Corrosión study of the passive film of amorphous Fe-Gr-Ni-( Si , P , B ) alloyŝ * ^

Amorphous Fe62CrioNÍ8X2o (X = P, B, Si) alloys in O.OIM HCl solution have been investigated by means of standard electrochemical measurements in order to evalúate their corrosión resistance. The study reveáis that the best corrosión behaviour is given by the Si containing amorphous alloy. X-ray photoelectron spectroscopy (XPS) and Auger electrón spectroscopy (AJES) have been employed to study the composition of the passive layers, formed on the surface of the different amorphous alloys. The results on Fe52CrioNÍ8SÍ2o show that a protective passive film, mainly consisting of oxidized chromium, greatly enhances its corrosión resistance.


INTRODUCTION
Rapidly quenched amorphous metal-metalloid alloys containing a certain amount of chromium are malcriáis that exhibit extremely high corrosión resistance (1)(2)(3)(4)(5)(6).They do not suffer pitting corrosión in acidic chloride solutions even under anodic polarization and also show a significantly high resistance to crevice corrosión (7)(8)(9).This corrosión behaviour depends on the presence of a protective passive film and, therefore, on its composition.
Passive films on amorphous iron-chromium base alloys are constituted mainly of hydrated chromium oxihydroxide (10).The corrosión resistance and protection of amorphous alloys improve highly with the content of this chromium oxihydroxide film (11).The high reactivity of amorphous alloys is the responsible of the chromium enrichment in the passive film, and their chemically homogeneous nature provides high passivating ability and corrosión resistance.For this reason, alloys which form stable passive films, most notably Crcontaining alloys, have been studied quite extensively (4, 8 and 12-14).
In the present work, we have performed electrochemical measurements along with X-ray photoelectron spectroscopy (XPS), and Auger electrón spectroscopy (AES) of amorphous Fe52CrioNÍ8X2o (X=P, B, Si) alloys, and investigated the role of the passive film composition on the corrosión behaviour of the samples.

EXPERIMENTAL
Alloy ingots were prepared by are melting in a water cooled copper crucible under He atmosphere.From these ingots, amorphous alloy ribbons of about 5 mm and 20-30 \xm thickness were prepared by the planar flow casting method.Ribbon compositions were checked by electrón probé microanalysis.The amorphous structure of the samples was confirmed by X-ray diffraction.Prior to the electrochemical experiments, the amorphous alloy ribbons were degreased in acetone and rinsed with distilled water.
Electrochemical measurements were made at room temperature using as electrolyte O.OIM HCl.An área of 2 cm^ of the amorphous alloy ribbons, acting as working electrodes, was exposed in this solution.A saturated calomel reference electrode and a stainless steel wire as counter electrode were used for the evaluation of the corrosión behaviour.
XPS and Auger spectra were recorded under ultra-high vacuum (UHV) conditions with a VG-CLAM hemispherical electrón energy analyzer equipped with a special lens to probé small áreas on the sample (< 1 mm^).Additionally, both an electrón gun with a spot size of 50 fxm and 3 keV acceleration energy as excitation source for the Auger spectra and a Mg X-ray source was used for the XPS spectra were used.The base pressure in the UHV-chamber during measurements was better than 10-9 nibar.Samples were cleaned by sputtering with an ion gun operating with 2 ¡xA sample current and an ion energy of 3 keV.

RESULTS AND DISCUSSION
Figure 1 shows the valúes of the corrosión intensity, i^^^^.versas testing time for three different amorphous alloys, Fe62CrioNÍ8X2o (X = P, B, Si).These data are obtained from the measurements of polarization resistance, R^.For this purpose, potential steps of 10 and 50 mV were applied and the intensity A/ was registered.Under a 10 mV step, the response in intensity for most of the samples diminished very fast to zero before any stationary stage was achieved.Furthermore, in some cases the obtained valúes approached the detection limit of the equipment, therefore it was necessary to apply a step of 50 mV.The valué of the polarization resistance R^ is deduced from the quotient R^= LEIM, where A£ is the step of the potential applied at the corrosión potential and A/ is the resulting cuiTent for a testing time of 60 s.After determining the polarization resistance, the corrosión intensities, Z^^,,,., could be immediately determined by applying the well-known Stern-time (days) FiG.1.-Corrosión current density, /^orr (A/cm^) as a function of time for the three studied alloys: Fe62CrioNÍ8X2o(X = P,B,Si).

FiG. 1.-Densidad de corriente de corrosión, i^^^^ (A/cm^) en función del tiempo para las tres aleaciones estudiadas: Fe^2^rjQNigX2o (X = P, B, Si).
Geary equation (i^^^ = B/R^), where J5 is a constant containing Tafel slope informatiorí.The valué of B for each amorphous alloy was previously deduced from the anodic and cathodic slopes of the polarization curves.
From the data represented in figure 1, we can clearly see that the lower corrosión intensity corresponds to Fe52CrioNÍ8SÍ2o amorphous alloy, presenting /^^j-r valúes at least one order of magnitude lower than the other alloys.Therefore, in a first step, with the polarization resistance method we observe that the amorphous alloy containing Si gives the best corrosión behaviour.
The corrosión resistance can be also evaluated by means of the altern current impedance method, also known as electrochemical impedance spectroscopy (EIS).For these measurements a sinusoidal potential variation with 10 mV amplitude was applied within a frequency range from 0.01 Hz to 64,000 Hz and the resulting current was collected and measured.Figure 2  behaviour, in agreement with the polarization resistance method, is given by the Fe52CrioNÍ8SÍ2o amorphous alloy, shown in the figure inset.Fe62CrioNÍ8B2o amorphous alloy shows a Nyquist plot semicircle that is ñot complete.In these cases 7?t valúes can approximately be obtained from the lower frequency impedance data.The Nyquist plot for Fe52CrioNÍ8P2o amorphous alloy exhibits two different ares, a small semicircle for high frequency valúes and another capacitive are, easier to extrapólate, for low frequency valúes.The presence of this two semicircles is related to the fact that different processes are taking place on the working electrode surface.As seen previously in the literature (15), this result suggests that the system can not be defined simply by a Randless circuit.The semicircle at high frequency valúes is associated with the formation of a layer on the working electrode surface.The semicircle at low frequencies is direcdy associated with the corrosión process.From the extrapolation of this last semicircle we can calcúlate the R^ parameter {R^ = 3.74 10^ Q cm2).This valué indicates that the formed layer is a low-protecting film.Table 1 resumes the valúes of R^ and R^ for these different studied alloys.As mentioned above, taking the R^ valúes it is possible to obtain the i^^^ data by applying the Stern-Geary equation.From the represented data it is clear the best corrosión behaviour shown by the Fe52CrioNÍ8SÍ2o amorphous alloy.
In order to clarify the nature of the different corrosión resistance valúes observed in these alloys we performed XPS and AES measurements with the samples showing the two extreme corrosión behaviours.Therefore, Fe52Cr|QNigSÍ2o and Vt(^2^^\()^'H^2Q alloy specimens, without any Figure 3 shows Cr 2p3/2 core-level emission of Fe52CrioNigSÍ2o after sputtering.As we mention above, the dashed subspectrum gives the signal corresponding to oxidized chromium Cr3+ at a binding energy (BE) of 576.7 eV.From figure 3 we can observe that there was no signal corresponding to elemental chromium, located at 574.5 eV BE.Therefore, it is clear from this Cr 2p photoemission spectrum that the chemical state of chromium in the passive layer formed on the surface of the Fe52CrioNigSÍ2o amorphous alloy is mainly a oxidized Cr^+ state and the content of elemental Cr is negligible.
Figure 4 represents the Si 2p3/2 emission where the same procedure was applied.The Lorentzian located at 103.5 eV BE, represented by the dashed curve, corresponds to Si in form of silica.From this figure we observe that there is no emission corresponding to elemental Si which should be It is important to mention that we did not observe any signal corresponding to Fe ñor to Ni 2p emissions after 10 min of sputtering, this is related to the fact that the passive layer consisted only of Cr^+ oxide and oxihydroxide with some amounts of sihca as it has been seen above.
After sputtering on Fe62CrioNigP2o amorphous alloys, the spectrum recorded over a wide binding energy región showed, in addition to Cr 2p signal, Ni 2p and Fe 2p emissions, contrary to the previous case.Cr 2p3/2 core-level emissions were measured as it is shown in figure 5. We analyzed this spectrum using two Lorentzians.The dashed subspectrum shows the signal corresponding to oxidized chromium Cr^+ located at 576.7 eV BE and the dotted subspectrum represents elemental chromium emission located at 574.4 eV BE.This figure indicates that in the case of Fe52CrioNigP2o the total Cr 2p emission is a mixture of elemental and oxidized chromium.The presence of elemental Cr in the passive layer is not expected and, con^equently, we assign the elemental component to emission from the amorphous alloy base material and not from the passive layer.
Figure 6 represents the Fe 2p3/2 emission for this sample.The dashed curve, located at 709. 5  Figure 7 shows the Ni 2p3/2 emission that was described by two Lorentzians.The dotted curve located at 852.7 eV BE represents the elemental Ni State and the dashed curve at 854 eV BE represents the oxidized state.From this figure we can see the presence of oxidized Ni in the passive layer, as well as an elemental Ni component coming from the alloy substrate.We measured the P 2p photoemission signal, which showed an emission line at approximately 130 eV, indicating that the P chemical state corresponds to elemental phosphorous without oxidized contribution.Since we observe that the passive layer on Fe52CrioNÍ8P2o is formed not only by oxidized Cr but also by oxidized Fe and Ni, this layer is less uniform and less resistant than that formed on Fe52CrioNÍ8SÍ2o.This explains why with the same sputtering time it vanishes faster and, that after 10 min sputtering we collected elemental Fe, Ni and P emissions from the substrate.
The XPS results indícate that the passive layer formed on the surface of Fe52CriQNÍ8SÍ2o contains oxidized Cr and silica without any Ni and Fe signáis, contrary to the case of the Fe52CrioNÍ3P2o surface, which also showed the presence of Fe and Ni oxides.The results are in agreement with the results of the electrochemical measurements since a higher content of Cr3+ in the passive layer indicates a better corrosión behaviour (10).For the P containing alloy, the presence of several oxides decreases the corrosión resistance of the passive layer which is less uniform and less resistant than the layers formed only by Cr oxide.
The AES results showed that for the Fe52CrioNÍ8SÍ2o amorphous alloy the amount of Cr and O remained almost constant when we performed 2 min sputtering cycles up to a total sputtering time of 10 min, with no traces of Fe and Ni.However, for Fe52CrioNÍ8P2o alloy, applying the same sputtering cycles up to a total sputtering time of 10 min, Cr, O, Fe and Ni emissions were observed.The Fe52CrioNigSÍ2o AES results agree with the XPS measurements suggesting that at different depths the passive layer formed on this amorphous alloy is constituted mainly by oxidized Cr.However, for the Fe52Cr|QNigP2o case, the results show that the passive layer is formed by oxidized Cr, Fe and Ni.

CONCLUSIONS
We have shown by applying electrochemical technics to Fe62CrioNÍ8X2o (X = P, B, Si) amorphous alloys that the best corrosión behaviour is given by the Si-containing alloy.The corrosión resistance is determined mainly by the presence of a protective film formed on the surface of the alloy.The composition of this film plays a fundamental role in this corrosión behaviour, as we showed by the XPS and AES measurements.For the case of Fe62CrioNÍ8SÍ2o the study revealed a passive layer formed by oxidized chromium with some amounts of silica and with no contribution of other metal oxides.On the other hand, Fe52CrioNÍ8P2o showed a passive layer formed by a mixture of metal oxides.These results are in agreement with the corrosión studies because a passive layer formed mainly by oxidized chromium enhances greatly the corrosión resistance.

TABLE L -
R^ and R^ valúes obtained after 15 days of immersion in 0.01 M HCl for the three studied alloys: Fe62CrioNÍ8X2o (X = P, B, Si) Valores de Rj y Rp depués de 15 días de immersion en HCl 0.01 M para las tres aleaciones amorfas estudiadas: Fe^2 ^^lo ^h^io (^ -^' ^' ^^) TABLA /.- eV BE, corresponds to oxidized Fe and the dotted curve at