USE OF SCANNING ELECTROCHEMICAL MICROSCOPY AS A PREDICTIVE TECHNIQUE IN A SALT FOG CORROSION TEST

Abstract

The use of the scanning electrochemical microscopy to predict the corrosion resistance results which would be obtained for a surface S1 having undergone an anticorrosion treatment if the surface S1 was subjected to a salt fog corrosion test, which use comprises an analysis of the surface S1 by scanning electrochemical microscopy.

Claims

1. A process for predicting corrosion resistance results which would be obtained for a surface S1 having undergone an anticorrosion treatment if the surface S1 was subjected to a salt fog corrosion test, the process comprising an analysis of the surface S1 by scanning electrochemical microscopy.

2. The process of claim 1, wherein the analysis of the surface S1 is carried out in feedback mode.

3. The process of claim 2, wherein the analysis of the surface S1 comprises a scanning of the surface S1 with a scanning electrochemical microscopy probe and wherein a tip of the probe is maintained at a constant distance d from the surface S1 during scanning the surface.

4. The process of claim 3, wherein the tip of the probe follows, during scanning the surface S1, a trajectory comprising one or more rectilinear portions.

5. The process of claim 4, wherein the tip of the probe follows, during scanning, a trajectory comprising several rectilinear portions which are parallel to each other.

6. The process of claim 3, wherein the analysis of the surface S1 comprises at least the steps consisting in: a) selecting a point directly above the surface S1; b) bringing, at the point selected in step a), the tip of the probe to the distance d; and c) scanning the surface S1 with the probe from the point selected in step a) by maintaining the tip of the probe at the distance d and measuring a current at the probe during scanning.

7. The process of claim 2, wherein the analysis of the surface S1 comprises positioning a scanning electrochemical microscopy probe at several points directly above the surface S1 and wherein a tip of the probe is located at a constant distance d from the surface S1 at each positioning point.

8. The process of claim 7, wherein the positioning points are randomly selected.

9. The process of claim 7, wherein the analysis of the surface S1 comprises at least the steps consisting in: a) selecting the probe positioning points; and b) bringing, for each positioning point, the tip of the probe to the distance d and measuring a current at the probe at each positioning point.

10. The process of claim 1, which comprises an use of an electrolyte comprising a redox mediator in the reduced state.

11. The process of claim 10, wherein the redox mediator is ferrocyanide, ferrocene, decamethylferrocene or ferrocene dimethanol.

Description

BRIEF DESCRIPTION OF FIGURES

[0096] FIG. 1, already commented on, schematically illustrates a typical example of SECM equipment as used in the laboratory.

[0097] FIG. 2 shows the results of a SECM analysis by mapping as obtained for a sample of a non-anodised Al2024 aluminium alloy.

[0098] FIG. 3 shows the results of an SECM analysis by mapping for a sample of an anodised Al2024 aluminium alloy.

[0099] FIG. 4 shows, seen from the front, a sample of a non-anodised Al2024 aluminium alloy as obtained after a 168-hour neutral salt fog test.

[0100] FIG. 5 shows, seen from the front, a sample of an anodised Al2024 aluminium alloy as obtained after a 168-hour neutral salt fog test.

[0101] FIG. 6 shows the results of SECM analyses by linescan as obtained for, on the one hand, a sample of a non-anodised Al2024 aluminium alloy (curve 1) and, on the other hand, a sample of an anodised Al2024 aluminium alloy (curve 2); in this Figure, the ordinate axis corresponds to the normalised current I/I.sub.inf measured at the probe, while the abscissa axis corresponds to the distance D, expressed in ?m, travelled in line by the probe.

[0102] FIG. 7 is an image taken under an optical microscope of a first sample of an Al2024 aluminium alloy having undergone an anticorrosion treatment by trivalent chromium conversion.

[0103] FIG. 8 is an image taken by SEM of a second sample of an Al2024 aluminium alloy having undergone an anticorrosion treatment by trivalent chromium conversion.

[0104] FIG. 9 shows the three polarisation curves, denoted 1, 2 and 3 respectively, obtained by subjecting the same sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, to three successive cyclic voltammetry tests, as well as the Tafel lines plotted from these curves; in this Figure, the ordinate axis corresponds to the logarithm of the absolute value of the current, denoted log|I|, while the abscissa axis corresponds to the potential, denoted E and expressed in volts, applied to the sample of aluminium alloy.

[0105] FIG. 10 shows, seen from the front, a sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, after having subjected this sample to three successive cyclic voltammetry tests.

[0106] FIG. 11 shows the polarisation curves, respectively 3 and 4, obtained by subjecting two different samples of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, to a cyclic voltammetry test, as well as the Tafel lines plotted from these curves; in this Figure, the ordinate axis corresponds to the logarithm of the absolute value of the current, denoted log|I|, while the abscissa axis corresponds to the potential, denoted E and expressed in volts, applied to the sample of aluminium alloy.

[0107] FIG. 12 shows the results of three SECM analyses by linescan, denoted 1, 2 and 3 respectively, having been carried out on the same area of a sample of an Al2024 aluminium alloy previously anticorrosion treated by trivalent chromium conversion; in this Figure, the ordinate axis corresponds to the normalised current I/I.sub.inf measured at the probe, while the abscissa axis corresponds to the distance D, expressed in ?m, travelled by the probe during each analysis.

[0108] FIG. 13 shows the results of an SECM analysis by mapping as obtained for a first sample of an Al2024 aluminium alloy previously anticorrosion treated by trivalent chromium conversion.

[0109] FIG. 14 shows the results of an SECM analysis by mapping as obtained for a second sample of an Al2024 aluminium alloy previously anticorrosion treated by trivalent chromium conversion.

[0110] FIG. 15 shows, seen from the front, a first sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, as obtained after a 168-hour neutral salt fog test.

[0111] FIG. 16 shows, seen from the front, a second sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, as obtained after a 168-hour neutral salt fog test.

[0112] FIG. 17 shows the results of an SECM analysis by mapping as obtained for a third sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion.

[0113] FIG. 18 shows the results of an SECM analysis by statistics as obtained for a third sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion; in this Figure, the ordinate axis corresponds to the number N of iterations while the abscissa axis corresponds to the normalised current I/I.sub.inf measured at the probe.

[0114] FIG. 19 shows, seen from the front, a third sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion, as obtained after a 168-hour neutral salt fog test.

DETAILED PRESENTATION OF PARTICULAR IMPLEMENTATIONS

Example 1: Predictions by SECM Analyses by Mapping on Reference Samples

[0115] Samples of two Al2024 aluminium alloys, respectively anodised and non-anodised, whose corrosion resistance is known and which can therefore be used as references, are subjected to SECM analyses by mapping, using the following operating conditions: [0116] probe: UME consisting of a platinum wire 12 cm long and 50 ?m in diameter sealed in a glass capillary; [0117] reference electrode: Ag/AgCl; [0118] counter electrode: gold; [0119] liquid electrolyte comprising potassium chloride as salt and ferrocene dimethanol (Fe(MeOH).sub.2) at 1 mmol/L as redox mediator; [0120] potential applied by potentiostat to the probe: 0.6 V; [0121] potential applied to the samples: none (OCP mode); [0122] distance d (probe tip/sample surface): 10 ?m; [0123] surface scanning speed by the probe: 10 ?m/s.

[0124] The anodised alloy has a protective layer made of aluminium oxide, about 8 ?m thick, which gives it high corrosion resistance.

[0125] The non-anodised alloy has, itself, a low corrosion resistance.

[0126] The obtained maps for samples of each of the two alloys are illustrated in FIGS. 2 and 3, FIG. 2 corresponding to a sample of the non-anodised alloy and FIG. 3 corresponding to a sample of the anodised alloy. On the right margin of each of these figures, there is represented a scale of the values of the normalised current, denoted I/I.sub.inf, measured at the probe, which corresponds to the ratio between the current I actually measured at the probe during the scanning of the surface of the samples and the current I.sub.inf measured at the probe when the tip of the probe is at infinity from the surface of the samples.

[0127] As shown in these figures, the obtained map for the non-anodised alloy sample shows that this sample has a uniformly high surface conductivity (with I/I.sub.inf>0.9) which may result in a sensitivity of the surface of this sample to the pitting corrosion, while the map obtained for the anodised alloy sample shows that this sample has, on the contrary, a uniformly low surface conductivity (with I/I.sub.inf<0.4) testifying a priori to the quality of passivation of the protective layer contained in this sample.

[0128] In order to check whether these maps allow predicting that salt fog tests will reveal that only the anodised alloy samples are compliant, samples of both types of alloy are subjected to salt fog tests under the following conditions: [0129] neutral salt fog (NaCl solution); [0130] temperature and pressure prevailing in the test chamber: 35? C.1 bar; [0131] NaCl concentration of the sprayed solution: 50 g/L; [0132] inlet flow rate of the solution sprayed into the test chamber: 1.8 L/h; [0133] flow rate of the collected sprayed solution (condensate): 2 mL/h; [0134] duration of the tests: 168 hours.

[0135] The results of these tests are presented in Table I below and in FIGS. 4 and 5, FIG. 4 corresponding to a sample of the non-anodised alloy and FIG. 5 corresponding to a sample of the anodised alloy.

[0136] It should be noted that a sample is considered to be compliant if it has less than 2.5 pitting/dm.sup.2 after 168 hours of exposure to salt fog according to the standard NF EN ISO 9227.

TABLE-US-00001 TABLE I Samples Number of pitting/dm.sup.2 Results Non-anodised Al2024 >50 non-compliant Anodised Al2024 <2 compliant

[0137] This table and these figures show that the results of the salt fog tests are in perfect agreement with the predictions obtained by the SECM tests in mapping mode, thus confirming the possibility of using the SECM as a predictive technique for salt fog tests.

[0138] It is important to note that the results of the salt fog tests required 7 days of waiting while the SECM analyses by mapping have, themselves, been carried out in a single half-day.

Example 2: Predictions by SECM Analyses by Linescan on Reference Samples

[0139] Samples of the two Al2024 aluminium alloys, respectively anodised and non-anodised, tested in the example 1 above are also subjected to SECM analyses by linescan in order to check whether this mode of implementation of an SECM analysis, which is faster than mapping, also allows reliably predicting salt fog test results.

[0140] These analyses are carried out using the same operating conditions as those indicated in the example 1 above.

[0141] The results are shown in FIG. 6.

[0142] In this Figure, the curve 1, which corresponds to the non-anodised alloy sample, shows a very high feedback (with I/I.sub.inf close to 1) testifying to the sensitivity of the surface of this sample to corrosion. It also reveals significant variations in the normalised current I/I.sub.inf, indicating the presence of corrosion spots.

[0143] Conversely, the curve 2, which corresponds to the anodised aluminium alloy sample, shows very low feedback (with I/I.sub.inf substantially equal to 0.3 over the entire distance travelled by the probe) as well as the absence of variations in the surface conductivity.

[0144] These results are therefore in perfect agreement with the maps shown in the example 1 above and demonstrate that the use of the SECM by linescan allows obtaining predictive results which are just as relevant as those obtained by a SECM by mapping.

Example 3 Predictions by SECM Analyses by Mapping on Samples which are Anticorrosion Treated by Trivalent Chromium Conversion

[0145] In order to demonstrate the industrial interest of the invention, a series of analyses (SEM, Tafel lines, salt fog tests, SECM analyses) is carried out not on reference samples as in Examples 1 and 2 above, but on two series of Al2024 aluminium alloy samplesreferred to as series 3 and 4 hereinafterhaving been anticorrosion treated by Cr.sup.|I| conversion.

[0146] Indeed, this type of anticorrosion treatment being still, to date, less well controlled than anodising, it leads to parts whose corrosion resistance varies from one batch of treated parts to the other, so that salt fog tests are still currently essential to check whether these parts are compliant.

3.1 Anticorrosion Treatment by Cr.SUP.|I| Conversion

[0147] The anticorrosion treatment by Cr.sup.|I| conversion is performed in three steps: [0148] a first step which consists in immersing the alloy samples for 5 minutes in a bath comprising 40 vol. % of Socosurf? 1858, 10 vol. % of Socosurf? 1806 and 50% demineralised water, this bath being maintained at 50? C. under stirring, then rinsing the samples with distilled water; [0149] a second step which consists in immersing the alloy samples for 10 minutes in a bath comprising 35 vol. % of Socosurf? TCS and 65 vol. % of demineralised water, this bath being maintained at 40? C. under stirring, then rinsing the samples with demineralised water; and [0150] a third step which consists in immersing the alloy samples for 5 minutes in a bath comprising 10 vol. % of Socosurf? PACS, 6 vol. % of hydrogen peroxide (H.sub.2O.sub.2) at 35% and 8 vol. % of demineralised water, this bath being maintained at ambient temperature under stirring, then rinsing the samples with demineralised water and, finally, drying them with compressed air.

[0151] At the end of this treatment, all samples have, in principle, a protective layer mainly made of chromium and zirconium oxides, about 200 nm thick.

3.2 Analysis of the Samples by Optical Microscopy

[0152] As shown in FIGS. 7 and 8, which correspond to images of the surface of a sample from the series 3 (FIG. 7) and the surface of a sample from the series 4 (FIG. 8), no difference between the surfaces of the two samples is detectable by optical microscopy. The anticorrosion layers of the two samples appear to have similar morphologies.

3.3 Method of Tafel Lines

[0153] Cyclic voltammetry tests are carried out on samples belonging to the series 3 and 4 in order to plot and exploit the Tafel lines since this characterisation method is regularly used in R&D to assess the current and the corrosion potential of a metal material.

[0154] These tests are carried out using the following operating conditions: [0155] working electrode: sample of Al2024 aluminium alloy which is anticorrosion treated by Cr.sup.|I| conversion; [0156] reference electrode: Ag/AgCl; [0157] counter-electrode: gold; [0158] electrolyte: KCl 0.1 mol/L; [0159] scanning speed: 10 mV/s.
* Successive Cyclic Voltammetry Tests on the Same Sample from the Series 4

[0160] Three cyclic voltammetry tests are carried out successively on the same sample from the series 4.

[0161] The polarisation curves log|I|=f(E) as well as the Tafel lines are then plotted.

[0162] The results are shown in FIG. 9 in which the curves 1, 2 and 3 correspond respectively to the first, second and third cycles.

[0163] As shown in this Figure, very different results are obtained from one cycle to another, which highlights an evolution of the surface of the sample during manipulation. The passivated layer is stimulated during the measurement by the polarisation of the sample, which generates modifications of its surface state.

[0164] These modifications are moreover visible to the naked eye, as shown in FIG. 10, which is a photograph of the surface of the sample taken at the end of the third voltammetry cycle and on which a blackened area is visible at the exact location where the measurement has been performed.

* Cyclic Voltammetry Tests on Samples from the Series 3 and 4

[0165] A cyclic voltammetry test is carried out on a sample from each of the series 3 and 4 with the aim of checking whether the method of Tafel line allows differentiating the corrosion resistance of the protective layers of these samples.

[0166] Again, the polarisation curves log|I|=f(E) and the Tafel lines are then plotted.

[0167] The results are shown in FIG. 11 in which the curves are respectively denoted 3 and 4 depending on the series to which the sample belongs.

[0168] As this figure shows, very similar results are obtained on the two samples (very close corrosion potential and current).

[0169] However, as highlighted in point 3.5 below, salt fog tests on these two samples lead to very different results. This highlights that the exploitation of the Tafel lines via a cyclic voltammetry is not an adequate method to differentiate between two samples having undergone the same type of anticorrosion treatment, but nevertheless having different corrosion resistances.

3.4 SECM Analyses

[0170] * SECM Analyses by Linescan on the Same Sample from the Series 4

[0171] A sample from the series 4 is subjected to three successive SECM analyses by linescan, the three analyses being carried out on the same area of the sample and using the same operating conditions as those described in the example 1.

[0172] The results are shown in FIG. 12 in which the curves 1, 2 and 3 correspond respectively to the first, second and third scans of the sample area by the probe.

[0173] As shown in this Figure, the curves 1, 2 and 3 are almost identical, which means that an SECM analysis does not induce any alteration of the surface of the sample.

* SECM Analyses by Mapping on Samples from the Series 3 and 4

[0174] SECM analyses by mapping are carried out on samples belonging to the series 3 and 4 using the same operating conditions as those described in the example 1.

[0175] The maps obtained for samples from each of the two series are illustrated in FIGS. 13 and 14, FIG. 13 corresponding to a sample from the series 3 and FIG. 14 corresponding to a sample from the series 4. On the right margin of each of these figures, a scale of the values of the normalised current I/I.sub.inf, which is measured at the probe, is represented.

[0176] These maps show that the sample from the series 4 has a surface conductivity which is uniformly higher (with I/I.sub.inf?0.7) than that of the sample from the series 3, which may result in a greater sensitivity of the surface of this sample to corrosion. The map of the sample from series 4 further reveals more conductive areas which are potentially sensitive to pitting.

[0177] The sample from the series 3 has, itself, a uniformly low surface conductivity (with I/I.sub.inf?0.4) which testifies to the passivation quality of the anticorrosion layer of this sample.

[0178] These obtained maps allow predicting that the results of salt fog tests will be different for the two series of alloys and that the alloy from the series 3 will have a corrosion resistance which is higher and assuredly of high quality while the alloy from the series 4 will present a lower corrosion resistance (this corrosion resistance must however be better than that of an untreated Al2024 alloy for which values of I/I.sub.inf in the range of 1 have been obtainedsee examples 1 and 2).

[0179] Consequently, the predictions are as follows: [0180] alloy from the series 3.fwdarw.compliant with the salt fog; [0181] alloy from the series 4.fwdarw.non-compliant with the salt fog.

3.5 Salt Fog Tests

[0182] Samples from the series 3 and 4 are subjected to 168-hour salt fog test under the same operating conditions as those described in the example 1 above.

[0183] The results of these tests are presented in Table Il below and in FIGS. 15 and 16, FIG. 15 corresponding to a sample from the series 3 and FIG. 16 corresponding to a sample from the series 4.

TABLE-US-00002 TABLE II Samples Number of pitting/dm.sup.2 Results Al2024 from the series 3 <2 Compliant Al2024 from the series 4 ~10 Non-compliant

[0184] This table and these figures confirm the predictions made above and, thereby, that the corrosion resistance diagnostics carried out by SECM are extremely reliable.

Example 4: Predictions by SECM Analysis by Statistics

[0185] An SECM analysis by statistics is carried out on a sample of an Al2024 aluminium alloy, previously anticorrosion treated by trivalent chromium conversion.

[0186] The used operating conditions are identical to those described in the example 1 with the slight difference that there is no scanning of the surface of the sample by the probe, but a positioning of the probe at different points located directly above the surface of the sample, the tip of the probe being 10 ?m from said surface at each positioning point. The points are randomly selected by a computer program developed for this purpose.

[0187] By way of comparison, an SECM analysis by mapping is also performed on the same sample.

[0188] The results of these two analyses are shown in FIGS. 17 and 18, FIG. 17 corresponding to the map obtained by the analysis in mapping mode and FIG. 18 corresponding to the classification of the values of the normalised current I/I.sub.inf measured at the probe during analysis in statistics mode.

[0189] These figures highlight the existence of a concordance between the results obtained by mapping and those obtained by statistics which, both, show that the surface conductivity of the sample is generally high, with values I/I.sub.inf between 0.7 and 0.8, which allow predicting that this sample will be considered non-compliant at the end of a salt fog test.

[0190] The sample is therefore subjected to a 168-hour salt fog test under the same operating conditions to those described in the example 1 above.

[0191] The results of this test are illustrated in FIG. 19 which shows a significant number of pitting (50/dm.sup.2), confirming the non-compliance of the sample and, thereby, the predictive value of the SECM analysis by statistics. 5

CITED REFERENCES

[0192] [1] Jiang Mei-Yan et al., Corrosion Science 2015, 92, 118-126 [0193] [2] Junsheng Wu et al., Materials 10(4), 2017, 426 [0194] [3] Dongdong Peng et al., Journal of Coatings Technology and Research 2016, 13, 837-850 [0195] [4] Yi Xiao et al., Arabian Journal for Science and Engineering 2018, 43(7), 3577-3584 [0196] [5] Peng Guangchun et al., Progress in Organic Coatings 2020, 140, 1-10 [0197] [6] WO-A-2020/012097