Reinforcement corrosion in cement- and alternatively-stabilised rammed earth materials

Abstract

Developing low-embodied energy building materials can significantly contribute to reducing global greenhouse gas emissions. However, these new building materials must be properly understood before they can or will be adopted by the construction industry. In this paper, we examine the corrosion of reinforcement embedded within rammed earth materials in both atmospheric and accelerated carbonation conditions. Six rammed earth materials with alternative stabilisers to cement were studied – three stabilisers combined with a substrate of either crushed limestone or recycled brick and concrete – and contrasted against cement-stabilised crushed limestone. The stabilisers included industrial by-products fly ash, ground granulated blast furnace slag and silica fume, combined with sodium hydroxide or hydrated lime as activators, or with no activator, to form alkali-, lime- and self-activated rammed earth, respectively. Likelihood of reinforcement corrosion was assessed via changes in reinforcement half-cell potential and matrix carbonation.Half-cell potential results for four cement-stabilised rammed earth mixes that underwent accelerated carbonation showed that while uncarbonated cement-stabilised rammed earth protected steel from corrosion, once carbonated, the electrochemical behaviour was distinctly different and reinforcement was found to corrode. Self-activated rammed earth carbonated fully within 29 weeks. Carbonation rates of alkali- and lime-activated rammed earth were sufficiently slow that the carbonation front would not reach reinforcement with a typical cover depth of 150 mm over a typical 50 year design life. However, the carbonation front will likely breach a cover of 50 mm in that time. Electrochemical half-cell potential measurements of alkali-activated rammed earth were similar for both accelerated carbonation specimens and those only exposed to atmospheric conditions.Alternative rammed earth materials equilibrated to atmospheric moisture and CO2 levels gave half-cell potential readings above -200 mV SCE: a value previously shown to be indicative of a negligible corrosion rate in cement-stabilised RE. However, destructive testing revealed that reinforcement embedded in all specimens was corroded, i.e. this value is not indicative of negligible corrosion in these materials. Results therefore indicate that mild steel should not be used in RE materials not containing cement. Rather, galvanised, stainless steel or fibre reinforced polymer reinforcements should be used to avoid reduction in service life due to corrosion.