/4'. ,a,^>^ ;\
u \JVvV3lJt
UNIVERSITY FARM
Digitized by the Internet Archive
in 2007 with funding from
IVIicrosoft Corporation
http://www.archive.org/details/chembeetsugarOOspenrich
WORKS OF G. L. SPENCER
PUBLISHED BY
JOHN WILEY & SONS.
A Handbook for Cane>gugar Manufacturers and their Chemists.
Containing a review of processes of cane-sugar manuiacture, practical instruction in sugar-house control, selected methods of analysis, reference tables, etc. Fourth Edition, Rewritten and En- larged. i6mo, viii-l-331 pages, 52 figures, morocco, $3.00.
A Handbook for Chemists of Beet-sugar Houses and 5eed>culture Farms.
Containing selected methods of analysis, sugar- house control, reference tables, etc., etc. i6mo, x+475 pages, 74 figures, morocco, $3 00.
A HANDBOOK
FOR
CHEMISTS OF BEET-SUGAR HOUSES
AND
SEED-CULTURE FAEMS.
CONTAINING
SELECTED METHODS OF ANALYSIS, SUQAR-^
HOUSE CONTROL, REFERENCE
TABLES, ETC., ETC,
BT
GUILFORD L. SPENCER, D.Sc.,
OF THE U. S. DEPARTMENT OF AGRICULTURE,
Author of "^ Handbook for Sugar Manufacturert.*
FIRST EDITION.
SECOND THOUSAUC
NEW YORK: JOHN WILEY & SONS. London: CHAPMAN & HALL, Limited. 1910.
Copyright, 1897,
BY
G. L. SPENCER.
THE SCIENTIFIC PRESS
ROBERT ORUMMOND AND COMPANY
BROOKLYN. N. V.
PREFACE.
At the time the writer's "Handbook for Sugar Manu- facturers" was published, 1889, the sugar industry of the United States was confined almost exclusively to the cane sections of the South. Sorghum was attracting attention in the North, with some prospect of success ; the beet in- dustry was represented by two factories in California and dismantled factories in several other States. The condi- tions at this time are quite different. The beet-sugar in- dustry bids fair to attain enormous proportions, and sor- ghum, for the present, at least, has given up the struggle.
Under these changed conditions there appears to be an opening for a book devoted exclusively to the sugar-beet, hence this work.
In the preparation of this book it is assumed that the reader is familia. • with many of the ordinary chemical manipulations, but the fact is recognized that on account of the short manufacturing season many factories are com- pelled to employ assistants whose chemical knowledge is somewhat limited.
In order to avoid repetition, methods of sampling are de- scribed in a special chapter.
It is appropriate to mention here some of the men through whose efforts the sugar-beet has been successfully introduced into the United States. Among these are Dr. William McMurtrie, who visited the beet-sugar districts of Europe in 1880 and published a very complete report on the industry. Dr. H. W. Wiley, Chemist of the U. S. De- partment of Agriculture, has labored incessantly for the promotion of sugar-manufacture in this country, and has
iii
226830
17 PREFACE.
published many able and exhaustive reports upon the sub- ject. Mr. E. H. Dyer, after repeated disappointments which would have discouraged the bravest advocates of the sugar-beet, succeeded in establishing the Alvarado factory in California, the pioneer of the successful American beet- sugar houses. Mr. Claus Spreckels, through his large in- vestments in the Watsonville, Cal., works, and the prestige of his renown as a successful sugar-manufacturer, has given the advocates of the industry great encouragement. The work of Mr. Henry T. Oxnard gave renewed impetus to beet-sugar manufacture, and has been of material value in demonstrating its financial success when backed by thoroughly scientific and systematic preparations. Many others have done much to encourage the culture of the sugar-beet. Among these may be mentioned Mr. Lewis S. Ware, of Philadelphia, who has for several years pub- lished a journal devoted to the sugar-beet without other compensation than the satisfaction of encouraging a new and promising industry.
I take this opportunity of acknowledging many refer- ences to methods and suggestions given me by Mr. Ervin E. Ewell, Assistant Chemist of the U. S. Department of Agriculture, and of thanking him for many courtesies.
G. L. Spencer.
Washington, D. C, 1897.
TABLE OF CONTENTS.
References are to pages.
SUGAR-HOUSE CONTROL. General Remarks, i. The Basis of Sugar-house Control, 2.
WEIGHTS AND MEASURES. System of Weights, 3. Net Weight of the Beets, 4. Measurement of the Juice, 5. Automatic Recording Apparatus, 5. Various Methods of Measuring the Juice, 7. Calculation of the Weight of the Juice, 7. Auto- matic Determination of the Weight of the Juice, 8. Measurement and Weight of the Sirup, 9. Measurement and Weight of the First Massecuite, II. Measurement and Weight of the Second Massecuite, etc., 12. Sugar Weights, 13.
ESTIMATION OF LOSSES OF SUCROSE.
Division of the Season into Periods, 13. Loss in the Exhausted Cos.
settes, 15. Loss in the-'Waste- water, 16. Estimation of the Losses in the
Diffusion, by Difference, 17. Loss in the Filter P«ess-cake, 18. Loss in
the Evaporation to Sirup, 18. Loss in the Vacuum pan, 18.
SUGAR ANALYSIS. OPTICAL METHODS. The Polariscope, 20. Half-shadow Polariscope, 20. Triple-field Polari- scope, 23. Laurent Polariscope, 2^. ^ransition-tint Polariscope, Soleil- Ventzke-Scheibler, 26. General Remaoks on Polariscopes, 27. Manip- ulation of a Polariscope, 27. The Polariscopic Scale, 29. Reading thff Polariscopic Scale, 30. Preparation of Solutions for Polarization, 31. Adjustment of the Polariscope, 32. Notes on Polariscopic Work, 33. Error due to the Volume of the Lead Precipitate, 35. Scheibler's Method of Double Dilution, 37. Sach's Method of determining the Volume of the Lead Precipitate, 38. Influence of Subacetate of Lead and other Sub- stances upon the Optically Active Non-sugars, 38,
SUGAR ANALYSIS. CHEMICAL METHODS. Determination of Sucrose by Alkaline Copper Solution, 41. Determina- tion of Sucrose in the Presence of Reducing Sugars, 42.
SAMPLING AND AVERAGING.
General Remarks on Sampling and Averaging, 43. Sampling Beets in
the Field, 44. Subsampling of Beets in Fixing the Purchase-price, 45.
Sampling Beets at the Diffusion-battery, 47. Sampling the Fresh Cos-
8tfttes, 48. Sampling the Exhausted Cossettes, 49. Sampling Waste-
VI TABLE OF CONTENTS.
waters, 48. Sampling Dififusion-juice, 49, Sampling Filter Press-cake, 49. Sampling Sirups, 49. Preservation of Samples, 49. Automatic Sampling Juices, 50. Sampling Sugars, 54.
DENSITY DETERMINATIONS. APPARATUS AND METHODS.
Notes on Density, 55. Brix and Baume Scales, 55 Automatic Appa ratus for the Determination of the Density of the Juice, 55. Hydrometers or Spindles, 56. The Westphal Balance, 58. Pyknometers, 60.
ANALYSIS OF THE BEET. The Direct Analysis, 62. Scheibler's Extraction Method, 92. Stam- mer's Alcoholic Digestion Method, 64. Pellet's Aqueous Method, Hot Digestion, 65. Pellet's Instantaneous Aqueous Diffusion Method, 67. Determination of the Reducing Sugar, 68. Notes on the Direct Methods of Analysis, 69. Rasps and Mills for the Reduction of the Beet, 69. Indirect Analysis, 71.
ANALYSIS OF THE JUICE. Determination of the Density, 74. Special Pipette for Measurements in (he Sucrose Determiflations, 74. General Method for Sucrose, 75. Notes on the Clarification of Samples for Polarization, 77. Remarks on th2 Reducing Sugars in Beet Products, 78. Gravimetric Determination o. Reducing Sugars, 78. Volumetric Determination of Reducing Sugars, 84. Notes on the Determination of Reducing Sugars, go. Determination of the Total Nitrogen, Albuminoids, 92. Determinatibn of the Total Solids, P3. Acidity, 95. Analysis of Carbonated juice, 95. Alkalinity, 96. Rapid Methods of Moderate Accuracy for the Alkalinity, 96. Methods (or the Total Calcium, 99. Free and Combined Lime and Alkalinity due io Caustic Alkalis, Pellet's Method, loi.
ANALYSIS OF THE SIRUP. Analysis of the Sirup, 102,
ANALYSIS OF THE MASSECUITES AND MOLASSES.
Determination of the Density, 102. Density by Dilution and Spindling, 103. Total Solids and Moisture by Drying, 103. Total Solids and Coefficient of Purity, Weisberg's Method, 104. Determination of Sucrose »nd Raffinose, Creydt's Formula, 106. Sucrose and Raffinose, Lindet's Method, 107. Sucrose and Raffinose in the Presence of Reducing Sugar, no. Sucrose in the Presence of Reducing Sugar, Clerget's Method, no. Determinations to be made in the Analysis of Massecuites and Molasses. tii. Scheme for the Analysis of Massecuites and Molasses, in. Alkalinity of Massecuites and Molasses, 112. Estimation of the Proportion of Crystal- lized Sugar, 112. Notes on the Estimation of Crystallized Sugar, 117.
ANALYSIS OF SUGARS.
Analysis of Sugars, 118, Notes on the Analysis of Sugars, Massecuites, »nd Molasses, 119.
TABLE OF CONTENTS. VU
ANALYSIS OF FILTER PRESS-CAKE. Determination of the Moisture, 120. Total Sucrose, 120. Free and Combined Sucrose, 122.
ANALYSIS OF THE RESIDUES FROM THE MECHANICAL
FILTERS. Determination of the Moisture and Sucrose, 122.
ANALYSIS OF THE WASH AND WASTE WATERS. Determination of the Sucrose, 123.
ANALYSIS OF THE EXHAUSTED COSSETTES. Indirect Method for Sucrose, 124.
DEFINITIONS OF THE COEFFICIENTS AND TERMS USED
IN SUGAR ANALYSIS.
Coefficient of Purity, True and Apparent, 126. Glucose Coefficient, or
Glucose per 100 Sucrose, 126. Saline Coefficient, 126. Proportional
Value, 127. Apparent Dilution, 127. Actual Dilution, 127. Coefficient o(
Organic Matter, 127.
DETERMINATION OF THE MARC. Determination of the Marc, 128.
VISCOSITY OF SUGAR-HOUSE PRODUCTS. Viscosity of Sirups, etc., 130.
CONTROL OF THE OSMOSIS PROCESS. Analytical Work, 135. " J
ANALYSIS OF SACCHARATES. Saccharates, 137. Determination of the Sucrose, Lime, Strontium, and Barium, 137. Apparent and True Coefficients of Purity, 138. Analysis of Mother Liquors and Wash-waters, 138.
EXAMINATION OF BONE-BLACK. Limited Use of Bone-black in Sugar Factories, 139, Revivification, 139. Weight of a Cubic Foot of Bone-black, 139. Sulphide of Calcium, 140. Moisture, 140. Decolorizing Power of the Bone-black, 140. Determina- tion of the Principal Constituents, 141.
ANALYSIS OF THE LIME-KILN AND CHIMNEY-GASES. Analysis of the Gas from the Lime-kiln, 142. Simple Apparatus for Determining the Carbonic Acid, 146^ Analysis of the Chimney-gases, 147.
ANALYSIS OF LIMESTONE. Preparation of the Sample, 148. Determination of the Moisture, 148. Sand, Clay, and Organic Matter, 148. Soluble Silica, 148. Total Silica, 149. Iron and Alumina, 150. Calcium, 151. Magnesium, 152. Carbonic Acid, 153. Sulphuric Acid, 156. Notes on the Analysis of Limestone, »56.
Vlli TABLE OP CONTENTS.
ANALYSIS OF LIME. Determination of the Calcium Oxide, 15Q. Unburned and Slaked Lime 159, Calcium Oxide, Degener-Lunge Method, 159. Complete Analysis 160.
ANALYSIS OF SULPHUR.
Estimation of Impurities, 161.
ANALYSIS OF COKE. Preparation of the Sample, 162. Determination of the Moisture, 162. Ash, 162. Sulphur, 162.
LUBRICATING OILS. Tests applied to Lubricating Oils, 164. Cold Test, 164. Viscosity Test, 164. Tests for Acidity and Alkalinity, 165. Purity Tests, 165.
ANALYSIS AND PURIFICATION OF WATER. Characteristics of Suitable Water, 167. Analysis, 167. Purification, 171.
SEED-SELECTION. General Remarks, 174. Distribution of the Sugar in the Beet, 177. Methods of removing the Sample for Analysis, 177. Analysis of the Sample, 179. Pellet's Continuous Tube for Polarizations, 183. Polari- scope with Enlarged Scale, 184. Pellet's Estimate of Laboratory Appa- ratus and Personnel required for a Seed-farm, 185. Chemical Method for the Analysis of Beet-mothers, 187.
SEED-TESTING. Beet-seed, 190. Sampling, 190. Moisture, 191. Proportion of Clean Seed, t9t. Number of Seeds per Pound or Kilogram, 191. Germination Tests, 192. Characteristics of Good Seed, 195.
MISCELLANEOUS NOTES. • Cobaltous Nitrate Test for Sucrose, 197. Test for Sucrose, using o- Napthol, 197. Nitrous Oxide set free in Boiling Sugar, 198. The Precipi- tate formed in heating Diffusion-juice, i<^8. Spontaneous Combustion of Molasses, 198. Calorific Value of Molasses, 198. Fermentation, 199. Melassigenic Salts, 201. Chemical Composition of the Sugar-beet, 201. List of Reagents suggested for the Treatment of Beet-juice, 203.
SUGAR-HOUSE NOTES. Diffusion, 207. " Gray " Juice, 208. Carbonatation, 208. Sulphuring, 2IO. Filter-pressing, Difficulties, 210. Lime-kiln, 211. Granulation of the Sugar in the Vacuum-pan, 214. Second and Third Massecuites, 215. Gray Sugar, 215.
SPECIAL REAGENTS. Alkaline Copper Solutions, 216. Normal Solutions, 217. Pure Sugar,
222. Subacetate of Lead, 223. Bone-black, 223, Hydrate of Alumina,
223. Indicators of Acidity and Alkalinity, 224.
REFERENCE TABLES, 226.
BLANK FORMS FOR USE IN SUGAR-HOUSE WORK, 301.
LIST OF ILLUSTRATIONS.
FIGURE PAGE
1. Sugar-beet, showing Method of Topping. 4
2. Automatic Recording Apparatus, Horsin-Ddon 6
3. Automatic Scale, Baldwin 8
4. Diagrams showing Operation of Baldwin's Scale 9
5. Apparatus for determining the Weight of a Unit Volume of
Massecuite ix
6. Half-shadow Polariscope 21
7. Double Compensating (Shadow) Polariscope — 22
8. Triple-field Polariscope 23
9. Diagram illustrating Triple-field Polariscope 24
10. Laurent Polariscope 25
11. White-light Attachment for Laurent Polariscope 25
12. Soleil-Ventzke-Scheibler Polariscope 26
13. Lamp for Polariscopic Work 29
14. Polariscopic Scale 30
15. Weighing Capsule v I- j" 3*
i6. Filtering Apparatus /T 32
17. Control-tube 35
18. Diagram showing Method of Removing a Sample from a Beet. 46
19. Boring-rasp 46
20. Details of Boring-rasp 46
21. Automatic Sampler. Coombs 51
22. Automatic Sampler, Horsin-Ddon 53
23. Sugar-trier . 54
24. Automatic Apparatus for Density Determinations 56
25. Brix Hydrometer 57
26. Method of reading a Hydrometer 57
27. Westphal Balance 59
28. Pykaometer 60
29. Soxhlet-Sickel Extraction Apparatus 63
30. Knorr's Extraction-tube ; 63
31. Pellet and Lomont Rasp, side view.. 65
32. Pellet and Lomont Rasp, end view '. 65
33. Pelle', and Lomont Rasp, view from above 66
34. Section, showing Method of Sampling a Sugar-beet 66
35. Sugar- flask 66
36. Cylindro-divider 70
ix
LIST OF ILLUSTRATIOifS.
FIGURE PACK
37. Neveu and Aubin^s Rasp 71
38. Pulp-press 7a
39. Special Pipette for Use in Sucrose Determinations 75
40. Filtering-tube 7^
41. Apparatus for controlling the Current in Electrolytic Depo-
sitions 80
42. Automatic Zero Burette .. 85
43. Wiley and Knorr Filter-tubes 86
44. Muffle for incinerating Sugars 91
45. Muffle for incinerating Sugars 91
46. Muffle for incinerating Sugars 91
47. Vacuum Drying-oven 94
48. Vivien's Tube for Control Analyses in the Carbonatation 98
49. Vivien's Apparatus for Crystallized Sugar Determination 114
50. Kracz Apparatus for Crystallized Sugar Determination 114
51. Pellet's Apparatus for Marc Determinations 129
52. Doolittle's Viscosimeter 131
53. Engler's Viscosimeter 133
54. Orsat's Apparatus for Gas Analysis. 143
55. Knorr's Carbonic Acid Apparatus 154
56. Schroetter's Alkalimeter 155
57. Vilmorin's improved White Beet 176
58. Kleinwanzlebener Beet 176
59. Diagram showing the Distribution of the Sugar in the Beet 177
60. Diagram showing the Distribution of the Sugar in the Beet 177
61. Diagram showing the Distribution of the Sugar in the Beet.... 177
62. Lindeboom's Sound , 178
63. Details of Boring-rasp 179
64. Hanriot's Apparatus 180
65. Sach's-Le Docte Apparatus for Determination of the Sucrose in
the Beet 181
66. Automatic Pipette 182
67. Pellet's continuous Polariscope-tube 183
68. Polariscope for use in Seed Selection 185
69. Enlarged Scale for a Polariscope 186
70. Filtering Apparatus 186
71. Numbered Clamp... < 186
72. Antomatic Pipette 189
73. Seed Sampling-disk 190
74. Apparatus for Seed-testing • • i95
HANDBOOK
FOR
SUGAR-HOUSE CHEMISTS.
SUGAR-HOUSE CONTROL.
1. General Remarks.— The control of sugar-house work requires the analysis of the various products at each stage of the manufacture, and the tabulation of the results. From the data supplied by the analyses, the weights and measures of the raw material and the products, the chemist endeavors to trace the l^ses. The sugar received by the factory, in the beets, is clrarged on one side of the account, and that in the products and known losses is credited on the other side. The two sides of this account never balance owing to small unavoidable inaccuracies in methods, and to losses which cannot be located or measured.
The question of the detection, location, and estimation of the losses of sugar in the processes of the manufacture is often very complicated, and its solution requires the highest degree of skill on the part of the chemist. As the processes become more complicated through efforts to extract the uttermost grain of sugar from the beet, the difficulties which beset the chemist increase.
In many houses it is impossible to trace the losses quan- titatively, through lack of tank-room, etc.
The slightest analytical error will sometimes result in figures of negative value and necessitate their rejection. The so-called "losses fror.i unknown sou^ce^s," "undeter- minable losses," and." mechanical losses,' are probably in
2 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
many cases the result of unavoidable errors in weights and measures or in sampling and analysis.
. If an apparent loss be too large to be attributable to a reasonable allowance for error, it is well to view its exist- ence with doubt, until it is verified by repeated observations.
The work of the chemist is further complicated in sugar- houses which treat the molasses by a saccharate process, es- pecially a lime process in which the saccharate is used in liming the juice.
The adjustment of the analytical instruments should be frequently verified. The calibration of graduated ware should be checked. {See pages 231 and 250.)
The chemical control of a sugar-house does not end with the tracing and location of losses; it is also necessary to control the processes of manufacture. Each product should be studied, and the influence of each of the processes on the yield of the sugar noted. Slight modifications in the treatment of the material at various stages of the manu- facture are often suggested by the work of the chemist, and result in an increased yield of sugar.
Analytical data should be promptly obtained and tabu- lated, also all manufacturing data. Blank forms are given in pages 302 et seq. for permanent records for the chemist's use. The comparison of the data obtained in one period with those of another will always raise the questions, "Why is the yield of sugar smaller in one period than in the other ? " and " Why are the losses greater or less this week than last ?"
The writer has always made it a practice, in the control of sugar-house work, to divide the season into periods of one week each, and estimate the yield and losses, so far as practicable, in each. {See 14.)
2. The Basis of Sugar-house Control. — It is evident that sugar-house control must begin at a stage where the amount of sugar entering the factory can be accurately determined. In order to include the diffusion it must begin with the weight of the beets. The weight of the beets cannot be deduced with accuracy from the aver- age volume of a- definite tyeiglji of cuttings as measured in the diffusers.
SUGAR-HOUSE CONTROL. 6
The objections to the use of the net weight as determined by the deduction of the estimated tare from the gross weight are (i) the element of uncertainty due to an estimate, and (2) that portions of the beet, for which a deduction is made in the tare, reach the diffusion-battery.
In those countries where the clean beets are weighed as they enter the cutters, by the government officials, the con- trol should begin with the cuttings. This affords the only strictly reliable method of checking the work of the diffusion- battery, since the losses at this stage must be the difference between the weight of sucrose in the beets, as determined by analysis of the cuttings, and that in the diffusion- juice.
In the absence of the weights of the beets as indicated above, the control of the general work of the factory must begin with the weight of the diffusion-juice.
It is very probable that the so-called "losses from un- known sources," "mechanical losses," and "undetermined losses" are largely due to errors in weights and measures, and inaccuracies in sampMng and analysis, rather than to actual losses. f
This suggests that all instruments and graduated ware be carefully checked, and that weights of the raw material be adopted, instead of gauging, where practicable.
Claassen,* a prominent German authority, recommends the automatic scale constructed by Reuther & Reisert, Hennef, Germany, for weighing the beets immediately before they are sliced. He states that this scale is prefectly reliable.
The eminent French sugar engineer Charles Gallois has devised an apparatus which insures accurate weights. This apparatus is so arranged that the small car in which the roots are weighed cannot leave the scale unless it contain the correct weight of beets.
WEIGHTS AND MEASURES.
3. System of Weights.— In view of the fact that all chemists employ the metric system in their analvtical work,
1 Zeit. RUbenzucker-Industrte, 1895, 1084.
4 HAi^DBOOK FOR SUGAR-HOUSE CHEMISTS.
and that manufacturers in this country still adhere to the English, it is necessary in a work of this kind to use both systems of weights and measures.
4. Net Weight of the Beets. — The beets as re- ceived at the factory have been topped with more or less care, and have variable quantities of earth and pebbles ad- hering to them. These conditions necessitate the careful determination of an allowance for tare.
As nearly an average sample of the roots as is practicable is selected. This sample should consist of as many beets as can be conveniently taken, the larger the number the
Fig. I.
better. This number may afterwards be reduced by sub- sampling by the method of " quartering."
Thq roots are weighed, then thoroughly washed, using a brush to remove adhering soil and rootlets, and are then dried. A cloth may be used for drying them, but where many samples are to be examined it is usually more con- venient to dry the roots by exposure to a free circulation of the air for a short time.
SUGAR-nOUSE CONTROL. 9
The next operation is the removal of the neck or crown, i.e., that portion of the beet from just below the lowest leaf- bud. The cut should be made at the line shown in Fig. i.
The roots are again weighed, the difference between this weight and the first being recorded as the tare. The number of beets included in the sample and their average weight should also be recorded.
The beets, which have been employed in determining the deduction for tare, conveniently serve as a sample for analy- sis when the roots are purchased upon a basis of their sugar content. These roots, however, would not be a satisfactory average for calculating the sugar entering the factory.
5. Measurement of the Juice.— At the present time, the diffusion process has replaced all others in the ex- traction of the juice from the beet. This process requires that definite volumes of juice be drawn from the battery for definite quantities of beets.
The juice is drawn into a measuring-tank which is alter- nately filled and emptied. If this measurement be made with accuracy \nd reliable samples of the juice be drawn, a basis is supplitd for subsequent control work. Unfortu- nately this measurement as usually made is only an approx- imation. Errors are introduced through variations in the temperature of the juice and the difficulty of closing the inlet-valve at the proper instant. Hence special apparatus is essential to accurate measurement. This apparatus should be so arranged that it is wholly or partly automatic in its functions.
Whatever the system of tank measurements, it is essen- tial that the measuring-tank be carefully calibrated by means of a known volume of water rather than by calcula- tion. A slight error in the calibration is multiplied many times before the end of the manufacturing season.
6. Measurement oftlie Juice— Automatic Re- cording" Apparatus. — The errors mentioned above may be reduced to a minimum by a careful supervision of the battery temperatures, the use of automatic recording apparatus, and overflow pipes.
The apparatus illustrated in Fig. 2, the invention of
6
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Horsin-D6on, is largely used in France. It consists essen- tially of a paper-covered cylinder revolved by clockwork. A float in the measuring-tank is connected, by means of a wire or chain, with a drum which revolves when the float rises or falls : on the shaft of the drum is a pinion which in revolving engages a rack ; this latter in turn is attached to a small arm which carries a pen. When the juice enters the tank the float lifts, revolves the drum, and by means of the motion transmitted through the rack and pinion the pencil traces a line on the paper-covered cylinder. The paper is divided vertically into 12 parts, corresponding to the
Fig. 2.
hours. These parts are subdivided into 5-minute spacca. The cylinder makes one revolution every twelve hours. The sheet of paper is ruled horizontally into spaces of such width that each represents a certain volume of juice. It is evident from an inspection of the figure that the apparatus will record irregularities in the operation of the diffusion- battery. As the lines traced by the pen bear an invariable ratio to the depth of the tank, the volume of the juice may be deduced from the height of the "peak" of the curve above the base-line. Bell signals, also operated by the float.
SUGAR-HOUSE CONTROL. 7
warn the battery-man when the tank is filled nearly to the required point, or is almost empty. A counter records the number of times the tank has been filled.
It is advisable to provide an overflow-pipe, to prevent drawing more than a certain volume of juice.
Similar apparatus, constructed by Rassmus, is employed in German sugar-houses.
The automatic recording apparatus is often of great value in locating irregularities which may lead to losses.
7. Various Methods of Measuring the Juice. — Probably the most reliable method of measuring the juice is that adopted by the Belgian Government in connec- tion with the excise. This method consists essentially of a tank provided with an adjustable overflow-pipe, and a device for returning the overflow liquor, the volume of which is very small, to the battery. The inlet and outlet are at the bottom of the tank.
Several automatic measuring-tanks, more or less reliable, have been devised. The valves in the better class of these are operated by^ydraulic, steam, or air pressure.
8. Calcuhimon of the Weight of the Juice from its Volume. — The reference tables given in this book, except when otherwise stated, are referred to a tem- perature of 17^° C. This number is that adopted in the German sugar-houses and in the cane-sugar factories of this country as a standard. In view of these facts it is convenient to refer all sugar-house measurements to this temperature. The observed density of the juice should also be reduced to 17^° C.
The mean temperature of the juice at the time of meas- urement should be noted and the volume corrected for temperature. The juice expands practically at the same rate as a water-solution of sugar, hence Gerlach's table may be used in figuring the corrected volume (23 j).
The weight of a cubic foot of pure water at 17^° C. (63^° F.) is 62.348 pounds; the weight of one U. S. gallon of water (231 cu. in.) at this temperature is 8.335 pounds. These numbers, multiplied by the density of the juice, give respectively the weight of one cubic foot and of one gallon of juice. The calculations are facilitated by the table258.
8
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
9. Automatic Determination of the Weight of the «Tuice.— It is preferable to determine the weight of the juice by actual weighing when practicable. The automatic scale shown in Fig. 3 and in the diagrams (i, 2,
Fig. 3.
3, and 4), Fig. 4, is the invention of John Paul Baldwin, and was devised especially for sugar-house purposes.
The machine consists essentially of a revolving drum mounted upon a suitable scale. The liquid enters through the central pipe and flows into one of the compartments of the drum. When the weight of liquid for which the scale is set has entered the compartment, the liquid is automat- ically diverted to the s^ond compartment, the l«ad io
SUGAR-HOUSE CONTROL.
which soon revolves the drum so that the weighed liquid runs into the receiver beneath. The drum continues to re- volve until it assumes its original position.
|
1 |
||
|
HJ |
||
|
^ |
||
|
< |
^\nLLm^^^^ |
|
|
\ |
^^ |
|
' s |
|
|
Ci |
t\ |
|
/ |
|
|
3 |
^ |
|
|
^ |
f |
|
|
*y 1 11 r |
K |
|
|
« |
H |
|
|
\ ^^j^^:^^^:^ — '—j |
||
|
1 |
||
|
SJ |
T |
Fig. 4.
A counter records the number of weighings. A cup removes a small sample of the liquid from each load and stores it in a bottle, as shown in Fig. 3.
10. Measurement and Weight of the Sirup.—
The sirup is pumped from the multiple-effect evaporator to storage-tanks. It is not always easy to obtain accurate measurements of the sirup in these tanks. Rectangular tanks should be thoroughly stayed with rods. In case the tanks are bulged or uneven, it may be necessary to calibrate them by running in a measured volume of water. If the tanks are of uniform sectional area from top to bottom, they may be fitted with gauge-glasses similar to the water-
10 HANDBOOK FOR SUGAR-HOtJSE CHEMISTS.
gauges on a steam-boiler, except that the tubes should be of larger diameter, and may be graduated to any convenient scale. A stop-cock should be provided to cut off communi- cation with the tank, and a second cock to drain the sirup into a sample-bottle. The contents of the tank should be thoroughly mixed before admitting sirup to the tube. The sirup so obtained constitutes the chemist's sample. An electric signal-bell should be arranged to notify the work- man in charge of the tanks and the chemist each time a tank is filled. The reading on the scale is taken, the tem- perature noted, and the contents of the tube stored for analysis. The density and volume supply the data for cal- culating the weight of the sirup. A correction must be made to reduce the observed volume of the sirup to the standard conditions stated in 8. The calculations are facil- itated by the table 258.
The coefficient of expansion of an average sample of the sirup should be determined by experiment. This coefficient approximates that of a pure sugar solution (236), which for most purposes is sufficiently near the truth.
In sugar-houses which make a practice of drawing the sirup into the vacuum-pan from the tank into which the liquor is being pumped from the multiple-effect, it is neces- sary to provide special measuring and sampling apparatus.
An automatic measuring tank, such as is sometimes used in connection with diffusion-batteries, can readily be adapted for the purpose. The measuring-tank proper is fitted with inlet and outlet valves operated by hydraulic or steam pressure. The valves are controlled by means of a float acting upon a suitable lever, which in turn opens and closes the water or steam ports. A small storage-tank is also provided, the outlet from which is operated by a float. This tank must be large enough to allow ample time for the drainage of the measuring-tank. The sirup delivery- pipes should dip below the surface of the liquor.
The sirup may also be weighed directly by means of an automatic scale (9). A scale used for this purpose requires careful inspection at frequent intervals, especially when weighing very dense sirups.
These methods of ascertaining the weight of the sirup
SUGAR-HOUSE CONTROL.
11
complicate the sampling. An automatic sampler should be used (see p. 50).
1 1 . Measurement and Weight of First Masse- cuite. — Few sugar-houses have the facilities for obtain- ing the direct weight of the massecuite. Results based upon measurements should be received with caution. When practicable the weight should be ascertained by weighing the massecuite in sugar-wagons or in tanks.
The massecuite as it flows from the vacuum-pan is filled with bubbles which it is practically impossible to remove, hence the difficulty in obtaining a reliable direct determina- tion of the density for use in the calculation from volume to weight.
It is difficult to gauge the massecuite in tanks and obtain accurate measurements. When the weight must be de- duced from such measurements, it is advisable that the weight of a unit-volume be determined by some simple method, such as the following :
The massecuite is sampled from time to time as it flows from ihe^p^n, and the small portions drawn are united in a tall brass or copper cylinder, as shown in section in Fig. 5. S nc - there are great variations in the deiiaiiy of the massecuite in different parts of the pan, it is essential that great care be exercised in this sampling. The rim of the cylinder should be ground, and provided with a strip of brass or glass(CC), which extends from side to side and sup- pprt^ a capillary tube, as shown at T T' in the figure. Pins {P /*) should be placed in the rim of the cylinder and project through the strip, to insure replacing the latter always in the same position.
Surround the cylinder, filled with masse- cuite (J/), with hot water, and remove as many of the air-bubbles ^s possible; cool, dry, and weigh . Place the strip C, carrying the capillary tube (7") upon the cylinder ; Fig. 5.
add water ( W) from a burette, being careful to cause as few waves as possible, until the capillary tube is reached. The
12 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
instant the water reaches the tube, it rises some distance by capillarity, and affords prompt means of ascertaining when the vessel has been filled to a certain point. If water slightly colored with phenolphthalein be used, the rise of the water may be observed with ease. The difference between the volume of the cylinder to the capillary tube and the volume of water added is the required volume of the massecuite.
It is evident that this method can only be used in a build- ing free from vibrations. Under proper conditions, a measurement to within two or three tenths of a cubic centi- metre can be made by this method in a large cylinder.
A convenient-sized cylinder is 8 centimetres in diame- ter by 25 centimetres in depth, holding approximately 1500 grams of massecuite.
In houses where the massecuite is run into large rectan- gular tanks or into small portable tanks, the volume may be roughly approximated by the above method; but where the various forms of " crystallizers with movement" are used, or the massecuite is run directly into the mixer, the weight can only be calculated from the analysis and the volume of the lower products. In order to estimate approx- imately the loss of sucrose at this stage when " boiling in " is practised, the analysis and volume of the molasses used must be known. It is not possible to do more than closely approximate the loss without knowing the actual weight of the massecuite.
12. Measurement and Weight of the Second Massecuite, etc. — With modern methods of boiling first- sugar, i.e., "boiling in" molasses on first-sugar, there is comparatively little of the lower grades of massecuite made. Such massecuite is usually boiled on a footing of grained massecuite and then run into motion crystallizers; low material' is often boiled to "string-proof." The weight may be estimated from that of a unit volume. The measure- ment may be made in the tank after the massecuite attains approximately the temperature of the hot-rcom. A correction for expansion should be made, or the weight of a measured volume at the temperature of measurement should be deter- mined.
ESTIMATION OF LOSSES. 13
13. Sugar-weights.— The sugar-weights should be reported to the chemist for tabulation and for his use in calculating the yield and losses.
ESTIMATION OF LOSSES AND THE DIVISION OF THE MANUFACTURING SEASON INTO PERIODS.
14. Division of the Season into Periods.— In
factories which suspend manufacturing operations every Sunday, it is a simple matter to divide the season into periods of one week each, but in other factories it requires a systematic scheme of estimates to do this.
.The following plan has given excellent results in the hands of the author, and is suggested : Sunday is a conven- ient time for beginning a period; for example, let each period begin at 6 a.m. that day. At six o'clock the chemist and his assistants pass through the sugar-house and meas- ure and estimate the quantities of materials in stock.
This inch^des the measurement of the juice and sirup; an estimate of tne juice and sirup in the multiple effect and of the massecuite in process in the vacuum pans; the measurement of the massecuite in the crystallizers, mixers and centrifugals, and an estimate of the sugar in the centrifugals, hoppers, granulators, etc. The last package serial mmiber must be noted, or the quantity of sugar produced to the moment of stock-taking must be ascertained by other means. Where crystallizers are em- ployed, as is now usual, the massecuite is most conveniently and accurately measured at the time of discharging it from the pans. The measurement is made in the crystallizer. Prompt measure- ment is necessary, since the massecuite expands as the crystalli- zation progresses. The various materials should be sampled and analyzed.
From the quantity of material and its composition, the sugar value or probable yield of sugar is calculated, using the formula given on page 14. It should be noted that using apparent purities in the calculation only approximate results are obtained, also that losses in manufacture are less from massecuite to sugar than from juice or sirup:
14 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
looP-BM . , 1 ^ ,1
X = —3 rj— = percentage yield of granulated sugar from the
material; P, is the polarization of the material; B, its degree Brix; and M, its coefficient of purity. To adapt the formula to the calculation of raw sugar, substitute the following expression for the denominator: p—{SM -r- 100), in which p is the polariza- tion of the sugar and 5 the percentage of dry matter it contains. This formula gives the total sugar value, whether the product is obtained in one or more operations.
15. Loss of Sucrose in the Exhausted Cos- settes (Pulp). — In the analysis of the exhausted cossettes, the percentage of sucrose is expressed in terms of the cossettes. In order to calculate the loss of sucrose, it -is necessary to know the weight of exhausted cossettes per 100 pounds of beets. This number can only be accurately determined by actuaUy weighing the cossettes from a defi- nite weight of beets. This is manifestly impracticable, hence the chemist must necessarily base his calculations upon the average of a few weighings made each season.
It is also evident that different diffusion-battery condi- tions result in differences in the percentage of exhausted cossettes. The depth of the diffuser, the working temper- ature, the condition of the beets, the thickness of the cos- sette, and the use of water-pressure only or water-pressure and compressed air, all have their influence upon the weight of exhausted cossettes produced.
In general, it is usually considered that 100 pounds of beets, when working by water- pressure only, produce ap- proximately 90 to 100 pounds of well-drained exhausted cossettes, and working with compressed air, 100 pounds of beets produce approximately 80 to 85 pounds of exhausted cossettes.
16. Loss of Sucrose in the Waste Water.— It is not practicable to measure the waste water in the diffu- sion process. In order to figure the loss of sucrose at this stage of the manufacture it is necessary that this quantity be known; hence, being unable to ascertain it by actual measurement, it must be determined approximately by calculation.
ESTIMATION" OF LOSSES. 15
t he total volume of the diffuser and its connections must be known, also the weight and specific gravity of the ex- hausted cossettes.
It is more convenient to use the metric system in these calculations.
Calculation. Let X = the required volume of waste water in hecto- litres; D = specific gravity of the exhausted cossettes; tV = the weight of the exhausted cossettes per diffuser
in kilograms; F = the net volume of the diffuser in hectolitres, i.f., the volume between the upper and lower strainers;
X ■=V =- = the waste water in the net diffuser
iooZ>
in hectolitres.
To obtajK the total volume of the waste water, add the
calculated volume of the "dead space," i.e.y the space
.;.bove and below the strainers and of the parts of the pipes
which drain into the diffuser.
Example.
(A diffusion-battery using water-pressure only.)
Volume of the diffuser (net), hectolitres 30
Weight of the exhausted cossettes per diffuser,
kilograms 1300
\Veight of fresh cossettes per diffuser, kilograms.. 1530
Specific gravity of the exhausted cossettes 0.984
Ter cent sucrose in the waste water .05
Volume of the " dead space," hectolitres 2.5
W 1300
X ~ V = 30 — — 30 — 13.2 = 16.8 hectolitres,
\ooD 98.4
and 16.8 + 2.5 = 19.3 hectolitres total waste water. This
water contains so little solid matter in solution that its
specific gravity may be considered to be i, hence 19.3 hecto-
1930
litres of the waste water weigh 1930 kilograms or X 100
1530
~ 126 kilograms per 100 kilograms of beets. 126 X .05 -;- 100
16 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
= .063 kilogram of sucrose lost per 100 kilograms of beets or .063 pound of sucrose per 100 pounds of beets.
The quantity of sucrose in the waste water is so small that a very considerable error in figuring the volume of the waste water has but little influence.
With a battery employing compressed air, the volume of the waste water is very small, and is determined by deduct- ing the volume of diffusion-juice drawn from the volume of the waste water as calculated above.
In the above example, assuming a " draw " of 115 litres of diffusion-juice per 100 kilograms of beets, using com- pressed air, the volume of the waste water would be calcu- lated as follows :
15.3 X 115 = 1759-5 litres = 17.595 hectolitres of juice drawn and 19.3 — 17.595 = 1.705 hectolitres of waste water = 170.5 kilograms, or 11. i kilograms per 100 kilograms of beets. The loss of sucrose would be 11. i X .05 -r- 100= .0056 kilogram per 100 kilograms of beets or .0056 pound per 100 pounds of beets.
17. Estimation of the Losses of Sucrose in the Diffusion by Difference. — If it were always practicable to ascertain the exact weight of the beets enter- ing the diffusers, the simplest method of estimating the loss of sucrose in the diffusion would be by deducting the su- crose obtained in the diffusion-juice from that present in the beets, as ascertained by direct analysis. There are several probable sources of error in this method when not based upon the actual net weight of the beets. The tare (3) includes that part of the neck of the beet which should be removed in the field, but which has been left through care- less topping ; this passes into the diffusion -battery and contributes its sugar to the juice. This sugar increases the quantity in the juice without being charged to the beet supplying it.
In brief, except in houses where the beets are weighed immediately before they are sliced, the only method of de- termining the losses in the diffusion is by direct gauging and analysis of the waste products. It is always advisable to make these analyses.
Many chemists consider that there is usually some loss
ESTIMATION OF LOSSES. 17
through decomposition of sucrose in the battery. Such loss has not been clearly proven.
There is probably not often an appreciable inversion of sucrose in the diffusion of beets, except when there aj:e long delays.
In the event of inversion the loss may be calculated by the formulae used in cane-sugar-houses, which were first proposed by Dr. Stubbs of Louisiana (263).
18. Loss of Sucrose iu the Filter Press-cake.— The weight of the press-cake per ton of beets X per cent sucrose in the press-cake -4- loo = pounds of sucrose lost per ton of beets. In sugar-houses in which it is not con- venient to weigh the press-cake the approximate weight may be estimated by the following method : Weigh several entire press-cakes and figure the average weight; multiply the average by the number of cakes per press. A record must be kept of the number of presses emptied. The average w^ght should occasionally be verified.
19. Los^ of Sucrose in the Evaporation to Sirup. — An examination of the ammoniacal waters from the multiple-effect apparatus will sometimes reveal the presence of sucrose. It is practically impossible to esti- mate this loss from the analyses of these waters, since the weight of the water is unknown and the percentage of su- crose small. The quantity of sucrose lost is best determined by the difference between the weight of sucrose in the purified juice and that in the sirup. To obtain the weight of sucrose in the purified juice otherwise than by direct analysis, the loss in the filter press-cakes and at the mechanical filters must be deducted from the weight of sucrose entering the house in the diffusion-juices.
The following are some of the sources of loss of sucrose in the evaporation : Priming, i.e., juice entrained with the vapors ; caramelization and decomposition of the sugar. The liquors should always be alkaline, hence there is no loss from inversion.
20. Loss of Sucrose in the Vacuum-pan.— The estimation of the loss in the granulation of the sugar in the vacuum-pan is difficult. The sources of loss are the same as those in the multiple-effect. If the weight of the masse-
18 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
cuite can be accurately ascertained (11), the loss can be de- termined with certainty, as the weight of sucrose in the sirup should balance that in the massecuite. The "boiling in" of molasses with first-sugar complicates the determina- tion in so far as it requires that the quantity and analysis of such molasses be known. The weighf of sucrose in the massecuite can be ascertained indirectly when boiling "straight strikes " from the weight of sugar obtained and the volume of molasses produced, the weight of sucrose in the "wash" used in the centrifugals being deducted. In the event of its not being convenient to gauge the molasses, the measurement may be made after concentration to sec- ond massecuite, the loss indicated being that of the two boilings.
SUGAR ANALYSIS. OPTICAL METHODS. 19
SUGAR ANALYSIS. OPTICAL METHODS.
APPARATUS AND MANIPULATION,
21. The Polariscope. — The instrument employed in the optical methods of determining cane-sugar and other sugars is termed a polariscope, or saccharimeter. This instrument depends in theory and construction upon the action of sugar upon the plane of polarization of light.
Polariscopes may be divided into two general classes, viz., shadow and transition-tint instruments. The shadow instruments may be subdivided into polariscopes employing white light.-^s from an ordinary kerosene lamp, and those employing monochromatic light, supplied by a sodium lamp.
The principal instruments in use are the half-shadow, triple-field and the transition-tint polariscopes. The shadow instruments are constructed for use with white light and with the yellow monochromatic light. The former are usually employed in commercial work, and the latter in scientific investigations.
The transition-tint instruments are being rapidly dis- placed by the shadow polariscopes, since these latter leave little to be desired in the matter of accuracy and conven- ience.
The reader is referred to the manuals of Wiley and others for the theory and construction of polariscopes.
A brief description of the polariscopes in general use will suffice for the purposes of this book.
22. Half-shadow Polariscope (Schmidt and Haensch). — The optical parts of this instrument are in- dicated in Fig. 6. At O there is a slightly modified Jellet-" Corny Nicol prism, at G is a plate of dextrogyratory quartz, at i? is a quartz wedge, movable by means of the screw M, and at /^ is a quartz wedge, fixed in position, to vrhich is attached the vernier. The scale is attached to the
20
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
movable wedge. These quartz wedges are of laevogyratory quartz. The parts G, E, and /^constitute the compensating apparatus, i.e.\ the apparatus which compensates for the deviation of the plane of polarization due to the influence of the solution of the optically active body placed in the
observation-tube as shown in the figure. At H is the ana- lyzer, a Nicol prism. Aty is the telescope used in making the observation, and K is the telescope and reflector for reading the scale. The two lenses, shown in the diagram at the extreme right, are for concentrating the rays •£
SUGAR ANALYSIS. OPTICAL METHODS.
21
light from the lamp and transmitting them in parallel lines to the polarizing Nicol prism.
The instrument above described is of the single compen- sating type.
The double compensating instrument is shown in Fig. 7. This polariscope differs from the single compensating in- strument in having two sets of quartz wedges of opposite optical properties and two scales and verniers.
The field of vision of the above instruments when set at
22
HANDBOOK FOK SUGAR-HOUSE CHEMISTS.
the neutral point is a uniformly shaded disk. If the milled screw controlling the compensating wedge be slightly turned to the right or left, one half the disk will be shaded and the other light. It is from this half-shaded disk that this type of instruments takes its name.
23. Triple - Field Polariseope (Schmidt and Haensch). — This instrument differs from the preceding in
having two small Nicol prisms placed in front of the polar- izer, as shown in Fig. 8. The field of the instrument is divided into three parts, i, 2, and 3 of the diagram, Fig. 9. This figure shows the arrangement of the Nicol prisms (i, II, III) and a diagram of the field of observation.
SUGAR AifALYSIS. OPTICAL METHODS.
23
When the scale is set at the zero point, no optically active body being interposed, the field is uniformly shaded ; in other posi- tions I is shaded and 2 and 3 are light, or vice versa. This arrange- ment permits a very high degree of accuracy in the adjustment of the field in polariscopic observations. Ac- cording to the experiments of Wiley ' this instrument is extremely sensitive and is capable of results but little in- ferior to those with the Landolt- Lippich apparatus. It is probably the superior, in point of accuracy, to other instruments designed for industrial work.
24. Liaurent Polariscope. — The Laurent "pi?3+^iscope (Fig. 10) is a half-shadow instrument. It was originally designed for use with a monochromatic flame, but these in- struments, as now made, are provided also with compen- sating apparatus for use with white light.
In the Laurent polariscope the analyzer is revolved by means of a milled screw, to compensate for the deflection of the plane of polarization by the sugar solution. The angular rotation is measured by means of a scale and ver- nier. This instrument is also provided with a second scale, termed the cane-sugar scale, on which the per cents may be read directly.
As stated above, the Laurent polariscope is also often provided with a compensating apparatus (Fig. 11), which permits the use of white light.
A distinctive feature of the Laurent instrument is the adjustable polarizer. This Nicol prism may be rotated through a small angle, thus permitting the sensitiveness of the instrument to be varied.
The polarized light is passed through a disk of glass,
1 Agricultural Analysis, 3, gi.
24 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Fig. iz,
SUGAR ANALYSIS. OPTICAL METHODS.
25
one half of which is covered with a thin plate of quartz,
thus producing the half-shadow feature of the instrument.
25. The Transition-tint Polariscope. Soleil-
Ventzke-Scheibler, — The tint polariscope, Fig. 12,
resembles in appearance the half-shadow instrument of Schmidt and Haensch. It differs from this in being pro- vided with an additional Nicol prism at A and a quartz plate B, which produce the color. The tint is varied by means of a spur-wheel and pinion, revolved by a rod with a milled head, Z. The optical parts at the front end of the
26 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
instrument are the same as in the Schmidt and Haensch half-shadow polariscope.
The field is colored, and when the instrument is set at the neutral point the tint is uniform. The sensitive tint for most eyes is a rose-violet.
26. Genertil Remarks upon Polariscopes.— The Laurent polariscope is very extensively used in France, and to but a limited extent in other parts of Europe and in this country. The tint instruments were formerly used almost exclusively, but have been largely replaced by the various forms of half-shadow polariscopes. Tint instru- ments, obviously, cannot be used by persons who are color-blind.
All polariscopes are made to receive observation-tubes of various lengths. The standard length is 200 millimetres.
There are several forms of polariscopes in addition to those described, but for industrial work it is unnecessary to mention others.
27. Manipulation of a Polariscope.— Having dis- solved the normal weight (28) of the material under examin- ation in water, clarify the solution as described in 30. Fill the observation-tube with a portion of the clarified solution, and pass the light froiii a suitable lamp into the instrument. The observer, with his eye at the small telescope y of the Schmidt and Haensch instruments, Figs. 6, 7, 8, and 12, or the corresponding part of the Laurent, will notice that one half the disk is shaded or more deeply colored, according to the kind of polariscope, provided the instrument is not set at the neutral point. The vertical line dividing the half-disks should be sharply defined; if not, the oculai' should be slipped backwards or forwards until a sharp focus is obtained. Turn the milled screw until the field appears uniformly shaded or tinted on both sides of the vertical line, and then read the scale (29). A little prac- tice will enable the observer to detect very slight differ- ences in the depth of the shadow or color and to attain great accuracy in this manipulation.
The manipulation of the triple-field polariscope is as described above except as to the position of the shadows (23).
SUGAR ANALYSIS. OPTICAL METHODS. 27
In the older models of polariscopes, the ray filter, a dichromate of potash crystal in an occular, should be used with very clear solutions. In recent models the filter consists of a glass cell containing the salt in solution. This filter is placed in a chamber in front of the polarizer, and the instrument should only be used with it in place.
The Laurent instrument is fitted with a device for vary- ing its sensitiveness. This is convenient in polarizing dark-colored solutions, since a slight change in the position of the lever which rotates the polarizer will increase the intensity of the light, though at the same time decreasing the sensitiveness of the instrument: and vice versa in polar- izing very clear light-colored solutions, the rotation of the polarizer in the opposite direction, through a small angle, increases the sensitiveness.
The double compensating Schmidt and Haensch polari- scopes are"\>rovided with two scales, one graduated in black and tRe other usually in red. The black scale is operated by a black milled screw, and the red scale by a brass screw. For ordinary work, set the red scale at zero and equalize the field with the black screw. To check the readings, remove the observation tube and equalize the field with the brass screw. The readings on the two scales should agree. To make a reading with laevorotatory sugar set the black scale at zero, and use the brass screw and red scale.
The manipulations of the tint instruments, as explained, are similar to those of the shadow polariscopes, except that a uniform tint must be obtained. The intensity of the tint varies with the position of the analyzer. The color is varied by turning the milled screw on the horizontal rod which revolves the regulator.
The Schmidt and Haensch shadow polariscopes, the Laurent with special attachment, and the tint instruments require a strong white light. A kerosene-lamp with duplex burner is usually employed. A gas-lamp such as shown in Fig. 8 is very convenient in many localities. The kerosene-lamp should be provided with a metal chimney.
Dr. Wiechmann uses the Welsbach light in his labo-
28
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
ratory at the Havemeyer & Elder refinery, Brooklyn, and finds it very satisfactory. Dr. Wiley of the U. S. Department of Agriculture has investigated the use of the light from acetylene gas for polariscopic purposes, and states that the readings obtained are very accurate and that the light is espe- cially convenient when polarizing very dark-colored solutions. This gas is readily and economically produced, in the small quantities required for polariscopic purposes in sugar-works, by the decomposi- tion of calcium carbide in water, in a suitable gas-holder. Incandescent electric-lamps, properly arranged for the diffusion of the light, yield ex- cellent results with white-light polari- scopes. In new models of polari- scopes the electric lamp is attached to the instrument. Lamps for mono- chromatic light are the Laurent gas- sodium and the Landolt gas-sodium lamps, and the Laurent eolipyle, burning alcohol. F'G. 13. M. Dupont ' has recently experi-
mented with various sodium salts for use in monochromatic lamps. He finds that sodium chloride and tribasic phosphate of sodium, melted together in molecular proportions, give excellent results and are in every way superior to sodium chloride alone.
28. The Polariscopic Scale. The Normal Weight. — The scales of polariscopes for use in industrial work are usually so divided that if a certain weight of the substance be dissolved in water and the solution diluted
* Bulletin de V Association des Chiniistes de France^ 14, 1041
SUGAR ANALYSIS. OPTICAL METHODS.
29
to lOO cc.,* and observed in a 20-centimetre tube, the read- ing will be in percentages of sucrose. This scale is termed the "cane-sugar scale," and the weight of material re- quired to give percentage readings is termed the "normal weight," or sometimes the " factor of the instrument."
In commercial work the divisions of the scale are often termed "degrees," especially in the polarization of sugars.
The normal weight for the German instruments is 26.048 grams, and for the Laurent 16.29 grams. The number given for the Laurent polariscope is that adopted by the 2^ Congres International de Chimie Appliqude, 1896.
29. Readings the Polariscopic Scale.— Having equalized the shadow or tint as directed in 27, examine the scale through the reading-glass. For example : Let the scale and vernier have the positions shown in Fig. 14.
I
20
30
40
u
I ,1.1,1,1
mill
I
m
10
10
Fig. 14. The zero of the vernier is between 30 and 31; record the lower number; note the point to the right at which a line of the vernier, the small scale, corresponds with a line of the scale, in this case at 7; enter this number in the tenths place. The completed reading is 30.7. The portions of the scale and vernier to the left of the zeros are used in the polarization of laevorotatory bodies. If the zero of the ver- nier correspond exactly with a division of the scale, the reading is a whole number.
If the normal weight of the material have been dissolved in a volume of 100 cc.^ and a 20-centimetre observation-tube have been used, the reading on the cane-sugar scale is the percentage of sucrose in the substance, provided other op- tically active bodies than sucrose are absent. The read-
.. * The flasks should be graduated to hold 100 grains of distilled water at i7i° C. and not to true cubic centimetres, for the S. and H. instruments, but to true cc. for the Laurent.
30 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
ings must be corrected for other weights of the substance than the normal,^ for other volumes than lOO cc, and for other tube lengths than 20 centimetres.
30. Preparatiou of Solutions for Polariza- tion.— Dissolve the normal or other convenient weight of the material in water. Add sufficient subacetate of lead to clarify the solution. It is difficult to specify the amount of the lead salt to use. If too little or too much be used, the solutions usually filter with difficulty and become turbid. With juice from immature beets the filtered solution will sometimes be perfectly clear and colorless when first ob- tained and in a few moments become too dark to polarize. Ih such cases the juice should be thoroughly mixed, with the lead solution and stand some time before filtration. Usually ID cc. of the dilute solution "^ subacetate of lead (207) or 2-3 cc. of the concentrated solution (208) will be sufficient for 100 cc. of beet juice. Sugars of high grade require only a few drops of the reagent. After adding the lead salt dilute to 100 cc, mix thoFougbly and filter. Reject the first few drops of the filtrate. Fill the observation- tube with a portion of the filtrate, and polarize as described in 27.
In sugar analysis the materials to be examined are most conveniently weighed in a nickel or German-silver capsule such as is shown in Fig. 15. A convenient filtering ar- rangement is illustrated in Fig. 16. ^ is a stem- less funnel; B is a. quar- ter-pint precipitating jar; (7 is a small cylinder. The stemless funnels may Fig. 15. be made of tinplate or
thin copper, planished. The latter, while more expensive, are preferable, as they are more durable. A plain cylin-
'-i*t*fie Tnternational Commission for Uniform Methods in Sugar Analysis has adopted a normal weight of 26 grams to be used with a flask holding 100 metric or true cubic centimeters; the polariscope Ts ptandardized at 20^ C.
SUGAR ANALYSIS. OPTICAL METHODS.
31
der is preferred by some chemists, as the funnel makes a close joint with the edge.
The advantage of the metal stemless funnels and the heavy glass precipitating jar or the lipped cylinder is the ease with which they may be washed and dried. The jar or cylinder is also a very convenient support for the funnel.
Stammer and Sickel advise the addition of at least four times the weight of the sucrose in the massecuite or mo-
FiG. i6.
lasses, of strong alcohol in preparing solutions for polariza- tion, and if the substance be alkaline to acidulate with acetic acid.' Herzfeld, as the result of his experiments, gives the same advice.^ Other equally prominent chemists con- sider the use of alcohol unnecessary and liable to lead to error.
31. The Adjustment of the Polariscope.— The scale of the polariscope is the only part which is liable to get out of position. Fill an observation-tube with water and make an observation. If the scale be properly adjusted the reading should be zero.
The method of adjusting the instrument to read zero under the above conditions is the same with all the Schmidt
* Revue Universelle de la Fabrication du Sucre, ad year, 578.
* Deutsche Zuckerind.y 1886, No. 24.
32 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
and Haensch polariscopes. A micrometer-screw, turned by means of a key, is arranged to move the vernier a short dis- tance. The field is equalized as usual by manipulating the milled screw. The micrometer-screw is then turned until the zeros of the scale and vernier coincide. The scale is moved through several divisions and the field then equal- ized as before. If after several trials the zeros be found not to coincide, the adjustment must be repeated, turning the micrometer-screw very little. It usually requires several trials to set the instrument to read zero. This adjustment is very fatiguing to the eye, which should be rested a few seconds between readings.
In adjusting the Laurent polariscopes to read zero, the lever U, Figs. lo and ii, is lifted to the upper limit; the oc- ular O is next focused on the vertical line which divides the field into halves ; the zeros of the scale and vernier are made to coincide by the screw G. The field should then be uniformly shaded if the instrument is in .adjust- ment ; if not in adjustment, equalize with the screw F. This adjustment should be tested as with the Schmidt and Haensch polariscopes, and repeated until satisfactory. All parts of the instrument should be kept very .clean, especially the exposed parts of the lenses. Chamois-skin is con- venient for cleaning the metal parts and pieces of clean old linen for the lenses. All the crown-glass lenses should be occasionally removed from the instrument and cleaned with alcohol and wiped with old linen. The Nicol prisms should not be removed from the instrument or disturbed. The micrometer screw near H^ Figs. 6, 7, 8, and 12, is for adjusting the analyzer, should the field be unevenly shaded. This adjustment should be left to an experienced workman. Should the prisms, etc., require adjustment owing to an accident to the instrument, it is advisable to send the polariscope to the dealer that he may have it re- paired by an expert.
32. Notes on Polariscopic AVork.— When solu- tions do not filter readily, the funnel employed should be covered with a glass plate to prevent evaporation.
The screw-caps of the observation-tubes should not bear heavily upon the cover-glasses, since glass is double-
SUGAR AKALYSIS. OPTICAL METHODS. 33
refracting under these conditions. It is preferable to use caps held with a bayonet-catch rather than screw- caps.
In making an observation, the eye should be in the optical axis of the instrument, and should not be moved from side to side.
The cover-glasses should be of the best quality of glass, perfectly clean and with parallel surfaces. A glass may be tested by holding it in front of a window and looking through it at the window-bars ; on turning the glass slowly, if the bars appear to move the surfaces of the cover are not parallel and the glass should be rejected. Old glasses which have become slightly scratched by repeated wiping should not be used.
The planes of the ends of the observation-tube should be perpendicular to the axis of the tubes. This may be tested by placing a tube, containing a sugar solution, in the instrument and making an observation. On revolving the tube in the trough of the polariscope, should the readings in different positions vary, the ends of the tube have not been properly ground.
The manufacturers of polariscopes have attained such precision in their methods that errors in the adjustment of the instruments or accessories are rarely found.
The polariscope should be used in a well-ventilated room from which all light, except that from the polariscope-lamp, is excluded. It is an excellent arrangement to have the lamp in an adjoining room and pass the light through a glass screen to the instrument. Late models of the Ger- man polariscopes have mirrors arranged to reflect the light to the scale. When the instrument is in a room adjoining the lamp-room, obviously the above arrangement cannot be used. A small gas-jet or a candle should never be used for lighting the scale. A convenient source of light is a half- candle-power incandescent electric lamp mounted near the scale and switched into the circuit by an ordinary push- button. The lamp may be operated by a two-cell accumu- lator or in circuit with a 32-candle-power incandescent lamp, the latter being outside the polariscope-room. Ord'Pary
34
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Leclanche cells are cheap and will answer for several hun- dred polarizations.
The instrument should be occasionally tested with pure sugar(206). or more conveniently with standardized quartz plates, to be obtained of the makers.
Messrs. Schmidt and Haensch construct a control-tube, Fig. 17, with which all parts of the scale may be tested. The sugar solution is poured into the funnel T' and flows into the tube as it is lengthened by turning the milled screw. The tube length is read on the scale N. The figure describes the tube sufficiently. ^*-^
Errors which may occur in the polarization, but not through faulty manipulation of the instrument, are indi- cated in the following paragraphs.
Fig. 17.
33. Error Due to the Volume of the L^ead
Precipitate. — The lead precipitate, formed in the clarifi- cation of the solutions, introduces errors in the polarization, some of which are probably offset by compensating errors, notably in the analysis of low-grade products.
An important error is that due to the volume of the lead precipitate. This question has been studied by a number of chemists, notably by Scheibler in Germany and Sachs in Belgium. Scheibler devised a simple method for the cor- rection of this error, which is commonly termed the *' method of double dilution." It was noticed by Rafey, Pellet, Commerson, and others that in low-grade products, the saline coefficient of which is large, there is apparently
SUGAR ANALYSIS. OPTICAL METHODS. 35
no error due to the volume of the precipitate, which is very large. They attributed this fact to an absorption of sucrose by the precipitate at the moment of its formation. Sachs' published an exhaustive paper on this question some years since, and demonstrated that there is no absorption of su- crose. He attributed the results with low products to the influence of acetates of potassium and sodium, formed with the acetic acid set free in the decomposition of the lead salt, upon the rotatory power of the sucrose. This view is strengthened by the fact that there is a very perceptible error, in the polarization of juices, due to the precipitate. The precipitate in juices contains but little of the acetates of potassium and sodium, whereas these salts are formed in considerable quantities in molasses and low products.
Sachs* experiments were made by increasing the concen- tration of the solutions instead of by dilution as practised by Scheibler. Sachs dissolved x grams of molasses in water, added sufficient subacetate of lead for clarification, completed the volume to loo cc. and polarized as usual. The quantity x increases from experiment to experiment by practically equal increments. Since the quantity of molasses is increased with each experiment, the volume of the precipitate must increase in the same ratio. An in- crease in the volume of the precipitate, if this were the only disturbing influence, should increase the polarization, since the volume of the solution is decreased and the con- centration is increased.
Letting x — the weight of molasses, and^ = the polari-
scopic reading, the ratio — should increase with each incre- ment of molasses if there be an error due to the volume of the precipitate, not compensated for by other influences. Sachs employed quantities of molasses ranging from 5 to 35 grams in 100 cc, and substituting the values of x and ^ in the ratio and reducing, obtained the following figures:
ist series : 1.906, 1.900, 1.900, 1.906, 1.896; 2d series : 2.14, 2.13, 2.14, 2.14
^fvtfe UniverselU de la Fabrication du Sucre, 1>45I<
36 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
The practically constant value of - shows that a minus
error or errors have fully compensated for that due to the volume of the precipitate. Sachs' deductions are given above.
A similar experiment with beet-juices gave the foUow-
y
ing values of -:
X
ist series : 0.5446, 0.5474, 0.5480, 0.5497TV 2d series : 0.5800, 0.5830, 0.5842, 0.5860.
It is thus shown that there is an increase in the ratio and an error due to the volume of the precipitate in the analy- sis of juices. The volume of the lead precipitate from 100 cc. of normal juice is approximately i cc.
It is not improbable that at least to some extent the so- called "losses from unknown sources" in sugar-house practice are due to errors in analysis which, with our present information, are unavoidable.
34. Error Due to the Volume of the Lead Precipitate — Scheibler's Method of Double Dilution. — The error due to the volume of the lead precipitate may usually be determined by Scheibler's' method.
To 100 cc. of the juice add the requisite quantity of subacetate of lead, complete the volume to 100 cc. and polarize as usual; a second portion of 100 cc. of the juice is treated with lead as above, diluted to 220 cc. and polarized.
Calculation. — Multiply the second reading by 2, subtract the product from the first reading, multiply the remainder by 2.2, and deduct this product from the first reading. The remainder is the required per cent sucrose.
• Zeit,^ Rubenzucker -Industrie^ 85, 1054.
SUGAR ANALYSIS. OPTICAL METHODS. 37
Example,
Degree Brix of the juice l8.
First polariscopic reading (no cc.) 57.6
Second polariscopic reading (220 cc). ....... 28.7
2 X 28.7 = 57.4; 57-6 — 57-4 = .2; 2.2 X .2 = .44; 57-6 - .44 = 57.16, = the corrected reading. By Schmitz' table, as described on page 76, we have
15.18 .03 .02
15.23 = required per cent.
In the application of this method to other products using the normal or multiple-normal weight, calculate as follows:
1st volume, 100 cc; 2d volume, 200 cc.
Multiply the second polariscopic reading by 2 and sub- tract the product from the first reading; multiply the re- mainder by 2 and subtract the product from the first reading. This remainder is the required per cent sucrose.
It is evident that this method requires extreme care in the polarization, since an error is multiplied.
35. Sachs' ' Method of Determining the Vol- ume of the Lead Precipitate.— Clarify 100 cc of juice with subacetate of lead as usual, using a tall cylinder instead of a sugar-flask. Wash the precipitate by decanta- tion, first using cold water and finally hot water. Continue the washing until all the sucrose is removed. Transfer the precipitate to a loo-cc sugar-flask and add the one-half normal weight (13.024 grams) of pure sugar, dissolve and dilute to 100 cc, mix, filter, and polarize, using a 400-mm. tube.
op. cit., 1, 451.
38 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Calculation.
Let P = the per cent of sucrose in the sugar;
jP' = the polarization of the solution in the presence
of the lead precipitate; X = volume of the lead precipitate.
-, lOOP' — lOOP _
■f hen X = ;; . X
Example,
tet P = 99.9;
P' = 100.77.
^. _ 100 X 100.77 — 100 X 99-9
1 fltftii X —
100.77
= .86 cc, the volume of the lead precipitate.
3^. Yniiueiice of Subacetate of Lead and Other Substances upon the Suj^rars and Optically Ac- tive Non-sugars 1 in Beet Products. — Sucrose.— The rotatory power of sucrose in aqueous solution is not modified by subacetate oi lead under the conditions which usually obtain in analysis. In the presence of a very large excess of the lead salt there is a slight diminution in the rotatory power; there iy a decided diminution in alcoholic solution in the presence of the lead salt.
Farnsteiner^ made the following observations relative to the influence of certain inorganic salts:
" With a constant relation of sugar to water, the chlorides of barium, strontium, and calcium cause a decrease in the rotation which continues to decrease as the salt is increased; calcium chloride causes a decrease, but when %he salt reaches a maximum further addition causes an
* The beet and beet products contain other substances which are opti- cally active in addition to those given here, but the quantities present are exceedingly small and would not appreciably influence the analytical results. The following optically active substances are also present : tar- taric acid, leucine, coniferine, and cholesterine.
2 Berichte deut. chem. Gesel.^ 33, isjo\ Journ. Chem. Soc, ^O, 283.
SUGAR ANALYSIS. OPTICAL METHODS. 39
increase which finally exceeds that of the pure sugar solution.
" If the relation of the sugar to that of the salt be kept constant, it is found that the addition of water causes in all cases an increase in the specific rotatory power, i.e., the action of the salts is lessened. The specific rotatory power is almost unaffected by varying the quantity of sugar with a constant relation between the salt and water. The chlorides of lithium, sodium, and potassium behave in a similar manner.
"An examination of the action of the same quantities of different salts shows that in the case of strontium, calcium, and magnesium the depression varies inversely with the molecular weight, and that the product of the two quanti- ties is approximately a constant. Barium chloride does not act in the same manner, but the chlorides of the alkalis show a similar relation. The relation, however, only holds good within each group of chlorides and not for two salts belonging to different groups."
The rotatory power of sucrose in water or alcohol solu- tion is not modified by the presence of nitrates of sodium and potassium even when the quantity of the nitrate amounts to as much as 50 per cent of the sucrose (E. Gravier).
In investigating the influence of the lead precipitate (33), Sachs found that the presence of acetate of potas- sium very perceptibly diminished the rotation. The diminu- tion was also noticeable with the sulphates of potassium and lead, but was not so marked with the corresponding sodium salts. Sachs also states that he has demon- strated that citrate of potassium, carbonate of sodium, and several other salts have an influence analogous to that of the acetates. The presence of free acetic acid reduces this influence in part. Sachs, in the same paper, urges that the use of tannic acid in decolorizing solutions is very objec- tionable, on account of the volume of the precipitate formed with the lead.
Dextrose. — The rotatory power of dextrose is not modi- fied, or, if at all, but very slightly, under ordinary analytical
40 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
conditions by either the subacetate or the neutral acetate of lead. See also Invert-sugar.
Levulose. — The rotatory power of levulose is very greatly diminished by the presence of subacetate of lead. Under certain conditions, a levulosate of lead is probably formed. This levulosate is precipitated in the presence of certain chlorides, in quantities more or less considerable according to the relative proportions of the salts, lead, and levulose. There is no precipitation by the normal acetate of lead (Gill, Pellet, Edson, Spencer).
Invert-sugar. Dextrose and Levulose. — In the presence of the salts formed in the decomposition of the subacetate of lead, dextrose and levulose are precipitated in part (Pellet, Edson). The influence of the basic lead salt on the rotatory power of levulose {see above), or the formation of a levulosate of lead of little optical activity, gives undue prominence to the dextrose and results in a plus error. In 1885 the author recommended the acidulation of solutions containing invert-sugar with acetic acid. This restores the normal or nearly the normal rotatory power to the levu- lose.
Acetic acid slightly lowers the rotatory power of invert- sugar; hydrochloric acid has an opposite effect. Sodic acetate and sodic chloride increase the rotation (H. A. Weber and Wm. McPherson). Sulphuric and hydrochloric acids increase the rotation; oxalic acid has no effect. The rotation increases as the quantity of mineral acid is increased.'
Raffi,nose. — The rotatory power of raffinose is greatly diminished in concentrated solution by subacetate of lead in large quantity, and not at all in dilute solution, especially in the presence of sucrose. The normal rotation is restored by slight acidulation with acetic acid (Pellet). Raffinose is precipitated by highly basic subacetate of lead as readily as with ammoniacal acetate of lead solution (Svoboda).
Asparagine. — Not precipitable by subacetate of lead, but is rendered dextrorotatory, instead of Isevorotatory, by the
* Gubbe, Bulletin Assoc. Chimistes de France^ 3, 131.
SUGA.R ANALYSIS. CHEMICAL METHODS. 41
lead salt. Asparagine is insoluble in alcohol, and in the presence of acetic acid is inactive (Pellet). In neutral and alkaline solution, laevorotatory; in presence of a mineral acid, dextrorotatory; in the presence of acetic acid the rotation is diminished and with lo molecules of the acid becomes o°, and with additional acid dextrorotatory (Degener).
Aspartic Acid. — From asparagine by the action of lime; the lime salt is soluble. In alkaline solutions aspartates are laevorotatory, and acid solutions dextrorotatory ; aspar- tic acid is precipitated by subacetate of lead.
Glutamic acid is dextrorotatory, and in the presence of subacetate of lead it becomes laevorotatory. Not precipi- tated by lead acetate except in the presence of alcohol.
Malic acid is laevorotatory. The artificial malic acid is optically inactive. Malic acid is precipitated by subacetate of lead.
Pectine ajid parapectine are dextrorotatory and are both precipitated by subacetate of lead, and the latter by normal acetate of lead.
36a. Bone-black Error. — The use of bone-black or animal chav- coal is to be avoided when possible, since it absorbs sucrose. The degreeof absorption varies with the charcoal from different sources.
Where the use of this substance is necessary for bleaching dark -colored samples it should be used in small quantities, pref- erably placing about 3 grams of a finely powdered char in a filter. The solution should be poured onto the charcoal in suc- cessive portions, each of which is allowed to drain from the char before adding the next. The first five or six portions of the filtrate should be rejected and subsequent fractions utilized in the analysis, as by this time the char can absorb no more sucrose.
Many chemists advise adding from 0.5 to 3 grams of finely powdered dry bone black, per 100 cc. of solution, to the material in the sugar-flask. After shaking the mixture thoroughly and letting it stand a few minutes, the char is removed by filtration. The charcoal should previously be tested with a solution of known sucrose content to ascertain its absorbent power, that a correction may be applied.
36b. Temperature Errors in Polarizations. — According to
42 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Wiley * a correction should be made for errors due to variations of temperature of 0.03 per cent, per degree above or below the normal temperature (17^° C. for the older and 20° C. for recent models of polariscopes). This correction is to be added for tem- peratures above the normal and sub^acted for those below it. This correction is designed to include all temperature errors, viz: changes in the quartz wedges, tube length, concentration and specific rotation. Wiechmann ^ and others deny the advisability of Wiley's correction.
Since the rotation of quartz is slightly changed by variations of temperature, it may be well to conduct polariscopic work re- quiring great exactitude, as in research work, at the normal temperature of the instrument.
CHEMICAL METHODS.
37. Detenu illation of Sucrose by Alkaline Copper Solution. — Dissolve a weighed quantity of the material in water and dilute to 50 cc. Invert by means of hydrochloric acid as described in 89. Transfer to a litre flask, cool, neutralize with caustic soda, and dilute to 1000 cc. The quantity of material to be used depends upon the method of further procedure selected.
It is, however, convenient to use 5 grams or a multi- ple of 5 grams and to dilute to a multiple of 100 cc. in order that the table of reciprocals on page 294 may be used for the calculations if a volumetric method be selected.
Determine percentage of invert-sugar by one of the methods in 72 or 73. Multiply the per cent invert-sugar by .95, since sucrose on inversion yields invert-sugar in the ratio 100 : 95.
38. Determination of Sucrose in the Presence of Keducing Sugars. — Determine the reducing sugar before inversion and after, as indicated in 37.
Calculation, — Per cent reducing sugar after inversion — per cent reducing sugar before inversion X -95 = the required per cent sucrose.
1 Ckjmptes-renduIV CoQgreslut. de Chimie Appliqa^e, 2, 143.
2/6wi. 1. 143.
SAMPLIKG AND AVEEAQING. 43
GENERAL ANALYTICAL WORK.
SAMPLING AND AVERAGING.
39. General Remarks on Sampling and Averaging. — Accurate sampling is essential to successful chemical control. The samples must be strictly represent- ative of the average composition of the substance or sub- sequent analytical work will be wasted.
The method of sampling should be by aliquot parts. This consists in drawing a definite quantity from each lot of the material, which must be the same aliquot part in each case.
Example. — Given four lots of sirup, A^ B, C, and D, from which an average sample is to be drawn. Let A = looo, B = 800, C= 500, and Z> = 200. Each of these lots differs in analysis. Manifestly a mixture of equal parts of A, B, C, and D would not be a true average sample, but a mixture of 10 parts of ^, 8 parts of B, 5 parts of C, and 2 parts of D would be a representative sample.
In calculating an average analysis from a large number of analyses the same principle must be applied.
Example. — Given the following per cents of sucrose, representing the analyses of the beets each d^y for a week: \l%\ 14%; 13^; 14.5^; 15^; 15.5^; i6^- The following num- bers of diffusers of beets were worked each day : 168; 144" 140; 150; 165; 160; 145 — a total of 1072. Required the mean percentage of sucrose in the beets.
Multiply each analysis by the number of diffusers of beets it represents, and divide the sum of the products by the total number of diffusers worked.
44 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
|
15 X 168 = 2520 14 X 144 = 2016 |
15,806 ~ = 14.74 = 1,072 ^ '^ |
|
13 X 140 = 1820 |
the mean per cent |
|
14.5 X 150 = 2175 |
sucrose for the week |
|
15 X 165 = 2475 |
|
|
15.5 X 160 = 2480 |
|
|
16 X 145 = 2320 |
1,072 15,806
It is obvious that the weight of juice obtained per day, or the weight of the beets worked, may be used as a factor in the above calculation and strictly accurate averages secured.
Usually, however, if the diffusers receive practically uniform charges of beets, the average analysis, as calcu- lated above, will approximate the true mean very closely.
40. Sampling Beets in the Field.— Beets grow- ing side by side may differ greatly in sugar content; the same is true of beets grown within a few feet of one an- other as well as from widely different parts of the field. This indicates the difficulty if not impossibility of selecting a strictly representative sample. In point of fact, samples selected in the field only approximately represent the general average.
A convenient plan for sampling in the field is as follows: When drawing the beets to the factory, take a definite number at random from each load until all the beets have been hauled; unite the subsamplesand proceed as indicated farther on.
If the sample is to be taken after the beets have been lifted and placed in piles, select a number of beets from each pile as above, or from every second or third pile, etc., and unite the subsamples. If the roots be still in the ground, lift a beet at definite intervals in the row, from every second, third, or fourth row as may be deemed best, and unite the subsamples as before.
The importance of the sample and the size of the field must determine the number of beets to be drawn, but this number should in any case be as large as practicable.
Having selected the beets, they should be sorted into three or four classes according to size and ranged in rows
SAMPLING AND AVERAGING. 45
in a convenient place, protected from the rays of the sun. The number of beets is now reduced by subsampling, tak- ing from each row in proportion to the nufrtber of beets in the row. For example, take every fifth or every tenth beet in the row. If the number of beets drawn in this way be too large, the subsample should be rearranged in rows and again subsampled.
41. Subsaniplin^ of Beets for Analysis in Fix- ing the Purchase Price.— As will be shown (p. 177), the sucrose is not uniformly distributed throughout the beet, and further the juice obtained by pressure from the same sample varies in composition with the pressure exerted and the state of division of the pulp. The more finely divided the pulp, and the heavier the pressure, the nearer the juice obtained approaches the mean juice in composition.
The proportion of juice in the beet varies from sample to sample, and often materially from the average (95^) ; hence the practice of employing a coefficient, e.g. .95, to calculate the percentage of sucrose in the beet from the analysis of the juice, should be discouraged. In the course of an en- tire season this may be just to the manufacturer, but un- doubtedly is an injustice in many cases to the producer of the beets.
If the indirect method of analysis be employed, the same models of rasp and press should be used by the chemists of the buyer and the seller. Further, the conditions of sam- pling and analysis should be the same in both laboratories.
The beets should be divided longitudinally into quarters or eighths, and an entire segment should be rasped. This insures the reduction of a portion of the beet in propor- tion to its size.
In many factories it is the custom to remove a small plug "or cylinder from each beet for the analysis. Owing to the unequal distribution of the sugar in the beet this method cannot be depended upon to give a strictlj^ repre- sentative sample, but experience has shown that the varia- tions from the true average sample are not great, provided the cylinder be taken in the proper direction. The method and direction of removing the cylinder are indicated in Fig. 18. The boring-rasp (Keil and Dolle) is well adapted for
46
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
removing a sample of pulp from each of a number of beets.
Fig
This machine, which is shown in Fig. 19, may be used in Pellet's instantaneous diffusion method (B2).
Fig 19. The beet is pressed carefully against the rasping-tool, which revolves at the rate of 2000 revolutions per minute.
Fig. 20. An opening in the rasp, which is sh«wn in detail i» Fig. 20, permits the pulp to pass into the tool, whish is hollow, and
SAMPLING AKD AVERAQIKG. 47
thence to the box shown in the figure. Practice is neces- sary in using this machine in order to produce a suitable pulp. The pulp from the first perforation should be re- jected.
It is evident that this machine does not remove a portion of pulp bearing a fixed relation to the size of the beet. This is essential in order that the analysis may represent the mean composition of the roots. The following method of sampling has been proposed by Kaiser' to obviate this difficulty.
The form of the beet is a cone the height of which is approximately three times the radius of the base, hence its volume is calculated by the formula nr^ = volume ; in other words, the volume of the beet increases as the cube of the radius of the base. For example, we have three beets whose radii are 4 : 5 : 6; their volumes are then in the ratio 4^ : 5' : 6', or 64 : 125 : 216. The beet whose radius is 4 should be per- forated once ; the second, whose radius is 5, should be per- forated (VY) 2 times, and the third, having a radius of 6, should be perforated (Vt) 3 times, and so on. Kaiser uses a scale which indicates the number of perforations to be made in each beet. Such a scale may easily be made which will show at a glance the number of times each beet should be perforated.
This method of sampling gives approximately correct re- sults, even if the relation between the radius of the base of the beet and its length be different from Kaiser's numbers. When practicable, in order to obtain a thoroughly reliable sample, it is advisable to divide the beets longitudinally and reduce an entire segment of each to a pulp suitable for a direct method of analysis.
After the sample of washed beets is received in the lab- oratory its weight should be noted, that a correction may be made for the loss of weight by drying prior to the an- alysis.
42. Sainpliiij? Beets at the Diffusion-battery. — Samples of beets can be drawn at the battery with mod- erate certainty of obtaining a fair average. In the various
* Deutsche Zuckerindustrte, Nov. i8q6.
48 HANDBOOK FOR SUGAK-HOUSE CHEMISTS.
manipulations from the field to the factory, including the transport and washing, the beets are pretty thoroughly mixed ; hence if a beet be taken at random at regular and frequent intervals, the united subsamples so drawn will be very nearly of the mean composition of the beets entering the sugar-house. It is not usually necessary to sample beets in this way, since the method given in the following para- graph is simpler and the sample drawn is more satisfactory.
43. Sampling- the Fresh Cossettes at the Diffusion-battery.— The proper time to sample the beets is after they have been sliced. A handful of the cos- settes should be taken from the elevator or drag at regular intervals and stored in a covered receptacle. Large granite- or agate-ware pails are very convenient for the purpose, as they can be easily inspected as to their cleanliness. It is not advisable to use a mechanical device to divert a part of the cossettes to the pail, since the sample so obtained is not usually a fair average.
The samples should be drawn at very frequent intervalsp if practicable every two or three minutes. In practice it is more convenient to take a small portion of the cuttings shortly after they begin to fall into the diffuser, a second when the diffuser is half filled, and a third before directing the cuttings into the next diffuser. The sample obtained in the manner described should be taken to the laboratory for immediate treatment. It is perfectly reliable, and if the beets be weighed immediately before entering the cutter, it may enter into the chemical control of the diffusion. It is necessary to keep the sample-pails scrupulously clean, using boiling water in washing them ; they should be large enough to contain the subsamples from two or three hours' work.
44. Sampling the Exhausted Cossettes.— The exhausted cossettes should be sampled in a similar manner to the fresh cuttings. 1 his sample may be taken from the elevator leading to the pulp-presses, and should be stored in a covered galvanized-iron pail having the bottom per- forated for drainage.
45. Sampling Waste Waters.— A definite volume of the waste water should be drawn from each diffuser
SAMPLING AND AVERAGING. 49
and these subsamples stored in a loosely stoppered bottle, with corrosive sublimate as a preservative.
46. Sampling Diffusion-juice, etc. — In sampling diffusion-juice, a definite volume should be drawn from each measuring tankful. This volume once decided upon should not be changed during the sampling period except there be a change in the volume of juice drawn into the tank, and then the sample should be changed in a like proportion. This is not easily accomplished, except by the use of an automatic sampler.
In sampling purified juices the same method should be observed.
47. Sampling Filter Press-cake.— In sampling the press-cake, small portions should be taken systemati- cally from different parts of the press, bearing in mind that parts of the cake contain more moisture than others, according to the kind of press. The number of presses filled should be recorded for use in estimating the weight of the press-cake and in averaging the analyses.
A very simple and satisfactory instrument for sampling filter press-cake is made from a small brass tube with a cutting edge at one end. Several cork-borers of the same diameter are more convenient than a single brass tube for this purpose.
In using this instrument small cylinders of the press-cake are cut out in precisely the same manner as one would bore a hole through a cork. The subsamples are left in the tubes until a sufficient quantity of material has been col- lected. Each subsample pushes its predecessor farther into the tube.
48. Sampling Sirups. — A method is recommended in 10 for the measurement of sirups. In this method gauge-tubes, similar to the water-gauges on steam-boilers, are used. The sirup should be thoroughly mixed in the tanks before admitting it to the tubes. If several tanks be used, a volume of sirup should be drawn from each in- cluded in the analytical period, as advised in lO.
49. The Preservation of Samples.— The sample of diffusion-juice is effectually preserved from fermentation by the addition of subacetate of lead. It maybe preserved
50 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
in this way several weeks or even months without percepti- ble change in the sucrose content. The most convenient preservative is mercuric chloride, i part to 5000 or 10,000 parts of juice. It is not advisable to store juices treated with mercuric chloride for a longer period than 24 hours. The advantage of the mercuric chloride is that it permits the usual determinations, viz., sucrose, total solids, ash, etc., to be made with the same sample, thus obviating the neces- sity of drawing a second sample as is usual when subacetate of lead is used.
In many houses it is the practice to store the samples a week before analysis, uniting those drawn from day to day. In such cases it is advisable to determine the density, solids, and ash from day to day, and store a portion of the juice with subacetate of lead for the sucrose determination. The use of mercuric chloride simplifies the work, and as it is used in such minute quantities it does not perceptibly affect the accuracy of the results.
When subacetate of lead is employed as a preservative, it should be added in the proportions required for the clari- fication of the juice, i.e., about 2-3 cc. of the concentrated solution (207). It is convenient to use the concentrated lead solution, and, when preparing for the polarization, to measure the mixed juice and lead solution and add suffi- cient water to increase the volume to 110% that of the juice. The per cent sucrose is then readily calculated by the use of Schmitz' table(p.285)from the degree Brix of the juice and the polariscopic reading.
The preservative must be thoroughly mixed with juice as each portion is added. This is easily accomplished when an automatic sampler is employed by letting the de- livery-tube dip to the bottom of the storage-bottle. The mouth of the bottle should be loosely plugged with cotton. When the sampling is by hand, it is advisable to use a wide- mouthed jar, provided with a cover, for the storage of the juice, and mix frequently. This facilitates the collection of the subsamples without the use of a funnel.
No preservative is required for sugar-house products other than the waste waters, juices, and sirups.
50. Automatic Samplings of Juices. — Automatic
SAMPLING AND AVERAGING.
51
samplers have for their object not only the relief of the chemist from this duty, but the drawing of samples which are probably more reliable than those obtained in any other way.
This problem is not a simple one in the case of sampling diffusion-juices at the measuring-tank. It is evident from the method of conducting the diffusion, that the juice re- ceived into the measuring-tank is not of uniform composi- tion. A sample drawn from the bottom of the tank will differ slightly from one drawn at the centre or near the top.
Coombs' Automatic Sampler. — The apparatus shown in Fig. 21 is the invention of Mr. F. E. Coombs, Chemist of
JUICE PIPE
A.— i TO I INCH VALVE. B,— STRONG RUBBER TUBE CON- NECTING PIPE LEADING FR0m"A"wITH C,— A GLASS T-TUBE|tO 7 INCHES INSIDE DIAMETER.
D, — SHORT ARM OF T. FROM WHICH THE SAMPLE IS TO BE LED INTO AN APPROPRIATE RECEIVER.
Fig. 21.
the Shadyside Plantation, Louisiana, and of the Esperanza Estate, Trinidad, B. W. I., throfligh whose courtesy this de- scription and illustration were supplied the author.
This apparatus is applicable to the sampling of liquids which are not too viscous to flow through small pipes. It may be used in sampling juice and sirup, and has proved quite reliable in practical work. It has the advantage of
52 HANDBOOK FOB SUGAR-HOUSE CHEMISTS.
being quickly set up wherever there is provision for re- turning a small quantity of overflow liquor to the tank.
Attempts to draw continuous samples of liquor from pipes by means of a small valve, depending upon the valve to regulate the flow of the sample, have usually failed, since the valve must be so nearly closed that fine pulp in the juice or, in the case of sirup, a mere change from a low to a high density clogs the opening and stops the sampling.
The flow must be sufficient to keep the valve free from obstruction. By the use of a T-tube, as shown in the figure, a strong current of liquor can be kept flowing through the pipe, and at the same time a small, continuous, easily regulated drip can be diverted into the sample- bottle.
In the figure, the apparatus is shown as arranged for drawing a sample of juice as it passes from the measuring- tank to the carbonatation. It is advisable to pass the juice through a distributing-tank in which the sampler is lo- cated, otherwise an arrangement must be provided for con- ducting the overflow to the carbonatation-tanks.
The sample-bottle at D rests upon a wooden shelf hung inside the tank by hangers of strap-iron which hook over the edge. It is apparent that when Z>, the short arm oi the T-tube, is in its lowest position it will give its maximuna discharge. By rotating the T-tube, which is of glass, in the strong rubber connecting-tube B to the position Z?', the drip will cease, all the liquor passing out at C, The posi- tion giving a sample of the required volume is readily ascertained by experiment. The sample, if juice, is pre- served as indicated in 49 ; sirups require no preservative.
With well-strained juice the drip is regular and there is rarely trouble from clogging.
It is evident, from the arrangement of the sampler, that the samples drawn, whether of juice or sirup, may be de- pended upon as being representative of the composition of the entire volume of the liquor.
It is necessary to connect the small pipe at the under side of the juice or sirup main, to insure a continuous flow, even when but little liquor is passing. The main should be tapped at its highest level, or on the discharge side of
SAMPLING AND AVERAGING.
53
that level, to avoid drawing liquor left in the pipe when the flow is temporarily stopped. The valve on the sampling- pipe should be placed as close as possible to the point where the main is entered.
The valve A should always be opened as widely as pos- sible to prevent clogging, but this must be regulated so that the current through the main arm of the T-tube shall not be too swift, since it will then act as an aspirator. For this reason it is advis- able to avoid extending the discharge-tube Z> below the level of the sample in the bottle, otherwise the entire contents may be lost.
Horsiti' Dean's Automatic Sampler. — This apparatus, shown in Fig. 22, consists of a three-way cock for con- necting a small standpipe alternately with the measur- ing-tank and the sample- bottle, and is operated by a suitable float.
This sampler is placed in- side the measuring-tank. It is so arranged that the volume of the sample drawn is proportionate to the quantity of juice in the tank. The discharge-pipe from the diffusion-battery should enter the measuring-tank at the bottom. The inlet to the sampler should be di- rectly over the inlet from the battery, if practicable, projecting into the pipe. If this precaution be not ob- ^^^' ^^"
served the sample drawn will not be a iair average.
54 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
2t is obvious that this sampler is not applicable in sam- pling sirups.
51. Sampling Sugars.— Sugars are best sampled by means of a "trier" or sound (Fig. 23). This in-
FlG. 23.
strument is so constructed that it may be plunged into a quantity of sugar and, on withdrawal, remove a sample representative of the sugar through which it has passed. The trier should be long enough to pass from end to end of the package of sugar, diagonally if necessary. The chemist must be guided largely by the grade of the sugar and the method of packing in drawing the sample. A portion should be dra^yn from every third, fifth, etc., package according to the size of the lot. The large sample should be well mixed, andiall lumps broken, then subsam- pled by quartering.
k
DENSITY DETERMINATIONS. ^5
DENSITY DETERMINATIONS. APPARATUS AND METHODS.
52. Notes on Density. — The expression "density" as used in this work is synonymous with " specific gravity," and is employed for brevity and convenience. Sugar chemists also frequently term the degree Brix or the degree Baume the " density " of the liquor. While this use of the word " density" is not strictly correct, it is sanctioned by general usage.
53. The Brix and Baume Scales.— The degree Brix is the percentage by weight of sucrose in a pure sugar solution. It is customary to consider the degree Brix as the percentage of total solid matter in the solution, and it is thus applied in sugar analysis. It is this feature of the Brix spindle which renders it more convenient than the Baum6 for sugar-house purposes. .
The Baume scale has no convenient relation with the percentage composition of any of the sugar-house products. The point to which it sinks in distilled water at the stand- ard temperature is marked zero ; the corresponding point in pure sulphuric acid of 1.8427 specific gravity is marked 66 degrees. Baume spindles are also graduated for den- sities above and below the limits mentioned, but this range is all that is ever required in sugar analysis.
There has been much confusion in the graduation of Baum6 spindles. The graduations should be checked by means of splutions of known density and under standard conditions (55). The use of these spindles is now com- paratively limited in sugar-house practice.
54. Automatic Apparatus for the Determi- nation of the Density of the Juice.— The density \h usually ascertained by means of a hydrometer, an instru- ment commonly termed a " spindle " in sugar-house practice. These instruments are usually graduated to degrees Brix or Baum6. The readings on the spindle are converted into terms of specific gravity or density by means of a table {see p. 256);
Automatic apparatus is used to some extent in the Euro- pean sugar-houses for the determination of the density.
56
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
One of the simplest forms of apparatus for this purpose is that devised by Langen and shown in Fig. 24.
The construction of this instrument is based upon the principle of communicating vessels. By suitable means, the small reservoir ^ is connected with the measuring-tank at the diffusion-battery; a portion of the juice from each charge drawn into the latter is deflected and passes through the reservoir into the tube S, and overflows at r. Inside the tube S is another tube, J^D, which terminates above in a fun- nel-shaped vessel and below in a flexible bulb in the tank H. The interior of this tube, including the bulb, is filled with water, whose height is registered upon a cylinder B by means of a float carrying a pencil, n. It is evident from an in- spection of this apparatus that the water in the inner tube will rise in proportion to the specific gravity of the juice surrounding and press- — ing upon the flexible rubber bulb. This rise in the level of the water is registered by the pencil, carried Fig. 24, by the float, upon the paper-cov-
ered cylinder. The cylinder is revolved by clockwork, making one revolution every twelve hours. The record may be in degrees Brix or Baum6 as preferred.
The variable temperatures of the juice have no influence upon the apparatus, provided the column of water be of the same temperature as the juice surrounding it. For this reason the tube /^is spiral at its lower end. Mr. Eugene Langen, the inventor of this instrument, has substituted a bundle of fine copper tubes for the spiral, jD. Foam and mechanical impurities do not affect the accuracy of the apparatus.
56. Hydrometers or Spindles. — These instru- ments are also termed " saccharometers " when specially graduated for use in the sugar industry. The density is
DENSITY DETERMINATIONS.
57
ascertained by noting the depth to which -BRix the spindle sinks in the liquid.
1 10 11 13 13 14 15 16 17 18 19 20 21
L
Hydrometers of the better grade, for use in sugar work, are of the shape shown in Fig. 25. Spindles of the best quality are of glass, and are usually provided with a fine thermometer. Instruments for rough work are made of metal or of glass, and without a thermometer.
A variety of systems of graduations is used in France, but in this country and in Germany the Brix and the Baum6 are the only scales employed in sugar work. In com- paring data obtained in French sugar-houses it is well to remember that percentages are usually expressed in terms of the volume of the solution instead of the weight.
In American and German sugar-houses, the standard temperature for the graduation ~ of instruments is 17^° C; in
France, etc., it is 15° C.
Since the tables for cal- culations are based on a temperature of 17^° C. it is advisable that all hydrom- eters be graduated at this temperature.
It is recommended that the hydrometers be gradu- ated to ^V Brix. The in- struments should be ar- ranged in sets of o" to 5°, 5° to 10°, 10° to 15°, 15° to 20°, and 20° to 25° Brix. Each should be provided with a delicate thermome- ter. The stem should be long, that the graduations may be read with certainty Fig. 26. and ease.
|
0-. 1-. i |
^ |
|
|
m |
1 |
|
|
tft^ |
^'. |
Wii |
|
ES |
2| |
^ |
58 HAN"DBOOK FOR SUGAR-HOUSE CHEMISTS.
Spindles should be tested from time to time, employing •standardized solutions of pure sucrose of the temperature at which the instrument was graduated, preferably at 17^° C. The strength of the sugar solution should be checked by means of the polariscope.
The method of reading the spindle is shown in Fig. 26. The reading at E, not R\ should be recorded as the observed density, and a correction should be made for variations in the temperature from the standard. A table is given on page 282 for the correction of the observed degree for varia- tions of temperature above and below 17^° C. It is advisable to make all readings at as nearly i7^°C.as may be practicable.
56. The WestpUal Balance.— The principle of this balance ' may be briefly stated as follows : A glass bob is so adjusted as to be capable cf displacing a given num- ber of grams, five for instance, of distilled water, at a given temperature when wholly immersed in the liquid and sus- pended by a fine platinum wire. The bobs may be gradu- ated for any temperature ; but for sugar work 17^° C. is most convenient, since this is the temperature usually em- ployed in preparing specific-gravity tables. For accurate work the temperature of the solution whose specific grav- ity is to be determined should be exactly that for which the bob was graduated. The balance is provided with several riders or weights. Two of these riders, (i) and (2), are each exactly the weight of the water displaced by the bob at the standard temperature, i7i°C. The other riders, (3), (4), and (5), are respectively one tenth, one hundredth, and one thousandth the weight of the first mentioned. When the weight (i) is hung on the hook at the end of the beam, and the bob is immersed in distilled water at 17^° C, the balance should be in equilibrium, the weight having the value i.ooo in this position. In case the balance be not in equilibrium under these conditions, provided the bob have been correctly graduated, the latter must be suspended from the hook and the adjusting-screw turned until the pointers are exactly opposite one another. The weights (2), (3), (4), and (5) are placed on the beam in addition to (i) for liquids
* Adapted from Bull. 13, Chem. Div., U. S. Dept. Agri.; also illustration.
DENSITY DETERMINATIONS.
59
heavier than water, and have the values .i, .oi, .001, and .ooor, respectively, when placed on the corresponding graduations of the beam, and for other graduations .300, .030, .003, .0003, etc. Each rider is provided with a hook
Fig.
from which additional weights may be suspended in the case of more than one falling upon the same graduation.
The method of using the balance is as follows : Dissolve a weighed portion of the material in water and dilute to a measured volume at 17^° C. ; for example, 25 grams to 100 cc. Suspend the bob of the balance, as described above, in this solution, and weight the beam with the riders until the balance is in equilibrium. Read off the specific gravity from the position of the weights on the beam. Example : 25 grams material dissolved and diluted to 100 cc. Position of the riders :
60
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
(i) at point of suspension of the bob = i.ooo
(2) not on the beam.
(3) at 7 =0.07
(4) at 9 = 0.009
Specific gravity = 1.079
The degree Brix corresponding to 1.079, '•'^•» the per cent solids in this solution, is 19, as given in the table, page 275. To obtain the weight of the solution, multiply 1.079 by 100 = 107.9 > hence the weight of solids in the solution js 107.9X19-^-100 = 20.5 grams = the weight of solid matter in 25 grams of the material. The per cent solids in the material, i.e., the degree Brix = 20.5 -^^ 25 X 100 = 82, and the corresponding specific gravity, obtained from the table,
/T^
Fig. 28. the density is to be determined.
is 1.4293. See 85 relative to the ac- curacy of this determination of the degree Brix.
57. Py kilometers. — Pyknome- ters are bottles so constructed that they may be filled with a definite volume of liquid. Knowing the weight of this volume, it may be compared with the weight of an equal volume of water, from which the density of the liquid is calculated. It is rarely necessary to use a pyknometer in the sugar in- dustry, the more rapid density deter- mination by the spindle being usually sufficiently accurate.
Pyknometers are made in a great variety of forms. One of the most convenient of these is shown in Fig. 28. The stopper is a fine thermom- eter ground into the neck of the bottle. The side tube provides an outlet for the excess of liquid when the stopper is put in place. The bottle should be filled at a somewhat lower temperature than that at which As the temperature
DENSITY DETERMINATIONS. 61
gradually rises to the desired point, the excess of liquid is blotted off. At the required temperature, the cap is placed in position, and receives any further liquid, which may be expelled from the bottle, as the temperature rises to that of the room. There is a minute opening in the cap for the escape of the air.
In sugar work, the specific gravity should be determined at 17^° C. for reasons already stated. The weight of the corresponding volume of water may be determined at room temperature and a correction be made to reduce it to the standard temperature, the tables on page 251 being used for this purpose. It is customary to express specific gravities
as follows : — '—:, 1.0705; the numbers above and below the
17.5 line being the temperatures at which the bottle was filled with water and the substance respectively.
Recently boiled and cooled distilled water should be used in density determinations.
To calculate the density of a liquid, divide the weight of a definite volume of it by the weight of an equal volume of water.
57a. Standard Temperature for Density Determinations. —
20° The International Commission has adopted — ^ C. as the stand- ard temperature for density determinations. The hydrometer should sink to the 0° mark in water at 20° C, and the cor- responding specific gravity of the water, referred to water at 4° C, is 0.998234.
As the adoption of this standard is very recent, most factories are equipped with instruments standardized at 17^° C. and with the corresponding tables, such as are given in this book.
62 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
ANALYSIS OF THE BEET.
58. The Direct Analysis of the Beet.— The Methods for the direct analysis of the beet may be divided into two general classes, according to the solvent used, viz. : (i) methods employing alcohol ; (2) methods employing water. The alcoholic methods have found most favor in Germany, and the aqueous methods in France.
Certain modifications of the Scheibler alcoholic method and Pellet's aqueous methods, hot and cold, are the most important of their classes, and are the only ones which will be described in this book. It is probable, judging from the published statements of many chemists, that these meth- ods are equally accurate if the instructions of their invent- ors be implicitly complied with. The alcoholic methods are usually considered the most scientific.
69. Scheibler's Alcoholic Method with Sox- hlet's Extraction Apparatus.— Various modifications of Soxhlet's apparatus are used to such an extent in chem- ical laboratories that an illustration. Fig. 29, and a brief description of it will suffice. The apparatus is so arranged that the vapors of the solvent, which is boiled in the flask by means of a hot-water bath, pass up through the tube B to the reflux condenser, and the solvent falls back into the extractor in which the material is placed. When a suffi- cient quantity of the solvent accumulates in the extractor, it is siphoned into the flask by the tube shown at the right. The substance is thus extracted with successive portions of the solvent.
A very convenient and efficient modification of this ap- paratus is the siphon extraction-tube devised by A. E. Knorr, shown in Fig. 30. The connections with the flask and condenser are made with corks as in the Soxhlet apparatus. Knorr's apparatus, as arranged for general purposes, dispenses with corks, but requires a special flask, which is not convenient for sugar analysis.
The siphon-tube S is sealed into the bottom of the tube
ANALYSIS OP THE BEET.
63
A and lies close to the wall so as to permit the insertion of the tube B containing the material. The lower end of B is closed with a perforated disk. A spiral of copper wire, C, pre- vents the tube A from closing the tube D.
This apparatus has the ad- vantage of extracting the SU' crose with a hot solvent. Other convenient modifications of Soxhlet's appara- tus are described by Wiley in his Agri- cultural Analysis,
In the direct an- alysis of the beet with the Soxhlet- Sickel apparatus, Fig. 29, proceed as follows for the ex- traction of the su- crose: Place a plug of absorbent cotton in the bottom of the tube, then introduce 26.048 grams of the pulped beet, or 2 X 16.29 grams, accord- ing to the polari- scope in use, press- ing the pulp lightly with a rod. Very small fragments of the beet may be used instead of pulp. Connect the extractor with the reflux condenser as shown. Place 75 cc. of 95 per cent alcohol in the flask and connect with the extractor as indicated in the figure; heat the flask in the water-bath and continue the extraction from half an
Pig. 29.
Fig. 30.
64 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
hour to two hours or more, according to the state of division of the sample. Use somewhat weaker alcohol if only 16.29 grams of pulp be taken. Cool and remove the flask, substituting a second containing 75 cc. of 75 to 80 per cent alcohol, and continue the extraction to ascertain whether the first extraction were complete.
Fill the first flask to the 100 cc. mark, after treating the sample with two or three drops r>f subacetate of lead solu- tion. Mix the contents of the flask, filter, and polarize. Having extracted the normal weight of pulp, the polari- scopic reading is the per cent of sucrose in the sample.
The extract in the second flask should also be polarized as a check upon the extraction.
Great care is essential in the polarization of alcoholic solutions. The least quantity of subacetate of lead, that will clarify the solution, should be used. The solution must be protected from evaporation during the filtration by a cover-glass. Avoid irregularities in the temperature of the solution in the observation-tube, due to the warmth of the hands; since the density of the solution in different parts of the tube will vary under such conditions, striae will form, rendering an accurate reading impossible.
The Scheibler method, as above described, differs from the original only in a few minor details, especially in the arrangement of the extraction apparatus. The Soxhlet ex- traction apparatus is much more effective than Scheibler's original instrument.
00. Stammer's Alcoholic Digestion Method/ — This method differs from that of Pellet described in 62 in details of manipulation and in the use of alcohol instead of water. The pulp must be reduced to a cream, in fact should be as finely divided as is required in the Pellet method (62).
Wash 26.048 grams of pulp into a flask graduated at 100.55 cc. with 92 per cent alcohol, add subacetate of lead for clarification, and dilute to the mark with the alcohol. The least quantity of the subacetate that will effect clarifica- tion should be used. Acetic acid is not required. Mix thor- oughly, and after allowing a few minutes for the digestion,
^ Zeit. Rubenzucker-Indu5irte,ZZ»aQ6. i-,-iy,'*f -Ju
ANALYSIS OF THE BEET.
65
filter and polarize, observing the precautions given in 59 relative to the polarization of alcoholic solutions.
A method similar to this, Rapp-Degener, employs hot digestion in a flask fitted with a reflux condenser.
61. Pellet's Aqueous Method. Hot Digestion. — Any good rasp may be used in the preparation of the pulp for this method. Pellet recommends the conical rasp
of Pellet and Lomont, as illustrated in Figs. 31, 32, and 33. There is frequently a depression in the side of the beet, as shown in section in Fig. 34. Since the segments OA and OB are not of equal sugar con- tent, two segments should be reduced to pulp, or, if the sam- ple include a large number of beets, a single segment of each may be pulped, taking care to present alternately the large and the small diameters of the beets to the rasp.
The special flasks shown in Fig. 35 are convenient for use in this method. Transfer 26.048 grams of the pulp to 4;he flask, F'g. 32.
using a little water to wash the weighing capsule and funnel, or, for the Laurent, employ 32.58 grams of pulp, i.e., 2 X normal weight. The flasks are graduated to contain 801.35 cc. for the Schmidt and Haensch and 201.7 cc for
i>^^»«»
66 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
the Laurent polariscopes, in order to compensate for the
Fig. 34.
Fig. 33- volume of the marc and the lead precipitate. Add 5 to 10 cc. subacetate of lead solution of 54.3° Brix (207) for the clarification. Approxi- mately 6 to 7 cc. are required per 26 grams of beet-pulp. This reagent should be run into the flask in advance of the beet-pulp. Add a few drops of ether to beat down the foam, then sufficient water to increase the volume of the solution to about 190 cc. Heat to So"" C. in a water- bath and maintain this temperature about 30 minutes, occasionally giving the flask a circular move- ment to facilitate the escape of the air from the pulp. Increase the volume of the solution from time to time during the heating, so that when the opera- tion is completed only a few drops of water will be required to complete the volume of the solution to the mark. After approxi- mately 30 minutes' heating, cool the flask and contents and add strong acetic acid to the solution to acidity, dilute to the graduation, mix and filter. The state of division of the pulp will govern the time of heating. In polarizing the filtrate, use a 400-mqj. observa- tion-tube, thus directly obtaining the per cent sucrose in the beet with the Schmidt and Haensch polariscope, or double this percentage if the Laurent instrument be used.
Fig. 35.
ANALYSIS OF THE BEET. 67
Pellet uses a special water-bath in this process that admits a considerable number of flasks at one time. The flasks are held in a rack and may all be removed from the bath at one time and plunged into cold water.
The solutions should be carefully protected from evap- oration by covering the funnel during filtration.
There has been much controversy relative to this method, especially among the German chemists. Many claim that it gives results that are too high, and other chemists of equal prominence and experience contend that it gives cor- rect results. Le Docte,' in a series of experiments, obtained percentages by hot digestion a few one-hundredths higher than by the cold diffusion method described below. The following method, using cold water, is usually preferred, provided a sufficiently fine pulp can be produced.
62. Pellet's Instantaneous Aqueous Diffu- sion Method. — The method as described by Pellet will be given first, and then a few of the various modifications. The author prefers the Sachs-Le Docte modification given on page i8i, which combines rapidity and accuracy.
Pellet's Original Method. — In following Pellet's original method the specifications as to the condition of the pulp and the quantity used must be strictly complied with in order to obtain satisfactory results.
For polariscopes whose normal weight is 26.048 grams wash this weight of pulp, with water, into a fiask graduated to hold 201.35 cc, or 25.87 grams into a 200-cc. flask. Run 5 to 7 cc. of subacetate of lead solution of 54.3° Brix (ii07) into the flask before washing in the pulp, and then thoroughly mix with the latter. Add several small portions of ether to beat down the foam. Rotate the flask to facilitate the escape of the air-bubbles. Add a few drops of acetic acid to acidulate the solution, complete the volume to the graduation, mix, filter, and polarize, using a 400-mm. observation-tube. The polariscopic reading is the per cent sucrose in the beet. With the Laurent instrument, use the normal weight of pulp and a flask graduated to hold 200.85 cc. The polariscopic reading,
> Sucrerit Beige ^ 585, 2451 273, 309.
68 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
using a 400-mm. observation-tube, is the per cent sucrose in the beet.
Success with this method demands (i) that the pulp shall be in a suitable state of division, neither too coarse nor too fine; (2) that no more pulp shall be used than indicated in the description of the method. If there be difficulty in removing the air occluded by the pulp, notwithstanding repeated additions of ether, the pulp is too fine. This may be remedied by altering the speed of the rasp. The occluded air is the source of error that requires greatest care to avoid.
Kaiser-Sachs Modification. — This method and the Sachs- Le Docte modification practically eliminate errors from the Pellet instantaneous diffusion method. Use flasks holding a little more than 200 cc. Also use the same quantities of pulp as indicated in the description of the original Pellet method. Run 5 cc. of subacetate of lead solution into the flask, then counterpoise the flask and contents on a balance. Wash the pulp into a flask and add sufficient water to make a total of 172 grams of water. Mix thor- oughly, filter, and polarize the solution in a 400-mm. tube. The polariscopic reading is the per cent of sucrose in the beet. According to Pellet, acetic acid should always be added. This agrees with the author's experience.
Sachs-Le Docte. — This method, which is fully described on page 181, differs from the above in adding the water and subacetate of lead from an overflow or automatic pipette. This insures a very accurate measurement, with extreme rapidity.
The finest attainable pulp should be used with both the Sachs-Le Docte and the Kaiser-Sachs methods.
03. Determination of the KecliiciDg- Sugar in the Beet. — Herzf eld's Modification of Claassen's Method. — Digest no grams of finely divided pulp, or preferably creamed pulp, in a 500-cc. flask with 10 to 15 cc. of dilute subacetate of lead solution, 3 grams of precipitated carbon- ate of calcium, and suflBcient water to nearly fill the flask. Digest 45 minutes at a temperature of 75° to 80° C. Cool and complete the volume to 500 cc, mix, and filter. If necessary, clarify 100 cc. of the filtrate with an additional
ANALYSIS OF THE BEET. 69
portion of subacetate of lead ; add carbonate of sodium in small excess to precipitate the lead, dilute to no cc, and filter. Determine the reducing sugar in the filtrate by one of the methods given in 72 and 73. The percentage of reducing substance in the beet is so small that no correc- tion need be made for the volume of the marc.
64. Notes on the Direct Methods of Analysis. — With the exception of Scheibler's alcoholic method, it is necessary to make an arbitrary allowance for the volume of the marc in the direct analysis of the beet. Pellet has based this allowance upon the mean of a large number of marc determinations, made under practically the conditions which obtain in his cold diffusion method. The error intro- duced through an arbitrary allowance for marc is very small, and even in extreme cases may be neglected.
There should be no delay in the analysis of the pulp. As soon as it is obtained it should be thoroughly mixed and protected from the air.
65. Rasps and Mills for the Reduction of the Beet. — The Cylindro-divider, Keil (Gallois and Dupont, Paris). — This machine, Fig. 36, as indicated by its name, consists esssentially of two deeply grooved cylinders which revolve in opposite directions. Nearly all of the pulp adheres to the cylinders, but little dropping into the drawer. The particles which fall, if too large, should be returned to the mill and the grinding should be continued until the pulp is uniformly divided. The mill should be driven at 120 revolutions per minute, either by hand or power.
Should the beets be unripe or unsound, the juice may separate and collect in the drawer. In this event, the pulp, when fine enough, should be removed from the cylinders and thoroughly mixed with this juice. The pulp will absorb the juice, and may then be sampled as usual.
This mill is designed for grinding cossettes and fragments of beets, and produces a pulp which may be analyzed by Pellet's instantaneous method.
Pellet and LomonVs Conical Rasp. — This machine as illus- trated in Figs. 31,32, and 33 is fitted with saw-blades and is not applicable in the instantaneous diffusion method. The
70
HANDBOOK FOR SUGAK-HOUSE CHEMISTS.
machine is also constructed with a cast-steel disk, which may be briefly described as a rotary file as cut for rasping wood.
Fig. 36.
This form is applicable in the above-mentioned diffusion method.
A little practice is necessary in the manipulation of this and certain other rasps in order to produce a suitable pulp.
Neveti and Aubin^s Rasp.^ — This rasp may be used in the
* Bulletin de r Association des CUimistes, 13, 31
ANALYSIS OF THE BEET.
71
reduction of beets, but not cossettes, to an extremely fine pulp for use in any of the direct methods of analysis or in the indirect method. The construction of the machine is shown in Fig. 37. It is driven by hand or power from 75 to 400 revolutions per minute.
Fig 37.
Additional rasps, designed especially for use in seed selection, are described in IGO and 1 (>1 .
66. Indirect Analysis of the Beet.— The indirect analysis, i.e., the analysis of the juice and calculation to terms of the weight of the beets, cannot be depended upon to supply data for the control of the factory. In order tb calculate the analysis of the beet from that of the juice, it is necessary to assume that the juice extracted by the press is of the same composition as the average of all the juice contained in the beet. Experience has shown that this is
72
HANDBOOK FOE SUGAR-HOUSE CHEMISTS.
not true, and that the juice obtained by moderate pressure differs materially from that obtained by heavy pressure. It also varies with the state of division of the pulp. There is also reason to believe that the beet contains water in which there is little, if any, sugar in solution. Further, in order to render an indirect method practicable, it is necessary to assume that the beet contains an average of a certain percentage of juice, and employ this percentage as a coefficient in reducing to terms of the beet. The fact
Fig. 38.
that the content of marc varies within rather wide limits is an argument against this method of analysis.
The indirect method is still employed in a large number
of sugar-houses, hence is described in this book.
The following is the usual method of procedure :
The sample is finely rasped by a suitable machine, such
as a special rasp or an efficient horseradish grater. The
pulp is placed in a small cotton bag and the juice is ex-
AKALYSIS OF THE BEET. •?$
pressed by means of a powerful press, such as that shown in Fig. 38. In operating the press as heavy pressure as possible is exerted by turning the upper wheel, then locking with the ratchet as shown in the figure, and completing the expression of the juice by means of the lower wheel. This press exerts a maximum pressure of nearly 2000 lbs. per square inch.
In order to closely approximate the true mean composi- tion of the juice, it is essential that the pulp be very finely divided and that as much pressure be exerted in expressing the juice as is practicable.
The analysis is made as indicated in 07 et seq.
In the indirect analysis, it is customary to assume that the beet contains a mean of 95 per cent of juice ; therefore to calculate the percentage to terms of the weight of the beet, multiply the per cents on the weight of the juice by 95 and divide by 100.
This method permits an approximate determination of the coefficient of purity of the juice which is not possible with a direct method and which is often of value.
74 HANDBOOK FOR SUGAR-HOUSE CHExMISTS.
ANALYSIS OF THE JUICE.
67. Determination of the Density.— The density is usually determined by means of a Brix spindle. The degree Brix may be converted into terms of the specific gravity for use in calculating the weight of the juice by means of the table, page 275 ; or a Baume spindle may be used and the readings converted into Brix and specific gravity by the above-mentioned table.
A cylinder is filled with a sample of the juice and is set aside for the escape of air-bubbles and to permit mechani- cal impurities to subside or rise to the surface. This time varies from a few minutes to half an hour. Care must be observed not to let the juice stand long enough for fermen- tation to set in. Those impurities which rise to the surface should be brushed off, and the spindle then floated in the juice. After allowing sufficient time for the spindle to reach the temperature of the juice, the scale is read as directed in 55 and illustrated in Fig. 26, and the tempera- ture of the juice is noted.
To correct for temperatures above or below 17^° C, the standard temperature at which these instruments are usually graduated in Germany and the United States, con- sult the table on page 282. It is advisable that the tem- perature of the juice when spindling be as nearly 17^° C. as practicable.
It is necessary that the density be determined with great care, since the result obtained is employed in calculating the weight of the juice at an important stage of the control work.
Other methods of determining the density are indicated in pages 55 to 61.
68. Sucrose Determination . Special Pipette for Measurements.— The method of preserving the samples will, to some extent, influence the preliminary work of the analysis.
ANALYSIS OF THE J DICK
75
The method of analysis indicated in 69 is usually more convenient when subacetate of lead is used as a preservative. If, however, mercuric chloride be employed in the sampling, the special pipette devised by the author is convenient, since the polariscopic reading is a multiple of the per- centage of sucrose.
This pipette is shown in Fig. 39. It is so gradu- ated that one need simply note the degree Brix of the juice, then fill the pipette to the corresponding degree marked on its stem. The graduations in- dicate the volume of juice, of corresponding den- sities, which weighs 52.096 grams, i.e., two times the normal weight.
The pipettes are usually graduated for ordi- nary work from 5 to 25 degrees Brix in tenths. It is recommended that, for control work, the pipettes be graduated with only a small range on each, and that there be an additional graduation as shown in the figure. The tubing should be of small internal diameter that the tenths may be the more easily read. The pipette as ordinarily made, without the additional mark near the outlet, should be graduated with a solution of approximately the viscosity of a sugar solution, of the mean de- gree Brix within the limits of the scale.
One should not blow into the pipette while emptying it, nor should the last portions of the juice be expelled in this way.
To calculate the percentage of sucrose, divide the polariscopic reading, with the German instru- ments, by 2. Pipettes for the Laurent instrument are graduated to deliver three times the normal weight (3 X 16.29 grams), hence the reading should be divided by 3. The juice should be measured at the temperature at which the degree Brix was determined.
69. Sucrose in tlie Juice . General Metliod. — When the test is not complicated by the use of a liquid preservative, the measurement may
/\
/
^6 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
be effected in a loo-iio cc. flask. To lOO cc. of the juice, sub- acetate of lead is added and the volume is completed to the no cc. mark with water. The percentage of sucrose is ascertained from the polariscope reading, and the degree Brix, with aid of Schmitz' table, page 285.
Method Employing Subacetate of Lead Solution as a Pre- servative — This method is applicable in preparing a composite sample representing a day's work. A measured volume of the lead solution is used, and at the end of the sampling period the sample, containing the lead, is measured and sufficient water is added to complete the volume to no per cent, of the juice. The method and calculations are best illustrated by the following exaniple:
Degree Brix of the juice as determined in duplicate sam- ples = 12.2. Measure the day's sample, plus the lead sub- acetate solution, subtract the number of cubic centimetres of the lead solution, and calculate the water to be added as shown below:
Volume of juice and lead solution. . . . 3750 cc. Volume of lead solution 75 "
Volume of juice 3675 "
Ten per cent of volume of juice
= one tenth of 3675 = 367.5 cc. '
Volume of lead solution = 75 "
Volume of water required = 292.5 '*
The total volume, i.e., 3750 -|- 292.5 = 4042.5 cc. = iio per cent of the volume of the juice (3675 cc).
Having diluted the juice and lead solution to 4042.5 cc, mix and filter off a few cubic centimetres, and polarize in a 20-centimetre tube:
Polariscopic reading = 38.3.
In Schmitz' table, in the column headed 12, the nearest degree Brix to the observed degree, and opposite 38, the integral part of the polariscopic reading, note the number 10.36; in the small table at the bottom of the page, opposite .3, the decimal part of the polariscopic reading, note the number .08, and add this to the number obtained above for
ANALYSIS OF THE JUICE. 77
the completed percentage: 10.36 + .08 = 10.44, t^^ P^*" ^^^^ sucrose in the juice.
Method Employing Dry Subacetate of Lead (Home's Method). — The storage and preservation of composite samples of juice and their subsequent analysis are greatly facilitated by Home's dry subacetate of lead method. This method was designed priinarily to eliminate the error arising from the displacement of a part of the solution by the lead precipitate; it not only accomplishes this end, but also greatly facilitates the analytical work by obviating the necessity of measurements.
For ordinary analytical purposes with juices, a small quantity of finely powdered anhydrous subacetate of lead is added to an indefinite volume of juice and the whole is thoroughly mixed by shaking. It is usually advisable to add also a small quantity of dry sharp sand with the lead. The sand is for the purpose of breaking up any portion of the imperfectly precipitated impurities that may be occluded by a coating of lead precipitate.
After thorough shaking, the mixture is poured upon a filter and the filtrate is polarized as usual. The per cent, sucrose is ascer- tained by dividing the observed reading of the polariscope by i.i and using the quotient in connection with Schmitz table, page 285, In this and similar calculations, the uncorrected degree Brix or the degree Brix at the temperature of the sucrose test, is used.
Manifestly the use of a table or extended calculation could be avoided in this method, by measuring a portion of the filtrate with a sucrose pipette as in 68, but the table is usually the more convenient.
Dr. Home's method is especially convenient in the preparation of composite samples of juices. A quantity of the dry lead estimated to be sufficient to defecate the entire sample, is placed in a jar or large bottle. A measured quantity of the juice is drawn from each measuring-tankful and added to this lead. The contents of the jar should be thoroughly mixed after each addition. The analysis is conducted as has been described above.
70. Notes on the Clarification of Samples for Polarization. — Too little subacetate of lead solution or a decided excess in the clarification may result in cloudy fil- trates, or solutions which filter too slowly. Experience will soon enable one to estimate the proper amount of the lead solution to use. Sufficient of the lead nalt must be used, not
78 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
only to produce a clear filtrate, but to precipitate all the matter precipitable by this reagent. This is essential, since the beet contains other optically active bodies than sucrose (36).
71. Remarks on the Reducing Sugars in Beet Products. — Beet juices and products, under normal con- ditions, do not usually contain more than traces of reducing sugars. There is a reducing substance present in small quantity, however, of which little is known. It is usually termed " Bodenbender's substance," from the name of the chemist who first reported its presence. There is little probability of inversion in the processes of manufacture, except at the diffusion-battery, since the liquors are always more or less alkaline. There is probably rarely any inver- sion in the diffusion process, except during very irregular work or in treating unsound beets. In view of these facts, the beet-sugar chemist is not often called upon to make reducing sugar determinations, except in the estimation of sucrose by the chemical inversion method. The methods of estimating reducing sugars are given quite fully in the following pages, for use in any work in which chemical methods may be required.
72. Determination of Reducing Sugars (Glu- cose, etc.). Gravimetric Methods.— In selecting a method for reducing sugars, the analyst should be guided by the probable composition of the material under examination.
Gravimetric Method for Material containing I percent or less of Invert-sugar ' and a High Percentage of Sucrose. — Dis- solve 20 grams of the material in nearly loocc. of water. If necessary, clarify with subacetate of lead {see 74), precipi- tate the excess of lead by means of sodium carbonate in small excess, complete the volume to lOocc, mix thoroughly and filter. This clarification is usually advisable. Place 50 cc. of Soxhlet's solution (192) in a beaker and add 50 cc. of the sugar solution. Heat slowly, taking about four minutes to reach the boiling-point, and boil two minutes. These directions should be strictly complied with. After the completion of the two minutes' boiling add 100 cc. of cold re-
^ The reducinff sugar of the beet and beet products is probably the re- sult of inversion of sucrose. The methods described for invert-sugar are applicable.
ANALYSIS OF THE JUICE.
79
cently boiled distilled water. Determine the copper, in the precipitate, by one of the following methods : (i) Filter im- mediately under pressure, using the filter-tube described below. The filter-tube, Fig. 40, consists of a 6-inch hard glass tube about | inch in diameter, into one end of which is sealed a tube about 3 inches long and of con- ^venient size for inserting into the stopper of the .filtering apparatus such as that shown in Fig. 49. A perforated platinum disk A, A' is sealed into the bottom of the large tube as a support for an as- bestos felt filter. To prepare the tube for filter- ing, place it in position in the stopper of the fil- tering apparatus, start the filter-pump, then pour water containing finely divided asbestos in suspen- sion upon the disk. The asbestos forms a film or felt; dry and weigh. Moisten the felt before commencing the filtration. A funnel should be used in pouring the liquid and precipitate into the filter-tiibe, to prevent the cuprous oxide from adhering to the walls of the tube, near the top. F'g- 40. Transfer all of the precipitate to the filter and wash thoroughly with hot water. After washing with water pass a few cc. of alcohol through the filter and finally a little ether. 'Dry the precipitate. Pass a continuous current of pure, dry hydrogen through the tube, at the same time gently heating the cuprous oxide, with a Bunsen burner, until it is completely reduced to the metallic state; cool in a current of hydrogen and weigh.
(2) Filter immediately after the reduction is completed, using a Gooch crucible. Wash the beaker and precipitate thoroughly with hot water, but without any effort to transfer the entire precipitate to the crucible. Wash the asbestos film and the adhering cuprous oxide back into the beaker, using hot dilute nitric acid. After the copper is all in solution, filter through a Gooch crucible, using a very thin asbestos film, and wash thoroughly with hot water. Add 10 cc. of dilute sulphuric acid, containing 200 cc. acid of 1.84 specific gravity, per litre, to the filtrate and evaporate it until the copper salt has largely crystallized. Heat care- fully on a hot iron plate or a sand-bath until the evolution
80
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
of white fumes. Add 8 to lo drops of nitric acid, specific gravity 1.42, and rinse into a platinum dish of 100 to 125 cc. capacity. Precipitate the copper on the dish by electrolysis. Wash the copper thoroughly with water before breaking the current ; remove the dish from the circuit, wash with alcohol and ether successively, and dry at a temperature that can easily be borne by the hand, cool and weigh. A beaker may be substituted for the platinum dish, the copper being deposited upon a platinum cylinder.
When a direct current is used in lighting the sugar-house, it is the most convenient source of electricity for the deposi- tion of the copper. The current must be passed through a resistance or regulator in addition to the lamp. A convenient and durable regulator is shown in Fig. 41. C is a glass tube partly filled with water slightly acidulated with sulphuric acid ; the wire A connects with a platinum wire sealed into the tube ; ^ is a glass tube through which a copper wire extends and connects with a platinum wire E sealed into this tube. The tube B may be slipped up or down, thus regulating the distance between the wires E and A and regulating the current. The twin wire M\s separated, severed, and one end, Z>, connected with the platinum dish in which the copper is to be deposited, and the other with the regulator i5, thence through the acidulated water and A with the platinum cylinder f\ suspended in the copper solution.
II (3) Collect the suboxide in a weighed Gooch
crucible, wash as indicated in (3), following the water first with a little alcohol, then with a few drops of ether. Place the crucible in a water oven and dry 30 minutes. Weight of suboxide of copper X -888 = weight of copper reduced. Fig. 41. Having determined the weight of copper
reduced by one of the above-described methods, ascertain from
ANALYSIS OF THE JUICE.
81
Herzfeld's table the per cent of invert-sugar corresponding to the weight of copper.
Herzfeld's Table for the Determination of Invert- sugar IN Materials Containing i Per Cent or Less OF Invert-sugar and a High Percentage of Sucrose.
|
Copper |
Copper |
Copper |
|||
|
reduced by lo Grams |
Invert- Sugar. |
reduced by 10 Grams |
Invert- |
reduced by 10 Grams |
Invert- |
|
of |
of |
sugar, |
of |
sugar. |
|
|
Material. |
Material. |
Material. |
|||
|
Milligrams. |
Per Cent, |
Milligrams. |
Per Cent. |
Milligrams. |
Per Cent. |
|
50 |
0.05 |
120 |
0.40 |
190 |
0.79 |
|
55 |
0.07 |
125 |
0.43 |
195 |
0.82 |
|
60 |
0.09 |
130 |
0.45 |
200 |
0.85 |
|
65 |
O.II |
135 |
0.48 |
205 |
0.88 |
|
70 |
0.14 |
140 |
0.51 |
210 |
0.90 |
|
75 |
0.16 |
145 |
0.53 |
215 |
0.93 |
|
80 |
0.19 |
150 |
0.56 |
220 |
0.96 |
|
85 |
0.21 |
155 |
0.59 |
225 |
0.99 |
|
90 |
0.24 |
160 |
0.62 |
230 |
1.02 |
|
95 |
0.27 |
165 |
0.65 |
235 |
1.05 |
|
100 |
0.30 |
170 |
0.68 |
240 |
1.07 |
|
105 |
0.32 |
175 |
0.71 |
245 |
1. 10 |
|
110 |
0.35 |
180 |
0.74 |
||
|
115 |
0.38 |
185 |
0.76 |
Gravimetric Method for Materials containing more than I Per Cent of Invert-sugar. — Prepare a solution of the material to l)e examined, in such a manner that it contains 20 grams in 100 cc. ; clarify and remove the excess of lead with a small excess of sodium carbonate {see 74). Prepare a series ef solutions in large test-tubes by adding i, 2, 3, 4, etc., cc. of this solution successively. Add 5 cc. of the Soxhlel solution (192) to each, heat to boiling, boil two minutes and filter. Note the volume of sugar solution that gives the filtrate lightest in tint but still distinctly blue. PJace twenty times this volume of the solution in a loo-cc. flask, dilute to the mark and mix well. Use 50 cc. of this solution for the determination, which is conducted as under the preceding method, for materials containing i per cent or less of invert-sugar, until the weight of copper is
82 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
obtained. For the calculation of the result use the follow- ing formulae and table of factors of Meissl and Hiller:
Let Cu = the weight of copper obtained; P = the polarization of the sample; IV = the weight of the sample in the 50 cc. of the
solution used for the determination; F= the factor obtained from the table for conver- sion of copper to invert-sugar;
— = approximate absolute weight of invert-sugar
= Z;
100 Z X -zzr = approximate per cent of invert-sugar =_y; fV
looP „ , . , ,
-— = J?, relative number for sucrose;
P + y
100 — P = /, relative number for invert-sugar;
CuP
W
= per cent of invert-sugar.
Z facilitates reading the vertical columns; and the ratio P to /, the horizontal columns of the table, for the purpose of finding the factor, P, for the calculation of the copper to invert-sugar.
Example. — The polarization of the sugar is 86.4, and 3.256 grams of it, W, are equivalent to 0.290 gram of copper. Then
Cu .200 ^
„ 100 100
^ X ;^ = . 145 X^;^^ = 4.45 =>•; I OOP _ 8640 _ _
Jh=7~ 86.4+ 4.45 ~^^*'~ '
100 — p = 100 — 95.1 = /= 4.9; ^: 7=95.1 :4.9.
By consulting the table it will be seen that 150 mg. in the vertical column are nearest the value of Z, 145 mg., and
ANALYSIS OF THE JUICE.
83
the horizontal column headed 95 : 5 is nearest the ratio R to /, 95.1:4.9. Where these columns meet we find the factor 51.2 which enters into the final calculation:
CuF W
.290 X 51-2 3.256
4.56 per cent of invert-sugar.
MEISSL AND KILLER'S FACTORS FOR THE DETERMINATION OF MORE THAN 1 PER CENT OF INVERT SUGAR.
|
Ratio of Sucrose to Invert-sugar = R'.I. |
Approximate Absolute Weight of Invert-sugar |
= Z. |
|||||
|
200 Milligr. |
175 Milligr. |
150 Milligr. |
125 Milligr. |
100 Milligr. |
75 Milligr. |
50 Milligr. |
|
|
Per Ct. |
Per Ct. |
Per Ct. |
Per Ct. |
Per Ct. |
Per Ct. |
Per Ct. |
|
|
0:100 |
56.4 |
55.4 |
54.5 |
53.8 |
53.2 |
53.0 |
53.0 |
|
10:90 |
56.3 |
55.3 |
54.4 |
53.8 |
53.2 |
52.9 |
52.9 |
|
20:80 |
.56.2 |
55.2 |
54.3 |
53.7 |
53.2 |
52.7 |
52.7 |
|
30:70 |
56.1 |
55.1 |
54.2 |
53.7 |
53.2 |
52.6 |
52.6 |
|
40:60 |
55.9 |
55.0 |
54.1 |
53.6 |
53.1 |
52.5 |
52.4 |
|
50:50 |
55.7 |
54.9 |
54.0 |
53.5 |
53.1 |
52.3 |
52.2 |
|
60:40 |
55.6 |
54.7 |
53.8 |
53.2 |
52.8 |
52.1 |
51.9 |
|
70 : 30 |
55.5 |
54.5 |
53.5 |
52.9 |
52.5 |
51.9 |
51.6 |
|
80:20 |
55.4 |
54.3 |
53.3 |
52.7 |
52.2 |
51.7 |
51.3 |
|
90:10 |
54.6 |
53.6 |
53.1 |
52.6 |
52.1 |
51.6 |
51.2 |
|
91 :9 |
54.1 |
53.6 |
52.6 |
52.1 |
51.6 |
51.2 |
50.7 |
|
92:8 |
53.6 |
53.1 |
52.1 |
51.6 |
51.2 |
50.7 |
50.3 |
|
93:7 |
53.6 |
53.1 |
52.1 |
51.2 |
50.7 |
50.3 |
49.8 |
|
94:6 |
53.1 |
52.6 |
51.6 |
50.7 |
50.3 |
49.8 |
48.9 |
|
95:5 |
52.6 |
52.1 |
51.2 |
50 3 |
49.4 |
48.9 |
48.5 |
|
96:4 |
52.1 |
51.2 |
50.7 |
49.8 |
48.9 |
47.7 |
46.9 |
|
97:3 |
50.7 |
50.3 |
49 8 |
48.9 |
47.7 |
46.2 |
45.1 |
|
98:2 |
49.9 |
48.9 |
48.5 |
47.3 |
45.8 |
43.3 |
40.0 |
|
99:1 |
47.7 |
47.3 |
46.5 |
45.1 |
43.3 |
41.2 |
38.1 |
The above methods have been taken, with a few changes in the wording and with additions, from Bulletin No. 46, U. S. Department of Agriculture.
Gravimetric Method using SoIda'inV s Solution.'^ — Place 100 to 150 cc. of Soldaini's solution (103) in an Erlenmeyer
Traits d'' Analyse des Matiires Sucrdes^ D. Sidersky, p. 148.
84 HANDHOOK FOR SUGAU-HOUSE CHEMISTS.
flask; boil five minutes; add a solution containing lo grams of the material previously clarified with subacetate of lead, if necessary, the excess of lead being removed with small excess of carbonate of sodium {see 74); boil five minutes. In boiling always use the naked flame. Having completed the reduction, remove the flask from the flame and add loo cc. cold distilled water. Filter immediately through a Gooch crucible and determine the copper in the precipitate by the electrolytic method, or collect the precipitate in a filter-tube, Fig. 40, and reduce in hydrogen. These methods are described on page 79.
The weight of metallic copper X 0.3546 -i- weight of the material used in the determination X 100 = per cent invert- sugar. It is claimed that this method is very exact and that invert-sugar can be determined to within .01 per cent with certainty.
73. Deteriiiiiiatioii of Reducing Sugars (Glu- cose, etc.). Volumetric Methods.—^ Modification of Violette's Method. — This is the rapid method used very generally in cane-sugar-houses. If always conducted under the same conditions as to dilution, method, and time of heating, the results are approximately correct and are comparable with one another.
Take a definite weight of the juice, a multiple of 5 grams is most convenient, varying this quantity with the amount of reducing sugar present, clarify with subacetate of lead, precipitate the excess of lead with small excess of carbon- ate of sodium {see 74), and dilute to 100 cc; mix and filter.
A sufficient quantity of the juice should be taken, if prac- ticable, to give a reading on the burette of approximately 20 cc. in the titration to be described. In seed selection, as will be explained, it is unnecessary to adhere strictly to these specifications, but in using this method with other products they should, as far as practicable, be complied with.
It is convenient in this work to use an automatic, zero burette, in measuring the copper solution. Such a burette as designed by Squibb is shown in Fig. 42. This burette is filled by suction, as with a pipette, applying the suc- tion at the mouthpiece shown at the end of the rubber
ANALYSIS OF THE JUICE.
85
tube. The reagent is drawn into the burette to a point a little above the zero mark, the mouthpiece is then released and the liquid siphons back into the reservoir, leaving the burette filled to exactly zero. A wash-bottle containing caustic soda solution should be connected with the air-inlet near the reservoir to prevent the entrance of carbonic acid. This is one of the most conven- ient of the many forms of auto- matic burettes. These burettes may be used with advantage in nearly all the measurements required in volumetric analysis, in the sugar-house laboratory.
Measure lo cc. of Violette's modification of Fehling solution (195) into a large thin glass test-tube, 1.5 X 9 inches and di- lute it with an equal volume of water. If the alkaline copper re- agent be prepared with the cop- per in one solution and the alkali in a second, use 10 cc. of each solution and omit the addition of the 10 cc. of water. Heat the reagent in the tube, over the naked flame of a lamp, to the boiling-point, then add a few cubic centimetres of the sugar solution, and boil two minutes. A sand-glass is convenient for use in timing the boiling. Repeat these operations until the blue color almost disappears, taking care to add the juice very gradually as this point is approached. After the first boiling, it is only necessary to boil the liquid a few seconds each time. Now add the juice, a drop or two at a time, until the blue color disappears. Filter off a small portion of the liquid, using a Wiley or Wiley-Knorr filter- tube, and proceed as described farther on.
Fig. 42.
S6
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Wiley's filter-tubes, Fig. 43, a, are made from glass tubing about one fourth inch in diameter and about ten inches in length. One end of the tube is softened in the flame of a lamp and then pressed against a block of wood to form a shoulder ; a piece of washed linen is stretched over this end and is held in place by means of a strong thread. In using these tubes the filter end is dipped into water in which very finely divided asbestos is sus- pended, and by suction, with the mouth, the cloth is covered with a film of this substance. Knorr's modification of these tubes is very convenient, and is preferred by many chem- ists. These filter-tubes, Fig. 43, <^, are of small diameter and are tipped with platinum-foil. The asbestos is applied as with the Wiley tubes. With the Wiley filter, the filtrate must be poured from the tube ; with the Knorr tube, the liquid is expelled through the platinum tip, after wiping off the asbes- tos with a cloth. These tubes should be dipped in dilute acid after use, then thor- oughly washed.
Many chemists prefer to remove a drop of the solution and place it on a piece of quan- titative filter -paper. The precipitate re- mains in the centre of the moistened spot with the filtered solution around it. A drop of ferrocyanide of potassium solution acidu- lated with acetic acid is placed adjacent to the first drop. There will be a coloration where the two solutions touch one another if there be still copper in solution.
If a portion of the solution be filtered off in one of the tubes above described, pour it into a few drops of acetic acid, to acidity, in a depression in a white porcelain test-plate ; the acid discharges the color from tha solution and neutralizes the alkali of the Violette's solution. Add a drop of* a dilute solution of ferrocyanide of potassium.
Fig. 43.
ANALYSIS OF THE JUICE. 87
yellow prussiate of potash ; a brown coloration shows the copper has not all been reduced, and that more juice must be added. The juice must be added very carefully as the test reaction diminishes in intensity, until finally all the copper is reduced, there being no further brown colora- tion. The burette reading is now made.
It is advisable to make a preliminary test to guide in the dilution of the juice and to show within a few tenths of a cubic centimetre the volume of juice required for the re- duction of the copper, and then add nearly all the sugar solution at one time in a final test.
A porcelain dish may be substituted for the large test- tube, but on account of the small surface exposed for evaporation, the latter is preferred. Calculations. W = the weight of juice in i cc. of the solution ; B = the burette reading ;
D . A • 0-05 X loo
Per cent reducing sugar = x = .
rr /\ Jj
When ^ is .05 gram the formula reduces to jc = —
H
or j; = reciprocal of the burette reading multiplied by 100.
A table of reciprocals is given on page 294 to simplify these calculations.
If a multiple of 5 grams of juice be diluted to 100 cc. for this determination, the reciprocal of the burette reading multiplied by 100 is the same multiple of the percent of re- ducing sugar.
If 5 grams in 100 cc. should prove a too-concentrated solu- tion, dilute to 200, 300, etc., and multiply 100 times the reciprocal of the burette reading by 2, 3, etc.
If 5 cc. or a multiple of 5 cc. of juice be used for the an- alysis, the above-mentioned method of calculation may be employed, but the value of x must be divided by the spe- cific gravity of the juice to reduce it to terms of the weight of the juice.
On account of the very small percentage of reducing sugar in beet-juices a much higher burette reading than 20 cc. may be necessary, even using the undiluted juice; fur- ther, for the same reason, it may be necessary to use only
88 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
5 cc. of Violette's solution. It is preferable in such cases to use a gravimetric method.
The accurate determination of reducing sugar by this method requires rapid work and considerable practice.
Sidersky's Volumetric Method, using Soldaini's Sohition} Standardize the Soldaini solution by means of a solution of invert-sugar containing 5 grams of the reducing sugars per litre. Proceed as in 73, except that the end reaction is judged by the disappearance of the blue color instead of by the ferrocyanide test. The method described in 73 is probably applicable, though Sidersky was guided solely by the disappearance of the blue color.
This method has the advantage of freedom from the source of error, due to the presence of sucrose, in the older method of Violette. For highly colored products, such as molasses, etc., Sidersky has modified his method as fol- lows: Dissolve 25 grams of the material in water, add suf- ficient subacetate of lead for clarification {see 74), dilute to 200 cc, mix and filter. To 100 cc. of the filtrate add 25 cc. of a concentrated solution of sodium carbonate, mix and filter; of this filtrate use 100 cc, corresponding to 10 grams of the material, for the reduction. Boil 100 cc. of Soldalni's solution five minutes in a flask over a naked flame, then add the sugar solution, little by little, continuing the heat- ing an additional five minutes. Remove the flask, add 100 cc. cold distilled water, and collect the precipitate upon an asbestos felt in a Gooch crucible, with the assistance of a filter-pump. Wash the precipitate with hot water until the wash-waters are no longer alkaline. Three or four washings are usually sufficient. Wash the cuprous oxide into an Erlenmeyer flask and add 25 cc. normal sulphuric acid (199) and two or three crystals of chlorate of potas- sium, then heat gently until the cuprous oxide is completely dissolved. Titrate the solution with a standard alkali solution (201), determine by diff^erence the volume of the acid saturated, and from this the amount of copper re- duced. It is preferable to use a half-normal solution of ammonia (201) for this titration, letting the sulphate of copper act as an indicator. Check the ammonia solution
* Train d' Analyse des Matures Sucrees, D. Sidersky, p. 150.
ANALYSIS OF THE JUICE. 89
against the normal sulphuric acid, using 2 cc. of a concen- trated solution of sulphate of copper as an indicator to 25 cc. of the ammonia. Continue the addition of the acid until the blue color disappears.
In making the titration proceed as follows: Cool the sulphate of copper solution, resulting from the treatment of the cuprous oxide with the normal sulphuric acid and chlorate of potassium, add 50 cc. half-normal ammonia solu- tion and titrate back with the normal sulphuric acid. The blue color disappears with each addition of the acid, but re- appears on stirring the solution so long as any unsaturated ammonia remains. When all the ammonia is saturated the color of the solution is no longer blue, but a faint green. Note the burette reading. Each cc. of the sulphuric acid is equivalent to .0317 gram of copper. Multiply the weight of copper by .3546, Bodenbender and Scheller's factor, to obtain the weight of reducing sugar (invert-sugar), or multiply the burette reading by .1124 to obtain the per cent reducing sugar.
Volumetric Permanganate Method.^ The saccharine strength of the solution should be approximately one per cent. The solution should be clarified as usual, and the excess of lead removed (74). Ten cubic centimetres of this solution are placed in a porcelain dish with a consider- able excess of copper solution (102). If the saccharine solution contain no sucrose, heat to the boiling-point and maintain this temperature until the reducing sugar is oxi- dized. When sucrose is present the temperature should not exceed 80° C, and the heating should be continued longer than at the higher temperature. There should be enough of the copper solution used to maintain a strong blue coloration at the end of the reaction. Ervin E. EwelP advises using the following modification of the method of determining the weight of copper reduced: Collect the precipitate on asbestos in a Gooch crucible, with the as- sistance of a filter-pump, and wash thoroughly with hot recently boiled distilled water. Transfer the asbestos, with
* Principles and Practice 0/ Agricultural Analysis^ H, W. Wiley, 3,
134- » Op. cit.y 136.
90 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
as much of the precipitate as possible, to the beaker in which the precipitation was made, beat it up with 25 to 30 cc. of hot recently boiled distilled water, and add from 50 to 75 cc. of a saturated solution of ferric sulphate in 20 per cent sul- phuric acid; pour this solution through the crucible to dis- solve adhering portions of the cuprous oxide. The precipi- tate must be well beaten up with the water to break all large lumps or there may be difficulty in effecting solution with the ferric salt. After the solution is complete, titrate with per- manganate of potassium of such strength that i cc. is equiv- alent to .01 gram of copper (203), or decinormal perman- ganate solution (202) may be used. In addition to stand- ardizing the permanganate solution with metallic iron or oxalic acid, as is usual for general purposes, it should be standardized, for this method, by titrations with copper, reduced by solutions of invert-sugar which have been stand- ardized by the gravimetric method (72). The invert-sugar value of I cc. of the permanganate solution is thus ascer- tained for use in calculating the percentage of reducing sugar in the material.
Ewell's modification of the permanganate method of determining the amount of reduced copper, is also recom- mended for use in the methods in 72,
74. Notes on the Deterniiiiation of Reducing Sng"ars. — Edson, Pellet and other chemists have shown that a part of the reducing substances in certain sugar- house products is precipitated by subacetate of lead, but not at all or to a very small extent with the normal acetate. Edson advises that the solutions be acidulated with acetic acid before filtering oft" the lead precipitate, and finds that acidulation practically obviates this source of error. The author's experience confirms Edson's observations. Born- trSger ' states that sodium sulphate is preferable to sodium carbonate for the precipitation of the excess of lead. Ac- cording to his experiments, an excess of the sulphate is less objectionable than of the carbonate. The carbonate is almost exclusively used by sugar-house chemists for the removal of the excess of lead.
* Zeit, Angew. Chem., 1892, 333.
ANALYSIS OF THE JUICE.
91
75. Determination of the A^\\,— Sulp hated Ash.—
)ty lo grams of the juice in a tared platinum dish. Add a
drops of concentrated sulphuric acid to moisten the
ndue, and heat over the flame of a lamp or in a muffle at
redness until the organic matter is charred, then in-
;ase the temperature to bright redness and heat until all
le carbon is consumed. In the event of too high a tem-
srature, the ash will melt and thus may vitiate the results.
The ash so obtained is termed the " sulphated ash," since
ertain of the mineral constituents are converted into
llphates by the acids. It is estimated that the average
icrease in the weight of the ash, due to the formation of
mlphates instead of carbonates, is lo per cent, hence a
>rrection of one tenth is customary to reduce the sulphated
fcSh to terms of the normal or carbonated ash.
Calculation. — Weight of ash X 9 = per cent normal ash.
tt is usually more convenient to measure lo cc. of the juice
lan to weigh lo grams. In such cases calculate as follows :
Weight of sulphated ash X 9 Specific gravity of the juice
The above method of incineration 1$ usually employed, since there is isually difficulty in the direct inciner- don of saccharine materials. Normal Ash. — The normal or car- bonated ash may be obtained by Bey- er's method, as follows : Dry lo grams, or lo cc, of the juice in a platinum dish, then heat carefully to caramelize the sugar, but not enough to char it; add 2 cc. benzoic acid solution, 25 grams benzoic acid in 100 cc. of 90 ^ alcohol, and warm gently to expel the alcohol. Char the sugar at a low heat, at the same time volatilizing the acid ; incinerate at a low red heat. The ash consists largely of alkaline carbonates, which, on exposure to the air, quickly absorb moisture. Cool the ash in a desiccator and weigh quickly.
per cent normal ash.
Fig. 44.
Fig. 45.
[;■
|
Fig. 46. |
||
|
1 |
4. |
|
|
1 T i 1 |
1 ©' |
... |
|
— , |
92 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
The weight of the ash -r- the weight of the juice X loo = per cent ash.
The following described muffle, devised by Schweitzer and Lungwitz,^ is effective, and may be cheaply constructed for sugar purposes.
In a French clay muffle a narrow slot is cut the length of the bottom, Fig. 44, a, b ; holes are drilled in the walls atr, </, Fig. 45, and heavy platinum wires are inserted. These wires are supports for a trough of platinum-foil, Fig. 45, w, x,y, z, upon which the dishes rest during the incineration. A hole is cut in the dome of the muffle at t, Fig. 46. The muffle is placed on a support and is heated by wing-top burners.
76. Determination of the Total Nitrogen Albuminoids. — The beet contains, in addition to albu- minoid matter, several nitrogenous substances classified as amido-compounds. Some of these substances may be read- ily separated, others require complicated analytical proc- esses. For an extended study of the nitrogenous bodies in agricultural analysis, Wiley's Principles and Practice of Agricultural Analysis is recommended. Allen gives methods for several of the amido-compounds in Vol. Ill, Part III, Commercial Organic Analysis. E. O. von Lippmann has published a very exhaustive study of the nitrogenous con- stituents of the beet-juice in Berichteder deutschen chemischen Gesellschaft, 29, 2645. A translation of this paper is pub lished in Bulletin de V Association des Chimistes de France^ 14, 6gi and 8iq. See also this book, page 201. It has long been customary in plant analysis to multiply the per cent of total nitrogen by 6.25 and term the product the per cent of "albuminoids." The figures obtained in this way are often of value in sugar-house work.
A modification of Kjeldahl's moist combustion process' may be conveniently employed for nitrogen determinations :
(i) The Digestion. — Ten cc. of the juice, dried in a small capsule, are brought into a 550-cc. digestion-flask with approximately .7 gram of mercuric oxide and 20 cc. of
^Journ. Am. Chem. Soc, 16, 151
» Adapted from Bulletin 46, Div. Chem., U. S. Dept. Agric.
ANALYSIS OF THE JUICE. 93
sulphuric acid. The flask is placed on a frame in an in- clined position, and heated below the boiling-point of the acid for from 5 to 15 minutes, or until frothing has ceased. If the mixture froth badly, a small piece of paraflSne may- be added to prevent it. The heat is then raised until the acid boils briskly. No further attention is required till the contents of the flask have become a clear liquid, which is colorless, or at most has only a very pale straw color. The flask is then removed from the frame, held upright, and, while still hot, potassium permanganate is dropped in care- fully and in small quantity at a time, till, after shaking, the liquid remains of a green or purple color.
(2) The Distillation. — After cooling the contents of the flask, add about 200 cc. of water, then a few pieces of gran- ulated zinc and 25 cc. of potassium-sulphide solution, 40 grams commercial potassium-sulphide in 1000 cc. water, shaking the flask to mix its contents. Next add 50 cc. of a saturated caustic-soda solution, free from nitrates, or suf- ficient to make the reaction strongly alkaline, pouring it down the side of the flask so that it does not mix at once with the acid solution. Connect the flask with the con- denser, which should be of block-tin, mix the contents by shaking, and distil until all the ammonia has passed over into the standard acid. The first 150 cc. of the distillate will generally contain all of the ammonia. This operation usually requires from 40 minutes to one hour and a half. The dis- tillate is then titrated with standard ammonia, using cochi- neal as an indicator, and the calculations are made as usual. Previous to use, the reagents should be tested by a blank experiment with sugar, which will partially reduce any nitrates present, which might otherwise escape notice.
77. Determination of the Total Solids.— The degree Brix is usually considered as representing the total solid matter in solution. An accurate determination of the total solids can only be made by actually drying the juice in an oven.
The problem of drying saccharine materials, to a constant weight, is not as simple as may appear at first glance. A number of methods have been devised for this purpose, two of which are given.
94 HAKDfiOOK FOR StJGAR-fiOUSE GfiEMlSTS.
.Carr and Sanborn'' s Method for Drying Sugar-house Prod- ucts ,-^T\i\s is a modification of the ordinary pumice-stone method. Prepare pumice-stone in two sizes. One size should pass a i-mm. sieve and the other should pass a 6- mm. sieve, circular perforations. Place a layer 3 mm. thick of the finer pumice-stone on the bottom of a small metal dish, and a layer of the coarse, 6 mm. to 10 mm. thick, upon the first layer, and dry and weigh. Tin caps for bottles are inexpensive and well adapted for use in this determination.
Fig. 47. Each dish is used but once, then thrown aside. Distribute about 5 grams of juice over the pumice-stone, weighing it accurately from a weighing-bottle. Dry this juice to a constant weight in a water-oven or in a vacuum-oven at no° C, making trial weighings at intervals of two hours. Calculation : Weight of solid matter -r- weight of juice em- ployed X 100 = per cent total solids.
Method of Drying Employing a Vacuum Apparatus. — This method was suggested to the author by that of Courtonne,' from which it differs in several important particulars, no- tably in the construction of the oven and drying-bottles. Courtonne heats the bottles by immersion in hot water.
^Manuel-Agenda des Fabricants de iSucre, MM. Gallois and Dupont, 1891, p. 215.
AKALYSIS OF THE JUICE. 95
The oven and bottles are shown in section in Fig. 47. The walls of the oven are double and are filled with plaster of Paris, C; the bottom is also double, the space* being filled with air. A fan, £>, driven by a toy engine, or other suit- able means, agitates the air inside the oven and insures a strictly uniform temperature in all parts.
The drying-bottles. A, are connected by means of short tubes with a central vacuum-pipe, JS, which is in turn con- nected with an ordinary filter-pump or the third pan of the triple-effect. Each bottle may be removed by closing the cock G without disturbing the others. A small trap, //, of glass, shown also in detail at the right of the oven, pre- vents any moisture which may condense in the tubes from falling back into the bottle.
The following procedure is advised : Place a quantity of small fragments of pumice-stone suflScient to absorb 5 cc. of juice, in a weighing-bottle, dry in the oven, cool, insert the glass stopper and weigh ; distribute a definite weight of the juice, approximately 5 grams, upon the pumice-stone. Insert the stopper, provided with the trap, in the bottle, and connect with the vacuum-pipe. A vacuum of 20 inches is usually all that is required, and in fact is preferable to a higher vacuum. The drying is usually complete in one hour; it is advisable to dry to a practically constant weight, weighing at intervals of one hour or more as may be con- venient. The calculations are made as in the preceding meihod.
This apparatus may also be used for drying in an inert gaa.
The per cent total solids by the spindle, the degree Brix, and the per cent total solids by drying, are employed in calculating the purity coeflScients or quotients (106).
78. Acidity of the Juice. — The normal juice of the beet and the diffusion-juice are always acid. This acidity is due to a number of organic acids. It is not often neces- sary to determine the acidity of the juice. This determi- nation is made by a titration with a decinormal alkali solution (201). It is somewhat difficult to determine the end reaction, since the color of the juice obscures the color of the indicator to some extent. Phenolphthalein is usually Employed as the indicator. Collier recommended the
96 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
use of logwood solution as an indicator in determining the acidity of sugar-cane juices, and in the author's experience it has been satisfactory.
The acidity may be expressed in terms of the number of cubic centimetres of normal alkali solution required to neu- tralize the juice or, for comparative purposes, more conven- iently as cubic centimetres of normal alkali per loo grams of sucrose or lOO degrees Brix.
79. Analysis of Carbonated Juices. — The methods of analysis of the purified juices are the same as for the raw juice, except that the carbonated juice must receive an additional treatment with carbonic acid to pre- cipitate all of the calcium. This is evidently necessary, since these analyses are made in part for the purpose of comparing the purity of these juices with that of the diffusion-juice before treatment.
80. Alkalinity of the Juice.— It is occasionally necessary to determine the total alkalinity of the juice after liming and before carbonatation; it is also necessary at very frequent intervals to determine the total alkalinity of the carbonated juices, in the control of the carbonatation process. In many factories an alkalimetric method is employed in ascertaining when to shut off the carbonic acid gas in the carbonatation of each tankful of juice.
The total alkalinity is usually expressed in terms of the grams of lime (CaO) per litre of juice, although the alkalinity is in part due to the presence of caustic alkalis.
Methods are usually employed, in the control of the car- bonatation of the juice, which are very rapid and well adapted to the use of unskilled employes, but which yield only moderately accurate results (81).
It is advisable that the rapid methods indicated be occa- sionally checked in the laboratory. This is necessary in order to know to what extent the results vary from the truth, that the carbonatation may be the more satisfactorily controlled.
81. Rapid Methods of Moderate Accuracy for
the Alkalinity, of Juices.— (i) Standard Add Solution. — Prepare a standardized solution of sulphuric acid contain-
AlfALTSIS OF THE JUICE. 97 '
ing 35 grams of the monohydrated acid (HjSO*) in looo cc. {See 200.) The strength of this solution is such that i cc. will neutralize 0.02 gram of lime (CaO).
This solution is used for limed juices and juice from the first carbonatation. A more dilute acid is employed for the titration of juice from the second carbonatation. This acid is prepared by diluting 100 cc. of the above standard acid to 1000 cc, and contains 3.5 grams of sulphuric acid in 1000 cc.
Indicators. — As great accuracy is not necessary in this determination, indicators which are more or less affected by carbonic acid may be employed. Among those most commonly used are neutralized corallin, phenolphthalcin, cochineal, etc. A few drops of the solution of the indicator are added to the juice, or in this class of analyses, with cer- tain indicators, more conveniently to the acid solution, when standardizing it, and before completing the volume to 1000 cc. {See^lS,)
Titration. — Measure 20 cc. of the juice into a porcelain dish or into a small Erlenmeyer flask. If the flask be used, it should be placed over a sheet of white paper or a por- celain slab during the titration.
Except in the case of the limed juice, before carbonata- tion, the liquor should be filtered.
Add a few drops of the indicator to the juice, if it be not already contained in the standard acid, and deliver the acid cautiously from a burette. Note the point when the alkalinity is saturated by the change in the color of the indicator, and read the burette.
Calculation. — I cc. of stronger acid solution neutralizes 0.02" gram of lime (CaO); hence for each cc. of acid used there is an alkalinity corresponding to 0.02 gram of lime per 20 cc. of juice, or to o.i gram per 100 cc. of juice, or I gram per litre of juice.
Example. — 20 cc. of juice required 2.2 cc. of the acid. .'. 0.02 X 2.2 X 50 = 2.2 grams lime per litre of juice, or the number of cc. of acid used = grams of lime per litre.
The calculations are the same when using the weaker acid with second carbonatation juices, except that i cc. of the acid <:ofresponds to 0.002 gram of lime.
/.
98
HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
(2) Vivien's Method. — This exceedingly convenient and simple method is employed very generally in France. Like the preceding method, it only gives approximately correct results. Vivien employs a solution of sulphuric acid con- taining a small quantity of phenolphthalein, of such strength that one volume of this acid will neutralize one volume of juice containing .05 gram of lime per litre, i.e.^ total alkalinity expressed as lime (CaO).
A specially graduated tube shown in Fig. 48 is used with this method. This tube is divided into six parts of equal _ volume. Each part except the bottom one is subdivided into five parts.
Acid Solution. — Prepare a standardized solu- tion of sulphuric acid containing 0.875 gram of the monohydrated acid (H2SO4) in 1000 cc; add a small quantity of phenolphthalein to the solu- tion before completing the volume to 1000 cc. Standardize by titration against decinormal alkali solution; 10 cc. of the alkali should neutralize 56 cc. of this solution.
Manipulations. — Fill the tube, Fig. 48, to the zero mark with juice; add the standardized acid cautiously, placing the thumb over the mouth of the tube and agitating from time to time. The solution turns red at the first addition of the acid, provided it be not added in excess; finally, when the acid is in very slight excess, the color disappears. The reading on the scale is next made. Every ten divisions correspond to an alkalinity due to i gram of lime per litre of juice, • 48- j^jj(j each division to o.i gram of lime (CaO.) per litre.
For second carbonatation juice, use a much more dilute acid; for example, one half or one fifth the strength of the above. In this case every ten divisions of the scale cor- respond to 0.5 gram or 0.2 gram of lime per litre.
It is evident that these methods are susceptible of many modifications, but for the purposes of this book those described are sufficient.
These methods must be used with caution in analyzing
25
so-
ls:
10—
5=:^
ANALYSIS OF THE JUICE. 99
the juice from the second carbonatation, for the reasons given below.
It is the practice in the second carbonatation to saturate all the lime; hence this process is often termed the " saturation." If this point be passed, the caustic sodium and potassium, which remain as such in the presence of the caustic lime, are converted into carbonates. This is wrong, from manufacturing considerations, and further it would be objectionable to leave lime unprecipitated. It is thus apparent that a process should be employed which will show the exact moment at which all the lime has been combined with the carbonic acid. In practice it is usual to ascertain, in the laboratory, approximately the alkalinity the juice should have when the lime has all been precipitated, and be guided by this in the control of the carbonatation.
The use of phenacetoline is said to be an advantage in this test. It is used in the cold. Degener recommends the use of a few drops of a i per cent solution of phenace- toline in alcohol.
82. Methods for the Deteriiiiiiatioii of the Total Calcium iii the Juice.— Gravimetric Method.— To ICO cc. of the juice add an excess of ammonium hydrate, heat to the boiling-point and filter, should there be a pre- cipitate. Wash the filter with hot water, add an excess of oxalate of ammonium to the filtrate, boil two hours, and let stand several hours ; collect the precipitate in a small quantitative filter and wash with dilute ammonia. The filter and contents are next transferred to a tared platinum crucible, partly dried and the filter charred at a low tem- perature, then ignited until the carbon is removed. Add a small quantity of sulphate of ammonia solution containing chloride of ammonia (see 136), dry at a moderate heat, and ignite at a high temperature. The residue consists of sul- phate of calcium (CaSOi). Cool in a desiccator and weigh. The weight of the calcium sulphate multiplied by .41158 is the weight of calcium oxide (lime) per 100 cc. of juice. This number is practically the percentage of calcium oxide (CaO) by weight in the juice, or the correct percentage is this number divided by the specific gravity of the juice.
100 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Fradiss' Volumetric Method.' — Treat loo cc. of juice as described under the preceding method. Decompose the oxalate of calcium with warm dilute sulphuric acid. The acid combines with the calcium and sets the oxalic acid free. The oxalic acid is determined by means of a i/io normal solution of permanganate of potassium (202).
Titrate the solution without filtering, maintaining a tem- perature of 60° to 80° C. The addition of the permanganate solution should be continued until a permanent pink color is produced.
Calculation. — Multiply the burette reading, the cc. per- manganate solution, by 0.0028 to obtain the weight of cal- cium oxide (CaO), or by 0.002 to obtain the weight of cal- cium (Ca). The numbers so obtained are the per cents by volume of the juice. Divide by the specific gravity of the juice to obtain the corresponding per cents by weight.
Soap Method. — This is an application of Clarke's soap test, used in estimating the hardness of water. The total percentage of calcium as calcium oxide (CaO) may be rapidly and closely estimated by this method. As used by the French it is more convenient for sugar-house purposes than the English method.
Chloride of Calcium or Barium Solution. — Dissolve 0.25 gram of pure chloride of calcium or 0.55 gram of pure crystallized barium chloride (BaCla + 2H3O) in water and dilute to I litre.
Special Burette. — The burette is so graduated that 2.4 cc. correspond to 23 divisions. The zero of the graduation is placed at the second division to allow for the quantity of soap solution required to produce a permanent lather with 40 cc. of distilled water; the 22 divisions correspond to o.oi gram of chloride of calcium dissolved in distilled water: hence a division or 1° corresponds to 0.00045 gram of the chloride in 40 cc, or 0.0114 gram per litre.
Special Bottle. — This bottle is graduated at 10, 20, 30, and 40 cc. Only two of these graduations, viz., at 10 and 40 cc, are used in sugar work.
Method of Making the Test. — Introduce 40 cc. of the cal- cium chloride or barium chloride solution into the special
* Bulletin de P Assoc, des Chimistes de France, 14, 22.
ANALYSIS OF THE JUICE. 101
bottle, and add the soap solution {see 186) little by little, with agitation, until a foam 5 mm. deep forms and persists during 5 minutes. The solution must be vigorously agi- tated by shaking the stoppered bottle after each addition of the soap. If the soap solution be of the correct strength, a volume corresponding to 22 divisions of the burette is required. The burette should always be filled to the division above the zero mark, and the reading should be from zero. If the reading be not 22°, add sufficient cold, recently boiled distilled water to dilute it to this strength.
To ID cc. of the juice in the special bottle, add sufficient cold, recently boiled distilled water to dilute it to 40 cc. Proceed as above, using the standarized soap solution. Multiply the number of *' degrees " read on the burette by 0.0228 to calculate the lime (CaO) per litre of juice. This method may be applied to the sirup, massecuites, and molasses, using i gram of the material diluted to 40 cc.
See page 171 relative to the influence of magnesia in this test. The presence of magnesia, resulting from dolomite in the limestone, may vitiate the results obtained. Parallel determinations by the soap and the gravimetric methods, or an examination of the lime, will show whether sufficient magnesia is present to render this process unavailable. This method is not applicable to the juice from the first car- bonatation.
83. Free and Combined Lime and Alkalinity Due to Caustic Alkalis. Pellet's Method.' — A. Determine the total alkalinity by titration with sul- phuric acid, using litmus as an indicator. The titration must be made at the boiling-point of the juice. Calculate the alkalinity as lime per 100 cc. of juice.
B. Add an equal volume of strong alcohol to a measured portion of the juice; the "free" lime is precipitated as an insoluble saccharate of lime; filter and determine the alka- linity of the filtrate operating upon an aliquot part; calcu- late as lime per 100 cc. of juice. This alkalinity is, however, due to sodium and potassium hydrates, but is expressed as lime for comparative purpos2S. -7
Fabrication du Suc*-e Per.uc^et, Pallet, etc, >18< joi.
102 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
C. The total lime is determined by one of the methods in 82, and is also expressed as lime per loo cc. of juice. The following example illustrates the calculations:
Example.
As Lime per loocc.
\A) Total alkalinity 0.027 gram.
{B) Alkalinity due to soda and potassa 0.021 "
(C) Total lime, including organic salts 0.023 "
Free lime (^ - ^) 0.006 "
Combined lime, /..?., lime salts (C—[^—j9]).. 0.017 '* \
ANALYSIS OF THE SIRUP.
84. Analysis of the Sirup.— The analysis of the sirup is conducted as that of the juice (67 to 83); the same determinations are made, the only variations being in the quantities of the material used for the analysis.
All the portions used for analysis should be weighed, not measured. This is necessary on account of the viscos- ity of the sirup.
: ANALYSIS OF THE MASSECUITES AND MOLASSES.
85. Deteriiiiiiation of the Density. — The deter- mination of the density of massecuites presents certain difficulties which cannot well be avoided, and which com- pel the acceptance of results which are not strictly accurate.
As has been explained, the degree Brix of a solution is the percentage, by weight, of pure sugar which it contains, but it is usually taken as the percentage of solid matter in the solution. The use of a spindle or pyknometer for the determination of the degree Brix, assumes the impurities in the solution, or the non-sucrose, to have the same specific gravity as sucrose. This assumption, unfortunately for the convenience of th,e caerni^i, is far from true, especially in the densei»products and in those from which a part of the sugar has been removed, viz., the second, third, etc., mas-
ANALYSIS OF THE MASSECUITES A^B MOLASSES. 103
secuites and the molasses. The mineral impuruic* Influ- ence the specific gravity very materially, since they differ so widely in specific gravity from the sugars. Since the proportion of inorganic non-sugar increases as one passes from the products of high purity to those of low purity, the difference between the apparent percentage of total solids, as indicated by the density, and the true percentage of total solids, becomes greater.
From this, it is apparent that calculations of the total solids in massecuites, etc., from the density of the product, must be accepted with caution, and then only for compara- tive purposes, when uniform conditions of analysis are maintained.
The methods by dilution and spindling are given in this book for calculating approximate coefl5cients, etc., and must not be assumed to give strictly accurate results. •'■ It is customary to term the degree Brix, as deduced from the specific gravity of the material, the "apparent degree Brix," or simply the "degree Brix"; the term "true or real degree Brix" is sometimes applied to the percentage
-of total solids, when this number is determined by actually
^drying the material in an oven.
86. Determination of the Density by Dilu- tion and Spindling. Apparent Degree Brix.— Dissolve 250 grams of the massecuite or molasses in water and dilute to 500 cc. Transfer a portion of the solution to a cylinder and determine its degree Brix. Calculate the degree Brix of the product used by the following formula :
Apparent degree Brix = — — ,
in which B is the degree Brix (corrected) of the solution, S/>. Gr. the specific gravity corresponding to the degree Brix of the solution before correction, V the volume of the solution, and ^ the weight of massecuite used.
The above formula reduces to the following if the weight and volume specified have been used :
Apparent degree Brix = 2 X Sp. Gr. X B.
The following is a very convenient modification of th^ above method :
104 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
Dissolve a definite weight of massecuite in an equal weight of water, mix the solution thoroughly, and spindle.
The degree Brix of the massecuite is two times the degree Brix of the solution. {See also 88, Weisberg's method.)
87. Determination of the Total Solids or Moisture by Drying.— The method of Carr and San- born, and the vacuum method given in 77, are recom- mended. In the latter case use i gram of the massecuite, and in both methods, after weighing the material, dissolve it in a small quantity of distilled water, in order to dis- tribute it evenly. In the Carr-Sanborn method, dilute the sample to content of about 20 to 30 per cent dry matter, using a weighed portion of water. Add such quantity of the diluted material to the pumice-stone, in the tared dish, as will yield approximately i gram dry matter.
88. Approximate Determination of the Total Solids and Coefficient of Purity of Massecuite, etc., by Dilution and Spindling. Weisberg's Method.' — This is the ordinary method by dilution and spindling, but conducted under certain definite conditions, under which a table of coefficients, deduced by Weisberg from a very large number of experiments, is used.
Weigh three times the normal weight, or any convenient multiple of the normal weight, of the massecuite and dissolve it in water; transfer the solution to a 300-cc. flask, or to a flask corresponding to the multiple of the normal weight of massecuite used, and dilute to the graduation. Mix the solu- tion thoroughly and determine its degree Brix, using a spin- dle graduated to twentieths of a degree. Transfer 50 cc. of the solution, corresponding to the half-normal weight of the massecuite, to a flask, clarify with subacetate of lead, dilute to 100 cc, mix and filter. Polarize the filtrate, and multiply the polariscopic reading by 2 to compensate for the dilu- tion. This gives the percentage of sucrose in the masse- cuite. In materials containing notable quantities of raffin- ose, etc., use the method of Creydt (89) to ascertain the per cent of sucrose in the massecuite. The methods of calculation are most conveniently explained by an example.
* Bui. Asfoc. ChimUtes de France, 14, 978.
ANALYSIS OF THE MASSBCVlTES AND MOLASSES. 105
Example and Formula for Calculations, Weight of massecuite {2\ times the normal) = 65.12 gram
Volume of the solution = 250 cc.
Degree Brix of the solution = B = 22
Specific gravity corresponding to the degree
Brix {see table page 275) = Z> = i. 09231
Polariscopic reading X 2 = ^ = 55-
Constant (normal weight -r- 100) = .26048
, ^ R X 0.26048 . , .., . ,
(i) — ^— = per cent sucrose in the diluted solu- tion, S\
(2) — X 100 = apparent coefficient of purity (106) of the solution and of the massecuite.
WEISBERG'S TABLE OF COEFFICIENTS.
|
Coefficient of Purity. |
Coefficients. |
• Apparent Coefficient of Purity. |
Coefficients. |
|
57 |
1.054 |
78 |
1.021 |
|
57.5 |
1.(^2 |
79 |
1.020 |
|
58 |
1.050 |
80 |
1.019 |
|
58.5 |
1.048 |
81 |
1.018 |
|
59 |
1.046 |
82 |
1.017 |
|
60 |
1.044 |
83 |
1.016 |
|
61 |
1.042 |
84 |
1.015 |
|
62 |
1.040 |
85 |
1.014 |
|
63 |
1.038 |
86 |
1.013 |
|
64 |
1.036 |
87 |
1.012 |
|
65 |
1.034 |
88 |
1.011 |
|
66 |
1.033 |
89 |
1.010 |
|
67 |
1.032 |
90 |
1.009 |
|
68 |
1.031 |
• 91 |
1.008 |
|
69 |
1.030 |
92 |
1.007 |
|
70 |
1.029 |
93 |
1.006 |
|
71 |
1.028 |
94 |
1.005 |
|
72 |
1.027 |
95 |
1.004 |
|
73 |
1.026 |
96 |
1.003 |
|
74 |
1.025 |
97 |
1.002 |
|
76 |
1.024 |
98 |
1.002 |
|
78 |
1.023 |
99 |
1.001 |
|
77 |
1.022 |
100 |
1.000 |
106 HANDBOOK FOR SUGAR-HOUSE CHEMISTS.
The letters have the values indicated in the statement of the example and in equation (i).
(3) Multiply the apparent coefficient of purity by the co- efficient corresponding to it in Weisberg's table to obtain the true coefficient of purity of the massecuite.
(4) The true per cent total solids of the massecuite is deduced by dividing its percentage sucrose by the true co- efficient of purity and multiplying by 100.
Substituting the values of -^and D in formula (i) we have
R X 0.26048 55 X 0.26048
= = 13.12 = S\
D I. 09231
and substituting the values of ^ and 5 in formula (2) we have
S 13.12
B^ ^°° ^~^^ X 100 = 59.64,
apparent purity of the massecuite; and from (3),
59.64 X 1.045 = 62.32,
the approximately true purity of the massecuite.
- - ' R
From (4), X 100 = 88.25, the approximately true per
cent of total solids in the massecuite.
In checking this method by actual drying of the above massecuite, Weisberg obtained a true purity of 62.02. Thiy sample was a very severe test of the method owing to the low purity of the massecuite,
Weisberg constructed his table from experimental data, obtained in the examination of massecuites produced with- out " boiling in " molasses, as is now practised to a consider- able exteiit. With massecuite obtained by "boiling in" molasses on first-sugar, it is possible that the method may not give as satisfactory results as indicated in the example.
89. Detenninatidii of Sucrose and RafRnose. Creyclt'sForillulse. — This is the official German method ;' it is that of Clerget, as published by the German Govern- ment, except that the acidulation of the solution for direct polarization is recommended. This method is not applicable in the presence of optically active bodies Other than sucrose and raffinose. Percentages of raffinose less than 0.33 cannot be determined with certainty by the inversion methods.
\ Zeit, Rubenzucker -Industrie, 38, 867.
AKALYSIS OF THE MASSECUITES AKD MOLASSES. 107
Dissolve the normal weight of the material in water, clarify as usual, and dilute-to loo cc. Filter, and polarize the filtrate at 20° C. Record the polarization as the " direct reading." It is recommended that this solution be slightly acidulated with acetic acid before diluting to 100 cc.
Dissolve 13.024 grams of the substance in 75 cc. of water, in a loo-cc. flask, and add 5 cc. hydrochloric acid containing 38.8 per cent of the acid, mix the contents of the flask by a circular motion, and place it on a water-bath heated to 70' C. The temperature of the solution in the flask should reach 67° to 70" C. in two and one half to three minutes. Maintain a temperature of as nearly 69" C. as possible for seven to seven and one half minutes, making the total time of heating ten minutes. Remove the flask and cool the contents rapidly to 20'^ C, and dilute the solution to 100 cc. If necessary treat the solution with i gram of dry bone-black (180) to decolorize it. Polarize in a tube provided with a lateral branch for the insertion of a thermometer. A tube, provided with a jacket, through which a current of water of 20° C. circulates, should be used. The invert reading should be made at 20° C, and be multiplied by 2. If a preliminary cal- culation, using the formula, per cent sucrose =0.7538 X sum of the direct and invert readings, give a percentage which is more than i per cent higher than the direct read- ing, raffinose is probably present, and the following formulae by Creydt should be used in making the calculations :
P= the direct reading, z.^?., the polarization before in- version ; /= the invert reading , multiplied by 2. S = the percentage of sucrose ; R = the percentage of anhydrous raffinose.
G.5i88iP-/ P-S
•J "—^ ;; 1 J^ — '::. — •
0.845 1.85
It is very important in this process that the time and temperature conditions be strictly complied with. The amount of material used should be varied, according to the nature of the substance, that the invert solution may have a concentration of approximately 13.7 grams rn loa cc.r i.e., the invert-sugar produced in the inversion of 13.024
108 HANDBOOK POR SUGAR-HOUSE CHEMISTS.
grams of sucrose. The value of the constants varies con- siderably with the concentration (j-^f 267).
90. DetermiDation of Sucrose and Raffinose. Lindet's Inversion Method as Modified by Courtonne. — Courtonne^ has slightly modified the method of Lindet'^ in order to facilitate the manipulations.
Dissolve the normal weight of the material in water and dilute to IOC cc. Transfer 50 cc. of this solution to a 50-cc. flask and add sufficient dilute subacetate of lead solution (20.7); acidulate with acetic acid; mix, filter, and polarize the filtrate. Increase the polariscopic reading one tenth and record as the direct reading (A).
Transfer 20 cc. of the original solution of the material to a 50-cc. flask, and add to it 5 grams of zinc-dust. The dust must be weighed. Heat the flask and contents by immer- sion in boiling water or in the steam from a water-bath. Add 10 cc. of dilute hydrochloric acid, in portions of about 2 cc. at a time, being careful that none of the liquid is lost through a too rapid addition of the acid. The portions of acid may be added as frequently as convenient. The dilute acid is prepared by adding an equal volume of distilled water to pure hydrochloric acid of 1.2 specific gravity.
In the original method of Lindet, it is specified to heat the contents of the flask on the boiling-water bath about 20 minutes. In the modified method, it is only necessary to heat a few minutes after the last addition of acid.
The quantity of acid is so gauged that a portion of the zinc is left undecomposed and occupies a volume of .5 cc, for which a correction must be made in the calculations.
After the inversion is