Corrosion Behavior of Porcelain Enamels in Water Tank Storage

Abstract

Recent updates to European Union directives on drinking water have extended safety limits to hot water, increasing the need to assess materials commonly used in water storage systems, such as porcelain enamel. This study investigates the interaction between enameled surfaces and aqueous environments, focusing on element release and microstructural alterations. The mass loss and chemical stability of the enamel were evaluated through a combination of surface characterization and Inductively Coupled Plasma (ICP) analysis. Time-resolved quantification of selected elements confirmed that all concentrations remained within EU regulatory thresholds. Additionally, the enamel was subjected to acidic and alkaline environments to explore the influence of pH on degradation mechanisms. Scanning electron microscopy (SEM) revealed that while the enamel undergoes surface-level modifications, the bulk structure remains intact. Notably, alkaline exposure had the strongest impact, dissolving needle-like calcium-rich structures and altering the surface more significantly than water or acid alone. These structures appear to facilitate localized corrosion once degraded. The correlation between surface morphology and elemental release dynamics highlights the critical role of microstructural features in determining long-term chemical resistance. Overall, the results underscore the importance of optimizing both the composition and structure of enamel coatings for applications involving prolonged contact with potable water.

1. Introduction

With the advent of the first water heaters in the early 1900s, these water-storage systems were not equipped with any corrosion prevention system. This caused serious problems related to corrosion events, especially in the presence of hard or aggressive water (with high concentrations of calcium Ca²⁺ and magnesium Mg²⁺ salts in the form of bicarbonates, sulfites, and chlorides) [1,2]. Later, from the 1940s to 1950s, with the increasing domestic spread of hot water, various protection methods began to be used, eventually leading to the principle of the sacrificial anode [3]. This method involves the use of a magnesium and zinc–aluminum anode [3], which protects the internal wall of the boiler through cathodic protection, with the anode corroding instead of the steel.

However, using cathodic protection alone is not sufficient as it would lead to excessive wear of the anode, requiring its frequent replacement. For this reason, the internal surface of boilers is traditionally protected with porcelain enamel coatings.

Since the advent of industrial enameling, porcelain enamel has proven to be an effective solution for protecting various components in contact with water or food. It has been widely used in the manufacturing of cookware, buckets, containers, storage tanks, and even water heaters [4]. The key advantages of porcelain enamel lie in its long-term protective properties, its high level of hygiene, and its chemical inertness, which make it a preferred alternative to organic coatings, especially in applications involving potable water or food contact [4,5]. However, one of the main limitations of enamel is its brittle nature, which makes it susceptible to mechanical damage, potentially leading to the formation of cracks and surface defects [5]. For this reason, the use of cathodic protection with a sacrificial anode, typically made of zinc or aluminum, is often combined with enamel coatings, ensuring continued protection of the underlying steel even in areas where enamel defects are present.

Recently, with the integration of domestic hot water systems, hygiene and durability have become crucial issues. In fact, the main problem with corrosion phenomena is the production of elements and ions potentially harmful to the human body. According to the latest European Union drinking water regulations [2], the limits for elements that may be present in domestic water are very strict. The limits recognized by the European Union (EU) and World Health Organization (WHO) for the elements composing sacrificial anodes are as follows: Mg < 50 mg/L; Al < 0.2 mg/L; Zn < 3 mg/L; Mn < 0.05 mg/L; Cu < 1 mg/L; Fe < 0.2 mg/L; and Ni < 0.05 mg/L [6].

The composition of porcelain enamels is primarily a blend of various oxides that contribute to their durability, corrosion resistance, adhesion to metal, chemical inertness, thermal and mechanical resistance, compatibility with potable water, and esthetic qualities. These compositions typically include a glass component with significant proportions of silica (SiO₂), alkali oxides (such as Na₂O and K₂O), and other oxides like alumina (Al₂O₃) and boron oxide (B₂O₃). The specific ratios of these components can vary based on the desired properties of the enamel.

Degradation of the enamel coating can also bring potentially dangerous ions into the solution. The diverse composition of porcelain enamels is designed to achieve heat resistance, essential for applications in cooking appliances [4], chemical resistance [5,7,8], to protect against degradation from organic materials and cleaning agents [5,9], and esthetic qualities.

Both corrosion protection methods discussed, such as magnesium anode and porcelain enamel, inherently have both positive and negative aspects.

Specifically, porcelain enamel has high initial cost production, as it requires a vitrification process, but it ensures high sanitary safety, although it can be damaged by thermal shocks [10] or if the water is too aggressive.

Although enamel has a glassy composition designed to be inert, due to osmotic reactions caused by the difference in ion and element concentrations between the enamel and the water in the boiler, there is a tendency for the components of the porcelain enamel to permeate into the water, thus contaminating it and entering the domestic water circuit. The deterioration of the enamel coating is related not only to the composition but also to the quality of the coating obtained. Increasingly, the enamel exhibits a suboptimal structure due to the need to reduce the firing temperature and time for economic reasons.

The aim of this research is to study which and how much of each element is released from a typical chemical composition porcelain enamel, designed to be applied as an internal protective coating for boilers. In this research on weight loss of porcelain enamels, surfaces analysis, microstructure, and cross-section investigation have been conducted in order to assess the sanitary safety of porcelain enamel in contact with drinking water.

2. Materials and Methods

2.1. Materials

The porcelain-enameled flat panels used are standard, measuring 105 mm × 105 mm. The final composition of the porcelain enamel is as follows in

Table 1. Chemical composition (% weight) of the porcelain enamel frit.

Compound	Quantity (wt. %)	Compound	Quantity (wt. %)
Li2O	1.8–2.5	SiO2	54.0–55.5
Na2O	10.8–11.8	ZrO2	3.0–3.5
K2O	1.0–2.1	TiO2	1.0–1.9
MgO	<0.2	Fe2O3	<0.2
CaO	4.5–6.5	F	1.2–2.0
BaO	2.2–3.0	MnO2	<0.2
ZnO	0.3–0.5	CoO	0.7–1.1
B2O3	10.0–12.5	CuO	0.2–0.5
Al2O3	2.0–3.5

. A DC04EK steel substrate, according to the EN10209–1998 standard [11], was specifically selected for its suitability for enamel deposition and was used as the metal substrate [4]. The low carbon content helps to prevent gas formation and surface defects during firing, ensuring optimal adhesion of the enamel.

After the panels were pre-treated by pickling with 10% hydrochloric acid to remove corrosion products and degreasing with acetone, the enamel was applied by spray deposition onto the steel substrate, followed by a thermal treatment to eliminate residual water. A final firing treatment was then carried out at a temperature of 840 °C for 7 min and 30 s, ensuring proper fusion and durable bonding between the enamel and the steel surface. Table 1 reports the chemical composition of the frit used to produce the enameled layer.

2.2. Characterization Methods

2.2.1. Surfaces and Cross-Section Investigation

Using a scanning electron microscope (SEM, Jeol JSM IT 300, JEOL, Tokyo, Japan), the samples were evaluated and characterized. Both the surface and cross-section were observed to identify the structure and the thickness of the coating. The EDS analysis permitted us to individuate the distribution of elements. The integrity of the samples was observed before and after exposure to water and aggressive chemical solutions.

2.2.2. X-Ray Diffraction

The surface composition of the porcelain enamel was analyzed using an X’Pert HighScore diffractometer (Rigaku, Tokyo, Japan).

2.2.3. Roughness

Surface roughness was measured using a MarSurf PS1 roughness gouge (Mahr GmbH, Göttingen, Germany) before and after water exposure to evaluate how distilled water alters the enamel surface morphology.

2.2.4. Weight Loss of Porcelain Enamels

The behavior of porcelain enamel samples was investigated when exposed to two types of water with different conductivity levels (pure water and ultrapure distilled water, with conductivity coefficients 6 µS and 1 µS, respectively) at a temperature of 95 °C. The aim of this experiment was to assess the mass loss of enamel due to water contact.

The test followed the standard procedure UNI-TR11358 [12], which consists of two exposure cycles of 48 and 840 h (for a total of 35 days), with intermediate weight measurements to evaluate the amount of material loss. After exposure, the samples were analyzed using scanning electron microscopy to assess how water-induced degradation affected the enamel structure.

2.2.5. Analysis and Quantification of Released Elements (ICP)

Following the same setup previously described [12], the samples were exposed to pure distilled water with conductivity of 6 µS at 95 °C for a total duration of 25 days. At predetermined time intervals (4, 4, 2, 5, and 10 days), the water was collected and replaced with fresh distilled water. This procedure was adopted to avoid reducing the leaching efficiency due to the increasing concentration of dissolved elements in the solution. It is pivotal to underline that accelerated laboratory tests do not fully reflect real operating conditions, due to the countless variables present in actual use. For this reason, all standards define specific testing conditions.

The collected water samples were then analyzed using ICP (PerkinElmer Avio^TM 550 Max, PerkinElmer, Waltham, MA, USA), enabling quantification of specific elements of interest (calcium, aluminum, cobalt, potassium, sodium, and silicon) released into the water during each time interval.

The quantification of metal release by ICP was conducted only using distilled water, as the aim of this study is to assess the resistance of enameled surfaces in contact with drinking water, in accordance with the new European regulations on drinking water. Acidic and alkaline tests were carried out to better understand the behavior of the enamel, as these tests are included in the standard procedures for evaluating the resistance of this type of coating. However, since the pH values used in these tests are extreme and do not reflect the conditions of drinking water, the focus was placed on the attack morphologies rather than on the quantification of metal release.

2.2.6. Chemical Resistance

Drinking water can exhibit varying pH levels depending on its source. Moreover, the corrosion of the magnesium anode, which provides cathodic protection to the steel in areas where enamel defects are present, leads to water reduction and a consequent local increase in pH. It is therefore necessary to assess the resistance of the enamel to solutions with different pH levels. The chemical resistance of the porcelain enamel surfaces to acidic and alkaline agents was evaluated following and adapting standard procedures provided by ASTM International, specifically ASTM C282-10-2015 [13] for acid resistance and ASTM C614-20 [14] for alkaline resistance.

To assess acid resistance, the enamel surface was exposed to a 10% citric acid solution for 15 min at room temperature. For alkaline resistance, a 5% tetra potassium pyrophosphate solution was used for 6 h at a temperature of 96 ± 0.2 °C. After exposure, tested samples were observed using scanning electron microscopy to evaluate the damaged surfaces.

3. Results and Discussion

3.1. Microstructure Evaluation

The first scanning electron microscopy observation aimed to examine the surface structure of the intact enamel. As shown in Figure 1, the undamaged surface displays porosity distributed to varying degrees across the entire area. These pores are common, as the enamel manufacturing process involves high temperatures that lead to the formation of gases coming from the lower layers. These gases generate bubbles, which in turn result in the formation of pores. These pores are typically present in the enamel layer [5]; however, there is no reduction in protection since the closed pores do not permit contact between the aggressive environment and the substrate.

The most noteworthy feature, however, is the presence of “needle-like structures”, which appear grouped and localized in specific areas, protruding from the enamel surface. Adjacent to these structures, darker regions are consistently observed, which instead lie beneath the enamel surface.

Further SEM analyses enabled a more in-depth investigation of the surface of interest. As shown in Figure 2, the composition of the previously mentioned fusion products was identified as calcium based. To determine the exact composition of these structures, X-ray diffraction analysis was carried out. However, due to the small quantity of the compound of interest and the amorphous nature of the surface, the resulting signal was too weak and noisy to allow the exact identification of the calcium-based structures’ compound. The darker regions observed in Figure 1 were identified as aluminum-rich areas, as then proved and shown in Figure 2, the image on the left. This accumulation may be attributed to local supersaturation or differences in solidification behavior during the cooling stage of the firing process, which could lead to phase separation or localized enrichment beneath the enamel surface.

Following the surface analysis, the cross-section of the sample was examined to evaluate the presence of this calcium structure in depth and also to have information about the thickness, the presence of defects, and the adhesion quality between the enamel layer and steel substrate. As shown in the two images in Figure 3, the enamel layer is characterized by a single-layer coating. Indeed, the abundant presence of pores is not unusual and is attributed to gas released from the underlying metal layer during the firing process. An interesting feature is the presence of dendritic structures (Figure 3, image on the right), observed at the interface between the enamel and the iron substrate. This dendritic growth is promoted by the thermal and chemical gradients at the enamel–metal interface. This presence also permits us to confirm a correct firing process, additionally obtaining a good adhesion of the layer with steel substrate. The thickness of the enamel layer is about 230 µm. No cracks or other defects are present in the layer.

3.2. Weight Loss of Porcelain Enamels Due to Contact with Water

Figure 4 summarizes the weight loss, expressed in g/m², of two samples exposed to pure distilled water and one sample exposed to ultrapure distilled water.

The two weight loss tests were performed in triplicate. The samples exposed to pure water showed consistent results, as indicated by the small standard deviation (0.151 g/m²). Overall, a significant difference in mass loss was observed between the porcelain enamels exposed to the two types of water. The first sample showed a weight loss of 1.42 g/m², while the samples in contact with ultrapure water exhibited an average weight loss of 5.16 g/m², with a relatively large standard deviation among replicates (0.946 g/m²).

This marked difference in weight loss is attributed to the purity of the water used. The purer the solvent, meaning it contains fewer dissolved ions, the higher its ability to extract elements from the enamel surface. This phenomenon is driven by the concentration gradient between the solvent and the enamel: the enamel tends to release ions into the solvent, therefore leading to mass loss. The study then proceeded by examining the enamels that had been exposed to pure distilled water.

3.3. Analysis and Quantification of Released Elements

The core experiment of this research, aligned with the overall objective of assessing the health safety of porcelain enamel, focuses on evaluating whether, and to what extent, the enamel releases elements into water. To assess this, the enamel was placed in contact with pure distilled water for a total duration of 25 days at a temperature of 95 °C. During the experiment, the water used for the leaching process was replaced at specific time intervals, allowing the calculation of the amount of material released from the enamel into the water after 4, 8, 10, 15, and 25 days.

The collected water samples were then analyzed using the ICP technique, in order to quantify the concentration, expressed in ppb (µg/L), of each specific element of interest. This method allows for the detection of trace elements at concentrations as low as parts per billion (ppb), making it suitable for assessing the leaching behavior of enamel coatings.

Figure 5 presents the graphs obtained from the ICP analysis for six elements: calcium, cobalt, sodium, potassium, aluminum, and silicium.

The released-element profiles reveal distinct extraction kinetics for the various enamel constituents, indicating different degradation mechanisms. Calcium exhibits a rapid initial release, which then progressively slows down. This trend suggests that calcium-rich structures, previously highlighted by SEM observation, are predominantly located near or on the surface of the enamel, and are therefore quickly solubilized upon exposure to the aqueous environment. Once these surface features are depleted, the release rate drops, likely reflecting the limited availability of calcium deeper within the matrix.

In contrast, sodium, potassium, and silicium show a more linear and sustained release over time, pointing to a progressive and continuous degradation of the glassy matrix itself. This constant extraction rate suggests that these elements are more homogeneously distributed throughout the glassy bulk and are released as the amorphous phase gradually dissolves. It is worth noting that potassium and sodium, in particular, are typically present in the glass network as modifiers and are associated with weaker chemical bonds compared to glass formers like silicium and aluminum. This structural role contributes to their earlier and more consistent release during matrix degradation.

Cobalt, on the other hand, displays a delayed release onset, with an acceleration phase once a defect is formed and the element becomes accessible, followed by a slowdown, likely due to the near-complete extraction of the cobalt and the need for deeper matrix degradation.

Similarly, aluminum shows a gradual increase in extraction rate, suggesting a more complex mechanism where matrix degradation may expose aluminum-rich regions or phases that were initially more protected.

These kinetic trends highlight the importance of investigating the morphology of enamel damage. Indeed, it is essential to understand whether the degradation is generalized or localized, and whether it is facilitated by structural features or porosity. Surface and cross-section analysis can provide critical insight into whether specific microstructural elements, such as needle-like calcium-rich phases or microcracks, serve as starting points for preferential attack, ultimately leading the evolution of damage throughout the enamel layer. A major drawback of this measurement was the impossibility to directly extrapolate the enamel degradation behavior to the actual performance of a water heater, as several parameters, such as enamel composition, tank geometry, and operating conditions (temperature, thermal stress, fluctuations in water composition, and pH) can vary significantly.

Table 2. Cumulative released elements from the enamel surface (95 °C, distilled water, 25 days).

Element	Cumulative Release (µg/L)	EU Drinking Water Limit (µg/L)
Ca	613	100 × 103
Co	21.75	N/A
Na	1226	200 × 103
K	408.3	10 × 103
Al	47.1	200
Si	14,247	1500 × 103

reports the total amount of each element released from the surface in contact with distilled water after 25 days. It is important to highlight that none of the analyzed elements reached or exceeded the limits set by the EU, confirming the safety of porcelain enamel as a protective layer in potable water systems. The detection limits for each element were 0.40 ppb for Ca, 0.67 ppb for Co, 0.65 ppb for Na, 1.25 ppb for K, 1.68 ppb for Al, and 6.50 ppb for Si. Prior to each measurement, calibration curves were generated using PerkinElmer multi-element standards (Multi-Element Calibration Standard 3, N9300233). Additional potential hazard elements, such as copper, manganese, and zinc, were considered. Indeed, the released-element analyses have been carried out. However, the values proved to be not significant, also due to the low concentration of these elements in the enamel.

3.4. Evaluation of Porcelain Enamels After Water Exposure

Following the weight loss tests, the enamel surfaces and cross-sections exposed to water were examined and evaluated.

As shown in Figure 6, the most significant change observed on the surface before and after water exposure is the disappearance of the needle-like calcium-rich structures, accompanied by the appearance of localized corrosion in the areas where these structures were previously present. The process appears to follow a sequence: initially, calcium is extracted, likely because it is more exposed and weakly bound at the surface, subsequently leading to the formation of surface defects. These defects may then act as initiation sites for a localized attack, accelerating the overall degradation process.

This hypothesis is supported by ICP data: calcium release is initially high but decreases over time, while aluminum release progressively increases. This trend suggests that as surface calcium is depleted, the underlying aluminum-rich zones become exposed to the external environment (Figure 7), contributing to continued degradation. Similarly, cobalt shows a delayed release, which seems to accelerate once structural defects form, potentially indicating deeper matrix degradation involving the main glassy component (silicium oxide).

It is therefore possible to conclude that the dissolution process is initially surface-driven (calcium extraction), then transitions into a defect-driven mechanism that promotes a localized attack and increased extraction rates of structural elements such as aluminum and cobalt.

Moreover, following exposure to pure water, localized damaged areas appeared along the enamel surface (Figure 6), corresponding to the boundaries of the enamel microstructure. It is likely that more soluble compounds are concentrated at these boundaries; once leached out by water, the borders become more pronounced and clearly visible.

Figure 8 shows a cross-sectional view in tilted mode, permitting us to observe at the same time the cross-section and the surface of the enamel after water exposure. As observed, the depth of the water-affected zone was measured, revealing that although material loss occurred, resulting in the formation of a surface cavity, the depth of this damage is very limited, measuring just over 8 µm. Notably, the presence of porosity, even at or near the surface, does not appear to significantly influence or intensify the degradation, suggesting that water attack proceeds independently of these microstructural features under the tested conditions.

Additionally, surface roughness was measured before and after water exposure to evaluate how material loss, particularly the dissolution of calcium-based needle-like structures, which may promote localized chemical attack, affects the enamel surface.

Figure 9 shows the roughness measurements, highlighting significant differences between the sample before and after water exposure. This change is in agreement with previous observations. The dissolution of calcium-rich surface structures appears to facilitate the formation of localized defects, from which further leaching of enamel components may occur. These zones of intensified attack contribute to the increased surface irregularity.

Although the difference in both Ra and Rz values is not dramatic, the trend suggests that surface degradation is beginning to develop, especially in regions where the original microstructure has been altered or removed. This supports the hypothesis that the presence and subsequent loss of specific microstructural features, such as the calcium-based needles, play a crucial role in initiating and propagating surface degradation.

3.5. Surface and Cross-Section Investigation After Acid Chemical Attack

Following the ASTM C282-10-2015 [13] standard, porcelain enamel samples were exposed to 10% C₆H₈O₇ citric acid to investigate the effects of acid attack on the enamel surface. After 15 min of exposure at room temperature, the samples were properly prepared for SEM analysis.

As shown in Figure 10, the image on the left, morphological boundaries appear across the entire surface. However, they are significantly less pronounced compared to those observed after exposure to water. It is well known that glassy matrices tend to exhibit lower chemical resistance under alkaline conditions, which may explain the more pronounced degradation observed in other tests. This aspect is well-known in the literature [15].

More interestingly, the needle-like calcium-based structures appear unaffected by the acid. As clearly visible in Figure 10, Figure 11, and Figure 12, these structures remain intact and are present beneath the surface, extending to a depth of approximately 10 µm. However, localized matrix degradation appears to occur around these acicular structures, potentially indicating that the surrounding glass phase is more susceptible to acid attack in those regions. This may result in partial exposure of the calcium structures beneath the surface, leaving them intact—an interpretation that remains a plausible hypothesis based on the observed morphology.

3.6. Surface and Cross-Section Investigation After Alkaline Chemical Attack

The final analysis focused on the exposure of the enamel to an alkaline species. This experiment is particularly relevant, as porcelain enamel is commonly used to protect the internal surface of water heaters, where it is typically paired with sacrificial magnesium anodes. Inspired by ASTM C614-20 [14], and aiming to simulate this condition under more aggressive circumstances, the enamel was exposed to a 5% K₄P₂O₇ tetra potassium pyrophosphate solution for 6 h at 96 °C. The goal was to evaluate how the enamel behaves in contact with alkaline species.

As shown in Figure 13, structure boundaries on the enamel surface are significantly more pronounced compared to those observed after exposure to acidic or neutral environments. Furthermore, as illustrated in Figure 13 and Figure 14, the needle-like calcium-based structures were almost completely dissolved. Similar to what was observed during water exposure, the dissolution of these structures resulted in the exposure of the underlying aluminum-rich layer.

In Figure 15, where the enamel surface was exposed to the alkaline agent, signs of attack are also visible in correspondence with surface porosities. This suggests either that the alkaline environment preferentially targets these pre-existing porous regions, or that the chemical degradation of the glassy matrix gradually exposes subsurface porosities, making them more evident and susceptible to further damage.

This behavior contrasts with the effect observed under pure water exposure, where the attack is mainly localized to the needle-like calcium-rich structures, leading to more limited and superficial damage. In acidic conditions, on the other hand, no significant surface alteration was detected, indicating a lower aggressiveness of the acid environment toward both the matrix and surface features of the enamel.

4. Conclusions

Porcelain enamel showed significant surface alterations under all tested conditions, especially in the presence of pure water and alkaline species, both representing harsher conditions than those typically found in real applications. Although visible degradation and elemental release were observed, no signs of structural or chemical instability were detected. The surface composition and microstructure of the enamel played a central role in the degradation process: needle-like calcium-rich structures were the first to dissolve, triggering localized attacks. In contrast, elements like aluminum and cobalt were released more gradually, indicating a progressive degradation of the glassy matrix. ICP analysis allowed for both the monitoring of release dynamics and the quantification of released elements, confirming that none exceeded the limits set by the European Union for potable water. These results emphasize the importance of optimizing the enamel’s composition and microstructural design to ensure durability in aqueous environments, particularly in contact with drinking water.