized corrosion defects (pits) with a diameter of 5 to 8 micrometers for the sprayed aluminum(Figure 19, a), from 7 to 17 μm for foil (Figure 19, b) were found on the pure aluminum. Thepresence of a vinylsiloxanenano layer on the surface leads to uniform corrosion (Figure 19. c). Nolocal defects were found after 10 days of testing, which also indicates inhibition of local metalcorrosion.

XSMA spectra of aluminum coated with a vinylsiloxanenanolayer (Figure 19. d) showed that after conducting tests on the surface, quality elemental composition of the surface layer is almost not changed compared to the original spectra. Thus, in the regions damaged by corrosion, the presence of silicon and carbon bands was established and the intensity of Si and С bands was close to those obtained immediately after the application of the nanolayer. This confirms the conclusion about the strong interaction between the vinylsiloxanenanolayer and the surface groups of aluminium and also speaks about the stability of the nanolayer to the action of water and corrosionactive components.

1. It was shown that adsorbing, silanes displace adsorbed water from the surface, occupying more than 2.5 adsorption places on the surface.
2. Aminosilan forms a densely packed layer vertically oriented to the surface at the initial stage of adsorption, the layer thickness is not less than 3 molecular layers at the solution concentration of 10-5 M.
3. It was found that in the case of vinylsilane there is a dissociative adsorption on the aluminium surface. The ethoxysilane molecule is hydrolyzed to form 1 silanol molecule and 3 ethyl alcohol molecules.
4. In the initial period of adsorption, VS molecules on the surface are horizontally oriented and do not interact with neighboring molecules in the region of fillings up to one monolayer.
5. Adsorption heats were calculated using different adsorption models. It was shown that theVS chemosorbes, forming a uniform, self-assembled, covalently bound with surface groups of metal vinylsiloxanenanolayer on the aluminium surface, the thickness of the nanolayer can be controlled by changing the processing conditions, e.g. the concentration of the vinylsilane solution.
6. It was estabhshed that an ordered vinylsiloxanenanolayer, up to 5 molecular layers tick effectively slows down uniform and localized corrosion of aluminium in chloride-containning electrolytes.
7. It was shown that after 10-day corrosion tests vinylsiloxanenanolayer was preserved on the aluminium surface, indicating its resistance to water and corrosive components.

This research was funded by the Basic research program of the Presidium of the Russian Academy of Sciences №4P, “Urgent Problems of Surface Physicochemistry and Creation of New Composite Materials. Nanostructured coatings for electronics, photonics, alternative energy sources and materials protection”

1. Shreir LL, Jarman RA, Burstein GT. Corrosion. Metal/Environment Reactions. Vol. 1 3-d Edition:.Newnes-Butterworths. London. Boston. 1994:1424.
2. Kaesche H. Die Korrosion der Metalle: physikalisch-chemischePrinzipien und aktuelleProbleme. Springer-Verlag. Berlin. New York. 1979.
3. Kuznetsov Yu I. Role of solution anions in depassivation of aluminum and inhibition of corrosion. Protection of metals. 1984;20:(3):359-372.
4. Marcus P, Maurice V, Strehblow HH. Localized corrosion (pitting): A model of passivity breakdown including the role of the oxide layer nanostructure. Corrosion Science. 2008;50(9):2698-2704.
5. Soltis J. Passivity breakdown, pit initiation and propagation of pits in metallic materials -Review. Corrosion Science. 2015;90:5-22.
6. Chalfoun D, Chocron М, Kappes MA, Rebak RB. Localized Corrosion of UNS A95052 Aluminum Alloy for Application in Multi-Effect Desalinator Plants. NACE – International Corrosion Conference Series. 2018;74(9):1023-1032.
7. Leuenberger M, Faller М, Richner P. Runoff of copper and zinc caused by atmospheric corrosion. Materials and Corrosion. 2002;53:157-224.
8. White Paper Strategy for a Future Chemical Policy of the Commission of the European Communities. Brussels, Directive 2011/65/EU of the European Parliament.
9. Nanna ME, Bierwagen GP. Mg-rich coatings: A new paradigm for Cr-free corrosion protection of A1 aerospace alloys. JCT Research. 2004;1:69-80.
10. Twite RL, Bierwagen GP. Review of alternatives to chromate for corrosion protection of aluminum aerospace alloys. Progress in Organic Coatings. 1998;33(2):91-100.
11. Pluddemann EP. Silane Coupling Agents. 2-nd Edition. Plenum Press. New York 1991.
12. Semenov VV. Preparation, properties and applications of oligomeric and polymeric organosilanes. Russ Chem Rev. 2011;80(4):313-339.
13. Petrunin MA, Nazarov AP, Mikhailovskii Yu N. Electrochemical and corrosion behavior of steel, magnesium, and aluminum primed with silanes. Protection of Metals. 1990;26:749-754.
14. Hayashi К, Saito N, Sugimura H, Taka O, Nakagiri N. Surface potential contrasts between silicon surfaces covered and uncovered with an organosilane self-assembled monolayer. Ultramicroscopy. 2002;91(1-4):151-156.
15. Russian State Standards. GOST 11069-200. Primary aluminium grade.
16. Heppel М, Cateforis E. Studies of copper corrosion inhibition using electrochemical quartz crystal nanobalance and quartz crystal immittance techniques. ElectrochemActa. 2001;46(24-25):3801-3815.
17. Mikhailovskii Yu N, Marshakov AI, Ignatenko VE, Pelrunin MA, Petrov NA, Bukhovtsev VM. Monitoring of the Corrosion State of Underground Pipelines with Corroding Resistors-Transducers. Protection of Metals. 2000;36:583-587.
18. Kotenev VA, Petnmin MA, Maksaeva LB, TsivadzeAYu. 3D Visualization of the Dissolution Products of a Metal in the Near-Electrode Layer at the Metal-Solution Interface. Protection of Metals and Physical Chemistry of Surfaces. 2005;41(6):507-520.
19. Petrunin MA, Maksaeva  LВ, Yurasova  ТA, Gladkikh NA, Terekhova  ЕV, et al. Adsorption of Vinyl Trimethoxvsilane and Formation of Vinyl SiloxaneNanolayers on Zinc Surface from Aqueous Solution. Protection of Metals and Physical Chemistry of Surfaces. 2016;52(6):964-971.
20. Szklarska S. Pitting and Crevice Corrosion. NACE International. 2005.
21. McCaffetry E. Introduction to Corrosion Science. Springer-Verlag. NY. 2010:575.
22. Horcas L, Fernandez R, Rodriguez JM, Colchero J, Herrero J, Baro. A.M.:WSXM: A software for scanning probe microscopy and a tool for nanotechnology. Review of Scientific Instruments. 2007; 78.013705-1 - 013705-7.
23. Kern P, Landolt D. Adsorption of an Organic Corrosion Inhibitor on Iron and Gold Studied with a Rotating EQCM. The Electrochemical Society. 2001;148:B228-B235.
24. Nuzzo RG, Fusco FA, Allara DL. Spontaneously organized molecular assemblies. 3. Preparation and properties of solution adsorbed monolayers of organic disulfides on gold surfaces. J American Chem. Society. 1987;109:2358-2368.
25. Land K. Adsorption of organic corrosion inhibitors on iron in the active and passive state. A replacement reaction between inhibitor and water studied with the rotating quartz crystal microbalance. ElectrochimicaActa. 2001;47(4):589-598.
26. Grabowsky SJ. Hydrogen Bonding—New Insights.. Springer. Dordrecht. The Netherlands. 2006:5.
27. El-Awady AA, Abd-El-Nabey BA, Aziz SG. Kinetic-Thermodynamic and Adsorption Isotherms Analyses for the Inhibition of the Acid Corrosion of Steel by Cyclic and Open-Chain Amines. The Electrochemical Society. 1992;139(8):2149-2154.
28. Flamini DO, Truebab М, Trasatti SP. Aniline-based silane as a primer for corrosion inhibition of aluminium. Progress in Organic Coatings. 2012;74:302-310.
29. Petnmin MA, Maksaeva LB, Yurasova TA, Terekhova EV, Kotenev VA, et al. The directional formation and protective effect of self-assembling vinyl siloxanenanolayers on copper surface. Protection of Metals and Physical Chemistry of Surfaces. 2012;48:656-664.
30. Kelly RG, Scully JR, Shoesmith DW, Buchheit   RG. Electrochemical techniques in corrosion science and engineering. Marcel Dekker, Inc. NY. 2002:426.