VITREOUS ENAMELS WITH IMPROVED CHIP RESISTANCE
Dechun Fu, A. W. Gee and William D. Faust
Appliance Division, Ferro Corporation, Cleveland, Ohio USA

Abstract
Improved chip resistance of vitreous enamels has been demonstrated by control of glass thickness, incorporation of ceramic toughening agents, and plasma applied ceramic interlayer.
Potentially stronger enamel systems for critical applications may be possible while maintaining the desirable properties of vitreous enamels.

Introduction
Introduction
Chipping of porcelain enamel on steel or cast iron upon impact has been aconcern of enamelers and suppliers to the porcelain enamel industry fordecades. This is because of the negative impression consumers develop aboutporcelain enamel if the product chips and then rusts due to steel exposure or ifthe appearance is marred by the presence of a patch of dark ground coat in themiddle of an otherwise smooth, light-colored coating.
This paper addresses some of the questions that arise out of this concern about chipping of porcelain enamel on steel: how can chip resistance be quantified, what are the factors affecting chip resistance, and how can chip resistance be improved?
Earlier studies describing the chip resistance of vitreous enamels have been subjects of Porcelain Enamel Institute Forum papers for many years(1).
Currently, a series of papers has been presented on the examination of the fracture surfaces(2) and factors in materials and design.
This paper describes a range of variables and processes which has shown rogressive improvement in the chip resistance of vitreous enamels. Glass thickness, the introduction of toughening agents to control the enamel compressive stresses, and deflection of cracks and substrate surfaces that have significantly different elasticity have potential to aid in control of crack propagation.
In 1999, a PEI report Factors Affecting the Chip Resistance of Enamels(3) concluded that the most significant factor in chip susceptibility was radius ofcurvature of an enameled part followed by ground coat firing temperature and cover coat thickness, thinner being better. The thickness of the enamel has always been a factor that in practical terms needs be kept to as thin as possible to increase flexibility. A study by Horton(4) concluded that thinner coatings also show improved torsion resistance.
Examination of the fracture surfaces has shown that normal vitreous enamels exhibit the same fracture surfaces as brittle solids such as glass(2). The fracturepatterns can assist with the identification of the initiation of fracture due to impact and the effect of inclusions.
The current "state-of-the-art" for enameling is thin coatings achieved by minimizing the enamel thickness through tight process control and using new materials such as electrostatic powder vitreous coatings for two coat - one fire application. However, chipping is still an inherent problem of vitreous enamel materials. Various methods of enhancing the performance of vitreous enamels were studied.

Background of materials and processes for study
Enamel chipping is the removal of material from the surface of a coated metal substrate as a result of impact.
In such an event, the impact velocity is usually low, about 3.5 meters/second or less and the impact energy small, about 2.7 Joules or less. The impact test adopted for this work involved dropping a 224-gram stainless steel ball from various heights directly onto a curved enamel surface (figures 1 and 2).
Hertzian and Vickers indentation, finite element analysis, high-speed photography, stress measurements, and fractography were utilized in analysis. Impact test results for enamels made with various toughening agents, multiple coats and thermal spray are summarized.

Experimental details
Test panels with the size 15.24 cm x 15.24 cm were made by spraying and firing laboratory milled enamel on interstitial-free steel (AK Steel) of 1.4 mm (0.056 inches) thickness. The panels were bent to a radius of curvature of 9.3 mm and cleaned by alkaline degrease prior to use.
To allow the use of toughening additives, the ground coat formula was adjusted by reducing the refractory content (silica and zircon) to ensure proper enamel-to-steel bonding on the fired panels. After milling to a proper fineness, the slips were adjusted to a specific gravity of 1.75 g/cc and the pick-up to 25 grams on a 15.24 x 20.32 cm flat panel following traditional enameling procedure. For ground coat applications, approximately 9 grams of slip was sprayed on the front of the panels and 6 grams on the backside.
After firing the ground coat at 838 °C for 4.5 minutes and cooling the panels to room temperature, 9 grams of cover coat slip was sprayed on the front side of the panels. The dried cover coat was then fired at 800 °C for 4.5 minutes. The bent panels were impact tested following the methodology of Faust, et al. (1998)(3). Average impact test values were taken from 8 data points.
The following three parameters were used to characterize the enamel chip resistance:

Figures 3 and 4 illustrate the test results on the panels.

Improving chip resistance by control of process variables for conventional enameling
Other means of improving chip resistance have been studied including the plasma coating of the metal with metallic and ceramic layers prior to enameling (Gee et al., 2000)(5).
To minimize chip size, including the possibility of no chip, the radius of curvature of drawn parts should be as large as possible.
The ground coat should be fired at as high a temperature as possible, the ground coat coefficient of thermal expansion (CTE) should be as high as possible, and cover coat thickness should be minimized.
To reduce the probability of any chipping for a given degree of impact, the total coating should be as elastic, thick, and strong as possible.

Evaluation of coating modifications
Various toughening agents were screened in ground coat and cover coat enamels.
The tested additives included fine particles, flakes, and whiskers.

Tested additives:

With these toughening agents, it was hoped that one or more of the followingtoughening mechanisms could be utilized:

Results verified the working of at least the first two mechanisms. Due to severe chemical reactions taking place between some toughening agents and the enamels during processing, some prepared panels were inadequate for impact testing. These included some metal powders, titanium diboride and SiC whiskers.

Indentation testing
Hertzian indentation was done on enamel surfaces to characterize the event of contact damage.
To determine subsurface damage mode, indentation was done along a joining line of two polished cross sections. Such sections were then separated for microscopic examination.
Stress-strain relationship was also determined by Hertzian indentation. Stainless steel spheres of several different sizes (R=3.18 to 12.69 mm) were used at a loading rate of 0.2 mm/second. Methodological details for these experiments were described by Pajares et Al.(6) (1996). Figures 5 and 6 illustrate the sequence of enamel cracking and chipping as determined by Hertzian indentation.
Displacement-sensitive indentation was performed using a hardness tester capable of measuring energy dissipation during the loading-unloading indentation cycle (Faber et al.(7) 1998). The loading rate was varied in the range of 200-2200 grams/second.
Figure 7 illustrates the effect of loading rate on enamel response to stress.

Testing results
Of the materials tested, mica and aluminum metal flakes showed some toughening. The threshold volume fraction is about 15% for aluminum and 30% for mica. Chips after impact become more symmetric with both of these materials. Unfortunately, the aluminum has a higher thermal expansion than the enamel.
Contracts more on cooling and leaves circumferential cracks around the metal particles, resulting in a porous coating. The aluminum also readily reacts with the enamels, as it is molten at enameling temperatures.
Severe outgassing was noticeable in the cover coat applied to aluminumbearing ground coat.
Mica addition to the ground coat has a more dramatic effect in preventing steel exposure than reducing chip size. In tested ground coat panels, mica addition effectively eliminated steel exposure. With about 30% volume percent of mica, the chip size was reduced more than 50% in size.
Further addition of mica, however, causes degradation of the enamel-tosteel adhesion that in turn induces larger chip size under impact.

Though zircon incorporation into the ground coat causes enamel-to-steel bond degradation which worsens the chip resistance of ground coat in a two coat - two fire system, zircon addition improve the chip resistance.
About 10% zircon volume fraction eliminated chipping at the 15 cm weight drop height and reduced the chip size at 30 cm and 61 cm drop heights.
This suggests that zircon's contribution to the beneficial increase in coating stiffness is more than offset its adverse effect on bond degradation. Thicker coatings should exhibit better chip resistance due to higher critical stress necessary for crack initiation and better containment of cracks within the coating. To verify the finding, test panels were made with multiple coatings.
Generally speaking, these panels showed better chip resistance than singly coated panels. Not only the threshold drop weight is raised, but also the probability of steel exposure is reduced.
However, multiple coatings require repetitive spraying and firing as well as excessive use of material.
To circumvent this problem while trying to achieve a similar effect, a single layer of ceramic oxide (e.g. ZrO2 or Al2O3) was used. These ceramic oxides typically have a Young's modulus several times that of porcelain enamel. In other words, a layer of ceramic oxide would be equivalent to several layers of enamel in terms of impact resistance.
To prove the concept, panels were made by thermally spraying alumina or zirconia followed by conventional processing of cover coat enamel. Impact test results show that such panels indeed have much better chip resistance than all other types of panels (Table 1).
Figure 8 shows the morphology of this type of ceramic interlayer.

Future directions
Based on the above results, thermal spray technology would merit further investigation. Though traditionally not used in the enamel industry, thermal spray technology is proven for producing reliable ceramic coatings for demanding aerospace applications.
It would be useful to conduct a thorough research on the economics involved in adapting thermal spray process to the manufacturing of enamel products and the feasibility of applying this technology to fabricate sanitary ware and other appliance parts.
Chip resistance of porcelain enamel can be significantly improved by incorporating toughening agents, multiple processing of conventional enamels, incorporating a ceramic interlayer, or combination of these approaches.

References
1.a. Smith, Paul L., "Chipping Resistance of Enamels", Proceedings of the Porcelain Enamel Institute Technical Forum, Vol. 3, 1938, pages 37-47, Vol. 4, 1939, pages 155-162, and Vol. 5, 1940, pages 130-133. 1.b. Peterson, A. and Andrews, A. I., "Relation of Metal Thickness, Enamel Thickness, and Bottom Radius to Impact Resistance of Porcelain Enameled Utensils," Journal of the American Ceramic Society, Vol. 28, No. 4, 1945, pages 102-109.
2. Faust, W. D., "Fractographic Examination of Porcelain Enamel Chipping Defects", Proceedings of the Porcelain Enamel Institute Technical Forum, Vol. 60, 1998, pages 69-79.
3. Faust, W. D., Gee, A.W., "Factors Affecting the Chip Resistance of Enamels", Proceedings of the Porcelain Enamel Institute Technical Forum, Vol. 61, 1999, pages 59-68.
4. Horton, M. A. and Stash, A., "Chip Resistance of Two-Coat/One-Fire Porcelain Enamels as Determined by Torsion Testing", Proceedings of the Porcelain Enamel Institute Technical Forum, Vol. 56, 1994, pages 1-6.
5. Gee, A.W., "Porcelain Enamels With Improved Chip Resistance", Proceedings of the Porcelain Enamel Institute Technical Forum, Vol. 62, 2000 (To be published).
6. Pajares, A. Wei, L., and Lawn, B.R., "Contact damage in plasma-sprayed alumina-based coatings", J. Am. Ceram. Soc., 79 (1996) 1907-1014.
7. Faber, B. Ya., Orlov, V. I. And Heuer, A. H., "Energy dissipation during high temperature displacement-sensitive indentation in cubic zirconia crystals", Phys. Stat. Sol.(a), 166 (1998) 115-126.

figure 1A - Schematic of test fixture and panel/box for impact testing

figure 1B - Schematic of Bent Test Panel

figure 2 - View of bent panel, untested

figure 3 - Bent panel, conventional enamel, 2 coat / 2 fire, impact test at 30 and 61 cm

figure 4 - Bent panel, plasma applied Al2O3, 2 coat / 2 fire, impact test at 45 and 61 cm

figure 5 - Sequence of enamel cracking and chipping as determined by Hertzian indentation

figure 6 - Subsurface damage mode analysis of enamel-steel bi-layer

figure 7 - Effect of loading rate on enamel response to stress

figure 8 - Morphology of ceramic interlayer