[Reprinted from The Journal of Analytical and Applied Chemistry, Vol. VI, No. 3. March, 1892.] [Contributions from the Chemical Laboratory of the University of Cincinnati.] XIvIII. EXAMINATION OF THE LEOYD METHOD FOR THE ASSAY OF ALKALOIDS.* In February, 1891, Prof. J. U. Floyd read before the Cincinnati Section of the American Chemical Society a paper on the analysis of alkaloids, which was presented later before the American Phar- maceutical Association at its annual meeting, published in pam- phlet form,f and printed more or less fully in several home and foreign journals. The method proposed, on account of its claims to rapidity and reasonable accuracy, evoked much criticism both friendly and adverse. The possession of a reliable method plac- ing in the hands of the chemist the means of analyzing easily and quickly the majority of the members of the group of the alkaloids, is of such prime importance, that we undertook the following study of the process with the view of ascertaining as definitely as possible the limits of accuracy in the case of the leading alkaloids, under varying conditions. The method proposed by Prof. Lloyd, and employed in all our experiments, is, briefly stated, as follows : 5 cc. of an alkaloidal solution are poured into a mortar, and a sufficient amount of a mixture of equal parts of ferric hydrate and sodium bicarbonate added to form a stiff paste. This magma is then triturated with 20 cc. of chloroform. Should so much alcohol be abstracted that the magma becomes pulverulent, a small amount of water or of a solution of glucose is added and triturated until a stiff paste is again formed, from which the chloform separates clearly with no trace of turbidity. The chloroform extract is decanted into a small porcelain dish and evaporated. The magma is then washed three times by trituration with chloroform. 10 cc. of chloroform are used each time and the wash- ings are decanted as before and evaporated in the same dish. The residue in the dish is covered with a 2 per cent, solution of sulfuric acid, digested for a few minutes on the water bath and filtered. The same diluted acid is used to wash the residue on the filter paper. By Herbert T. Nichols and Thos. H. Norton. *Read before the Cincinnati Section of the American Chemical Society, Jan. 28, 1892. fCopies of a revised edition of this pamphlet are gladly furnished to those interested in the matter, by application to the author, Prof. J. U. Idoyd, Court St., Cincinnati. 2 THE ASSAY OF ALKALOIDS. The clear solution of the sulfate of the alkaloid, thus obtained, is rendered slightly alkaline by ammonia, and rotated in a separa- ting funnel with io cc. of chloroform for about two minutes. This operation is repeated four times, the chloroform being drawn off each time. Tfce solution of the alkaloid in chloroform is then evaporated to dryness in a small platinum dish or on a watch-glass over the water-bath. While still on the water-bath the residue is stirred with the sharp point of a knife, to ensure the total removal of the chloroform. After cooling in the desiccator, the dish or watch-glass is weighed with its residue of pure alkaloid. The chief advantage and essential peculiarity claimed for the process just described is the total avoidance of any emulsion in the chloroform extract from the magma obtained by the treatment with ferric hydrate and sodium bicarbonate at the outset. The remaining steps in the process are practically identical with those hitherto used in similar analytical determinations. In the study of the method in question, we have extended our experiments over the following subjects, viz: 1. Purity of the chloroform and other reagents used. 2. Insolubility of ammonium sulfate in chloroform. 3. Completeness of the extraction of the alkaloids from the magma. 4. Completeness of the extraction of alkaloids from alkaline so- lution by rotary agitation with chloroform. 5. Complete assays of pure alkaloids and of their fluid extracts. 6. Increase in weight of certain alkaloids after solution in chlo- roform and subsequent evaporation. 7. Substitution of aluminum and chromium hydrates for ferric hydrate. 8. Limits of precaution to be observed in weighing. 9. Amounts of moisture present in pure commercial alkaloids. A. Experimentae Part by H. T. Nichols. Chloroform.—The purity of the chloroform used should be tested by evaporating about io cc. on the water-bath. If pure enough for the purpose in view, it should leave absolutely no residue. I. PURITY OF THE REAGENTS USED. THE ASSAY OF ALKALOIDS. 3 This precautionary test is especially necessary in the case of chlo- roform which is not of recent manufacture. Where many deter- minations of this nature are made, a large proportion of the chloro- form can be saved by evaporation beneath an inverted funnel lead- ing to a condenser through which a current of air is drawn. If rubber connections are used, redistillation is of course, necessary. Ferric hydrate and sodium bicarbo?iate.—These should be pure, or at least free from all substances capable of being dissolved by chloroform. The reagents used in the following analyses were triturated with chloroform, and the decanted liquor left no residue. II. INSOLUBILITY OF AMMONIUM SULFATE IN CHLOROFORM. As no statements with regard to the solvent powers of chloro- form on ammonium sulfate are on record, it seemed desirable to decide the question by actual experiment. Several grams of ammonium sulfate were triturated with 20 cc. of chloroform, and filtered. The filtrate evaporated to dryness on the water-bath left no weighable residue. This experiment was repeated five times with the same result. 10 cc. of a 2 per cent, solution of sulfuric acid were made alkaline by ammonia and agitated with 20 cc. of chloroform, in a separat- ing funnel. The chloroform was then drawn off and evaporated to dryness on the water-bath. This was repeated five times, like- wise without obtaining any residue. Insolubility of ammonium acetate in chloroform.—In this connec- tion a few experiments were made to ascertain whether ammonium acetate is soluble in chloroform. 10 cc. of acetic acid were made alkaline by ammonia and agitated in a separating funnel, with 20 cc. of chloroform. The chloroform was then drawn off and evaporated to dryness over the water-bath on a watch-glass. This operation was repeated five times, and gave in no instance a residue. The experiment was altered, by allowing a few drops of the solution of ammonium acetate to run through with the chloroform. This when evapo- rated to dryness yielded no residue. The operation was repeated five times with like results. The observation has already been made that ammonium acetate is decomposed at ioo° C. 4 THE ASSAY OF ALKALOIDS. This observation was confirmed by three experiments in which solutions of ammonium acetate were evaporated to dryness, and left no residue. III. COMPLETENESS OF THE EXTRACTION OF ALKALOIDS PROM THE MAGMA BY CHLOROFORM. Three series of tests were made to ascertain whether any appre- ciable amount of alkaloid remained in the magma after four extrac- tions with chloroform. Several magmas containing nux vomica were extracted in the usual manner. In no case was the intensely bitter taste peculiar to this alka- loid perceptible after the third washing with chloroform. A weighed amount of cinehonin, one part of which is soluble in over three hundred parts of chloroform, was treated as before. The residue obtained from the chloroform extract was almost equal to the amount used, showing that even in the case of this difficultly soluble alkaloid, scarcely any portion was retained by the magma. A further test of the delicacy of the process was shown by the following experiment: o.ooooi gm. of a mixture of brucin and strychnin was put through the assay and the residue on the watch-glass, though not visible, still yielded the bitter taste of the alkaloids. IV. COMPLETENESS OP the EXTRACTION OF ALKALOIDS FROM ALKALINE SOLUTIONS BY ROTARY AGITATION WITH chloroform. Experiments were made to find the number of rotations neces- sary for the complete extraction of an alkaloid when the solution of the alkaloid and ammonium sulfate is rotated with chloroform, i gm. of an alkaloid was dissolved in acidulated water, and the solution was made up to ioo cc. or a i per cent, solution, io cc. of this solution were used each time, containing in each case, therefore, o.i gm. Each portion was rendered slightly alkaline by the addition of ammonia, put into a separating funnel, and rotated with chloro- form. The dried residue obtained from the chloroform drawn off after each period of rotation was weighed separately. The rotations were made with the circling motion suggested in his article by Prof. Eloyd, and precautions were observed to pre- THE ASSAY OF ALKALOIDS. 5 vent the liquids from being shaken together. The average time of each period of rotation was about two minutes. Caffe'in.—Amount used, .1 gm. First Second Third Fourth Total Amount Rotation. Rotation. Rotation. Rotation. Recovered. A . . . . .0886 gm. .0048 gm. .0016 gm. .0000 gm. .0950 gm. B . . . . .0826 “ .0120 “ .0032 “ .0006 “ .0944 “ C . . . . .0708 “ .0222 “ .0048 “ .0006 “ .0984 “ D . . . . .0846 “ .OO92 “ .0020 “ .OOOO “ .0958 “ E . . . . .0832 “ .0098 “ .0024 “ .0006 “ .0960 “ Average, .0820 “ .0096 “ .0028 “t .0004 “ ,0959 “ Amount Recovered after each Successive Extraction. The greatest loss occurred in B, .0056 gm. The least loss was in C, .0016 gm. The average loss was .0140 gm., or 4 per cent. Brucin.—Amount used, .1 gm. A mount Recovered after each Successive Rotation. First Rotation. Second Rotation. Third Rotation. Total Amount Recovered. A . . . . .1170 gm. .0084 gm. .0008 gm. .1262 gm. B . . . .1104 “ .0040 “ .OOOO “ .1152 “ C ... . .1026 “ .0042 “ .0004 “ .1072 “ T> ... . .1038 “ .0036 “ .OOOO “ .1068 “ n ... . ,.1004 “ .0036 “ .OOOO “ .1040 “ Average, .1068 “ .0048 “ .0002 “ .1120 “ Average total gain, 12 per cent. The variations in total gain ranged from 4 to 26.2 per cent. It will be noticed that the total amount has increased over the original amount 12 per cent, on an average. As will be seen later in the section on the increase in weight of certain alkaloids after treatment with chloroform, this result was to be expected. Strychnin.—Amount used, .1 gm. A mount Recovered after each Successive Rotation. First Second Third Total Amount Rotation. Rotation. Rotation. Recovered. A . . . . .0810 gm. .0032 gm. .0000 gm. .0842 gm. B . . . . .0948 “ .0026 “ .0000 “ .0974 “ C . . . . .0898 “ .0016 “ .0000 “ .0914 “ D . . . . .0840 “ .0024 “ .0000 “ .0864 “ E . . . . .0906 “ .0028 “ .0002 “ .0936 “ Average .0880 “ .0025 “ .0000 “ .0906 “ Average total loss, 9.4 per cent. The variations in the total loss ranged from 2.6 to 15.8 per cent. 6 THE ASSAY OF ALKALOIDS Atropin.—Amount used, . i gm. First Rotation. Second Rotation. Third Rotation. Total Amount Recovered. A . . . .0926 gm. .0032 gm. .0000 gm. .0958 gm. B . . . . .0786 “ .0116 “ .0032 “ •0934 “ C . . . . .0928 “ .0024 “ .0000 “ .0952 “ D . . . . .0916 “ .0030 “ .0000 “ .0946 “ n ... . .0890 “ .0062 “ .00X0 “ .0952 “ Average, .0889 “ •0053 “ .0008 “ .0948 “ Amount Recovered after each Successive Rotation. Average total loss, 5.2 per cent. The variations in total loss ranged from 4.2 to 6.6 per cent. Aconitin.—Amount used, .1 gm. First Rotation. Second Rotation. Third Rotation. Total Amount Recovered. A . . . . .0960 gm. .0040 gm. .0006 gm. . 1006 gm. B . . . . .1066 “ .0042 “ .0002 “ .IIIO “ C . . . . .0998 “ .0036 “ .oop4 “ .1038 “ D . . . . .0926 “ .0036 “ .0002 “ .0964 “ E . . . . .0923 “ .0030 “ .0000 “ .0966 “ Average, .0977 “ .0037 “ .0003 “ .IO17 “ A mount Recovered after each Successive Rotation. Average total gain, 1.7 per cent. The variations in the total loss or gain ranged from 3.6 per cent, loss to 11 per cent. gain. The latter high figure, obtained in but instance, is to be considered as probably due to error. Quinin.—Amount used, . 1 gm. A mount Recovered after each Successive Rotation. First Second Third Total Amount Rotation. Rotation. Rotation. Recovered. A . . . . •0996 gm. .0052 gni. .0000 gm. .I048gm. B . . . . .0974 “ .0022 “ .0002 “ .0998 “ C . . . . .0990 “ .0026 “ .0002 “ .1018 “ D . . . . .0966 “ .0040 “ .0006 “ .1012 “ E . . . . •0954 “ .0020 “ .OOOO “ .0974 “ Average, .0976 “ .OO32 “ .0002 “ .1010 “ Average total gain, 1.2 per cent. The variations in the total loss or gain ranged from 2.6 per cent, loss to 4.8 per cent. gain. THE ASSAY OF ALKALOIDS. 7 Cinchonin.—Amount used, .1 gm. Amount Recovered after each Successive Rotation. First Rotation. Second Rotation. Third Rotation. Total Amount Recovered. A . . . . .0796 gm. .0156 gm. .0003 gm. •0955 gm. B . . . . .0862 “ .0070 “ .0004 “ .0936 “ C . . . . .0836 “ .0036 “ .0006 “ .0878 “ D . . . . .0762 “ .0112 “ .0006 “ .0880 “ E . . . .0802 “ .0114 “ .0004 “ .0920 “ Average .0812 “ .0102 “ .0005 “ .0919 “ Average total loss 8.1 per cent. The variations in the total loss ranged from 4.5 to 12.2 per cent. Cinconin is soluble, with difficulty in chloroform, and there- fore the amount of alkaloid obtained in the first rotation is less than in the case of other alkaloids. Cincliotiidin.—Amount used, .1 gm. First Rotation. Second Rotation. Third Rotation. Total Amount Recovered. A . . . . .0950 gm. .0036 gm. .0004 gm. •°99° gm- B . . . . .0924 “ .0026 “ .0004 “ •0954 “ C . . . . .0924 “ .0024 “ .0006 “ •0954 “ D . . . . .0926 “ • OOIO “ .0000 “ .0936 “ E . . . . .0890 “ .0012 “ .0000 “ .0902 “ Average .0923 “ .0022 “ .0003 “ .0948 “ A mount Recovered after each Successive Rotation. Average total loss, 5.2 per cent. The variations in the total loss ranged from 1 to 9.8 per cent. SUMMARY. The highest average gain was in the ease of brucin, amounting to 12 per cent.; and the lowest average gain was in the case of quinin, amounting to 1.2 per cent. The lowest average loss was in the case of caffein, amounting to 4 per cent., and the highest average loss was in the case of strychnin, amounting to 9.4 per cent. A fourth agitation yielded in but few instances a weighable amount. v. complete assays of fluid extracts, of THEIR pure ALKALOIDS AND OF MIXTURES OF BOTH. As the method in question is of special applicability to the analysis of fluid extracts, a study was made of the degree of uni- formity attainable in this field. 8 THE ASSAY OF AEKAEOIDS. Fluid extracts are alcoholic or hydro-alcoholic solutions of the soluble constituents of vegetable products, those of alkaloidal drugs containing the alkaloids in natural combination. To test the method most thoroughly, parallel determinations were made, first of a fluid extract, next of the same fluid extract with a known amount of its proper alkaloid added, and thirdly of a standard solution of the alkaloid dissolved in alcohol and water. In this manner the fluid extracts and alkaloids of nux vomica, belladonna, guarana, ipecac, aconite root, and henbane were as- sayed. The fluid extracts used were those obtained in commerce. Nux Vomica.—Five samples of the same fluid extract were analyzed : Percentages Obtained. I. II. III. IV. V. Average. 1.39% 1.50% 1.45% 1.46% 1.39% I-44/& .125 gm. of strychnin and .125 gm. of brucin were added to 50 cc. of the fluid extract, and 5 cc. were used in each assay. The results were as follows : Percentages Obtained. I. II. III. IV. V. Average. Theory. 2.03% 1.98% 1-95% 2.00 e mixture of pure brucin and strychnin was dis- solved in 50 cc. of alcohol and water, making a 1.5 per cent, solution. The results of the analysis were as follows : Percentages Obtained. I. II. III. IV. V. Average. Amount Used. 1.60% 1.51.48 % 1-47% i-52% 1-52% 1.50% Average increase, .02 per cent., or 1.3 per cent, of the alkaloid used. Belladonna.—The fluid extract of belladonna was assayed with the following results, all from the same sample : Percentages Obtained. I. II. III. IV. V. Average. 0.50% 0.55% 0.54% 0.52% 0.50% 0.52% The greatest variation from the average was .03 per cent., or 6 per cent, of the weight of the alkaloid. THE ASSAY OF ALKAEOIDS. 9 To ioo cc. of the same sample of fluid extract, i gm. of pure atropin was added, and the following results were obtained : Percentages Obtained. I. II. III. IV. V. Average. Theory. 1.49% 1.46% 1.46% 1-49% 1.52% 1.48% I.52# In analyzing the pure alkaloid, it was found that a solution of atropin in alcohol and water deteriorated so rapidly that the same solution could not be used for comparative assays. The atropin was therefore weighed directly each time. The results were as follows : Per Cent, of Atropin Used. I. II. III. IV. V. 1.12% 1.20% 1-49% 1.39% 0.88% Per Cent, of Atropin Obtained. I. II. III. IV. V. 1.04% 1.15% 1.46% 1-36% 0.88% Guarana.—The fluid extract of guarana can be analyzed by di- rect determination from the chloroform extract obtained from the magma. The results were as follows : I. II. III. IV. V. Average. 3.84% . 3.86% 3.84% 3.71% 3-78% 3-8i% 1 gm. of caffein was dissolved in 50 cc. of the fluid extract of guarana, and the strengthened extract was assayed with the following results : I. II. III. IV. V. Average. Theory. 5.68% 5-66% 5-56% 5-52% 5-40% 5-56% 5-8i % The average loss is 4 per cent, of the alkaloid. 4 gms. of caffein were dissolved in 100 cc. of alcohol and water, forming a 4 per cent, solution, which was analyzed as follows : I. II. III. IV. V. Average. Theory. 3.82% 3.76 % 3-80% 3-83% 3-86% 3.81% 4.00% Average loss, .19 per cent., or 4.7 per cent, of the alkaloid used Ipecac.—A fluid extract of ipecac gave the following results : I. II. III. IV. V. Average. 1.42% i-37% i-52% i-47% i-52% 1.46% In this case the greatest departure from the average was 7 per cent, of the weight of alkaloid present. The alkaloid emetin is of such a delicate and unstable nature 10 THE ASSAY OF ALKALOIDS. that it was considered useless to carry out the check experiments employed in the cases of the preceding alkaloids. Aconite Root.—The assays of the fluid extract of aconite root gave the following results : I. II. III. IV. V. Average. 0.41% 0.42% O.41 % 0.44% 0.42% 0.42% In these analyses the greatest variation is 5 per cent, of the alkaloid present. It was not deemed expedient to make experiments with alka- loidal addition to the fluid extract of aconite root or of hyoscy- amus, on account of the instability of aconitin and hyoscyamin. Hyoscyamus.—The fluid extract of hyoscyamus was assayed as follows : I. II. III. IV. V. Average. 0.10% 0.10% 0.09% 0.09% 0.10% 0.10% VI. INCREASE IN WEIGHT OF CERTAIN ALKAEOIDS AFTER SOEUTION IN CHLOROFORM AND SUBSEQUENT EVAPORATION. An interesting and important fact noticed in connection with this work was the alteration in weight of some alkaloids after so- lution in chloroform, and subsequent evaporation. This altera- tion, usually a gain in weight, was most noticeable in the case of brucin. The real nature of this change was not ascertained. As, how- ever, it has a marked bearing on the value of the process in ques- tion, a series of experiments as to the extent of this variation was carried out in the following manner. A weighed amount of an alkaloid varying from o. i to o.oi gm. was dissolved in chloroform, the solution placed upon a tared watch-glass, and then evaporated on the water-bath to dryness. In the case of those alkaloids which were left in a gummy condi- tion, the residues were stirred with the point of a pen-knife. The glass was then cooled in a desiccator, over calcium chlorid, for about ten minutes, and quickly weighed. Cajfein.—Caffein, unlike most alkaloids, leaves a crystalline residue. The variations noted were inconsiderable, except where small amounts were used. THE ASSAY OF AEKAEOIDS 11 The figures obtained were as follows : Amount of Caffeiin Used. 1. 11. hi. IV. V. Average. O.I gm. . . . O. % 0. % 0. % CO f —0.6% -o-3% 0.05 “ . . ■ —r-9 —2.9 0. —0.9 —2.1 —i-5 0.01 “ . . 0. 0. —3-o 0. 11. + 1.6 Percentages of Loss or Gain. Brucin.—The results obtained were as follows : Amount of Brucin Used. 1. II. hi. IV. V. Average. O. I gm. . . . 6.6% 9-3% 9.8% 9.4% 8.9% 8.9% O.05 “ . . ■ 8.3 6.6 6-3 9.8 II.0 8.4 O.OI “ . . • 6.3 3-2 4.4 10.0 4-9 5-5 Average, 7.6 % Percentages of Gain in Weight. It will be noticed that the range of variation increases as the amount of the alkaloid used in the evaporation decreases. Strychnin.—The results obtained were as follows : Amount of Strychnin Used. I. II. III. IV. V. Average. o.i gm. . . 0.9% —6.5% —4-7 % 2-7 % —1.8# 0.05 “ . . . 0.0 —1.6 0.0 —0.7 i-3 0.4 Average, —°-7% Percentages of Loss or Gain. Aconitin.—The results obtained were as follows : Percentages of Gain. Amount of Aconitin Used. 1. 11. hi. IV. V. Average. O. I gm. . . . 6.7 % 5-8% 4.1% 5-o% 5-9% 5-5% O.05 “ . . • i-7 8.0 2.0 1-3 2.4 3-i O.OI “ . . 0.0 2.0 2.6 i-9 3-o i-7 * Average, 3-i% Atropin.—The results obtained were as follows : Percentages of Gain. Amount of Atropin Used. I. II. III. IV. V. Average. O. I gm. . . • 0.3% 0.2% 0.2% 0.4% 2.2 % 0.6% O.05 “ . . • 0.7 0.0 i-5 1.0 1.8 1.0 O.OI “ . . 0.0 4.0 *13.0 0.0 0.0 2.4 Average, 1.3% an error. 12 THE ASSAY OF ALKALOIDS. Quinin.—The results obtained were as follows : Amount of Quinin Used. I. II. III. IV. V. Average. o.l gm. . . . 0.4% —0.2% 1.1% 3-4% i-3% 1.2