Volume I Part 21
_Hist., &c._ The great importance of being able accurately to determine the strength of spirits in the United Kingdom, on account of the high duties levied on them, has induced the Government authorities, at various times, to investigate the subject. In 1790, the matter was referred to Sir C. Blagden, then Secretary to the Royal Society, who inst.i.tuted an extensive series of experiments to determine the real specific gravities of different mixtures of alcohol and water. The results of his labours and researches were put forward, with 'Gilpin's Tables,' in 1794, but no practical measures appear to have been taken in consequence. In 1832 a committee of the Royal Society, at the request of the Lords of the Treasury, examined into the accuracy of the Tables, and the construction and application of the instrument (SYKE'S HYDROMETER) now used by the Revenue officers, on which they reported favorably, and declared that they were sufficiently perfect for all practical and scientific purposes. The errors introduced into calculations of the strength of spirits by these tables were found to be quite unimportant in practice, and did not, in any one instance, amount to unity in the fourth place of decimals. This method adapts the specific gravity as the test of the strength of spirits, and is founded on the fact that alcohol is considerably lighter than water, and that (with proper corrections for condensation and temperature) the sp.
gr. regularly increases, or decreases, according to the relative proportions in which the two are mixed.
Several other methods of alcoholometry have been proposed, founded upon--the variations in temperature of the vapour of alcohol of different strengths--the heat involved by its admixture with water--its dilatation by heat--the tension of its vapour--the insolubility of carbonate of potash in alcohol--its volatility, boiling point, &c. &c., the more important and useful of which are noticed further on. The method adopted by the Boards of Inland Revenue and Customs is, however, the one which is almost exclusively employed in trade and commerce in Great Britain, not only on account of its simplicity and correctness, but for the purpose of the results exactly coinciding with the results obtained by the Revenue officers.
METHODS OF ALCOHOLOMETRY.
1. _Methods based_ on the _specific gravity_, or _per-centage strength_, by VOLUME:--
[Ill.u.s.tration]
_a._ With SYKES' HYDROMETER. _Revenue system._ The _engraving_ below represents Sykes' hydrometer, as made by Mr Bate, under the directions of the Commissioners of Inland Revenue and Customs. It consists of a spherical ball or float, with an upper and lower stem, and is made of bra.s.s, which (in the more expensive instruments) is usually coated with gold, to prevent corrosion from damp, and the acidity so generally present in spirituous liquors. The upper stem (A) is about four inches long, and is divided into ten parts, each of which contains five subdivisions. There are nine movable weights of the form _b_, of different sizes, numbered respectively 10, 20, 30, &c., to 90, each of which represents so many of the princ.i.p.al divisions of the stem, as its number indicates. In use, one of these weights is slipped on to the lower stems; and thus, by means of them, the instrument acquires a range of above 500 divisions, or degrees, extending from the Revenue 'standard alcohol' (sp. gr. 825) to water. It is so formed as to give the sp. gr. with almost perfect accuracy, at 62 Fahr. When loaded with the weight 60 it sinks in proof spirit to the line marked (P) on the narrow edge of the stem at 51 Fahr.; and, by further placing the square weight or cap (also supplied with the instr.) on the top of the upper stem, it floats exactly at the same point in distilled water. This weight or cap is found to weigh 4366 grs., which is practically 1-12th of the total observed weight of the instrument, and its poise 60, and hence shows the difference between the gravity of proof spirit and water, as explained hereafter. The whole is fitted up in a neat mahogany case, accompanied with a thermometer, and a book of tables containing corrections for temperature, &c.--_Process._ A gla.s.s tube of the form of fig. _B_ is filled to about the mark (_a_) with the sample for examination; the thermometer is then placed in the liquor, and stirred about for two or three minutes (observing not to breathe upon the gla.s.s, nor hold it in the hand), and the temperature noted. The hydrometer is next immersed in a similar manner, and gently pressed down in the liquor to the 0 on the stem with the finger; it having been previously loaded with any one of the nine weights that will cause it to float with the surface of the spirit at some point on the graduated part of the scale.
The indication at the point cut by the surface of the liquor, as seen from below, added to the number of the weight with which the float is loaded, gives a number which must be sought in the hook of Tables, which is always sold with the instrument. In this book, at the page headed "Temperature as observed by the Thermometer," and against the part of the column appropriated to the given indication (weight), will be found the strength per cent., expressed in degrees over or under proof, by VOLUME, in whole numbers or decimal parts. In reading off the indication, to ensure accuracy, it is necessary to allow for the convexity of the liquor at the part where it immediately rests against the stem.
_Obs._ In an instrument requiring so much care and skill in its manufacture the purchaser should be careful to procure a perfect one. A very slight blow, friction from continual wiping with a rough cloth, and other apparently trivial causes, tend to injure so delicate an instrument.
The shape of the weights occasionally vary; some being intended to be attached to the hydrometer at the bottom of the spindle, and others to rest on its top. The first plan is, perhaps, the best, as it tends to make the instrument float with greater steadiness in the liquor; but, at the same time, it renders its adjustment by the maker a matter of greater difficulty.
In employing this instrument, the Revenue officers are instructed to take the nearest degree above the surface of the mercury, when it stands between any two degrees of the thermometer; and the division on the scale of the hydrometer next below the surface of the liquid, when it cuts the stem between any two lines; thus giving the difference in favour of the trader in both cases.
By means of the _Table_ at page 64 the hydrometer indication, or the degrees over or under proof, of the Revenue system, may be converted into 'real specific gravities,' by mere inspection; and the corresponding 'per-centage richness' in alcohol of any sample may be found, either by WEIGHT or VOLUME.
The specific gravities in this table are such as, on being referred to Gilpin's Tables, will give the expressions of proof strength answering to the whole indications of the Revenue hydrometer. Intermediate values at fifths of indications may be had by taking proportional differences between the nearest tabular numbers. Thus, to find the specific gravity that should stand opposite to Indication 706, we first obtain the difference between the densities standing in a line with Indications 70 and 71 respectively, and then say, as 1 : 06 :: 00192. 00115, and 94135 + 00115 = 94250, the specific gravity required.
_b._ With GLa.s.s ALCOHOLOMETERS. That of Tralles, and most others of a like description (as made in England), gave the per-centage strength, by VOLUME, with tolerable accuracy, at the standard temperature of 60 Fahr.
Gay-Lussac's ALCOoMETRE, which closely resembles that of Tralles, is adjusted for the temperature of 59 Fahr. (15 Cent.). All of these, to give at once accurate results, must, of course, be employed at the 'normal temperature' of the instrument. As, however, in practice, the experiment cannot be conveniently performed at any 'fixed' temperature but only at that of the atmosphere, it is obvious that certain corrections are constantly required in order to obtain results of any value. Perfect accuracy requires that table for every variation of the thermometer, founded on actual experiments, should accompany each instrument; as, without them, tedious and difficult calculations are necessary, which, in the hurry of the cellar and laboratory, or by persons inexpert at figures, are not easily performed. A series of such Tables were prepared by Gay-Lussac, and, with his instrument, are those which are almost exclusively used in France. For rough purposes, in the absence of Tables or nicer calculations, it may be useful to know that, for commercial spirits, at ordinary temperatures, a variation of--
By VOLUME, 5 Fahr. is equal } 100% of Alcohol; } 1794% of Proof to (about) } or (about) } spirit.
1 " " 020% " 0359% "
5 Cent. " 180% " 3229% "
1 " " 036% " 0646% "
By WEIGHT, 5 Fahr. is equal } 080% of Alcohol; } 162% "
to (about) } or (about) } 1 " " 16% " 32% "
5 Cent. " 143% " 29% "
1 " " 28% " 58% "
TABLE I.--_Showing the Densities and Values of Spirits at 60 Fahr., corresponding to every Indication of Sykes' Hydrometer._
+-----------+---------+----------+--------------------+
Per Cents. of
Sykes'
Absolute Alcohol.
Hydrometer
Strength
Specific +----------+---------+
Indication.
per cent.
Gravity.
By
By
Measure.
Weight.
+-----------+---------+----------+----------+---------+
O.P.
0
670
81520
9528
9278
1
661
81715
9478
9208
2
653
81889
9431
9142
3
645
82061
9384
9078
4
636
82251
9333
9007
5
627
82441
9280
8936
6
618
82622
9229
8867
7
609
82800
9177
8799
8
600
82978
9125
8730
9
591
83151
9074
8663
10
582
83323
9023
8596
11
573
83494
8972
8530
12
564
83661
8921
8465
13
555
83827
8870
8400
14
546
83993
8817
8333
15
537
84153
8767
8270
16
527
84331
8710
8199
17
517
84509
8651
8126
18
507
84680
8595
8058
19
497
84851
8539
7989
20
487
85022
8481
7919
21
476
85205
8419
7844
22
466
85372
8361
7774
23
456
85537
8304
7707
24
446
85700
8247
7639
25
435
85878
8185
7566
26
424
86055
8121
7492
27
413
86229
8059
7419
28
402
86402
7997
7347
29
391
86574
7934
7275
30
380
86745
7871
7203
31
369
86915
7808
7132
32
357
87099
7740
7054
33
345
87282
7671
6977
34
334
87450
7608
6906
35
322
87627
7541
6832
36
310
87809
7472
6755
37
298
87988
7403
6679
38
285
88179
7329
6598
39
273
88355
7260
6523
40
260
88544
7186
6443
41
248
88716
7117
6368
42
235
88901
7043
6289
43
222
89086
6969
6210
44
209
89268
6895
6132
45
196
89451
6821
6053
46
183
89629
6747
5976
47
169
89822
6667
5892
48
156
89997
6593
5815
49
142
90182
6514
5734
50
128
90367
6434
5652
51
114
90551
6354
5570
52
100
90732
6274
5489
53
86
90913
6194
5409
54
71
91107
6109
5323
55
56
91299
6024
5238
56
42
91479
5943
5157
57
27
91666
5858
5073
58
13
91839
5778
4994
U.P.
59
03
92037
5686
4904
60
19
92228
5596
4817
61
34
92408
5510
4733
62
50
92597
5419
4646
63
67
92798
5322
4553
64
83
92984
5230
4465
65
100
93176
5136
4376
66
117
93367
5039
4284
67
135
93586
4934
4186
68
153
93758
4831
4090
69
171
93949
4729
3996
70
189
94135
4629
3904
71
208
94327
4520
3804
72
227
94518
4409
3703
73
247
94709
4296
3601
74
267
94899
4182
3498
75
288
95092
4063
3392
76
310
95288
3940
3282
77
332
95484
3810
3168
78
356
95677
3676
3050
79
381
95877
3532
2924
80
406
96068
3390
2801
81
433
96259
3241
2673
82
461
96457
3077
2532
83
491
96651
2908
2388
84
522
96846
2731
2238
85
555
97049
2539
2077
86
590
97254
2341
1911
87
625
97458
2139
1742
88
660
97660
1941
1578
89
694
97857
1746
1416
90
728
98057
1551
1256
91
761
98261
1358
1097
92
792
98452
1185
956
93
823
98657
1004
808
94
852
98866
828
665
95
880
99047
683
548
96
907
99251
525
420
97
933
99448
380
303
98
959
99658
231
184
99
982
99851
997
793
100
...
100000
...
...
+-----------+---------+----------+----------+---------+
This Table {above} has been copied, by permission, from Loftus's 'Inland Revenue Officer's Manual,' and its correctness verified by W. H. Johnston, Esq., Surveying General Examiner.
TABLE II.--_Table for finding the Specific Gravity of any Spirit at 60 Fahr., when the Specific Gravity at any other Temperature is given._
Water taken as 1000.
+-----------------+----------+-----------------+----------+
Correction
Correction
Specific gravity.
for each
Specific gravity.
for each
degree.
degree.
+-----------------+----------+-----------------+----------+
810 to 820
475
910 to 920
434
+-----------------+----------+-----------------+----------+
820 " 830
473
920 " 930
424
+-----------------+----------+-----------------+----------+
830 " 840
472
930 " 940
406
+-----------------+----------+-----------------+----------+
840 " 850
471
940 " 950
381
+-----------------+----------+-----------------+----------+
850 " 860
471
950 " 960
340
+-----------------+----------+-----------------+----------+
860 " 870
466
960 " 970
269
+-----------------+----------+-----------------+----------+
870 " 880
460
970 " 980
165
+-----------------+----------+-----------------+----------+
880 " 890
456
980 " 990
090
+-----------------+----------+-----------------+----------+
890 " 900
450
990 " 1000
084
+-----------------+----------+-----------------+----------+
900 " 910
442
+-----------------+----------+-----------------+----------+
Thus, by making the proper ADDITION to the apparent strength per cent., when the observed temperature is BELOW the normal temperature of the instrument, or a corresponding SUBTRACTION, when it is ABOVE it, the strength of the sample may be determined sufficiently near for all practical purposes.
The following Table, taken from Loftus's 'Inland Revenue Officer's Manual,' will be found of great value in making these corrections, and has the merit of being easily applied.
An example will show how this Table is to be used.
_Example._--If a quant.i.ty of spirit is of the sp. gr. 894 at 73, what will be its sp. gr. at 60?
Here the sp. gr. being between 890 and 900, we must add 450 for each degree of temperature between 73 and 60. The sp. gr. at 60 would, therefore, be 894 + (450 13) = 89985. When the temperature is below 60, the correction for each degree must be subtracted. When, however, very accurate results are desired, and the necessary Tables are not accessible, the sample for trial must be brought to the normal temperature of the instrument, in the manner explained under HYDROMETRY.
_c._ From the SPECIFIC GRAVITY. The temperature having been taken by a thermometer, and the specific gravity ascertained by any of the usual methods, but preferably by means of an accurate gla.s.s hydrometer, it merely becomes necessary to refer to Table I, where, against the number expressing the specific gravity, the alcoholic content per cent., by volume, of the sample examined, will be found for 60 Fahr., subject to the corrections just referred to, when the temperature is either above or below this point.
If the precise specific gravity sought cannot be found in the _Table_, the difference between it and the next greater specific gravity must be taken for the numerator of a fraction, having for its denominator the difference between the greater and the next less specific gravity in the table. This fraction, added to the per-centage of alcohol in the fourth column of the table, opposite the greater sp. gr., will give the true per-centage sought. Thus, the sp. gr. 96051 is not in the table, and the next greater number is 96068; the former must, therefore, be deducted from the latter, and the difference (17) put as the numerator of the fraction, having for its denominator 191, the difference between 96068 and 95877. The fraction (17/191) 089, so found, added to the per-centage strength opposite 96068 in the third column, gives 33989 as the true per-centage of alcohol in the given sample.
The per-centage by volume may be converted into per-centage by weight, by multiplying the former by 793811, the sp. gr. of absolute alcohol, and dividing the product by the sp. gr. of the sample. The quotient is the number of pounds of alcohol in 100 pounds of the given spirit.
Thus:--Suppose 1000 grains by measure of alcohol to weigh 95092 grains, and to contain (see Table I) 4063 per cent. by volume of absolute alcohol, what per cent. by weight does the sample contain?
793811 4063 = 3225254093, and this product divided by 95092 = 33917, the true per-centage by weight of absolute alcohol in the sample.
2. Method based on the specific gravity, or per-centage strength by WEIGHT:--
The specific gravity is ascertained and the Table used in precisely the same manner as in the "method by volume," already described.
The per-centage by weight may be converted into per-centage by volume, by multiplying the former by the sp. gr. of the sample, and dividing the product by the sp. gr. of absolute alcohol. This is merely the reverse of the operation described above.
_Obs._ The preceding methods of alcoholometry, as well as all others depending on the sp. gr. refer to UNSWEETENED SPIRITS only; and are inapplicable to those holding sugar in solution, or any other organic matter capable of altering the sp. gr. For sweetened spirits, fermented worts, wine, beer, &c., one or other of the following processes must be adopted:--
3. Other methods, adapted to either SWEETENED or UNSWEETENED SPIRITS, Tinctures, Fermented Liquors, &c.--
_a._ By DISTILLATION as originally proposed by M. Gay-Lussac. 300 parts of the liquor under examination (measured in a graduated gla.s.s tube) are placed in a retort or small still, and a quant.i.ty exactly equal to one third (_i.e._, 100 parts), carefully drawn over; a graduated gla.s.s tube[13] being used as a receiver, and the operation stopped as soon as the distillate reaches the hundredth degree. The 'alcoholic strength' of the distilled liquor is then ascertained by any of the usual methods, and the result divided by three, when the per-centage of alcohol in the original liquor is at once obtained. If, from want of attention, more than 100 parts should be distilled over, the number which expresses the relation of the volume of the distilled product to the original bulk of the liquor tested, must be employed as the divisor. Thus, if 106 parts of liquor have distilled over (instead of 100), containing 33% of alcohol, the 300 must be divided by 106, which gives 283, and the 33% by this 283, which gives 1166%, the true proportion of alcohol in the original liquor. The strength at 'proof' may be calculated from this in the usual way.
[Footnote 13: Mulder, in his 'Chemistry of Wine' recommends this receiver to be shaped like a bottle, with its neck, or tubular part, bent at right angles above the line of its scale; and that it should be set in the centre of a gla.s.s jar kept filled with very cold water.]
To ensure accurate results, the acidity (if any) of the liquor must be neutralised with carbonate of sodium, prior to distillation. It is also advisable to add 8% or 10% of common salt to the liquor in the retort or still; this, by raising the boiling point, causes the whole of the spirit to pa.s.s over into the receiver before the distillate has reached the required measure. This applies more particularly to weak liquors. With those of greater strength (as the stronger wines), it is better to distil over 150 parts, and divide the result by 2 instead of 3. To liquors stronger than 25% by volume of alcohol, or above 52% to 54% under proof, add about an equal volume of water to the liquor in the still, and draw over a quant.i.ty equal to that of the sample tested; when the alcoholic strength of the distillate gives, without calculation, the true strength sought. To liquors stronger than 48% to 50% (14 to 12 u. p.), add thrice their bulk of water, and do not stop the process until the volume of the distillate is double that of the sample tested, when the per-centage obtained must also be doubled. In each case a proportionate quant.i.ty of salt is employed.
REVENUE METHOD. The following is the method adopted in the Inland Revenue and Customs Laboratories for the estimation of the per-centage of alcohol in wines, liqueurs, &c. A measure flask is filled up to a mark on its neck, with the wine, which is then carefully transferred to a distilling flask or retort, the traces of wine remaining in the former vessel being rinsed out with small quant.i.ties of distilled water, and the rinsings added to the wine in the latter vessel. About two thirds of the contents of the retort are then distilled over into the clean measure flask, and made up to the original bulk with distilled water, at the same temperature as the sample was previous to distillation. The strength is then taken by Sykes' hydrometer, and this (if u. p.) deducted from 100, gives the per-centage of proof spirit in the wine. Thus:--
Strength of distillate = 746 u. p. = 254 per cent. proof spirit.
_b._ From the TEMPERATURE of the VAPOUR, as originally proposed by Groning. The bulb of a thermometer is thrust through a cork into the head of the still, or other vessel employed, and the temperature of the vapour in which it is immersed being noted, is sought in the following table:--
TABLE III.--_Showing the Alcoholic Content, by_ VOLUME_, of Boiling Spirits, and of their Vapour, from the Temperature of the latter, as observed by a Thermometer._ By GRoNING.
+---------------+-------------------+-------------------+----------------+-------------------+------------------+
Temperature of
Alcoholic content
Alcoholic content
Temperature of
Alcoholic content
Alcoholic content
the Vapour.
of the Distillate
of the Boiling
the Vapour.
of the Distillate
of the Boiling
Fahr.
per cent.
Liquid per cent.
Fahr.
per cent.
Liquid per cent.
+---------------+-------------------+-------------------+----------------+-------------------+------------------+
1700
93
92
1898
71
20
1718
92
90
1920
68
18
1720
91
85
1940
66
15
1728
90-1/2
80
1964
61
12
1740
90
75
1986
55
10
1746
89
70
2010
50
7
1760
87
65
2030
42
5
1783
85
50
2054
36
3
1808
82
40
2077
28
2
1830
80
35
2100
13
1
1850
78
30
2120
0
0
1874
76
25
+---------------+-------------------+-------------------+----------------+-------------------+------------------+
This method is admirably adapted to the purposes of the distiller and rectifier, as it furnishes a ready means of approximately determining the strength of the spirit pa.s.sing over, at every part of the process of distillation, as well as that of the wash left in the still.
_c._ From the BOILING POINT, as originally proposed by M. l'Abbe Brossard-Vidal. This method is founded on the fact, that the boiling points of mixtures of alcohol and water, unlike water alone, are scarcely disturbed by the addition of saline, saccharine, or extractive matter within certain limits. It hence offers a ready means of determining the proportion of alcohol present in spirits, wines, fermented liquors, &c., with sufficient accuracy for all ordinary purposes. In applying it, a thermometer, with a large bulb and a narrow bore, and a movable scale graduated from 180 to 212 Fahr., is usually employed. Before using it as an alcoholometer, it is set, with its bulb immersed, in a small metallic boiler (bra.s.s or copper) containing distilled water, which is then raised to the boiling-point, and the 212 of the scale accurately adjusted on a level with the surface of the mercury, should it vary from that point.
This is necessary on account of variations of atmospheric pressure causing corresponding variations of the boiling-points of liquids. It is then ready for several hours' operations, and, generally, for an entire business day, without further adjustment. The little boiler is next filled with the liquor to be examined, and the lamp again lighted. The temperature as shown by the scale of the instrument at the commencement of full ebullition being ascertained, may be sought in one of the following _Tables_, against which the alcoholic content of the liquor will be found (nearly).
TABLE IV.--_Exhibiting the_ BOILING POINTS _of Mixtures of Alcohol and Water of the given strengths._ By GRoNING.
