Abstract
The corrosion of AISI Type 316L stainless steel (316L SS) liner plates in the flash chambers of a multistage flash (MSF) desalination plant, located on the Arabian Gulf coast was investigated. The 316L SS liner plates developed severe corrosion within six years of operation. This study was conducted to develop an understanding of the mode and causes of corrosion of the liner plates, and to determine the effect of heat treatment (annealing or heat effect during welding) and temperature of salt solution on corrosion of the liner plates. Specimens of the liner plates were studied in as-received (AR) condition and after being heat treated (HT) at 900°C in air and air-cooled to room temperature. Electrochemical techniques were used to measure the corrosion of the specimens. Scanning electron microscope (SEM) installed with energy dispersive (ED) X-ray diffraction capability was used for identification of compositional and structural changes in the specimens during heat treatment and corrosion. The results showed that:(1) Commercial grade 316L SS is susceptible to pitting, crevice and grain boundary corrosion under the operating conditions in the desalination plant. The heat-affected-zone (HAZ) had larger grains and corroded more severely than other parts of the liner plates. (2) The liner plates had randomly distributed inclusions containing Ti, Cr, Mo, Mn, and S in the structure. (3) Measurement of the corrosion rate, and (4) Metallographic investigation of the AR and HT samples.
The test samples were machined from 2 mm thick 316L SS liner plates, in circular shape with 16 mm diameter. The samples were prepared from weld zone including HAZ and normal liner plates. Some of the samples, both from HAZ and plate, were studied in as-received condition. The remaining of the liner plate samples were investigated after a heat treatment to simulate welding conditions in the SS samples. Table 1 shows elemental analysis of 316L SS samples used in this investigation. Five samples were prepared for metallographic examination:
Two samples, one from 316L SS plate and one from HAZ, were studied in as-received condition.
One polished cross-section at HAZ was prepared by mounting the sample in transparent epoxy and polishing to 0.05 microns surface finish. It was studied for penetration depth of intergranular corrosion.
One of the AR plate samples was mounted in transparent epoxy, and polished to 0.05 microns surface finish, and etched with modified Marble Reagent (ie.20g CuSO4 + 50ml H2SO4 + 100ml HCl and 100ml water) for austenitic stainless steels.
One of the HT samples, annealed at 900°C for 20 h and cooled in air, was investigated after polarization scan in 5% NaCl solution at 60°C.
All of the specimens (AR, HAZ and HT) used for electrochemical testing were polished with SiC paper to 600 grit finish. They were then degreased with acetone, washed with distilled water, and dried in air.
The heat treatment was designed to simulate, as closely as possible, the conditions that HAZ is subjected during welding. The 316L SS plate samples were first solution annealed at 900°C for 20h or 120h in air in a laboratory furnace to ensure dissolution of carbide precipitates, if any, and cause appreciable grain growth. Following the solution annealing the samples were cooled in air to simulate the cooling rate during welding and induce sensitization at grain boundaries. The samples were then polished to 0.5 micron surface finish prior to electrochemical tests.
(3) The corrosion rate of the AR samples of the liner plates was the lowest. It was followed by corrosion rates of HT and HAZ samples respectively in ascending order.
