search  current discussion  categories  glazes - specific colors 

ti blues in the abstract truth

updated sun 16 nov 97

 

Karl P. Platt on thu 13 nov 97

This is a long post and one I hope will be useful in developing better
and stable Ti Blues. This color is frequently referenced here as being
quite fickle. It is and I've sought to explain a little bit as to why
this is the case. Hopefully there'll be a few clues in here to
facilitate your being able to obtain repeatable results.

There's another version of a Ti Blue post I made here several years back
stored on the SDSU Ceramic's page -- the address of which I don't have
here at the moment.

Regards,

KPP


Maybe it s time to revisit Ti Blue................ Besides, it s the end
of a glorious, crisp warm day up here in the barraco de computador 100
meters above Praia Mole after two weeks of #&$%-ing chuva . Yes, I
got to use my sunga today -- he said "sem vergonha."

Ti is added to glass/glaze either as the oxide TiO2 or in some mineral
compound, such as the stuff they call rutile.

The oxide TiO2 is used abundantly to make things look WHITE paper or
toothpaste or house-paint are a couple of examples. As such it s cheap
and clean as commercially pure materials go.

TiO2 has also been long introduced into glazes combined with iron in a
mineral substance called rutile. We ll consider this raw material a bit
more further below, but this substance should not be confused with the
real Rutile, which is actually a crystalline form of TiO2.

TiO2 can exist in three different forms : Rutile, Anatase and Brookite.
At room temperature Rutile is the stable form. Above 800 C or so Anatase
is favored. Anatase is not easy obtain at room temperature you can t
just heat and cool TiO2 and end-up with pure Anatase at the end of the
process it s a bit more involved. Anatase can be present as part of
the crystalline material present in ceramic glazes. Brookite has also
been spotted in glazes.

As an aside, Anatase, deposited as thin films (by exotic methods) on
glasses and elemental Si has been found to have very interesting
photochemical properties. More to the point, Anatase decomposes water
into hydrogen and oxygen by simple exposure to the sun.

TiO2 is added to glazes for several reasons

A. To Produce Opacity.

B. To Modify Colors

C. To Produce Color

TiO2 has been added to produce opacity in glazes since deep antiquity.
This is owing to the low solubility of TiO2 in the glassy material
presiding in glazes. To the extent that TiO2 can be dissolved in glaze
it has potent effects on the colors given by many of the metal oxides (
Cu, Fe, U etc....) dissolved in glass to produce color.

The low solubility of TiO2 in SiO2 glasses is given to the fact that Ti
ion is a lot bigger than Si ion and it would rather sit between 6 oxygen
ions instead of the 4 oxygens preferentially offered by the arrangement
we find prevalent in Si based glass (glaze). As a structure resembling
TiO2 has been found in SiO2-TiO2 glasses, it s obvious that the Ti ion
manages to satisfy itself between only 4 oxygens. There are limits as
to how many maladapted Ti ions can stuff their way in between only 4
oxygens in the SiO2 based glaze. Ti ions really want 6 oxygens and Si
really doesn t want to give them over. This is the source of the limited
solubility of TiO2.

Compared to an SiO2 grouping in the glaze, TiO2 in the represents a
weak spot in the glassy material. Owing to this weakness, TiO2 is added
to pure quartz as a flux in making fused quartz, where it lowers
viscosity markedly in teeny additions.

Another effect of the weak spots TiO2 puts in the glaze is that it
provokes crystallization. This happens for a couple reasons. As with
sugar in water TiO2 is more soluble in hot glaze than cold glaze. If the
amount of TiO2 is only slightly above the amount the glaze can hold, and
the glaze is decently homogeneous, the crystals formed are very small
and uniformly distributed organizations of some thousands of atoms.
These little organized regions are called nuclei as they present a place
from which further crystalline material can grow.

The material that proceeds from the TiO2 nuclei can either be a bigger
TiO2 crystal or some other type of crystal such as willemite (ZnSiO4)
in ZnO saturated glazes i.e. crystalline glazes. What crystal forms
depends on whether TiO2 or some other substance is more saturated in the
glaze at a given temperature.

All of this, however, is of limited interest to us in considering Ti
Blue colors as Ti blue colors do not depend on the presence of
crystalline material in the glaze crystals are present as a bonus or a
bane depending on what you re trying to acheive.

The blue color is given by the reduced form of Ti oxide Ti2O3. Ti does
not like to be reduced and to develop this color special conditions are
required of the glaze.

In order to obtain Ti2O3 Blue, the glaze must contain a good bit of B2O3
and it must be fired in reduction.

B2O3 rich glaze is preferred because B2O3 presents glassy material
looser in structure than SiO2 glassy material, which facilitates
reduction.

The looser structure also makes for runny glazes (as a rule) and this
frequently leads to crazing and/or glaze running off the pot. To an
extent this can be controlled through adding a bit more Al2O3 to the
glaze, but this has strict limits.

We should also recall from above that the presence Ti reduces the
viscosity of SiO2 glasses, too, and contributes to increased fluidity
and expansion.

The Ti blue color is quite sensitive to the concentration of Ti2O3
developed in the glaze by reduction. Diluting the concentration of
Ti2O3 a little reduces the color by a lot sorry I can t put a graph
here, but it shows a dramatic relationship between the intensity of the
color and the concentration of Ti2O3. As such one wants a glaze which
will dissolve as much Ti as possible. Again, the looser Borosilicate
glaze facilitates this by possessing a structure more friendly to the
oresence of Ti2O3.

It is unwise to add Ti in any form other than TiO2 even if they re
cheaper.

The stuff sold as rutile to potters has large amounts of iron (Fe) in
with the TiO2 and is notoriously variable in composition. Remember, it
only takes a very small reduction in the amount of Ti2O3 present to
cause the blue color to fade drastically. Add the Ti as TiO2 and there s
never any question as to how much Ti you start out with. If you really
like the effects given by the combination with iron (Fe) with Ti (Pill
Bottle Brown) , add some iron to the glaze, too.

For the same class of reasons, adding B2O3 to the glaze should be done
by using one of the well-known frits. The use of mineral borates which
are both notoriously inconsistent in composition and water soluble,
opens you up to a whole range of problems you don t need to fool with.
These problems range from having your glaze slip highly thixotropix --
at best with the consistency of glue -- to having your Ti Blue fail to
develop as a consequence of erratic composition in the glassy material
formed.

Reduction needs to be fairly strong to develop the Ti2O3 Blue usually
a good bit stronger than for other reduction colors. For this reason, Cu
Reds fired together with Ti2O3 blues are often muddy when Ti Blue is
well developed or vice versa, where the Ti Blue is weak and the Cu Red
is nice. Of course you can strike a balance between the two, too and
obtain nice effects. And then again, some people like muddy Cu reds.

An interesting sidebar to all of this would be that adding a percent or
two of phosphate (P2O5) to the glaze can strongly influence the color
making it both more blue and more sharply defined.

Here s why for the real glaze weenie -- non-glaze weenies can skip this
paragraph:

The blue color is due to the electronic transition of the 3d1 electron
of the Ti+3 ion between the 2T2 and 2E energy levels. This occurs at
about 18,000 cm^-1 in phosphate glasses and 21,000 cm^-1 in Borate
glasses. In Phosphate glasses , the 18,000 cm-1 transition is much
more more distinctly defined (the color is more sharply defined) and the
color shifts to shorter wavelengths the color has a greater blue
component. Also, there's a another absorption peak in either glass
related to the Jahn-Teller effect -- related to the degenerate ground
state of the Ti-O groupings -- at 14000 cm^-1 which is more sharply
resolved in the presence of P2O5.

Adding a little P2O5 to the glaze to encourage these effects isn t a
wholly bad idea. You don t lose anything with respect to Ti solubility
or reducibility and the potential effect is nice your color will be
more saturated and somewhat more sharply defined. However, P2O5 has a
number of quirks. First, it is volatile and a lot of it will vaporize
away in the firing -- heavily fritted glaze helps here by facilitating
solution of the P2O5 ASAP.. Second, P2O5 doesn t mix at all with SiO2
glass and not real well with B2O3 glass. Since the TiO2 Blue glaze will
be a borosilicate glaze (containing both B2O3 and SiO2) adding P2O5 in
any notable amount will result in milkiness and/or a more pronounced
tendency to crystallize, and a reduction in viscosity.

The effects of P2O5 are worth fooling around a little bit to obtain. As
a production glaze effect, P2O5 effects have the main disadvantage of
being sensitive to the amount to P2O5 left in the finished
glaze remember, it s volatile. In all but the most well defined
circumstances it ll be tricky to obtain strictly repeatable results
but this can be an effect, too.


To recapitulate:

The intensity of a Ti2O3 blue is *highly* sensitive to the amount of
Ti2O3 in the finished glaze

The amount of Ti2O3 in the finished glaze depends on how much you put
added to the batch in the first place, and the degree of reduction to
which the glaze was exposed during the firing.

Further, the glaze needs to contain a good bit of B2O3. B2O3 facilitates
the reduction and lends to an acidic chemical environment which also
favors the development of Ti2O3 I didn t go into this much, but will
if anyone cares to know. Suffice it to say for now that Ti2O3 (more
specifically the Ti+3 ion) is favored in acidic environments as has
been shown in aqueous solutions (wet chemistry). In this light the
amount of B2O3 can be seen to have a critical effect on Ti2O3
development and you re better off to have some certainty over how much
B2O3 is in the glaze -- i.e. avoid ghastly borate.

To guarantee the amount of Ti present in the glaze is consistent, it is
best to add it in as the commercially available oxide instead of the
mineral alternatives which are well known to be fugitive.

Reduction is a critical step in the development of the color. The
reduction required for Ti2O3 blues is generally stronger than for other
reduction colors (Cu Red, for example) Insufficient reduction will yield
pale colors given a diminished quantity of Ti2O3 in the finished glaze
TiO2 is colorless. All that said, this furnishes an excellent example as
to where a metering orifices have a huge advantage over oxygen analyzers
which cannot measure reduction (and which cost a lot and which wear out
and which require batteries which also wear out)

Again, using mineral borates is asking for trouble. It s no secret that
I have a strong aversion to putting substances like Ghastly Borate into
glazes. Ti2O3 Blues, with their narrow requirements for development, are
a real good example as to when to avoid using things like Ghastly Borate
in glaze.

I haven t said anything about the profound effects Ti has on other
colorants and will save these for some other time. However, it s worth
noting that Ti can be used to expand your pallet quite a little.

OK, I think it s time to go eat a fish the anchova looked particularly
good today when I was in Barra da Lagoa and a little trip to Bar do Neni
is in order. Nene is very much a native to this island ( a Manezinho, as
they re called) and has an amazing sotaque or accent. Really, it would
be better referred to as a dialect and it s a riot to talk with him. As
he d say : Queresh, queresh. Nao queresh deish .

And with that, I m outta here............

KPP marvelling at the teeny waves today.

Grimmer on fri 14 nov 97

Karl,
Interesting article. Thanks. Sounds like TiO2 Blew glaze bases share a
lot of characteristics with Cu Red glaze bases (low alumina, low
viscosity, borosilicate, etc), no?
You've mentioned metering orifices once or twice before: Could you
elaborate?

Thanks,

steve grimmer
marion illinois

ps are you in Brasil or Portugal?

Karl P. Platt wrote:
>
> ----------------------------Original message----------------------------
SNIP!
> The blue color is given by the reduced form of Ti oxide Ti2O3. Ti does
> not like to be reduced and to develop this color special conditions are
> required of the glaze.
>
> In order to obtain Ti2O3 Blue, the glaze must contain a good bit of B2O3
> and it must be fired in reduction.
>
> B2O3 rich glaze is preferred because B2O3 presents glassy material
> looser in structure than SiO2 glassy material, which facilitates
> reduction.
>
> The looser structure also makes for runny glazes (as a rule) and this
> frequently leads to crazing and/or glaze running off the pot. To an
> extent this can be controlled through adding a bit more Al2O3 to the
> glaze, but this has strict limits.
>
> We should also recall from above that the presence Ti reduces the
> viscosity of SiO2 glasses, too, and contributes to increased fluidity
> and expansion.
>
SNAP!

> Reduction is a critical step in the development of the color. The
> reduction required for Ti2O3 blues is generally stronger than for other
> reduction colors (Cu Red, for example) Insufficient reduction will yield
> pale colors given a diminished quantity of Ti2O3 in the finished glaze
> TiO2 is colorless. All that said, this furnishes an excellent example as
> to where a metering orifices have a huge advantage over oxygen analyzers
> which cannot measure reduction (and which cost a lot and which wear out
> and which require batteries which also wear out)
>
CLIP!
> And with that, I m outta here............
>
> KPP marvelling at the teeny waves today.

Karl P. Platt on fri 14 nov 97



Dan C Tarro wrote:

> I would ask your help on one simple issue and maybe it should be assumed.
> There is some discussion on reduction cooling, will this help to prevent
> the re oxidation of the Ti in the glaze? or isn't it a problem after it
> has reached the stage of reduction?
>

Maintaining reduction on cooling can't hurt. However, it's not clear that it
is necessary.

> One more, The iron in the rutile ore, does this play any part in the
> formation of the blue? Could a good blue glaze be produced with just the
> use of Ti alone?
>
>

Iron plays no role at all in the development of the Ti blue -- the color can
be developed in the absence of Fe.

Karl P. Platt on sat 15 nov 97



On Fri, 14 Nov 1997, Grimmer wrote:

> ----------------------------Original message----------------------------
> Karl,
> Interesting article. Thanks. Sounds like TiO2 Blew glaze bases share a
> lot of characteristics with Cu Red glaze bases (low alumina, low
> viscosity, borosilicate, etc), no?
> You've mentioned metering orifices once or twice before: Could you
> elaborate?
>
> Thanks,
>
> steve grimmer
> marion illinois
>
> ps are you in Brasil or Portugal?
>


Steve --


Cu reds aren't particularly sensitive to the bas glaze in which they're
made. For example, it is possible to obtain Cu reds in a simple bead of
borax glass (as mineralogists do in their tests) or in a cone 15
electrical porcelain glaze.

I was thinking about writing something about Cu reds and maybe this
weekend will allow time. Cu reds are sensitive to the presence of things
like Sn and Zn, but aren't too picky about the glassy material they live
in. The glassy material can affect "reoxidation" on cooling, but this is
actually a pretty rare thing to have happen. Usually the firing wasn't
done right in the first place. Reduction is the critical variable in
processing.....ah, but this gets real long real fast and I can't go into
it now.

Metering orifices are simply orifices placed in a gas or air line. They're
mounted so that by hooking-up a $45.00 magnahelic gauge (or a U-tube,
ifyou prefer) and reading the pressure drop across the orifice you can
determine the volume of gas (or air) passing through the orifice to within
a couple percent. Metering Orifices cost about $60.00 - don't need
batteries, etc,etc


Estou no Brasil

Abracos.

KPP