washings being poured through a weighed Gooch crucible. As little of the precipitate as possible is transferred to the filter. When the washing is completed, the precipitate is transferred to a weighed picnometer, which is filled to the mark and weighed. The difference between the first and final weighings gives the total weight of the lead precipitate.

The density of the precipitate is found in the following way:
Let c weight of precipitate transferred in decanting,

A weight of water in picnometer when filled,
B=weight of water and precipitate in picnometer,

C=weight of precipitate in picnometer found by difference be-
tween second and third weighing of Gooch crucible,
C+W total weight of precipitate, and

Density=C/A—(B—C).

The total volume of the precipitate is then its total weight divided by its density, thus

Volume=W/D=C+C/[A— (B—C)].

If care is taken to avoid any considerable loss of precipitate during the decantation, the determination may be shortened by neglecting the small quantity of precipitate lost in this way. The washed precipitate may be transferred directly to the picnometer, which is filled and weighed. The picnometer is then emptied directly upon a weighed Gooch filter. The volume of the precipitate is then the weight of water displaced in the picnometer by the precipitate, or A-(B–C). If the weight of the precipitate lost in the decantation does not exceed a few percent, the shorter method is satisfactory.

Horne dry-lead method.-A method intended to avoid rather than correct for the effect of the volume of the lead precipitate has been proposed by Horne [5]. Dissolve the sample in water and make up to 100 ml before adding the clarifer. Add a minimum quantity of dry basic lead acetate until sufficient clarification is obtained. Or, as in careful work with a lead solution, add the solid in successive small amounts until precipitation is almost complete. It is evident that it is necessary to stop short of complete precipitation because an excess of the solid, which does not produce a corresponding precipitate, serves to swell the volume of solution and a corresponding error is introduced. Horne has been able to show that by this method the volume of the solution is very approximately that indicated.

(4) EFFECT ON SUCROSE.-It is often erroneously stated that basic lead acetate has no effect on the rotation of sucrose. The experiments of Bates and Blake [9] show that errors in rotation caused by excessive amounts of basic lead acetate solution are of equal importance to the other errors in saccharimetry. These authors have found, figure 35, that an excess of 2 ml causes a diminution of polarization of 0.10° S; 1 ml, 0.12° S; 2 ml, 0.11° S; 3 ml, 0.09° S. The rotation reaches a minimum value when an excess of 1 ml is present. It returns to its initial value when 6 ml in excess has been added and continues to increase linearly with the amount of lead solution added. If as much as 50 ml is present, the rotation is then increased by a whole degree Ventzke. This source of error is avoided if the minimum quantity of lead solution necessary to clarify is added.

(5) EFFECT ON LEVULOSE.-By the action of basic lead acetate on levulose, the direct polarization may be considerably disturbed. This effect may occur from two causes. A soluble combination of lead and levulose may be formed which has a lower specific rotation than levulose or the lead-levulose compound may be actually precipitated from solution. The result in either case is an increase in dextrorotation or a higher polarization. Prinsen Geerligs has shown that basic acetate of lead precipitates levulose when the same solution contains salts which are capable of producing insoluble compounds with lead.

FIGURE 35.-Influence of lead acetate on normal sugar solution.

The combinations between lead and levulose are very easily broken up by a slight acidification. Acetic acid is sufficiently effective, but many other acids have been used for this purpose. Sulfur dioxide, tannic acid, and, as is frequently claimed, a solution of alum is acid enough to decompose this rather loose combination. In any case, only a slight excess of acid should be present.

(d) BASIC LEAD NITRATE (HERLES SOLUTION)

Herles solution [10] is prepared by dissolving 100 g of solid NaOH in 2 liters of water, and a second solution is prepared by dissolving 1 kg of neutral lead nitrate in 2 liters of water. Upon mixing equal volumes of the two solutions, basic lead nitrate is precipitated, the reaction being expressed by the equation

2Pb(NO3)2+2NaOH=Pb(NO3)2.Pb(OH)2+2NaNO3.

The precipitate is washed free of soluble impurities and mixed with water to a cream for use in clarification. The clarification may also be performed by forming the basic lead nitrate in the solution to be polarized. This is done by first adding a measured quantity of the above lead nitrate solution, 1 to 10 ml, according to the darkness of the sample, and then, after mixing, adding an exactly equal quantity

of the alkaline solution. An excess of alkali must be avoided. The mixture is then shaken and made to volume. The latter procedure gives the better clarification but introduces a considerable quantity of soluble salts, which may affect the polarization. The defects of the basic nitrate are, in general, those of the basic acetate. The volume of the precipitate is even greater because of the bulk of the solid clarifier. The precipitation of reducing sugar is even more marked than in the case of the basic acetate.

(e) HYPOCHLORITE (ZAMERON METHOD)

The hypochlorite solution is prepared by grinding 625 g of dry bleaching powder in a mortar with 1 liter of water. The mass is squeezed out in a sack and the extract filtered through paper. The filtered solution, about 700 to 800 ml of about 18° Baumé, is preserved in a stoppered dark-glass bottle in darkness. To perform the clarification, a few milliliters of the hypochlorite solution is added to the sugar solution and then a few milliliters of neutral lead acetate solution. The reagent usually causes a slight rise in temperature so that the solution should be readjusted to the temperature of the polariscope before making to volume. This method of clarification is very effective, and if no great excess of the reagent is employed, the reducing sugars are unaffected. The volume of the precipitate, which is increased because of the presence of insoluble lead chloride, is the main fault of this method.

(f) HYDROSULFITE

Sodium hydrosulfite is prepared by the reaction of zinc, sodium bisulfite, and sulfuric acid according to the formula

2NaHSO2+Zn+H2SO1=ZnS2O1+Na2SO4+2H2O.

The zinc hydrosulfite is changed to the sodium compound by the reaction

ZnS2O1+Na2CO3=Na2S2O4+ZnCO3.

The sodium hydrosulfite is salted out from solution by means of sodium chloride and dehydrated by warming with strong_alcohol. The compound is then dried in a vacuum at 50° to 60° C. This substance is produced commercially under the names of Blankit and Redo. It is frequently used in sugar manufacture for bleaching massecuites and, in dissolved form, as a wash for whitening sugar in centrifugal machines. To prepare a solution for polarization, a quantity of alumina cream is added and then a few crystals of hydrosulfite, 0.1 to 1 g, according to the color of the solution. The solution is made up to volume, shaken thoroughly, and filtered. As the clarified solutions occasionally redarken, they should be polarized immediately. The clarifying action, according to Weisberg [11], is due to free sulfurous acid and nascent hydrogen. The reduction by the latter leaves compounds which may be reoxidized and cause a redarkening of the solution.

Another hydrosulfite derivative (sodium sulfoxylate-formaldehyde) known as Rongalite accomplishes a permanent clarification but it is slower and less effective than Blankit.

903232 O-50-10

The defects of hydrosulfite as clarifiers are, in addition to the frequent redarkening, their effect on reducing sugars, the possible separation of finely divided sulfur, and their ineffectiveness in discharging the color of caramel bodies. Bryan [12] states that the rotation of dextrose is lowered by hydrosulfites and finds evidence of the formation of a laevo oxysulfonate. No immediate effect is observable upon sucrose or fructose, but sucrose is apparently inverted by a prolonged action. These clarifiers have not come into general use in analytical work, but nevertheless they are unique in that they produce no volume error.

(g) BONE CHAR

In cases where neither alumina cream nor lead subacetate is capable of producing a clear solution, recourse may be had to bone black. Bone black, for analytical purposes, may be prepared by treating the granular material used in sugar refining with a slight excess of hydrochloric or nitric acid until all of the mineral matter is dissolved. The treated char is washed with boiling water, dried at 120°, and finely powdered and bottled. The more completely the material is freed from mineral matter, the more effective is its action for analytical purposes. Bone black probably owes its clarifying action to the very large surface which is caused by its porosity.

The most serious error accompanying clarification with bone char is caused by its tendency to absorb sugar and thus give abnormally low readings. For this reason, most official methods of clarification exclude bone black as an agent. It is difficult to make a correction for the amount of sugar absorbed, because it varies with the composition and concentration of the sample and the condition of the bone char. In order to avoid the error arising from the absorption of sugar, the absorption coefficient may be determined under the approximate conditions of the analysis or the solution may be made up to volume and filtered through a column of bone black, the first third of the filtrate being rejected.

4. REFERENCES

[1] G. H. Hardin and F. W. Zerban, Louisiana Planter 73, 388 (1924). [2] F. J. Bates and F. P. Phelps, Bul. BS 10, 537 (1914) S221.

[3] Official and Tentative Methods of Analysis of the Association of Official Agricultural Chemists, 5th ed., p. 490 (1940).

[4] R. F. Jackson, Bul. BS 11, 331 (1914) S232.

[5] W. D. Horne, J. Am. Chem. Soc. 26, 186 (1904).

[6] F. Sachs, Z. Ver. deut. Zucker-Ind. 30, 229 (1880).

[7] G. L. Spencer and G. P. Meade, A Handbook for Cane-Sugar Manufacturers and Their Chemists, 7th ed., p. 224 (1929).

[8] C. Scheibler, Z. Ver. deut. Zucker-Ind. 25, 1054 (1875).

[9] F. J. Bates and J. C. Blake, Bul. BS 3, 105 (1907) S52.

[10] F. Herles, Z. Zuckerind. Böhmen 13, 559 (1888); 14, 343 (1889); 21, 189 (1896).

[11] J. Weisberg, Centr. Zuckerind. 15, 975 (1906).

[12] A. H. Bryan, Bul. Bur. Chem. No. 116.

VIII. CLERGET METHOD

1. INTRODUCTION

The direct polariscopic reading of a sugar solution is the resultant rotation of all optically active substances present, and is conse

quently a correct measure of the sucrose only when the other substances present have no effective rotatory power. If other optically active substances are present, the direct polarization must be supplemented by a second observation in which the rotations of these substances are kept constant while that of sucrose is subjected to a change which can be measured and is known to be an exact function of the quantity of sucrose. This change is brought about by the hydrolysis of sucrose to invert sugar. The change of rotation of the normal solution of pure sucrose is known as the Clerget divisor. The divisor is not a constant but its numerical value is influenced by concentration, temperature, and impurities. The hydrolysis or inversion can for analytical purposes be effected by either the enzyme, invertase, or by hydrochloric acid.

In its simplest form applicable to the ideal case, where nothing but the rotation of sucrose is altered, the Clerget formula is

in which P and P' are the direct and invert polarizations, respectively, of the normal solution; C, the basic value of the Clerget divisor at 20° C; ac, the change in the value of the divisor with concentration of substance; and b, its change for each degree rise in temperature.

The Clerget formula is frequently applied in the above form without assurance that the fundamental condition is fulfilled, namely, that the rotations of all substances except sucrose remain unaltered in the two polarizations. In general, it is true that for samples containing high percentages of sucrose and small quantities of invert sugar the method yields reliable results and that for low-grade products it yields results which are sometimes sufficiently accurate for the purposes at hand.

Of the two commonly employed hydrolytic agents, invertase is the superior because of its highly selective action on the sucrose group and because it is without effect on the rotations of other substances occurring as impurities in sugar samples. Its disadvantages are its relatively high cost, the considerable labor required for its preparation, the uncertainty that the preparation has retained its activity, and, except under certain conditions, the long time required for the completion of the hydrolysis.

Hydrochloric acid, on the other hand, involves negligible expense and is capable of completing the hydrolysis in any desired period of time by merely regulating the temperature of reaction. However, it is not selective but hydrolyzes any glycosidic group. Moreover, it influences the rotatory power of invert sugar and many other impurities occurring in natural products.

2. ACID METHODS

(a) BASIC VALUES OF THE CLERGET DIVISOR

In devising the method, Clerget [1] in 1849 took 50 ml of a normal sucrose solution in a 50- to 55-ml flask, added 5 ml of "pure and fuming" hydrochloric acid, and after mixing, placed the flask in a water bath so regulated that 10 minutes were required to raise the temperature of the solution to 68° C. Upon attaining this temperature the flask was removed, cooled rapidly to 20°C, and the solution polarized.