The contents of the casserole are continuously stirred until the sugar has dissolved, and the stirring rod is removed. If the size of the flame has been properly adjusted the solution should start to boil in about 5 minutes after being first put over the flame. At this point an inverted watch glass is placed over the casserole; otherwise the sugar will be apt to crystallize as the evaporation proceeds.

After the heating has continued precisely 15 minutes from the time when the casserole was first placed over the flame, the watch glass is removed and the solution is then thoroughly and constantly stirred without a moment's interruption until the boiling point has reached exactly 350° F (177° C), the thermometer being used as a stirring rod. The casserole is than instantly removed from the flame and its contents are as rapidly as possible emptied upon a polished copper slab 14x14x4 inches in size. In a few minutes the candy becomes brittle and can be broken up for any tests it is desired to make.23

The Hooker procedure is deficient chiefly in reproducibility. Experiments conducted at this Bureau indicate that most of this variability can be eliminated through the use of more definite specifica

COPPER NO.168.W.G.

OAK

FIGURE 89.-Copper casserole as used in Hooker's procedure.

The drawing is adapted from the illustration presented by Murphy [5].

tions for the apparatus and procedure and by the omission of handstirring of the boiling sirup [4]. As a result of this study, a method has been developed which is designated the "National Bureau of Standards simple barley sugar test.'

2. NATIONAL BUREAU OF STANDARDS METHOD FOR BARLEY SUGAR

The Bureau method varies from the Hooker procedure in several respects, as summarized later in this discussion. All of these modifications are important to the convenience and uniformity of the test procedure, as also to the precision and reliability of the results, especially as obtained by different operators, and particularly as obtained in different laboratories. The following is the improved procedure.

22 Most candy and sugar technologists regard commercial sucrose as being the "weaker" the greater the degree of hydrolysis which it undergoes upon conversion into hard candy under standardized conditions [2, 4, 5, 9, 10, 11]. Hooker's procedure originally was intended to indicate, through direct polarization of the resulting candy project, the relative "strength" of any lot of sugar, conceived as the apparent resistance of the sugar to hydrolysis under such conditions. In Hooker's laboratory no other tests of the candy product were made. In later development of candy test methods, various other observations have been applied to the candy product as criteria of the quality of the sugar. When Osborn [2], in 1912, introduced candy test methods as a means of control of quality in beet sugar production, substituting porcelain casseroles for those of copper, he immediately gave heed to the rates of crystallization of the individual candy products and also devoted special attention to the color of the candy and to the tendency of the sirupy mixture to "foam" upon first coming to a boil in the candy test. He finally omitted the direct polarization in most cases. Empirical means of roughly estimating in quantitative form, the tendencies of the sugar to cause foaming and caramelization were developed somewhat later by Proffitt [4].

(a) PROCEDURE

1. Weigh 250.00 +0.05 g of the sugar sample on a dry basis and make up to a weight of 350.00 +0.05 g with cold equilibrium water 24 in a 600-ml chemical-resistant glass beaker (provided with a thermometer clasp, as specified on p. 375) previously tared with the thermometer and the notched watch-glass cover.

2. With the thermometer used as a stirring rod, loosen the compacted crystals from the bottom of the beaker, and mix them evenly with the liquid to form a homogeneous suspension. Continue the stirring until the crystals easily remain in suspension for some time, or until the temperature has ceased falling.

3. Replace the thermometer in its clasp (to avoid injury of the thermometer and the dripping of sirup), and set the beaker upon its supports in the stove with the burner alight and with the flame in correct stable adjusment. Immediately resume stirring of the mixture with the thermometer to prevent settling of the undissolved crystals.

4. On the instant the starting temperature of 30° C (86° F) is indicated, note "zero time" in the heating cycle (as for example, by the starting of a stop watch or a second counter).25 Proceed with the stirring until a temperature of 70° C (158° F) is indicated, when crystals of fine or medium granulated sugar ordinarily should be practically all dissolved.26 Clasp the thermometer at its proper working level, which will bring the bottom of the bulb to the position indicated in the drawing (fig. 90). Immediately cover the beaker with the watch. glass as indicated in the drawing. Observe the sirup closely as the boiling point is approached 27 for its behavior during the transition interval between quiescent heat absorption and steady boiling is an important criterion of the probable stability of the sugar in storage as well as in candy making. There should be no considerable increase in the apparent volume of the sirup, not more than a trace of foamy scum should collect on the surface, and the transition interval should last not more than 30 seconds.28 Steady boiling usually is barely established when the thermometer indicates a temperature about 6° C (11° F) higher than it would indicate in boiling water in the same beaker. This is assumed to be the initial boiling temperature of the sirup.

5. Allow the sirup to boil under cover until its indicated temperature is just 120° C (248° F) [4] (which in a 23-minute cooking interval occurs approximately 15 minutes after the start of heating). On the instant this temperature is indicated, lift the watch glass at one edge and tip it to drain most of the adhering condensate down the side of the beaker into the boiling sirup. Do not disturb the sirup in any other manner. Place the wet watch glass into a drying oven at a temperature of about 105° C (221° F) without washing, for it may retain a small amount of sirup.

24 Equilibrium water is defined on p. 271.

25 If temperature-gradient data are desired, record elapsed time against a previously prepared schedule of temperatures. Such gradients are a valuable means of interpreting results in certain cases.

26 If grains still can be felt with the thermometer or can be seen, remove the beaker from the stove and proceed with the stirring without further heating until the refractory grains have disappeared. Return the beaker to the stove, and stir the sirup again until the temperature of 70° C is regained. Do not include the time consumed in these operations in the recorded duration of the cooking interval.

27 Remove the watch glass during the transition interval only if dew formation on the undersurface interferes with observation, or if its removal is required for the measurement or the quelling of foam. (See b (1), p. 373).

28 If a layer of sudsy foam accumulates on the surface of the sirup, proceed as indicated under b (1), p 373.

6. On the instant a temperature of precisely 176.7° C (350° F) is indicated, remove the beaker from the stove, simultaneously noting the time (as by the stopping of the timer device), and dump its contents all at once onto the center of the cooling slab. This operation should be performed in such a manner as to form a nearly circular plaque. When the stream breaks, turn the beaker to an angle of about 45° and allow the sirup to drain into the center of the plaque for 102 seconds. Lift the beaker with a turn which brings the threads of sirup into its mouth, and set it upon a quenching sheet of 4-inch aluminum or copper to prevent charring of the adhering sirup in the bottom of the beaker. Cover the beaker with the dried watch glass now taken from the oven and, when cooled to room temperature, weigh the whole assembly.

7. Five minutes after the pouring of the candy, strip the plaque from the water-cooled slab and place it in a desiccator or dry-air closet until it has cooled to room temperature. When the plaque is cool, proceed as indicated under b (2), below.

(b) OBSERVATIONS FOR THE INTERPRETATION OF RESULTS

(1) FOAM NUMBER.-When a layer of sudsy foam accumulates upon the surface of the sirup as it approaches the transition interval, the apparent volume of the mixture probably will expand suddenly at initial boiling [2, 4]. To obtain a roughly quantitative statement of the phenomenon, proceed as follows:

Observe the number of seconds elapsing from the instant the rapid expansion begins until the moment the foam subsides to its minimum volume, and record the result as "duration of foaming." Observe also the distance in centimeters from the surface level of the quiescent, nonfoaming sirup to the highest level reached by the general surface of the foam on expansion, convert the result to percentage of increase in volume, and record the number as the "volume of foam." The product of these two numbers is the "foam number" [4, 6].

Space is available in the beakers for the accommodation of but little more than a doubling of the apparent volume of the sirup with the specified size of sample. If the expansion much exceeds this value, at the standard rate of heating, it may result in the overflowing and loss of a portion of the sirup, with consequent fouling of the stove and burner. If this misfortune threatens, reduce the rate of heating; do not disturb the flame but hold the vessel by hand at a greater distance above the flame, or hold it intermittently over the flame until all danger of loss is past. Do not include any prolongation of the heating which is involved in the quelling of foam as a part of the cooking interval.

(2) EXAMINATION OF THE PLAQUE.--Notice the general appearance of the candy, especially with respect to color, transparency, inclusion of bubbles and specks, etc., and state the results in qualitative form. Weight of the plaque. When the plaque has cooled to room temperature, weigh it with a precision of ±0.05 g.

If

Projected area of the plaque.-Measure the projected area of the plaque, preferably by following the periphery of the disk itself with the stylus of a planimeter specially designed for such purposes. this means is not feasible, first trace the outline of the plaque upon paper and later measure the area enclosed by the tracing. Express the results in square centimeters.

Slump or specific area of the plaque.-Divide the number representing the projected area by the number of grams in the weight of the plaque, and record the quotient as the "slump", or specific area, expressed as square centimeters per gram. The numerical value of this quotient, under uniform conditions of testing, increases with increasing weakness of the sugar [13].

Mean thickness of the plaque.-Measure the thickness of the plaque at a minimum of 20 random points distributed evenly over the plaque, preferably by means of a dial-type thickness gage. In lieu of this, pieces of the plaque may be measured with an ordinary micrometer. The plaque is thicker the stronger the sugar from which it was prepared (with uniform conditions of testing) [13].

Color measurement of the solid candy.-Estimate the transmission of light of 560 mu wave length (or of three or more wave lengths, if preferred) at several clear points on the plaque or at clear points on several pieces. Compute the transmissions for layers of exactly 5 mm thickness. To minimize surface reflections, observe the pieces while freshly immersed in a water-clear, slightly supersaturated sirup of pure sucrose contained in a wide, open-top colorimeter cell of about 1-cm length (on the optical axis). Employ any suitable instrument (chap. XIX), and correct the results for any perceptible absorption of light by the immersion sirup [13].

Yield and overweight of the candy. The total yield, y, of the candy is the sum of the weight of the plaque and the weight of the candy adhering to the boiling-vessel, thermometer, and cover. The "overweight" in grams is the difference expressed by (y-wt of dry substance introduced), which, in all standard tests, is (y-250). It represents approximately the weight of the residual water retained by the candy, partly combined as water of hydrolysis, partly as solvent or solute in the supercooled solution. In standard tests it is expressed as a percentage of the sample weight simply by multiplying by the factor 0.4. Breaking of the plaque. The relative brittleness or toughness of the plaque can be estimated by breaking the plaque with a suitable impact device. Ordinarily, it is simply broken by hand. In either case, the pieces are placed promptly in a container, such as a glass-top mason jar which can be closed to protect the candy from the moisture of the atmosphere. If the candy is to be examined in aqueous solution, the container should be tared.

(3) EXAMINATION OF THE CANDY IN AQUEOUS SOLUTION.-From the net weight, w, of the candy stored in the container, and the yield of the candy per gram of the dry substance introduced, compute the weight of dry substance introduced (see above) equivalent to the contents of the container, as sw/y, where s represents the weight of dry substance in the sample (250.00 g in all standard tests). Add equilibrium water until the total weight of the contents is precisely double this quantity, i. e., 2sw/y. Cover the container securely and without the application of any heat, shake or rotate it, preferably by mechanical means, until the candy is completely dissolved. Use this stock solution for whatever tests or analyses are to be made on the aqueous solutions, keeping in mind the fact that 1 g of the original dry substance of the sample is represented by 2 g of the stock solution.

(c) APPARATUS

(1) BOILING-VESSELS.-Glass beakers, Griffin form, 600-ml capacity, of chemical resistant, low-expansion Pyrex or equal glass must be used as the boiling-vessels. Each beaker is to be provided with a notched watch-glass cover and with a quickly detachable combination handle and thermometer support with quick-opening spring-actuated clasp, as indicated in figure 90. For lightness and cleanliness, the handle and thermometer support are constructed mainly of aluminum. For uniformity of results, the thermometer bulb should be located precisely as indicated in the drawing, and the burner should be of precisely the form and arrangement indicated in figure 92. Since

Consists of 600-ml chemical-resistant-glass beaker provided with a notched water-glass cover, thermometer, and handle set carrying the thermometer support with spring-actuated clasp. The aluminum handle set is clamped onto the beaker by means of the hinged hoop bands which are secured by closing the handle shell, jackknifewise, over the straight protruding ends of the bands. It is retained at a fixed level on the beaker by the V-shaped wings which bear upon the sides of the lip of the beaker, and by the short hook which rests in the notch of the lip.

variations in the thickness of the beaker bottoms cause relatively large variations in the duration of the cooking interval with any given adjustment of the burner and gas supply, beakers which are to be used interchangeably on a particular adjustment of the heating equipment should be carefully matched as to weight and as to heat transmission. A counterpoise, preferably adjustable, should be provided for the taring of each assembly. Since the handle sets are so easily removed, and should be removed while the glass parts are being washed, all weighings may be made without the handle sets. Thus consecutive tests can be run without loss of time by providing a sufficient number of matched beakers and thermometers with but two handle sets.