Effect of Magnesium on Glaze Melt at Cone 6

[Min]

This thread is jumping off from this one where the subject of the efficacy of magnesium as a flux in a cone 6 glaze came up and the prospect that  the magnesium alumino-silicate eutectic is unlikely to form at cone 6. It's probably best to read that thread for background before this one.

To test whether or not magnesium acts as a flux at cone 6 I'm going to run a test with the  glaze below. Purpose of the test is to see what effect the MgO has on the melt. I know the first version of it does melt well at cone 6 and results in a clear gloss glaze. For the second version I have removed the talc  (that  supplied nearly all  the magnesium). I have rebalanced the silica and alumina but left everything else the same. Testing if the removal of the talc effects the melt fluidity of the glaze. In theory the first glaze should have more fluidity if the MgO is acting as a flux.

There is 0.1 MgO left but this is as low as I can get it without messing around with the other materials. It's a bit of a wonky recipe insofar as materials used and totals as it was part of an experiment I was doing blending different recipes.

@BobMagnuson, does this look like a suitable recipe candidate for this test?

Also, if anyone has the eutectics for a boron calcium magnesium silica alumina melt that would be terrific! Or even one with boron magnesium alumina silica. I found some for boron + magnesium here.  The last column is in degrees C.

 

 

[BobMagnuson]

I think this is an excellent recipe for the test.  My guess is that the modified glaze will be somewhat more fluid, because MgO generally increases the viscosity of the melt.  A negative effect of removing the MgO will be that the modified version will be more likely to craze.

Thanks for bringing up magnesium borate!   When I added borate eutectics to EuCal I didn't find magnesium at all.  That's on me.  It should be added, and it does explain how a fair amount of magnesium can exist in a transparent cone 6 glaze.  But there's no silica in magnesium borate, so it still doesn't explain how it might be acting as a flux.  

[Hulk]

The 1170 °C/2138°F eutectic I'm seeing in Bloomfield* is "Calcia" - the "Magnesia" (MgO) is 2471°F.

Any road, very interesting; my first read into anything beyond the general idea**.
My low expansion liner glaze clear (clouded with Zircopax, so clear-ish) has .279 MgO, Unity, from talc.
Before I started adding Zircopax, it came up clear.
The Pottasia, Soda, Lithia and Calcia eutectic temps are all below my Cone 5 target, but the Baria and Magnesia are well above Cone 5.
Are there more layers of complexity with multiple eutectic happenings? Looks like it's melting well.

Ah, the Thornton article includes a eutectic with MgO and Boric Acid that's below 1000°C, per Min's post.

Interesting!
 

*Bloomfield_May17CM.pdf (lindabloomfield.co.uk)

**things that melt at a higher temperature by themselves can melt at lower temps when mixed with certain other materials.

Edited October 2, 2023 by Hulk
edit

[Bill Kielb]

Fairly high boron levels in both, but especially the second recipe. Holding boron constant might be a good thing here.

Edited October 2, 2023 by Bill Kielb

[Min]

3 hours ago, Bill Kielb said:

Fairly high boron levels in both, but especially the second recipe. Holding boron constant might be a good thing here.

I thought about that but my logic was if the magnesium isn't working as a flux then it won't come into play as such in either recipe so wanted to keep the other variables at a constant if that is indeed the correct scenario.

[Bill Kielb]

Not sure. The Katz and Carty research indicates boron requirements are fairly linear and predictable. I mention because 0.43 BO approaches the requirements to melt a glaze at 04. I am not aware if the data are flux specific though. Interesting !

https://ceramicartsnetwork.org/docs/default-source/uploadedfiles/wp-content/uploads/2008/10/tf-boroninglazes-0912.pdf

[Min]

45 minutes ago, Bill Kielb said:

Not sure. The Katz and Carty research indicates boron requirements are fairly linear and predictable. I mention because 0.43 BO approaches the requirements to melt a glaze at 04. I am not aware if the data are flux specific though. Interesting !

https://ceramicartsnetwork.org/docs/default-source/uploadedfiles/wp-content/uploads/2008/10/tf-boroninglazes-0912.pdf

Yup, I'm familiar with boron levels but for the purposes of this experiment boron levels are not the issue. I would rather just change one variable at a time. Not looking at the hardness, durability etc of this glaze, just the effect the magnesium has on it. 

If you have a recipe for a clear gloss that contains magnesium sourced from talc I'm more than happy to add that to the testing I'll do this week.

[Bill Kielb]

29 minutes ago, Min said:

If you have a recipe for a clear gloss that contains magnesium sourced from talc I'm more than happy to add that to the testing I'll do this week.

Only one I can think of is Marcia’s matte, a Stull matte so you would likely want to gloss it up. Looks like Si needs to go to 20 g though to get there or at least close. Just going by a 2018 studio paper of the progression and tested path from matte to gloss for the glaze. Thanks!

 

 

 

Edited October 3, 2023 by Bill Kielb

[Min]

Sure, let me know how much silica you want me to add. Just to confirm it's this one? https://glazy.org/recipes/19734

I did try that a few years (as published) ago and it was really unsuccessful for me, cutlery marked really badly and stained from the sumi ink.

[Bill Kielb]

That’s the one - amtalc. It looks like 20g of silica gets you near 7:1 which progressed from very matte to satin to gloss. The recipe was made to dial in a level of desired matte / gloss just by raising the silica, just because everyone has a different concept of matte / satin. The excerpt above looks like it finished testing at 20g / 100 as glossy enough to finalize the progression and results. You likely could go higher as well, just never went any further and not intended as a gloss glaze.

This clear along with Bills hard candy clear (high gloss) were a pair of glazes developed with additional boron to eliminate issues over heavy underglaze and certain underglaze colors. The additional boron seemed to help a great deal and allowed the brush artists freedom to paint, layer etc…. It was tested at cone 5 and appeared to melt well, never lower nor its upper end. It is so stiff, it never moved for us but was used at cone 6 with good results.

Edited October 3, 2023 by Bill Kielb

[Min]

 

I made a comment in the thread leading up to this one in regards to Linda Bloomfield and her statement that magnesium starts fluxing at 2138F/1170C. This temperature was pulled from her Feb 2016 Ceramics Monthly article titled Phase and Eutectics. I contacted her and asked where she got her data from for that number. She very kindly emailed me back and said it was from Hamer and Hamer The Potter's Dictionary of  Materials and Techniques. I can't do a screen shot of the pages for copyright reasons but if you have a copy the "Eutectic mixture" and "Magnesia" entries cover it thoroughly and is where Bloomfield got her 2138F figure from. 

In a very condensed nutshell Hamer and Hamer say given calcium is in just about every glaze with magnesium, (though the magnesia-alumina-silica eutectic is at a higher temp than calcia-alumina-silica), in practice, the glaze usually starts to melt closer to the calcia-alumina-silica eutectic at 1170°C. Potassium and sodium will also be present in most glazes, so these will bring the melting temperature down even further. The degree to which magnesia acts as a flux varies. Below 1170C it acts as a refractory, from 1190C through 1230C fluxing is "most marked". When used as an auxiliary or secondary flux, at under 0.25 molar  it can greatly increase glaze fluidity.

@BobMagnuson, if you don't have a copy of that text send me a pm and we can figure out a way for me to get those sections to you. If you still would like for me to run those tests I'm happy to do so but I really don't think it necessary given the info from Hamer and Hamer. It also meshes with my experience using a fair bit of magnesium in my low expansion glazes. (plus I have 16 people coming for dinner this weekend)

Edited October 3, 2023 by Min

[BobMagnuson]

I was finally able to lay my hands on a copy of Hamer and Hamer "The Potters Dictionary...", and read the section on Magnesium Oxide.  The amount of information they present is really quite impressive.  I don't want to say they're wrong about MgO being an auxiliary flux producing "late fluidity" between 1190 and 1230 C (roughly cone 6 to 8.)  I'm just looking for examples demonstrating that statement.  If we can find examples where MgO is clearly an active flux, perhaps we can begin to understand how it works.

My thinking goes like this:  A flux, in ceramics, is something that lowers the melting point of silica.  All of the other R2O and RO oxides (Li2O, K2O, Na2O, CaO, ZnO, SrO, and BaO) do this by forming alumino-silicate eutectics melting in our typical temperature range.  The MgO alumino-silicate eutectic point, on the other hand, is about 1350 C, or around cone 13 or 14.  There is a magnesium borate "eutectic" that melts much lower, but it doesn't have anything to do with silica.  And in that binary system, I would say that boron is the flux and magnesium is the refractory.  It will melt and become an important part of the glaze, but it doesn't necessarily help melt the silica.

Even if MgO CAN be an active flux at cone 6 to 8, it doesn't always.  I would bet most of us have used glazes in this temperature range where MgO is an opacifier. So if Hamer and Hamer are correct, there must be specific conditions for it to become a flux.

There are possible explanations for this, including: there could be a simple MgO eutectic mixture that we haven't discovered yet.  Or there could be a complex eutectic mixture formed when MgO dissolves in the primary R2O-RO-alumino-silicate melt.  (ZnO, for example, requires CaO to form its low melting eutectic.) At this point, we just don't know.

So, I would like to find glaze recipes where MgO is believed to be an active flux.  Then, if MgO, and only MgO, is removed from the recipe, the glaze should be less mature, or at least less fluid.

The more important question here, in my opinion, is that if MgO is not acting as a flux, becoming part of the clear glass part of a glaze, does it belong in the RO flux column of our glaze calculators?  If we take it out of the RO column, the R2O in the R2O:RO ratio is higher than we thought and perhaps the glaze is less durable than expected.

[Hulk]

Thornton's article lists three Boron Magnesium eutectics; this is the lower temperature one (last column is °C)

MgO·B2O3

Magnesium oxide and boric oxide

MgO 36.6, B2O3 63.4

988

If I'm reading right, the Boron melts much much earlier.
If I'm reading right ii, my low COE liner glaze has enough boron to melt with the MgO...

My low COA liner glaze looked clear to me - before I started adding Zircopax to it.

Now I'm curious and interested (again); what's melting the seventy + moles of Silicon? almost eight moles of Alumina?
I'm not getting the maths.

P.S. Hope the dinner was smashing Min!
...was just trying to remember the last time we had that many guests

 

Edited October 12, 2023 by Hulk
P.S.

[Babs]

Just absorbing what I can. Hamer and Hamer, I think, names Mg as a "catalyst " at  lower temps and a flux at higher...have to look up the definitions of those terms in chemistry.

Lay talk, catalyst "encourages" a reaction, flux allows reaction to occur at a lower temp?

[Min]

12 hours ago, BobMagnuson said:

There are possible explanations for this, including: there could be a simple MgO eutectic mixture that we haven't discovered yet.  Or there could be a complex eutectic mixture formed when MgO dissolves in the primary R2O-RO-alumino-silicate melt.  (ZnO, for example, requires CaO to form its low melting eutectic.) At this point, we just don't know.

I would suggest that we do if fact know this is what is happening. We don't use oxides in isolation with just SIO2. From the Hamer and Hamer in the "Eutectic Mixture" entry starting with the paragraph that begins "Eutectic mixtures in glazes are not a simple matter of one oxide with silica but a case of each oxide reacting with every other...." then the continues for a few paragraphs down to "...would be a transparent, if coloured, glass."

In a nutshell what they are saying in Hamer and Hamer is the oxides don't work in isolation, one oxide will form a eutectic with another then once that is complete the other oxides will get involved in the melt. Adding heat excites the molecules as does adding more than one other type of oxide / molecule. Bloomfield also confirmed that CaO + MgO with SiO2 will create a eutectic that melts well below that of MgO + SiO2 alone. 

12 hours ago, BobMagnuson said:

Even if MgO CAN be an active flux at cone 6 to 8, it doesn't always.  I would bet most of us have used glazes in this temperature range where MgO is an opacifier. So if Hamer and Hamer are correct, there must be specific conditions for it to become a flux.

Yes. Excess MgO isn't going to be taken into the melt, just as excess Al2O3 (or many other oxides) won't be and will result in a matte or opacified type glaze. But the fact that we can readily get a transparent glaze that contains MgO does indicate it is taken into the melt and is acting as a flux within certain boundaries.

(@Hulk, dinner was a success, house was a disaster zone afterwards though!)

[Hulk]

Interesting! Thanks to all contributing to this thread.

My guess, given the amount of Boron in my glazes (that contain MgO), the MgO is (likely) all melting, early, per the previously mentioned eutectic.
However, fully melted may not necessarily mean "active flux" as in lowering the melting point of Silica and Alumina.

I'm still reading "refractory" as not melting/melted.

Words!
From "DOE Explains" article, "A catalyst is a substance that speeds up a chemical reaction, or lowers the temperature or pressure needed to start one, without itself being consumed during the reaction."
Chlorophyll (green in plants) is certainly the first catalyst I'd heard of; the bead coatings in catalytic converters also comes to mind - cars were being fitted with them some forty-eight years ago...

The term "flux" as used in ceramics/pottery for glaze, clay, etc. may be different from the worlds of metal - soldering, brazing, welding, smelting.
...err, certainly is different! Wikipedia has a nice long article on Ceramic Flux; MgO is listed.
The term in medicine (never mind!), math, physics, general state/status, magnetics/electrics - and more! - each somewhat distinct.
 

 

[High Bridge Pottery]

On 10/12/2023 at 2:48 AM, BobMagnuson said:

I was finally able to lay my hands on a copy of Hamer and Hamer "The Potters Dictionary...", and read the section on Magnesium Oxide.  The amount of information they present is really quite impressive.  I don't want to say they're wrong about MgO being an auxiliary flux producing "late fluidity" between 1190 and 1230 C (roughly cone 6 to 8.)  I'm just looking for examples demonstrating that statement.  If we can find examples where MgO is clearly an active flux, perhaps we can begin to understand how it works.

My thinking goes like this:  A flux, in ceramics, is something that lowers the melting point of silica.  All of the other R2O and RO oxides (Li2O, K2O, Na2O, CaO, ZnO, SrO, and BaO) do this by forming alumino-silicate eutectics melting in our typical temperature range.  The MgO alumino-silicate eutectic point, on the other hand, is about 1350 C, or around cone 13 or 14. 

 

Here's an experiment where MgO is fluxing silica and possibly better than CaO, not sure what temperature. Transactions of the American Ceramic Society Volume 17, page  236. Copyright 1915.

 

 

Edited October 24, 2023 by High Bridge Pottery

[High Bridge Pottery]

Bigger triangle 

[neilestrick]

In my experience, MgO only contributes to opacity via crystal growth causing matteness, which in my book does not make it an opacifier. I would only consider it to be an opacifier if it can make a glaze go opaque without affecting the surface qualities, like zircopax does.

I have used several glazes over the years that will fire clear and glossy when cooled quickly, but matte and opaqe when cooled slowly, due to the magnesium in the glaze. I have also used magnesium-containing glazes that are perfectly clear no matter how they are cooled. It all depends on the percentage of magnesium and fluidity of the glaze. The clear I am currently using contains more magnesium than is recommended by limit formulas, but it's still incredibly clear and glossy.

[Min]

Magnesium zirconium silicate acts as an opacifier,  don’t know how much the magnesium plays into the opacifacation though. 
https://digitalfire.com/material/magnesium+zirconium+silicate

[Bill Kielb]

9 hours ago, neilestrick said:

In my experience, MgO only contributes to opacity via crystal growth causing matteness, which in my book does not make it an opacifier. I would only consider it to be an opacifier if it can make a glaze go opaque without affecting the surface qualities, like zircopax does.

Agreed, my observation as well over time. In fact the Marcia’s matte above is a clear matte and becomes clear glossy with only the addition of silica. It’s part of the exercise to show that a true Stull matte will go glossy based in the Si:Al ratio, all other components remaining the same.

Definitely not saying it can’t be, but if so …….. likely under specific conditions.

[Hulk]

This thread (and the thread it spawned from) piqued, err, re-piqued my interest in eutectics, generally, and also the role of MgO in mid-range glazes, specifically.

...then I thought it best to let it ride for a while; time's up, for forgetfulness looms!

The triangles (and pyramids), surely, don't capture the complexity of even the simpler glazes.
The articles I'd like to read co$t.
Others have, over centuries, conducted tests, lots of tests*...

The Excel workbook EuCal_ver_1_91 indicates there's less excess Alumina and Silicon in my liner glaze when 6.3% talc (Texas) ingredient is included; I believe the differences are significant.
With talc .33 and 3.24, Alumina and Silicon, respective;
W/O talc, .51, 4.36

Why?

Lots of work went into that EuCal!

The talc in my liner glaze, it's for lowering COE.
Per my testing, it helps with that.

Eutectics, hmm. Jess McKenzie, "The point where the liquid solidifies is more to the point than where the dried slurry of mixed powders liquifies."
hmm
See "liquid phase sintering," "Quaternary Piercing Point," "deformation eutectic," Ian Currie's book, particularly the MgO chapter, and lots of other stuff. ...then get back to work!

Frits and mixtures of different powders - expect them to behave differently, for they Are Different.

Melt, ye glaze, then hang on the pot without moving (much) over a range of temperature - so I have some wiggle room with the heat work - there, d'ya hear me, Glaze Spirits? Bond to the clay, provide a smooth sound wear resistant surface, be beautiful (at least interesting, err, not hideous?), inert, and uncracked.
 

*test, that is "where it's at!"
 

 

 

 

Edited October 31, 2023 by Hulk
because, and spelling

[PeterH]

2 hours ago, Hulk said:

Eutectics, hmm. Jess McEnzie, "The point where the liquid solidifies is more to the point than where the dried slurry of mixed powders liquifies."

If you have it handy I would welcome a reference.

Does your" hmm" indicate that you have a problem with the statement?

PS

I think that this statement is a reflection of the fact that phase diagrams are "funny" in the sense that they represent what happens given sufficient time to reach stability. 

The classic example is that of carbon, a part of the phase diagram is given in:
https://opengeology.org/petrology/8-igneous-phase-diagrams-and-phase-equilibria/

Diamond is thermodynamically unstable at pressures less that 10s of kilo-bars. So at normal atmospheric pressure of 1 bar it apparently shouldn't exist! 
... given sufficient time it will convert to graphite ... but it persists over geological timescales.

[Hulk]

Peter, I'd copied the quote to the notes I was keeping whilst reading up on eutectics; I wanted to credit the source.
I'm not finding it again, almost sure it was from potters.org; sorry I haven't found it, I did look!

"hmmm" on account of several notions it points to (imo):

  Several authors point out that dissimilar particles in a dried slurry behave differently on heating than powder ground up from previously melted/solidified material of same proportions.

  Typical glaze slurry may have several eutectic "things" happening, much more complex than the three legs of a pyramid.

  Liquidation and solidification likely occur over a range of temperature change and takes time (like butter, as others have pointed out), which is probably good.

As for "Why?" does my liner glaze with the talc have less excess Silica and Alumina, per EuCal, than the "same" glaze without talc (MgO) my guess is that untangling the math would answer.
From there, testing could prove.

   

[Min]

Throwing out a thought that might simplify this debate. 

Q: what colour response does cobalt give in a high magnesium glaze? (including midrange)

A: shades of purple

If the magnesium wasn't entering the melt as a flux then no purple.

Dave Finkelnburg article on cobalt purples and magnesium.