Effects of Coke and Formulation Variables on Fracture of Bench Scale Prebaked Anodes for Aluminum Smelting
- PDF / 693,391 Bytes
- 6 Pages / 612 x 792 pts (letter) Page_size
- 80 Downloads / 240 Views
ing conditions, prebaked anodes are not prone to cracking. Only factors such as excess carbon consumption and electrical resistivity need to be considered in selecting raw materials and formulation. However, at some smelters anode cracking is a problem of considerable magnitude. The devastating effect of large quantities of cracked anodes can overshadow and contribute to high excess consumption and electrical resistance. Several types of anode cracking can occur due to thermal and mechanical stresses. Thermally induced cracks can result in spalling of corners from anodes shortly after being put into service. Vertical cracks emanating from the stub hole within a few days after the anode is put into service are caused by me chantcalor thermomechanical-induced stresses. This is caused by pressure exerted by the expanding stub and possibly augmented by differential thermal expansion between the hotter anode bottom and cooler top. A similar crack sometimes occurs during rodding. Although cracking problems can at times be alleviated by engineering changes, such as redesign of stub holes, or changes in anode setting procedures, such modifications are not always economically or technically feasible. Improvement of the intrinsic resistance of the anode carbon to cracking is often the most promising avenue. Brown and Rhedey" discussed this subject in some detail but dealt With only a limited number of cokes and a single aggregate sizing. An earlier paper" by Harvey and Van Dyne was limited to one coke. GENERAL Work was carried out on a variety of calcined delayed cokes including petroleum cokes having a wide range in bulk densities, gilsonite coke, and coal tar pitch cokes. Some coke properties are given in Table 1. Since coal tar pitch remains the dominant binder used in prebaked anode manufacture, binder was limited to coal tar pitch (110°C softening _point, 12 pct
quinoline insoluble). Aggregate sizings and binder levels used in each set of experiments are described below. Anode fabrication procedure involved the blending of two-kilogram samples of aggregate with pitch for 30 min in a 3.8 liter (1 gal) sigma-blade mixer heated at 140°C. Green specimens, 50.8 mm (2 in.) in diam and approximately 150 mm (6 in.) long, were formed in a mold preheated to 140°C by application of 27.6 MPa (4000 psi) pressure. Green specimens were packed in calcined coke and baked under a nitrogen purge at an upheat rate of 25°C/h to 1135°C and held at that temperature for ten hours. Anode green and baked apparent densities and volume change during baking were measured. Since the green apparent densities were not particularly informative, they were not included in this report. To determine resistance to mechanical stress, flexural strength was measured using a room temperature, four-point loading test (ASTM C78- C4). Most values reported are averages of four determinations. Although stresses in various cases may in fact be tensile, compressive or shear, it is believed that differences among various carbons would be in about the same relative ord
Data Loading...
