Kervegant Chapter XVII Analysis of Distillery Materials:
Pages 429-452
CHAPTER XVII
ANALYSIS OF DISTILLERY MATERIALS (1)
(1) Spencer (G. L.) – Manual de fabricantes de azucar de cana y quimicos azucareros, New-York, 1922.
Deerr N.). Cane Sugar chap. XXV. London, 1921.
Determination of density
Determination of density can be done by three different methods: volumetric flask method, hydrostatic balance method, and hydrometer method. The latter is the most used in industrial practice, because of its simplicity and speed, although it gives results that are often less accurate than the first two methods.
Volumetric flask method.
The weight of the liquid to be tested is compared with that of an equal volume of water, by means of the pycnometer. The pycnometer, washed with water, alcohol, and ether, and well dried, is first weighed empty, then full of boiled distilled water, and finally full of sugar solution. By deducting from the weight of the full pycnometer that of the empty pycnometer, we have the weight in the air of the contents. Density is obtained by dividing the weight of the sugar solution by the weight of the same volume of distilled water.
The simplest picnometer is a small 50 cc bottle, but there are more advanced models (Sprengel, Boot, etc.). Boot’s, consisting of two concentric flasks between which the vacuum is made, is convenient for use in hot countries, because it has the advantage of preserving the cooled liquids, without any change of temperature and without condensation of water on the outer walls.
[What Kervegant goes into detail about are vacuum jacketed pycnometers. I’ve have a nice pycnometer collection, but I’ve never actually seen a vacuum version. New models are astoundingly pricey now that pycnometery is not commonly practiced. Luckily, density of sugar solutions may require fare less precision than that of alcohol determinations.]
For accurate results, it is important that the determinations be made at a given temperature: usually 15° C, temperature used in France, or 20° C, normal temperature adopted by the International Congresses for the determination of physicochemical constants. As these temperatures are not suitable in the tropics to operate quickly, N. Deerr advises to proceed as follows: Determine the weight of the water in the pycnometer for every tenth of a degree between 20 and 30°. Examine the sugar solution at the laboratory temperature and divide the weight obtained by that of the water at the same temperature. The error in admitting an equal expansion for water and sweet solutions between the above limits is very small and only occurs at the third decimal of the Brix degree.
Hydrostatic balance.
We compare the weights of a float taken in water and taken in the sugar solution.
The Mohr hydrostatic balance, modified by Westphal, is usually employed. This consists of a rod, supporting a flask with a counterweight and a needle. The right arm of the beam is divided into ten equal parts with numbered notches, intended to receive weights in the form of riders. At point 10 is a hook, to which is attached a platinum wire with a float equipped with a thermometer. This float plunges into a test tube containing the liquid to be tested; a knot made on the platinum wire indicates how far the thread should go down into the liquid. There are 4 riders: the biggest weighs as much as the weight hooked in 10 and intended to compensate for the loss of the float in the water; the other 3 weigh respectively 1/10, 1/100 and 1/000 of the first.
[There are great demonstrations of these devices on Youtube.]
For liquids heavier than water, the weight hooked in 10 being recognized as insufficient, the other weights are placed on the notches, starting with the heaviest, so as to bring the balance back. For liquids less heavy than water, remove the weight placed at 10 and operate with the other weights as above. We read the density directly on the balance. The decimal digits follow one another in the order of the hanging weights, the value of each being given by the number of the notch.
The pycnometer and Westphal balance give the ratio of two weights taken in the air at the temperature of the observation, 15° C for example. The specific weight with respect to the water is obtained at 4°, by applying the formula:
where P is the weight in the air at 15° of the sugar solution and P’ the weight of distilled water.
[Pretty sure you need this correction because the specific gravity of distilled water is only 1.000 at 4°C.]
Hydrometer method.
We observe the equilibrium position of an empirically graduated instrument, called a hydrometer, which is immersed in the liquid to be tested.
The densimeter-control, used in France, gives the densities at 15° C compared to pure water at 4° C and in a vacuum, or, in other words, the weight of one liter of liquid at 15° C (compared to vacuum). The hydrometer has divisions in degrees and tenths of a degree. If n is the regulated degree of a solution the actual density will be:
When the observation temperature is different from 15 °, it is necessary to correct the density, by means of a special table.
To obtain the density of thick liquids (syrups, molasses), the Baumé hydrometer is often used. This instrument, invented by Antoine Baumé in 1768, was the subject of various modes of graduation.
Originally, for the heavy Baumé, the 0 was placed at the upper part of the stem, at the point of outcropping in distilled water at the temperature of 12°5, and the point 15 at the outcrop in a 15% solution of sea salt. The distance between these two points was divided into 15 equal parts and continuously marked down the stem.
[Salt’s solubility is roughly 360 g/L (and over twice the density of water) so at 15% solution is pretty close to the saturation point.]
The Baumé aerometer currently manufactured in France marks 0 in a liquid having at 15 ° the same density as water at 4 °.
Point 66 is obtained by immersing the instrument in pure sulfuric acid monohydrate with a specific gravity of 15/4 ° of 1.8427. The intermediate space is divided into 66 equal parts. The modulus of the apparatus is 144.32, and one goes from the Baumé degree to the specific weight 15°/4°, by applying the formula:
[This is very good information about sulfuric acid monohydrate that was not readily apparent on the web elsewhere. This chemical is important to analysis techniques in previous chapters.]
The hydrometer adopted by the United States Bureau of Standards is graduated at 20 ° C with respect to water at 20 ° C and a module of 145.
[module or modulus refers to the absolute or highest value of the scale.]
Formerly widely used, the Baumé hydrometer, whose arbitrary graduation is meaningless for to saccharine richness of solutions, tends to be replace more and more, in sugar houses and distilleries, by saccharometers, which indicate the percentage of dissolved solids, assuming at the same time that all substances act on the density of the liquid as pure sucrose would do.
[Saccharometers are often used on stillage, but it is hard to say what those measures can consistently imply.]
This percentage of dry matter is generally referred to as Brix degree or, more rarely today, Balling degree. When determined by means of the saccharometer, it is apparent Brix and, when it is by oven drying, the actual Brix. Unless otherwise specified, the term “Brix degree” always applies to the apparent extract. In the case of pure sugar solutions, apparent brix and real brix are obviously identical.
The saccharometers have been subjected to various graduation modes. The devices used in Germany and America indicate the quantity (in grams) of dissolved solids in 100 grams of solution (Brix-weight). The normal temperature of graduation, originally 17°5, is currently 20°. In France, the hydrometers are graduated at 15°. The Dupont saccharometer indicates the Brix-weight and the Vivien saccharometer Brix-volume, that is to say the amount in gr of dissolved solids in 100 cc.
The dissolution of sugar in water gives rise to a concentration of the mixture, it was necessary to experimentally determine the relationships between densities and sugar weights of pure sugar solutions (work by Balling, Gerlach, Plato). There are various concordance tables (Brix, Gerlach, Schelbler, Plato, Saillard, etc.), showing some variations, depending on the normal temperature adopted and the experimental data from which the authors were inspired. Correction tables are used for the case where the temperature of the observation is different from the normal temperature of the apparatus used. The correction will be all the more exact as we get closer to the normal temperature.
We reproduce hereinafter the table of concordance between the densities and the richness of the sucrose solutions, established by the “Bureau of Standards” of the United States (1), for the normal temperature of 20° C and according to the experimental data from Dr. Plato (2). Note also that Sidersky (3), in order to meet the needs of sugar factories located in tropical countries, has recently calculated a table for the normal temperature of 28°/4°, also taking as a basis the work of Plato:
(1) U. S. Bureau of Standards. Circ. N° 44, p. 151, 1918.
(2) Wiss. Abb. der Kaiserlichen Normal-Eichungs-Kommission, 190. [seems to be missing last year of its date]
(3) Bull. Ass. Claim. LII, 432, 1935.
Degrees Brix, densities and Baumé degrees of sugar solutions
Brix degree correction for different temperatures from normal (20° C) (1)
(1) U. S. Bureau of Standards Circ. N° 19, p. 25, 1919.
Taking density of sweet solutions.
Certain precautions must be taken in the preparation of products whose density is to be determined.
Cane juice from mills carries solids (bagasse particles, earthy matter) in suspension as well as air in emulsion. It is important to filter it with gauze and let it sit for a while, to allow heavy materials to settle and light parts to rise to the surface. The density is taken on the intermediate liquid. A container having 2 or 3 times the volume of the specimen can be used for decantation when taking density or using a tap about 5 cm from the bottom.
Molasses, because of their viscosity, are often diluted with an equal volume of water, and the result is multiplied by 2. This way of proceeding is not very precise because, by mixing with water, the molasses undergo a greater contraction, under equal content in dry matter, than that of pure sugar solutions and variable from one molasses to another: the result is therefore a little higher than the direct specific weight.
One can also take the degree Baumé or Brix without dilution, operating at a relatively high temperature (between 40 and 60° C), and leaving the hydrometer for some time at rest before reading. But it is preferable, if one wants to obtain precise results, to use the following method, which gives the real density.
Sidersky Process — The first step is to remove air from the molasses by heating in a water bath about 100 g of molasses contained in a 150 cc bohemian vase. After having skimmed the molasses, pour it all hot into a 50 cc bottle, calibrated in advance, which is filled, by means of a long and obliquely cut up funnel, up to 1 or 2 cm below the line of measurement, taking care not to let the molasses run on the inside walls of the neck above the mark. After filling, the bottle is cooled and weighed: by subtracting from the gross weight the tare of the bottle, we have the weight of the molasses. The flask is then placed under a Mohr burette filled with water or any other transparent liquid, and the water is slowly poured until the level of the water is flush with the 50 cc line. We read from burette the volume of the water used, which is subtracted from 50 to have the volume occupied by the molasses. By dividing the weight of the molasses by its volume, the density is obtained. If the operation was done at a temperature different from the normal temperature, make the necessary correction.
[This technique of using small amounts of distilled water to measure small amounts of a volume below a calibration line is also practiced on single droplets of essential oils to find their density. For molasses, I don’t think you need a precision pycnometer so you can probably use a cheap 50 ml volumetric flask.]
The previous method has given rise to the following criticisms: during heating of the molasses, an appreciable decomposition takes place, which results in a decrease in the density of the product (Newkirk). At the same time, large quantities of air in emulsion can remain in the melasse (1), especially since to obtain an average sample it is important to mix the product well. To remove air and gas from molasses, Newkirk proposed the use of a special pycnometer, put into communication with a vacuum machine. The air bubbles gather in a chamber, which is removed when the release of gases, facilitated by the vacuum, is finished.
(1) Some cane molasses require several days of heating to be demulsified (Pellet).
[These days vacuum degassing may superceded by ultrasonic degassing. Degassing by vacuum is harder than you’d think and a there can be a lot of volume expansion that would have to be accounted for then then taken up by the previously described burette technique.]
Determination of water and dry matter
The evaluation of water is usually made by heating the test material in a drying oven at a temperature of 100-105° to constant weight.
The results thus obtained are not very precise, especially in the case of molasses. Effected by heating are, indeed, entrained products other than water: dissolved carbon dioxide, volatile acids, etc. There is also a decomposition of certain constituents of molasses, particularly (according to Prinsen-Geerligs) glucinates, or bodies produced by the action of lime on reducing sugars. N. Deerr found, in the case of cane molasses heated at 100° for 10 hours, that at a loss of weight of 100 corresponded 98.7 of water, 1.0 of volatile acids and of CO2 and 0.3 of undetermined. For cane juice, weight loss was represented by 99.7% water and 0.3% volatile acids.
In order to eliminate, or more exactly to reduce, the importance of the error resulting from cooking, it has been proposed to carry out desiccation below atmospheric pressure at 70° C in a vacuum. It has also been recommended to replace drying by distillation with a water immiscible liquid, toluene for example (Bidwell and Sterling). But during the boiling with toluene (boiling point 110 ° 7), as well as at temperatures below 70°, it seems that some molasses compounds are still affected.
[Toluene is now the preferred method and protocols exist for it that are also used for the moisture content of grains. There is also a special distillation head to collect the water.]
As sugary substances become very viscous towards the end of the desiccation process and give up with difficulty the last traces of water, it is indicated, to reduce the time needed, to use dividing materials realizing a larger surface of evaporation: pumice stone, quartz sand, filter paper.
[This advice is described cryptically, but I think he is asking for material of high surface area to seed boiling and reduce bumping. If there are no seed points, the material can super heat very easy without actually boiling and giving up the water vapor. The degree to which bumping is prevalent can be quite surprising. Many of these liquids are 20% free water. Sometimes for honeys they stop counting sugar content and starting counting free water content because it is correlated to stability and other factors like crystalization.]
We can proceed as follows: Place in a flat capsule about 10 gr of freshly calcined pumice and a small glass stirrer. After drying in the oven and cooling under the desiccator, tare carefully. Then introduce into the capsule 3 g of molasses, then slowly pour via pipette hot distilled water to dilute the test portion; mix everything evenly in the capsule with the stirrer. Dry in the oven at 70 ° C until constant weight.
[I’m not sure what calcined pumice is, but I suspect its a cleaning step to ensure it is neutral. Pumice is essentially natural glass foam.]
In the case of cane juice, sufficiently exact results are obtained by operating the desiccation at a temperature of 100-105°.
Total dry matter (actual extract or real Brix) is obtained by subtracting from 100 the quantity of water evaluated. It is not possible to go from the apparent extract, determined by means of special tables starting from the density or directly by means of a saccharometer, to the actual extract, because of the compositional variations presented by the non-sugars.
For the determination of total solids, the refractometer (ABBE refractometer or better Zeiss sugar refractometer) is also employed. The results obtained are as accurate as those provided by the desiccation method, when the products examined do not contain suspended solids and have a low proportion of non-sugar.
Assay of ashes
The determination of mineral matter is by direct incineration (carbonate ash), or by incineration after destruction of the organic material by means of sulfuric acid (sulphated ash). It can also be performed electrometrically, with a precision as satisfactory as the previous methods, if we have well built devices.
Carbonated ash. — 5 to 10 g of material are introduced into a platinum capsule of 50 to 100 cc, which is slowly heated in a muffle oven until no more gas is released, then we bring it to dark red until we get white ashes. It is then moistened with a small amount of Ammonium carbonate, the excess carbonate is removed, heated again moderately, and weighed.
[I’m not too sure how the excess is removed? One of these processes may be valuable to understand as we explore dunder that has accumulated a lot of ash. Eventually it may be needed to troubleshoot stuck fermentations and odd results.]
Sulphated ash. — Add to the sample of material placed in the capsule a few cc of sulfuric acid, heat gently until carbonization of the organic matter, then bring to dark red until disappearance of all the carbon. Cool down, add a few drops of sulfuric acid (to convert the sulphides that could have come from the decomposition of sulphates by carbon) and place the capsule in the muffle oven, heating very slowly to avoid splashing. Cool and weigh.
Sulphated ash is often translated into carbonate ash, multiplying the number by the coefficient 0.9, proposed by Scheibler. However, it is clear from the work of many authors that this coefficient is much too low: Spencer, for example, observed that in the final molasses of Cuba the sulphated ash level was 17 to 21% higher than carbonate ash.
The weight of the carbonate ashes does not, moreover, correspond much more exactly than that of the sulphated ashes to the quantity of the actual mineral salts in the sugary matters. The bases that these contain are in fact combined, for the most part, with mineral acids (HCl, SO4H2) or organic acids, and not in the state of carbonates. On the other hand, as Browne and Gamble have shown, during direct incineration, variable and uncontrollable losses occur in Cl, S, etc., so that the carbonate ash method provides more irregular results than sulphated ash.
As, moreover, the determination of carbonated ash presents difficulties of execution, the complete combustion of sugars in the presence of certain salts being very difficult (the latter, by melting, includes particles of carbon and subtracts them from the action of fire), it is practiced only for research work. In routine analyzes, the sulphated ash method is always used, which is easier and gives more concordant results.
Evaluation of sugar substances
The sugars found in rhummerie raw materials are sucrose, glucose and levulose, with small amounts of glutose and mannose. The methods of assaying these sugars fall into two broad groups: optical methods (which allow the determination of sucrose) and chemical methods (which give the amounts of reducing sugars and, indirectly, sucrose).
At least in the case of molasses, methods of analysis must be regarded as having a conventional character. The results provided by different chemists may vary quite substantially unless they are placed in exactly the same conditions for the preparation of solutions and the conduct of operations. The figures obtained, expressed as sucrose and invert sugar, do not represent in an exact manner the quantities of these sugars actually present, as a result of the intervention of various causes of error. Nor do they permit a rigorous estimation of the value of the raw material in the distillery, because of the presence of infermentable reducing sugars, usually referred to as glutose, which exist in relatively high proportions in certain types of molasses (particularly those of Java and the Hawaiian Islands).
In general, polarimetric methods give less accurate figures than chemical methods. With some molasses, the results obtained are concordant, but in many cases optical methods give lower results (up to 1.5%) than chemically (Davis) (1). In industrial practice, moreover, it is possible to content oneself with processes capable of giving, with the various raw materials examined, comparable figures, making it possible to calculate the theoretical yield of alcohol and to control the manufacture.
(1) Int. Sug. J. XL, 186, 1938.
[I am very much interested in knowing how much potential alcohol I am adding to complicated baticións where the end density is not strongly represented by the sugar content.]
A. — Preparation of sugar solutions for analysis
Sugar solutions are usually too messy and too strongly colored to be examined as they are with a polarimeter, so it is important to submit them to a prior defecation. This operation unfortunately introduces various causes of error, which can influence the results of the analysis quite significantly.
[I have learned this lesson with measuring total dissolved solids of coffee with a refractometer. Samples of coffee or espresso must be filtered with a syringe filter. The line seen on the refractometer goes from hazy and indefinite to significantly more definite.]
The most commonly used defecating agent for the polarimetric examination of molasses is lead acetate or liquid lead acetate, recommended for the first time by Clerget. This product is obtained by dissolving in neutral distilled water at a gentle temperature for 8 to 10 hours (CH3.COO)2 Pb, and pulverized litharge, PbO. There are different formulas of preparation: according to the digestion time and the respective proportions of acetate and litharge employed, a reagent is obtained whose composition, as well as the action on the optically active constituents of the sugar solutions, vary.
[I have heard of this, but I’m not sure if there is a more modern, safer alternative. Even though lead is toxic, I’m wondering if this is prepared at quantities small enough that it is fairly safe and not horribly toxic to dispose of.]
Lead sub-acetate ensures a good discoloration of impure sugary solutions: it precipitates pectins, organic acids, several nitrogenous coloring matters, and a part of pentosans and substances resulting from the action of heat on sugars (caramel).
A first cause of error in the use of this reagent is due to the insoluble precipitate which is formed and whose volume reduces that originally occupied by sucrose. The sugar solution is consequently more concentrated, which increases the polarization. Several methods have been prepared to correct this error. Horn, for example, uses, instead of the liquid sub-acetate, lead sub-acetate in the form of anhydrous salt, added to the solution after filling the flask to the mark. This process is common in American sugar factories, for the control of manufacture.
On the other hand, basic lead acetate acts on the constituents of sugar solutions. Levulose, in the presence of certain chlorides and salts which form insoluble compounds with lead, is precipitated partly in the form of lead levulosate. Glucose is also precipitated, but less than levulose. As a result, the rotation of a solution containing invert sugar tends to increase to the right. The optically active nitrogenous materials are also modified: in the presence of basic sub-acetate, asparagine and aspartic acid, for example, levorotary become dextrorotary, while glutamine and glutamic acid from dextrotary become levorotary. The importance of the action on reducing sugars and polarizing nitrogenous substances depends on the basicity of the reagent and the amount of reagent used. Also, in the various operating techniques it is often recommended to carefully avoid using an excess of lead sub-acetate. But it is obvious that this minimum amount is a function of the chemist’s personal equation, some adding just enough reagent to allow polarimetric reading and others seeking maximum discoloration.
Various products that do not have the disadvantages of lead sub-acetate have been advocated on a number of occasions.
Neutral lead acetate does not affect the optical activity of reducing sugars, but its bleaching power is much lower than that of the sub-acetate. Although it is suitable for the defecation of weakly colored cane juice, it does not give satisfactory results for the polarimetric examination of molasses. It is used quite frequently in the treatment of molasses analyzed by chemical methods; Excess lead is then removed by means of carbonate, oxalate or sulphate of soda.
Lime hypochlorite has been proposed by Zamaron. Pellet and Pairault particularly recommend it for the treatment of strongly colored cane molasses. The reagent can be prepared by introducing into a flask 200 gr of hypochlorite with 500 cc of water, shaking occasionally and filtering. 25 to 40 cc of this solution would be sufficient to discolor the normal weight of molasses. The experimental data are too few to allow specifying the value of the reagent. According to Pairault, it would not influence direct polarization.
Animal black gives clear filtrates from the well discolored. Unfortunately, it absorbs sugar and this absorption is variable according to the commercial brands of blacks. With well-prepared carbons, and more particularly with activated carbon (Norit, Darco), whose bleaching power is high, the absorption is however relatively low. They are therefore often used to complete the discoloration of products previously treated with lead acetate.
Ammonia magnesia acetate, according to Ishida, is an excellent defecation for cane juice. This reagent destroys the optical activity of asparagine and glutamic acid, while it breaks down glucose and levulose very little.
Kielselguhr, the crème of alumina (obtained by treating a solution saturated with alum with ammonia and before the alumina by decantation until elimination of the sulphates) precipitates the colloidal materials and give clear, but not discolored filtrates. They are usually used as adjuvants of other defecating products. Kielselguhr is sometimes used alone, especially in the evaluation of reducing sugars by chemical methods.
[IKielselguhr is also known as diatomaceous earth which is a common filter aid.]
Zinc powder, in the presence of hydrochloric acid, gives rise to hydrogen, which reduces some of the coloring substances and causes some discoloration of the solutions. According to Herles, good discoloration can be obtained by using a large quantity of powder, but then the inversion constant is modified. This reagent, which does not appear to modify the polarization of sugars or other optically active products, is sometimes used to decolorize sugar solutions after inversion.
B. — Optical methods
Optical methods are based on the determination of the rotation of a polarized light beam, passing through a column of sugar solution. This rotation is proportional, at least between certain limits, to the concentration of the solution and to the length of the column. The apparatus used for the examination of the sugars, or saccharimeters, differs from the polarimeters only by the mode of graduation.
Normal weight. [Ifluence or load may be more appropriate than weight or possible metric?]
The weight of sucrose which gives, on the arbitrary scale of the saccharimeter, the point 100 is said normal weight.
In the French apparatus (saccharimeters of Laurent, Pellin, etc.), point 100 corresponds to the deviation produced by a quartz plate with parallel faces, 1 cm thick, cut perpendicularly to the optical axis. Biot, the creator of the sugar polarimetry, had found that this same deviation was given by a length of 20 cm of a solution containing, per 100 cc, 16.47 g of pure and dry sucrose. The normal weight was later found equal to 16.35 grams by Clerget, 16.19 grams by Girard and Luynes, and finally 16.29 grams by Mascart and Benard (1899) and Pellat (1901). It was shown later that the work of the latter experimenters did not lead to a normal weight of 16.29, but to 16.26 gr. The Decree of 24 March 1938 set the normal weight at 16.269 ± 0.002 gr, weighed in the air with weights of brass. Each saccharimetric degree of the French scale then corresponds to 0.16269 gr of sugar in 100 cc of the solution examined.
In the German apparatuses (Schmidt and Haensch, Peters, Bausch and Lomb), point 100 corresponds to the deviation produced by a solution of pure sugar at a density of 1.700 (at 17°5), examined in a tube of 20 cm. This solution contains, for 100 cc Mohr, 26,048 gr of sucrose. When the Morh liter was replaced in German laboratories by the metric liter, the normal weight per metric liter was reported at 20°C, which squeezed it down to 26 grams adopted by the “International Commission for the Unification of methods of analysis “in 1897 and currently used in most of the sugars countries (Ventzke scale).
Bates and Jackson, following very detailed experiments conducted in 1916 at the US Bureau of Standards, found that a solution of 26 grams of pure sugar in 100 cc, observed at 20° C in a tube of 20 cm did not give the point 100 of the scale Ventzke, but marked only 99.895, which carried the normal weight to 26.026 gr. The “International Commission for the unification of the methods of analysis” has, in its 8th metting (Amsterdam 1933), adopted this figure for the saccharimeters graduated according to the Ventzke scale and retained as international scale that giving exactly the point 100 with a normal weight of 26 gr of sugar.
Variation of the rotatory power of sucrose.
The rotatory power of sucrose is influenced by various factors. Optically inactive salts found in sweet raw materials (chlorides, sulphates, phosphates, carbonates, acetates of Na, K, Ca, etc.) reduce the rotation, while formaldehyde increases it (Farnsteiner). However, this action is weak enough to be negligible in current practice. Lead sub-acetate decreases the rotational power appreciably only if it is used in excess.
The concentration of the sugar solution affects rotary power, but only slightly.
The most important factor involved in polarimetric observations is temperature. This increases the rotatory power of quartz plates and decreases that of sucrose. The result, in the case of a normal solution of pure sucrose, is an increase of 0°03 Ventzke per degree of temperature above the normal temperature of graduation. In the case of commercial sugars, the correction is often done using the formula:
where P20 is the polarization at the international normal temperature of 20° C and Pt the polarization at the temperature of the observation t.
This formula loses its value when sucrose is accompanied by various impurities, including levulose. Browne (1) has observed that at about 80° Ventzke, the action of temperature on impurities compensates for the reduction in sucrose rotation, so that no correction is necessary. But below 80° Ventzke, there is an increase of the real polarization when the temperature rises. The author has proposed the following correction formula:
(1). Handbook of sugar analysis. 1912.
This formula applies only to products of average composition; it can give results far removed from reality with aberrant samples.
The only way to get accurate results is to operate at the normal temperature of the device. Since in the tropics it is very difficult to achieve these temperatures, it is important at least to operate at constant temperatures (especially in the case of double polarization), to have comparable figures, and preferably during the cool hours of the day.
Inversion.
If the solution contains only sucrose, direct polarimetric examination, or direct polarization, gives the amount of sugar contained in the examined liquid. But when this sugar is accompanied by other optically active bodies (reducing sugars, etc.), which is normally the case for products of interest to the distillery, it is necessary to use the double polarization method, attributed to Clerget.
Let x be the polarization of sucrose and y that of the other active substances, the direct polarization d will be given by the formula d = x + y. If the sucrose is hydrolyzed in equal parts of glucose and levulose, the polarization after inversion, or inverse polarization i, will be: i = ax + y By subtracting the two equations, we will have:
Rotation of the invert sugar decreases, as shown by Clerget, with the rise of the temperature. The formula, corrected for this factor, becomes (t being the temperature at which the reading is made):
or, as it is expressed more generally:
The factor (1 – a) 100 is known as the Clerget constant, or inversion coefficient. Since inverting sucrose has a left rotation, the value of a will be negative and (1 – a) greater than unity i will also be negative, unless the value of y is high enough to compensate for rotation to the left invert sugar. So, we often give Clerget’s formula the form:
where A is the right reading before inversion and B is the left turn after inversion.
For the validity of this method of analysis, it is important on the one hand that the operation of the inversion be done in such a way that we always find the same value for a; and on the other hand, that the value of y remains the same before and after the inversion.
In fact, as Gubbe pointed out in 1883, the rotatory power of invert sugar varies with the sugar concentration and the degree of acidity of the sugar solution. The heating mode itself influences rotation (Jackson and Gillis). It follows that the different operating techniques will each have their own inversion coefficients and that, for each of them, these coefficients will vary with the sugar content.
At the same time, the presence of basic lead salts in the defecated solutions greatly reduces the rotatory power of levulose, which is partially precipitated as insoluble combinations. The addition of acid restores the rotation power. The left rotation of the pre-existing reducing sugars is thus increased after inversion, which gives different values for y in the forward bias and in the reverse bias. The rotatory power of active nitrogenous substances is also modified: asparagine and aspartic acid, for example, which are dextrorotatory in alkaline solution, become levorotary in acidic medium. The quantity of these substances is, however, low in cane products, and their influence, according to some authors (Jackson and Gillis), could be neglected, while according to others it would be quite noticeable in some cases (Zerban).
To eliminate the cause of error resulting from modification, during the hydrochloric inversion, of the optical activity of substances other than sucrose. Saillard advocated, in 1912-13, the neutral double polarization method. This method, which has been repeated with some modifications by Jackson and Gillis, is based on the principle that equivalent quantities of soluble salts exert respectively the same action on the polarization of sucrose, invert sugar and active nitrogenous substances. For the solution used for the direct polarization, a quantity of NaCl or KCl equivalent to HCl used to obtain the inversion, while HCl present in the solution for the reverse polarization is neutralized exactly with sodium hydroxide or potassium hydroxide. In the pure sugar solution used to establish the inversion coefficient, the same amount of salts (NaCl or KCl) which exist in the molasses test portion are also produced.
Instead of an acid, invertase can be used to effect the inversion. The diastase inversion method, first advocated by Kjeldahl and used by O’Sullivan, Hudson, Ogilvie, etc., is the method most recommended for exact evaluation of sucrose (Zerban). However, it is less suitable for routine analyzes because it requires more time and more careful precautions than hydrochloric inversion.
Operational methods.
Clerget method, — a) Direct polarization: Weigh 32.538 gr (twice the normal French weight) of molasses and transfer it with hot water into a 200 cc graduated flask. Defecate with 15-20 cc of lead acetate at 28° B, cool; make up to 200 cc with water; shake ; filter; polarize in a tube of 20 cm.
b) Polarization after inversion. — In a 50-55 cc flask, introduce 50 cc of the liquid and filter; add 5 cc of pure HCl at 22° B and mix well while turning. Put in a water bath, which is heated to reach 68-70° inside the flask for 10-12 minutes. Remove the flask from the bath and leave to cooling at room temp to 40°; then cool to 20° in a stream of cold water. Fill to 55 cc with water. Add, if necessary, a little animal black washed with acid, then with water and dried. Shake, filter, polarize in a 20 cm tube. Apply the formula:
Hersfeld method — a) Direct polarization: Weigh 26 gr. of molasses and transfer them to a 200 cc flask; defecate with the amount of lead sub-acetate needed; make up to 200 cc, shake, filter and polarize.
b) Polarization after inversion — Transfer 13 gr. molasses to a 100 cc flask and add 75 cc of distilled water (without defecating), then 5 cc of 38% HCl (density 1.188 at 17°5). Mix everything well. Place the flask in a preheated water bath at 70° C and hold it there for 5 minutes from the time its contents mark 69°. Extract the flask from the bath and immerse in a stream of cold water until cooled to 20°. Make up to 100 cc with distilled water. Add a little animal black, if necessary, then shake, filter and polarize. Apply the formula:
The Clerget method is used in France and Herzfeld method in Germany for commercial analyzes. Since, in this case, we are dealing with conventional analyzes, we apply the inversion coefficients which relate to the normal solution (France) and semi-normal solution (Germany) of pure sucrose. In the case of scientific analyzes, it would be important to use the coefficients corresponding to the true sugar content, which were established by Saillard for the French method and by Hersfeld for the German method. However, it is preferable to use the following methods, the Clerget method applied to molasses giving too low a sucrose content.
Saillard method with dual neutral polarization — a) Direct polarization. — Put 4 times the normal French weight of neutralized molasses in a 200 cc flask; defecate with a sufficient amount, but not excess, of lead sub-acetate, make up to 200 cc: shake, filter (K filtrate). Take 50 cc of K fitrate; add in KCl equivalent of HCl which will be used for the inversion, complete to 100cc; put a little animal black if necessary; shake, filter, polarize at 20°.
b) Polarization after inversion — Take 50 cc of filtrate K; invert with 6.8 cc of HCl at 22° B; neutralize with potash; cool to 20°; bring to 100 cc; put a little animal black, if necessary; shake, filter, polarize at 20°. Apply the formula:
To determine the inversion coefficient C : a) Prepare a pure sugar solution of the same polarization as the molasses solution (filtrate K). Take 50 cc of the pure solution; add in KCl the equivalent (q = 6.8 + s) of HCl at 22° B, where s is the amount of HCl equivalent to the sulfuric acid contained in the sulphated ash; complete to 100 cc; shake, polarize. Let A be the reading.
b) Take 50 cc of the pure solution; add 5 cc of HCl at 22° B; mix, invert, cool, add (q – 5) cc HCl or the equivalent amount of KCl; neutralize with potash; complete to 100 cc; shake, polarize. Let B’ be the reading.
c) Take 50 cc of the pure solution; make up to 100 cc with water; shake, polarize at 20 ° Let A” be thereading.
The inversion coefficient C is deduced from the formula:
Jackson and Gillis method No IV. — Direct polarization. Prepare a normal solution of the product to be analyzed and defecate with the necessary amount of dry basic lead acetate, avoiding excess, mix well and filter. Place in a 100 cc flask, 50 cc of the filtrate; add 20 cc of distilled water, then 10 cc of a solution of NaCl (at 231.5 gr per liter); complete at 100 cc and polarize.
b) Inverse polarization. — Introduce 50 cc of defecated liquid into a 100 cc flask; add 20 cc of distilled water. Place the flask in a bain-marie until the temperature reaches exactly 65°. Remove the flask from the bath, add 10 cc of HCl (density 1.1029 at 20°/4°); mix, by gyrating the flask, and leave to cool 30 minutes at ambient temperature. When cool, make up to 100 cc and polarize.
A special table shows the values of the inversion coefficient for the various sugar concentrations (sum A + B) and the corrections to be made for observations made at temperatures above 20° C. The sucrose level is given by the formula:
This method, widely used in American countries, has been specially studied for the analysis of cane products, generally poor in optically active nitrogenous substances. As it neglects, contrary to the Saillard method and the Jackson and Gillis No II method, the action of these materials on polarization, it is not suitable for the analysis of beet molasses, rich in asparagine and aspartic acid.
C. – Chemical methods
The chemical methods for the determination of sugars are based on the property possessed by certain sugars, so-called reducing agents (glucose, levulose, lactose, maltose, etc.), to reduce by boiling cupric salts into cuprous salts. This property, used by Trommer to differentiate grape sugar from cane sugar, was for the first time applied to chemical analysis by Barreswill (1842) and generalized by Fehling.
Chemical methods can only directly evaluate reducing sugars, but since it is easy to convert sucrose into invert sugar, they can be used indirectly for the determination of cane sugar. If one makes a first assay on a complex sweet solution as it is presented, and a second assay on the same solution after inversion, one will obtain, on the one hand, the pre-existing reducing sugars in the solution (direct reducers), on the other hand the sum of these and the reducing agents produced by the inversion of sucrose (total reducing agents (1)). The difference between the total and direct reducers, multiplied by 0.95, will represent the amount of sucrose existing in the original solution.
(1) The inversion is done in the same way as in optical methods. However, according to Davis, the Herzfeld process gives, with some molasses, somewhat higher figures than the invertase method, which seems to indicate the presence of carbohydrates other than sucrose (pentosans, dextrins…) hydrolyzable by strong acids. The invert solution should be neutralized by addition of Fehling liquor.
Fehling had found in his experiments that one equivalent of dry glucose reduced exactly 10 equivalents of copper sulphate, taking 5 equivalents of oxygen. He recommends a liquor containing per liter: 34.64 gr of pure and dry SO4Cu, 150 gr of neutral potassium tartrate and 75 to 86 gr of NaOH. 10 cc of this liquor are reduced exactly by 10 cc of a sweet solution containing 5 g of anhydrous glucose per liter.
Most of the formulas proposed for the preparation of cupric liquors retained the above concentration for copper sulphate. On the other hand, the proportion of NaOH varies between 50 and 150 g per liter. When the reducing sugars are not accompanied by sucrose, the more alkaline cupric liquors, which give a sharper and faster reduction, may be employed to advantage. In the opposite case, there is an attack of sucrose, all the more accentuated as the proportion of this sugar is greater. Hereinafter the composition of one of the liquors most often used today.
Soxhlet Liquor — a) Dissolve 34.639 gr of crystallized copper sulphate (SO4Cu.5H2O) in distilled water, dilute to 500 cc and filter through asbestos; b) Dissolve 173 gr of Seignette salt and 50 gr of NaOH in water, dilute to 500 cc, let stand for 2 days and filter through asbestos. Mix the two solutions immediately before use with 10 cc of liquor equivalent to 0.05 gr of invert sugar.
In fact, the amount of copper reduced by the Fehling liquor depends on various factors: a) composition of the liquor, particularly as regards the sodium content and the pH; (b) duration and speed of boiling; (c) presence of sucrose, lead salts and lime. Even the surface area of the beakers in which the reaction takes place has a certain influence.
In order to remedy the causes of error that result, it is first of all necessary to titrate the cupric liquor with respect to a pure solution of glucose or invert sugar, rather than to deduce the measure according to the content of the copper solution. A correction coefficient will thus be calculated, which will be applied to the figures of the tables accompanying the methods of analysis and which make it possible to pass from the reduced quantity of copper to the reducing sugar content. It was placed rigorously in the conditions which preceded the setting of the copper liquor.
[Something about this makes me think that you need to be processing vast amounts of molasses to have the economies of scale to make all this worth it.]
A correction must also be applied when the sugar solution contains sucrose. The attack of this sugar is all the more pronounced as the reaction temperature is higher and the copper liquor more alkaline. For specific reaction conditions, it varies with the proportion of sucrose relative to reducing sugars and is all the more marked that this proportion is higher. Special tables (Meissl and Hiller, Munson and Walker, Lane and Eynon, etc.) have been established which permit correction, but it is important that the procedure followed, as well as the quantities of material taken for analysis , are exactly the same as those for which the tables were calculated. Some authors have also proposed to make the correction by incorporating into the reducing sugar solution used for the Fehling liquor titration the proportion of sucrose which is usually found in the samples to be analyzed.
The duration and speed of heating must be carefully adjusted, and the sweet solution to analyze diluted to contain between 0.1 and 0.5% of sugars.
Finally, the reducing power of invert sugar is significantly reduced in the presence of lead salts (Pellet, Edson) and lime salts (Eynon and Lane). It is therefore necessary either not to use lead sub-acetate for defecation, which can be replaced by Kieselguhr, or to eliminate lead salts by means of phosphoric acid or sodium phosphate. As for the calcium salts, it is easy to get rid of them by adding, when lead acetate is not used, a little Na oxalate (0.10 gr per gr of molasses) to the kieselguhr, before operating filtration (Lane and Eynon). Cook and Mc Allep advise, to remove both the salts of Pb and Ca, and to use, per gram of molasses, 1 cc of a solution containing 7 gr of phosphate and 3 gr of oxalate of K per 100 cc. It is important not to decalcify the molasses without at the same time removing the lead salts; otherwise, one would find figures that are substantially too high for reducing sugars (Davis).
The chemical methods for the determination of reducing sugars can be classified into two groups: volumetric methods and gravimetric methods.
Volumetric methods.
It is determined how much cubic centimeters of a sugar solution is needed to completely reduce a known volume of cupric liquor, the end of the reaction being indicated by the disappearance of the blue coloring of the liquor.
This method has the advantage of simplicity and speed; it is therefore the commercial method par excellence. But it requires the use of careful precautions and corrections, if we want to obtain accurate results.
Since the end of the reaction is difficult to grasp, the use of various indicators to minimize the chemist’s personal judgement has been advocated. For example, a little liquid can be filtered rapidly, neutralized with acetic acid and a drop of K-ferrocyanide solution added, which gives a brown coloring if copper remains in solution. Another more common method, is to take a drop of liquid with a stirrer, and place it on a white plate in contact with a reagent formed of 1 gr of ferrous ammonia sulphate and 1.5 gr of ammonium sulphocyanide in 10 cc of water and 2.5 cc of HCl: in the presence of copper salts, an intense red coloration occurs. It has also been recommended to add 1% methlyene blue to the Fehling liquor, which becomes discolored when the reduction of the cupric salts is complete.
Other authors formally disapprove of the use of any reagent to check the end of the reaction, because a certain amount of copper can remain in colloidal solution, once the reaction is complete. Pellet writes in particular: “We have shown that complete discoloration can take place while leaving copper in solution, but in the minimum state, and if, in consequence of various circumstances, there are at least traces of copper in solution, this oxidizes in contact with the air and by cooling, so that the reagents such as K ferrocyanide still detect copper while the operation is complete”.
Among the different modes of operation that have been proposed, and which differ quite little between them, we will note the following:
French method. — Dilute the sweet solution (the composition of which was determined by a first rough test), so that it contains about 0.5% invert sugar. Introduce 10 cc of Fehling’s liquor into a flask, heat on a wire cloth, and as soon as the liquid begins to boil, drop into the balloon drop by drop and without stopping, so as not to interrupt the boiling, the sugar solution contained in a graduated burette, until the discoloration of the liquor is complete. The operation must be conducted as quickly as possible. For more accurate results, it is advisable to do several successive tests and take the average of different readings.
Lane and Eynon method (1). — A solution of the sample to be analyzed containing 0.25 to 0.80 gr. of glucose per 100 cc, if the amount of sucrose is relatively low (molasses), or 0.1 to 0.5 gr, if sucrose is in high proportion (juices, syrups, etc.). Remove the calcium salts by adding Na or K oxalate (0.10 gram per gram of molasses or 0.5 cc of a 10% solution of K oxalate) and filter with kieselguhr.
(1) Int. Sug. J. XXV, 142, 1923.
[This is a reminder that molasses is poor in sucrose because the vast majority was removed.]
In a 300-400 cc Erlenmeyer flask, add 10 or 20 cc of Soxhlet liquor. Add cold enough the amount of sugar solution needed to carry out almost complete reduction of the copper, so that there is only to use 0.1-1 cc later to complete the titration. Heat the mixture until boiling, on a wire cloth and keep gently boiling for 2 minutes. Without interrupting the boiling, add 3-5 drops of 1% methylene blue, then complete the titration, pouring the sugar solution dropwise, until the indicator has faded; the operation must be completed after 3 minutes of uninterrupted boiling in all.
To determine the approximate volume of sugar solution needed, do a preliminary test as follows. Add 10-20 cc of Soxhlet liqueur, 15 cc of sugar solution and heat to boiling. After 15 seconds of boiling, rapidly pour 1 cc of sugar solution, boil for 10 seconds, and so on until almost complete disappearance of the blue color of the solution. Then continue boiling for 1-2 minutes, add the indicator and finish the titration by adding sugar solution drop by drop.
The amount of reducing sugars, in mg per 100 cc, is given by the formula:
where f is the amount of invert sugar in mgr corresponding to 10 cc of Soxhlet liquor and t the number of cc of sugar solution used.
Since the factor f varies with the proportion of sucrose present, the authors have established the following correction table:
The Lane and Eynon method, often used in Anglo-Saxon countries, for the determination of reducing sugars in cane molasses, gives very regular results (Lever). According to Davis, when the rate of reducing agents exceeds 20%, there is a close agreement between the results obtained and those of the gravimetric method of Brown, Morris and Millar. But when the proportion of reducing agents in molasses is minimal, the Lane and Eynon method would tend to give too low numbers.
Gravimetric methods.
The copper oxide precipitate produced by a known quantity of sugar solution acting on a large excess of cupric liquor is collected, and from the weight of the precipitate obtained, the amount of reducing agents contained in the sugar solution is deduced from tables.
Filtration of the precipitate is usually done on asbestos placed in a Soxhlet tube or in a porcelain Gooch creuset.
The proto-oxide of copper can be evaluated in the form of copper oxide, after desiccation to constant weight; cupric oxide, after calcination; or copper, by reduction in hydrogen or by electrolysis in nitric solution. It can also be volumetrically evaluated with potassium permanganate or potassium iodide and sodium hyposulphite.
The most accurate results are given by the electrolytic and iodometric processes. However, according to Meade and Harris, the figures obtained in the case of cane molasses are almost identical, whether the evaluation is in the form of cupric oxide, metallic copper, or iodometrically.
There are many procedures operating under the gravimetric principle. They are distinguished by the composition of the cupric liquor used, by the reaction conditions (temperature, duration of heating) and by the titration mode of the copper oxide precipitate.
The Herzfeld method, used in Germany, has lost much of the prestige it has enjoyed for a long time, because, during the boiling of the liquid, there is an overheating of varying importance affecting the amount of sucrose attacked and consequently , that of copper reduced. It tends to be replaced almost everywhere by more precise or easier methods: Schoorl (Java), Munson and Walker (United States), etc. (1). We will describe only those methods which have been more specifically studied for the analysis of cane molasses.
(1) Report of the Proceedings of the 9th Session of the International Commission for the Unification of Methods of Sugar Analysis.
Bull. Ass. Chim. LIV. 303. 1937.
Munson and Walker Method. — Place 25 cc of each solution of Soxhlet liquor in a 400 cc beaker, add 50 cc. of the sweet solution. Heat on an asbestos cloth placed on a Bunsen burner, so that boiling occurs after 4 minutes. Keep the beaker covered with a watch glass for the duration of the heating. Filter the solution hot on an asbestos layer deposited in a Gooch crucible, using suction. Thoroughly wash the precipitate of cuprous oxide with water at the temperature of 60° and weigh it directly in the state of copper, or measure the quantity of copper reduced by one or the other of the methods indicated previously.
A special table gives the amounts of reducing sugars corresponding to reduced copper and sucrose ratios. The Munson and Walker method is widely used in America for the determination of reducing sugars in cane mixtures, glucose syrups and, more rarely, in raw cane sugars.
Schoorl method (2). — Treat a solution of 6 gr of molasses with 15 cc of a solution of neutral lead acetate (10%); bring to 250 cc and filter. Take 50 cc of the filtrate and remove the excess lead with 5 cc of a solution containing in 100 cc, 7 g of PO4HNa2, 12H20 and 3 g of C2O4K2. H2O; bring to 100 cc and filter.
(2) Handboek methoden van onderzoek bij die Java – Suikerindustrie, p. 254.
[This is also helpful on the disodium phosphates.]
Add 500 cc of this last filtrate to an Erlenmeyer flask containing 50 cc of Soxhlet liquor and one or two pieces of washed and calcined pumice. Heat the flask on a wire cloth resting on asbestos cardboard with a hole in the center, so that the liquid boils in 4 minutes, then boil gently for another 2 minutes exactly. After cooling rapidly, without stirring, treat the liquid with 25 cc of a 20% KI solution and 35 cc of SO4H2 (1 vol of concentrated acid in 5 vol of water). Titrate the released iodine with N / 10 sodium hyposulphite, using 3-4 cc of a 1% starch solution as an indicator. Deduct the volume of hyposulphite used, and that found in a blank test, using 50 cc of water instead of the sugar solution.
A special table, of which we give an extract below, indicates the percentage of reducing sugars corresponding to the reduced copper and the sucrose richness of the molasses.
This method is used in Java for the determination of reducing sugars in cane molasses.
Brown, Morris and Millar method (1). — Use a non-defecated sugar solution containing 1 gr of molasses per 100 cc and a cupric liquor consisting of 34.639 gr of crystallized copper sulphate, 173 gr of Seignette salt and 65 gr of anhydrous NaOH per liter.
(1) Browne – Handbook of sugar analysis. 1912, p. 420.
Place 50 cc of freshly prepared copper liqueur in a 250 cc beaker and heat in a gently boiling water bath. When the solution has reached the temperature of the water, add 50 cc of the sugar solution; cover the beaker with a watch glass and continue heating for 12 minutes. Collect the precipitate of copper oxide in an asbestos-filled Gooch crucible, wash it with 200 cc of boiling water and finish with alcohol; drying out. Transform the precipitate into cupric oxide, placing the Gooch crucible inclined in a larger ordinary crucible, and warm at first gently and to finish strongly, on a Teclu burner (but without torch).
Correct the weight of cupric oxide by subtracting the figure obtained by doing a blank test with water.
When the weight of copper oxide is less than 0.245 gr, the amount of reducing sugars in the molasses is obtained by multiplying by the factor 81.5. For higher amounts of cupric oxide, the percentage of reducing agents can be found by means of the table below:
Determination of acidity and alkalinity.
The acidity (or alkalinity) of a solution can be expressed either by the total amount of H atoms (or OH oxhydryl groups, in the case of alkalinity) in the liquid: total acidity (or alkalinity) or titratable acidity; either by the quantity of H atoms existing in the state of electrical dissociation: real acidity or ionic acidity. It is the latter that is important to consider in biochemical phenomena, these being directly related to the percentage of ionization.
Titrimetry. [titration]
The total acidity is usually measured, by neutralizing a determined volume of the solution, using an alkaline liquor of known composition, in the presence of a coloring material as an indicator. Similarly, for the determination of the total alkalinity, the solution is saturated with a titrated acidic liquor, in the presence of a colored indicator.
In the analysis of distillery raw materials, decinormal sodium hydroxide is generally used as an alkaline liquor and phenolphthalein as an indicator. [decinomral is 1/10 normal?]
The turning point is not the same for all indicators. Tournesol, once commonly used, turns cold at pH 5-8; it is sensitive to CO2, which must be removed if it exists in liquids during evaluation. Phenolphthalein is colorless at pH 8.2 pink at pH 9 and red at pH 10. It has recently been proposed that methyl red in alcohol solution at 0.1%, insensitive to CO2, and phenol red, or phenolsulfonephthalein, in solution at 0.02%. The first is red in acidic medium and yellow in alkaline medium (limits of pH 4.4 – 6.0); the second yellow in an acidic medium and red in the alkaline medium (limits of pH 6.6 – 8.0).
In the case of juice or musts, 10 cc of the liquid to be analyzed is added to a porcelain dish and diluted with 25 cc of phenolphthalein-neutral water. 2 or 3 drops of 1% phenolphthalein neutral solution are added and the decinormal soda liquor, contained in a graduated burette, is allowed to flow until a persistent pinkish coloring. The number of cc of alkaline liquor necessary to obtain this result, multiplied by 0.49, gives the acidity, expressed in gr of sulfuric acid per liter of solution. To have acidity in acetic acid, multiply by the factor 0.60.
The coloring of molasses and vinasse is generally too dark to allow acidity (or alkalinity) to be measured by the above process. Pairault advises to operate via touch, as follows:
[touch seems almost like dabbing before the era where you could wet standardized papers.]
Dilute 20 cc of vinasse to 200 cc with water, then saturate with normal soda. The end of the operation is indicated by the touch on sensitive turmeric paper, prepared with the fresh root. When saturation is obtained, a small red circle forms around the point touched; beyond it, the point touched itself is red. Use a very fine glass rod.
[This is very clever but is from the era before we can use pH meters to find our endpoint in colored liquids.]
We can also use the method proposed by Gille (1) for the determination of acidity in wines:
(1) Ann. Falsif. XXV, 146, 1932.
Prepare a 0.02% solution of phenolsulfonephthalein: dissolve 0.02 gr of dyestuff in previously distilled distilled water, add 1.1 cc of N20 soda and make up to 1.000 cc. Place in each scoop of a porcelain plate 2 drops of the solution. Titrate 10 cc of the liquid to be tested and, when the coloring material turns, take a drop of wine and touch one of the cups of the plate with a glass stirrer.
[SOS, this is getting hard to follow]
The indicator, as we approach neutrality, begins to turn. The successive color changes made are sensitive to a drop and allow you to stop the titration at the precise pH that you want to achieve. The author chose pH 7.4, because at this ionic acidity it is certain to have completely salified the weakest acids, and it is easy, on the other hand, practically to make, for comparison, the orange-red coloring of pH 7.4.
Ionimetry. [pH]
It is agreed to express the ionic concentrations as gram-ions, or as gram-ion fractions, per liter of solution. As they are very small numbers, Sörensen proposed to use conventionally the logarithm of the inverse of the concentration, and to designate it by the symbol pH, ie:
The pH of absolutely pure distilled water was found to be 7. This figure represents neutrality, that is, the equality between the concentrations of H-ions and OH-ions. When the concentration of H ions exceeds that of OH ions, the solution is acidic and the pH is less than 7. When, on the contrary, there is predominance of OH ions in the solution, the solution is alkaline and the pH is greater than 7.
The pH of the solutions can be determined by colorimetric methods or electrometric methods.
Colorimetric methods. — These methods are based on the use of so-called indicator substances whose hue changes for a determined H-ion concentration. The transition from one color to another is not abrupt, but in gradual and insensitive variations. There is, therefore, for each indicator, not a point, but a turning zone, and, within this zone, there is a range of tones corresponding to a series of consecutive pH’s.
Among the various series of dyes proposed, that of Clark and Lubs is the most used. It has the following composition:
These indicators are most often used at a concentration of 0.02%, for those who turn red, and 0.04% for those who turn blue.
The pH measurement is made by comparing the staining of the test solution, to which a specified amount of a suitable indicator has been added, with that of a series of standard solutions of known pH, which have received the same dose of indicator. Test liquid and standard solutions are usually placed in Pyrex glass test tubes, 15 cm long and 1.5 m in diameter, into which 10 cc of liquid plus 8-10 drops of indicator are added.
Standard solutions, or buffer solutions, are constituted by mixtures, in varying proportions, so as to produce a pH range, of 2 or more electrolytes, most often one is strong and the other weak. These mixtures are sufficiently dissociated so that hydrolysis is practically zero, but insufficiently to bring about rapid changes in pH, when the proportions of their components are varied and the dilution (1) is modified to a certain extent. There are various buffer mixtures (mixtures of Sörensen, Palitzsch, McIvaine, Kolthoff, Clark and Lubs, etc.), for the composition and preparation of which we refer to the special works.
(1) In non-buffered solutions, the pH is modified by the slightest influences, such as the action of CO2 in the air, the alkali yielded by the glass, etc.
Sealed tubes containing permanent solutions of calibrated colors are commercially available. However, it is difficult to have identical indicator solutions to those used for the tubes. In addition, these standards can deteriorate in the long run, especially when they are poorly protected against heat and light. It has also been sought to achieve color ranges by means of colored glasses. In the Hellige comparator, for example, ebonite discs carrying a series of colored glass plates, graduated 0.2 in 0.2 pH are used.
If the solution to be studied is colorless or slightly colored, it is used as is. Otherwise, it should be diluted with distilled water. The liquids used in the distillery (juice, molasses, etc.) are sufficiently “buffered” to allow dilution of 1: 3 or 1: 5 (2). Avoid the use, for clarification, of filter paper, which is generally acidic and may alter the pH of the test solution.
(2) Von Stieglitz, by varying the concentration of Queensland’s final molasses, between 0.1 and 0.75 gr per cc, observed that the pH difference, determined by means of the quinhydrone electrode, oscillated, for the two concentrations above, between 0.32 and 0.62 depending on molasses. [My takeaway is quite a lot of dilution, but not a lot of pH movement]
It is also appropriate, when the solution examined is cloudy or colored, to juxtapose in turn to each standard solution a tube of natural liquid with no added indicator. To facilitate this operation, a “comparator” is usually employed. Walpole’s comparator, the simplest, consists of a wooden block pierced with holes, in which one can place the test tubes, with a side view to examine them by transparency.
Colorimetric methods have the advantage of being simple and easy to execute. They are unfortunately tainted with many sources of error, some physical (dilution of the indicator, temperature, illumination), other chemicals (presence of alcohol, mineral salts, proteic and colloidal substances). By using standard solutions whose pH has been electrometrically controlled and taking the necessary precautions (3) during the manipulations, an accuracy of the order of 0.1 pH can be obtained. If you want more accurate results, you have to use electrometric methods.
(3) In particular, it is important to operate under the same lighting and temperature conditions, on cold solutions with the same concentration of the indicator, avoid contact of the solutions with hands, not to use standards whose color has been altered by exposure to air, light or heat, etc.
Electrometric methods. — These methods are based on the variation of electrical tension caused by the ionic concentration of a solution which is introduced into a cell or electrolysis circuit. They give figures of great precision and are convenient for the study of colored solutions or disorders. Unfortunately, they require quite complicated equipment and manipulations that have kept them away from common practice for a long time.
Various potentiometer models are commercially available, using the hydrogen electrode or the quinhydrone electrode for measuring the H-ion concentration. Among the most employed, we can mention that of Trenel in Europe and that of Leeds and Northrup in America…