SELL : STUDIES ON CITRAZINIC: ACID. 1447 XCIII. --Studies on Citmziwic: acid. Part IV. By WILLIAM JAMES SELL, M.A., F.I.C. Idroduct ion. MANY years since, it was shown by Tiemann and Reimer (Bey., 1876, 9, 423, 824) that by the interaction of sodium hjdroxide and chloro- form with phenol a mixture ot' the sodium derivatives of salicyl and paroxybeuzoic aldehydes was produced. Later, Tiemann and Lewy (Be?.., 1877, 10, 2216) applied the same reaction to dihydroxy- benzene derivatives, notably to resorcinol, fi-om which they obtained both a monaldehyde and a dialdehyde. This method, as is well known, has since been recognised as a general one for introducing the aldehyde group into the liydroxyl derivatives of benzenoid hydrocarbons. Up t o the present time, however, no attempt seems to have been made to apply this reaction to derivatives of pyridine, although these compounds have many interesting relationships with benzene derivatives.The pyridine derivative chosen for the application of this inter- action was citrazinic acid, a substance to which the following con- stitutional formula (I) was assigned by Hofmann and Behrmann (Ber., 1884, 17, 2681), but which, as aliown in this series of papers1448 SELL: STUDIES ON ClTRAZINlC ACID. (Trans., 1893, 1035; 1894, 28, 828) behaves in many respecte as having the tautomeric one (11). SOOH YOOH C C \N.. \/ N Assuming for our present purpose that citrazinic acid has the symmetrical constitution assigned to it by Hofmann and Behrmann, i t will be noticed that each of the hydrogen atoms is in the ortho- position with regard to the hydroxyl group on the same side of the ring, and para to that on the opposite side.If, therefore, the same rule of displacement, applies to pyridine derivatives as is observed in those of benzene, we should expect only one derivative containing a, single a,ldehycle group t o be possible, although of course there is not only the possibility, but the probability, of the introduction of two aldehyde groups into the molecule. These compounds would be represented by the formula YOOH C COOH /\ 11. OHC+' g*CHO . HO*C C*OH "/ Working on the lines indicated above, the following compounds, (1) The disodium salt of the monaldehyde, formed to the extent of (2) The monaldehyde acid. (3) The oxime of the monaldehyde acid. (4) The phenylhydrazine salt of the hydrazone.amongst others, have been isolated and analysed. more than 60 per cent. of the citrazinic acid taken. EXPERIMENTAL. so dim^ Salt of the Monalclehyde. Following the directions given by Tiemann and Lewy (Bey., 1877, 10, 2216) in their research on resorcinol, that, to diminish the pro- portion of bye-products a less amount of the hydroxy-compaund should be taken than is the case with phenol, the following p ~ o - portmioils of the materials were found to work well in practice, no citritzinic acid being left unacted on. 25 grams of citrazinic acid,SELL: STUDIES ON CITRAZINIC ACID. 1449 200 of soda, and 125 of chloroform were mixed i n a cnpacisus flask connected with a reflux condenser, and heated by a small flame for about six hours, the termination of the action being shown by the colour which from the first is of a deep reddish-purple, gradually fading to a yellowish-brown.The filtered liquid on Lcing saturated with carbon dioxide gave a, copious precipitate of crystalline matter which, after draining by the aid of the pump and recrystallisation, consisted of the disodium salt of the monnldehyde. This crptallises from warm water in cream-coloured, nodular masses composed of opa,que needles containing 2H20. If, however, the solution, saturated at the ordinary temperature, be cooled down, the crystals deposited are large, faintly coloured, transparent prisms with 5H20. As a rule, however, the solution, after concentration, remains in a condition of supersaturation, and, on baing induced to crystallise at the tempera- ture of the laboratory (about 15O), deposit,s chiefly the form with 2H20, mixed, however, with a small percentage of the other form.A typical specimen of the substance deposited under these conditions gave the following numbers on analysis. 0.175 lost at, 182' 0.027 H20. 0.3595 gave 0.1865 Na,SO,. Na = 16-80. 0.231 ,) 0.121 ,> Na = 16.96. 0.2125 ,, 9.71 C.C. moist nitrogen at 15' and 783 mm. N = 5.32. C,H3N06Na2,2H20 requires N = 5.32 ; Na = 17-49 ; H,O = 13.68 p.c. The transparent, faintly yellow crystals, which are somewhat efflorescent in dry air, gave the following numbers on analysis. 0.473 lost 0.132 at 180". H20 = 27.90. 0.324 gave 0.1465 Na2S04. Na = 14.62. 0.727 ,, 0.320 ,, Na = 14.25. 0.251 ,, 9.7 C.C. moist nitrogenat 16' and 758.8 xnm.N = 4-49. C,HsN05Na2,5H20 requires N = 4-51 ; Na = 14.51 ; H,O = 28.39 P.C. H20 = 13.42. 0.1875 ,, 8.71 C.C. ,, ,, 19' ,, 785 ,, N=5*51. 0-2155 ,, 8-45 C.C. 3 9 ,, 757 ,, N = 4.54. The Monaldehyde acrid. When an aqueous solution of the sodium salt is mixed with excess of hydrochloric acid in the cold, B precipitate is produced consisting of spherical groups of needles which, after washing and drying, are faintly yellowish ; it is slightly soluble in cold, more readily in hot water, as also in alcohol, ether, and acetone. If, however, heat is applied in effecting solution, the liquid becomes coloured, and the recrystallised product is invariably so ; this i s owing to the occurrence of a certain amount of decomposition with formation of a purple-red 5 3 2I450 SELL: STUDIES ON CITRAZINIC ACID.substance, the nature of which is at present under investigation. On the application of heat, the crystals lose their water of crystallisation at about 130-140°, and at higher temperatures become purple-red, and finally blacken without melting. The substance does not restore the colour to a solution of magenta decolorised by sulphurous acid (Schiff's reaction), and does not reduce Fehling's solution. The silver salt may be precipitated from a solution of the normal sodium salt and dried at 100" without change, but if the solution be first rendered alkaline by sodium hydroxide, metallic silver is formed on heating. The precipitated aldehyde gave the following results on analysis. The numbers refer to the undried substance, unless otherwise stated. I.0.5504 lost 0.0489 a t 127'. HzO = 8.88. 11. 0.490 ,, 0.390 ,, HZO = 8.86. 111. 0.2310 gave 0.3885 CO, at 0.057 H20. C = 45.86; H = 2-74. IV. 0.202 ,, 13.12 C.C. nitrogen at 15' and 770 mm. N = 7.70. V. 0.3435 ,, 21% C.C. ,, 16.5"and 759 mm. N = 7-30, Determinations I11 and IV were made on the dried substance, V on the undried substance. C7H5N06 requires C = 45.90; H = 2.73; N = 7.65; whilst C7H,N0,,2Hz0 requires N = 6-96, H,O = 8.95 per cent. The Oxime. A specimen of the pure sodium salt, weighing 2 grams, was dis- solved in water, mixed with rather more than the calculated quantity for 1 mol. of hydroxylamine hydrochloride and a slight excess of sodium carbonate, and, after a few hours, the mixture was strongly acidified with acetic acid ; a copious, pale yellow precipitate, eonaist- ing of fine needles, was thus obtained which, after washing with water and drying in air, formed specimen (1).Another specimen was prepared from 5 grams of sodium salt, using two molecular proportions of the hydroxy lamine salt with the calcu- lated amount of sodium carbonate, allowing the mixture t o stand over night, and precipitating by a slight excess of dilute hydrochloric acid. This is sample (2). A third specimen (3) was prepared exactly as in (2), except that only one molecular proportion of hydr- oxylamine salt was employed. The determination I was made with specimen ( l ) , I1 and I11 with (2), and IV with (3). I. 0.3037 lost 0,0252 at 97". H2O = 8.29. 11. 0.2275 ,, 0.0187 in a vacuum.H,O = 8.26. 111. 0.21 gave 23-2 C.C. nitrogen at 15' and 770 mm. N = 13.11. IT. 0.16 ,, 17.75 C.C. nitrogen at 15' and 770 mm. N = 13.16. Calculated for oxime with 3 mol. of water, N = 12.96 ; H20 = 8.33 p.c.ACTION OF ACIDIC OXIDES ON SALTS OF HYDROXY-ACIDS. 1451 Yhenylhyd~azi?te Salt of the Hydrazone. The effect of phenylhydrazine acetate on a solution of this alde- hyde acid is not t o give the simple hydrazone, but, as would be expected, its very stable phenylhydrazine salt ; to prepare this, the aqueous solution of the sodium salt is acidified with acetic acid and mixed with a solution of phenylhydrazine acetate, when the solution immediately becomes semi-solid from the deposition of this salt in long, yellow needles. These were collected, and, after washing with hot water, in which they are but very slightly soluble, recrystallised from methylated spirit.The crystals, dried in a vacuum, gave the following results on analysis. The samples analysed were from two different specimens of the sodium salt, No. 1 from one specimen, and the others from a different one. 0,1392 gave 22.6 C.C. nitrogen at 18' and 755.0 mm. N = 18-62, 0.1135 ,, 17.75 ,, 17" ,, 766.4 mm. N = 18.28. 0.1195 ,, 18-75 ,, 18' ,, 766.4 mm. N = 18.22. CIgH,,N,04 requires N = 18.37 per cent. Before concluding, it may be remarked that one of the most interesting of the subsidiary products of this reaction, of which only a partial study has as yet been made, is a very stable and sparingly soluble acid which is formed to the extent of about 5 per cent,of the citrazinic acid taken. Analysis indicates its empirical formula to be C4H3NO2, but a study of its salts and nitro-derivative renders it certain that some multiple of this would more correctly represent its moiecu- lar formula. I n conclusion I have much pleasure in expressing my thanks to Mr. T. E. Tadman, B.A., Scholar of Queen's College, for his assistance in the analytical work recorded in this paper. Univemity Laboratory, Cambridge.