Open Collections

Fouling of heated stainless steel tubes with ferric oxide from flowing water suspensions

University of British Columbia

The fouling behaviour of ferric oxide (hematite) particles suspended in water flowing through 0.343 inch i.d. type 304 stainless steel tubes was experimentally investigated. Independent variables studied, using micron and submicron size particles, were ferric oxide concentration (15 - 3750 ppm), tube Reynolds No. (10090 - 37590) and heat flux (0 - 92460 BTU/ft²-hr). For selected runs, fouled tubes were sectioned and the fouling deposit subjected to "in situ" chemical analysis by means of an electron microprobe. During the fouling process, measurements were made of local and average thermal resistance as a function of time. The resulting fouling curves fell into three distinct categories, depending on the particle concentration and the mode of operation: (I) At ferric oxide concentrations below 100 ppm, no thermal fouling could be detected over experimental periods of up to 14 days. Microprobe examination of such tubes showed spotty deposits. (2) At ferric oxide concentrations of. 750 ppm and higher, using mixed size particles, measurable thermal fouling occurred at a steadily decreasing rate, similar to the asymptotic type behaviour reported previously in other fouling systems. In the present study, the asymptotic condition was achieved after about four hours of operation. Prolonged operation resulted in a sudden decrease in fouling resistance at localized positions on the test section, followed by refouling of the whole test section. (3) If the suspension was circulated through the test section at zero heat flux for approximately eight hours and then heating started, the tube commenced fouling thermally at a constant rate considerably greater than the previous decreasing rates. Microprobe results showed the deposits to contain, in addition to iron and oxygen, significant amounts of nickel and chromium. Chemical composition profiles typically showed nickel and chromium concentration gradients from the wall inwards, concentrations varying from the highest values at the wall to zero at the deposit-fluid interface. A test section used for a series of fouling trials, when examined under an electron microscope, was found to contain small but distinct pits. A hypothesis is presented according to which the fouling behaviour of water suspended ferric oxide on stainless steel is controlled by the rate at which crevice corrosion of the stainless steel occurs. The corrosion products precipitate within the initially loose deposit structure and thus serve to stabilize this structure. The corrosion rate is in turn controlled by the oxygen reduction rate at unfouled areas on the tube wall. Experiments specifically designed to test this hypothesis, such as increasing the unfouled area in an attempt to accelerate the corrosion rate, and removing oxygen with a scavenger in order to decrease the rate, gave results entirely consistent with the hypothesis. Mathematical models based on the hypothesis are explored.

Text

2011-03-04

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/32039

Chemical and Biological Engineering

University of British Columbia

Graduate