The ultrastructure of bacterial granules that were maintained in an upflow anaerobic sludge bed and filter reactor was examined. The reactor was fed a sucrose medium, and it was operated at 35°C. Scanning and transmission electron microscopy revealed that the granular aggregates were three-layered structures. The exterior layer of the granule contained a very heterogeneous population that included rods, cocci, and filaments of various sizes. The middle layer consisted of a slightly less heterogeneous population than the exterior layer. A more ordered arrangement, made up predominantly of bacterial rods, was evident in this second layer. The third layer formed the internal core of the granules. It consisted of large numbers of Methanothrix-like cells. Large cavities, indicative of vigorous gas production, were evident in the third layer. On the basis of these ultrastructural results, a model that presents a possible explanation of granule development is offered.

Transgranular stress corrosion cracking in oil and gas linepipe was studied by testing full-scale pipes in soil environments. Three pipes made of Grade 359 (X-52) and Grade 414 (X-60) steels were used. Sixteen cracks were prepared on the external surface of each pipe, and crack growth was monitored and studied as a function of the pressure level and pressure fluctuation, the latter being described by the minimum to maximum stress ratio (R) and the loading rate.The overall results suggest strong mechanical effects on crack growth. Crack growth rates were found to be affected by both the pressure level and the level and rate of pressure fluctuation. For the Grade 414 pipe, reducing pressure fluctuation by increasing R from 0 . 6 to greater than 0.97 reduced the crack growth rates to below 5*1 O'9 mm/s, the detection limit of the crack depth measurement system used. Similar effects of pressure fluctuation were also seen for the Grade 359 pipes.Two main approaches are being adopted by researchers working on transgranular pipeline SCC in their efforts to simulate the cracking environment: (1) using "model" solutions such as NS4, a solution consisting of, per litre, 0.122 g KC1, 0.483 g N aH C 03, 0.093 g CaCI2 and 0.131 g M gS04-7H20 (Harle et al, 1995) and high purity water (Psaila-Dombrowski and Van Der Sluys, 1994) in testing pre-cracked compact tension specimens and (2) using actual soil environments in testing small-scale (Wilmott, et al, 1996) and full-scale pipes (Zheng et al, 1996). While the formulation of the NS4 solution was based on the field study conducted by TCPL, the use of pure water as the test medium was probably associated with the diluteness of the liquid found under the coating disbondment. The use of actual soil taken

Scanning electron microscopy revealed that collapsed extracellular polymeric substances (EPS) surrounded bacteria present in granular sludge. Treatment of granular sludge with whole-cell antiserum and staining with polycationic ferritin demonstrated that bacteria were enveloped by extensive EPS. Antibody stabilization permitted a visualization of the EPS which more closely resembled its natural hydrated state. The EPS was seen to completely fill the intercellular spaces in the microcolonies. Both pure and mixed microcolonies were observed to be enclosed by EPS. The presence of these large amounts of EPS indicates that this extracellular layer is important in maintaining the structural integrity of granular sludge.

No abstract