The moisture remaining will in general amount to less than 0.003 percent.

TABLE 44.-Time required at various temperatures to form caramel equivalent to 0.01 percent of invert sugar

Inulin can be prepared most satisfactorily from dahlia tubers, or in somewhat lower yield from chicory, or in much lower yield from jerusalem artichokes. The most suitable time for the extraction is early autumn, but in the case of dahlias the large clumps of tubers are usually divided for replanting in the late spring and the "blind" tubers which are unsuitable for planting are then available for inulin extraction. Even as late as early summer inulin can be obtained from healthy tubers.

Comminute the tubers or roots in a food chopper or similar appliance and express the juice with a tincture press, using, if necessary, a small proportion of water to complete the extraction. Heat the juice to 60° or 70° C and add milk of lime to about pH 8. Filter and adjust the pH to 7.0 with oxalic acid. Heat to 70° to 80° C, add active carbon, and filter. Allow the filtrate to stand quiescent overnight, during which time the inulin separates in the form of small spheroids. Slightly increased yields can be obtained by placing the filtrate in a refrigerator and still greater yields by completely freezing it and allowing it to thaw at low temperature. Filter the separated inulin with suction and wash with abundant quantities of cold water.

The moist cake of inulin has at this point a dry-substance content of 35 to 40 percent. It can be purified by dissolving in hot water to a concentration of about 15 to 18 percent, treating with carbon, refiltering, and allowing to separate again by chilling. Filter and wash with cold water.

For many purposes it is a satisfactory procedure to spread the moist inulin cake on a glass plate and allow it to dry in air. When dried in this way, it takes the form of hard hornlike masses which can be pulverized readily in a mortar or ball mill. The air-dried substance contains about 10 percent of water of hydration and a small quantity of inorganic impurities. The latter can be diminished to negligible proportions by repeating the processes of solution and chilling, or by electrodialysis.

To avoid the formation of hornlike masses, the moist inulin cake can be washed with 95 percent alcohol, followed by absolute alcohol. The substance, however, has a tendency to retain alcohol [7], which conceivably displaces in part the water of hydration.

A procedure described by Tollens and Elsner [8] is probably suitable for less pure materials than fresh dahlia juices. The juice is expressed

in the presence of calcium carbonate, fermented at 25° C. with bakers' yeast, defecated with lead acetate, and filtered. After removal of the excess lead with hydrogen sulfide, the filtrate is frozen and thawed to cause the separation of inulin.

Inulin crystallizes in the form of doubly refracting sphero-crystals whose crystalline structure is indicated by X-ray patterns [9]. In aqueous solution inulin has a specific rotation, [a] = -39 to -40 referred to anhydrous substance.

If heated in glycerine, or glycol, or even in water solution and precipitated from these solutions by alcohol, inulin occurs in a form soluble in cold water [7, 10]. This soluble form is also produced by the action of carbon dioxide [11].

(b) PREPARATION OF LEVULOSE

Levulose can be prepared from inulin, or invert sugar, or directly from the expressed juices of inulin-bearing plants. The sugar can, with some difficulty, be crystallized directly from hydrolyzed inulin, but from the two latter sources it must, after hydrolysis, be isolated by means of the insoluble calcium levulate, CaO.C,H120.xH2O.

(1) HYDROLYZED INULIN.35-To 100 g of hydrated inulin add 400 g of water and agitate, preferably with a motor-driven stirrer, until the inulin is "swollen" to a uniform paste. Heat to nearly boiling and allow the inulin to dissolve completely, adding more water if necessary. Reduce the temperature to 60° C and make the solution 0.08 N with sulfuric acid. Maintain the solution at 60° for 2%1⁄2 hours, preferably following the hydrolysis by polariscopic observation of samples withdrawn from time to time. Neutralize by agitating vigorously with barium carbonate in excess or by cooling and titrating to exact neutrality with barium hydroxide, preferably leaving the solution very slightly acid (that is, acid to bromthymol blue). Filter with carbon and evaporate in a vacuum to a thick sirup. Add absolute alcohol and evaporate again to a sirup to remove the remaining water. Extract the sirup with several portions of hot absolute alcohol. Allow the combined extracts to stand for 12 to 24 hours and decant the solution from the sirupy residue which forms. Inoculate with levulose crystals and allow the crystallization to become complete (usually within 2 or 3 days). The yield of sugar is usually relatively small.

(2) PREPARATION OF INVERT SUGAR.-Dissolve 250 g of pure cane sugar in 846 ml of water and heat to 70° C in a water bath. Add 20 ml of 5 N hydrochloric acid, mix, and allow the solution to remain at this temperature for 35 minutes. Cool and neutralize the solution with dilute sodium hydroxide, avoiding local alkalinity. The solution whose volume can be measured now contains 263.16 g of invert sugar, one-half of which is levulose.

(3) HYDROLYSIS OF PLANT JUICES.-Juices extracted from inulinbearing plant tubers or roots contain from 12 to 20 percent of dry substance depending upon the composition of the original juice and upon the quantity of maceration or diffusion water used for extraction. The polysaccharides contained in the juice are hydrolyzed,

as This method is essentially that of M. Hönig, S. T. Schubert, and L. Jesser [Monatsh. 8, 529 (1887) 9,562 (1888); Ber. deut. chem. Ges. 20, 721 R and 21, 663 R]; and Wohl, [Ber. deut. chem. Ges. 23, 2084 (1890)) but modified by the substitution of more definite directions for the hydrolysis of inulin (Jackson and Goergen, BS J. Research 8, 29 (1929) RP79].

preferably with sulfuric acid. It has been found impracticable to supply exactly defined conditions for this hydrolysis, since varying types of juice require variations of conditions. These variations arise from the fact that the acid is in great part rendered inactive by the buffering action of certain impurities in the juice. Even when the pH of different types of juice is made apparently the same, differences in the rate of hydrolysis occur. It is, therefore, necessary to follow the hydrolysis by means of polariscopic or reducing-sugar analysis of samples withdrawn from time to time. Typical hydrolysis data are given in table 45, in which N is the normality of sulfuric acid calculated from the amount of acid actually added; CH+, the normality of hydrogen ion calculated from the pH measurement (quinhydrone electrode); k, the velocity constant in terms of common logarithms and minutes; and "time 99 percent," the time required for 99-percent hydrolysis.

The last two measurements were made on the same juice, and it is evident that the velocity is not proportional to the apparent hydrogenion concentration, the explanation probably being that the pH measurement is subject to errors caused by impurities in the juice. No difficulty is encountered, however, if the hydrolysis is followed analytically. The time required for 99-percent hydrolysis can be diminished at will by raising the temperature of reaction.

When the hydrolysis is complete, the solution is cooled to room temperature or below and treated with milk of lime to pH 7 to 8 with vigorous agitation and finally filtered.

The filtered juice is analyzed for total reducing sugar and levulose. In precise work the Lane and Eynon titration and a method selective for levulose are used (see p. 205). In approximate work, which is satisfactory in most cases, the Lane and Eynon titration method and the direct polarization are used (see the Mathews formula, p. 223).

(4) PRECIPITATION OF CALCIUM LEVULATE [12].-Prepare a milkof-lime suspension, magnesia-free, of 18 to 20 percent calcium oxide content and calculate the volume or weight required to contain an amount of calcium oxide equal to 45 or 50 percent of the levulose in the juice. Dilute either the juice or milk of lime so that when the equivalent quantities are mixed, the levulose shall be approximately 6 percent of the total mixture.

The details of procedure for precipitating the calcium levulate have been directed to the end of obtaining relatively large discrete crystals rather than the fine interlacing needles produced in the original method of Dubrunfaut. The precipitation preferably is made in a vessel equipped with a motor-driven stirrer and immersed in a cold bath. The following procedure illustrates the principal features of the

method and is subject to obvious variations with respect to details. Transfer to the precipitation vessel about one-tenth of the total quantity of milk-of-lime suspension and add one-tenth of the total volume of sugar sirup, allowing the latter to run in at a slow regular rate. It is advisable to examine the precipitate under the microscope from time to time and ascertain that the calcium levulate has formed discrete crystals. When properly prepared, they take the form of elongated prisms. Add successively the remaining portions of milk of lime, each being followed by the respective portion of sirup. Under favorable conditions the levulate crystals will be 0.1 to 0.2 mm in length and can be readily and rapidly separated from the waste water by filtration methods.

The yield of levulose by this process is dependent upon many circumstances. In order to obtain high yields, the temperature of precipitation must be in the neighborhood of 0° C. Increased temperature causes a considerable diminution in the yield of levulose. Thus, in a series of experiments on hydrolyzed artichoke juices, Jackson and Mathews [13] obtained a recovery of 82.3 percent at 3.2° C. 76.2 at 10.7° C, and but 64.0 at 16.0° C.

Typical quantitative data on the precipitation of calcium levulate from various juices are shown in table 46.

It will be observed that the "Total recovered" of each product (lines 16, 22, and 28) approximates the amount taken in the reaction mixture (lines 12, 18, and 24). The yield of levulose (line 33) varies inversely as the solubility of calcium levulate in the juice. The solubility of calcium levulate is greatly increased in the presence of nonlevulose sugars. Thus, in experiment 1, the levulate in the waste. water, expressed as levulose (line 13), is held in solution largely by the dextrose. In the remaining experiments, it is dissolved both by dextrose and by other nonlevulose sugars.