A ChatBot gives jaw-dropping glaze recipe advice - Jan 2025

It is January 1, 2025. I asked ChatGPT a question (in relation to the Ovo Perfect Matte glaze recipe). The conversational nature of its answer and the detail of its analysis and recommendations indicate what appears to be a depth of understanding. Its response is far beyond what the average potter or technician could give. Or even the material suppliers! This is the future. Here is my prompt:

Pottery glazes employing Gillespie Borate as a substitute for Gerstley Borate often do not turn out with the same greenish crystalized surface. They fire more blue. Do you know why? Its answer was repetitive, but here are the highlights:

The difference in the color and surface quality when substituting Gillespie Borate for Gerstley Borate in pottery glazes can be attributed to the distinct chemical compositions of these two materials. Here are the key factors contributing to the observed differences:

Chemical Composition

Gerstley Borate is a natural source of boron with a complex and inconsistent composition that includes borates, calcium, magnesium, sodium, and other trace minerals. This natural mix often leads to the formation of characteristic greenish, crystallized surfaces due to specific interactions during firing.
Gillespie Borate, on the other hand, is a synthetic or blended material designed to replicate the fluxing properties of Gerstley Borate, primarily focusing on its boron content. However, it lacks the same trace minerals and natural variability, which changes its behavior in glazes.

Color Development

The color difference, such as the blue hues you observe, may be due to the absence or reduced levels of trace elements like iron and magnesium in Gillespie Borate. These elements in Gerstley Borate can affect color responses, particularly in combination with other oxides.
Additionally, Gillespie Borate's cleaner chemistry may enhance the presence of other colorants in the glaze, such as cobalt or copper, which could intensify blue tones.

I gave a couple of recommendations that were imprecise but then asked: "Would you like a detailed suggestion for reformulating your glaze to address this issue?" Two hours later I responded:

Yes. Here is the recipe:
custer feldspar 39.6
epk 26.2
gerstley borate 13.3
wollastonite 12.8
whiting 8.1

It then said:

Reformulation with Gillespie Borate
Gillespie Borate typically acts as a direct substitution for Gerstley Borate by weight, but due to its lack of certain trace elements, some compensations are necessary.

Adjust for Crystallization and Color
Add Magnesium Carbonate (MgCO₃): Gerstley Borate contains magnesium, which can encourage crystallization. Adding 1-2% MgCO₃ can help.
Add Red Iron Oxide (RIO): To mimic the trace iron content that may influence the greenish tones, add 0.25-0.5% RIO.

Maintain Proper Flux Balance
Gillespie Borate is slightly more predictable as a flux. Ensure you monitor the melting behavior. If the glaze becomes overly fluid, consider reducing the amount of feldspar slightly (e.g., to 38%) and compensating with silica to retain the SiO₂/Al₂O₃ balance.

Adjusted Recipe

Custer Feldspar: 38.0
EPK (Kaolin): 26.2
Gillespie Borate: 13.3
Wollastonite: 12.8
Whiting: 8.1
Magnesium Carbonate: 2.0
Red Iron Oxide: 0.3

Testing and Observations

Firing Schedule: Test with a slower cooling cycle or a hold during the cooling phase (e.g., hold at 1900°F for 30 minutes) to encourage microcrystal formation.
Adjustments Post-Test: If the glaze still lacks the desired surface, consider tweaking the MgCO₃ or adding a small amount of zinc oxide (e.g., 1%) to enhance crystallization.

It then asked: "Would you like assistance with a recalculated chemical analysis of this revised glaze?"
I answered "yes" and it said:

It seems I cannot directly calculate the glaze analysis here due to a missing module. However, I can guide you through the process manually or with tools like Insight-Live or Glazy.org.

Here’s how to approach it:
1. Key Steps in Analysis

Determine the oxide contributions of each material (e.g., SiO₂, Al₂O₃, CaO, B₂O₃, etc.) based on their chemical composition.
Calculate the weight percentage of each oxide in the glaze.
Normalize the molar amounts to a standard unity formula.

2. Oxide Contributions of the Materials

Here’s a summary of the main oxides in your materials:

Custer Feldspar: High in SiO₂, Al₂O₃, and K₂O.
EPK: Primarily Al₂O₃ and SiO₂.
Gillespie Borate: B₂O₃ with some CaO.
Wollastonite: CaO and SiO₂.
Whiting: Pure CaO.
Magnesium Carbonate: MgO.
Red Iron Oxide: Fe₂O₃.

3. Tools for Calculation

You can use the following tools to calculate the revised unity molecular formula:

Glazy.org: Input your materials and percentages to get a breakdown.
Insight-Live: Provides detailed oxide analysis.
Spreadsheet: Use a glaze calculation spreadsheet with the oxide formulas of your raw materials.

Would you like me to provide the oxide breakdowns for the raw materials so you can plug them into a calculation tool?

Related Information

This glaze not working with Gillespie Borate

These two mugs employ the same cone 6 pottery glaze recipe, the high-feldspar calcium matte (Ovo Perfect Matte). Like other mattes, it is high in calcium carbonate/wollastonite and kaolin but has no silica. But the one on the left has 13% Gerstley Borate while the one on the right uses Gillespie Borate. Gerstley Borate is a complex material, one that mother nature has uniquely endowed. It is a brown powder, a mix of two calcium borate minerals, ulexite and colemanite. And it is plastic, very plastic, from a hyper-fine particled hectorite clay. And trace minerals. Gillespie Borate, by contrast, is a white powder, a synthetic blend attempting to replicate the obvious melting and physical properties of Gerstley Borate. It has, what some call, "a cleaner chemistry", enabling it to enhance rather than muddy whatever colorants are present. Any borate can melt well and foster crystallization, but Gerstley Borate is a mix of two borates that have different melting temperatures and patterns, this encourages phase separation and thus variegation in the aesthetic (its sub-micron clay particles may also act as catalysts).
What could be done? Add some iron to dirty-up the material, if well dispersed in the slurry 0.25% might be sufficient. While Gillespie also has MgO, it might not be in the same form. A 1-2% addition of magnesium carbonate could help. Even a small percentage of hectorite.

By Tony Hansen
Follow me on

Got a Question?

Buy me a coffee and we can talk