Size distribution of huntite-borate crystals grown by spontaneous crystallization from flux
- PDF / 503,428 Bytes
- 4 Pages / 576 x 792 pts Page_size
- 16 Downloads / 185 Views
re cooled to room temperature over 5-10 h. All temperature parameters are given in Table I. After the crystal growth process, crystals were removed from the melt by dissolving excess flux in hot HNO 3 and water, and then were dried. The huntite structure for each batch of crystals was checked by x-ray powder diffraction analysis. The grown crystals with the huntite structure were mostly colored transparent hexagonal rods as shown in Fig. 1. Since cooling rates for all melt compositions were too high, the optical quality of the prepared materials was not as high. Moreover, as the given growth procedures shown in Table I were only preliminary ones, other growth parameters also were not optimized. The crystals contained many cracks which were formed during either crystal growth or melt solidification. The as-grown crystals were separated into different size fractions using meshes with hole sizes from 0.037 to 1.680 mm. Each fraction was weighed precisely. The fractions with sizes less than 0.037 mm were not detected. The maximum size of the crystals for all dimensions was about 2.0 mm. The weights of each fraction for all crystal compositions are presented in Table I. For each fraction with size Si < S < Si-1, the size interval AS,- and the mean size value S* were calculated using the following formulas: AS ; = Si-i - Si 5 ; = (Si + 5,-_i)/2
(1) (2)
The weights given in Table I were recalculated to wt. % for comparisons between different melts and crystals. In this case, the total weight of crystals for each composition (3) and the weight of crystals with size S, < S < Si-i (4) can be represented as follows, respectively:
W = \
W(S) dS = 100%
(3)
•Si-i
W(S)dS a)Permanent
address: General Physics Institute, Vavilov Street 38, Moscow, 117942, Russia.
3028
J. Mater. Res., Vol. 9, No. 12, Dec 1994
http://journals.cambridge.org
Downloaded: 15 Mar 2015
(4)
Assuming that the size distribution inside each interval, Si < S < Si-i, was linear, it was possible to © 1994 Materials Research Society IP address: 138.251.14.35
Communications
TABLE I. Melt compositions (mol %), crystal growth parameters, and weight of each size fraction Wt (St < S < S«-i), 1» for size S measured in mm. 5-1
Melt No. Yb 2 O 3 Er 2 O 3 Nd 2 O 3 Bi 2 O 3 A12O3 Ga 2 O 3 Cr 2 O 3 B2O3
1.12 2.53
0
76.69 1200-1250 1000
3.69
0
4.58
0.0076 0.4589 0.6614 0.7954 1.027 1.4388 3.3049 3.3518 2.0322 1.273 0.6825 0.0361
0.0172 0.3129 0.7816 1.0973 1.7177 2.9748 6.5492 5.7232 3.2125 2.435 2.2657 0.3112 0.188 0.1439
0.0129 0 15.0825
0
27.7302
calculate the weight fraction W* (wt. %) at the centers of each interval S* per AS = 0.001 mm as follows: (5)
The assumption about linear distribution inside the intervals was reasonable, because the number of meshes used was sufficiently large. These calculated data are presented in Fig. 2 as W* (wt. %) dependence on S*. It is interesting to note that using the logarithm axis, the dependence mentioned for all compositions looks like almost symmetrical functions. It was interesting to represent these da
Data Loading...
