Studies on Localized Corrosion

Studies on Localized Corrosion and Desalination Treatment of Simulated Cast Iron Artifacts

Ouyang Weizhen and Xu Chunchun

A simulated occluded cell was utilized to study the localized corrosion in seawatcr and desalination treatment in alkaline solution of simulated cast iron artifacts. After the simulated occluded cell galvanostatic test, high acidity and enrichment in chloride ions in the occluded cell were demonstrated as a result of an autocatalytic process. In addition, scanning electron microscopy (SEM) and energy dispersive X-ray analysis (EDX) indicated the degree of the localized corrosion and the enrichment in chloride ions. An investigation was also made of the chemical changes within the occluded cell during the desalination treatment; the results showed that the pH values and the amount of chloride ion removed from the occluded cell increased with the treatment time. The decrease in chloride of the corroded surface of the specimen was confirmed by EDX. The research appeared to prove that the simulated occluded cell is capable of providing useful information on the localized corrosion and desalination treatment of simulated cast iron artifacts.

INTRODUCTION

Nowadays the storage of cast iron artifacts recovered from marine sites has received extensive attention. One of the most serious problems facing conservators and conservation scientists is how to stabilize iron artifacts against further corrosion under favourable environmental conditions. The ingress of chloride ions is of particular concern because they are aggressive and can promote corrosion [1, 2J. Since they have deleterious effects on corrosion performance, chloride ions must be removed [3]. Therefore, it is worthwhile investigating the mechanism of localized corrosion and desalination treatment of cast iron artifacts.
The localized corrosion of cast iron artifacts is a complex electrochemical process, with the anodic and cath-odic reactions occurring at spatially separate points [4]. During corrosion, the iron-containing phases are corroded but the graphite remains as an interlinked three-dimensional network which retains the original shape of the artifact [1]. It has been agreed that the solution within the graphitized region of corroded marine cast iron is essentially a ferrous chloride solution, with a concentration of 1 M or greater and pH 4.8 [5]. Local acidification and chloride ion enrichment could be explained by the model developed by Turgoose [6]. which was in some ways similar to crevice or pitting corrosion on iron exposed to a solution containing chloride ions. Within these occluded cavities, the presence of chloride ions indeed contributes to a corrosion cycle (7). The autocatalytic process is a feature common to the localized corrosion of cast iron artifacts in seawater. Detailed studies of the autocatalytic process will certainly help in understanding the localized corrosion process. To date, there have been no direct techniques to measure either the pH values or the chloride concentration in the pore solution.
On the other hand, the removal of chloride ions is absolutely essential for the conservation of iron artifacts. The principal aim of any treatment of archaeological iron of marine origin is to eliminate chloride from the corrosion products. It this is done successfully, the development of the corrosion process due to chloride ceases and the rate of corrosion is reduced to a more manageable level, where conventional methods to counter corrosion can operate efficiently. When artifacts are soaked in a treatment solution, the hydroxide ions diffuse into the corrosion product reacting with chloride-containing compounds to produce free chlonde ions, and then chloride ions diffuse out from the pores of high chloride ion concentration into the treatment solution [8J. The chloride ion diffusion tends to be relevant to the processes occurring in solution-filled pores. At present, there is no consensus as to the optimum treatment and one reason for this appears to be that there is no reliable method of assessing the ongoing process during treatment. Consequently, a thorough knowledge of the chloride ion migration in occluded pores during the treatment is essential if desalination methods are to be carried out successfully.
It would be of great practical interest to have a physical model which enabled the study of the microscopic mechanism of the localized corrosion of cast iron artifacts and chlonde ion migration during corrosion and desalination. The method using a simulated occluded cell is a convenient way to reveal the chemical changes within an occluded cell, and has been confirmed by many researchers [9], including Mars G. Fontana and Marcel Pourbaix. Experimental data, combined with the physical description of the localized corrosion process, should prove useful for developing more versatile, phenomenological corrosion prediction and conservation for cast iron artifacts.
In this paper, it is intended to investigate the changes of chemical states within the occluded cells of simulated cast iron artifacts in 3.5 wt% NaCl solution and 2.5 wt% NaOH solution by use of a simulated occluded cell. In addition, scanning electron microscopy (SEM) and energy dispersive X-ray analysis (EDX) were adopted to analyse corrosion morphology and components.

EXPERIMENTAL DESIGN

The specimens were made of grey cast iron to simulate cast iron artifacts. A simulated occluded cell apparatus was used in the study of localized corrosion and desalination treatment. (A full description of the simulated occluded cell is provided in the Appendix.)
A simulated occluded cell galvanostatic test was carried out in 3.5 wt% NaCl solution, which was the medium to simulate seawater corrosion. The glass container was filled with the solution and its pH was adjusted to 7, though seawater is normally in the range 8.2-7.8. At room temperature, the solution was injected into the occluded cell. An anodic current density of
1 mA-cm 2 was applied to the occluded specimen to simulate the couple current in and out of the occluded cell. The chloride ion concentration and pH in the simulated occluded cell were monitored at various time intervals. During desalination treatment the simulated occluded cell, after 48 hours of galvanostatic tests, was immersed in a glass container containing 2.5 wt% sodium hydroxide. The pH values and the chloride contents in the occluded cell at different time intervals were monitored in order to determine the progress of chloride ion release. Details of the experimental procedures are described in the Appendix. Specimens after tests were photographed by SEM and analysed by EDX.

RESULTS AND DISCUSSION

After the simulated occluded cell galvanostatic tests, the pH values in the occluded solution over time under 1 mA-cm-2 anodic current density are shown in Figure 1. It can be seen that the pH dropped sharply within the first hour from 7 to 5.68, then tended to decrease gradually. After 72 hours of accelerated corrosion, the pH value dropped to 3.86.
The reason for this was that localized corrosion took place inside the occluded cell. The localized corrosion inside the occluded cell resulted from the establishment of differential oxygen cells and was continued by the stimulation of the autocatalytic process which promoted local acidification and chloride ion enrichment. Local acidification that occurred inside the occluded cell involved dissolution of the anode and hydrolysis of the dissolved metal ions. Cations dissolved from the anode

Figure 1 Fall in pH in the occluded cell with time with a 1 mA-cm-2 current density passing through the specimen.

found difficulty in diffusing outwards because of the greater immobility of the solution in the occluded cell compared to the bulk solution. In consequence, their concentrations increased. The accumulation and hydro-lysis of Fe2+ resulted in local acidification in the occluded cell. The hydrolysis reactions are given below [6]:

The generation of free acid lowered the pH value and, inversely, the increase in acidity accelerated localized corrosion. The effect in the occluded cell of acidification causing accelerated localized corrosion activity is called the ‘autocatalytic effect’ [10]. As anodic current passes through the occluded specimens, the autocatalytic process continues at a steady rate in the occluded cell, and the pH value gradually decreases.
On the other hand, chloride ions migrated inward from the bulk solution simultaneously with local acidification. Figure 2 is representative of the quantity of chloride ion migration in the occluded cell at various time intervals under 1 mA-cm-2 anodic current density. It was found that the chloride ion migration increased in a linear relationship with the increase of time.
As noted previously, dissolution of the anode resulted in an increase in Fe2+ ion concentration. In order to balance the excess positive charge produced in the occluded cell, negative chloride ions have to migrate inwards from the bulk solution to maintain electrical neutrality [7], which causes chloride ion enrichment in the occluded cell. Moreover, the autocatalytic process

keeps chloride ions migrating inwards, and therefore chloride ion concentration increases with time. As long as the cast iron specimen is corroding and producing Fe2+ ions, chloride ions will diffuse in and concentrate in the occluded cell.
As a result of the autocatalytic process inside the occluded cell, the acidity- (via hydrolysis) and chloride concentration (via migration) increased with time. The relationship between pH and chloride ion concentration m the occluded cell at 1 mA.cm-2 current density is presented in Figure 3. It is evident that the pH value in the occluded cell dropped rapidly from 7 to below 5, and thereafter decreased linearly, indicating that the decrease in the pH value was accompanied by an increase in chloride ion concentration.

Figure 3 Relationship between pH value and the chloride ion concentration in the occluded cell.

Based on the results above, the pH value below 5 was proportional to the chloride ion concentration. The following empirical equation was derived:

pH = 6.82-2.71 [Cl-] R = 0.9910

where [Cl-] was the chloride ion concentration in the simulated occluded cell and R. was the confidence in the fitting equation.
SEM micrographs of specimens after galvanic testing tor 24 and 48 hours at 1 mA-cm-2 anodic current density-are presented in Figure 4. As can be observed, the surface patterns differed from one another. Figure 4a exhibited a rust laver with a uniform, adherent and con-

Figure 4 SEM micrographs of cast iron under an anodic current density of 1 mA.Cm-2 for (a) 24 hours and (b) 48 nours:;(c) and (d) show the detailed surface morphology of sample 4b

tinuous structure. In contrast, Figure 4b demonstrated a more severe attack on the rust layer. The insoluble structure, and micro-cavities and cracks, can be clearly observed in Figures 4c and 4d.
The more severe attack on the rust layer in Figure 4b was attributed to the autocatalytic effect, which promoted acidity and chloride ion concentration, thus accelerating the dissolution of cast iron with longer exposure time. The insoluble structure m Figure 4c, which was flake-like, indicated that the residual graphite formed a three-dimensional network. In addition, micro-cavities and cracks could be seen in Figure 4d. The formation of these micro-cavities and cracks was evaluated in relation to the acid produced and the formation of the solid products, which could give rise to local stresses and cause dissolution of other corrosion products, thereby chemically assisting the propagation of cracks and aiding the disintegration of the rust layers [11]. The general conclusion was that the autocatalytic effect exerts a greater influence on the sample after the 48-hour test than on the sample after the 24-hour test.
In order to obtain a better understanding ot outward diffusion of chloride ions, the simulated occluded cell is also used in the desalination treatment. Monitoring the pH in the occluded cell over time provides more details during chloride removal, which is shown in Figure 5. The pH value in the occluded cell rises gradually with the square root of the treatment time.
The reason tor the increase in the pH value is the increase m hydroxide ion concentration in the occluded cell. The inward flow of the hydroxide ions is due to the concentration gradient between inside and outside oí the occluded cell. In general, the high mobility of hydroxide ions and their concentration in the treatment solution maintain the inward diffusion of hydroxide ions, thus the pH value increases gradually over time.
As a result of the ingress of hydroxide ions, chloride ions diffuse outwards from the occluded cell into the washing solution. After immersion, the amount of chloride ion removed from the occluded cell increased with the square root of time as shown in Figure 6. It can also be seen that the results are divided into two stages.
It is generally accepted that two steps might occur when a marine iron artifact is soaked in the treatment solution [8]. One is the reaction of dissolution of the ‘FeOCl’ to produce free chloride ions; the other is the movement of the chloride ions from their point of production inside the corrosion product to the bulk wash solution. After galvanic testing for 48 hours at 1 mA.cm-2 anodic current density, the specimen released Fe2+ ions into the occluded solution. It has been reported that chloride ions present in the iron corrosion products are trapped within the lattice structure of the various iron oxyhydroxides; among them, ferric oxychloride (FeOCl) is the dominant chloride-containing corrosion product on cast iron [12-14|. In the case of the galvanic test, FeOCl, which is not stable m air, may be present as an intermediate phase during the corrosion process due to lack of oxygen in the occluded cell and be converted to ß-FeOOH once exposed to air. North has proposed

Figure 5 Relationship between pH in the occluded cell and the square root of the treatment time.

Figure 6 The amount of chloride ion removed as a function of the square roof of the treatment time.

the mechanism by which chlorides in the corrosion products are converted to free chloride ions during chloride removal. The reaction is given by [15]:

The hydroxide ions diffuse into the occluded cell and react with chloride-containing compounds to produce free chloride ions. Then the chloride ions migrate outwards because of the concentration gradients between inside and outside of the occluded cell, and the removal of chloride should be achieved.
The data in Figure 6 indicate that there may be two stages involved in the washing. The initial stage is controlled by the reaction ot dissolution of the FeOCl and in turn this is controlled by the outward diffusion of chloride ions. In stage I. it is due to the rapid ingress ot hydroxide ions in the occluded cell, which allows the dissolution of the FeOCl, and the more chloride ions present in the occluded solution the more chloride ions can be transported. Thus, the amount of chloride ion removed rapidly increases. In stage II, the amount of chloride ions removed increases gradually because of the decrease in the concentration gradients ot hydroxide ions between inside and outside of the occluded cell. The reaction of dissolution of the FeOCl may not be a significant factor in determining chloride release rate. The outward diffusion of chloride ions may then play an important role in the washing. As the rate-controlling process is generally outward diffusion ot chloride ions in the washing method, the rate of chloride ion release from the occluded cell into the wash solution follows the diffusion law [8]. According to this law, the diffusion equation predicts that the amount of chloride ion removed varies linearly with the square root of the treatment time. Such a line is shown in Figure 6, and the agreement between theory and experiment indicates that diffusion is the rate-controlling process in stage II.
The release rate of chloride ions can be obtained by analysis of chloride ion removal data measured in the occluded solution, which has seldom been used in the literature. The chloride ion extraction rate at a specific time is defined by the instantaneous slope at that time, which is the derivative of the amount of chloride ions removed from the occluded cell with respect to time [3]. Figure 7 is a plot of the chloride ion extraction rate as a function of treatment time.

Figure 7 The chloride ion extraction rate from the occluded cell with treatment time.

It was tound that the chloride ion extraction rate decreased rapidly at the beginning and then became almost stable, with a minor decrease with the increase of washing time. By monitoring the amount of chloride ion extracted and constructing a graph from this information, the chloride ion extraction rate can be used to evaluate the effectiveness of different treatment solutions by using the same occluded cell. Moreover, the chloride ion extraction rate as a guide to the progress of the treatment can provide a signal to change the treatment solution when the curve levels out into a plateau region. Having achieved an understanding of the process of chloride removal, it can be applied to increase the desalination efficiency.

X-ray energy dispersive spectrometry (EDX) of samples before and atter the 72-hour desalination treatment is shown in Figure 8. Figure 8a shows the chemical composition of the specimen after the 48-hour galvanic test at 1 mA-cm-2 anodic current density and Figure 8b is the EDX analysis of the corroded specimen atter the 72-hour desalination treatment which followed the application of a 1 mA-cm-2 anodic current density for 48 hours. The presence of Fe. Mn, Si, P, S and Cl could be observed. Comparing these two pictures. Figure 8a was rich in chloride and Figure 8b was low in chloride.
Figure 8 Chemical composition of specimens before and after the treatment analysed by EDX. (a) EDX analysis of the cast iron with 1 mA.cm-2 applied anodic current density for 48 hours, (b) EDX analysis of the corroded specimen after 72-hour desalination treatment.

The EDX analysis in Figure 8a provided evidence that the chloride ion enrichment was the result of the autocatalytic process inside the occluded cell. Figure 8b provided intormation that the decrease in chloride was due to the diffusion of the chloride ions from the occluded cell to the washing solution during the treatment.

CONCLUSIONS

A simulated occluded cell was used to study the localized corrosion and the desalination treatment of simulated cast iron artifacts. It is a convenient method to reveal the chemical changes within an occluded cell, which can not only study the mechanism ot the localized corrosion of cast iron, but can also investigate chloride ion migration in and out of the occluded cell during localized corrosion and desalination treatment.
In the present work, the changes of chemical state for localized corrosion of simulated cast iron artifacts in 3.5 wt% NaCl solution were studied. It was found that as an anodic current was passed through the cell, the pH value inside the cell initially fell quickly and then decreased gradually. Meanwhile, the chloride ions migrated into the occluded cell. The quantity of chloride ion migration increased with the time that the current was flowing. The results of the SEM analysis showed acceleration of dissolution of cast iron with the time of application of the anodic current, and EDX analysis indicated the enrichment in chloride ion.
A study was made of the chemical changes within the simulated occluded cell after the 72-hour desalination treatment which was applied to samples that had first been corroded at 1 mA-cm-2 anodic current density for 48 hours. During chloride removal, the pH value in the occluded cell gradually increased over time. In addition, the amount ot chloride ions removed from the occluded cell increased with the treatment time. The chloride ion extraction rate initially decreased rapidly and then became almost stable with the washing time. Furthermore, the EDX analysis disclosed the change of composition ot chlonde after the treatment.
It is evident that the simulated occluded cell could help in understanding the degree of localized corrosion and the progress of the treatment of cast iron artifacts, thereby enabling their safe and effective preservation.

ACKNOWLEDGEMENTS

The authors would like to thank the National Key Technologies R&D Program of the 10th Five-Year Plan Period for financial support (Contract No. 2001BA805B01). This work was also supported by the State Key Laboratory for Metallic Corrosion and Protection. Thanks are also extended to Dr L.S. Selwyn for her kind help.

APPENDIX: EXPERIMENTAL

A simulated galvanostatic occluded cell apparatus [16], as shown in Figure 9, was adopted to study the localized corrosion.

Figure 9 Schematic diagram of the simulated occluded cell: 1 bulk specimen, 2 occluded specimen. 3 simulated occluded cell. 4 magnetic stirrer.

As can be seen, the simulated occluded cell made of hard glass was in the centre of a glass container, the effective volume of which was 1.5 mL. The cell was separated from the bulk solution by a glass tube (1.5 x 15 mm), which was filled with filter-paper scraps to retard diffusion and convection between the occluded and the bulk solutions. The specimen was inserted into the cell through a rubber stopper. The outer end of the specimen was connected to the positive terminal of a set of batteries, the negative terminal of which was connected to the external bulk specimen of graphite.
The specimens used were grey cast iron to simulate cast iron artifacts, so the corrosion process would be similar in soine ways to that observed in the field. The chemical composition of specimens (wt%) was: C 3.00, Si 1.84, Mn 0.82, P 0.098, S 0.089. The surfaces of the specimens were wet-polished with silicon carbide paper to grade 1000. The samples were then rinsed with deionized water, degreased with CP-grade acetone (propanone) and stored in a desiccator filled with nitrogen until they were ready for testing. The exposed area of each specimen was 25 mm2, and the remaining surface was shielded with silicone paste. The exposed area ratio between the occluded and bulk specimen was about 1:100. NaCl (AR grade) and deionized water were used to make up a 3.5 wt% solution, which was the medium to simulate seawater corrosion. The glass container was tilled with 2 L of the solution and its pH was adjusted to 7. At room temperature, about 1.5 mL of the bulk solution was injected into the occluded cell. An anodic current density of 1 mA.cm-2 was applied to the occluded specimens to simulate the couple currents in and out of the occluded cell via resistances. After various time intervals the tests were stopped, and the occluded solution was removed for analysis. The pH value of the occluded solution was recorded with a pH meter (pHs-25 type) at ambient temperature. The chloride ion concentration was obtained with a Metrohm (model 751 Titrino) potentiometric titrator.
The desalination treatment was conducted in 2.5 wt% NaOH solution. After a 48-hour galvanic test at 1 mA.cm-2 anodic current density, the simulated occluded cell was immersed m a glass container containing the sodium hydroxide. The outer end of the specimen was directly connected to the external bulk specimen of graphite. The pH values and the chloride ion contents in the occluded cell at different time intervals were monitored.
After exposure in the simulated occluded cell galvanostatic tests at 1 mA.cm-2 anodic current density for 48 hours and the subsequent desalination treatment for 72 hours, specimens were rinsed in deionized water and then dried in hot air and stored in a container filled with nitrogen. The microstructural characterizations of the samples were performed with a Cambridge (model S250-MK3) scanning electron microscope, fitted with a Link (model AN-10000) energy dispersive spectrometer.

REFERENCES

1 North. N.A.. and MacLeod. I.D.. ‘Corrosion oí metals’, in Conservation of Marine Archaeological Objects, ed. C. Pearson, Butterworths. London (1987) 68-98.
2 Dalard. F.. Gourbeyre, Y.. and Degrigny. C, ‘Chloride removal from archaeological cast iron by pulsating current’, Studies in Conseivation 47 (2002) 117-121.
3 Selwyn, L.S.. McKmnon. W.R.. and Argyropoulous, V., ‘Models tor chloride ion diffusion in archaeological iron”, Stadia in Conservation 46 (2001) 109-120.
4 Walker. R.. ‘The corrosion and preservation of iron antiques’, Journal of Chemical Education 59(11) (1982) 943.
5 North. N.A., “Corrosion products on marine iron’. Studies in Conservation 27 (1982) 75-83.
6 Turgoose. S.. ‘The corrosion ot archaeological iron during burial and treatment’, Studies in Conseivation 30 (1985) 13-18.
7 Selwyn. L.S.. Sirois, P.J., and Argyropoulous, V., ‘The corrosion of excavated archaeological iron with details on weeping and akaganéite’. Studies in Conservation 44 (1999) 217-232.
8 North. N.A.. and Pearson. C ‘Washing methods for chloride removal trom marine iron artifacts”. Studies in Conservation 23 (1978) 174-186.
9 Zuo. J.. jin. Z.. Sun. 1C. Xu. Y.. Feng. X.. ‘Accelerating eüect and cntical pH value ot occluded cell corrosion within pits, crevices, or stress corrosion cracks’. Corrosion 44(8) (1988) 539-543.
10 Wei, IS.. Qin. Y.. Lin. K.. Lin. Y.. Caí. Y.. Chen. X.. Shi. M. Jian. C The Theory and Application ot Metal Corrosion. Chemical Industry Press. Beijing (1984) 150.
11 Turgoose, S.. “Post-excavation changes in iron antiquities’. Studies in Conservation 27 (1982) 97-101.
12 North. N.A., and Pearson. C. ‘Alkaline suliite reduction treatment ot manne iron’, in ICOM Committee for (Conservation 4th Triennial Meeting. Venice (1975) 75-13, 3-1 to 75 13 3-14.
13 North. N.A.. and Pearson. C ‘Thermal decomposition oí FeOCl and marine cast iron corrosion products’. Studies in Conservation 22 (1977) 146-157.
14 Gilberg. M.R., and Seeley. N.}.. ‘The identity of compounds containing chloride ions m marine iron corrosion products: a critical review’. Studies in Conservation 26 (1981) 50-56.
15 North. N.A.. and Pearson. C ‘Investigations into methods for conserving iron relics recovered from the sea’, m Consen-atiou in Archaeology and the Applied Arts. I!C. London (1975) 173-181.
16 Xu. C Fu. X.. and Liu, Y., “Chemical changes within occluded corrosion cell of mild steel in NaCl solution’, Journal of Chemical Industry and Engineering 50(4) (1999) 258-262.

AUTHORS

OUYANG WEIZHEN completed his master’s degree in corrosion and protection from Beijing University of Chemical Technolog)1 in 1996. He studied for a doctorate in 2(102. He is an associate professor and his research focuses on the corrosion and conservation problems related to cast iron. Address: Beijing University of Chemical Technology. 15 Bcisanhuau East Road, Beijing 100029, PR. China. Email: ouyangwz@grad.buct.edu.cn

Xu CHUNCHUN has an engineering degree (1964) from Beijing University ot Chemical Technology. She is a professor at the University and Vice-Secretary General and Director, Chinese Society of Corrosion and Protection (CSCP). She has conducted research in corrosion science. Address as for Ouyang. Email: chunchunxu@263.net

Résumé – Une cellule fermée simulée a été utilisée pour étudier la corrosion localisée dans l’eau de mer et le traitement de dessalement en solution alcaline pour des objets eu foute simulés. Après un test galvauostatique de ¡a cellule, on a pu observer, comme résultat d’un processus autocatalytique, une forte acidité et un enrichissement en ions chlorure dans la cellule. Par ailleurs, les analyses au MEB/EDS indiquaient le niveau de corrosion localisée et l’enrichissement eu ions chlorure. Une investigation a également été menée sur les changements chimiques à l’intérieur de la cellule fermée pendant le traitement de dessalement; les résidtats montrent que les valeurs du pH et la quantité d’ions chlorure éliminés de la cellule augmentaient avec la durée du traitement. La diminution de la quantité d’ions chlorure à la surface corrodée de l’échantillon a été confirmée par EDS. La recherche semble prouver que la cellule fermée simulée est eu mesure de fournir des informations utiles sur ¡a corrosion localisée et sur le traitement de dessalement d’objets en fonte.

Zusammenfassung – Eine Simulation einer Einschlußzclle wurde verwendet, um die lokale Korrosion in Meerwasser und die Entsalzung durch die Behandlung mit alkalischeu Lösungen au Schmiedeeisen zu untersuchen. Gemäß einem galvanostatischen Test konnte ein holier pH-Wert und eine Anreicherung von Chloridionen in der Eiuschlusszelle beobachtet werden, welche auf autokatalytisclie Prozesse zurückgeführt werden. Das Ausmaß der Korrosion und der Chloridanreicherung wurden durch Untersuchungen mit Hilfe der Rastcrclektronenmikroskopie I energiedispersiveu Röutgenmikroaualyse (REM/EDX) bestätigt. Weitere Untersuchungen betrafen die chemischen I eranderungen in der Einschlusszelle während der Entsalzung. Dabei stiegen der pH-Wert und der Grad der Chloridentjernung mit der Behandlungsdauer. Letzteres wurde mit EDX bestätigt. Die Untersuchungen legen nahe, dass anhand der simulierten Eiuschlusszelle nützliche Informationen über lokale Korrosionen und Entsalzungseffektc an Schmiedeeisen gewonnen werden können.

Resumen – Se empleó una simulación de célula cerrada con el fin de estudiar, tanto la corrosión localizada por efecto del agua de mar, como los tratamientos de desalinizacióu en disoluciones alcalinas para artefactos simulados de hierro fundido. Según mostró el test, en la célula galvanoestática cerrada se manifestó una elevada acidez y un enriquecimiento de iones cloruro, todo ello resultado del proceso autocatalítico. Adicioualmente, micrografías obtenidas por microscopía electrónica de barrido (SEM) y energía dispersiva de rayos X (EDS) mostraron el grado de corrosión localizada y el aumento de iones cloruro. Se realizó además una investigación sobre los cambios químicos ocurridos en el interior de la célula cerrada durante el tratamiento de desalinización; los resultados mostraron que los valores del pH y la cantidad de iones cloruro eliminados aumentaban cu el transcurso del tratamiento. La disminución de los cloruros en la superficie corroída de las muestras se confirmó mediante EDS. La investigación parece probar que la célula cerrada simulada es capaz de suministrar información muy útil sobre corrosión localizada y tratamientos de desalinización en objetos simulados de hierro fundido.