Abstract:
At the present time, combined cycle systems for power generation (e. g.
IGCC), offer increased efficiency of power generation and lower
environmental emissions, specifically C02, SOxg, and NO,,, as well as being
adaptable to most fossil fuels. Economic factors, such as the cost of the
materials must be considered. Materials influence the service lifetime in the
required operational environment. Solid fuels like coal and biomass produce
different combustion environments containing a range of contaminants that,
when they reach their melting points, may cause accelerated corrosion,
affecting directly the service life time of the gas turbine constructional
materials. This accelerated corrosion is known as Hot Corrosion.
The aim of this study was to develop, an understanding of the influence
of these environmental factors on rate of hot corrosion of modem turbine
materials, i. e. the single crystal alloys CMSX4 the SC2
,
both uncoated and
PtAl coated that are needed for a gas turbine blade and vanes operating in a
range of hot corrosion environments expected in an lGCC plant.
To achieve this aim, a series of laboratory corrosion tests was planned
to simulate the same corrosion environment as in industrial high temperature
gas turbine operation. Following established procedures for corrosion testing,
samples were exposed in a controlled atmosphere furnace to a mix of gases
(air/SO241CI) with a cyclic exposure time of 50 and/or 100h duration. Each
cycle, samples were removed to be recoated with an alkali salt mixture to a
total exposure time of 500h and or 1000h. Cross sections were examined by
SEM/EDX to identify the mode of hot corrosion attack. To quantify the rate of
corrosion, samples were measured pre-exposure and post-exposure, and this
corrosion data was statistically assessed.
Finally, from this quantitative data, life prediction models were
developed to describe/predict the onset of hot corrosion and the corrosion rates
observed under different gas compositions, and various deposition fluxes, both
at typical type I and type II hot corrosion temperatures in terms of incubation
and propagation periods. Separate models have been developed for the two
single crystals superalloys: CMSX4 and SC2, in both the uncoated and
platinum aluminide coated condition. The goodness of fit as defined by the
regression coefficient varies from 0.88 to 0.99 for the propagation models at
700 and 900'C. The incubation models are as precise at 7001C but less precise
at 9001C with regression coefficients of 0.78-0.94.
I
