Journal of The Americanlining are exposed to temperatures and atmospheres sufficient to cause formation of a dense layer during the latter portion of the blow and during turndown for sampling and tapping. Either dissolution of a previously formed layer or of MgO condensed from the vapor phase into the unsaturated liquid slag would account for the apparent lack of a dense layer in the reaction zone of brick removed from the BOF. However, the point to bear in mind is the overall similarity in microstructural features of laboratory and service specimens.A review of all information on the relative wear rates of burned impregnated and pitch-bonded brick indicates that the most significant difference that would account for the higher wear rate of the pitch-bonded brick in high-wear areas is the very porous weak zone at the refractory-slag interface. This zone results from internal oxidation-reduction reactions and occurs independent of formation of a dense layer. If the dense layer forms. it bridges coarse particles; the strength of the interface depends on the degree to which the coarse particles are anchored in the carbon-bonded matrix.When the dense layer can form, its retention in service for burned and pitch-bonded brick is contingent on MgO saturation of the slag. In this respect, process parameters of the operation should avoid large volumes of low-lime/silica slags. VII. SummaryThe laboratory data and theoretical analysis of the present study show that internal oxidation-reduction reactions will occur in all types of carbon-bearing BOF refractories under steelmaking conditions. The rate of these reactions increases with increasing temperature and decreases with increasing pMg and/or pC0 at constant temperature. Above 2700°F in an oxidizing atmosphere, a dense MgO layer forms and slows the rate of internal reactions. Subsequent reaction appears to be limited by diffusion across the dense layer. This layer dissolves quickly in low-lime/silica slags undersaturated with Ceramic Society-Bell et al. Vol. 55, No. 1 MgO but is quite impervious to high-lime/silica slags easily saturated with MgO. Porosity and low-lime/silica material accumulate in the decarburized region behind the hot face independent of formation of a dense layer. The alteration in the distribution of porosity and impurities must result in a decrease in high-temperature strength of the affected area adjacent the hot face. This phenomenon would account for the faster rate of wear of pitch-bonded brick in high-wear areas of the BOF. , "Carbon-ME0 Reactions in BOF Refractories," Amer. Ceram. SOC. Bull., 50 [7] B. Brezny and R. A. Landy, "Effects of Heat and Silicate

No abstract

No abstract

HE present work is a continuation of high-temperature T creep recovery experiments made on rutile single crystals oriented for ( 1 TO) [OOl] def~rmation.'-~ Creep specimens were prepared and tested a s reported previously.' Experiments were conducted in air' and vacuum.In air, creep recovery was studied at IOOO", 1020°, and 1040°C with reduced stresses, up, of 500, 2500, 5000, and 7500 psi. The original creep stress, u~, was 10,000 psi. The variation of recovery index,' n, with recovery time at 1020°C for each reduced-stress level investigated is shown in Fig. 1. For a given relatively shont recovery time, a higher degree of recovery of creep-resistant substructure is obtained at higher reduced-stress levels. For longer recovery times, however, the maximum attainable value of n (n,nc,s) decreases as the reduced-stress level increases (Fig. 2, curve ( A ) ) . No noticeable metallographic changes (dislocation density, dislocation wall spacing) were observed in the plateau region of the reduced-stress treatment, in which n did not vary with recovery time.The activation energy for the recovery of creepresistant substructure decreases as the reduced-stress level increases (Fig. 2 , curve ( B ) ) and strongly suggests that the recovery mechanism is stress-assisted as well as thermally activated. Metallographic examination of etchdpit configurations revealed that dislocation walls became more distinct, the separation of walls increased, and the densities of dislocations between the walls decreased during the portion of the reduced-stress treatment in which the recovery index increased with time. The process primarily responsible for recovery during this treatment is the migration of subboundaries of dislocation walls through the crystal, sweeping up barriers to dislocation motion in their path.In vacuum, creep recovery experiments were conducted under lo-' torr a t lOOO", 1020°, and 1040°C. Compressive specimens' were crept under a stress of 0,,=3000 psi to a strain tc=0.051 early in the secondary stage of creep and then allowed to recover for varying times under a reduced stress of ur=300 psi. Recovery was detected by the increased creep strain which occurred on reapplication of u~. The recovery index,' n, is plotted vs time of recovery, t,, on log-log coordinates in Fig. 3. The mean activation energy for recovery of creep-resistant substructure of vacuum-reduced rutile is 119,-000 cal/mol, whereas that for recovery of creepresistant substructure in air' is 135,000 cal/mol. An error of 5% was estimated for both activation energies for recovery. A stress of 10,000 psi was required to reach a creep strain' of 0.046 for a particular temperature-compensated time, Oc ( 8 , = t exp ( -H,/RT), where H, is the activation energy for creep). In vacuum about the same creep strain was reached under a stress of 3000 psi for the same temperature-compensated time. Creep of Rutile at High Temperatures," ibid., 54 [7] 359-60 (1971). ' W. M. Nirthe and J. 0. Brittain, "High-Temperature Steady-State Creep in Rutile," ibid., 46 [9] 411-1...

No abstract

N. Y. ' W. M. Hirthe and J. 0. Brittain, "Dislocations in Rutile as Revealed by the &&Pit Technique," J . Amere Ceram. sot., 45 [111 54h54 (1962).

The recovery of creep resistant substructure in CoO single crystals was studied at temperatures of 1000, 1050, and 1100 ~ C in air. Compressive creep specimens were crept under a stress of 13.8 MN m -2. to a strain early in the secondary stage of creep, then allowed to recover for varying periods of time under a reduced stress of 0.69 MN m -2 . Recovery was detected by increased amounts of creep strain which occurred upon reapplication of the 13.8 MN m -2 stress. An apparent activation energy of 71.7 -+ 10 kcal mol -a was obtained for the recovery process. Experimental evidence suggests that the primary recovery mechanism involves the climb of dislocations within subgrain boundaries.

No abstract