Masters Theses

Gary L. Riner

Date of Award

12-1993

Degree Type

Thesis

Degree Name

Master of Science

Major

Metallurgical Engineering

Major Professor

Carl D. Lundin

Committee Members

Ray Buchanan, Eugene E. Stansbury

Abstract

Microbiologically influenced corrosion (MIC) has been identified as a serious problem in a number of industries including power generation, chemical, pulp and paper, gas transmission, and shipbuilding. The National Association of Corrosion Engineers (NACE) estimates that MIC contributes several billion dollars in costs to industry each year. The MIC phenomenon is complex and has been observed in sea water, brackish water, fresh water, petroleum products, and even in some deionized water systems. Materials such as carbon steels, low-alloy steels, stainless steels, copper-nickel alloys, and high nickel alloys have been shown to be susceptible to MIC. The purpose of this study was to determine the basic mechanisms by which MIC attacks austenitic stainless steel weldments (AISI types 304L and 316L) and to determine potential postweld treatments that may mitigate MIC. In order to study these effects, two field test sites were chosen; (1) the Tennessee River in Knoxville, Tennessee, where MIC has been shown to be a problem for the Tennessee Valley Authority (TVA) fossil and nuclear plants; and (2) the Guadalupe River in Victoria, Texas, where MIC of stainless steel weldments has occurred in a Dupont Chemical Facility. The methodology employed in this study consisted of utilizing test cells containing thirteen welded stainless steel coupons with various postweld treatments and conducting electrochemical measurements as a function of time to determine if and when MIC occurred. A 2- 1/2 inch diameter stainless steel pipe string consisting of welds with treatments similar to those in the test cells also was set up to evaluate MIC in an actual pipe weld configuration. Two sets of test cells were installed at the Tennessee River site, one in April 1992 and the second in August 1992. Both sets were disassembled in September of 1993 and none of the test coupons revealed any MIC. The test coupons in one of the four cells demonstrated low open-circuit potentials. These low potentials were contributed to a depletion of available oxygen on the surface of the test coupons. Visual observations of that cell revealed a thick coating of slime adhered to the coupon surfaces. Microbial analysis on the surfaces of the coupons revealed the presence of sulfate reducing bacteria (SRBs), anaerobic bacteria and aerobic bacteria. Microbial assays performed on samples from each of the cells identified a consortia of bacteria thought to be conducive to MIC. Possible reasons why no MIC was observed in the test cells as opposed to piping systems in the TVA plants are: (1) the cathode to anode area ratio on the coupon surface once it was covered with biofilm may have been insufficient to set up an oxygen-concentration corrosion cell; (2) the kinetics of MIC are not well established and the time for the corrosion process to occur may have been insufficient; (3) the water temperature of the Tennessee River varied throughout the year from 14 to 25° C, as opposed to water inside plant piping that may sit stagnant at temperatures in excess of 32°C for long periods of time thus enabling the bacteria to grow more rapidly and the corrosion process to occur more quickly; and (4) the probability of obtaining all the critical parameters necessary for MIC to occur on 13 linear inches of weld in a test cell is far less than that in a plant with thousands of linear feet of weld metal. If a certain oxide film, crevice, microstructure, flow rate, water chemistry, etc., is necessary to promote MIC, the probability of having the precise parameters occur coincidentally with a reduced size weldment is low. A more detailed discussion of these findings is found in the conclusion of this report. The Tennessee River pipe string assembly, installed in August of 1992, was also disassembled in September of 1993 and no MIC was observed. The pipe contained a layer of sediment (approximately one inch deep) which contained Asiatic clams and a wide consortia of bacteria. No explanation other than those mentioned above could contribute to the reason there was no MIC. Similar test coupons and pipe string have been field exposed on the Guadalupe River in Victoria Texas since March of 1993. To date, there have been no indications of MIC from the electrochemical measurements or visual inspections on either the coupons in the test cells or the pipe weldments. These test samples will be field exposed for one year or until MIC occurs.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS