Eng.-techn. I. V. Gebler

Some Data on Bergenia¹)

Among Siberian tanning plants, badan (Saxifraga crassifolia), a saxifrage inhabiting the mountainous regions of Western and Eastern Siberia, is of outstanding interest and real industrial significance—because of its considerable tannin content, its wide distribution, and the comparative ease with which it can be collected.

Both the leaves and the root of this plant contain tannins, but its technical value must be seen chiefly in the root, since the leaf, in comparison with the root, has negligible weight and, in mass procurement of badan, the leaf dries up, falls off, and is scattered; thus, in the end, after piling at the collection sites, after transfers and reloadings, the mass of prepared badan consists exclusively of root alone²), unless, of course, special measures are taken to preserve the leaf, such as sorting at the collection site, which would require additional labor for gathering the badan and thereby increase its procurement cost.

Although badan, as a tanning plant, attracted attention quite a long time ago, there are no exhaustive studies on the character of its tannins and on the data for extraction; tanning with it is carried out, moreover, apparently most of all in Irkutsk, while Baikal badan has been studied especially little; data on it in the literature, it seems, are wholly absent. In Irkutsk I was able to find only three incomplete analyses of the root: two in the Laboratory of the Bacteriological Institute and one in the Laboratory of the Trans-Baikal Railway; these analyses—

¹) The present work was carried out during the 1922–23 academic year in the laboratory of chemical technology of fibrous and dye substances of the Tomsk Technological Institute, by order of the Siberian Bureau of the Supreme Council of the National Economy. The study of badan as a tanning plant has so far not once yielded exhaustive data concerning the conditions of its extraction, the stability of its tannins in solutions under the influence of high temperature, etc. In particular, the question of the significant content of starch in badan roots, which has recently been subject to some doubt, and also of how a greater or lesser content of starch in the extracted material may affect changes in the operating processes of extract production, on one or another complexity of the technical equipment of production, on the composition and properties of the extract, and consequently on its suitability for the purpose of tanning, as well as the production of solid extracts readily soluble in water, are connected with a preliminary study of questions of purification and clarification of badan diffusion juices, enrichment of them with tannins, the influence on them of sulfitation, etc. Already a priori one might have expected that all the processes just named would have very great and important significance for the preparation of high-quality extracts from badan; nevertheless, it seemed necessary to verify experimentally that concentrated badan extracts would give good results in tanning heavy leather.

All these assumptions were fully confirmed by the present investigation; the author’s quite well-founded conclusions are undoubtedly of great scientific and practical interest, since they fully illuminate the conditions under which one may design a badan-extract industry in Siberia. Prof. V. I. Minaev.

²) This I had occasion to observe during large-scale procurement of badan at the Irkutsk tanneries.

Table 1.

The most recent study known to me is the following: information about it is found in Prof. Povarnin’s book Tanning Materials, where the average composition of bergenia root is given as follows: at 13% moisture: tannins, 20.5%; non-tannins, 20%; sugar, 5.2%; ash, 5%; there is an indication of a high starch content, up to 40% (according to Harr and Maillard). In No. 13 of the Bulletin of the Main Administration of Leather Industry for 1921, in Prof. Povarnin’s article “New Data on Bergenia,” an extract obtained from bergenia by extraction with acetone and methyl alcohol is described, and the results are also given of the determination of carbohydrates in the root (free glucose, 5.42%; glucosidic glucose, 6.61%; starch, about 7%). In an article by the same author, “On Little-Known Domestic Tanning Materials” (Bulletin of the All-Russian Society of Tanners, 1915, No. 22), information is given on bergenia, while the ease of collecting this root and its use for tanning by the peoples of Eastern Siberia are noted.

Finally, numerous analyses of bergenia root (47 analyses) and leaves (17 analyses) were made by V. P. Markov and are given in his article: “Study of Tanning Plants of the Central Altai” (Proceedings of the Tomsk Technological Institute, Vol. 43, Issue 1). The data available in the literature are very insufficient for drawing practical conclusions and a definite judgment about bergenia as a material for extract production. In order to supplement this gap at least somewhat, I carried out several experiments on the extraction of bergenia and on the use of the resulting extract for tanning hides under laboratory conditions. The results of these experiments are given below.

For the work, Altai bergenia was taken—a root that had long been kept in the laboratory; the moisture content of the root was 8.50%. The root was ground in a laboratory root crusher, yielding a coarse powder with a certain content of larger pieces up to 4 mm across. The ground material was subjected to extraction in a Koch apparatus with steam at 1 meter; a 20 g sample of the ground root was pre-soaked in water for 12 hours; the rate of withdrawal of the extract was 1 liter in 3 hours; this condition proved necessary and sufficient for complete extraction of the substances soluble at \(t = 100^\circ\text{C}\), which was established by evaporation of the last portions of the extract and by reaction with ferric chloride (very weak coloration). In all, four extracts were prepared at different temperatures, namely at 100°, 75°, 50°, and 20°; the departure from the official method used in studying tanning materials for tannin content consisted in carrying out the extraction from beginning to end at one and the same temperature, which appears necessary in the present case, considering not the analysis of the material but the clarification of the conditions of extraction and the nature of the extracts obtained.

All extracts were examined by the usual methods for their content of extractive substances and tannins; the results obtained are given in the following table.

Table 2.

T = tannins; NT = non-tannins.

If, from these data, one calculates the content of substances in the root taken (at 8.56% moisture), the figures obtained will be as follows:

Table 3.

Taking into account that at 100° extraction was complete, the tables presented give a clear idea of the shortfall of extractive and tanning substances that occurs in the case of extraction at lower temperatures; moreover, as can be seen, the very character of the extracts also changes, since the content in them of non-tanning substances, on the contrary, [[unclear: line damaged]] and the ratio T/NT, i.e., simultaneously with the temperature, the quality of the juices obtained falls. These ratios, characterizing the success of extraction as a function of temperature, may be clearly represented graphically in the appended diagram, where, on the chosen scale:

Curve EE corresponds to the content of extractive substances in the extract
“ TT ” “ tannins
“ NN ” “ non-tannins
“ DD expresses the ratio T/NT and characterizes the quality of the extracts.

Thus the only temperature suitable for extracting badan root is one close to 100°, and the proposal that, in order to avoid possible dextrinization of starch (which in fact does not occur), extraction of badan should be carried out at a low temperature of about 30°C—proposals that are expressed in the literature (V. G. Varavin, Tanning Materials, p. 159)—prove to be erroneous. Until the proper extraction temperature was established, in subsequent studies we always took extracts at 100°C, which we take as normal.

All extracts, with the exception of the 4th (30°), are quite turbid at ordinary temperature, owing to insoluble substances falling out of the solutions below 50°C, which are completely transparent at this temperature; the greatest quantity of insolubles is found, as is evident (Table 2), in the 1st extract (100°), and amounts in this case to 28.66% of the weight of dry substances. The insoluble matter settles extremely slowly; over the course of many days the extracts remain turbid, and their clarification does not occur even over an excessively long period; these observations were made with an extract placed in a cylinder with a ground-glass stopper (filled to the top) and [[unclear: text cut off]].

…impregnated with toluene. Filtering the insoluble matter is difficult: the precipitates at first pass through the densest paper filters (Schleicher—Schüll No. 590); later the pores of the filter become clogged and filtration does not proceed even under pressure. It was best accomplished through an asbestos filter in a Fresenius tube, with strong rarefaction in the receiver; in precisely this way the separation of insoluble precipitates was carried out (at \(t = 16^\circ\mathrm{C}\)).

Insoluble or sparingly soluble (at higher temperatures) substances, often found in extracts of tanning plants and having the character of tannins, bear various names: “insoluble,” “red precipitate,” “anhydrides,” and, finally, “phlobaphenes.” It is accepted that these substances are oxidized or condensed tannins, or products of their dehydration, or, finally, compounds of altered (or unaltered) tannins with some non-tannin substances: sugars, acids, pectic substances, etc.; accordingly, in all probability, they arise from the amphoteric (alcohol-acid) function of the hydroxyl groups of tannins. Sometimes phlobaphenes are contained in the plant itself, so to speak, in finished form (natural phlobaphenes); in other cases their formation has been observed in extracts when the latter are heated for a more or less long time, and then their content may be, generally speaking, variable. Characteristic of phlobaphenes is their property of giving compounds with sodium bisulfite that are soluble at any temperature and that penetrate into leather; this property is widely used in technology for preparing soluble extracts and is carried out under the special name of “sulfitization.”

To ascertain the character of the insoluble bergenia, the following experiments were made.

1. The extract (500 cu. cm.), containing in 100 cu. cm. 1.676 g of dry substances and 0.472 g of insolubles, was boiled with a reflux condenser for 8 hours; after each 2-hour interval the liquid was cooled to \(16^\circ\mathrm{C}\) (laboratory temperature), a quantitative portion of 50 cu. cm. was taken and filtered in the above-described manner through an asbestos filter; 25 cu. cm. of the clear filtrate (the first portions were discarded) were evaporated, and from the residue dried at \(105^\circ\mathrm{C}\) the amount of soluble substances was determined; the insolubles were determined by difference. The results, recalculated to 100 cu. cm. of extract, are as follows:

Table 4.

1. The same extract, by evaporation in vacuum in a water bath, was successively thickened and, for various concentrations corresponding to a reduction of the volume by 2, 4, and 8 times in relation to the initially taken portion, the amount of dry substances and insolubles was determined at the same temperature, \(16^\circ\mathrm{C}\). The figures thereby obtained are given in Table 5.

Table 5.

Thus, within the limits of the unavoidable experimental errors, the amount of insoluble matter may be considered constant; it does not change in either case, i.e., when the extract is heated without access of air, and also with full access of air at various concentrations; the amount of insoluble matter is always, on the average, 28% of the weight of the dry substances. The observation made gives an indirect indication that the insoluble substances are contained in the plant itself and are hardly formed during extraction.

3. For the determination of tannins in the insoluble matter, the filtered precipitate was washed on an asbestos filter with a small quantity of cold water, then repeatedly with boiling water until the precipitate was completely dissolved; the solution was adjusted to a dry-substance content of approximately 0.5 g per 100 cc, and in it, after cooling, tanning substances were determined by the usual method with hide powder. It should be noted that precipitation from solutions on cooling does not occur at once, and the liquid, in the supercooled state, remains transparent for some time, then begins to opalesce and, finally, becomes completely turbid owing to the formation of a precipitate; this circumstance made it possible to take quantitative aliquots and to carry out the necessary determinations each time with transparent liquids at normal temperature.

The figures obtained were as follows:
200 cc of extract (100°) were taken, with a dry-substance content of 1.9240 g per 100 cc. Filtered at 16°C; the precipitate was dissolved in hot water, the volume brought to 100 cc; of the latter solution there was taken:

a) for determination of dry substances: 25 cc.
result: 0.1320 gram (A).

b) for detanning with hide powder: 25 cc.
dry residue after detanning: 0.0422 g (B).

\[ \% \text{ tannins} = \frac{(A-B)100}{A} = 68.03 \]

of the weight of the insoluble substance dried at 100°.

4. The same solution which served for the determination of tannins was tested for reducing sugars directly and after hydrolysis. In the first case the solution was decolorized with basic lead acetate, the excess lead was precipitated with sodium sulfate, and the filtrate from PbSO₄ was subjected to Fehling’s reaction: reduction of copper oxide did not occur and, consequently, no reducing sugars were found. Another portion of the solution was boiled with sulfuric acid (5 g H₂SO₄ per 100 cc of liquid) with a reflux condenser for three hours, the acid liquid was neutralized with barium carbonate to a neutral reaction and then prepared and tested as before; here too no reduction reaction occurred and the Fehling number = 0. These data, together with other reactions taken for the substance accompanying the tannins in the insoluble bergenia in an amount of about 32%, give the following characterization of it: 1) it does not reduce Fehling’s solution, 2) it is not hydrolyzed by acids with formation of reducing sugars, 3) it is not precipitated by acids, 4) it is not precipitated by salts of heavy metals, 5) it gives a flocculent precipitate with caustic alkalis, soluble in acetic acid.

Having established the tannin nature of the insoluble bergenia, we have the right to call them, according to the existing terminology, phlobaphenes. A preliminary experiment on the behavior of these phlobaphenes toward sodium bisulfite gave a positive result: the bisulfited extract did not show the presence of insoluble matter, remaining transparent at any temperature.

For a more detailed elucidation of the bisulfiting reaction, important for extract production, a corresponding experiment was set up with sufficient precision; for this purpose the extract was taken as a whole, such a circumstance—

[[unclear: beginning of line]] more closely corresponds to practical conditions. As the bisulfite we used a laboratory preparation of dry acid sodium sulfite, containing \(NaHSO_3\), in which the content was determined at \(75.32\%\). A \(1/2\%\) solution of this salt was added in various amounts to 25 cc of extract containing 1.676 g of dry matter in 100 cc; test tubes with such samples were closed with a stopper, into the opening of which a long tube was inserted (serving as a reflux condenser), and were placed for 10 hours in a boiling water bath. After cooling and standing for 12 hours, the liquid in the test tubes was assessed for clarity, then filtered, and the amount of solubles was determined in the filtrate; in the case of completeness of the reaction, this obviously should not be less than 0.419 g per 25 cc of the extract taken. An exact correspondence of the amount of solubles with the amount of dry substances in general, however, cannot be attained here, owing to the addition of bisulfite, the amount of which that entered into the reaction is rather difficult to take into account. The experiment is presented in the following table:

Table No. 6.

Thus, the amount of \(NaHSO_3\) necessary and sufficient for complete bisulfiting—in the sense of converting into solution the phlobaphenes contained in samples 6 and 7—amounts in the present case to \(13.48\%\) of the weight of the dry substances. In other words, when sulfiting extracts, which is usually carried out on solutions obtained after the first boiling-down, when the content of dry substances has been brought to \(40\%\), the consumption of \(37^\circ\) Bé liquid bisulfite will be approximately 16–17 poods per 100 poods of the sulfited 40% product.

Thus, in badan root there are two suites of tannides, which differ mainly in the unequal solubility of these in water; or, more precisely, one group of tannides is fully soluble at all temperatures, giving solutions of high concentration, whereas the solubility of the tannides (or compounds of tannides) of the other group changes depending on the temperature, increasing on heating; at a temperature of \(15\text{–}16^\circ C\) these substances are completely insoluble and are quantitatively present in the precipitate, as is evident from the fact that the filtrate obtained by separating this precipitate at \(16^\circ C\) always remains transparent and does not give those reactions which are characteristic of the second group of tannides.

An independent investigation of tannides, in view of the little-elucidated nature of them and the absence of strict systematization, presents great difficulties; therefore, in the study of tannides one usually confines oneself to individual reactions carried out according to Procter’s scheme, which gives a series of empirical signs that make it possible to compare and identify tannides of different origin. Such characteristic reactions were applied to tannides of badan of both groups and are summarized in Table 7, without indicating the conditions under which they are carried out, as these are usual and sufficiently well known.

Table 7.

The reaction with a solution of $\mathrm{SnCl_2}$ in conc. $\mathrm{HCl}$ is very characteristic and clearly expressed; it may serve as a qualitative reaction for the tannide of badan; in general, a bright fuchsin-red coloration appears in this case instantly, even in very dilute solutions.

Lime water quantitatively precipitates the tannides of badan; the filtrate from such a precipitate gives no tannide reactions; the precipitate is soluble in carbonates, with formation of $\mathrm{CaCO_3}$; the solution, evidently containing in this case tannates of the alkalis, does not give tannide reactions; the same precipitate dissolves completely on boiling in a solution of acid sodium carbonate, and all reactions characteristic of tannides are detected in such a solution, except for

does not precipitate in the form of an acid sulfite salt. Such precipitation of tannin substances by means of Ca(OH)\(_2\), with subsequent dissolution of the precipitate in sodium bisulfite, is of some interest as a method for separating tannins from all extraneous substances not precipitated by calcium oxide in the cold, and makes it possible to obtain solutions containing only bisulfite tannins that tan hide; this could be used in preparing mixtures of tanning substances of definite character and composition for particular cases of tanning.

To determine the decomposability of bergenia tannins under the influence of heating, 100 c.c. of a normal extract, containing 0.802 g of dry substances and 0.420 g of tannin, were boiled under a reflux condenser for 8 hours, after which the tanning substances were determined by the usual method; their content proved to be 0.4019 g per 100 c.c. of the boiled extract, i.e. the loss of tannin in this case was 4.30%. This circumstance must be borne in mind when extracting the roots, and the liquid should not be brought to boiling; on the other hand, as was indicated above, lowering the temperature to 75° causes an even greater loss of tannins, owing to incomplete extraction; ultimately, the most advantageous extraction temperature should be about 90° C.

As for the nontannins present in the root, the following determinations were made to clarify their character:

1) reducing sugars in the normal extract,

2) the same after inversion,

3) the Fehling number in relation to the root after inversion.

For the first two determinations an extract was taken containing 1.924 g of dry substances in 100 c.c.; the liquid was decolorized and in general prepared in the same way as was indicated in the investigation of the insolubles; the actual determination of sugars was made with Fehling’s solution by the volumetric method: 10 c.c. of Fehling’s solution were placed in a test tube, to which the solution under examination was added from a burette, and the whole was boiled for 5 minutes; by this test with different quantities of the solution, the amount needed for reduction of the copper was determined with an accuracy of 1 c.c.; after finding this limit, 10 c.c. of Fehling’s liquid were placed in each of 10 test tubes, and the solution under examination was added in such quantities that the difference between successive test tubes was 0.1 c.c.; the test tubes were placed in a boiling water bath for 10 minutes, after which the test tube in which complete reduction of the copper oxide had occurred was determined, and the result was recalculated as glucose; it was thus found that 100 c.c. of extract (1.924 g of dry substances) contain 0.4315 g of glucose, or, in relation to the root (moisture 8.50), 8.97%. After hydrolysis the figure obtained was approximately the same, which indicates the absence of starch (soluble) in the solution.

To determine the substances in the root that undergo hydrolysis with formation of reducing sugars, a sample of the latter (8.50 moisture) of 5 g, previously soaked in water, was boiled with 5% sulfuric acid for 3 hours under a reflux condenser; after inversion, the filtered liquid was neutralized, and the determination was then carried out as before, the solution being adjusted to contain approximately 1 g of dry substances per 100 c.c.; it was found that 1 g of material, after deducting the sugars passing into solution and counted in the extract as glucose, reduces 41.8 c.c. of Fehling’s liquid.

Comparing the data obtained on bergenia root, we give a table, in average figures, characterizing this material:

Table 8.

For a closer acquaintance with the properties of bergenia extract, the same material with which the above-described work was carried out was subjected to extraction. In an amount of 2 kilograms, the root was extracted with distilled water in glass jars placed in a water bath at a temperature close to boiling, under a pressure of 1 meter. The extract obtained at 4°B was evaporated to 15°B in a flat porcelain dish on a water bath; the temperature of the evaporated liquid did not rise above 85°C. The condensed juice was sulfited with an amount of sodium bisulfite calculated on the basis of the preceding determinations, in a flask on a boiling water bath for 10 hours. The sulfited product was then evaporated in vacuo at a temperature of about 80°C¹ until the extract became so dry that it crumbled into powder. The yield of such extract was 44.5% of the weight of the root with 8.5% moisture; it readily dissolved in cold water.

Its analysis gave:

In % of the weight of the extract:

The amount of dry substances and the approximate tannin content in solutions of such an extract in relation to their specific gravity are given in Table 9.

Table 9.

An experiment in rapid laboratory tanning with bergenia on half-stock was carried out with a hide of medium weight, obtained in the form of rawhide at one of the Tomsk tanneries; it was not possible to obtain heavier leather.

The rawhide was washed in water, then, for neutralization, placed for 1 hour in a \(1/2\%\) solution of hydrochloric acid, washed again, and put into the tanning liquor. The weight of the hide taken for this was 75.5 g. The tanning process is shown in the scheme given. Table 10.

Table 10.

The tanned leather was washed with water, then rotated for 15 minutes in a \(1 \tfrac{1}{2}\%\) solution of oxalic acid; at the same time the dark-brown color of the grain changed to a beautiful yellow-brown; after this the leather was again washed, wrung out on rollers, lightly greased with vorvan’ on the grain side, and dried in the laboratory. The results are as follows: the tanning was satisfactory, the leather dense and firm; the gain is determined from the following ratios.

\[ \text{Gain in \%}=\frac{(26.9-14.9)100}{14.9}=81.5. \]

It is still premature to draw a definitive conclusion about the material under investigation, for this requires trials of the use of bergenia at factories for various cases of tanning, and not with it alone, but also in mixtures with other tanning agents.

However, on the basis of the data obtained, even now the following conclusions may be drawn:

1. Bergenia is one of the few Siberian tanning materials suitable for extract production, owing to its large area of distribution, ease of collection, considerable tannin content, and simplicity of extraction.

2. The large quantity of phlobaphenes, which, as is believed, are necessary for tanning sole leather and determine the weight of the leather, makes it possible to assume the suitability of bergenia for this purpose; for the same reason, sulfiting of the extract is, generally speaking, necessary, although alongside the sulfited extract an unsulfited extract may also be prepared, suitable for weak liquors; strong unsulfited liquors, owing to the presence of insoluble substances, have the consistency of jelly and therefore cannot be used.

3. For vat tanning, bergenia must be regarded as of little suitability, since the release of tanning substances into cold water is too incomplete, and in the tanning material, in this case, a significant quantity of tannin will remain, which will constitute a partial loss of it.

4. The comparatively high value of the Fehling number for the root after boiling it with acid indicates a considerable content of starch (and perhaps of other carbohydrates) that does not pass into solution and remains in the tanning material; consequently, the possibility of utilizing the latter by fermentation is not excluded.

For the article by P. V. Golovin, “Some Data on Bergenia”

![Graph: Increase in substances in 100 g of air-dry matter vs. extraction temperature. The vertical axis reads “Increase in substances in 100 g of air-dry matter”; the horizontal axis reads “Extraction temperature in degrees C.” Curves are marked with the letters F, K, T, N, M, and B.]

1. It was not possible, with the apparatus available, to attain a high vacuum. ↩