web analytics
Thermal shock resistant cement - the eternal search for the ideal mortar - Forno Bravo Forum: The Wood-Fired Oven Community


No announcement yet.

Thermal shock resistant cement - the eternal search for the ideal mortar

  • Filter
  • Time
  • Show
Clear All
new posts

  • Thermal shock resistant cement - the eternal search for the ideal mortar

    Not sure if this is the right section but. Please excuse the waffle! It's Monday morning at work and my haphazard brain gets carried away...



    First thing's first - Health & Safety;

    This mix contains Fly Ash, which is an industrial by-product from burning coal, and which in addition to being a respiratory hazard (silicosis), contains heavy metals.

    Exactly what heavy metals & what quantities will be dictated by where the coal was originally sourced from.


    Here are just a few links discussing toxicity. You will find some groups / pages which really talk up the toxicity, and which to me don't appear balanced - at the end of the day it is for you to do the research and decide if the benefits outweigh the risks for your application.





    From Wikipedia;


    Fly ash contains trace concentrations of heavy metals and other substances that are known to be detrimental to health in sufficient quantities. Potentially toxic trace elements in coal include arsenic, beryllium, cadmium, barium, chromium, copper, lead, mercury, molybdenum, nickel, radium, selenium, thorium, uranium, vanadium, and zinc.[46][47] Approximately 10% of the mass of coals burned in the United States consists of unburnable mineral material that becomes ash, so the concentration of most trace elements in coal ash is approximately 10 times the concentration in the original coal.[48] A 1997 analysis by the United States Geological Survey (USGS) found that fly ash typically contained 10 to 30 ppm of uranium, comparable to the levels found in some granitic rocks, phosphate rock, and black shale.[48]

    In 2000, the United States EPA said that coal fly ash did not need to be regulated as a hazardous waste.[49] Studies by the USGS and others of radioactive elements in coal ash have concluded that fly ash compares with common soils or rocks and should not be the source of alarm.[48] However, community and environmental organizations have documented numerous environmental contamination and damage concerns.[50][51][52]

    A revised risk assessment approach may change the way coal combustion wastes (CCW) are regulated, according to an August 2007 EPA notice in the Federal Register.[53] In June 2008, the United States House of Representatives held an oversight hearing on the Federal government's role in addressing health and environmental risks of fly ash.[54]

    Exposure concerns

    Crystalline silica and lime along with toxic chemicals represent exposure risks to human health and the environment. Fly ash contains crystalline silica which is known to cause lung disease, in particular silicosis, if inhaled. Crystalline silica is listed by the IARC and US National Toxicology Program as a known human carcinogen.[55]

    Lime (CaO) reacts with water (H2O) to form calcium hydroxide [Ca(OH)2], giving fly ash a pH somewhere between 10 and 12, a medium to strong base. This can also cause lung damage if present in sufficient quantities.

    Material Safety Data Sheets recommend a number of safety precautions be taken when handling or working with fly ash.[56] These include wearing protective goggles, respirators and disposable clothing and avoiding agitating the fly ash in order to minimize the amount which becomes airborne.

    The National Academy of Sciences noted in 2007 that "the presence of high contaminant levels in many CCR (coal combustion residue) leachates may create human health and ecological concerns".[1]


    TL/DR; Potential refractory mix - calcium aluminate cement (CAC) & Class F Fly Ash (a pozzolan) - that works at our temperatures; https://www.bnl.gov/isd/documents/81800.pdf

    A comparison of the heat resistance of calcium aluminate & ordinary Portland cement at temperatures relevant to pizza ovens; https://www.sciencedirect.com/scienc...50061815300490

    Let me preface this with the observation that literally thousands of ovens have been built with home brew mortar made from normal Portland cement, lime, sand & ground fire brick, and we have as far as I have read, zero reports of failure. The main structural integrity of our ovens comes from the dome shape, and they are build with some realistic expectation that the mortar will not last forever. But a properly built dome will none the less stand the test of time due to its inherently stability. In addition, *I am not a cement chemist* - but I do have a chemistry background.

    I would not recommend experimenting with mortar recipes when constructing a brick oven - testing should happen in a non-critical environment!

    Lots of discussion had been had on finding the 'ideal' mortar / cement for oven purposes. The prevailing view is that our particular environment is a difficult one for both ordinary Portland cement (OPC) and refractory cements such as calcium aluminates (CA). This is because the temps we might reach (500 C) significantly damage OPC, and while CA is slightly better, we can't get our ovens hot enough to sinter the mortar together (more aorund 1000 C).

    So we are in this funny weak spot in terms of temperature. In addition we are also often cornered with are shrinkage and cracking which could lead to weakening of the mortar.

    From what I have read, one of the main ways that cement gets damaged through heating is that the water it contains is driven off - first the general dampness, which pushed out too quickly (heating too fast) will crack cement - and then the chemically bound water which forms part of the actual cement structure.

    The changing mineralogy when the bound water leaves makes for weaker cement, as the new mineral phases are inherently less strong. Additionally, in our temperature range lime (CaO) is formed.

    If the cement stayed at this temperature this might not be so bad, but upon cooling, the lime absorbs water again, forming new minerals - which expand as they form! So, the cycle of heating and cooling gives not just a purely physical expansion and contraction, but a chemical / mineralogical phase change expansion and contraction - and this is where major cracking and strength loss occurs.

    Of course, the more heating / cooling cycles, the worse the effect. So you can see that for a pizza oven, hitting 500 C repeatedly and cooling to ambient - outside where it's often damp - we really are in a difficult spot for the mortar.

    All of this of course highlights how good the dome is as a structure! And that is where we should plan to derive our overall strength from.

    In terms of mortar we do things like using minimum cement / lots of refractory / inert fillers - sand & fire brick - which are unaffected by our heating and cooling cycles. Using larger grained fillers reduces shrinkage (see grog in pottery) but can make the mortar harder to work with. Using fine fillers can make the mortar easier to work with, but increases the water demand (water to dry ratio) of the mix, which decreases strength - less water makes for stronger cement. Ideally, fillers with a range of grain sizes are used, which give a balance between workability & minimising water use / shrinkage. I suspect that using lots of filler also gives more opportunity for water vapour to escape, reducing the chance of it cracking the cement.

    In addition to using refractory fillers (as an aside, our temperatures would not be considered 'refractory' by some), we use 'refractory' cements that are have more stable chemistries at high temperature, such as CA cement. The issue we have, as noted above, is that CA cement is actually only marginally better than OPC at our temps (see linked article) - it's much better at high temps, but we don't get there.

    So then I found this article looking at cements for geothermal wells, an environment that reaches the same temps as our ovens and is subject to heating & cooling cycles in the presence of water. They really thrash the mortar in this article - it gets abused in a way our ovens never would - cycles of 500 C then quenching in water.

    Maybe this is an avenue worth exploring for a new cement / binder system for our mortar; 50/50 Class F fly ash & high purity CA cement, initiated with sodium silicate. Initially with our standard refractory fillers - and then maybe we can explore other fillers from there.

    The articles I linked have great information on the chemistry of cements at temperatures that are actually relevant to us, well worth a read!
    Last edited by billyben; 01-13-2019, 11:11 PM.

  • #2
    Interesting article. Looks like adding silica activated flyash improves the robustness of CAC considerably.
    The TGA shows that CAC has a couple of fairly significant weight loss events at our operating temps, and adding flyash gets rid of the phase changes/decomposition events and substitutes a more gradual and ultimately smaller weight loss.

    The SEM pictures show we definitely can't get hot enough to form any useful crystal structure.

    I'd love to see what homebrew does in a TGA, but the guys we send our stuff to only use about 30mg. I sincerely doubt that 30 mg would actually be a representative sample. I suppose if i made a mix without the sand I might get a decent cement only sample.
    The lime and cement are quite finely ground, and I used unfired clay that should be quite tiny particles.

    The only other question is does the mix have to be autoclave? Can't really do that when you build the oven. I couldn't spot what pressure they used for the autoclave either.


    • #3
      Hey wotavidone!

      You work with the cement industry IIRC? Can you tell us any more about your job / the company?

      I assume the autoclave is just at atmosphere pressure, I can't see it stated otherwise. I know they steam cure pre-cast slabs for optimum performance, but both of these approaches are definitely a departure from how an oven goes about things in any case.

      I have obtained 40 kg of class C fly ash (http://www.goldenbay.co.nz/products/everplus/) and some CAC (https://www.calucem.com/products/ist.../istra-40.html) so I will be doing some experiments. I also have a kiln I can do temperature testing in.

      My materials differ from the above in that I have class C - not class F fly ash - which means mine has a significant proportion of lime in it (and so is self cementing), and lower 'quality' CA cement, which is lower in alumina and has a significant proportion of iron oxide in it, which I believe retards setting.

      My source for the idea that the Fe2O3 retards setting is this informative paper on fly ash in concrete;


      So far I have done two experiments;

      1) I had honeycombing on my table top when I removed the form, so I wanted to go around and patch it up. I used a mixture of 1:1:1 OPC : FA : Coarse sand ('builders mix" that I sieved to <3 mm).

      The fly ash seem to make it extra workable - smooth, creamy & easy to put a nice smooth surface on. Hardened well, no perceived delay of or speeding up of setting.

      2) Does it delay setting of my CAC (Istra 40)? - Made a 1:1 CAC : Very Fine Sand compared with a 1:1:1 CAC : VFS : FA.

      The FA containing mixture seemed to harden in a similar, but ever slightly quicker time. Firming up in 3-6 hours, proper hard overnight. Was much more workable, seemed to help hold the sand in suspension (would expect this from most fine clay like additives though I guess).
      Last edited by billyben; 01-13-2019, 11:08 PM.


      • #4
        No, I'm in the lead, copper, zinc, silver, gold, tellurium, cadmium, sulphuric acid industry.
        My limited understanding of refractories comes from my chemistry background and our experiences with refractory linings and high temperature blast furnace slags.
        Oh, and the company refractories primer.
        Which also says that CAC is weak at pizza oven temps.
        I believe some Fe2O3 is required for cement. I base this on discussions with people I know who are in the cement manufacturing game.

        Anyway, it seems flyash and CAC looks to be a pretty good system. Not sure why they say sodium silicate is needed to activate it.


        • #5
          For the record, I tried using sodium (meta)silicate with 40 % alumina CAC, Class C Fly ash and alumina cement and it severely retarded setting. So yeah, I agree, what are they up to with the sodium silicate? They do go into the chemistry in the paper, but it didn't work for me.

          Other geopolymers with microsilica (fly ash, silica fume etc) I've seen using sodium hydroxide or sodium hydroxide and sodium silicate as initiators.

          I kilned three samples at 565 C; (1) OPC concrete, (2) CAC / fly ash / fire clay / alumina sand (Shinagawa FG-30 from New Zealand), (3) CAC / fly ash / alumina sand and found that the CAC / fly ash / alumina sand mix (3) by far retained the highest strength.
          Last edited by billyben; 02-26-2019, 01:23 AM.


          • #6
            At 565C the OPC would probably be pretty much killed. The sample containing the clay would not be sintered at that temp, itís really then acting as a fine aggregate using up more power of the cementious material., so that leaves the last sample as your strongest, not surprised. Proprietary castables keep their recipes pretty secret so itís difficult to tell what they contain exactly, but they have been formulated to perform.Fly ash is added to concrete to improve flow and reduce water. A most desirable quality of any castable is its workability and correct water addition is only part of the answer. Apart from CACís tendency to go off really fast, it is also very temperature dependant and also quite thixotropic. The fly ash also acts to make the mix separate on vibration if the proportion gets too high, in the same way that excessive superplasticizer does, and CAC seems to be way more sensitive to it than Portland. You may end up making and firing thousands of samples by the hit and miss approach.You should also add some fine polypropylene fibres to your mix to assist safe water elimination. Proprietary castables usually have these already in the mix.
            Lastly you can use citric acid to retard setting, I use chilled water instead, I live in the tropics and the excessive heat is a problem. You may need warm water if the weather is cool.
            Kindled with zeal and fired with passion.


            • #7
              Fly ash is pretty much a mixture of oxides. It will behave a bit like cement.
              Addition of lime, calcium hydroxide,will make it stronger.


              • #8
                Originally posted by wotavidone View Post
                Fly ash is pretty much a mixture of oxides. It will behave a bit like cement.
                Addition of lime, calcium hydroxide,will make it stronger.
                The combination of lime and CAC is incompatible as it acts as an accelerant resulting in insufficient working time. It may be possible to retard the reaction, sugar works well, to balance the lime effect, but not tried it.
                Last edited by david s; 02-26-2019, 12:50 PM.
                Kindled with zeal and fired with passion.


                • #9
                  Originally posted by david s View Post

                  The combination of lime and CAC is incompatible as it acts as an accelerant resulting in insufficient working time. It may be possible to retard the reaction, sugar works well, to balance the lime effect, but not tried it.
                  ? I wasn't talking about adding lime to cac?


                  • #10
                    Originally posted by wotavidone View Post

                    ? I wasn't talking about adding lime to cac?
                    Ok, but billy ben has used CAC in two of his three samples. So I wanted to point out the problem with the combination of the two materials.
                    Kindled with zeal and fired with passion.


                    • #11
                      This abstract describes the reverse action of both sugar and citric acid on setting times for fly ash based geopolymers....complicated stuff. Itís little wonder modern science isyet to unlock the ancient Romanís secrets of their concrete technology. I prefer to use temperature control via water added to control setting time. That way itís not interfering with the chemistry.
                      Last edited by david s; 02-27-2019, 04:56 PM.
                      Kindled with zeal and fired with passion.


                      • #12
                        I did not use any lime in my mix.

                        I did not find the fly ash addition, at equal weights to the CAC, to retard or accelerate setting. It set in pretty much the same time. I read about the potential effects and experimented with this before I started using it on the oven.

                        I did notice some of the separation you mention - it was re-mixed immediately prior to application.

                        My CAC (Istra 40) was not of a super fast setting type. Even working with normal tap water on warm days (>25 C) we got about an hour of workability out of it, so more than enough when mixed in small batches. In the tropics at >30 C, might be a different story!

                        Istra 40 is:


                        Chemical composition %
                        SiO2 6
                        Al2O3 38Ė42
                        Fe2O3 13Ė17
                        CaO 36Ė40
                        MgO < 1,5
                        SO3 < 0,4
                        Note that the CaO in this is not 'lime' as we know it - it's all chemically bound up, the %'s we see are just a way of describing the system.

                        Fly ash was analysed by XRF and found to contain minimal heavy metal component, As was measured at 26.4 ppm, which divided into the mortar mix by 6 times (i.e. 1:6) gives approx 4.4 ppm, lower than would be found in many gardens. I've the rest of the analysis if anyone is interested.

                        Just got the insulation on the other day, by my partner in crime is away for a few weeks, so curing fires, and an assessment of performance, is still a month away at best.


                        • #13
                          That's really high in iron oxide. The castable I use has only 0.8% although that is a mix of CAC, aggregates and other goodies. It should not be a problem though at the temperatures we fire to. My understanding is that iron oxide operates as a flux on the silica content of the mix at higher temperatures >1100 C creating unwanted eutectic effect.

                          Make sure you keep the rain off the oven in the meantime.
                          Kindled with zeal and fired with passion.