Document KRgB6ywM35NQZ5vokQzeq94Lo

KE1J1 YADA 661 sppreciably from umplc to sample as shown in Ftp. 6. fringes corresponding In 200, but, in praoicc, iliii w.i% This is one of the new facts found by the lattice image not possible because of the very poor tonlnisl of this melhod. In the ease of the Coalinga sample, for in fringe system compared with the much higher contrast stance, the percentage of the fibrils showing parallel of the 7-3 A fringes, crossing them. Diffraction pat fringes is about 20%. The synthetic sample showed a terns corresponding to ortho-chrysotile were rarely ! similar feature, as can be seen in Fig. 6. On the other obtained. The results obtained on these para- and i hand, the sample from Quebec which was studied in ortho-chrysolite fibrils will be reported in detail else ! the previous work (Yada, 1967) gave a fairly high pcr- where. | centage of 94% for the fibrils showing the parallel j fringes, though its histogram is not included here. The (c) Growth patterns samples from other sources gave intermediate values. Lattice images suggesting some new information It is seen that the angular distribution of the tilted 4-5 about the growth mechanism were found at the growth A fringes shows a peak value at 2-3* and ranges up to end of the fibrils. Fig. 7 shows an example obtained about 10 depending on the source of the samples. from the Coalinga sample in which (a) is a low magni (h) Paro-chrysotile and ortho-chrysotile fication image and (A) its enlargement It is seen that the surface of the fibrils is quite clean on the molecular The existence of a small number of para-chrysotile scale, without any amorphous material, and that the fibrils (Whittaker, 1956 c) was confirmed in certain end of the fibril has a conical shape formed by steps samples (Globe and Tasmania) by observing the 4-5 A of two or three sheets rolled,up together. It is also seeif : fringes nearly perpendicular to the fibre axis, as was that this conical fibril is considerably thicker at the : reported in the previous paper (Yada, 1967). bottom than at the top (the lower darkened area in In principle, identification of ortho-chrysotfle (Whit-' (a)}, which suggests a discontinuous step-growth as taker, 1956; Zussman, Brindley ft Comer, 1957) described later. Such a conically stepped structure was would be possible by observing the 2*6 A transverse also observed inside a fibril as seen in Fig. 8 These observations seem to support the coupled dislocation mechanism proposed by Jagddzinski ft Kunze (1954). In the case of synthetic chrysolite, another type of cone-shaped fibrils was found in which the 7-3 A fringes on opposite sides of a fibril are not parallel, as shown in Fig. 9.- These two sets of 7-3 A fringes make an angle of 3 for fibril A and of 5-5* for B. The 4-5 A fringes in the middle of these fibrils are also tilted at a fairly large angle, to the fibre axis, though their fringe contrast is poor. This is illustrated in the drawing (6) which traces the fringe systems of A and B in (a). It is noted that the tilling of the two sets of the 4-5 A fringej arc asymmetrical about the fibre axis with an angular dilTcrcncc of a few degrees. This type of growth pnitcrn seems to correspond to the oonc-in-conc pattern which was reported by Boles ft Comer (1957). HH <*> Observation front the direction parallel to the fibre axis The general feature that most fibrils are hollow but a few arc solid, was seen in the cross section of the samples from all the sources examined. Further ob servations revealed the following. Fig- 6. Angular distribution of fhs frequency of tha tilted 41A Magst. (a) Step-growth Fig. 10 shows the cross section of a rather thick fibril from Coalinga indicating a two-step growth. The two or three-step growth was also observed in relatively thick fibrils in the samples from other sources. The frequency of such two- or three-step growth is different from sample to sample, and is particularly high in the specimen from Tasmania as shown in Fig. 11. It is seen that fibrils thicker than about 350 A show traces of the stepwise growth. In the analysis of the distribu tion of the fibril diameter, therefore, auch a stepwise growth should be taken Into consideration. wwprt:..'.' 1 'r .' *. VV-. r *- ,. ROOD OSS) PRODUCED BY FORD