and place the filled instrument in a water thermostat held at 20° C. Allow to remain in the thermostat for a sufficient time to reach the temperature of 20° C. Adjust the volume by removing the excess water which has exuded from the capillary stem, fit the ground-glass cap in place, and remove the instrument from the bath; wipe dry with a clean cloth and after allowing to stand for 15 to 20 minutes, weigh. Reduce the weight in air of the contained water to the weight in vacuo. Obtain the volume of the picnometer by dividing the weight in vacuo of the water content by the density of water at 20° C, 0.9982343. After emptying and drying the picnometer, fill it with the sample in question, and ascertain the weight of the contained sample at 20° C. Reduce the weight in air of the contained sample to the weight in vacuo. Divide the weight of contained sample in vacuo by the volume of the picnometer to obtain the true density, d20.

-E

(2) METHOD OF NEWKIRK.-The accurate determination of the density of blackstrap molasses is difficult due to the high viscosity of the material and to the presence of included and dissolved gases. Newkirk [7] has designed a special picnometer for making the determination (fig. 42). It consists of a bottle, C, fitted with an enlargement at the top, B, ground optically flat and closed by another optical flat, A. An expansion chamber, D, is ground to the bottle and fitted with a vacuum connection, E. Το avoid loss of water due to evaporation under reduced pressure, the connecting tube is fitted with a stopcock, F, so that when the proper vacuum has been reached the apparatus can be closed off from the vacuum source. In using the picnometer, the expansion chamber, after lubrication of all joints with molasses, is placed on the bottle. The molasses to be analyzed is allowed to flow into the bottle and into the expansion chamber until the latter is about one-third full. The vacuum line is then connected and the pressure reduced until the gas expands into visible bubbles. The apparatus is then closed off by turning the stopcock, F, and the whole placed in a thermostat and allowed to remain until the temperature has reached equilibrium and all of the bubbles have collected in the expansion chamber. The expansion chamber is removed and the volume fixed by carefully sliding plate A over surface B. The picnometer is then removed from the thermostat, wiped clean, placed in the balance case, and weighed. The weight of the contained sample is corrected to vacuo and compared with the weight of an equal volume of water at 4° C in vacuo.

с

FIGURE 42.Newkirk picnometer.

(3) WEIGHT PER GALLON OF MOLASSES. Since, in commercial transactions, molasses is sold both by volume and by weight, the determination of the weight per gallon is of considerable importance. The method employed at the National Bureau of Standards and adopted by the United States Customs Service is as follows:

A special 100-ml calibrated volumetric flask with a neck of approximately 8 mm inside diameter shall be used. Weigh the flask empty and then fill it with molasses, using a long-stem funnel reaching below the graduation mark, until the level of the molasses reaches the lower end of the neck of the flask. The flow of molasses may be stopped by inserting a glass rod of suitable size into the funnel so as to close the stem opening. Remove the funnel carefully to prevent the molasses coming in contact with the neck, and weigh flask and molasses. Add water almost up to the graduation mark, running it down the side of the neck to prevent mixing with the molasses. Allow to stand several hours or overnight to permit the escape of bubbles. Place the flask in a constant temperature water bath at 20° C for a sufficient time for it to reach the temperature of the bath, then make to volume at that temperature, with water. Weigh. Reduce the weight of the molasses to vacuo and calculate the density.

Example.

Weight per gallon determined at 20° C.

Weight of flask, 37.907 g.

Weight of flask and molasses, 167.148 g.

Weight of flask, molasses, and water, 174.711 g.

167.148 g-37.907 g=129.241 g=weight of molasses (in air with brass weights). 174.711 g-167.148 g 7.563 g=weight of water (in air with brass weights). Calculating volume of water from weights in air at 20° C.

Divide weight of water in air by weight of 1 ml of water in air at 20° C (table 106, p. 612), 7.563/0.99718-7.584 ml.

Volume of flask at 20° C, 100. 060 ml

7. 584 ml

92. 476 ml volume of molasses.

=

To reduce weight of molasses to vacuo: 129.241/8. 4=15.4 ml

77.1X0.0012

(8.4

volume of brass weights.

density of brass weights).

92.5 ml volume of molasses (approximate).

15.4 ml

volume of weights.

77.1 ml net volume of air displaced.

0.093 g, buoyancy correction to be added to weight of molasses.

(0.0012 g weight of 1 ml of air at 760 mm at 20° C).

(c) BY MEANS OF DIRECT-READING BALANCE

The weight per gallon of molasses may be determined directly by means of a torsion balance designed by H. J. Bastone, of the American Sugar Refining Co. It is fitted with two beams, one a double beam for taring the sample bottle, and the other a recording beam graduated in pounds per gallon from 10.80 to 12.05 in 1/100 pound per

gallon. A bottle or picnometer for containing the molasses is furnished with the balance. The performance of the balance has been investigated by Snyder and Hammond [8], who deemed it more convenient and accurate to fix the volume by sliding a flat glass disk seated on the flat polished top of the perforated stopper. Johnson and Adams [9] and Newkirk [7] have shown the accuracy of thus fixing the volume. By this procedure only a very thin film of molasses remained between the disk and the top of the stopper. The calibration of the volume of the bottle and the direct determinations were made in the same manner.

This balance seems entirely satisfactory for most determinations of the weight per gallon of molasses, provided the usual precautions are taken to allow the foam to subside and occluded gases to escape. Any direct-reading balance will be found useful in commercial work and routine testing as compared with other methods, since results may be obtained without resort to tedious calculations.

(d) BY MEANS OF ANALYTICAL BALANCE

Another method of determining the specific gravity of sugar solutions is based on the well-known principle of Archimedes that a body immersed in a liquid is buoyed up by a force equal to the weight of the displaced liquid. In making a determination, a glass sinker or bulb weighted with mercury is suspended from the arm of a balance by means of a fine platinum wire. It is weighed in air, immersed in distilled water, and immersed in the sugar solution. The specific gravity is calculated from the equation

S specific gravity of sugar solution,
A weight of sinker in air,

B weight of sinker in distilled water,
C=weight of sinker in sugar solution.

By reducing the weights in air to weights in vacuum and taking into account the temperature, the density of the air, and the atmospheric pressure, the true density may be calculated.

The so-called Westphal balance employs the same principle as the above and is so graduated that the specific gravity is obtained directly from the readings of the riders on the beam. A more detailed discussion may be found in textbooks on physics.

5. DENSITY OF SUCROSE SOLUTIONS

The density of sucrose solutions of different concentrations has been the subject of extensive study by investigators over a period of many years. Among the early workers in the field were Balling [1], Niemann [10], Brix [2], Gerlach [11], Scheibler [12], and others. A number of tables were published, varying with respect to the temperature, and expressed in terms of true density or of specific gravity. In 1900 the German Normal-Aichungs-Kommission published density tables of sucrose solutions [13] based on the very precise determinations of

323414°-42--18

F. Plato in collaboration with J. Domke and H. Harting. These tables include the percentages of sucrose by weight for density at 20°/4° C, table 113, p. 626, and specific gravities 15° C/15° C and to C/15° C. These tables are considered the most accurate available and are universally accepted as standards. Using the Plato tables as a basis, numerous other tables have been computed, giving such values as grams of sucrose per 100 g of solution (Brix), grams of sucrose per 100 ml of solution, etc. Among the more widely used of these are the tables of Sidersky, Les densités des solutions sucrees á differentes temperatures, Paris, 1908.

6. REFERENCES

[1] C. J. N. Balling, Gahrungschemie 1, pt. 1, p. 119.

[2] A. Brix, Z. Ver. deut. Zucker-Ind. 4, 304 (1854).

[3] A. Baumé, Elements de Pharmacie (Paris, 1797).

[4] C. F. Chandler, Mem. Am. Nat. Acad. Sci. 3, I, 63-73 (1884).

[5] F. J. Bates and H. W. Bearce, Tech. Pap. BS (1918) T115.

[6] Official and Tentative Methods of Analysis of the Association of Official Agricultural Chemists, 5th ed., p. 485 (1940).

[7] W. B. Newkirk, Tech. Pap. BS 13 (1920) T161.

[8] C. F. Snyder and L. D. Hammond, Tech. Pap. BS 21 (1927) T345.

[9] J. Johnson and L. H. Adams, J. Am. Chem. Soc. 34, 566 (1912).

[10] C. von Niemann, Erdmann's J. ötonomische tech. Chem. 15, 106.

11] Th. Gerlach, Dingler's Polytech. J. 172, 31; Z. Ver. deut. Zucker-Ind. 13, 283 (1863).

[12] C. Scheibler, Neue Z. 25, 37, 185 (1890).

[13] F. Plato, J. Domke, and H. Harting, Z. Ver. deut. Zucker-Ind. 50, 982, 1079 (1900).

XIV. REFRACTOMETRY

1. GENERAL

The refractive index of a pure sucrose solution is an accurate measure of the concentration of dissolved substance. The first refractometer table appears to have been prepared by Ströhmer [1], who showed the relation between the refractive index and the specific gravity of sugar solutions. Stolle [2] determined the refractive indices of solutions of sucrose, dextrose, levulose, and lactose, using the Pulfrich refractometer. He found but little variation in the refractive index of solutions of the different sugars at the same concentration.

Tolman and Smith [3], using an Abbe refractometer, found that solutions of a number of sugars and related substances in equal percentage composition by weight gave approximately the same refractive index. They recommended the use of the refractometer in the determination of soluble carbohydrates in solution and pointed out the advantages of the method from the standpoint of speed, ease of manipulation, and the small amount of sample required.

To expand the usefulness of the refractometer, Geerligs [4] showed that the refractive indices of impure sugar solutions, even when the impurities consisted of mineral salts, yielded far more reliable measures of total dissolved substance than the determinations by densimetric

measurement.

Hugh Main [5] prepared a table of refractive indices of sucrose solutions and demonstrated the applicability of the Abbe refractometer in sugarhouse work. Schönrock [6], using methods of high precision, determined the indices of sucrose solutions for concentra

tions ranging from 0 to 66 percent. Landt [7] redetermined the indices for concentrations from 0 to 24 percent and obtained values in almost perfect agreement with the Schönrock values for that range. In 1936 the International Commission for Uniform Methods of Sugar Analysis, realizing the need for a standard table of refractive indices of sugar solutions, adopted such a table [8]. This table, known as the "International Scale (1936) of Refractive Indices of Sucrose Solutions at 20° C," is constructed from the values of Schönrock-Landt (1933) up to 24 percent, of Schönrock (1911) from 24 to 66 percent, and of Main from 71 to 85 percent. The values 67 to 71 were obtained by extrapolation of the Schönrock values as a straight-line curve to meet the Main value at 71. This table in an expanded form is given in table 122, p. 652.

A similar table, based on the same data for use with the tropical model refractometer, standard at 28° C, is given in table 124, p. 658. Temperature corrections are made by means of tables 123 and 125.

2. PRINCIPLES OF THE ABBE REFRACTOMETER

N

K

M

ια

When a ray of light travels through a homogeneous medium, its path is a straight line. However, when the ray passes from one medium to another at an angle oblique to the surface of separation of the two media, the direction of the ray changes abruptly at the surface of separation. In addition to the portion of the ray which penetrates the second medium and is bent or refracted, a portion of the light is reflected. According to Snell's Law, the sine of the angle of refraction bears a constant ratio to the sine of the angle of incidence for all angles of incidence, the value of the ratio depending on the nature of the two media at the surface

OB

B

m

/R

N'

FIGURE 43.-Refraction of light.

of separation at which the refraction takes place, and also on the wave length of the incident light. This law is usually expressed as

where n is the index of refraction; i, the angle of incidence; and r, the angle of refraction. This is illustrated by figure 43.

Suppose AB represents the surface of separation between two media, say, air above and water below, and that a ray of light having the direction 10 is incident to AB at 0. Let NON' be the normal to the surface of separation, AB. A portion of the light of the ray 10 is reflected in the direction OK; the other portion is refracted in the direction OR. The angle IOM, or a, is the angle of incidence, and angle ROm, or B, is the angle of refraction.