Glass which is not free from strain is doubly refractive, and on this account should not be used between the polarizing and analyzing systems of a polariscope. It is therefore of great importance that cover glasses be made of optical glass and thoroughly annealed. A strain in the cover glass is in effect a rotation, a cause of many baffling discrepancies that occur in polarizations. There is no infallible method of detecting this trouble after the glass is in use, the only safe procedure being to use cover glasses which have been tested, and to put as little pressure as possible on the screw caps. After a setting has been made, it is advisable to rotate the tube in the trough of the instrument; if the halves of the field show variations in intensity, strain exists in one or both of the glasses. Strains may be caused by poorly fitting rubber washers. These should be made from the best quality rubber, soft and flexible. They should be made of such size that they lie flatly and evenly in the cap with no marginal elevation. Once a glass has been strained, it should not be used for several days or until test shows the absence of strain. All cover glasses must have plane, parallel surfaces, free from scratches, and should never be less than 1 mm thick. A thickness of 1.5 to 2 mm is preferable. The necessity for optically perfect glasses has not received the attention its importance demands. The cover glasses used in the laboratories of the National Bureau of Standards for Bates-type polariscope tubes conform to the following specifications: "Cover glasses shall be made from clear, colorless optical glass, thoroughly annealed, free from strain, and shall show no optical rotation or double refraction when observed in a precision polarimeter. The surfaces shall be plane and parallel and be free from scratches. The edges shall be slightly rounded to prevent chipping. Plane parallelism shall be within 5'; thickness 1.85 ±0.15 mm; diameter 23.2 0.2 mm." The National Bureau of Standards will accept polariscope tube cover glasses for test. 3. VOLUMETRIC FLASKS** (a) SPECIFICATIONS (1) MATERIAL AND ANNEALING. The material should be the best quality of glass, transparent, and free from bubbles and striae. It should have small thermal hysteresis and should adequately resist chemical action. All flasks should be thoroughly annealed before being graduated. (2) DESIGN. The cross section of the neck must be circular, and the shape of the flask must be such as to admit of complete emptying *See test fee schedule 423, p. 553. **See test fee schedule 241, p. 558. and drainage from the whole interior surface at the same time. The bottom of the flask should be slightly concave upward and should be of sufficient size to enable the flask to stand on a surface inclined at an angle of 15° to the horizontal. The neck must be cylindrical for at least 1 cm on each side of every graduation mark, but may be enlarged in the form of a bulb between graduation marks (for example, Giles flasks). At the graduation mark the inside diameters of the neck of the flask must be within the limits given in table 9. TABLE 9.-Diameters of necks and tolerances for volumetric flasks (3) GRADUATION MARKS.-The graduation marks must be of uniform width, finely but distinctly etched, must be perpendicular to the axis of the flask, and must extend completely around the neck. On flasks having a capacity of 100 ml, or less, the graduation mark shall be not less than 3 cm from the upper end, nor less than 1 cm from the lower end of the neck, and on flasks having a capacity of more than 100 ml, the graduation mark shall be not less than 6 cm from the upper end, nor less than 2 cm from the lower end of the neck. (4) UNIT OF VOLUME. The unit of volume employed is the liter, which is defined as the volume occupied at the temperature of its maximum density (4° C) by a quantity of pure water having a mass of 1 kg. The water is under a pressure of 760 mm of mercury, and the weighings are reduced to vacuo. The one-thousandth part of the liter, called the milliliter (ml), is also employed as a unit of volume. (5) STANDARD TEMPERATURE.-Twenty degrees centigrade has been adopted by the Bureau as the standard temperature for volumetric apparatus, and an extra charge is made for testing apparatus for use at other temperatures. (6) INSCRIPTIONS.-Each flask must bear, in permanent and legible characters, the capacity in liters or milliliters, the temperature at which it is to be used, the method of use, i. e., whether to contain or to deliver, and an identification number. In the case of flasks with stoppers, the stopper must bear the same number as the flask, or, if standard interchangeable grindings are used, they must bear the proper identification marks. A suitable arrangement of the inscription is as follows: No. 134 100 ml at 20° C. (b) TOLERANCES The tolerances for flasks of various sizes are shown in table 9. (1) PRECISION STAMPS.-Flasks tested by the National Bureau of Standards and found to comply with the foregoing specifications will be given the official precision stamp of the Bureau. The stamp consists of the letters "NBS" and the year in which the test is made, surrounded by a circle. Thus for the year 1941 the stamp will be NBS The stamp will be placed on the neck of the flask, above the graduation mark. (2) SUGAR-TESTING FLASK (BATES). The type of flask used in the sugar laboratories of this Bureau and in the United States Customs Service is shown in figure 30, No. 1. It is especially designed for sugar polarizations and is described in the Customs Regulations, 1931, chapter XI, Sampling, Weighing, and Testing of Sugars, Sirups, and Molasses, as follows: "The flasks shall have a height of 130 millimeters, the neck shall be 70 millimeters in length and have an internal diameter of not less than 11.5 millimeters and not more than 12.5 millimeters. The upper end of the neck shall be flared, and the graduation marks shall be not less than 30 millimeters from the upper end and 15 millimeters from the lower end of the neck." All flasks shall be standardized to contain 100 ml at 20° C. These specifications permit a maximum internal diameter of the neck of 12.5 mm, which exceeds the maximum allowed in the National Bureau of Standards specifications for 100-ml precision flasks by 0.5 mm. However, since the sugar flask is used almost exclusively for polariscopic work, it is tested as precision volumetric apparatus, and if the tests show it to be in accordance in all respects with the specifications given above, it will be given the precision stamp of this Bureau, described under (b) (1) above. In the Bates type of flask the neck, while smaller than that of the ordinary 100-ml sugar flask, is made slightly larger than the Bureau's requirements for a flask of this size in order that the neck shall not become clogged when the sugar is being introduced into the flask. The upper end is flared to facilitate pouring. The height of only 130 mm minimizes the dilution of the solution by moisture on the upper part of the neck and gives a length which readily permits the flask to be closed with the thumb while the forefinger rests on the bottom, thereby facilitating a thorough mixing by shaking, with no loss by spilling. (3) SPECIAL. A number of flasks designed for special purposes are shown in figure 30. Flask 1 is the Bates-type sugar flask, specifications of which are given above. This flask is used extensively in the United States Customs Service and elsewhere in routine sugar analysis. Flask 2 is the Kohlrausch flask with a funnel neck to facilitate the transfer of the solid sugar without loss. Flask 3 is typical of the double-graduated flasks and is used in inversions and clarification methods where one volume is to be diluted to a different volume. The usual sizes are 50-55, 100-110, 200-220, and 500-550 ml. Flask 4 was designed at the National Bureau of Standards for precision work. The inside diameter of the neck is 5 to 6 mm. The enlarged portion of the neck permits complete mixing of the contents of the flask without the solution coming in contact with the grinding or the thumb until the mixing is complete. Before making to volume, the bulb is dried inside by a current of filtered air. The glass tube is ground to fit the grinding of the flask and also the glass stopper of the tubulated water-jacketed polariscope tube, which permits the transfer from flask to polariscope tube without exposure to the air and the consequent evaporation of the solution. In addition to the 100-ml mark on the neck of the flask, there are supplementary graduations above and below the mark. With this flask it is possible to make the calibration and reproduce the volume of solution with an accuracy of 0.002 ml. Flasks 5 and 6 are special flasks used chiefly in the determination of the density of molasses and sirups. (4) CORRECTION TABLES.-Tables 106 and 107, p. 612, are given for the convenience of those who wish to verify the graduation of volumetric apparatus. More complete data for this purpose will be found in National Bureau of Standards Circular C19. 4. THERMOMETERS* (a) GENERAL Thermometers [1] should be of good design and workmanship and should be made of suitable materials, with special attention to the thermometric properties of the glasses used. Detailed specifications, covering the necessary items for the various types of thermometers suitable for use in polarimetric measurements, have been prepared and are available upon request. A sample form of the specifications is printed in this circular, page 382. *See test-fee schedule 311, p. 554. 323414°-42-9 Mercury-in-glass thermometers are calibrated to agree as closely as possible with the International Temperature Scale adopted in 1927 and now in general use. On this scale, which conforms with the Centigrade Thermodynamic Scale as closely as possible with presentday knowledge, temperatures in the interval 190° to 660° C are defined in terms of the resistance of a standard form of platinum resistance thermometer calibrated at basic fixed points. The International Temperature Scale is defined in NBS Research Paper RP22 [2]. It is highly desirable that a fixed point appear on the scale of a mercury-in-glass thermometer. The ice point (0° C or 32° F) or steam point (100° C or 212° F) is convenient. The volume of the bulb of a mercury-in-glass thermometer is known to change with time and use by amounts which must be taken into consideration if the best results are to be obtained. By checking the fixed-point reading from time to time, these changes may be determined and allowed. for. For example, if the ice-point reading is found to be higher or lower than the previous reading, all other readings on the thermometer will be higher or lower by the same amount. Thermometers provided with graduated metal backs should, in addition, have graduations engraved on the glass stem. The thermometer should be securely and firmly fastened to the back. If the thermometer is of the inclosed-scale (Einschluss) type, such fiducial mark should be placed on the outer glass tube. Thermometers should comply in all respects with the applicable specification. (b) ICE BATHS [3] Because of the changes in bulb volume, the ice bath, which provides a convenient means of determining the amount of such changes, is very important. The most convenient form consists of a widemouthed thermos bottle or Dewar flask, filled with a mixture of shaved ice and distilled water, or water obtained by the melting of the ice. Other containers may be used, but are likely to be less convenient because of the more rapid melting of the ice. Clear ice is considered sufficiently pure for the purpose and is readily obtainable. The ice may be shaved by means of a small plane, resembling a carpenter's plane, or other appropriate mechanism. There should be enough water in the mixture to make it soft or slushy, but not enough to cause the ice to float. Excess water which accumulates may be conveniently removed by means of a glass siphon, ending in a rubber tube with a pinch cock. Precautions should be taken to wash the ice and to avoid contaminating it in handling. 5. WEIGHTS* (a) SUGAR It is generally advantageous to have special 26- and 13-g weights for weighing out sugar samples for direct polarization. For this work, the Bureau recommends that the ordinary screw-knob type of weight be avoided in favor of a strictly one-piece weight in which the knob forms an integral part of the weight. Gold- or platinum-plated Tobin bronze weights have been found satisfactory. Recently, *See test fee schedule 226, p. 557. |