Soda ash dishwasher test confusing results

[Marie Lu]

Hi everyone, 

I am new to this forum and wanted to share some results from the soda ash dishwasher test I ran to test my own glazes which is leaving me quite confused. (Sorry for any grammar mistakes, I am not a native speaker ;)). 
 

So I have been working really hard on developing my own glazes and finally, after a lot of trial an error, came up with a couple I really like. Before sending them to a lab now I thought I should do the lemon and soda ash test. So a made a 5% soda ash solution and boiled/ simmered my pieces for 6 hours.

After that they were dull and white from the soda ash. When rinsed they would look completely normal again but as soon as the water dries off - dull again. 
 

i really did not expect these results because I tried to pay attention to R20, RO, Silica and Alumina levels. Maybe someone can check my recipe to see if I missed something or is this whole soda ash test simply not a good test to determine the durability of a glaze? I am feeling a little down about it all right now, so any help is greatly appreciated. 
 

I attached the base recipe for you guys. The colorants for the glaze are: 

A) 1.2 % nickel carbonate 0,5% cobalt oxide, 8% zircopax, 5% rutile 

B ) 0,5 cobalt oxide 7% zircopax 5% rutile 

C) 5% titanium dioxide 5% tin 3,2 % red iron oxide 

i fire to cone 6 in my electric kiln 

Thanks for your help, Marie 

 

[Bill Kielb]

Not all glazes with a decent R2O:RO are winners. A decent flux ratio indicates things could be durable, a bad flux ratio indicates likely not. I think of it as a good looking weld can be bad, but rarely is a bad looking weld good. No absolute here though. Dishwashers or Alkali testing can be very damaging to glaze surfaces. A change in visible surface gloss likely indicates significant  wear. I am not sure how this translates to dishwasher cycles though.

Pictures would be great, gloss meter even better and of course long term dishwasher for a perspective of actual longevity. One Thing I would try though is: you may have a glossier finish with a solid 7:1 to 8:1 Si:Al ratio. I think I would try this as well as subject some other known durable glazes to your alkali test and see how their surfaces perform.

Your test is for surface erosion not necessarily for leaching. Surfaces will erode chemically and physically by stacking, cutlery, etc.

Edited September 27, 2023 by Bill Kielb

[Callie Beller Diesel]

Hi and welcome to the forum!

What I’m really finding in my own testing is that colourants have a much stronger affect on the possible durability of the end result than is often assumed. I had a similar experience where a base glaze has pretty good numbers on paper, but certain unexpected colourant additions made some tiles fail a vinegar test (iron!?). My flux ratios and silica balance are similar to what you have listed, and I’m currently running some line blends to see if I can keep the same colour response while increasing the silica for durability. 

Dishwasher tabs do contain a lot of alkaline chemicals, including sodium carbonate (soda ash), that will etch all dish ware over time, not just pottery. It dissolves the silica. So testing ware in a way for dishwasher durability in a short amount of time that isn’t also too heavy handed is tricky, I think. I think the digitalfire recommendations that it looks like you followed may be too harsh.

 I’m going to tag @Minbecause I think she had different recommendations about the concentration of the soda ash solution, but I can’t remember the numbers she used. 

 

 

[Min]

Hi Marie and welcome to the forum!

23 minutes ago, Callie Beller Diesel said:

I’m going to tag @Minbecause I think she had different recommendations about the concentration of the soda ash solution, but I can’t remember the numbers she used. 

 

My experience with the 5% soda ash test when testing the black glaze Licorice from the book Mastering Cones 6 Glazes by Hesselberth and  Roy  led to a degradation of the glaze gloss level. I then went on to leave a test piece in the dishwasher for several months. Real world dishwasher tested did not show any degradation. I do think the soda ash test is too harsh a measure. Note it is within a hair from the "ideal" flux ratio. I did contact Roy about this and he didn't recall doing the soda ash test with this glaze.

If your glaze can take it increasing the silica and alumina in a formula will improve durability  regardless of flux ratios.

Regarding the "ideal" flux ratio of R2O: RO of 0.30:0.70, have a read of this snippet from Matt Katz's 2016 article Glossed Over: Durable Glazes from NCECA 2016

note the test for this article with a 0.1:0.9 ratio was "surprisingly robust" and the need for further research regarding the effects of silica:alumina ratio and the use of colourants. I don't think we are at an absolute conclusion yet in regards to flux ratios. They do not stand alone insofar as durability measures.

[Marie Lu]

Thank you fights for the quick answers! @Min I have made similar observations, the glaze that showed significant less gloss after the 6 hour boil has also been in the dishwasher almost every day for about 6 month now without any change in appearance. So I also feel like this test might just be too harsh. I wish I would have a put a regular porcelain mug from the store in too just to see how a commercially made one would hold up. 
 

@Callie Beller Diesel thanks for the warm welcome. Interesting that colorants do seem to have such a strong effect, even with good base recipes. Makes you really think that you always need to run tests even if it looks good on paper. Fingers crossed that you will be successful with the line blends. 
 

@Bill Kielb I agree that not all glazes are winners even when they look good on paper. I feel like though that maybe this test is too harsh because other glazes (variegated slate blue from mastering cone 6 glazes and a stable glossy glaze) also suffered through this test. Going more into the direction of glossy would be a good idea but I am trying to create a satin finish with the 6:1 ratio + slow cool. Thanks again for your time and Input 

[BobMagnuson]

I may be a lone wolf on this idea, but, I think the magnesium alumino-silicate eutectic is unlikely to form at cone 6, and therefore is unlikely to be part of the glass phase of the glaze.  So does MgO belong in the flux column?  Maybe not.  If you take it out, the R2O/RO ratio becomes 0.38/0.62, which would indicate a much less durable glaze.  The magnesium could be just an opacifier in this case.

[Min]

@BobMagnuson, how about if the magnesium is supplied by a frit, do you think it would be more likely to form a magnesium alumina silicate eutectic at ^6? Lone wolves ideas are more than appreciated chiming in on this, I've always thought there are more variables than just the R2O:RO ratio alone. What about when a slow cooled matte glaze has some of the calcium/strontium/magnesium etc precipitate  out of the glassy matrix, does the calculated flux ratio still apply, I don't think it does, at least not in the matting oxides entirety at least.

[Marie Lu]

@BobMagnuson hi Bob, thanks for your response. Could you go a little bit more into detail about what you mean by the  magnesiumalumino-silicate eutectic ? Sorry I have not heard about this yet. And when you say you don’t think it would form at cone 6, does this mean that when fired to a higher temp this eutectic would form? Thank you ,  Marie 

[Marie Lu]

Hi @Min, I am trying to make sense of the things you and bob are saying. Could you maybe explain why some of the strontium, magnesium, calcium would precipitate out of the glassy matrix in a slow cooled matte? 

[Min]

There are different methods of creating a matte glaze. Some matte glazes employ more than one mechanism to do so. The mechanism and oxides used will create different types of matte, some will feel butter soft (like a good magnesium matte), others will feel dry (like one created with barium), some cutlery mark easily (often these are high alumina mattes that don't use a secondary mechanism or oxide to create the matte) and so forth. Oxides that when used in higher amounts can cause matting are magnesium, barium, strontium, titanium, zinc, calcium and alumina.

There are also underfired glazes that will be matte but this is simply because they haven't been fired hot enough to melt properly.

Slow cooling a glaze is one method of achieving a matte glaze if the glaze is high in one or more of the matting agents.  Particular oxide(s)  combine with silica and precipitate out to form a microcrystalline glaze. These micro crystals will scatter the refracted light causing a matte surface. In effect what is happening is the glaze is devitrifying as it cools.  Since these matting oxide(s) + silica micro crystals have a different composition from the remaining glaze I would argue the calculated R2O:RO flux ratio is no longer valid. 

There are exceptions to the slow cooling method of achieving a matte glaze, barium carb for example will cause a matte glaze regardless of the cooling schedule. Adding excess alumina oxide will also cause a matte as an excess of it will simply not be taken into the glaze melt. 

 

[BobMagnuson]

I think of all glazes as being made up of two parts: the clear glass base and the modifiers.  The clear glass base is made up of a primary eutectics (usually those of K2O, Na2O and CaO).  The modifiers are any oxides that don't fit within the eutectics.  They might be dissolved in the base, or they might be suspended in it.  Some might be dissolved in the base when it's melted, then crystalize out during cooling.  The eutectics themselves never crystalize...they form the clear glass phase.  The modifiers are usually colorants and opacifiers.

Some oxides, like CaO, SrO, or BaO are usually fluxes that form eutectics, but if either SiO2 or Al2O3 is short, they will simply be opacifiers.  I previously used an example of a white glaze that was opacified by an excess of CaO.  It could be turned into a clear glaze by adding EPK to fill out the CaO-Al2O3-SiO2 eutectic.

A ceramic flux is an oxide that lowers the melting point of silica, usually by forming an alumino-silicate eutectic.  Since the melting point of the MgO-Al2O3-SiO2 eutectic is somewhere around cone 13, I'm not at all sure that magnesium ever acts as a flux, regardless of whether it comes from dolomite, talc, or a frit.  It's certainly an important glaze ingredient for its other properties. It's just not a flux.  But I could be wrong about this.  If you disagree, show me a glaze that would be less mature if you remove the MgO.

For some background and my method of calculation, see Techno File: Using Eutectics, February 2018 issue of Ceramics Monthly.
 

[Marie Lu]

Hi @BobMagnuson, thank you so much for your in depth response. If I got it right then regarding my glaze problem I maybe should take out the magnesium for my calculation, calculate the base again so I get a better balance of the R20/RO ratio and then add magnesium only for other reasons. Would you suggest a cutoff/limit regarding the Mg0 so a functional glaze that can be put on dishes is still produced? 

[BobMagnuson]

I don't have a maximum MgO for you, but I would start in the 0.2 to 0.4 range.

[Min]

Interesting comments about magnesium. So in your view is what Linda Bloomfield says about magnesium being an active flux above 1170C / 2138F incorrect? Also, do you have the melting point for a CaO + MgO + SiO2  Al2O3 eutectic? 

[BobMagnuson]

Min. Thanks so much for your questions!  The whole reason I presented my EuCal spreadsheet in the first place was to start a discussion on this way of looking at glaze calculation.  I would never say that Linda Bloomfield is incorrect, just that I have not been able to find the data to back it up.  That's not to say it isn't out there, just that I haven't found it.  There are zillions of phase equilibrium studies out there, many behind paywalls beyond my reach. As to MgO-CaO-Al2O3-SiO2 eutectics, the only ones I have been able to find, so far, have liquidus temperatures above 1350C.  I'll keep looking though.  And I'm still looking for a glaze where MgO is an active flux.

[Min]

Super interesting for sure! My clear has .25 magnesium (from frit) in the unity formula. I’ll try and remove it to get rid of the magnesium and use a magnesium free frit(s) to keep frit percentage the same (given Frits in general melt easier than most raw materials) and see what happens. If you have a suggestion for any glaze test comparisons let me know and I’ll run those too. I’m firing (^6) next week.

another edit: I'm going with a MgO matte (with slow cooling) that used talc to supply MgO instead. 

I have stumbled across an article from NASA where they used  calcium, magnesium (plus silica  & alumina) to make a glass, I seem to recall the melting point wasn’t exceptionally high, I’ll try and find the article tomorrow. I think there was a phase diagram in the article.

edit:  @Marie Lu, we have gone off topic a bit here. Debating over the effect of magnesium in regards to fluxing at^6, I do think it acts as an auxiliary flux but open to discussion and experimentation.

Edited October 1, 2023 by Min

[Marie Lu]

@Minno worries at all, really interesting topic. Maybe you can tag me somehow if you share the Tests you will run, I would be interested in hearing about it. Also thank you for taking the time to responding in depth about matte glazes. I am also thinking about running some tests, bringing up the CaO and cutting back on the MgO in my glaze and see if this will affect the melt. Would this mean that a glaze containing higher amounts is Mg0 will always appear not fully glossy because of that fact that it does not melt? 

[Min]

11 hours ago, Marie Lu said:

Would this mean that a glaze containing higher amounts is Mg0 will always appear not fully glossy because of that fact that it does not melt? 

Have a look at this article, it goes into how MgO works at mid range acting as a flux at around ^6 compared to lowfire where it acts more as a refractory agent causing the glaze to opacify. Yes, there is a limit to how much of it will be taken into the melt.

I'll start a new thread on how MgO effects glaze melt.