A puzzling glaze result

[Kelly in AK]

I’ve been pondering this one a while, I had a few pots do this a year or two ago and it’s always puzzled me. I’m not after repeating it, I’ve evolved the glaze into something different, but it still fascinates me. If I ever wanted to chase this I would have no idea how to think about it. 

The glaze was:

Frit 3134-  50 
Frit 3124-  30  
OM4 -  20 
add 10% zircopax, 3% copper carbonate

It’s fired to cone 03 and reduction cooled to 1500° F (propane). It usually goes all red, but in some parts of the kiln stays blue- green. Not unexpected. The way it changes color here in such distinct places surprised me. If anyone has thoughts or explanations I’m delighted to hear them. 

[Babs]

So you brushed a second layer on inside of bowls? Possible you used another glaze there? Quite crazy, eh? Or a slip over the glaze?

[Bill Kielb]

8 hours ago, Kelly in AK said:

t’s fired to cone 03 and reduction cooled to 1500° F (propane). It usually goes all red, but in some parts of the kiln stays blue- green. Not unexpected. The way it changes color here in such distinct places surprised me. If anyone has thoughts or explanations I’m delighted to hear them. 

One possibility - is the atmosphere. To maintain reduction, especially copper reds one needs to provide a rich carbon monoxide atmosphere.  If sufficient oxygen is allowed, the the piece simply reoxidizes to whatever extent it can. I should add, small a small amount of tin is usually a very important component for copper red glaze recipes and can have a pronounce effect on copper reds. Not sure if there is any in this recipe though. 

The example picture below contains test tiles mapped at various places throughout the kiln. These mapped tiles along with the pressurization schedule for the kiln often provide a way to ensure future firings are reduced evenly for that kiln with that specific pressurization / atmosphere. Copper red is very sensitive to reoxidizing so often a nice way to test.

 

Edited July 9, 2023 by Bill Kielb

[Russ]

Im with Bill. Ive seen this alot over the years with high boron copper reds in a supposed reduction atmosphere. Ive often wondered if the green is produced right after the firing while the glaze is still quite fluid but susceptible to an introduction of oxygen.

[Kelly in AK]

It’s a single glaze @Babs, poured in and out, then the rim dipped. What gets me isn’t the color change, that’s not a surprise. It’s the angular nature of it. I do suppose it’s got something to do with varying thickness of the glaze. 

[PeterH]

30 minutes ago, Kelly in AK said:

It’s a single glaze @Babs, poured in and out, then the rim dipped. What gets me isn’t the color change, that’s not a surprise. It’s the angular nature of it. I do suppose it’s got something to do with varying thickness of the glaze. 

It looks to me like the angular nature could be a crack pattern showing. I've no idea why different regions separated by cracks should show different colours.

[Kelly in AK]

Yes Peter, definitely cracks. They’ve melted together, some on the rim are filled with glaze showing as red lines on the mottled background.

And, to note, the pieces rang like a bell out of the kiln. They’re in my cupboard, not in the “for sale” pile. The smaller ones still ring, this one (cereal bowl) now has that dull thud that means its days are numbered. 

[Bill Kielb]

Maybe …………… They almost seem to be layered, so the difference in thickness likely influences the non reduced color. A very cool overlay effect! Interesting also, usually (not always) the interior of things can be reduced more fully because they are in a less turbulent air stream. Pretty cool effect though.

[Babs]

Were they subject to a 'draft', unsealed  top vent? Oxidising the top shelf, wierd, would you take an order? :-)))))

[Kelly in AK]

They were on the bottom shelf of a downdraft kiln. The trickiest place to get a good reduction (in my kiln, anyway). Babs, when I figure out how to do this on purpose you’ll get first pick.  Thank you for the compliment. 

[PeterH]

Puzzled brainstorming.

Wild idea for the colours, but not the cracks, fuming copper pooling in the bowl?

 I don't claim to fully understand copper reds. But fumed copper is presumably more likely to be present near the surface (and blue/green?), while there is a persistent argument that (some? all?) of the red is down below the surface.
The Dual Mechanisms of Tin Oxide in Copper Red Glazes https://glazy.org/posts/168150

On second thoughts ...

We seem to have a "paint-by-numbers" of regions: each  of more-or-less uniform colour; with the sections delineated by cracks that almost certainly weren't there when the colours were formed.

Large local differences in composition might explain the positioning of the cracks, but it's hard to see how that could happen. Could it be caused by differences in surface texture/crystallisation? That might add surface reflections (colour), and also influence crack initiation/propagation (position). 

Any chance of a specular-reflection free image of the same location?

PS >And, to note, the pieces rang like a bell out of the kiln. 

Can you remember if they had the angular colour pattern before they cracked?
Yes - the colour differences are controlling the position of the cracks? Surface texture influencing both?
No - the cracks are causing the colour differences? Is it conceivable that the release of stress caused by the crack changes the surface texture?

Edited July 10, 2023 by PeterH

[Babs]

On 7/10/2023 at 6:57 AM, Kelly in AK said:

It’s a single glaze @Babs, poured in and out, then the rim dipped. What gets me isn’t the color change, that’s not a surprise. It’s the angular nature of it. I do suppose it’s got something to do with varying thickness of the glaze. 

That is what brought me to broad brushstrokes.

[Babs]

15 hours ago, Kelly in AK said:

They were on the bottom shelf of a downdraft kiln. The trickiest place to get a good reduction (in my kiln, anyway). Babs, when I figure out how to do this on purpose you’ll get first pick.  Thank you for the compliment. 

Air sucking back into kiln from leaky damper?

[Min]

Looks like a massive version of a snowflake glaze pattern. 

[Kelly in AK]

Thank you for that article @PeterH, copper red is fun and fickle and I’m no better at understanding it after all these years. I can usually manage to get it nowadays though. I gave a shot to trying to get some better photos, but the mojo wasn’t with me and they’re no better than what I posted. When I have a few minutes to play with the camera I’ll try to oblige. 

 @Min, it’s exactly what I first thought of when I saw it, big snowflake glaze.

On another note, some of the reds I get are very much at the surface. It’s the reduction cooling. They make it through the dishwasher fine, but they’ll scuff right off with a light sanding. 

[PeterH]

8 hours ago, Min said:

Looks like a massive version of a snowflake glaze pattern. 

I think you've got it.

Horizontal cracks within the glaze (acting as reflectors) would affect the colour (by reducing the distance light travels in the glaze) even if the colour is uniformly distributed in the glaze.

The small amounts of colourants in copper-reds might also be significant.
Snowflake Crackle https://ceramicartsnetwork.org/ceramic-recipes/recipe/Snowflake-Crackle
Color
The first thing I noticed when I did color tests with the usual colorants and opacifiers (copper carbonate, red iron oxide, chrome oxide, stains, Zircopax Plus, etc.) was that, as the color improved, the crazing ceased. This was because all oxides have expansion/contraction rates and adding them changed the CTE of the glaze enough to stop the crackle effect. I ran more tests with very low levels of colorants (under 1%), which kept the crazing yet still produced a nice color.

PS Do you have any similarly fired pots which  didn't show this effect? An ice-water stress test might be interesting.

PPS I wanted to suggest that the relative freedom from cracks near the edge of the bowl was a significant clue, but couldn't word it right.
So here is a picture of the effected -- with a smaller crackle size and hence smaller outside ring.
I've been testing a Snowflake crackle glaze. Here it is over coloured slip
https://www.reddit.com/r/Pottery/comments/8jjxxi/ive_been_testing_a_snowflake_crackle_glaze_here/

Edited July 11, 2023 by PeterH

[Kelly in AK]

Yep. I think that nails it! 

[Min]

What does your KNaO normally run at in your other glazes?

[Kelly in AK]

They’re mostly variations of 3134+3124+clay, then some coloring oxides and/or zircopax. The white liner I use is (% analysis) Na2O 6.11, K2O 0.81. A glaze that behaves similarly to the one I originally posted (I haven’t seen the odd fractured color patterns in this one…yet) has Na2O 5.03, K2O 1.45. Now, a caveat with those numbers is an ingredient in both glazes is my local clay. I use the generic term “red clay” as a substitute ingredient in Glazy, so I don’t really know how accurate the analysis is.

A third glaze I’ve used that has no local clay in it comes out at Na2O 6.34, K2O 0.35

[Min]

So if we just look at the original recipe in this thread and your white liner the combined sodium + potassium are roughly about the same. My snowflake crackle (my current avatar) has the combined KNaO at 0.62 (in unity). Granted I fire at ^6 but if you look at all the recipes in Peter's link above or any other snowflake recipe the KNaO is much much higher than your original recipe here. My hunch is there is something else going on here. Another reason I'm thinking this is because of the 3% copper carb and the 10% zircopax. Colourants really do reduce the snowflake crazing patterning, which is a good demonstration of the power of some colourants to reduce or eliminate crazing. Just for curiosity's sake it might be fun to try the original base glaze w/without colourants, apply it really thick and see what happens.

[PeterH]

I was looking at some of the colours in the original pictures and somewhat reluctantly came to the conclusion that they might indicate that the colour of the glaze varied with depth. Which of course some people have been saying for a long time about some copper reds.

This made we wonder what would happen if you applied two differently coloured layers of snowflake crackle glaze. Would you get interestingly coloured patterns?

Edited July 12, 2023 by PeterH

[Bill Kielb]

2 hours ago, PeterH said:

Which of course some people have been saying for a long time about some copper reds.

I think my best guess initially was layers which would seem to lead to color and transparency of the layer. Reduction could bring into play: copper concentration, level of oxidation / reduction,  etc….. , not necessarily a direct function of copper reds but their tendency to be very red when fully reduced and often green / grey when re oxidized.

I suspect layers of different colored glaze would behave in a subtractive manner and level of transparency would also significantly affect the color viewed.

Definitely an interesting piece though.

[Min]

4 hours ago, PeterH said:

This made we wonder what would happen if you applied two differently coloured layers of snowflake crackle glaze. Would you get interestingly coloured patterns?

I've done this with a copper glaze layered over one with a vanadium blue stain, didn't really blend, which ever one was the final coat over shadowed the colour below. 

[Kelly in AK]

I want to thank everyone for their ideas and critical thinking on this. I appreciate it. Again, it’s not something I’m going for, just something that happened and I wondered about. 

Yeah, it’s not a snowflake thing. I was hoping for an explanation where I didn’t have to imagine too much. But you’re right @Min, the copper and zircopax don’t support that and furthermore it’s not applied thick at all. 

But it is uneven. I pour glaze in, then out, rotating the piece. It produces an area that’s thicker on the pour side. There’s also a leading edge to the pour that produces a distinct difference in thickness. Then I dip the rim. It’s all very fluid and organic, but there’s some complexity I hadn’t thought of. Considering thickness is one variable for a copper red glaze, well, there’s a lot of variation in thickness here. 

My current thinking is in firing the glaze cracks on the way up, like in glazes that crawl, just not so extreme. The shape and size of cracks are determined in part by the thickness of the glaze. I reduce for around 30 minutes starting at 012, before the it begins to fuse. Those cracks create separate plates of glaze and based on thickness, reduction affects each one individually. As the glaze melts they fuse back together.

I reduction cool it to 1500° and that reliably produces reds for me, even if I fire a neutral atmosphere the whole time. Not sure how it fits, I just have a hunch it plays a role.

Here’s a different glaze doing not the same thing, but I believe something related. 

[PeterH]

But why in this glaze - and in Snowflake Crackle - do the cracks mainly appear to propagate parallel to the glaze surface rather than normal to it?

There only seems to be two options: somehow the stress in normal to the glaze surface, or something is causing the propagating crack to change direction from normal to parallel to the glaze surface.  ‡

PS The usually physical model for "fast" crack propagation just doesn't seem to apply here.†
Hierarchical crack pattern as formed by successive domain divisions. I. Temporal and geometrical hierarchy
https://tinyurl.com/3yu6wx6y

† There is another model for "slow" cracks (where several may be propagating at the same time), for example mud drying.

... and the less relevant
Hierarchical crack pattern as formed by successive domain divisions. II. From disordered to deterministic behavior.
https://tinyurl.com/26w8ap4p

‡ Maybe there is some link with conchoidal fractures ( which explains how flint-knapping works)?

† There is another model for "slow" cracks (where several cracks may be propagating at the same time), for example mud drying.

PS Irrelevant to the present discussion, but an example of complex 3D cracks.
Evolving fracture patterns: columnar joints, mud cracks and polygonal terrain
https://royalsocietypublishing.org/doi/epdf/10.1098/rsta.2012.0353