IMPROVEMENT OF ENAMEL SURFACES BY SOL-GEL COATING
Marta Krzyzak, Günther Heinz Frischat, Peter Hellmold
Institut für Nichtmetallische Werkstoffe, Technische Universität Clausthal Zehntnerstr. 2a, D-38678, Clausthal-Zellerfeld, Germany

Abstract
Sol-gel coatings composed of SiO2, SiO2-TiO2, SiO2-ZrO2 and SiO2-TiO2-ZrO2, respectively, were prepared on different technical enamels as substrates. Thin (from 80 - 300 nm) and thick (up to 2 μm) films were deposited by the dip and spray coating methods. After heating at 500 °C the coatings showed a good adhesion to the enamels. The chemical durability of the coated in comparison to the uncoated enamels was tested against acidic and basic attack in accordance with DIN ISO standards. The experimental results show that the thin SiO2 coatings increase the chemical resistance of the coated enamels against attack of oxalic acid by a factor of 4 to 22 compared to the uncoated ones. Especially efficient were SiO22films of 1 μm thickness, based on MTEOS-TEOS solutions. SiO22films do not protect the enamels against NaOH attack; however, a SiO2-ZrO2 coating on non-resistant low-network former enamel significantly increases its stability against this basic attack. SNMS in-depth profiles show that in this case Ca and Ba diffuse from the enamel substrate into the sol-gel layer, obviously stabilizing it.

Introduction
Enamel coatings can improve the chemical and mechanical durability of steel. The temperature resistance limits this enhancement; the relatively low maximum temperature steel can be fired at and by the compositions of the enamels. To gain improvements beyond that limit, additional sol-gel coatings can be applied. Currently, sol-gel coating is a mature technology to improve surface properties of a wide range of substrate materials, including glasses, ceramics, plastics and metals, e.g. [1-4]. However, the development of sol-gel coatings on enamels is challenging, because enamels are considered to be heterogeneous systems, which possess more complex compositions than other materials.
The aim of this work was to prepare SiO2, SiO2-TiO2, SiO2-ZrO2 and SiO2-TiO2-ZrO2 coatings on various enamels by the sol-gel route and to study their performance and behaviour as barriers against chemical corrosion both in acidic and basic solutions. Thin (up to 300 nm) and thick (up to 2 μm) coatings were applied by using dip and spray coating techniques. The role of chemical and morphological parameters of the enamels on the formation of cracks in the coatings was investigated as well. The adhesion and mechanical properties of the coatings were determined by cross cut, tape and rubber tests. It was also aimed to investigate the influence of coatings with regard to the cleanability of enamels.

Experimental
Substrates

Five different 10 cm x 10 cm technical enamels on steel sheets were used as substrates. These enamels had various chemical compositions, roughness and homogeneities. The chemically most instable enamel contained about 45 wt % network formers, whereas the more stable enamels had between 65 - 71 wt % of them. The other main components of the enamels used were network modifiers and intermediate oxides, depending on the type of enamel, see Table 1. A technical silicate float glass was used as a reference sample.

Table 1. Chemical compositions of substrates
Network
Formers
(wt %)
Network
Modifiers
(wt %)
Intermediate
Oxides
(wt %)
Non-resistant enamel for dishes
45.6
22.4
32.0
Resistant enamels:
65.2
15.8
19.0
Baking appliances enamel
Sanitary enamel
65.7
17.0
17.3
Household enamel
69.3
19.0
11.7
Resistant enamel for dishes
71.0
19.7
9.3

Thin film preparation
The sol mixtures for the thin films were prepared starting from tetraethoxysilane (TEOS; Aldrich), tetra butyl orthotitanate (Fluka) and zirconium (IV) propoxide (70 wt % in propanol; Fluka). The starting solution for the SiO2 film was prepared by mixing and stirring TEOS with 2-propanol, distilled water and 1 N HNO3 or 1 N HCl [5]. To prepare solutions for TiO2-SiO2 and ZrO2-SiO2 coatings, the starting solution for SiO2 films was mixed with the second solution of Zr- or Tiprecursors, stabilised with ethyl acetoacetate by stirring for 30 min [6]. The sol concentration was adjusted by adding 2-propanol as a solvent. Coatings were obtained by dipping the samples into the sol using the different enamels. The withdrawal speed varied between 1 and 5 mm/s. The coated samples were fired in the furnace at 500 °C for 1 h in air. This process was repeated in the case of multilayer coatings.
For the spray coatings a solution of TEOS, ethanol, H2O and acetic acid was used. After keeping the solutions two days at room temperature, the pH value was adjusted to ≈2 using nitric acid. This solution was diluted with ethanol and 1.3-butanediol (Fluka), which posses a comparably high boiling point and reduces evaporation losses of the droplets during spraying [7]. The spray gun (Profi IDB700, Revell, Bünde, Germany) was located 5-30 cm from the substrate and could be moved in x-direction, while the substrate could be moved in y-direction. The coated samples were dried for 1 h at 100 °C and heated for 1 h at 500 °C.

Thick film preparations
For the preparation of the thick films methylethoxysilane MTEOS (Fluka) was mixed with TEOS and alternatively with H2O or SiO2 sol [8], see Table 2. Concentrated HNO3 or HCl were added to adjust the pH value to about 1 - 2. After 5 minutes, the sol was cooled on ice. A single-step dip coating was used. After deposition the coating on the enamels, the samples were dried for 1 h at 100 °C and heated for 1 h at 500 °C.

Table 2. Chemical compositions of the films
Coating
Composition [mol %]
Thin Films: SiO2
SiO2-TiO2
SiO2-ZrO2
SiO2-TiO2-ZrO2

100
50:50
50:50
50:25:25

(TEOS)
(TEOS, tetra butyl orthotitanate)
(TEOS, zirconium (IV) propoxide)
(as above)

Thick Films: SiO2
SiO2
60:40
55:25:20
(MTEOS:TEOS)
(MTEOS:TEOS:SiO2 sol)

Film characterisation
Since the surfaces of the enamels were too uneven to measure the thickness of the sol-gel coatings on them, films prepared under identical conditions on the reference glass were scratched and the scratch depth was measured with a profilometer (Tencor P-1).
The appearance of cracks in the coatings was observed by optical and by scanning electron microscopes (SEM). In-depth profiles of the films were measured by the secondary neutral mass spectrometry (SNMS) technique, using an INA 3 Leybold spectrometer.
The corrosion experiments with the coated enamels were performed in accordance with the DIN ISO 2742 and DIN ISO 4533 standards in boiling aqueous solutions of citric or oxalic acid at 95 – 98 °C and in 1 N NaOH solution at 80 °C. The adhesion and the mechanical properties of the coatings were determined according to the cross cut test (DIN 53151), the tape test (DIN 58196-6) and the rubber test (DIN 58196-5). The influence of coatings on cleanability of enamels was investigated by a rubbing test (samples soaked with salted milk and rice and heated at 220 °C, in accordance with [9] and by a combination of an exposure in boiling water and a rubbing test (samples soaked with tomato ketchup, Heinz, and heated at 250 °C, in accordance with [10]).

Results and discussion
Both optical and SEM microscopes revealed uniform, homogeneous and crack- and macro porefree SiO2 films (TEOS, dip-coating) on all enamels used up to 140 nm thickness. However, these thin films did not coat larger defects of the enamels. For film thickness > 140 nm their behaviour strongly depended on the morphology and the compositions of the enamels.

Fig. 1 Optical micrographs of 250 nm thick SiO2 films on various enamels: (a) non-resistant enamel for dishes – 45.6 wt % of network formers, (b) sanitary enamel – 65.7 wt % of network formers, (c) baking appliances enamel - 65.2 wt % of network formers.

Fig.1 displays micrographs of SiO2-dip coated enamels with a film thickness of ≈ 250 nm. The surface layer of the low-network former content coated enamel appears quite uniform without cracks, whereas the coatings on the other enamels show cracks, whose formation largely depend on the network former contents, heterogeneities and roughness of the enamels. To obtain crackfree coatings for the corrosion tests and for an appropriate comparison of the results, the thickness was optimised to 90 nm.
By using TEOS and MTEOS in the dip solution uniform, homogeneous, crack- and macro pore-free SiO2 films of thickness up to 1 μm could be obtained on all the enamels, and by addition of SiO2 sol the thickness could be extended even up to 2 μm.

Using the spray-coating process also crack-free, uniform, homogeneous films of ≈ 100 nm (± 20 nm) thickness was obtained. The critical thickness of coatings on various enamels and nfluences of spraying parameters (movement, air pressure, distance substrate-spray gun) were not investigated.
The thin and thick coatings showed a good adhesion to the enamel substrates (class Gt0, Gt1 in the cross test and class K2 in the tape test) and no abrasion in the rubber test.

Acidic attack
The corrosion experiments were used to evaluate the efficiency of coatings as protective barriers, by comparing coated and uncoated samples. Weight changes per unit area were calculated. The chemical resistance of the substrates was tested in 6 wt % boiling citric acid at 95 °C for 2.5 h (non-resistant enamel) and for 24 h (resistant enamel). Due to the very low weight changes of the resistant enamels (< 1.0 g/m2), the results were difficult to compare to the coated enamels. Hence the chemical durability of the resistant enamel was performed once again in 3 wt % boiling oxalic acid at 95-98 °C for 24 h (resistant enamels) and for 10 min (non-resistant enamel). Fig. 2 shows uncoated and SiO2 -coated enamels after such an attack.

Fig. 2 Comparison of an uncoated and an SiO2-coated resistant enamel after an attack of a boiling solution containing 3 wt % of oxalic acid for 24 hours at 95 - 98 °C: (a) uncoated, (b) SiO2-coated.

Figs. 3 and 4 display the influence of the number of layers/film thickness on the chemical durability of the enamels. The chemical durability in case of the non-resistant enamel increased with the number of layers and in case of the resistant enamel no further improvement was determined. This effect is due to the formation of cracks in the coatings on the resistant enamel. The non-resistant enamel was up to 270 nm crack-free and the chemical durability increased with the thickness.

Fig. 3 Chemical resistance of uncoated and SiO2-coated (TEOS, dip-coating) non-resistant enamel in a boiling solution containing 6 wt % of citric acid at 95 – 98 °C for 2.5 h.

Fig. 4 Chemical resistance of uncoated and SiO2-coated (TEOS, dip-coating) resistant enamel in a boiling solution containing 3 wt% of oxalic acid at 95 – 98 °C for 24 h.

Crack-free 90 nm thick SiO2 coatings (TEOS, dip-coating) increase the chemical stability by a factor of 4 to 22 compared to the uncoated substrates, see Fig. 5, and it is also shown that the mass loss of the coating is independent of the substrates.

Fig. 5 Chemical resistance of various uncoated and SiO2-coated (TEOS, dip-coating) substrates in a boiling solution containing 3 wt % of oxalic acid at 95 – 98 °C for 24 h (resistant enamels)
and for 10 min (non-resistant enamel).

SiO2coatings of ≈ 1 μm thickness, prepared from MTEOS-TEOS solutions, were especially efficient against acidic environments. In most cases there were even no measurable mass losses.
This strong improvement of the chemical resistance by SiO2 coatings (on the basis of TEOS and of TEOS-MTEOS) is due to the low solubility of pure silica in acidic solutions [11].

Basic attack
Fig. 6 shows the results of the corrosion experiments in 1 N NaOH solution for 5 h at 80 °C. An improvement of the stability by the 90 nm thick coatings could be found only in the case of the lownetwork former enamel, however, the SiO2 coatings were dissolved completely in any case.

Fig. 6 Chemical resistance of various uncoated and SiO2-coated substrates in a solution of 1 N NaOH for 5 hours at 80 °C.

Fig. 7 illustrates a comparison of the mass losses of the resistant and non-resistant enamels, covered by various 90 nm thick coatings. The alkali resistance of the coated substrates strongly depends on the compositions of the enamels. Thus, in the case of the chemically more resistant enamel, no clear improvement could be obtained, whereas a SiO2-ZrO2 coating on the chemically non-resistant enamel increases the alkali resistance significantly.

Fig. 7 Comparison of the chemical resistance of various coatings on resistant and non-resistant enamel in a solution of 1 N NaOH for 5 h at 80 °C.

Fig.8 a shows the in-depth profiles of the SiO2-ZrO2 coatings on the non-resistant enamel before firing. After heating the concentrations of Ca and Ba are strongly enhanced in the SiO2-ZrO2 coating, see Fig. 6 b. Obviously a diffusional transport from the enamel to the coating has occurred. Since practically no corrosion took place, the intensities of the components in the layer did not depend on whether layers before or after corrosion were analysed. In this enamel the contents of Ca ≈ 9 wt %, Ba ≈ 5,5 wt%, Zr ≈ 12 wt % are higher than in the resistant enamel (Ca ≈ 2.8 wt %, Ba, Zr - trace elements), see Fig.8 c. The quantity of Ca and Ba ions in the resistant enamels is too low for stabilizing the SiO2-ZrO2 films, even if there is a tendency to form of calcium and / or barium zirconates [12]. Also the movability of enamel components is lower than in the nonresistant enamel, due to the higher transformation temperature. Fig.6 d shows the in-depth profiles of a SiO2 coating on non-resistant enamel. In this case Ca and Ba strongly diffused from the enamel into the layers without stabilizing it.

Fig. 8 SNMS in-depth profiles of films on enamels: (a) an SiO2-ZrO2 coating on a non-resistant enamel before heating, (b) an SiO2-ZrO2 coating on a non-resistant enamel after heating for 1h at 500 °C, (c) an SiO2 coating on a non-resistant enamel after heating for 1h at 500 °C, (d) an SiO2-ZrO2 coating on a resistant enamel after heating for 1 h at 500 °C.

Cleanability test
Selected experiments investigating the cleanability were performed on uncoated and coated enamel. According to the adhesion ability of the tested food, specifying salted milk with rice and Heinz ketchup heated at 220 °C and 250 °C, there was no difference found between the coated and uncoated enamels. Just at lower temperatures (200 °C) the experiments indicate a lower adhesion on the sol-gel coated enamel.

Conclusions
Sol-gel derived coatings of SiO2,SiO2-TiO2 and SiO2-TiO2-ZrO2 compositions with thickness between 90 nm and 2 μm were plated on several technical steel enamels. Corrosion tests with the coated enamels were performed in a boiling aqueous solution of oxalic acid at 95 - 96 °C and in a 1 N NaOH solution at 80 °C. Crack- and macro pore-free SiO2 coatings increase strongly the chemical resistance of all the coated substrates against the acidic environment, independent of the compositions of the enamels. SiO2-ZrO2 coatings significantly improved the chemical stability of non-resistant enamel against basic attack, obviously due to a diffusion process of Ca and Ba from the enamel substrate into the sol-gel coating during firing.

Acknowledgements
This work was founded by the Arbeitsgemeinschaft industrieller Forschungsvereinigungen (AiF), Köln under the auspices of the Deutsche Email Verband (DEV), Hagen, utilizing resources provided by the Bundesminister für Wirtschaft, Bonn.

References
[1] J. D. Mackenzie and E. Bescher, J. Sol-Gel Sci. Technol. 19, 23 (2000).
[2] T. P. Chou, C. Chandrasekaran, S. J. Limmer, S. Seraji, Y. Wu, M. J. Forbess, C. Nguyen, and G. Z. Cao : J. Non-Cryst. Solids 280 (2001), 153.
[3] O. de Sanctis, L. Gomez, N. Pellegri, C. Parodi, A. Marajofsky, and A. Duran : J. Non-Cryst. Solids 121 (1990), 338.
[4] S. P. Mukherjee, W. H. Lowdermilk: J. Non-Cryst. Solids 48 (1982), 177.
[5] W. Beier, M. Meier, G. H. Frischat: Glastech. Ber. 85 (1985), 97.
[6] U. Wellbrock, W. Beier, G. H. Frischat: J. Sol-Gel Sci. Technol. 147&148 (1992), 350.
[7] C. Löser, C.Rüssel: Glastech. Ber. 72 (2000), 270.
[8] H. Schmidt, G. Jonschker, S. Goedicke, and M. Mennig: J. Sol-Gel Sci. Technol. 19 (2000), 39
[9] CEN/TC 262/WG 5 N 184.
[10] CEN/TC 262/WG 5 N 082.
[11] H. Scholze, Glas: Natur, Struktur und Eigenschaften (Springer-Verlag Berlin Heidelberg New York, 1977), p.265.
[12] F.M. Misselwitz, Zur chemischen und mechanischen Resistenz zirconiumhaltiger Silicatwerkstoffe, PhD Thesis, Martin Luther Universität Halle-Wittenberg 1993, p.41.

 

 

The International Enamellers Institute
Tel +39-02-38103333  Fax: +39-02-38103288
All rights reserved - Best View in 800x600 pixels - Powered by Archi3