~~

~~

The distribution and types of non-metallic inclusions in rimming steel ingotsobtained by four different teeming practices have been investigated microscopically.The different teeming practices were: (1) ordinary, (2) slag covering, (3) coveredtop of ingot mould during teeming, and (4) same as in (3), but immediately afterthe completion of teeming, a heavy steel plate was placed on top of the moltenmetal. Quantitative results indicate that the types and distribution of non-metallicinclusions in these ingots are basically similar, except that oxide networks alonggrain boundaries were observed in ingots obtained by using the last three teemingpractices. Several interesting phenomena are revealed and discussed. Suggestionsto improve the teeming practice are given.

By means of a simple wire-wound furnace, it was possible to purify aluminumand tin by the zone-melting method. The principle of this technique was givenand the mechanical properties of aluminum and tin before and after successivetreatments were determined. It was shown that with increasing purity, the yieldvalue as well as the cocfficient of strain hardening decreases in the tensila ex-periment. During a micro-creep test at a constant load, the amount of initialcreep and the subsequent creep rate increase with increasing purity. Effects ofpurity on the characteristics of recrystallization were also noted.

Industrially pure copper was subjected to ordinary tensile test. Some extraor-dinary type of fracture was found. Several specimens show the same results thatthe fracture can not he classified as "cup and cone" but appears as spirals asshown in Fig 6 and 8 in the chinese text. The fracture starts at center of thespecimen with a small irregular contour having scattered local cones, and thenseems to propagate by tearing off a wedge-shape-sectioned volume of materialalong a plane spiral until the surface is reached. The angle of the wedge neverappears anywhere close to π/2, this implied that the two wedge surfaces are notthe maximum principle shearing planes, therefore this type of fracture can notbe interpreted on the basis of the maximum shear stress criterion.

The pole-figures, earing behavior and directional properties of cold-rolled andannealed commercial pure aluminum strips were investigated. It was found thatthe ideal orientations of cold-rolled texture can be best described as (110) [112] +(112) [111] which yielded ears of drawn cups at the position 45° from the roll-ing direction. The recrystallization texture of specimen annealed at 500℃ exhibit-ed a strong (100) [001] + (146) [211] texture superimposed on the retained de-formation texture (110) [112] + (112) [111] and the earing position in the drawncups changed from 4 ears at 45°to 6 ears at 30° and 90° from the rolling direc-tion. When annealing temperture was raised to 600℃, a rather strong cubic tex-ture, (100) [001], was developed and the earing position changed again from 6ears at 30° and 90° to 4 ears at 0° and 90° from the rolling direction. The surface texture of cold-rolled aluminum strips was found to be (100) [110]and a transition zone between the surface texture and the main texture was observed. The mechanical properties of cold-rolled and annealed aluminum strips atvarious angle to the rolling direction were determined. It was found that the earingposition in the drawn cups generally occured at the directions where the approx-imate true stress, σ_R, is minimum as observed in copper. A mechanism of the development of a principal texture, (146) [211], wasproposed.

In order to improve the accuracy of spectrochemical analysis of high speedsteels, a systematic study of the effect of different excitation conditions and di-fferent micro-structures of the samples on the accuracy of analysis has been made. A statistical study of the experimental error of single determinations showedthat the accuracy of a spark source was higher than that of an arc source, how-ever, this varied much with the different line pairs chosen. The slope (b) of thecalibration curves changed with current in an arc source and the value of b cameto a maximum when the current was in the neighborhood of one ampere. Apossible explanation of the appearance of such a maximum was given. In a sparksource it was found that the slope of the calibration curves of chromium or vana-dium decreased with an increase of the spark gap, but this was insignificant inthe case of tungsten. The effect of micro-structure of samples on calibration curves has also beenstudied. The standard samples of different structures (cast, hot forged and anneal-ed) were prepared for the present investigation. A parallel shift of calibrationcurves due to difference in micro-structure was observed in an arc source but notin a spark source. This type of shift can be explained by the facts obtained fromour previous work on the effect of micro-structure of steel samples on spectroche-mical analysis.

~~