The Natural Organic Colouring Matters
By
Arthur George Perkin, F.R.S., F.R.S.E., F.I.C., professor of colour chemistry and dyeing in the University of Leeds
and
Arthur Ernest Everest, D.Sc., Ph.D., F.I.C., of the Wilton Research Laboratories; Late head of the Department of Coal-tar Colour Chemistry; Technical College, Huddersfield
Longmans, Green and Co.
39 Paternoster Row, London
Fourth Avenue & 30th Street, New York
Bombay, Calcutta, and Madras
1918
(Tekstiin lisätty kappaleita lukemisen helpottamiseksi. // Some paragraphs added to the original text for making reading easier.)
Kuvia (kemialliset kaavat) puuttuu // Most of the illustrations (of chemical formulas) included.
The employment of natural dyestuffs dates back to remote antiquity, and we have no knowledge when their tinctorial properties were first discovered and what was the original method of their application. It is to be presumed that the first steps in the art of dyeing resulted from the accidental staining of the skin or fabric by a vegetable material, and were followed by the use for this purpose of extracts of this and of other similar products which possessed colour.; In this manner certain fairly permanent effects due to a real attraction between colouring matter and fibre would come to light, and a substantive natural dyestuff would thus be revealed. On the other hand, substantive dyestuffs occur but rarely in nature, and it was not until the discovery of the art of mordanting, known, however, at a very early date in Hindustan and the Far East, that any considerable advance in real dyeing became possible. The mordants at first employed consisted evidently of the naturally occurring sulphates of aluminium and iron, the former being the most valuable for this purpose. Indeed Bancroft in his "Philosophy of Permanent Colours" (1813) remarks that the discovery of alum was one of the most important events in the history of dyeing. A method became thus available for the utilisation of the numerous adjective or mordant dyestuffs with which nature abounds, and moreover varied coloured effects of a really permanent character, previously unknown except perhaps in the case of indigo dyed materials, could now be produced by the dyer. A knowledge of the use of these salts gradually spread from India and Persia to Egypt, thence to Greece and Rome, and ultimately farther westward, of which matters an interesting account is given by Bancroft (loc. cit.). Important, again, for the progress of dyeing was the discovery of America, which resulted in the introduction into Europe of valuable dyestuffs previously unknown, such as logwood, brazil-wood, and cochineal.
The variety of natural products which have been and still are employed for dyeing is large, though the use of many of these has been confined to the uncivilised or semi-civilised countries in which they abound. Those which have acquired economic importance ares comparatively few in number, and may be said to represent the survival of the fittest since no better natural dyestuffs are known, or are likely to be discovered in the future. These comprise:
Logwood
Alkanet
Young fustic
Brazil-wood
Cochineal
Quercitron bark
Sanders-wood
Lac-dye
Persian berries
Barwood
Turmeric
Old fustic
Camwood
Annatto
Safflower
Caliaturwood
Orchil ?
Indigo
Madder
Weld
Woad?
Without doubt, this list could have been usefully extended in the past by the inclusion of certain of the native Indian dyes, such as Chay root and Morinda root, etc., had better methods of cultivation been adopted and a more exact knowledge of their tinctorial properties been obtained.
Whereas the majority of these dyes, owing to the advent of artificial colours, are now of limited importance, considerable quantities of logwood, old fustic, Persian berries, and catechu still find application, and though the use found for the remainder is trivial in the extreme compared with that of former days, it cannot be said that any have entirely disappeared from the market. A revival indeed of the employment of natural dyes has occurred at the present time, due to the scarcity of the artificial colouring matters, but this cannot be of a permanent character. Owing to the easy solubility of the natural colouring matters, especially in presence of other extractive matter Derived from the plant, which permits of their ready absorption by tfie fabric during the dyeing process, and to the fact that the majority yield with mordants colours of a more or less permanent character, as a class they suffered but little in general estimation during the twenty years which succeeded the advent of the artificial dyes. Madder up to that time the most important of all colouring matters, was, however, soon superseded by the synthesis on a commercial scale of its main colouring matter, and the gradual introduction of the brighter and more easily applied azo dyestuffs caused a steady decline in the employment of other of these natural products. Indigo, the most important survivor of the vegetable group, is now produced artificially, and there can be little doubt that the virtual extinction of the remaining members of this class which still find application is merely a matter of time.
In addition to the commercially important natural dyestuffs, there are a very large number, which on account of their inferior tinctorial strength, and for other reasons, are now practically unused. Among these are to be found kermes, the oldest dyestuff on record, dyer's broom, green ebony, onion skins, alder bark, saw-wort, and walnut skins, all of which were of service in the past, more especially to the home dyer in whose district they were available. Even at the present day heather and a species of lichen known as "crottle"are still employed in the outlying districts of Scotland and Ireland, and the use of other of these secondary natural dyestuffs may still prevail in like case to a minor extent throughout Europe.
Of the Eastern dyestuffs which come under this category there are a considerable variety, including, in addition to those previously mentioned, the root bark of the Ventilago madraspatana, the bark of the Myrica nagi or rubra, the flowers of the cotton plant and of the Butea frondosa (Tesu flowers), the Cedreela toona and Thespasia lampas, Kamala, Waras, pomegranate rind, Lokao, and the buds of the Sophora japonica. In certain of our museums, again, specimens of numerous so-called natural dyestuffs are to be found, mostly originating from the East, which merely dye or stain fabrics indefinite shades of a brown or brownish- red colour. Many of these no doubt contain red phlobaphens formed by an alteration of a catechol tannin, whereas probably in others the small amount of dye originally present has suffered decomposition.
Though natural products which are, or have been, employed for dyeing fabrics are extremely numerous, they represent but a small portion of those vegetable products in which colouring matter is present. Indeed it is hard to find a plant which, taken as a whole, does not possess the property of dyeing with mordants. That this is the case has been well known for a considerable time, and in the older treatises on dyeing, more especially Bancroft's "Philosophy of Permanent Colours" (1813) and the "Matieres tinctoriales" of Leuchs (1829), many plants possessing this character are described. Very generally the mere trace of dye present gives yellow shades on aluminium mordant, and this whilst far too weak in character to have interested the dyer of the past, has now little interest for the chemist owing in general to its identity with one or other of the substances present in some more readily available commercial product.
Though there is no rule by which the existence of a dye in this or that portion of a plant can be predicted with any certainty, the colouring matter in general is most prolific in the leaf or flower, and absent in most cases from the fruit, stem, bark, and root. On the other hand, when present in the latter, these are usually rich in colouring matter, as is the case with Persian berries, the dyewoods, quercitron bark, and madder. Colourless or yellow flowers capable of dyeing, such as white clover and the yellow primrose, give yellow shades on aluminium mordant, due to the presence of flavone or flavonol glucosides, and these, frequently contained also in red, blue, or violet flowers, are a cause of the green shades they give on aluminium mordant, which otherwise in their absence would be dyed blue as the result of the anthocyanin present. In but rare cases is the yellow dye present responsible for the yellow tint of the flower, for this, usually in the form of glucoside, is more or less colourless. The tint of the latter, in fact, is frequently due to carotin or a similar substance which like this is insoluble in water and devoid of dyeing property. Occasional instances are, however, met with in which the colour of the petal appears to be due to the presence of a yellow acid potassium salt of the flavone glucoside. Very interesting was the discovery by the late Dr. Hugo Miiller that flavone is the main constituent of the "farina" or "flour" which accumulates on the leaves of certain varieties of the primula.
As the history of the chemistry of this subject is given in detail later on in respect of each individual dyestuff, many points of considerable interest are purposely omitted in the following brief general statement. During the early part of the last century numerous investigations on the nature and general reactions of the colouring matters present in the natural dyestuffs were carried out, and as the subject was then of considerable technical importance, much space was given to an account of the results in the older manuals of dyeing. These, now difficult of access, and indeed out of print, were Bancroft's "Philosophy of Permanent Colours," 1813; Berthollet, "On Dyeing," translated by Ure, 1824; "Matieres tinctoriales," Leuchs, 1829; "Leçons de chimie appliquee a la teinture," Chevreul, 1830; " Traite des matieres colorantes," Schützenberger, 1867; and Crookes' "Dyeing and Calico Printing," 1874. Points of interest are also to be found in the still older volume by Hellot on "The Art of Dyeing Wool, Silk, and Cotton," 1789.
The most important of this early work was due to Chevreul, who in 1810 isolated haematoxylin from logwood and brazilin from brazil-wood, in 1814 morin from old fustic, and at about the same period luteolin from weld, fisetin (then termed fustin) from young fustic, quercitrin from quercitron bark, and ellagic acid from gallnuts. In addition to these substances, which are described as crystalline, Chevreul prepared a crude bixin from annatto and proved that a compound capable of developing indigo, rather than indigo itself, is present in the indigofera. Almost at the same time (1818) Pelletier and Caventou obtained carminic acid from cochineal, whereas to Robiquet and Colin, in 1826, we owe the first isolation of alizarin from madder, followed a year later by the discovery of purpurin in the same plant. To the first-named chemist is also to be ascribed the detection, in 1849, of orcin from the Variolaria dealbata, a variety of lichen employed for the manufacture of orchil. In addition to these colouring matters of commercial importance, others less valuable tinctorially were discovered about this period, among them being berberine (1826), datiscin (1816), gentisin (1827), and rutin (1842). Indeed it may be said that by the end of 1860 few, if any, of the natural tinctorial products readily available had escaped attention. The majority of the colour preparations of these older workers were crystalline, and occasionally chemically pure, as, for instance, the carminic acid of Warren de la Rue (1847), and the methods then devised for their isolation have often proved to be of considerable service to later investigators.
In 1847 a more critical study of madder was commenced by Schunck (Ann. Chem. Phann., 66, 176), with the result that it was soon evident that this root contained in addition to alizarin and purpurin also a small amount of a complex mixture of yellow crystalline substances now known to be anthraquinone derivatives. Again, it became apparent that certain at least of these compounds were present in the root as glucosides, and eventually the true glucoside of alizarin, the ruberythric acid of Rochleder (1851) termed rubianic acid by Schunck, was isolated in a pure condition. Schunck, from an analysis of the potassium salt of ruberythric acid, deduced for alizarin the formula C14H10O4, and this was subsequently altered by Graebe and Liebermann to C14H8O4.
The preparation of alizarin and purpurin from madder, at best a very tedious process, was simplified to some extent somewhat later by the appearance on the market of preparations designed for calico printing containing the colouring matter in a more concentrated form. Such were the commercial "purpurin" and "green and yellow alizarins" of Kopp (1864), the essential feature in their manufacture being the extraction of the ground root with aqueous sulphurous acid.
To within the last fifteen years this operation was still carried out in France, the "purpurin" thus obtained yielding a lake the shade of which was difficult to obtain in other ways.
Alizarin, considered to be at first a derivative of naphthalene, was subsequently recognised by Graebe and Liebermann, who employed Baeyer's method of zinc-dust distillation, as a derivative of anthracene, and the suspicion that it was in reality a dihydroxy-anthraquinone was confirmed, as is well known, by its synthesis in 1868. The production of alizarin on the manufacturing scale no doubt gave hopes of a similar commercial success with other important natural colouring matters, but this, except in the case of indigo, has not been realised, and has been, indeed, unnecessary owing to the production of artificial colours that can be more easily applied to fabrics.
In 1865 Baeyer commenced the long series of researches which led to the synthesis of indigotin, and in 1880 his well-known method for its production from ο-nitrocinnamic acid was announced. The accounts of these and other researches which have culminated In the manufacture of artificial indigo, are so fully given elsewhere that their repetition here is unnecessary.
Before 1890 little real advance was made in the determination of the actual structure of other commercial natural colouring matters, and, indeed, in many cases their correct formulæ were still in doubt.
Carminic acid, the colouring matter of cochineal, first obtained crystalline by Schützenberger in 1867, had, however, received considerable attention, and the production from it of tri-nitrococussic acid (trinitrocresotinic acid) by Warren de la Rue (1847), of coccinin by Hlasiwetz and Grabowski in 1869, and the bromcarmines by Will and Leymann in 1885, have proved of considerable value to later workers. That progress here was slow can now hardly be wondered at, because of all the natural colouring matters hitherto submitted to exhaustive investigation, carminic acid, perhaps, has appeared the most elusive in disclosing the true nature of its structure. Thus whereas the very able work of Miller and Rohde in 1897 on the bromcarmines suggested the probability that carminic acid was derived from naphthalene, in the same year Liebermann and Voswinkel from a study of its oxidation products preferred then to consider it as a derivative either of hydrindene or bishydrindene, although as the result of further work in 1904 a naphthacene quinone constitution appeared more probable to these authors. Finally, Dimroth (1909), who studied the more gentle degradation of carminic acid, although at first inclined to regard this colouring matter as a naphthalene compound, has, after a series of brilliant investigations (1913), described in the sequel, proved that it is derived from anthraquinone and that both the kermessic acid of kermes and laccainic acid of lac dye contain a similar nucleus.
An interesting, though now unimportant, Indian pigment is the Puiri or Indian yellow, a compound deposited from the urine of cows fed on the leaves of the mango tree, and which is the source of the somewhat feeble dye euxanthone. Euxanthone, though of little interest to the dyer, is historically important in that it was the first of the many natural yellow dyestuffs of which the constitution was determined. A study of this compound by Baeyer in 1870 indicated that it was derived from benzophenone, and, largely as the result of his work, Salzmann and Wichelhaus, seven years later, assigned to it the constitution of a dihydroxy-diphenylene ketone oxide. Diphenylene ketone oxide, subsequently termed xanthone by v. Kostanecki, was discovered by Kolbe and Lautermann in 1860, who prepared it by the interaction of phosphorus oxychloride and sodium salicylate, though the method devised by the late Sir W. H. Perkin in 1883, which consists in distilling acetic anhydride and salicylic acid, is the most convenient. The products of the hydrolysis of euxanthone are resorcin and hydroquinone carboxylic acid, and by a reversal of this process, that is distilling these substances with acetic anhydride, euxanthone was synthesised by Graebe (1889) and later by v. Kostanecki and Nessler (1891).
The commencement of a new era in the investigation of the natural yellow colouring matters dates from the patient study by Herzigof quercetin (1884), derived from quercitron bark, and fisetin, which is present in young fustic. Although the older formula C24H16O11, assigned to quercetin by Liebermann and Hamburger in 1879, was employed in his earlier papers, C15H10O7 was ultimately (1891) proved to be correct. It was subsequently pointed out by Perkin and Pate (1895) that quercetin, fisetin, and other yellow colouring matters yield with acids well-defined salts, the analysis of which indicates with some certainty their molecular weight, - as in the case of quercetin which forms salts of the type C15H10O7HCl. Of these two colouring matters Herzig first in 1891 determined the constitution of fisetin; this resulted from a study of the products (fisetol diethyl ether and protocatechuic acid diethyl ether) which he obtained by the gentle hydrolysis of fisetin tetraethyl ether. The fact that fisetol proved to possess the constitution led to the conception that fisetin was in reality the tetrahydroxy phenyl pheno γ-pyrone. Quercetin by analogy was represented as hydroxy fisetin. About the same period (1893) v. Kostanecki submitted to examination chrysin, C18H10O4, a very feeble colouring matter, which Piccard, in 1864, had isolated from the buds of the common poplar.
From the properties of this substance and the fact previously observed by Piccard that when hydrolysed it yields acetophenone and phloroglucinol, v. Kostanecki represented it as a dihydroxy phenyl pheno γ-pyrone.
Such a compound on hydrolysis would be expected to give, first, trihydroxy benzoyl acetophenone and then acetophenone, and phloroglucinol carboxylic acid, the latter subsequently passing into phloroglucinol. On the other hand, the hydrolysis could evidently also take place in another manner with formation of phloracetophenone and benzoic acid, and indeed phloroglucinol and acetic acid, derived from the former, and benzoic acid are also obtained from it in this way.
An interesting fact observed by v. Kostanecki with chrysin and euxanthone, and by Herzig with quercetin, is that when alkylated with alkyl iodides in the well-known manner the hydroxyl in the ortho position to the carbonyl group is not attacked. Again, the alkylated product, though still containing a free hydroxyl group, is insoluble in aqueous alkali, but gives by means of alcoholic potash a potassium salt which is hydrolysed by water. This behaviour is to a certain extent possessed by all hydroxyketones, and is evidence of an hydroxyl in this position. Perkin has shown, however, in the case of luteolin, quercetin, and the analogously constituted colouring matters myricetin and quercetagetin, that by employing excess of the reagents fully alkylated products can be obtained, and this no doubt will generally prove to be the case.
v. Kostanecki designated the mother substance of chrysin flavone, whereas the mother substance of fisetin which contains a hydroxyl attached to the γ-pyrone nucleus he termed "flavonol". From the period of 1895 onwards, a considerable number of natural yellow colouring matters have been examined, many of which have been proved to belong to the flavone or flavonol groups, and there can be no doubt that of all the natural dyes, these are much the most widely distributed in nature.
In 1898 Emilewicz, v. Kostanecki, and Tambor announced the synthesis of chrysin, employing for this purpose a series of reactions which represent a reversal of the scheme of hydrolysis outlined above. Thus ethyl benzoate condensed with phloracetophenone trimethyl ether gives 2.4.6 trimethoxybenzoylacetophenone, and this by treatment with hydriodic acid is demethylated, and passes into chrysin.
Other methods of synthesis were subsequently applied to chrysin by v. Kostanecki and his co-workers, and in 1899 flavone itself was prepared, followed in 1900 by apigenin (parsley) and luteolin (weld). Somewhat later a method was devised for the artificial preparation of flavonols, and in 1904 fisetin, quercetin, and kaempferol were synthesised by v. Kostanecki and his co-workers, morin being similarly obtained in 1907.
Kuvat puuttuvat.The course of the reactions employed may be illustrated as follows:
When ο-hydroxyacetophenone is condensed with benzaldehyde 2 hydroxybenzilidene acetophenone (2 hydroxy chalkone) is obtained.
This on boiling in alcoholic solution with dilute sulphuric acid by absorption and elimination of water is converted into dihydroflavone (flavanone).
By the action of amyl nitrite and hydrochloric acid iso-nitrosoflavone is produced and this dilute sulphuric acid converts into the ketone, which subsequently passes into the flavonol.
Interesting is the fact pointed out by Perkin and Hummel in 1904 that a chalkone and flavanone exist side by side as glucosides, and to a small extent in the free condition, in the flowers of the Butea frondosa. These compounds which are named Butin and Butein possess the following constitutions: [KUVA PUUTTUU]
and were synthesised in the form of their trimethyl ethers. Thus resacteophenone monomethyl ether, condensed with protocatechuic aldehyde dimethyl ether forms butein trimethyl ether, and from this the corresponding butin derivative is readily produced by the action of dilute alcoholic sulphuric acid.
A colouring matter known for many years past, and, in fact, among those examined by Chevreul, is ellagic acid, which though possessing somewhat feeble dyeing properties, gives a very fast shade on a chromium mordant. In the form of ellagitannic acid, it appears to be present in all plants which yield ordinary gallotannin, and is produced in considerable amount when certain tannin extracts, such as those of divi-divi and myrobalans, are subjected to fermentation, Merklein and Wöhler assigned to it, in 1845, the formula C14H5O8, and it is interesting as the first natural yellow colouring matter to have been synthetically prepared. This was accomplished by Löwe in 1868, who obtained it as a product of the interaction of gallic and arsenic acids, though since then other and better oxidation processes have been devised. Whereas Schiff, as early as 1879, suggested two constitutions for ellagic acid, one of which is now known to be correct, it was considered by Barth and Goldschmiedt, in the same year, that as ellagic acid when distilled with zinc-dust yields fluorene, it is most probably a fluorenone derivative. Graebe, however, in 1903 pointed out that diphenyl-methylolid treated in this way gives fluorene, and accordingly again proposed the constitution for ellagic acid given in Schiff's paper and referred to above.
That this is correct was established by Perkin and Nierenstein in 1905, who pointed out that the first product of the hydrolysis of ellagic acid is in reality pentahydroxy-diphenyl-methylolid. Though ellagic acid is at present the only known naturally occurring member of this group, other hydroxy derivatives of diphenyldimethylolid can be obtained, by the oxidation of hydroxybenzoic acids other than gallic acid, and by the more energetic oxidation of gallic acid itself.
For several years after the time of Chevreul little work of importance was carried out with haematoxylin, the colouring principle of the commercially important logwood, though in 1842 Reim proposed for it the formula C16H14O6, which is now known to be correct. Brazilin, the very similar colouring principle of brazil-wood, was examined by Liebermann and Burg in 1876, and the formula C16H14O5 assigned to it, and the many properties it possesses in common with haematoxylin indicated its probable relationship to this latter substance. This, as the result of the work of W. H. Perkin (junr.) and his pupils, has ultimately proved to be the case. It had long been known that haematoxylin and brazilin were not the actual colouring matters of logwood and brazil-wood, but that to develop this property an oxidation ("ageing") process was necessary. Haematoxylin thus yields haematein, as shown by Reim in 1871, and that brazilin behaves in the same way with formation of brazilein was pointed out by Liebermann and Burg in 1876.
These substances were subsequently, in 1882, isolated in the pure crystalline condition by Hummel and A. G. Perkin, and from the fact that they, respectively, possessed the formulæ C16H12O6 and C16H12O5, their simple relation to haematoxylin and brazilin appeared evident. The study of the constitution of these compounds received its first impetus from the work of Schall and Dralle in 1888, who, by the more energetic oxidation of an alkaline solution of brazilin with air, obtained β-resorcylic acid and a substance which Schall in 1 894 considered was the pheno γ-pyrone derivative (dihydroxychromone).
That this supposition was correct was proved by Feuerstein and v. Kostaneckiin 1899, and this in conjunction with the fact that Herzig, in 1898, and Gilbodyand Perkin, almost simultaneously, had obtained evidence in this compound of a catechol nucleus, led the former investigators to suggest the following as the formula for brazilin [KUVA PUUTTUU].
In 1899 there appeared the first of an elaborate series of papers by W. H. Perkin and his pupils on the constitution of haematoxylin and brazilin, and as a result an important series of acids obtained from both brazilin trimethyl ether and haematoxylin tetramethyl ether by oxidation were described. The constitutions of these acids was determined by synthesis, and these results in the face of alternative suggestions by v. Kostanecki and Lampe (1902) and Herzig and Pollak in 1906, resulted in 1908 in the formulæ of Perkin and Robinson for both brazilin and haematoxylin which are now accepted as correct.
Of interest is the Hæmatoxylon africanum more recently discovered in Africa by Pearson, which differs from the Hæmatoxylon campcachianum (logwood) in that in place of the haematoxylin present in the latter wood, it contains a small amount of a red colouring matter, which judging by its properties, is brazilin itself (A. G. Perkin, private communication).
Though more allied to the tannins, and, indeed, itself largely employed in the tanning of leather, catechu is of considerable service for dyeing purposes, due to an alteration during these processes in the catechin it contains. Catechin was first isolated in 1832 by Nees van Esenbeck, but its true formula, C15H14O6, was only correctly ascertained in 1902 by v. Kostanecki and Tambor, and Perkin and Yoshitake simultaneously. At the same time the latter authors observed that the catechins of Gambier catechu, and Acacia catechu, though isomeric are distinct substances, that present in the latter being now termed aca-catechin. Whereas Perkin suggested that these catechins were probably reduction products of quercetin, which is always associated with them in the plant, v. Kostanecki and his co-workers produced evidence of a cumaran nucleus in catechin and considered the following expression to be more correct: [KUVA PUUTTUU].
A synthesis catechin compound has not yet been effected, and further work appears necessary to confirm the above formula; moreover, an explanation is still required of the nature of the interesting change by which catechin so readily yields catechutannic acid, the true tanning principle.
As already indicated, the natural substantive dyestuffs comprise but a small group, and have always been of somewhat minor importance, owing to the fact that the shades they yield are either extremely fugitive, or less permanent than those obtainable by other methods. Among these are to be found safflower, annatto, barberry, turmeric, and the insoluble red woods, sanders-wood, barwood, camwood, and caliatur-wood, and it is of interest to note that these latter and also turmeric possess in addition to their substantive character the property of dyeing with mordants.
Of the members of this class the main interest has hitherto centred round turmeric, whose colouring matter, curcumin, was first isolated by Vogel in 1842. Though obtained crystalline by Daube in 1870 it was not until 1897 that Ciamician and Silber ascertained that its molecular weight is represented by C21H20O6 rather than by the older expression C14H14O4. The problem of its constitution was more recently attacked (1910) by Melobedzka, v. Kostanecki, and Lampe, and there is now little doubt that this colouring matter is an unsaturated β-diketone of the following formula: [Kuva puuttuu]
Safflower, at one time highly esteemed as the source of a very beautiful though expensive and fugitive shade of red, whilst still employed in the East, is extinct as a commercial dyestuff in this country. It contains both a red and yellow colouring matter; the latter of little interest, though both have been detected as the cause of the red and yellow dye of certain Egyptian mummy cloths. Carthamine, the red colouring matter first isolated by Schlieper in 1846, received little attention until 1910, when Kametaka and Perkin succeeded in obtaining it in a crystalline condition. Though possessing a more complex formula than curcumin, a certain resemblance exists between these colouring matters, as regards the simple nature of their decomposition products, and there is reason to suppose that they may be structurally related. Definite evidence, however, of the constitution of this latter, of the bixin of annatto still employed in this country for colouring foodstuffs, and of the santalin of sanderswood is still lacking, although the recent work of Cain and Simonsen (1912) has added considerably to our knowledge of this latter compound.
An elucidation of the true nature of the chemistry of the natural indigo process, long overdue, has been rendered clear by the researches of Hoogewerff and ter Meulen. Though Schunck in 1858, as is well known, by his work on woad and the Polygonum tinctorium, established in this connection points of considerable importance, his indican, though rich in colouring principle, was amorphous, and of an unstable nature. On the other hand, there is evidence that the colouring principle of woad which he describes is distinct from that present in the indigoferæ and the Polygonum tinctorium, and, owing to its instability, has hitherto baffled all attempts to isolate it in a pure condition. It has now been ascertained that pure indican derived from the indigofera, and which is in reality a glucoside of indoxyl, crystallises readily, and is quite stable in ordinary circumstances, though by the action of dilute acids and of a special ferment present in the plant, it is readily hydrolysed with formation of glucose and indoxyl. By means of a very simple method devised by Perkin and Bloxam in 1907 very large quantities of indican can be readily isolated, and its use has permitted the study in considerable detail of this interesting glucoside.
Closely connected with indigo, as is now proved to be the case, is the so-called purple of the ancients, also known as Tyrian purple, for many ages probably the dye most esteemed above all others. The colouring matter exists in the reduced condition, in very small amount in certain molluscs, as a yellowish-green fluid or solution, which on exposure to air rapidly develops a purple colour. The process by which it was applied to fabrics by the ancients is unknown, but it probably consisted in breaking up the molluscs under water and employing the supernatant liquid for dyeing. According to Crookes ("Calico Printing," II) molluscs of this character were employed for dyeing purposes in Bristol about 1663. The subject appears to have been first chemically investigated by Bizio in 1833, and the result of this and of later investigations pointed to a considerable resemblance between this colouring matter and indigotin. Schunck in 1879 described in detail the properties of this substance of which from 400 molluscs he obtained 7 milligrams and was successful in obtaining it in a crystalline condition by sublimation.
To those who have examined these animals, the difficulty and unpleasantness involved in obtaining a sufficient amount of the colouring matter for investigation appeared almost unsurmountable, but these were overcome with consummate skill by Friedlander in 1906. As the result of this work, which is described in detail in the following pages, Friedlander ascertained that this colouring matter contains bromine and is in reality a dibromindigotin of the constitution [KUVA PUUTTUU]
Of special interest in recent years has been the results of the investigations of Willstätter with Everest (1913) and others on the nature of the red and blue colouring matters of flowers and of fruits which are termed anthocyanins. Though in the past certain berries containing this class of substances have been employed to a slight extent for dyeing with mordanted fabrics, and it has been known that red and blue flowers give in the same way bluish or more generally greenish shades on aluminium mordant, these colours are fugitive and the subject has hitherto possessed more interest for the biochemist than for the dyer. Beyond the determination of the general properties of these colouring matters, and the fact that whereas blue flowers contain them in a neutral, and red flowers in an acid condition, no evidence of importance as to their structure had been forthcoming until the work of Willstätter and Everest on the blue colouring matter of the corn-flower appeared. As a result of this and later investigations, it is now known that these anthocyanins are always present in the plant as glucosides, and that from these by hydrolysis with acid the free colouring matter termed an anthocyanidin may be isolated in the form of its oxonium salt. Both classes of compounds are crystalline and are not so unstable as their behaviour in the coloured petal would lead one to expect. These compounds are in reality derivatives of benzopyranol and thus are closely connected with the flavones, or flavone glucosides, of which they may be regarded as reduction products. This relationship will be evident on comparing the formula of Pelargonidin (from the Pelargonium zonale) as hydrochloride 3.5.7 trihydroxy 2 p-hydroxy phenyl 1.4 benzopyranol anhydro hydrochloride with the flavonol kaempferol (Delphinium consolida).
A detailed description of these interesting compounds will be found in a later chapter.
The tannins, at one time employed to a considerable extent in the production of a black or grey colour on fabrics, by means of iron salts, and which even now take part in the black dyeing of silk, are distinct from and, indeed, are not usually classified among dyestuffs. This arises from the fact that, except in the case of the iron mordant, they do not, as a rule, dye fabrics which have been mordanted with the other metallic compounds used in practice, though, on the other hand, gallotannin is remarkable in that with titanium mordant it produces a bright yellow colour. Whatever view may be taken on this point, tannins are so largely employed as assistants in the dyeing operation, as for instance in the fixation of basic colours on cotton, and appear generally to bear some relationship to the yellow colouring matters with which they are usually associated in the plant, that their description naturally falls within the scope of a work of this character. Perkin, indeed, has pointed out that, in those cases which have been investigated, the tannin or tannin principle, or should there be two, one of these and the yellow colouring matter of the plant, contain either identical phenolic nuclei or at least one phenolic nucleus in common. Thus gallotannin appears to invariably accompany myricetin (pyrogallol nucleus) and a catecbol tannin, quercetin (catechol nucleus), whereas catechin and quercetin and cynanomaclurin and morin respectively occurring together contain identical phenolic nuclei.
Though the tannins are very widely distributed in nature, and by their general reactions can be grouped into three main classes, our knowledge of these compounds is slight, except as regards gallotannin. Löwe in 1867 and Schiff in 1871 claimed to have synthesised this compound, and the latter chemist proposed for it the digallic acid structure which was generally accepted for many years. It is now known as the result of the work of Walden (1899) and others, that Schiffs synthetical product, whatever it may have been, is quite distinct from the natural tannin. Among the various suggestions subsequently made as to the constitution of gallotannin, that of Nierenstein, who considered it to consist of a mixture of digallic and leucodigallic acids (1907), appeared most plausible, and indeed in 1910 this author described the preparation of digallic and in 1912 that of leucodigallic acid from gallotannin. In 1912 again Nierenstein came to the conclusion that gallotannin is more complex than he had previously supposed, and that it may consist of a polydigalloyl leuco digalloyl anhydride. Fischer and Freudenberg, however, in 1912 took an entirely different view of the constitution of tannin, and proved that the pure substance when hydrolysed with acid, always gives in addition to gallic acid a small amount of glucose. As a result they consider gallotannin to be a compound of dextrose with five molecules of w-digallic acid of the nature of a pentacetyl derivative, which will not only account for its very high molecular weight but also for its optical activity. These chemists were successful in synthesising both p and m-digallic acids of which the latter is concerned in the structure of tannin and among the many interesting proofs given in their elaborate work on this subject for their suggested constitution of gallotannin is their synthesis of penta-penta- methyl digalloyl glucose. This though differing slightly in optical activity from the penta-methyl gallotannin of Herzig (1905) very closely resembles it, and there can be little doubt the difference observed is due to the fact that both products contained a mixture of stereo-isomerides.
In concluding this brief sketch of the work which has been carried out upon the more important natural dyestuffs, it is hardly necessary to point out that the subject, even in its main features, is far from being exhausted. Not only is this so in respect of the dyes already enumerated, and with those of unknown constitution which are described later on, but in addition there exist without doubt, more especially in certain flowers, numerous natural colouring matters, evidently members of new chemical groups, which yet remain to be isolated. From a study of these no direct technical advantage can be anticipated, but on the other hand an elucidation of their nature will not only assist the biochemist in the problems he has to face, but will add materially to our present knowledge of the effect of constitution on colour.
In the succeeding chapters the natural dyestuffs are grouped according to the constitution, where known, of their main tinctorial constituents, and where members of two widely distinct groups, as for instance, those derived from flavone and anthraquinone, exist side by side in the same plant, the description of the plant will be found under that heading which from its present or past uses appears the more suitable.
The present known chemical groups to which the natural dyestuffs belong are given below, in the order in which the subject is treated in this volume:
