Page images




The advantages of quartz fibers as suspensions, principally their small elastic fatigue, great strength, and the possibility of drawing very fine threads, have led to their use in a great many important investigations where fine suspensions and a steady zero point are required.

In my work, suspensions were needed, able to carry a load of 500 grams and more and at the same time having as small an elastic fatigue as possible. Naturally fused quartz was selected as the substance promising the best results. Such fibers must be rather thick, from 0.1 to 0.2 mm in diameter. Since it was desired to draw them at least 1 meter in length, it required the melting of a rather large bead free from air bubbles. The process of building up quartz rods is known to be rather tedious. Shenstone's method is simpler, and consists of heating the crystals to about 1,000° C. and suddenly quenching them in cold water. The crystals after such treatment are not shattered, and do not break when brought into the oxy-hydrogen flame. A stick of the proper dimensions is easily formed, containing, however, a large number of air bubbles. To remove these is a very tedious task and particularly exasperating, because quartz at the high temperature of the blowpipe flame is quite volatile, about one-half of the mass evaporating during the process.

Whichever method is employed it means a considerable loss of time, if a great many fibers of the dimensions necessary for this work have to be drawn.

Boys, in his first paper on this subject, mentions his experiments on a great many minerals, of which only a few, however, could be drawn into fibers, and these he says were far inferior to those made from

a Shenstone: Nature, 64, p. 65; 1901.
o Boys: Phil. Mag., 23, p. 489; 1887.

fused quartz. I have also tried a great number of substances, all silicates of magnesium: Enstatite, Olivine, Serpentine, and Meerschaum, without success. Only the Amphibol-Asbestos, Mg, CaSi,0,2, in the ordinary mineral form, as well as when specially prepared for chemical purposes, and the Steatite or Soapstone, Mg, H.Si,O2, give clear beads before the oxy-hydrogen blowpipe. These, under proper precautions, can be drawn easily to fibers of the desired dimensions. The soapstone is especially easily worked and gives fibers of practically the same elastic properties as those of quartz. Since the method of making them is so simple and requires little time, a short description may be of general interest. The soapstone when heated to a high temperature becomes exceedingly hard, and is used commercially under the name of “lava” for making apparatus intended to be able to stand high temperatures-for example, tips of gas-burners. While it can not be melted before an ordinary blowpipe, it does so before an illuminating gas-oxygen jet and forms a clear bead, usually of a greenish tint, due to the presence of a trace of iron. The original soapstone or the “lava" will do equally well, the most convenient form being small cylindrical sticks. (I obtained such cylinders 3 mm in diameter and 7 cm long, giving colorless transparent beads about 5 mm in diameter, from the Chattanooga Sunlight Lava Manufacturing Company.) If the flame of the blowpipe is too long or the mixture of the gases not well adjusted, the substance will boil violently, while with a small, quiet flame it melts without boiling, a slight development of gases being noticeable only at the upper, cooler part of the bead. After taking the pearl out of the flame the thread is drawn, its thickness depending upon the temperature and rapidity with which it is drawn. Very fine fibers can thus be secured. The fibers should not be heated after being drawn. In a Bunsen burner, for instance, they will immediately become white and break to pieces.

Elastic fatigue.-It is very well known that fine quartz fibers show hardly any elastic fatigue, a but with thicker fibers of 0.1 to 0.2 mm diameter the effect of elastic fatigue became quite apparent. Careful experiments showed that small twists, or twists of small amplitude, do not affect the zero point, while larger twists, being continued for several minutes, would displace the zero point. The apparatus used allowed a turning of the torsion head, while the lower end of the fiber was held in its original position. After release the zero could be redetermined in a few seconds and the slow disappearance of the effect of elastic fatigue observed. With fibers 70 cm long and 0.1 mm diameter

a Boys: Phil. Mag., 23, p. 496, 1987; Threlfall: Phil. Mag., 30, p. 113; 1890.

the displacement, after release from a torsion of 360° lasting five minutes, amounted to as much as 2 in 3,000. Fibers made of steatite showed about the same effect, which is, however, very much smaller than any other substance experimented with, and about one-half to one-third of that shown by steel or phosphor bronze.

For these experiments the quartz and steatite fibers had been silvered at their ends, then coppered, and finally soldered into large brass pins clamped securely in the supports. Thus the elastic fatigue observed can not be attributed to an effect of the material with which fibers were fastened to the other parts of the instrument.

Tensile strength.— The tensile strength of very thin quartz fibers is quite large, 10x10' dynes per cm’, but it decreases appreciably with increasing thickness.a My experiments show the same results. From the following table it is seen that the tensile strength of the steatite fibers is at least as large as that of quartz. For comparison the values given by Boys (B) are added. The decrease of tensile strength for the thicker fibers is very striking. For the thickest fibers it is, however, still as large as that of brass.

[blocks in formation]

For thick rods the tensile strength is considerably smaller, as shown recently by Schulze, who found a value of 0.6 X 10' dynes cmo for a rod having a cross section of 0.272 cmo.

a Boys: Phil. Mag., 30, p. 116; 1890.
b Schulze: Ann. d. Phys., 14, p. 384; 1904.

Coefficient of simple rigidity.The torsional coefficient was determined by the well-known method of vibration. The diameter and length of the fibers were carefully measured, a cylinder made of Tobin bronze was suspended by the fibers, and the period of the torsional vibrations determined by means of a chronometer. Then

ΙΙ n=

Since it became apparent that different fibers of the same substance gave slightly different values, and since I measured the diameter of the fibers only with an accurate micrometer, I did not take the trouble to correct for the coefficient of expansion.a

The mean of a number of experiments gave for quartz n=3.4X 10; for the steatite fiber a value from 2.6 to 4.2 x 1011. Also in this respect the new fiber almost exactly equals fused quartz. The temperature coefficient of the torsional constant of quartz is given by Threlfall as +0.000133; by Barnettas +0.000115;' I obtained +0.000149. Steatite does not show the remarkable property of becoming more rigid at higher temperatures. Its temperature coefficient is negative, a= -0.000193. The results obtained are given in Table II and plotted in fig. 1.

[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small]
[blocks in formation]

Coefficient of expansion of fused steatite.— The following experiments were made by Mr. L. G. Hoxton, of the Bureau of Standards, using a rather thick fiber of fused steatite.

The length between two fairly well defined marks 110.4 mm apart was compared at different temperatures with that of the standard nickel-steel decimeter No. 43, belonging to the Bureau and furnished with a certificate from the Bureau International des Poids et Mesures, which gives its linear coefficient as a=0.96 x 10-6. In the following table the differences between the length of the fiber and the standard are marked F-NS. In the third column is given the linear coefficient of expansion, counting from the lowest temperature, and in the last the number of observations made for each temperature.

[merged small][merged small][merged small][merged small][ocr errors][merged small][merged small][merged small][merged small][merged small][merged small][merged small][merged small]
« PreviousContinue »