J. Sumner

Fireside issues in advanced power generation systems

Abstract The requirements to supply increasing quantities of electricity and simultaneously to reduce the environmental impact of its production are currently major issues for the power generation industry. Routes to meeting these challenges include the development and use of power plants with ever increasing efficiencies coupled with the use of both a wider range of fuels and technologies designed to minimise CO2 emissions. For fireside hot gas path components, issues of concern include deposition, erosion and corrosion in novel operating environments and increased operating temperatures. The novel operating environments will be produced both by the use of new fuel mixes and by the development of more complex gas pathways (e.g. in various oxyfired or gasification systems). Higher rates of deposition could significantly reduce heat transfer and increase the need for component cleaning. However, degradation of component surfaces has the potential to be life limiting, and so such effects need to be minimised. Materials and operational issues related to these objectives are reviewed.

Assessment of the performance of scanning capacitance microscopy for n-type gallium nitride

Six GaN-based test structures were grown on sapphire to assess the effectiveness of scanning capacitance microscopy (SCM) in measuring n-type carrier concentrations in GaN. These structures were designed to test the carrier concentration and spatial detection limits of the technique using data obtained from secondary ion mass spectroscopy for comparison. It has been found that, where nonintentionally doped (nid) spacers are present between doped layers, SCM reliably and repeatably distinguished between different carrier concentrations for levels ⩾2×1017cm−3 and detected layers with thicknesses ⩾25nm. Where no nid spacers exist in samples, the spatial resolution improves to 10nm. However, the exact quantitative response of SCM is seen to be highly dependent on the particular probe used, necessitating the use of calibration standards for the quantification of unknown samples. As an example of the application of such a calibration standard, we attempt to quantify the carrier concentration in the conductive l...

Scanning capacitance microscopy studies of unintentional doping in epitaxial lateral overgrowth GaN

Four samples were grown by epitaxial lateral overgrowth (ELOG) using different magnesium precursor fluxes during the coalescence stage. These samples were studied in cross section using scanning capacitance microscopy and scanning electron microscopy. The resulting images revealed the existence of several differently doped regions in the ELOG structures including n-type doping in the GaN initially grown through the ELOG mask’s windows, and undoped GaN after coalescence had been completed. In addition, samples for which a magnesium precursor flux had been present during the coalescence of the GaN stripes also exhibited p-type doping. From an analysis of the spatial distribution of the unintentional doping in the ELOG material, it has been possible to propose that the incorporation of the n-type dopant was slower on the (0001) facet than the {112¯2} facets. This facet dependent difference in incorporation rates also helps to explain the nonuniformity in thickness of n-type conductive layers seen at the GaN/...

Performance of thermal barrier coatings in industrial gas turbine conditions

Abstract The effect of aerofoil geometry on the oxidative degradation mechanisms experienced by thermal barrier coatings (TBCs) used on industrial turbine blades has been investigated. Modified aerofoil-shaped samples (CMSX4 coated with high-velocity oxy-fuel sprayed AMDRY 995 and air plasma sprayed TBC) were oxidised at five temperatures in furnaces from 900 to 1000°C. Scanning electron microscopy and energy dispersive X-ray analysis were used to characterise details of the microstructural evolution of the thermally grown oxide and to monitor inter-diffusion between the bond coating and substrate. Additionally, a novel non-destructive examination technique (flash thermography) was used to detect and track the spread of cracks beneath the TBCs. Multiple samples cracking in identical locations suggested an effect of geometry in the failure of coatings. Furthermore, it was observed that coating curvature influenced spinel formation.

Morphological changes of InGaN epilayers during annealing assessed by spectral analysis of atomic force microscopy images

During annealing, the morphologies of thin InGaN epilayers have been observed to change from a terraced structure to a network of interlinking InGaN strips separated by troughs. This change in morphology may contribute to high efficiencies in some GaN-based light emitting diodes (LEDs) if the InGaN is exposed to elevated temperatures without a protective GaN capping layer. Here, we investigate the changes in morphology which occur when InGaN epilayers are annealed at their growth temperature under NH3, N2, and a small H2 flux. We observe that while the layers initially roughen, more extended anneals lead to the surface becoming smooth and terraced once again. Power spectral density analysis of atomic force microscopy data is used to show that the dominant mechanism for roughening is loss of material from pre-existing pits, while the dominant smoothening mechanism is surface diffusion. This mechanistic analysis may be relevant to the growth of InGaN quantum wells in LED structures.

Hot corrosion resistance of gas turbine materials in combusted syngas environments

Abstract To reduce CO2 emissions, there is interest in a new generation of industrial gas turbines operating in advanced integrated gasification combined cycles (IGCC) with the option for pre-combustion CO2 removal systems. These gas turbines may be fired on syngas, cleaned H2 rich syngas, or natural gas, and this will alter the hot corrosion experienced by components in the hot gas path. Deposit recoat laboratory tests have looked at the response of 11 state-of-the-art materials systems under the expected IGCC conditions. MTData has been used to calculate the optimal deposits under the temperature and gas conditions for these tests. Studies of metal loss have helped assess quantitatively the resistance of the different materials systems, while microscopy techniques have given information on the degradation mechanisms experienced. Rene 80 has been selected to demonstrate the data from these tests and shows the most significant metal loss under partially cleaned syngas conditions.

High temperature oxidation and corrosion of gas turbine materials in burner rig exposures

Abstract With the introduction of novel fuels, which may contain high levels of trace impurities including sulphur and alkali metals, industrial gas turbines are operating with increasingly corrosive combustion environments. To investigate the effect that this, coupled with the higher combustion gas temperatures needed to increase power plant efficiency, has on current state-of-the-art gas turbine component materials, three burner rig exposure tests have been run. The tests evaluated the effects of fly ash, gas moisture and gas temperatures on the alkali sulphate induced hot corrosion of CM247LC, Haynes 230, IN939 and IN728LC. Type I (sulphidation and internal damage), type II (pitting) and mixed mode hot corrosion were observed under different test conditions; however, the presence of fly ash appeared to reduce the levels of hot corrosion. CM247LC, with its high Al content improving oxidation resistance, showed less resistance to hot corrosion than the other, higher Cr content, alloys.

High temperature oxidation and corrosion of gas turbine component materials in burner rig exposures

Abstract To meet environmental, legislative and commercial targets, gas turbines must operate with increasingly high gas temperatures and fuels with increased contaminant levels. A series of three burner rig tests have been used to evaluate the effects of fly ash, gas moisture and gas temperatures on alkali metal induced hot corrosion in the metal temperature range of 700 – 960°C on three uncoated materials (Haynes 230, IN939, and IN738LC) and one coated system (IN738LC/HVOF SV21). It has been found that the specific burner rig test conditions impact upon the severity of samples’ corrosion and oxidation, with samples exposed to impacting fly ash demonstrating reduced hot corrosion. However, type II hot corrosion (pitting) and type I hot corrosion (internal damage and sulfidation) have been observed under all test conditions. Generally IN939 was more resistant to hot corrosion over both high and low temperatures than IN738LC or Haynes 230. SV21-coatings on IN738LC provided improved resistance to both type I and II hot corrosion.

Impact of Deposit Recoat Cycle Length on Hot Corrosion of CMSX-4

Hot corrosion causes significant problems for both aerospace and power generation industries, where the combination of high temperature, corrosive gases, and contaminants severely limits component operating lifetimes in gas turbine hot gas streams. Multiple laboratory testing methodologies exist to study this hot corrosion, and these can be affected by a range of variables. This paper investigated the impact of varying deposit recoat cycle length when using the ‘deposit recoat’ testing method. CMSX-4 samples were exposed to simulated type II (pitting) hot corrosion conditions, with the same overall deposit load (averaged across the total exposure run), but different deposit recoat cycles. Post-exposure, samples underwent dimensional metrology analysis to compare metal loss resulting from different deposit recoat cycle lengths. Results for CMSX-4 suggest very small differences in corrosion losses, indicating CMSX-4 hot corrosion datasets obtained from deposit recoat experiments with different deposit recoat cycle lengths can be compared with confidence.

Cautionary note on use of focused ion beam sectioning as technique for characterising oxidation damage in Ni based superalloys

Abstract Previous observations on Ni based superalloys, obtained through the use of focused ion beam (FIB) sample preparation and imaging, have reported the presence of subsurface voids after oxidation. In this present study, oxidised specimens of the Ni based superalloy, RR1000, were subjected to conventional sample preparation as well as both dual and single beam FIB preparation, with the aim of re-examining the previous observations of subsurface void formation. It is clear from FIB preparations that features previously interpreted as networks of voids have been demonstrated to be internal oxides by varying the sample tilt angles and imaging signal using either secondary electrons (SEs) or secondary ions (SIs). Conventional preparation methods illustrate the presence of subsurface alumina intrusions and the absence of voids, supporting previous evidence. The positive identification of voids and oxides by FIB can be complex and prone to misinterpretation and thus, the use of several imaging conditions and tilt angles must be used, along with conventional preparation methods, to confirm or refute the presence of ‘voids’ underneath oxides.