| Abstract |
When pure Portland oil well cements are heatedhydrothermally, the calcium silicate hydrate bindercrystallises. The phases formed depend on several factorswhich include temperature and Ca/Si ratio. Without addedsilica, alpha di-calcium silicate hydrate, áC2SH, forms alongwith Ca(OH)2 between 120 and 200°C, while above 200°C,phases such as reinhardbraunsite, C5S2H, and jaffeite,C4S3H3, form together with Ca(OH)2. These high Ca/Siratio phases are weak, porous binders which undergo“strength retrogression” where strength decreases with time.In the 1950’s, Kalousek showed that by adding ~40% silicaflour (quartz) by weight of cement (BWOC), high strength,low Ca/Si ratio phases formed, tobermorite around 150°Cand xonotlite above 200°C. These phases avoid strengthretrogression and high additions of silica flour to oil wellcements have formed the basis of all geothermal cementformulations used in recent years. Research into using otherforms of silica to replace quartz has not been successful informing tobermorite.However, both tobermorite and xonotlite carbonate veryreadily in the CO2 rich fluids of New Zealand geothermalfields and become porous. The carbonated layer readilydissolves if the concentrations of CO2 in the surroundingfluid make it mildly acidic so the steel casing is no longerprotected and will corrode. In contrast, carbonation of theweak high Ca/Si ratio phases forms an impermeablecarbonation sheath around the samples slowing the rate ofcarbonation. By adding only 20% silica flour, the Ca(OH)2is removed and although the final strength of the áC2SHbinder formed is lower than that of tobermorite, the ongoingloss in strength is reduced, and the binder hasreasonable carbonation resistance.Recently, a natural hydrothermal amorphous silica fromNew Zealand, Microsilica 600 (MS600), has been used incement formulations that have shown marked advantages incarbonation resistance. Unlike with crystalline quartz, wehave found that additions of MS600 of around 20% BWOC,give an amorphous calcium silicate hydrate which does notcrystallise at hydrothermal temperatures. This binder hasgood strength that does not decrease over 6 months, evenwhen heated to 300°C. Laboratory autoclave experiments inCO2 saturated water (6 bar) show carbonation is limited to athin sheath which increases slowly with time. Theformation of this sheath can be improved by addition of asmall amount of Ca(OH)2 to the formulation. It appears thatMS600 reacts rapidly with Ca(OH)2 before the hydratingcement can carbonate and the resulting product does notcrystallise, so strength retrogression does not occur. |