Symmetry of pairs of hydrogen atoms.

R. Hubner and A. Petermann (Ann., 1869, 149, p. 129) provided the proof of the equivalence of the atoms 2 and 6 with respect to 1. From meta-brombenzoic acid two nitrobrombenzoic acids are obtained on direct nitration; elimination of the bromine atom and the reduction of the nitro to an amino group in these two acids results in the formation of the same ortho-aminobenzoic acid. Hence the positions occupied by the nitro groups in the two different nitrobrombenzoic acids must be symmetrical with respect to the carboxyl group. In 1879, Hubner (Ann., 195, p. 4) proved the equivalence of the second pair, viz. 3 and 5, by starting out with ortho-aminobenzoic acid, previously obtained by two different methods. This substance readily yields ortho-oxybenzoic acid or salicylic acid, which on nitration yields two mononitro-oxybenzoic acids. By eliminating the hydroxy groups in these acids the same nitrobenzoic acid is obtained, which yields on reduction an aminobenzoic acid different from the starting-out acid. Therefore there must be another pair of hydrogen atoms, other than 2 and 6, which are symmetrical with respect to 1. The symmetry of the second pair was also established in 1878 by E. Wroblewsky (Ann., 192, p. 196).

Orientation of Substituent Groups.—The determination of the relative positions of the substituents in a benzene derivative constitutes an important factor in the general investigation of such compounds. Confining our attention, for the present, to di-substitution products we see that there are three distinct series of compounds to be considered. Generally if any group be replaced by another group, then the second group enters the nucleus in the position occupied by the displaced group; this means that if we can definitely orientate three di-derivatives of benzene, then any other compound, which can be obtained from or converted into one of our typical derivatives, may be definitely orientated. Intermolecular transformations—migrations of substituent groups from one carbon atom to another—are of fairly common occurrence among oxy compounds at elevated temperatures. Thus potassium ortho-oxybenzoate is converted into the salt of para-oxybenzoic acid at 220 deg.; the three bromphenols, and also the brombenzenesulphonic acids, yield m-dioxybenzene or resorcin when fused with potash. It is necessary, therefore, to avoid reactions involving such intermolecular migrations when determining the orientation of aromatic compounds.

Such a series of typical compounds are the benzene dicarboxylic acids (phthalic acids), C6H4(COOH)2. C. Graebe (Ann., 1869, 149, p. 22) orientated the ortho-compound or phthalic acid from its formation from naphthalene on oxidation; the meta-compound or isophthalic acid is orientated by its production from mesitylene, shown by A. Ladenburg (Ann., 1875, 179, p. 163) to be symmetrical trimethyl benzene; terephthalic acid, the remaining isomer, must therefore be the para-compound.

P. Griess (Ber., 1872, 5, p. 192; 1874, 7, p. 1223) orientated the three diaminobenzenes or phenylene diamines by considering their preparation by the elimination of the carboxyl group in the six diaminobenzoic acids. The diaminobenzene resulting from two of these acids is the ortho-compound; from three, the meta-; and from one the para-; this is explained by the following scheme:—

NH2 NH2 NH2 NH2 NH2 NH2 ^ ^ ^ ^ ^ ^ / NH2 / NH2 / COOH / / / COOH /COOH / /NH2 /NH2 HOOC /NH2 / v v v v v v COOH COOH NH2 / , / , v v NH2 NH2 NH2 ^ ^ ^ / NH2 / NH2 / / /NH2 / v v v COOH NH2

W. Korner (Gazz. Chem. Ital., 4, p. 305) in 1874 orientated the three dibrombenzenes in a somewhat similar manner. Starting with the three isomeric compounds, he found that one gave two tribrombenzenes, another gave three, while the third gave only one. A scheme such as the preceding one shows that the first dibrombenzene must be the ortho-compound, the second the meta-, and the third the para-derivative. Further research in this direction was made by D.E. Noetling (Ber., 1885, 18, p. 2657), who investigated the nitro-, amino-, and oxy-xylenes in their relations to the three xylenes or dimethyl benzenes.

The orientation of higher substitution derivatives is determined by considering the di- and tri-substitution compounds into which they can be transformed.

Substitution of the Benzene Ring.—As a general rule, homologues and mono-derivatives of benzene react more readily with substituting agents than the parent hydrocarbon; for example, phenol is converted into tribromphenol by the action of bromine water, and into the nitrophenols by dilute nitric acid; similar activity characterizes aniline. Not only does the substituent group modify the readiness with which the derivative is attacked, but also the nature of the product. Starting with a mono-derivative, we have seen that a substituent group may enter in either of three positions to form an ortho-, meta-, or para-compound. Experience has shown that such mono-derivatives as nitro compounds, sulphonic acids, carboxylic acids, aldehydes, and ketones yield as a general rule chiefly the meta-compounds, and this is independent of the nature of the second group introduced; on the other hand, benzene haloids, amino-, homologous-, and hydroxy-benzenes yield principally a mixture of the ortho- and para-compounds. These facts are embodied in the "Rule of Crum Brown and J. Gibson" (Jour. Chem. Soc. 61, p. 367): If the hydrogen compound of the substituent already in the benzene nucleus can be directly oxidized to the corresponding hydroxyl compound, then meta-derivatives predominate on further substitution, if not, then ortho- and para-derivatives. By further substitution of ortho- and para-di-derivatives, in general the same tri-derivative [1.2.4] is formed (Ann., 1878, 192, p. 219); meta-compounds yield [1.3.4] and [1.2.3] tri-derivatives, except in such cases as when both substituent groups are strongly acid, e.g. m-dinitrobenzene, then [1.3.5]-derivatives are obtained.

Syntheses of the Benzene Ring.—The characteristic distinctions which exist between aliphatic and benzenoid compounds make the transformations of one class into the other especially interesting.

In the first place we may notice a tendency of several aliphatic compounds, e.g. methane, tetrachlormethane, &c., to yield aromatic compounds when subjected to a high temperature, the so-called pyrogenetic reactions (from Greek [Greek: pyr], fire, and [Greek: gennao], I produce); the predominance of benzenoid, and related compounds—naphthalene, anthracene, phenanthrene, &c.—in coal-tar is probably to be associated with similar pyrocondensations. Long-continued treatment with halogens may, in some cases, result in the formation of aromatic compounds; thus perchlorbenzene, C6Cl6, frequently appears as a product of exhaustive chlorination, while hexyl iodide, C6H13I, yields perchlor- and perbrom-benzene quite readily.

The trimolecular polymerization of numerous acetylene compounds—substances containing two trebly linked carbon atoms, —C:C—, to form derivatives of benzene is of considerable interest. M.P.E. Berthelot first accomplished the synthesis of benzene in 1870 by leading acetylene, HC:CH, through tubes heated to dull redness; at higher temperatures the action becomes reversible, the benzene yielding diphenyl, diphenylbenzene, and acetylene. The condensation of acetylene to benzene is also possible at ordinary temperatures by leading the gas over pyrophoric iron, nickel, cobalt, or spongy platinum (P. Sabatier and J.B. Senderens). The homologues of acetylene condense more readily; thus allylene, CH:C.CH3, and crotonylene, CH3.C:C.CH3, yield trimethyl- and hexamethyl-benzene under the influence of sulphuric acid. Toluene or mono-methylbenzene results from the pyrocondensation of a mixture of acetylene and allylene. Substituted acetylenes also exhibit this form of condensation; for instance, bromacetylene, BrC:CH, is readily converted into tribrombenzene, while propiolic acid, HC:C.COOH, under the influence of sunlight, gives benzene tricarboxylic acid.

A larger and more important series of condensations may be grouped together as resulting from the elimination of the elements of water between carbonyl (CO) and methylene (CH2) groups. A historic example is that of the condensation of three molecules of acetone, CH3.CO.CH3, in the presence of sulphuric acid, to s-trimethylbenzene or mesitylene, C6H3(CH3)3, first observed in 1837 by R. Kane; methylethyl ketone and methyl-n-propyl ketone suffer similar condensations to s-triethylbenzene and s-tri-n-propylbenzene respectively. Somewhat similar condensations are: of geranial or citral, (CH3)2CH.CH2.CH:CH.C(CH3):CH.CHO, to p-isopropyl-methylbenzene or cymene; of the condensation product of methylethylacrolein and acetone, CH3.CH2.CH:C(CH3).CH:CH.CO.CH3, to [1.3.4]-trimethylbenzene or pseudocumene; and of the condensation product of two molecules of isovaleryl aldehyde with one of acetone, C3H7.CH2.CH:C(C3H7).CH:CH.CO.CH3, to (1)-methyl-2-4-di-isopropyl benzene. An analogous synthesis is that of di-hydro-m-xylene from methyl heptenone, (CH3)2C:CH.(CH2)2.CO.CH3. Certain a-diketones condense to form benzenoid quinones, two molecules of the diketone taking part in the reaction; thus diacetyl, CH3.CO.CO.CH3, yields p-xyloquinone, C6H2(CH3)2O2 (Ber., 1888, 21, p. 1411), and acetylpropionyl, CH3.CO.CO.C2H5, yields duroquinone, or tetramethylquinone, C6(CH3)4O2. Oxymethylene compounds, characterized by the grouping >C:CH(OH), also give benzene derivatives by hydrolytic condensation between three molecules; thus oxymethylene acetone, or formyl acetone, CH3.CO.CH:CH(OH), formed by acting on formic ester with acetone in the presence of sodium ethylate, readily yields [1.3.5]-triacetylbenzene, C6H3(CO.CH3)3; oxymethylene acetic ester or formyl acetic ester or [beta]-oxyacrylic ester, (HO)CH:CH.CO2C2H5, formed by condensing acetic ester with formic ester, and also its dimolecular condensation product, coumalic acid, readily yields esters of [1.3.5]-benzene tricarboxylic acid or trimesic acid (see Ber., 1887, 20, p. 2930).

In 1890, O. Doebner (Ber. 23, p. 2377) investigated the condensation of pyroracemic acid, CH3.CO.COOH, with various aliphatic aldehydes, and obtained from two molecules of the acid and one of the aldehyde in the presence of baryta water alkylic isophthalic acids: with acetaldehyde [1.3.5]-methylisophthalic acid or uvitic acid, C6H3.CH3.(COOH)2, was obtained, with propionic aldehyde [1.3.5]-ethylisophthalic acid, and with butyric aldehyde the corresponding propylisophthalic acid. We may here mention the synthesis of oxyuvitic ester (5-methyl-4-oxy-1-3-benzene dicarboxylic ester) by the condensation of two molecules of sodium acetoacetic ester with one of chloroform (Ann., 1883, 222, p. 249). Of other syntheses of true benzene derivatives, mention may be made of the formation of orcinol or [3.5]-dioxytoluene from dehydracetic acid; and the formation of esters of oxytoluic acid (5-methyl-3-oxy-benzoic acid), C6H3.CH3.OH.COOH, when acetoneoxalic ester, CH3.CO.CH2.CO.CO.CO2C2H5, is boiled with baryta (Ber., 1889, 22, p. 3271). Of interest also are H.B. Hill and J. Torray's observations on nitromalonic aldehyde, NO2.CH(CHO)2, formed by acting on mucobromic acid, probably CHO.CBr:CBr:COOH, with alkaline nitrites; this substance condenses with acetone to give p-nitrophenol, and forms [1.3.5]-trinitrobenzene when its sodium salt is decomposed with an acid.

By passing carbon monoxide over heated potassium J. von Liebig discovered, in 1834, an interesting aromatic compound, potassium carbon monoxide or potassium hexaoxybenzene, the nature of which was satisfactorily cleared up by R. Nietzki and T. Benckiser (Ber. 18, p. 499) in 1885, who showed that it yielded hexaoxybenzene, C6(OH)6, when acted upon with dilute hydrochloric acid; further investigation of this compound brought to light a considerable number of highly interesting derivatives (see QUINONES). Another hexa-substituted benzene compound capable of direct synthesis is mellitic acid or benzene carboxylic acid, C6(COOH)6. This substance, first obtained from the mineral honeystone, aluminium mellitate, by M.H. Klaproth in 1799, is obtained when pure carbon (graphite or charcoal) is oxidized by alkaline permanganate, or when carbon forms the positive pole in an electrolytic cell (Ber., 1883, 16, p. 1209). The composition of this substance was determined by A. von Baeyer in 1870, who obtained benzene on distilling the calcium salt with lime.

Hitherto we have generally restricted ourselves to syntheses which result in the production of a true benzene ring; but there are many reactions by which reduced benzene rings are synthesized, and from the compounds so obtained true benzenoid compounds may be prepared. Of such syntheses we may notice: the condensation of sodium malonic ester to phloroglucin tricarboxylic ester, a substance which gives phloroglucin or trioxybenzene when fused with alkalis, and behaves both as a triketohexamethylene tricarboxylic ester and as a trioxybenzene tricarboxylic ester; the condensation of succinic ester, (CH2.CO2C2H5)2, under the influence of sodium to succinosuccinic ester, a diketohexamethylene dicarboxylic ester, which readily yields dioxyterephthalic acid and hydroquinpne (F. Herrmann, Ann., 1882, 211, p. 306; also see below, Configuration of the Benzene Complex); the condensation of acetone dicarboxylic ester with malonic ester to form triketohexamethylene dicarboxylic ester (E. Rimini, Gazz. Chem., 1896, 26, (2), p. 374); the condensation of acetone-di-propionic acid under the influence of boiling water to a diketohexamethylene propionic acid (von Pechmann and Sidgwick, Ber., 1904, 37, p. 3816). Many diketo compounds suffer condensation between two molecules to form hydrobenzene derivatives, thus [alpha,gamma]-di-acetoglutaric ester, C2H5O2C(CH3.CO)CH.CH2.CH(CO.CH3)CO2C2H5, yields a methyl-ketohexamethylene, while [gamma]-acetobutyric ester, CH3CO(CH2)2CO2C2H5, is converted into dihydroresorcinol or m-diketohexamethylene by sodium ethylate; this last reaction is reversed by baryta (see Decompositions of Benzene Ring). For other syntheses of hexamethylene derivatives, see POLYMETHYLENES.

Decompositions of the Benzene Ring.—We have previously alluded to the relative stability of the benzene complex; consequently reactions which lead to its disruption are all the more interesting, and have engaged the attention of many chemists. If we accept Kekule's formula for the benzene nucleus, then we may expect the double linkages to be opened up partially, either by oxidation or reduction, with the formation of di-, tetra-, or hexa-hydro derivatives, or entirely, with the production of open chain compounds. Generally rupture occurs at more than one point; and rarely are the six carbon atoms of the complex regained as an open chain. Certain compounds withstand ring decomposition much more strongly than others; for instance, benzene and its homologues, carboxylic acids, and nitro compounds are much more stable towards oxidizing agents than amino- and oxy-benzenes, aminophenols, quinones, and oxy-carboxylic acids.

Simple oxidation.

Strong oxidation breaks the benzene complex into such compounds as carbon dioxide, oxalic acid, formic acid, &c.; such decompositions are of little interest. More important are Kekule's observations that nitrous acid oxidizes pyrocatechol or [1.2]-dioxybenzene, and protocatechuic acid or [3.4]-dioxybenzoic acid to dioxytartaric acid, (C(OH)2.COOH)2 (Ann., 1883, 221, p. 230); and O. Doebner's preparation of mesotartaric acid, the internally compensated tartaric acid, (CH(OH).COOH)2, by oxidizing phenol with dilute potassium permanganate (Ber., 1891, 24, p. 1753).

Chlorination and oxidation.

For many years it had been known that a mixture of potassium chlorate and hydrochloric or sulphuric acids possessed strong oxidizing powers. L. Carius showed that potassium chlorate and sulphuric acid oxidized benzene to trichlor-phenomalic acid, a substance afterwards investigated by Kekule and O. Strecker (Ann., 1884, 223, p. 170), and shown to be [beta]-trichloracetoacrylic acid, CCl3.CO.CH:CH.COOH, which with baryta gave chloroform and maleic acid. Potassium chlorate and hydrochloric acid oxidize phenol, salicylic acid (o-oxybenzoic acid), and gallic acid ([2.3.4] trioxybenzoic acid) to trichlorpyroracemic acid (isotrichlorglyceric acid), CCl3.C(OH)2.CO2H, a substance also obtained from trichloracetonitrile, CCl3.CO.CN, by hydrolysis. We may also notice the conversion of picric acid, ([2.4.6]-trinitrophenol) into chloropicrin, CCl3NO2, by bleaching lime (calcium hypochlorite), and into bromopicrin, CBr3NO2, by bromine water.

The action of chlorine upon di-and tri-oxybenzenes has been carefully investigated by Th. Zincke; and his researches have led to the discovery of many chlorinated oxidation products which admit of decomposition into cyclic compounds containing fewer carbon atoms than characterize the benzene ring, and in turn yielding open-chain or aliphatic compounds. In general, the rupture occurs between a-keto group (CO) and a keto-chloride group (CCl2), into which two adjacent carbon atoms of the ring are converted by the oxidizing and substituting action of chlorine. Decompositions of this nature were first discovered in the naphthalene series, where it was found that derivatives of indene (and of hydrindene and indone) and also of benzene resulted; Zincke then extended his methods to the disintegration of the oxybenzenes and obtained analogous results, R-pentene and aliphatic derivatives being formed (R- symbolizing a ringed nucleus).

When treated with chlorine, pyrocatechol (1.2 or ortho-dioxybenzene) (1) yields a tetrachlor ortho-quinone, which suffers further chlorination to hexachlor-o-diketo-R-hexene (2). This substance is transformed into hexachlor-R-pentene oxycarboxylic acid (3) when digested with water; and chromic acid oxidizes this substance to hexachlor-R-pentene (4). The ring of this compound is ruptured by caustic soda with the formation of perchlorvinyl acrylic acid (5), which gives on reduction ethidine propionic acid (6), a compound containing five of the carbon atoms originally in the benzene ring (see Zincke, Ber., 1894, 27, p. 3364) (the carbon atoms are omitted in some of the formulae).

Cl Cl Cl ^ ^ ^ ^ // OH Cl // O Cl // /OH Cl / ==> ==> CO ==> CO ==> / CO2H / / OH Cl2 / O Cl2 / Cl2 / v v v v Cl2 Cl2 Cl2

(1) (2) (3) (4)

CCl CH2 // / Cl C CO2H CH CO2H ==> Cl C CH CCl2 CH3

(5) (6)

Resorcin (1.3 or meta dioxybenzene) (1) is decomposed in a somewhat similar manner. Chlorination in glacial acetic acid solution yields pentachlor-m-diketo-R-hexene (2) and, at a later stage, heptachlor-m-diketo-R-hexene (3). These compounds are both decomposed by water, the former giving dichloraceto-trichlor-crotonic acid (4), which on boiling with water gives dichlormethyl-vinyl-a-diketone (5). The heptachlor compound when treated with chlorine water gives trichloraceto-pentachlorbutyric acid (6), which is hydrolysed by alkalis to chloroform and pentachlorglutaric acid (7), and is converted by boiling water into tetrachlor-diketo-R-pentene (8). This latter compound may be chlorinated to perchloracetoacrylic chloride (9), from which the corresponding acid (10) is obtained by treatment with water; alkalis hydrolyse the acid to chloroform and dichlormaleic acid (11).

OH O O ^ ^ ^ / Cl / Cl2 Cl2/ Cl2 ==> ==> // OH H / O HCl / O v v v Cl2 Cl2 (1) / (2) (3) / v v HO2C.CCl2.CHCl.CCl2.CO.CCl3 HO2C.CCl:CH.CCl2.CO.CHCl2 (6) (4) v HO2C.CCl2.CHCl.CCl2.CO2H.CHCl3 CO2 + CHHC:CH.CO.CO.CHCl2 (7) (5) - v + CO.CCl2 ClOC.CCl:CCl.CO.CCl3 <== CO (8) (9) + CCl=CCl / v HO2C.CCl:CCl.CO.CCl3 ==> HO2C.CCl:CCl.CO2H + CHCl3 (10) (11)

Hydroquinone (1.4 or para-dioxybenzene) (1) gives with chlorine, first, a tetrachlorquinone (2), and then hexachlor-p-diketo-R-hexene (3), which alcoholic potash converts into perchloracroylacrylic acid (4). This substance, and also the preceding compound, is converted by aqueous caustic soda into dichlormaleic acid, trichlorethylene, and hydrochloric acid (5) (Th. Zincke and O. Fuchs, Ann., 1892, 267, p. 1).

OH O COOH CO2H CO2H ^ ^ ^ ^ / // Cl / Cl2 ClC// Cl2 ClC// CCl2 ClC CCl2 ==> ==> ==> ==> + ClC CHCl / Cl /Cl2 ClC /CCl ClC /CCl v v v v CO2H OH O O CO

(1) (2) (3) (4) (5)

Phloroglucin (1.3.5-trioxybenzene) (1) behaves similarly to resorcin, hexachlor [1.3.5] triketo-R-hexylene (2) being first formed. This compound is converted by chlorine water into octachloracetylacetone (3); by methyl alcohol into the ester of dichlormalonic acid and tetrachloracetone (4); whilst ammonia gives dichloracetamide (5) (Th. Zincke and O. Kegel, Ber., 1890, 23, p. 1706).

OH O ^ ^ > (3) Cl2C.CO.CCl2.CO.CCl3.CO2 / CL2/ CL2 / ==> / ==> (4) Cl2HC.CO.CHCl2.CH3O2C.CCl2.CO2.CH3 HO / OH O / O v v CL2 > (5) Cl2C.CONH2

(1) (2)

Reduction in alkaline solution.

When phenol is oxidized in acid solution by chlorine, tetrachlorquinone is obtained, a compound also obtainable from hydroquinone. By conducting the chlorination in alkaline solution, A. Hantzsch (Ber., 1889, 22, p. 1238) succeeded in obtaining derivatives of o-diketo-R-hexene, which yield R-pentene and aliphatic compounds on decomposition. When thus chlorinated phenol (1) yields trichlor-o-diketo-R-hexene (2), which may be hydrolysed to an acid (3), which, in turn, suffers rearrangement to trichlor-R-pentene-oxycarboxylic acid (4). Bromine water oxidizes this substance to oxalic acid and tetrabrom-dichloracetone (5).

OH O HOOC /OH // / Cl2CC // Cl2/ O HCl2C CO COOH -> -> -> / H /H2 HC /CH2 HC /CH2 / / / / Cl CCl CCl

(1) (2) (3) (4)

-> (5) Cl2BrC.CO.CBr3 + HO2C.CO2H

The reduction of o-oxybenzoic acids by sodium in amyl alcohol solution has been studied by A. Einhorn and J.S. Lumsden (Ann., 1895, 286, p. 257). It is probable that tetrahydro acids are first formed, which suffer rearrangement to orthoketone carboxylic acids. These substances absorb water and become pimelic acids. Thus salicylic acid yields n-pimelic acid, HOOC.(CH2)5.COOH, while o-, m-, and p-cresotinic acids, C6H3(CH3)(OH)(COOH), yield isomeric methylpimelic acids.

Resorcin on reduction gives dihydroresorcin, which G. Merling (Ann., 1894, 278, p. 20) showed to be converted into n-glutaric acid, HOOC.(CH2)3.COOH, when oxidized with potassium permanganate; according to D. Vorlander (Ber., 1895, 28, p. 2348) it is converted into [gamma]-acetobutyric acid, CH3CO.(CH2)3.COOH, when heated with baryta to 150-160 deg.

Configuration of the Benzene Complex.—The development of the "structure theory" in about 1860 brought in its train an appreciation of the chemical structure of the derivatives of benzene. The pioneer in this field was August Kekule, who, in 1865 (Ann., 137, p. 129; see also his Lehrbuch der organischen Chemie), submitted his well-known formula for benzene, so founding the "benzene theory" and opening up a problem which, notwithstanding the immense amount of labour since bestowed upon it, still remains imperfectly solved. Arguing from the existence of only one mono-substitution derivative, and of three di-derivatives (statements of which the rigorous proof was then wanting), he was led to arrange the six carbon atoms in a ring, attaching a hydrogen atom to each carbon atom; being left with the fourth carbon valencies, he mutually saturated these in pairs, thus obtaining the symbol I (see below). The value of this ringed structure was readily perceived, but objections were raised with respect to Kekule's disposal of the fourth valencies. In 1866 Sir James Dewar proposed an unsymmetrical form (II); while in 1867, A. Claus (Theoretische Betrachtungen und deren Anwendung zur Systematik der organischen Chemie) proposed his diagonal formula (III), and two years later, A. Ladenburg (Ber., 2, p. 140) devised his prism formula (IV), the six carbon atoms being placed at the six corners of a right equilateral triangular prism, with its plane projections (V, VI).

CH CH CH // / / HC// CH HC/ CH HC/ CH / * / HC /CH HC /CH HC /CH / / / CH CH CH

I Kekule II Dewar III Claus

HC -CH CH CH / / / / / HC /- -CH CH HC CH / / / HC CH HC CH / / / / / HC --/ CH / / / CH CH CH

IV V VI / v Ladenburg

Objections to Kekule's formula.

One of the earliest and strongest objections urged against Kekule's formula was that it demanded two isomeric ortho-di-substitution derivatives; for if we number the carbon atoms in cyclical order from 1 to 6, then the derivatives 1.2 and 1.6 should be different.[13] Ladenburg submitted that if the 1.2 and 1.6 compounds were identical, then we should expect the two well-known crotonic acids, CH3.CH:CH.COOH and CH2:CH.CH2.COOH, to be identical. This view was opposed by Victor Meyer and Kekule. The former pointed out that the supposed isomerism was not due to an arrangement of atoms, but to the disposition of a valency, and therefore it was doubtful whether such a subtle condition could exert any influence on the properties of the substance. Kekule answered Ladenburg by formulating a dynamic interpretation of valency. He assumed that if we have one atom connected by single bonds to (say) four other atoms, then in a certain unit of time it will collide with each of these atoms in turn. Now suppose two of the attached atoms are replaced by one atom, then this atom must have two valencies directed to the central atom; and consequently, in the same unit of time, the central atom will collide once with each of the two monovalent atoms and twice with the divalent. Applying this notion to benzene, let us consider the impacts made by the carbon atom (1) which we will assume to be doubly linked to the carbon atom (2) and singly linked to (6), h standing for the hydrogen atom. In the first unit of time, the impacts are 2, 6, h, 2; and in the second 6, 2, h, 6. If we represent graphically the impacts in the second unit of time, we perceive that they point to a configuration in which the double linkage is between the carbon atoms 1 and 6, and the single linkage between 1 and 2. Therefore, according to Kekule, the double linkages are in a state of continual oscillation, and if his dynamical notion of valency, or a similar hypothesis, be correct, then the difference between the 1.2 and 1.6 di-derivatives rests on the insufficiency of his formula, which represents the configuration during one set of oscillations only. The difference is only apparent, not real. An analogous oscillation prevails in the pyrazol nucleus, for L. Knorr (Ann., 1894, 279, p. 188) has shown that 3- and 5-methylpyrazols are identical.

Ladenburg's formula.

The explanation thus attempted by Kekule was adversely criticized, more especially by A. Ladenburg, who devoted much attention to the study of the substitution products of benzene, and to the support of his own formula. His views are presented in his Pamphlet: Theorie der aromatischen Verbindungen, 1876. The prism formula also received support from the following data: protocatechuic acid when oxidized by nitrous acid gives carboxytartronic acid, which, on account of its ready decomposition into carbon dioxide and tartronic acid, was considered to be HO.C(COOH)3. This implied that in the benzene complex there was at least one carbon atom linked to three others, thus rendering Kekule's formula impossible and Ladenburg's and Claus' possible. Kekule (Ann., 1883, 221, p. 230), however, reinvestigated this acid; he showed that it was dibasic and not tribasic; that it gave tartaric acid on reduction; and, finally, that it was dioxytartaric acid, HOOC.C(OH)2.C(OH)2.COOH. The formation of this substance readily follows from Kekule's formula, while considerable difficulties are met with when one attempts an explanation based on Ladenburg's representation. Kekule also urged that the formation of trichlorphenomalic acid, shown by him and O. Strecker to be trichloracetoacrylic acid, was more favourably explained by his formula than by Ladenburg's.

Baeyer's researches.

Other objections to Ladenburg's formula resulted from A. von Baeyer's researches (commenced in 1886) on the reduced phthalic acids. Baeyer pointed out that although benzene derivatives were obtainable from hexamethylene compounds, yet it by no means follows that only hexamethylene compounds need result when benzene compounds are reduced. He admitted the possibility of the formulae of Kekule, Claus, Dewar and Ladenburg, although as to the last di-trimethylene derivatives should be possible reduction products, being formed by severing two of the prism edges; and he attempted to solve the problem by a systematic investigation of the reduced phthalic acids.

CO C.OH C.OH (1)OH / // // / H2C/ CH.CO2Et HC// CH.CO2Et HC// C.CO2Et / EtO2C.(6) - - (5)H or > EtO2C.(3) - - (2)OH EtO2C.HC /CH2 EtO2C.C /CH2 EtO2C.C /CH / / / / / CO C.OH C.OH (4)H

I II III IV

Ladenburg's prism admits of one mono-substitution derivative and three di-derivatives. Furthermore, it is in accordance with certain simple syntheses of benzene derivatives (e.g. from acetylene and acetone); but according to Baeyer (Ber., 1886, 19, p. 1797) it fails to explain the formation of dioxyterephthalic ester from succinosuccinic ester, unless we make the assumption that the transformation of these substances is attended by a migration of the substituent groups. For succinosuccinic ester, formed by the action of sodium on two molecules of succinic ester, has either of the formulae (I) or (II); oxidation of the free acid gives dioxyterephthalic acid in which the para-positions must remain substituted as in (I) and (II). By projecting Ladenburg's prism on a plane and numbering the atoms so as to correspond with Kekule's form, viz. that 1.2 and 1.6 should be ortho-positions, 1.3 and 1.5 meta-, and 1.4 para-, and following out the transformation on the Ladenburg formula, then an ortho-dioxyterephthalic acid (IV) should result, a fact denied by experience, and inexplicable unless we assume a wandering of atoms. Kekule's formula (III), on the other hand, is in full agreement (Baeyer). This explanation has been challenged by Ladenburg (Ber., 1886, 19, p. 971; Ber., 1887, 20, p. 62) and by A.K. Miller (J.C.S. Trans., 1887, p. 208). The transformation is not one of the oxidation of a hexamethylene compound to a benzenoid compound, for only two hydrogen atoms are removed. Succinosuccinic ester behaves both as a ketone and as a phenol, thereby exhibiting desmotropy; assuming the ketone formula as indicating the constitution, then in Baeyer's equation we have a migration of a hydrogen atom, whereas to bring Ladenburg's formula into line, an oxygen atom must migrate.

The relative merits of the formulae of Kekule, Claus and Dewar were next investigated by means of the reduction products of benzene, it being Baeyer's intention to detect whether double linkages were or were not present in the benzene complex.

To follow Baeyer's results we must explain his nomenclature of the reduced benzene derivatives. He numbers the carbon atoms placed at the corners of a hexagon from 1 to 6, and each side in the same order, so that the carbon atoms 1 and 2 are connected by the side 1, atoms 2 and 3 by the side 2, and so on. A doubly linked pair of atoms is denoted by the sign [DELTA] with the index corresponding to the side; if there are two pairs of double links, then indices corresponding to both sides are employed. Thus [DELTA]^1 denotes a tetrahydro derivative in which the double link occupies the side 1; [DELTA]^{1.3}, a dihydro derivative, the double links being along the sides 1 and 3. Another form of isomerism is occasioned by spatial arrangements, many of the reduced terephthalic acids existing in two stereo-isomeric forms. Baeyer explains this by analogy with fumaric and maleic acids: he assumes the reduced benzene ring to lie in a plane; when both carboxyl groups are on the same side of this plane, the acids, in general, resemble maleic acids, these forms he denotes by [GAMMA]cis-cis, or shortly cis-; when the carboxyl groups are on opposite sides, the acids correspond to fumaric acid, these forms are denoted by [GAMMA]cis-trans, or shortly trans-.

By reducing terephthalic acid with sodium amalgam, care being taken to neutralize the caustic soda simultaneously formed by passing in carbon dioxide, [DELTA]^{2.5} dihydroterephthalic acid is obtained; this results from the splitting of a para-linkage. By boiling with water the [DELTA]^{2.5} acid is converted into the [DELTA]^{1.5} dihydroterephthalic acid. This acid is converted into the [DELTA]^{1.4} acid by soda, and into the [DELTA]^2 tetrahydro acid by reduction. From this acid the [DELTA]^{1.3} dihydro and the [DELTA]^1 tetrahydro acids may be obtained, from both of which the hexahydro acid may be prepared. From these results Baeyer concluded that Claus' formula with three para-linkings cannot possibly be correct, for the [DELTA]^{2.5} dihydroterephthalic acid undoubtedly has two ethylene linkages, since it readily takes up two or four atoms of bromine, and is oxidized in warm aqueous solution by alkaline potassium permanganate. But the formation of the [DELTA]^{2.5} acid as the first reduction product is not fully consistent with Kekule's symbol, for we should then expect the [DELTA]^{1.3} or the [DELTA]^{1.5} acid to be first formed (see also POLYMETHYLENES).

The stronger argument against the ethylenoid linkages demanded by Kekule's formula is provided by the remarkable stability towards oxidizing and reducing agents which characterizes all benzenoid compounds. From the fact that reduction products containing either one or two double linkages behave exactly as unsaturated aliphatic compounds, being readily reduced or oxidized, and combining with the halogen elements and haloid acids, it seems probable that in benzenoid compounds the fourth valencies are symmetrically distributed in such a manner as to induce a peculiar stability in the molecule. Such a configuration was proposed in 1887 by H.E. Armstrong (J.C.S. Trans., 1887, p. 258), and shortly afterwards by Baeyer (Ann., 1888, 245, p. 103). In this formula, the so-called "centric formula," the assumption made is that the fourth valencies are simply directed towards the centre of the ring; nothing further is said about the fourth valencies except that they exert a pressure towards the centre. Claus maintained that Baeyer's view was identical with his own, for as in Baeyer's formula, the fourth valencies have a different function from the peripheral valencies, being united at the centre in a form of potential union.

It is difficult to determine which configuration most accurately explains the observed facts; Kekule's formula undoubtedly explains the synthetical production of benzenoid compounds most satisfactorily, and W. Marckwald (Ann., 1893, 274, p. 331; 1894, 279, p. 14) has supported this formula from considerations based on the syntheses of the quinoline ring. Further researches by Baeyer, and upon various nitrogenous ring systems by E. Bamberger (a strong supporter of the centric formula), have shown that the nature of the substituent groups influences the distribution of the fourth valencies; therefore it may be concluded that in compounds the benzene nucleus appears to be capable of existence in two tautomeric forms, in the sense that each particular derivative possesses a definite constitution. The benzene nucleus presents a remarkable case, which must be considered in the formulation of any complete theory of valency. From a study of the reduction of compounds containing two ethylenic bonds united by a single bond, termed a "conjugated system," E. Thiele suggested a doctrine of "partial valencies," which assumes that in addition to the ordinary valencies, each doubly linked atom has a partial valency, by which the atom first interacts. When applied to benzene, a twofold conjugated system is suggested in which the partial valencies of adjacent atoms neutralize, with the formation of a potential double link. The stability of benzene is ascribed to this conjugation.[14]

Physico-chemical methods.

Physico-chemical properties have also been drawn upon to decide whether double unions are present in the benzene complex; but here the predilections of the observers apparently influence the nature of the conclusions to be drawn from such data. It is well known that singly, doubly and trebly linked carbon atoms affect the physical properties of substances, such as the refractive index, specific volume, and the heat of combustion; and by determining these constants for many substances, fairly definite values can be assigned to these groupings. The general question of the relation of the refractive index to constitution has been especially studied by J.W. Bruhl, who concluded that benzene contained 3 double linkages; whereas, in 1901, Pellini (Gazetta, 31, i. p. 1) calculated that 9 single linkages were present. A similar contradiction apparently exists with regard to the specific volume, for while benzene has a specific volume corresponding to Claus' formula, toluene, or methylbenzene, rather points to Kekule's. The heat of combustion, as first determined by Julius Thomsen, agreed rather better with the presence of nine single unions. His work was repeated on a finer scale by M.P.E. Berthelot of Paris, and F.C.A. Stohmann of Leipzig; and the new data and the conclusions to be drawn from them formed the subject of much discussion, Bruhl endeavouring to show how they supported Kekule's formula, while Thomsen maintained that they demanded the benzene union to have a different heat of combustion from the acetylene union. Thomsen then investigated heats of combustion of various benzenoid hydrocarbons—benzene, naphthalene, anthracene, phenanthrene, &c.—in the crystallized state. It was found that the results were capable of expression by the empirical relation C{a}H{2b} = 104.3b + 49.09m + 105.47n, where C{a}H{2b} denotes the formula of the hydrocarbon, m the number of single carbon linkings and n the number of double linkings, m and n being calculated on the Kekule formulae. But, at the same time, the constants in the above relation are not identical with those in the corresponding relation empirically deduced from observations on fatty hydrocarbons; and we are therefore led to conclude that a benzene union is considerably more stable than an ethylene union.

Mention may be made of the absorption spectrum of benzene. According to W.N. Hartley (J.C.S., 1905, 87, p. 1822), there are six bands in the ultra-violet, while E.C.C. Baly and J.N. Collie (J.C.S., 1905, 87, p. 1332; 1906, 89, p. 524) record seven. These bands are due to molecular oscillations; Hartley suggests the carbon atoms to be rotating and forming alternately single and double linkages, the formation of three double links giving three bands, and of three single links another three; Baly and Collie, on the other hand, suggest the making and breaking of links between adjacent atoms, pointing out that there are seven combinations of one, two and three pairs of carbon atoms in the benzene molecule.

Stereo-chemical Configurations.—Simultaneously with the discussions of Kekule, Ladenburg, Claus, Baeyer and others as to the merits of various plane formulae of the benzene complex, there were published many suggestions with regard to the arrangement of the atoms in space, all of which attempted to explain the number of isomers and the equivalence of the hydrogen atoms. The development of stereo-isomerism at the hands of J. Wislicenus, Le Bel and van 't Hoff has resulted in the introduction of another condition which formulae for the benzene complex must satisfy, viz. that the hydrogen atoms must all lie in one plane. The proof of this statement rests on the fact that if the hydrogen atoms were not co-planar, then substitution derivatives (the substituting groups not containing asymmetric carbon atoms) should exist in enantiomorphic forms, differing in crystal form and in their action on polarized light; such optical antipodes have, however, not yet been separated. Ladenburg's prism formula would give two enantiomorphic ortho-di-substitution derivatives; while forms in which the hydrogen atoms are placed at the corners of a regular octahedron would yield enantiomorphic tri-substitution derivatives.

The octahedral formula discussed by Julius Thomsen (Ber., 1886, 19, p. 2944) consists of the six carbon atoms placed at the corners of a regular octahedron, and connected together by the full lines as shown in (I); a plane projection gives a hexagon with diagonals (II). Reduction to hexamethylene compounds necessitates the disruption of three of the edges of the octahedron, the diagonal linkings remaining intact, or, in the plane projection, three peripheral linkages, the hexamethylene ring assuming the form (III):



In 1888 J.E. Marsh published a paper (Phil. Mag. Ṿ, 26, p. 426) in which he discussed various stereo-chemical representations of the benzene nucleus. (The stereo-chemistry of carbon compounds has led to the spatial representation of a carbon atom as being situated at the centre of a tetrahedron, the four valencies being directed towards the apices; see above, and ISOMERISM.) A form based on Kekule's formula consists in taking three pairs of tetrahedra, each pair having a side in common, and joining them up along the sides of a regular hexagon by means of their apices. This form, afterwards supported by Carl Graebe (Ber., 1902, 35, p. 526; see also Marsh's reply, Journ. Chem. Soc. Trans., 1902, p. 961) shows the proximity of the ortho-positions, but fails to explain the identity of 1.2 and 1.6 compounds. Arrangements connected with Claus' formula are obtained by placing six tetrahedra on the six triangles formed by the diagonals of a plane hexagon. The form in which the tetrahedra are all on one side, afterwards discussed by J. Loschmidt (Monats., 1890, II, p. 28), would not give stereo-isomers; and the arrangement of placing the tetrahedra on alternate sides, a form afterwards developed by W. Vaubel (Journ. Pr. Chem., 1894[2], 49, p. 308), has the advantage of bringing the meta-positions on one side, and the ortho- and para- on opposite sides, thus exhibiting the similarity actually observed between these series of compounds. Marsh also devised a form closely resembling that of Thomsen, inasmuch as the carbon atoms occupied the angles of a regular octahedron, and the diagonal linkages differed in nature from the peripheral, but differing from Thomsen's since rupture of the diagonal and not peripheral bonds accompanied the reduction to hexamethylene.

We may also notice the model devised by H. Sachse (Ber., 1888, 21, 2530; Zeit. fur phys. Chem., II, p. 214; 23, p. 2062). Two parallel triangular faces are removed from a cardboard model of a regular octahedron, and on the remaining six faces tetrahedra are then placed; the hydrogen atoms are at the free angles. This configuration is, according to Sachse, more stable than any other form; no oscillation is possible, the molecule being only able to move as a whole. In 1897, J.N. Collie (Journ. Chem. Soc. Trans., p. 1013) considered in detail an octahedral form, and showed how by means of certain simple rotations of his system the formulae of Kekule and Claus could be obtained as projections. An entirely new device, suggested by B. Konig (Chem. Zeit., 1905, 29, p. 30), assumed the six carbon atoms to occupy six of the corners of a cube, each carbon atom being linked to a hydrogen atom and by single bonds to two neighbouring carbon atoms, the remaining valencies being directed to the unoccupied corners of the cube, three to each, where they are supposed to satisfy each other.

Condensed Nuclei.

Restricting ourselves to compounds resulting from the fusion of benzene rings, we have first to consider naphthalene, C10H8, which consists of two benzene rings having a pair of carbon atoms in common. The next members are the isomers anthracene and phenanthrene, C14H10, formed from three benzene nuclei. Here we shall only discuss the structure of these compounds in the light of the modern benzene theories; reference should be made to the articles NAPHTHALENE, ANTHRACENE and PHENANTHRENE for syntheses, decompositions, &c.

Naphthalene.—Of the earlier suggestions for the constitution of naphthalene we notice the formulae of Wreden (1) and (2), Berthelot and Balls (3), R.A.C.E. Erlenmeyer (4) and Adolf Claus (5).

// // / / / // / // // / / / // / // _CH2 // / / / // / / CH / / -C / / / / / // / / / / / / // / CH / / / / / //

(1) (2) (3) (4)

/ / / / / / / / / / / /

(5)

The first suggestion is quite out of the question. C. Graebe in 1866 (Ann. 149, p. 20) established the symmetry of the naphthalene nucleus, and showed that whichever half of the molecule be oxidized the same phthalic acid results. Therefore formula (2), being unsymmetrical, is impossible. The third formula is based on Dewar's benzene formula, which we have seen to be incorrect. Formula (4) is symmetrical and based on Kekule's formula: it is in full accord with the syntheses and decompositions of the naphthalene nucleus and the number of isomers found. In 1882 Claus suggested a combination of his own and Dewar's benzene formulae. This is obviously unsymmetrical, consisting of an aliphatic and an aromatic nucleus; Claus explained the formation of the same phthalic acid from the oxidation of either nucleus by supposing that if the aromatic group be oxidized, the aliphatic residue assumes the character of a benzene nucleus. Bamberger opposed Claus' formula on the following grounds:—The molecule of naphthalene is symmetrical, since 2.7 dioxynaphthalene is readily esterified by methyl iodide and sulphuric acid to a dimethyl ether; and no more than two mono-substitution derivatives are known. The molecule is aromatic but not benzenoid; however, by the reduction of one half of the molecule, the other assumes a benzenoid character.

If [beta]-naphthylamine and [beta]-naphthol be reduced, tetrahydro products are obtained in which the amino- or oxy-bearing half of the molecule becomes aliphatic in character. The compounds so obtained, alicyclic-[beta]-tetrahydronaphthylamine and alicyclic-[beta]-tetrahydronaphthol, closely resemble [beta]-aminodiethylbenzene, C6H4(C2H5).C2H4NH2, and [beta]-oxydiethylbenzene, C6H4(C2H5).C2H4OH. If [alpha]-naphthylamine and [alpha]-naphthol be reduced, the hydrogen atoms attach themselves to the non-substituted half of the molecule, and the compounds so obtained resemble aminodiethylbenzene, C6H3.NH2(C2H5)2 and oxydiethylbenzene, C6H3.OH(C2H5)2. Bamberger's observations on reduced quinoline derivatives point to the same conclusion, that condensed nuclei are not benzenoid, but possess an individual character, which breaks down, however, when the molecule is reduced.

It remains, therefore, to consider Erlenmeyer's formula and those derived from the centric hypothesis. The former, based on Kekule's symbol for benzene, explains the decompositions and syntheses of the ring, but the character of naphthalene is not in keeping with the presence of five double linkages, although it is more readily acted upon than benzene is. On the centric hypothesis two formulae are possible: (i) due to H.E. Armstrong, and (2) due to E. Bamberger.

/ / / / / / / / - - / X / / / / - - / X / / / / / / / /

(1) (2)

In the first symbol it is assumed that one of the affinities of each of the two central carbon atoms common to the two rings acts into both rings, an assumption involving a somewhat wide departure from all ordinary views as to the manner in which affinity acts. This symbol harmonizes with the fact that the two rings are in complete sympathy, the one responding to every change made in the other. Then, on account of the relatively slight—because divided—influence which would be exercised upon the two rings by the two affinities common to both, the remaining four centric affinities of each ring would presumably be less attracted into the ring than in the case of benzene; consequently they would be more active outwards, and combination would set in more readily. When, as in the formation of naphthalene tetrachloride, for example, the one ring becomes saturated, the other might be expected to assume the normal centric form and become relatively inactive. This is absolutely the case. On the other hand, if substitution be effected in the one ring, and the affinities in that ring become attracted inwards, as apparently happens in the case of benzene, the adjoining ring should become relatively more active because the common affinities would act less into it. Hence, unless the radical introduced be one which exercises a special attractive influence, substitution should take place in preference in the previously unsubstituted ring. In practice this usually occurs; for example, on further bromination, [alpha]-bromonaphthalene yields a mixture of the (1.4) and (1.5) dibromonaphthalenes; and when nitronaphthalene is either brominated, or nitrated or sulphonated, the action is practically confined to the second ring. The centric formula proposed by Bamberger represents naphthalene as formed by the fusion of two benzene rings, this indicates that it is a monocyclic composed of ten atoms of carbon. The formula has the advantage that it may be constructed from tetrahedral models of the carbon atom; but it involves the assumption that the molecule has within it a mechanism, equivalent in a measure to a system of railway points, which can readily close up and pass into that characteristic of benzene.

Anthracene and Phenanthrene.—These isomeric hydrocarbons, of the formula C14H10, are to be regarded as formed by the fusion of three benzenoid rings as represented by the symbols:—

/ / / / / / / / / / / / / / / / / / / / / /

Anthracene Phenanthrene

In both cases the medial ring is most readily attacked; and various formulae have been devised which are claimed by their authors to represent this and other facts. According to Armstrong, anthracene behaves unsymmetrically towards substituents, and hence one lateral ring differs from the other; he represents the molecule as consisting of one centric ring, the remaining medial and lateral ring being ethenoid. Bamberger, on the other hand, extends his views on benzene and naphthalene and assumes the molecule to be (1). For general purposes, however, the symbol (2), in which the lateral rings are benzenoid and the medial ring fatty, represents quite adequately the syntheses, decompositions, and behaviour of anthracene.

/ / / / / / / / / / / / X X / / / / / / X X / / / / / / / / / / / /

(1) (2)

Phenanthrene is regarded by Armstrong as represented by (3), the lateral rings being benzenoid, and the medial ring fatty; Bamberger, however, regards it as (4), the molecule being entirely aromatic. An interesting observation by Baeyer, viz. that stilbene, C6H5.CH:CH.C6H5, is very readily oxidized, while phenanthrene is not, supports, in some measure, the views of Bamberger.

/—— / / / / / / / / / / / / / / / / / / / // //

(3) (4)

Heterocyclic Compounds.

During recent years an immense number of ringed or cyclic compounds have been discovered, which exhibit individual characters more closely resembling benzene, naphthalene, &c. than purely aliphatic substances, inasmuch as in general they contain double linkages, yet withstand oxidation, and behave as nuclei, forming derivatives in much the same way as benzene. By reduction, the double linkages become saturated, and compounds result which stand in much about the same relation to the original nucleus as hexamethylene does to benzene. In general, therefore, it may be considered that the double linkages are not of exactly the same nature as the double linkage present in ethylene and ethylenoid compounds, but that they are analogous to the potential valencies of benzene. The centric hypothesis has been applied to these rings by Bamberger and others; but as in the previous rings considered, the ordinary representation with double and single linkages generally represents the syntheses, decompositions, &c.; exceptions, however, are known where it is necessary to assume an oscillation of the double linkage. Five- and six-membered rings are the most stable and important, the last-named group resulting from the polymerization of many substances; three- and four-membered rings are formed with difficulty, and are easily ruptured; rings containing seven or more members are generally unstable, and are relatively little known. The elements which go to form heterocyclic rings, in addition to carbon, are oxygen, sulphur, selenium and nitrogen. It is remarkable that sulphur can replace two methine or CH groups with the production of compounds greatly resembling, the original one. Thus benzene, (CH)6, gives thiophene, (CH)4S, from which it is difficultly distinguished; pyridine, (CH)5N, gives thiazole, (CH)3.N.S, which is a very similar substance; naphthalene gives thionaphthen, C8H6S, with which it shows great analogies, especially in the derivatives. Similarly a CH group may be replaced by a nitrogen atom with the production of compounds of similar stability; thus benzene gives pyridine, naphthalene gives quinoline and isoquinoline; anthracene gives acridine and [alpha] and [beta] anthrapyridines. Similarly, two or more methine groups may be replaced by the same number of nitrogen atoms with the formation of rings of considerable stability.

Most of the simple ring systems which contain two adjacent carbon atoms may suffer fusion with any other ring (also containing two adjacent carbon atoms) with the production of nuclei of greater complexity. Such condensed nuclei are, in many cases, more readily obtained than the parent nucleus. The more important types are derived from aromatic nuclei, benzene, naphthalene, &c.; the ortho-di-derivatives of the first named, lending themselves particularly to the formation of condensed nuclei. Thus ortho-phenylene diamine yields the following products:—

/ N / N / N / / / / / / CH N S / / / / / / / / / / NH , / NH , / N ,

Benzimidazole Azimidobenzene Benzpiazthiole

/ NH / NH / N / / / / / / CH >CO or C.OH / / CH / / / / / // / NH / NH , / N

/ Benzimidazolone Quinoxaline

In some cases oxidation of condensed benzenoid-heterocyclic nuclei results in the rupture of the heterocyclic ring with the formation of a benzene dicarboxylic acid; but if the aromatic nucleus be weakened by the introduction of an amino group, then it is the benzenoid nucleus which is destroyed and a dicarboxylic acid of the heterocyclic ring system obtained.

Heterocyclic rings may be systematically surveyed from two aspects: (1) by arranging the rings with similar hetero-atoms according to the increasing number of carbon atoms, the so-called "homologous series"; or (2) by first dividing the ring systems according to the number of members constituting the ring, and then classifying these groups according to the nature of the hetero-atoms, the so-called "isologous series." The second method possesses greater advantages, for rings of approximate stability come in one group, and, consequently, their derivatives may be expected to exhibit considerable analogies.

As a useful preliminary it is convenient to divide heterocyclic ring systems into two leading groups: (1) systems resulting from simple internal dehydration (or similar condensations) of saturated aliphatic compounds—such compounds are: the internal anhydrides or cyclic ethers of the glycols and thioglycols (ethylene oxide, &c.); the cyclic alkyleneimides resulting from the splitting off of ammonia between the amino groups of diamino-paraffins (pyrrolidine, piperazine, &c.); the cyclic esters of oxycarboxylic acids (lactones, lactides); the internal anhydrides of aminocarboxylic acids (lactams, betaines); cyclic derivatives of dicarboxylic acids (anhydrides, imides, alkylen-esters, alkylen-amides, &c.). These compounds retain their aliphatic nature, and are best classified with open-chain compounds, into which, in general, they are readily converted. (2) Systems which are generally unsaturated compounds, often of considerable stability, and behave as nuclei; these compounds constitute a well-individualized class exhibiting closer affinities to benzenoid substances than to the open-chain series.

The transition between the two classes as differentiated above may be illustrated by the following cyclic compounds, each of which contains a ring composed of four carbon atoms and one oxygen atom:

CH2.CH2 CH2.CO CH2.CO CH.CO CHCH O O O O O CH2.CH2/ CH2.CH2/ CH2.CO/ CH.CO/ CHCH/

Tetramethylene Butyrolactone. Succinic Maleic Furfurane. oxide. anhydride. anhydride.

The first four substances are readily formed from, and converted into, the corresponding dihydroxy open-chain compound; these substances are truly aliphatic in character. The fifth compound, on the other hand, does not behave as an unsaturated aliphatic compound, but its deportment is that of a nucleus, many substitution derivatives being capable of synthesis. Reduction, however, converts it into an aliphatic compound. This is comparable with the reduction of the benzene nucleus into hexamethylene, a substance of an aliphatic character.

True ring systems, which possess the characters of organic nuclei, do not come into existence in three-and four-membered rings, their first appearance being in penta-atomic rings. The three primary members are furfurane, thiophene and pyrrol, each of which contains four methine or CH groups, and an oxygen, sulphur and imido (NH) member respectively; a series of compounds containing selenium is also known. The formulae of these substances are:

CHCH CHCH CHCH CHCH O S Se NH CHCH/ CHCH/ CHCH/ CHCH/

Furfurane. Thiophene. Selenophene. Pyrrol.

By substituting one or more CH groups in these compounds by nitrogen atoms, ring-systems, collectively known as azoles, result. Obviously, isomeric ring-systems are possible, since the carbon atoms in the original rings are not all of equal value. Thus furfurane yields the following rings by the introduction of one and two nitrogen atoms:

CH==N N==CH N==N O O O CHCH/ CHCH/ CHCH/

Isoxazole. Oxazole. Diazo-oxides.

CHN NCH NCH O O O CHN/ CHN/ NCH/

Furazane. Azoximes. Oxybiazole.

Thiophene yields a similar series: isothiazole (only known as the condensed ring, isobenzothiazole), thiazole, diazosulphides, piazthioles, azosulphimes and thiobiazole (the formulae are easily derived from the preceding series by replacing oxygen by sulphur). Thiophene also gives rise to triazsulphole, three nitrogen atoms being introduced. Selenophene gives the series: selenazole, diazoselenide and piaselenole, corresponding to oxazole, diazo-oxides and furazane. Pyrrol yields an analogous series: pyrazole, imidazole or glyoxaline, azimide or osotriazole, triazole and tetrazole:

CH==N N==CH N==N NH NH NH CHCH/ CHCH/ CHCH/

Pyrazole. Imidazole. Azimide.

N=CH N==N NH NH N=CH/ N=CH/

Triazole. Tetrazole.

Six-membered ring systems can be referred back, in a manner similar to the above, to pyrone, penthiophene and pyridine, the substances containing a ring of five carbon atoms, and an oxygen, sulphur and nitrogen atom respectively. As before, only true ring nuclei, and not internal anhydrides of aliphatic compounds, will be mentioned. From the pyrone ring the following series of compounds are derived (for brevity, the hydrogen atoms are not printed):

C C C N C / / / / / C / C C / C C / N C / C C / C C / C C / N C / C C / C N / N / / / / / O O O O O

Pyrone Ortho-oxazine Meta-oxazine Paroxazine Azoxazine or pentoxazoline

Penthiophene gives, by a similar introduction of nitrogen atoms, penthiazoline, corresponding to meta-oxazine, and para-thiazine, corresponding to paroxazine (para-oxazine). Pyridine gives origin to: pyridazine or ortho-diazine, pyrimidine or meta-diazine, pyrazine or para-diazine, osotriazine, unsymmetrical triazine, symmetrical triazine, osotetrazone and tetrazine. The skeletons of these types are (the carbon atoms are omitted for brevity):

N / / / / / / / N / / / N / / / / / / N N N N

Pyridine Pyridazine Pyrimidine Pyrazine

/ / / / N / N / N N / N / N / / / / N N N / Triazines

N N / / / N N / / N / N / / N N

Osotetrazone Tetrazone

We have previously referred to the condensation of heterocyclic ring systems containing two vicinal carbon atoms with benzene, naphthalene and other nuclei. The more important nuclei of this type have received special and non-systematic names; when this is not the case, such terms as phen-, benzo-, naphtho- are prefixed to the name of the heterocyclic ring. One or two benzene nuclei may suffer condensation with the furfurane, thiophene and pyrrol rings, the common carbon atoms being vicinal to the hetero-atom. The mono-benzo-derivatives are coumarone, benzothiophene and indole; the dibenzo-derivatives are diphenylene oxide, dibenzothiophene or diphenylene sulphide, and carbazole. Typical formulae are (R denoting O, S or NH):

/ / / / / / / / / / / / / , / / / R R

Isomers are possible, for the condensation may be effected on the two carbon atoms symmetrically placed to the hetero-atom; these isomers, however, are more of the nature of internal anhydrides. Benz-oxazoles and -thiazoles have been prepared, benz-isoxazoles are known as indoxazenes; benzo-pyrazoles occur in two structural forms, named indazoles and isindazoles. Derivatives of osotriazol also exist in two forms—azimides and pseudo-azimides.

Proceeding to the six-membered hetero-atomic rings, the benzo-, dibenzo-and naphtho-derivatives are frequently of great commercial and scientific importance, [alpha]-pyrone condenses with the benzene ring to form coumarin and isocoumarin; benzo-[gamma]-pyrone constitutes the nucleus of several vegetable colouring matters (chrysin, fisetin, quercetin, &c., which are derivatives of flavone or phenyl benzo-[gamma]-pyrone); dibenzo-[gamma]-pyrone is known as xanthone; related to this substance are fluorane (and fluorescein), fluorone, fluorime, pyronine, &c. The pyridine ring condenses with the benzene ring to form quinoline and isoquinoline; acridine and phenanthridine are dibenzo-pyridines; naphthalene gives rise to [alpha]- and [beta]-naphthoquinolines and the anthrapyridines; anthracene gives anthraquinoline; while two pyridine nuclei connected by an intermediate benzene nucleus give the phenanthrolines. Naphthyridines and naphthinolines result from the condensation of two pryridine and two quinoline nuclei respectively; and quino-quinolines are unsymmetrical naphthyridine nuclei condensed with a benzene nucleus. Benzo-orthoxazines, -metoxazines and -paroxazines are known: dibenzoparoxazine or phenoxazine is the parent of a valuable series of dyestuffs; dibenzoparathiazine or thiodiphenylamine is important from the same aspect. Benzo-ortho-diazines exist in two structural forms, cinnolin and phthalazine; benzo-meta-diazines are known as quinazolines; benzo-para-diazines are termed quinoxalines; the dibenzo-compounds are named phenazines, this last group including many valuable dyestuffs—indulines, safranines, &c. In addition to the types of compounds enumerated above we may also notice purin, tropine and the terpenes.

V. ANALYTICAL CHEMISTRY

This branch of chemistry has for its province the determination of the constituents of a chemical compound or of a mixture of compounds. Such a determination is qualitative, the constituent being only detected or proved to be present, or quantitative, in which the amount present is ascertained. The methods of chemical analysis may be classified according to the type of reaction: (1) dry or blowpipe analysis, which consists in an examination of the substance in the dry condition; this includes such tests as ignition in a tube, ignition on charcoal in the blowpipe flame, fusion with borax, microcosmic salt or fluxes, and flame colorations (in quantitative work the dry methods are sometimes termed "dry assaying"); (2) wet analysis, in which a solution of the substance is treated with reagents which produce specific reactions when certain elements or groups of elements are present. In quantitative analysis the methods can be subdivided into: (a) gravimetric, in which the constituent is precipitated either as a definite insoluble compound by the addition of certain reagents, or electrolytically, by the passage of an electric current; (b) volumetric, in which the volume of a reagent of a known strength which produces a certain definite reaction is measured; (c) colorimetric, in which the solution has a particular tint, which can be compared with solutions of known strengths.

Historical.—The germs of analytical chemistry are to be found in the writings of the pharmacists and chemists of the iatrochemical period. The importance of ascertaining the proximate composition of bodies was clearly realized by Otto Tachenius; but the first systematic investigator was Robert Boyle, to whom we owe the introduction of the term analysis. Boyle recognized many reagents which gave precipitates with certain solutions: he detected sulphuric and hydrochloric acids by the white precipitates formed with calcium chloride and silver nitrate respectively; ammonia by the white cloud formed with the vapours of nitric or hydrochloric acids; and copper by the deep blue solution formed by a solution of ammonia. Of great importance is his introduction of vegetable juices (the so-called indicators, q.v.) to detect acids and bases. During the phlogistic period, the detection of the constituents of compounds was considerably developed. Of the principal workers in this field we may notice Friedrich Hoffmann, Andreas Sigismund Marggraf (who detected iron by its reaction with potassium ferrocyanide, and potassium and sodium by their flame colorations), and especially Carl Scheele and Torbern Olof Bergman. Scheele enriched the knowledge of chemistry by an immense number of facts, but he did not possess the spirit of working systematically as Bergman did. Bergman laid the foundations of systematic qualitative analysis, and devised methods by which the metals may be separated into groups according to their behaviour with certain reagents. This subdivision, which is of paramount importance in the analysis of minerals, was subsequently developed by Wilhelm August Lampadius in his Handbuch zur chemischen Analyse der Mineralien (1801) and by John Friedrich A. Gottling in his Praktische Anleitung zur prufenden und zurlegenden Chemie (1802).

The introduction of the blowpipe into dry qualitative analysis by Axel Fredrik Cronstedt marks an important innovation. The rapidity of the method, and the accurate results which it gave in the hands of a practised experimenter, led to its systematization by Jons Jakob Berzelius and Johann Friedrich Ludwig Hausmann, and in more recent times by K.F. Plattner, whose treatise Die Probirkunst mit dem Lothrohr is a standard work on the subject. Another type of dry reaction, namely, the flame coloration, had been the subject of isolated notices, as, for example, the violet flame of potassium and the orange flame of sodium observed by Marggraf and Scheele, but a systematic account was wanting until Cartmell took the subject up. His results (Phil. Mag. 16, p. 382) were afterwards perfected by Robert Wilhelm Bunsen and Gustav Merz. Closely related to the flame-colorations, we have to notice the great services rendered by the spectroscope to the detection of elements. Rubidium, caesium, thallium, indium and gallium were first discovered by means of this instrument; the study of the rare earths is greatly facilitated, and the composition of the heavenly bodies alone determinable by it.

Quantitative chemistry had been all but neglected before the time of Lavoisier, for although a few chemists such as Tachenius, Bergman and others had realized the advantages which would accrue from a knowledge of the composition of bodies by weight, and had laid down the lines upon which such determinations should proceed, the experimental difficulties in making accurate observations were enormous, and little progress could be made until the procedure was more accurately determined. Martin Heinrich Klaproth showed the necessity for igniting precipitates before weighing them, if they were not decomposed by this process; and he worked largely with Louis Nicolas Vauquelin in perfecting the analysis of minerals. K.F. Wenzel and J.B. Richter contributed to the knowledge of the quantitative composition of salts. Anton Laurent Lavoisier, however, must be considered as the first great exponent of this branch of chemistry. He realized that the composition by weight of chemical compounds was of the greatest moment if chemistry were to advance. His fame rests upon his exposition of the principles necessary to chemistry as a science, but of his contributions to analytical inorganic chemistry little can be said. He applied himself more particularly to the oxygen compounds, and determined with a fair degree of accuracy the ratio of carbon to oxygen in carbon dioxide, but his values for the ratio of hydrogen to oxygen in water, and of phosphorus to oxygen in phosphoric acid, are only approximate; he introduced no new methods either for the estimation or separation of the metals. The next advance was made by Joseph Louis Proust, whose investigations led to a clear grasp of the law of constant proportions. The formulation of the atomic theory by John Dalton gave a fresh impetus to the development of quantitative analysis; and the determination of combining or equivalent weights by Berzelius led to the perfecting of the methods of gravimetric analysis. Experimental conditions were thoroughly worked out; the necessity of working with hot or cold solutions was clearly emphasized; and the employment of small quantities of substances instead of the large amounts recommended by Klaproth was shown by him to give more consistent results.

Since the time of Berzelius many experimenters have entered the lists, and introduced developments which we have not space to mention. We may, however, notice Heinrich Rose[15] and Friedrich Wohler,[16] who, having worked up the results of their teacher Berzelius, and combined them with their own valuable observations, exerted great influence on the progress of analytical chemistry by publishing works which contained admirable accounts of the then known methods of analysis. To K.R. Fresenius, the founder of the Zeitschrift fur analytische Chemie (1862), we are particularly indebted for perfecting and systematizing the various methods of analytical chemistry. By strengthening the older methods, and devising new ones, he exerted an influence which can never be overestimated. His text-books on the subject, of which the Qualitative appeared in 1841, and the Quantitative in 1846, have a world-wide reputation, and have passed through several editions.

The quantitative precipitation of metals by the electric current, although known to Michael Faraday, was not applied to analytical chemistry until O. Wolcott Gibbs worked out the electrolytic separation of copper in 1865. Since then the subject has been extensively studied, more particularly by Alexander Classen, who has summarized the methods and results in his Quantitative Chemical Analysis by Electrolysis (1903). The ever-increasing importance of the electric current in metallurgy and chemical manufactures is making this method of great importance, and in some cases it has partially, if not wholly, superseded the older methods.

Volumetric analysis, possessing as it does many advantages over the gravimetric methods, has of late years been extensively developed. Gay Lussac may be regarded as the founder of the method, although rough applications had been previously made by F.A.H. Descroizilles and L.N. Vauquelin. Chlorimetry (1824), alkalimetry (1828), and the volumetric determination of silver and chlorine (1832) were worked out by Gay Lussac; but although the advantages of the method were patent, it received recognition very slowly. The application of potassium permanganate to the estimation of iron by E. Margueritte in 1846, and of iodine and sulphurous acid to the estimation of copper and many other substances by Robert Wilhelm Bunsen, marks an epoch in the early history of volumetric analysis. Since then it has been rapidly developed, particularly by Karl Friedrich Mohr and J. Volhard, and these methods rank side by side in value with the older and more tedious gravimetric methods.

The detection of carbon and hydrogen in organic compounds by the formation of carbon dioxide and water when they are burned was first correctly understood by Lavoisier, and as he had determined the carbon and hydrogen content of these two substances he was able to devise methods by which carbon and hydrogen in organic compounds could be estimated. In his earlier experiments he burned the substance in a known volume of oxygen, and by measuring the residual gas determined the carbon and hydrogen. For substances of a difficultly combustible nature he adopted the method in common use to-day, viz. to mix the substance with an oxidizing agent—mercuric oxide, lead dioxide, and afterwards copper oxide—and absorb the carbon dioxide in potash solution. This method has been improved, especially by Justus v. Liebig; and certain others based on a different procedure have been suggested. The estimation of nitrogen was first worked out in 1830 by Jean Baptiste Dumas, and different processes have been proposed by Will and F. Varrentrapp, J. Kjeldahl and others. Methods for the estimation of the halogens and sulphur were worked out by L. Carius (see below, S Organic Analysis).

Only a reference can be made in this summary to the many fields in which analytical chemistry has been developed. Progress in forensic chemistry was only possible after the reactions of poisons had been systematized; a subject which has been worked out by many investigators, of whom we notice K.R. Fresenius, J. and R. Otto, and J.S. Stas. Industrial chemistry makes many claims upon the chemist, for it is necessary to determine the purity of a product before it can be valued. This has led to the estimation of sugar by means of the polarimeter, and of the calorific power of fuels, and the valuation of ores and metals, of coal-tar dyes, and almost all trade products.

The passing of the Food and Drug Acts (1875-1899) in England, and the existence of similar adulteration acts in other countries, have occasioned great progress in the analysis of foods, drugs, &c. For further information on this branch of analytical chemistry, see ADULTERATION.

There exists no branch of technical chemistry, hygiene or pharmacy from which the analytical chemist can be spared, since it is only by a continual development of his art that we can hope to be certain of the purity of any preparation. In England this branch of chemistry is especially cared for by the Institute of Chemistry, which, since its foundation in 1877, has done much for the training of analytical chemists.

In the preceding sketch we have given a necessarily brief account of the historical development of analytical chemistry in its main branches. We shall now treat the different methods in more detail. It must be mentioned here that the reactions of any particular substance are given under its own heading, and in this article we shall only collate the various operations and outline the general procedure. The limits of space prevent any systematic account of the separation of the rare metals, the alkaloids, and other classes of organic compounds, but sources where these matters may be found are given in the list of references.

Qualitative Inorganic Analysis.

Dry methods.

The dry examination of a substance comprises several operations, which may yield definite results if no disturbing element is present; but it is imperative that any inference should be confirmed by other methods.

1. Heat the substance in a hard glass tube. Note whether any moisture condenses on the cooler parts of the tube, a gas is evolved, a sublimate formed, or the substance changes colour.

Moisture is evolved from substances containing water of crystallization or decomposed hydrates. If it possesses an alkaline or acid reaction, it must be tested in the first case for ammonia, and in the second case for a volatile acid, such as sulphuric, nitric, hydrochloric, &c.

Any evolved gas must be examined. Oxygen, recognized by its power of igniting a glowing splinter, results from the decomposition of oxides of the noble metals, peroxides, chlorates, nitrates and other highly oxygenized salts. Sulphur dioxide, recognized by its smell and acid reaction, results from the ignition of certain sulphites, sulphates, or a mixture of a sulphate with a sulphide. Nitrogen oxides, recognized by their odour and brown-red colour, result from the decomposition of nitrates. Carbon dioxide, recognized by turning lime-water milky, indicates decomposable carbonates or oxalates. Chlorine, bromine, and iodine, each recognizable by its colour and odour, result from decomposable haloids; iodine forms also a black sublimate. Cyanogen and hydrocyanic acid, recognizable by their odour, indicate decomposable cyanides. Sulphuretted hydrogen, recognized by its odour, results from sulphides containing water, and hydrosulphides. Ammonia, recognizable by its odour and alkaline reaction, indicates ammoniacal salts or cyanides containing water.

A sublimate may be formed of: sulphur—reddish-brown drops, cooling to a yellow to brown solid, from sulphides or mixtures; iodine—violet vapour, black sublimate, from iodides, iodic acid, or mixtures; mercury and its compounds—metallic mercury forms minute globules, mercuric sulphide is black and becomes red on rubbing, mercuric chloride fuses before subliming, mercurous chloride does not fuse, mercuric iodide gives a yellow sublimate; arsenic and its compounds—metallic arsenic gives a grey mirror, arsenious oxide forms white shining crystals, arsenic sulphides give reddish-yellow sublimates which turn yellow on cooling; antimony oxide fuses and gives a yellow acicular sublimate; lead chloride forms a white sublimate after long and intense heating.

If the substance does not melt but changes colour, we may have present: zinc oxide—from white to yellow, becoming white on cooling; stannic oxide—white to yellowish brown, dirty white on cooling; lead oxide—from white or yellowish-red to brownish-red, yellow on cooling; bismuth oxide—from white or pale yellow to orange-yellow or reddish-brown, pale yellow on cooling; manganese oxide—from white or yellowish white to dark brown, remaining dark brown on cooling (if it changes on cooling to a bright reddish-brown, it indicates cadmium oxide); copper oxide—from bright blue or green to black; ferrous oxide—from greyish-white to black; ferric oxide—from brownish-red to black, brownish-red on cooling; potassium chromate—yellow to dark orange, fusing at a red heat.

2. Heat the substance on a piece of charcoal in the reducing flame of the blowpipe.

([alpha]) The substance may fuse and be absorbed by the charcoal; this indicates more particularly the alkaline metals.

([beta]) An infusible white residue may be obtained, which may denote barium, strontium, calcium, magnesium, aluminium or zinc. The first three give characteristic flame colorations (see below); the last three, when moistened with cobalt nitrate and re-ignited, give coloured masses; aluminium (or silica) gives a brilliant blue; zinc gives a green; whilst magnesium phosphates or arsenate (and to a less degree the phosphates of the alkaline earths) give a violet mass.

A metallic globule with or without an incrustation may be obtained. Gold and copper salts give a metallic bead without an incrustation. If the incrustation be white and readily volatile, arsenic is present, if more difficultly volatile and beads are present, antimony; zinc gives an incrustation yellow whilst hot, white on cooling, and volatilized with difficulty; tin gives a pale yellow incrustation, which becomes white on cooling, and does not volatilize in either the reducing or oxidizing flames; lead gives a lemon-yellow incrustation turning sulphur-yellow on cooling, together with metallic malleable beads; bismuth gives metallic globules and a dark orange-yellow incrustation, which becomes lemon-yellow on cooling; cadmium gives a reddish-brown incrustation, which is removed without leaving a gleam by heating in the reducing flame; silver gives white metallic globules and a dark-red incrustation.

3. Heat the substance with a bead of microcosmic salt or borax on a platinum wire in the oxidizing flame.

([alpha]) The substance dissolves readily and in quantity, forming a bead which is clear when hot. If the bead is coloured we may have present: cobalt, blue to violet; copper, green, blue on cooling; in the reducing flame, red when cold; chromium, green, unaltered in the reducing flame; iron, brownish-red, light-yellow or colourless on cooling; in the reducing flame, red while hot, yellow on cooling, greenish when cold; nickel, reddish to brownish-red, yellow to reddish-yellow or colourless on cooling, unaltered in the reducing flame; bismuth, yellowish-brown, light-yellow or colourless on cooling; in the reducing flame, almost colourless, blackish-grey when cold; silver, light yellowish to opal, somewhat opaque when cold; whitish-grey in the reducing flame; manganese, amethyst red, colourless in the reducing flame. If the hot bead is colourless and remains clear on cooling, we may suspect the presence of antimony, aluminium, zinc, cadmium, lead, calcium and magnesium. When present in sufficient quantity the five last-named give enamel-white beads; lead oxide in excess gives a yellowish bead. If the hot colourless bead becomes enamel-white on cooling even when minute quantities of the substances are employed, we may infer the presence of barium or strontium.

([beta]) The substance dissolves slowly and in small quantity, and forms a colourless bead which remains so on cooling. Either silica or tin may be present. If silica be present, it gives the iron bead when heated with a little ferric oxide; if tin is present there is no change. Certain substances, such as the precious metals, are quite insoluble in the bead, but float about in it.

4. Hold a small portion of the substance moistened with hydrochloric acid on a clean platinum wire in the fusion zone of the Bunsen burner, and note any colour imparted to the flame.

Potassium gives a blue-violet flame which may be masked by the colorations due to sodium, calcium and other elements. By viewing the flame through an indigo prism it appears sky-blue, violet and ultimately crimson, as the thickness of the prism is increased. Other elements do not interfere with this method. Sodium gives an intense and persistent yellow flame; lithium gives a carmine coloration, and may be identified in the presence of sodium by viewing through a cobalt glass or indigo prism; from potassium it may be distinguished by its redder colour; barium gives a yellowish-green flame, which appears bluish-green when viewed through green glass; strontium gives a crimson flame which appears purple or rose when viewed through blue glass; calcium gives an orange-red colour which appears finch-green through green glass; indium gives a characteristic bluish-violet flame; copper gives an intense emerald-green coloration.

5. Film Reactions.—These reactions are practised in the following manner:—A thread of asbestos is moistened and then dipped in the substance to be tested; it is then placed in the luminous point of the Bunsen flame, and a small porcelain basin containing cold water placed immediately over the asbestos. The formation of a film is noted. The operation is repeated with the thread in the oxidizing flame.

Any film formed in the first case is metallic, in the second it is the oxide. The metallic film is tested with 20% nitric acid and with bleaching-powder solution. Arsenic is insoluble in the acid, but immediately dissolves in the bleaching-powder. The black films of antimony and bismuth and the grey mottled film of mercury are slowly soluble in the acid, and untouched by bleaching-powder. The black films of tin, lead and cadmium dissolve at once in the acid, the lead film being also soluble in bleaching-powder. The oxide films of antimony, arsenic, tin and bismuth are white, that of bismuth slightly yellowish; lead yields a very pale yellow film, and cadmium a brown one; mercury yields no oxide film. The oxide films (the metallic one in the case of mercury) are tested with hydriodic acid, and with ammonium sulphide, and from the changes produced the film can be determined (see F.M. Perkin, Qualitative Chemical Analysis, 1905).

Wet methods.

Having completed the dry analysis we may now pass on to the wet and more accurate investigation. It is first necessary to get the substance into solution. Small portions should be successively tested with water, dilute hydrochloric acid, dilute nitric acid, strong hydrochloric acid, and a mixture of hydrochloric and nitric acids, first in the cold and then with warming. Certain substances are insoluble in all these reagents, and other methods, such as the fusion with sodium carbonate and potassium nitrate, and subsequent treatment with an acid, must be employed. Some of these insoluble compounds can be detected by their colour and particular reactions. For further information on this subject, we refer the readers to Fresenius's Qualitative Analysis.

The procedure for the detection of metals in solution consists of first separating them into groups and then examining each group separately. For this purpose the cold solution is treated with hydrochloric acid, which precipitates lead, silver and mercurous salts as chlorides. The solution is filtered and treated with an excess of sulphuretted hydrogen, either in solution or by passing in the gas; this precipitates mercury (mercuric), any lead left over from the first group, copper, bismuth, cadmium, arsenic, antimony and tin as sulphides. The solution is filtered off, boiled till free of sulphuretted hydrogen, and ammonium chloride and ammonia added. If phosphoric acid is absent, aluminium, chromium and ferric hydrates are precipitated. If, however, phosphoric acid is present in the original substance, we may here obtain a precipitate of the phosphates of the remaining metals, together with aluminium, chromium and ferric hydrates. In this case, the precipitate is dissolved in as little as possible hydrochloric acid and boiled with ammonium acetate, acetic acid and ferric chloride. The phosphates of aluminium, chromium and iron are precipitated, and the solution contains the same metals as if phosphoric acid had been absent. To the filtrate from the aluminium, iron and chromium precipitate, ammonia and ammonium sulphide are added; the precipitate may contain nickel, cobalt, zinc and manganese sulphides. Ammonium carbonate is added to the filtrate; this precipitates calcium, strontium and barium. The solution contains magnesium, sodium and potassium, which are separately distinguished by the methods given under their own headings.

We now proceed with the examination of the various group precipitates. The white precipitate formed by cold hydrochloric acid is boiled with water, and the solution filtered while hot. Any lead chloride dissolves, and may be identified by the yellow precipitate formed with potassium chromate. To the residue add ammonia, shake, then filter. Silver chloride goes into solution, and may be precipitated by dilute nitric acid. The residue, which is black in colour, consists of mercuroso-ammonium chloride, in which mercury can be confirmed by its ordinary tests.

The precipitate formed by sulphuretted hydrogen may contain the black mercuric, lead, and copper sulphides, dark-brown bismuth sulphide, yellow cadmium and arsenious sulphides, orange-red antimony sulphide, brown stannous sulphide, dull-yellow stannic sulphide, and whitish sulphur, the last resulting from the oxidation of sulphuretted hydrogen by ferric salts, chromates, &c. Warming with ammonium sulphide dissolves out the arsenic, antimony and tin salts, which are reprecipitated by the addition of hydrochloric acid to the ammonium sulphide solution. The precipitate is shaken with ammonium carbonate, which dissolves the arsenic. Filter and confirm arsenic in the solution by its particular tests. Dissolve the residue in hydrochloric acid and test separately for antimony and tin. The residue from the ammonium sulphide solution is warmed with dilute nitric acid. Any residue consists of black mercuric sulphide (and possibly white lead sulphate), in which mercury is confirmed by its usual tests. The solution is evaporated with a little sulphuric acid and well cooled. The white precipitate consists of lead sulphate. To the filtrate add ammonia in excess; a white precipitate indicates bismuth; if the solution be blue, copper is present. Filter from the bismuth hydrate, and if copper is present, add potassium cyanide till the colour is destroyed, then pass sulphuretted hydrogen, and cadmium is precipitated as the yellow sulphide. If copper is absent, then sulphuretted hydrogen can be passed directly into the solution.

The next group precipitate may contain the white gelatinous aluminium hydroxide, the greenish chromium hydroxide, reddish ferric hydroxide, and possibly zinc and manganese hydroxides. Treatment with casutic soda dissolves out aluminium hydroxide, which is reprecipitated by the addition of ammonium chloride. The remaining metals are tested for separately.

The next group may contain black nickel and cobalt sulphides, flesh-coloured manganese sulphide, and white zinc sulphide. The last two are dissolved out by cold, very dilute hydrochloric acid, and the residue is tested for nickel and cobalt. The solution is boiled till free from sulphuretted hydrogen and treated with excess of sodium hydrate. A white precipitate rapidly turning brown indicates manganese. The solution with ammonium sulphide gives a white precipitate of zinc sulphide.

The next group may contain the white calcium, barium and strontium carbonates. The flame coloration (see above) may give information as to which elements are present. The carbonates are dissolved in hydrochloric acid, and calcium sulphate solution is added to a portion of the solution. An immediate precipitate indicates barium; a precipitate on standing indicates strontium. If barium is present, the solution of the carbonates in hydrochloric acid is evaporated and digested with strong alcohol for some time; barium chloride, which is nearly insoluble in alcohol, is thus separated, the remainder being precipitated by a few drops of hydrofluosilicic acid, and may be confirmed by the ordinary tests. The solution free from barium is treated with ammonia and ammonium sulphate, which precipitates strontium, and the calcium in the solution may be identified by the white precipitate with ammonium oxalate.