Corrosion testing chemical plant

Most chemical plant processes are complex. Materials of construction for process equipment and piping must be selected with care to assure corrosion-free operation of those plants. In a well designed plant, materiids selection is based on a number of factors prior service history, field in-plant corrosion tests, pilot plant corrosion tests and laboratory corrosion tests, in that relative order of usefiihiess. often, actual service history regarding materials performance is not available because plants kept poor records, plant personnel were not trained to evaluate corrosion failures, or, in the case of new processes, because there is no service history. The same can be said for field in-plant corrosion tests. Pilot plant testing is usually the best alternative in such cases, but such facilities are typically operated only for short runs and are not well suited for gathering long-term corrosion performance data. 

Corrosion testing data. The pitfalls in corrosion testing and the test methods are described in Chapter 19, but several points need underlining from experience in the design of chemical plants ... 

Butt, L. T. and Wright, D. C. (1980) Use of Polymers in Chemical Plant Construction (Applied Science). Champion, F. A. (1967) Corrosion Testing Procedures 3rd edn (Chapman Hall). 

Because sulfuric acid and halogen are very corrosive, selection of the structural materials is an important issue. Screening tests have been carried out using test pieces of commercially available materials at GA [29], JAEA [30,31], etc. As for the gas-phase environment of the H2S04 decomposition step, some refractory alloys that have been used in conventional chemical plants showed good corrosion resistance. Figure 4.13 shows one of the experimental results of Alloys 800 and —600 obtained under gas-phase sulfuric acid decomposition environments at 850°C. Gas compositions in the upstream and downstream... 

Indications from both microscopic analyses of metallic particles from corrosion tests and evidence from the Oklo natural reactors indicate that performance assessment calculations should not assume 99Tc is easily mobilized. It is entirely inappropriate to use "Tc release as a marker for fuel corrosion because Tc is not located in the fuel matrix. The TEM examinations of corroded e-particles have shown that Mo is preferentially leached from these phases, a behaviour that is similar to the one observed at Oklo. It is interesting to note that laboratory dissolution of billion-year old 4d-metallic particles for a chemical analysis required a corrosive mix of peroxide and acid (Hidaka Holliger 1998) similar to the experience at SNF reprocessing plants. It is doubtful that the oxidation potential at the surface of an aged fuel will be sufficient to move Tc(0) from the e-metal particles. 

Corrosion rates are expressed in terms of inches per year of surface wastage and are used to provide a corrosion allowance in the design thickness of equipment such as vessels and pipes. Operators will often use data based on historical experience from plant operations to aid them in determining appropriate corrosion allowances. Alternatively, corrosion charts are widely available that give corrosion rates for many combinations of materials of construction and process fluids, and normally a range of values will be provided for various process temperatures. In some instances, particularly where there is a mixture of chemicals present, appropriate data may not exist and corrosion tests may be necessary in order to determine the suitability of equipment. Operators should be able to demonstrate the use of corrosion allowances in equipment specification and design. The sources of data used should be traceable. 

In a chemical plant nine aluminum alloys were tested for their resistance to corrosion. Four locations were chosen inside the chemical plant for this experiment and at each such place one plate of all nine alloys was placed. Exposure to chemical corrosion lasted one year. After the experiment four researchers examined the plates randomly and gave a mark from 1 to 10 to each plate depending on the observed resistance to corrosion. The experiment was aimed at asserting which of the offered aluminum alloys had the best resistance to corrosion at one or at all locations of the chemical plant. It was also interesting to see how much the researchers agreed or disagreed in their estimates of resistance to corrosion. Thus, the experiment had included three factors nine aluminum alloys, four locations and four researchers. Experimental results are shown in the following table ... 

PPG Lake Charles was one of the first chemical plants to openly discuss problems that users experience by improperly specifying and testing safety relief valves. Expertise on new safety valves could be obtained from manufacturers sizing methods were documented by various designers and users, but no one published practical SRV testing and SRV corrosion problems. 

Piping systems may be relatively simple or complex and may convey utility services and chemical fluids that can be highly flammable, highly toxic, corrosive, erosive, or any combination of these properties. Piping systems within chemical plants handling hazardous Iluids must be designed, fabricated, tested, operated, and maintained consid-... 

Pilot plant tests and laboratory corrosion tests under simulated plant conditions will help in the selection of suitable materials if actual plant experience is not available. Preliminary tests can be carried out by inserting coupons of different materials into an apparatus that is known to resist corrosion before testing plant components. This reduces the likelihood of component failure and possible release of chemicals during testing. Care is needed in the interpretation of laboratory tests. 

Krystow, P. E., Evaluating Materials of Construction in Pilot Plant Corrosion Tests, in Scaleup of Chemical Processes. Conversion from Laboratory Scale Tests to Successful Commercial Design Bissio, A. Kabel, R. L Eds. Wiley New York, 1985 pp. 620-654. 

The materials of construction of a chemical plant must be resistant to their contents over a long period of time and testing of samples of the construction materials in the reaction media will be carried out to ensure that corrosion does not lead to escape of reaction materials or failure of plant components. This work is usually carried out in collaboration between development chemists and chemical engineers. 

Evaluation and Testing. There is no laboratory test available to predict corrosion protection performance of a new coating system. Suppliers and end users of coatings for such applications as bridges, ships, chemical plants, and automobiles have collected data correlating performance of different systems over many years. These data provide a basis for selection of current coatings... 

---