Modeling steel corrosion under supercritical CO2 conditions

Corrosion in the presence of CO₂ was studied under supercritical conditions (high pressure and moderate temperature) using different carbon steels and various corrosion resistant alloys (CRA's). An objective of this work was, among others, to put the results in perspective relative to various CO₂ corrosion models that have been developed and published over the years. In particular, the NORSOK, FREECORP, and OLI models were used for comparison, while others were discussed as well. The systems investigated were (a) supercritical CO₂ (SC CO₂) saturated with water (no separate water phase), (b) a water mist phase in equilibrium with SC CO₂, (c) a water/brine phase in equilibrium with SC CO₂. It was found that the OLI model best simulates the experimental corrosion rates observed in system (c) where coupons were corroded in the aqueous phase at high velocities in the rotating cage. The varying susceptibility of different carbon steels to corrosion under these conditions is highlighted. CRA steels are comparatively more resistant.     

Influence of chromium,molybdenum and cobalt on the corrosion behaviour of high carbon steels in dependence of heat treatment

High chromium martensitic steels are designed to provide high corrosion resistance in combination with high strength. Some of these steel grades contain primary carbides for improving the wear resistance, e.g. the steel 440C. The present paper mainly deals with the effect of chemical composition and microstructure on the corrosion properties. Different experimental alloys were produced in the shape of small ingots. The influence of the alloying elements chromium, molybdenum, cobalt, and carbon on the corrosion properties was studied. The results can be summarized as follows: Chromium and molybdenum improve the corrosion resistance, however, only the content of these elements in solid solution in the steel matrix is effective. In case of cobalt the corrosion resistance decreases. The reason is the interaction between cobalt and carbon and its effect on the chromium content in the steel matrix. The calculated pitting resistant equivalent number of high chromium martensitic steels is only limited valid, because there is a major effect of carbide precipitation on the corrosion behaviour. Further investigations were focused on the heat treatment. Especially the effect of the tempering temperature of these steels was studied. The tempering temperature is most relevant for secondary hardening carbide precipitation, which lowers the chromium content of the matrix with detrimental influence on the corrosion properties. The carbide precipitation and chromium distribution was characterized by means of energy filtered transmission electron microscopy (EFTEM).     

Evaluation of high‐temperature corrosion life using temperature gradient corrosion test with thermal cycle component in waste combustion environments

In energy conversion facilities such as boilers and pyrolysis reactors that burn waste, fossil fuel, biomass and other materials, many corrosion factors, such as gas temperature and fluctuations in gas temperature, the condition of deposits, and gas composition, influence the high‐temperature corrosion rate of materials in a complex manner. In order to evaluate the corrosion behavior and corrosion life time of materials rapidly and accurately, a temperature gradient corrosion test (TGT) was developed and applied to waste incineration environments. The TGT with thermal cycle was carried out without corrosive deposits in order to investigate the basic effect of gas temperature and fluctuation on corrosion behavior, especially in terms of the stability of protective oxide layers. As a result, it was clarified that gas temperature enhances the environmental factors such as the deposition rate of ash and penetration of corrosive species, while thermal cycle mainly enhances material factors such as the breakdown of protective scales. Furthermore, three application case studies of TGTs for corrosion life estimation were carried out with a comparison between TGT and field corrosion data. From these studies, a good correspondence of corrosion rates between TGTs and field tests was obtained. This correspondence of corrosion rate was explained by the breakdown of the protective oxide layer and the penetration of corrosive species in a corrosion front. The reproducibility and applicability of TGTs in the evaluation of the corrosion life time of materials were clarified based on the consideration of various corrosion mechanisms.     

Which are the right test conditions for the simulation of high temperature alkali corrosion in biomass combustion?

The use of renewable energy sources for the generation of electricity has gained much interest in recent years. Biomass has the potential of being a CO₂ neutral energy source that could provide a significant proportion of the ever‐increasing energy demand. The easiest and most straightforward utilization of the energy content of biomass is direct combustion. The thermal energy of the biomass released in the combustion process is utilized by heating water or steam to produce electricity or heat or both (CHP). Whereas the heat production requires only modest temperatures in the water or steam circuit, producing electricity with high efficiency is not possible without high steam parameters (temperature and pressure). The heat market being quite saturated, the economic potential is in the electricity market. High heat transfer surface temperatures coupled with biomass fired steam generators have resulted, however, in serious corrosion of the heat transfer surfaces, especially in the hottest section of the convection superheaters. The type of corrosion found in biomass boilers is not encountered in fossil fuel fired boilers and the mechanism causing it is not fully understood. The development of new alloys that could withstand these harsh environments would benefit tremendously if the test conditions in the laboratory tests could be chosen so that they adequately resemble the corrosion environment in real boilers. Currently the high corrosion rate is believed to be caused by gaseous KCl that condense on the heat transfer surfaces. While KCl is certainly found in the corroded superheater tubes and probably has an important role in the corrosion reactions with the alloy, the formation of KCl on the cooled surface can also be heterogeneous. In this paper a discussion on the effect of alkali hydroxides, especially KOH, is presented. Biomass fuels have normally a high content of alkali metals and a low content of sulfur and chloride. The excess alkali will produce alkali hydroxides in the combustion environment. Alkali hydroxides then react with CO₂ in the flue gases to form carbonates as the flue gases are cooled. The reaction with CO₂, is however, very temperature dependent. The equilibrium being completely on the K₂CO₃ side with a gas temperature below 700 °C and completely on KOH side with a gas temperature above 900 °C. The hottest superheaters are normally located in the area where the flue gas temperature is 850°C–1000 °C. This makes KOH condensation on the tubes possible and subsequent heterogeneous reactions with HCl, SO₂ and CO₂ in molten phase forming KCl, K₂SO₂ or K₂CO₃. Although KOH is not thermodynamically stable at typical tube surface temperatures, a continuous flux condensing from the flue gases results in a corrosion environment on the tube where its activity has to be taken into account. Therefore it is suggested that KOH, either in gaseous or molten phase should be included in the laboratory test environments used for the testing of alloys for biomass combustion applications.     

Laboratory investigation on the condensation and corrosion rates of top of line corrosion in carbon steel: a case study from pipeline transporting wet gas in elevated temperature

A glass cell was designed to simulate the condition for top of line corrosion encountered in wet gas transportation pipelines. Aqueous solution of 3 wt-% NaCl saturated with CO₂ at atmospheric pressure was employed. Effect of temperature gradient in the formation of condensation and its rate was investigated. API 5L Grade X65 carbon steel material was used as the working electrode for the experiment. The condensation rate was measured for the temperature gradient of 20°C, 30°C and 50°C under atmospheric condition for 24?h duration of the experiment. The corrosion rate of the specimens was measured using weight loss and Linear Polarisation Resistance (LPR) techniques. The LPR probe was immersed in the collected condensed water from the experiment to calculate the corrosion rate. The measured corrosion rate from the weight loss technique was in agreement with the corrosion rate measured from LPR. The corrosion rate measurement was repeated with the addition of 1000?ppm of pH modifying agent. This study indicates that the corrosion rate of pipeline and piping when subject to temperature gradient of 50°C and above is very high and alarming.     

High temperature corrosion mechanisms of certain new TiAl-based intermetallic alloys in an aggressive H2/H2O/H2S environment at 850 °C

High temperature corrosion behaviour of three TiAl-based intermetallic alloys - Ti-44Al-8Nb-1B, Ti-46Al-8Nb-1B and Ti-48Al-2Nb-2Cr-1B (at.%) - was studied in an environment of H₂/H₂S/H₂O yielding p_S2 ∼ 6.8 × 10⁻¹ Pa and p_O2 ∼ 1.2 × 10⁻¹⁵ Pa potentials at 850 °C. The kinetic results obtained by a discontinuous gravimetric method indicate that increase in Al and Nb concentrations led to enhanced high temperature corrosion resistance, the corrosion resistance decreasing in the order: Ti-46Al-8Nb-1B > Ti-44Al-8Nb-1B > Ti-48Al-2Nb-2Cr-1B. The scale development studies using SEM, TEM, EDX, WDS and XRD confirmed the formation of a multilayered scale on all materials. An outer layer consisting of TiO₂ existed beneath which an Al₂O₃ layer was present. Then a layer of TiO₂ formed again, below which an Al-enriched NbAl₃ was observed. A TiS layer was found beneath the NbAl₃ layer. The formation of TiS led to the development of a NbAl₃ band between the multilayered scale and the substrate.