Page images
PDF
EPUB

had corroborated, the French Sugar Scale would from that time on have been in agreement with the International Sugar Scale which was adopted by the International Commission for Uniform Methods of Sugar Analysis at Amsterdam in 1932. Mascart and Benard [9] gave 66.54° as the specific rotation of sucrose, and 16.284 g weighed in vacuo as the normal weight. Through an error in converting this weight in vacuo to weight in air with brass weights, they arrived at the value 16.29, confirming Sidersky's calculation. If this error is eliminated, Mascart and Benard's value becomes 16.27 g.

Also Pellet [10] gave [a]=66.536, and 16.285 g in vacuo for the normal weight. The latter corresponds to 16.27 g weighed in air with brass weights.

In 1933, 38 years after his original calculation, Sidersky [8] accepted the specific rotation of 66.54° given by Schönrock [12] as the best average value and again calculated the normal weight and found the value to be 16.282 g. This he said differed by only a negligible amount from 16.29 g. Sidersky appears to have overlooked the fact that the specific rotation calculated from Schönrock's formula, which he accepted, is that based upon weights in vacuo, and hence that the normal weight which he obtained is the normal weight in vacuo. Corrected to weighings in air with brass weights, his figure becomes 16.272 g. If 21.667° be used, instead of 21.67° which Sidersky used, this figure is still further reduced to 16.270 g.

It has also been shown [11] by calculation from the best available constants that a French normal weight of 16.269 g would be necessary to bring the French Sugar Scale and the scale of the International Commission for Uniform Methods of Sugar Analysis into equivalence. This normal weight exactly (considering that the tolerance of experimental error is set at 0.002) equals that calculated by Sidersky, namely 16.270 g, after due attention is paid to the obviously necessary correction for buoyancy of the air and the 100° French Sugar Scale point rotation taken as 21.667°, instead of 21.67°, and also is the same as that actually indicated by the data of both Mascart and Benard and of Pellet.

Since the 100° point of the French Sugar Scale is fixed by definition as a rotation of 21°40' or 21.667°, it is advisable to use the latter figure to replace the less accurate value of 21.67° now used. Paralleling the definition of the International Sugar Scale, the following statement holds for the corrected French Sugar Scale:

(1) Normal quartz plate 100° French Sugar Scale-21.667 ±0.002° (X=5892.5 A) at 20° C.

(2) The graduation of the saccharimeter shall be made at 20° C. 16.269 ±0.002 g of sucrose dissolved in water, and the volume made up to 100 ml, all weighings to be made in air with brass weights, the completion of the volume and the polarization to be made at 20° C, on an instrument graduated at 20° C, gives a reading of 100° French Sugar Scale.

Quoting from Bates and Phelps [11]:

From a consideration of the foregoing it is apparent that many of the difficulties and uncertainties which have been attributed to the French Sugar Scale never had an actual existence. They have arisen through an error in adopting a normal weight which the experimental evidence did not fully justify. It should be a simple procedure to eliminate them. It is therefore suggested that the value 16.269±0.002 g weighed in air with brass weights be adopted as the official

normal weight for the French Sugar Scale. When this is done the present official normal weight of 16.29 g will have been discarded. Should this be done, not only will the International Sugar Scale and the French Sugar Scale be brought into exact agreement but there will result a long-overdue recognition of the accuracy of the experimental work of those French investigators whose work was interpreted as providing a value of 16.29 g for the normal weight. It is a fortuitous circumstance that brings the two scales into complete correspondence, and it is worthy of note that these investigators, with the limited facilities available at the early date at which the work was carried out, obtained values for the specific rotation and the normal weight of sucrose identical with those obtained by the use of modern instruments and methods. The adoption of the normal weight, 16.269 g, as the official normal weight for the French Sugar Scale would be an important development of great benefit to the sugar industry of the world.

This has now become an accomplished fact. By an official decision of March 26, 1938, the French Minister of the Budget ratified the change from 16.29 to 16.269 0.002 g for the official French normal weight, to be effective September 1, 1938 [13].

(b) THE GERMAN, OR VENTZKE, SCALE

Ventzke [3], in 1842-43, proposed a method for establishing the sugar scale which would make the use of a chemical balance unnecessary. The 100° point of his scale was established by the rotation of a sucrose solution having a specific gravity of 1.100 at 17.5° C referred to water at 17.5° C. However, it was soon found that the specificgravity method was unsatisfactory in use. Accordingly the weight of sucrose in 100 ml of Ventzke's solution, namely 26.048 g, was taken as the normal weight and the specific-gravity method was abandoned. Many of the old instruments were made for use with 26.048 g in 100 ml.

When the Mohr sugar flask came into general use in 1855, polariscope builders began determining the 100° point by the use of this flask and the same normal weight, 26.048 g, as before, which was of course an entirely different basis, since 100 Mohr cc equals 100.234 ml. Many saccharimeters in use today were constructed on this basis, which may be more fully stated as follows:

The 100° point (Ventzke) is determined by the rotation in a 200-mm tube of a solution containing 26.048 g of sucrose, weighed in air with brass weights, in 100 Mohr cc at 17.5° C, the temperature of the quartz wedges, as well as the polarization temperature, being 17.5° C. From 1855 to 1900 practically all saccharimeters except those using the French Sugar Scale had their 100° points determined on the basis given by this last definition of the Ventzke Scale.

(c) THE INTERNATIONAL SUGAR SCALE

The Ventzke Scale, although in general use for many years, has never been fully understood by polariscopists generally. This has led to much confusion and to the use of 100-ml flasks on instruments standardized for use with the Mohr flask. In addition, 17.5° C is well below the temperature of the average laboratory. Because of these and other considerations, the International Commission for Uniform Methods of Sugar Analysis at the Paris meeting in 1900 recommended the use of a new definition of the 100° point [14], based upon "true cc" and a standard temperature of 20° C. The change to 20° C necessitated a change in the normal weight in order to keep the physical dimensions of the new scale comparable with those

323414-42- -7

of the Ventzke Scale. Correcting for the change in the specific rotation (-0.000184), the expansion of a glass tube (+0.000008), quartz wedge (-0.000130), and metal scale (-0.000018), the new weight was calculated to be 26.01+ g. It was thought advisable to ignore the small fraction and use the round number 26.00 as the normal weight, and the Commission officially so decided [14]. Owing to the absence of a more suitable term, and in order to divorce it as completely as possible from confusion with the Ventzke Scale, the new scale was referred to as the International Sugar Scale. The International Sugar Scale then was defined at the Paris meeting as follows: "The graduation of the saccharimeter shall be made at 20° C, 26.00 g of sucrose dissolved in water and the volume made up to 100 metric cc and polarized in a 200-mm tube. All weighings are to be made in air with brass weights, the completion of the volume and the polarization are to be made at 20° C. This will determine the 100° point."

The advantages of the new scale were at once appreciated. It has been adopted by the National Bureau of Standards, the United States Treasury Department, the Physikalisch-Technische Reichsanstalt, the Institut für Zucker-Industrie and also by the makers of saccharimeters.

(1) HERZFELD-SCHÖNROCK VERSION. Following this meeting of the Commission, Herzfeld [15] and Schönrock [16] (1901-04), on the basis of the above definition of the International Sugar Scale, standardized a number of quartz control plates on the saccharimeter and then measured their optical rotation for sodium light. They found, as the average of 10 plates, that a quartz plate which read 100° on the quartz-wedge saccharimeter with white light, filtered through bichromate, read 34.657° on a circular-scale polariscope for sodium light. This value, 34.657, became known as the Herzfeld-Schönrock conversion factor. By the use of this factor, optical rotations of quartz plates determined with sodium light could be converted into saccharimeter degrees and the standardization of saccharimeters effected without resorting to the use of a pure sucrose solution, which is always difficult to prepare. This version of the International Sugar Scale-namely the one which for all practical purposes had its 100° point set by the rotation in the saccharimeter of a quartz control plate of such thickness that its rotation for sodium light was 34.657has frequently been referred to as the Herzfeld-Schönrock scale. This scale was the practical formulation of the International Sugar Scale as defined at the meeting in 1900 of the Commission and remained in practically world-wide use for many years.

(2) BATES-JACKSON VERSION.-At the meeting of the International Commission for Uniform Methods of Sugar Analysis in 1912 in New York, Bates reported [17] that work at the National Bureau of Standards indicated that the Herzfeld-Schönrock scale was not quite correct. A committee was appointed to investigate the question and report on it at the next meeting. The war intervened and the scheduled meeting was not held.

In 1916 Bates and Jackson [18] published their work on the redetermination of the 100° point of the saccharimeter, wherein it was shown that a normal solution of pure sucrose read only 99.895° on the Herzfeld-Schönrock scale, and consequently the 100° point should be where the 99.895° point then was, and that the corresponding quartz control plate read 34.620 for sodium light and 40.690 for mercury green. Because there seemed little likelihood, due to international

conditions, that the Commission would again meet for some time to come, the values found by Bates and Jackson were officially adopted by the National Bureau of Standards and the United States Treasury Department without waiting for the reconvening of the Commission, and were used in all of their subsequent work. Their lead was followed by makers of saccharimeters and by others generally throughout the sugar world. This version of the International Sugar Scale became widely known as the Bates-Jackson scale. The publication of this work stimulated great activity in the field, and although its validity was at first questioned in certain quarters, yet later, as more and more data were accumulated by world-wide investigators, it was found to be correct. In fact, when the Commission finally met in 1932 in Amsterdam [19], after a lapse of 20 years, it was found that the average of all the values determined by investigators of international prominence working in various parts of the world was exactly that found by Bates and Jackson and reported in 1916.

(3) AMSTERDAM VERSION (1932) [19]. At this meeting in 1932 the Commission adopted, under subject 1, the following resolutions: (a) It is recommended that this Commission adopt a standard scale for the saccharimeter and that this scale be known as the "International Sugar Scale.” Rotations expressed in this scale shall be designated as degrees sugar ( S).

(b) It is recommended that the polarization of the normal solution (26.000 g of pure sucrose dissolved in 100 ml, and polarized at 20° C in a 200-mm tube, using white light and the dichromate filter as defined by the Commission) be accepted as the basis of calibration of the 100° point on the International Sugar Scale. (c) It is recommended that the reading of the normal sugar solution on the Herzfeld-Schönrock Scale be accepted as 99.90° S.

(d) It is recommended that the following rotations shall hold for the normal quartz plate of the International Sugar Scale:

Normal Quartz Plate = 100° S=40.690° ± 0.002 (λ=5461 A) at 20° C. 1° (X=5461 A) == 2.4576° S.

Normal Quartz Plate = 100° S=34.620° ± 0.002 (λ=5892.5 A) at 20° C. 1° (X=5892.5 A)=2.8885° S.

} (17) }(18)

(e) It is recommended that the Commission suggest that new saccharimeters be graduated in accordance with the International Sugar Scale and be inscribed by the manufacturers with the phrase, "International Sugar Scale.' In the case of existing instruments graduated on the Herzfeld-Schönrock scale, it shall be permitted either to change the saccharimeter scale or to use a weight of 26.026 g in 100 ml.

(f) It is recommended that the method of purification of sucrose for use in fixing the 100° point on the saccharimeter scale, which was adopted at the Paris session of the Commission (in 1900), be subjected to further study.

(g) It is proposed that the recommendations (a) to (e) shall come into effect on September 1st, 1933.

At first reading it might appear that (b), (c), and (d) are three different definitions of the 100° point which might or might not be concordant. A little consideration, however, will show that (b) is a general restatement of the fundamental basis of the saccharimeter scale, while (c) is a recognition of the essential correctness of a particular numerical value which had been arrived at through the application of the general definitions of (b) to the then existing sugar scale. (c) thus transfers the general definition (b) to the actual physical instruments used for sugar testing and is the concrete physical expression, as determined by experiment, of the fundamental definition given in (b).

In (d) there is set up an equivalent secondary or working standard, based upon the numerical value in (c), whereby saccharimeter scales may be checked or standardized without recourse to the difficult procedure of preparing and using a normal solution of pure sucrose. In fact, having once determined the correct quartz control-plate equivalent (conversion factor), the original definition of the saccharimetric scale could be lost or discarded without affecting the graduation or standardization of saccharimeters. The 100° point could at any time be set or checked by the simple expedient of using a quartz control plate whose absolute rotation is 34.620° for sodium light or 40.690° for mercury light of wave length 5461 A. Since the values of such plates are not subject to change, they are far more convenient for the calibration and checking of saccharimeter scales than the normal solution of sucrose.

It is thus seen that (d) is the practical expression of (b) and (c) in terms of the absolute rotation of a quartz control plate and supplies the means whereby saccharimeter scales may be readily duplicated or checked.

(4) CORRECTION OF SACCHARIMETERS TO THE INTERNATIONAL SUGAR SCALE. In Resolution 1(e) adopted by the International Commission for Uniform Methods of Sugar Analysis in 1932 in regard to the method of correcting saccharimeters to the new scale [19], two alternatives are provided: (1) ". . . it shall be permitted either to change the saccharimeter scale or" (2) "to use a weight of 26.026 g in 100 ml." The second alternative, namely changing the normal weight by a slight amount, is objectionable from the standpoint of introducing additional complications and is therefore to be discouraged.

The first alternative is much to be preferred, namely to change the saccharimeter scale. This does not mean that the existing scale must be removed and regraduated or replaced. It is sufficient to change the scale by the simple procedure of recalibrating it in terms of the desired scale by the use of standardized (International Sugar Scale) quartz control plates and applying the small scale corrections so obtained to all polarizations made in the subsequent use of the instrument. It is worthy of note that scale corrections resulting from inaccuracies in the wedges and other optical parts, including residual inaccuracies in the scale itself, may in many instances be nearly as large as the corrections referred to above and for accurate work must be taken into account either by the use of a calibration table or chart or by the use of a standard quartz control plate. No additional inconvenience is involved, therefore, if the corrections due to change of scale are included with those due to residual inaccuracies in the construction of the instrument.

3. ADDITIONAL CONSTANTS OF THE QUARTZ-WEDGE

SACCHARIMETER

(a) ROTATION RATIOS FOR QUARTZ AND SUCROSE SOLUTIONS

The ratios of the rotations in circular degrees of quartz and of sucrose solutions for two wave lengths have been determined as follows: [18]

[blocks in formation]
« PreviousContinue »