densation product. In a vacuum, as before, the syrup, after several weeks, changed to a resin, which decomposed instantly with water, giving up to the latter, HCl=25.4 per cent. On exposure to the air, the resin first deliquesced, yielding a clear, mobile liquid; as more moisture was absorbed, this became turbid, and finally solidified, owing to the separation of a condensation product. The experiment was repeated, the gas being passed until all the clot had redissolved; the syrup, treated as before, gave a clear resin having the properties described above, and yielding to water HCl=24.5 per cent.

Expt. 12.-Dry hydrogen chloride had no visible effect on dry "methylenecarbamide," but when the latter was moistened and the passage of the gas continued, the solid gave place to a colourless syrup, gradually hardening in a vacuum. To water, the solid product gave up HCl=27.3 per cent., a condensate separating, in which was found N-37.85 per cent. (the sample of "methylenearbamide" taken for experiment contained N-380 per cent.). When added to dilute alkali hydroxide or to a solution of ammonia, the chloro-derivative behaved exactly as towards water, hydrogen chloride being eliminated and neutralised by the excess of alkali present, whilst a condensation product was instantly deposited.

Like results were obtained when carbamide was dissolved in formalin, previously saturated with hydrogen chloride (or mixed with excess of fuming hydrochloric acid), and when "methylenecarbamide" was dissolved in cold, fuming hydrochloric acid, the product in all cases being a syrup, behaving as described above. From the foregoing results, it is evident that carbamide, formaldehyde. hydrogen chloride, and water together produce a balanced system, the equilibrium of which depends on the concentrations of the various components.

Solid carbamide hydrochloride, when exposed to the fumes obtained by rapidly distilling paraformaldehyde, did not appear to undergo any change.

Methylolcarbamide.

The somewhat elaborate procedure recommended by Einhorn and Hamburger (loc. cit.) for the preparation of this substance is unnecessary; by exposing for a day or two in an exhausted desiccator a solution of carbamide (1 mol.) in formalin (1 mol.); just neutralised by dilute alkali hydroxide, a viscid, crystalline paste was obtained, from which, after a couple of recrystallisations from alcohol, methylolcarbamide was isolated in brilliant, flattened prisms melting at about 110° (corr.). (Found: N=30.9; C2H6O2N2 requires N=31·11 per cent.).

The aqueous solution was odourless, and gave with sodium nitroprusside and phenylhydrazine no reaction for formaldehyde, and with neutral mercuric nitrate no reaction for carbamide. On acidification, however, formaldehyde was detected, both prior to condensation and long after the process appeared to be complete.

Dimethylolcarbamide.

i

A neutralised solution of carbamide (1 mol.) in a little more than 2 mols. of formalin, when treated as above, yielded dimethylolcarbamide; it had essentially the properties described by Einhorn and Hamburger, including the relatively slow condensation by acids. In moderately concentrated solutions, this process requires approximately minutes, where seconds are needed in the case of the monomethylol analogue. Otherwise, its behaviour in an aqueous solution was the same as that of methylolcarbamide.

10

The fused material soon began to decompose (m. p. about 123°). water and formaldehyde escaping, and a white, amorphous solid being left; this, when heated alongside a pure specimen of the compound, CHON1, melted at the same moment as the latter, and with similar effervescence and browning (255°, uncorr.; both introduced into the bath at 235°). By condensing an aqueous solution with dilute hydrochloric acid, the same product was obtained as by heating; it gave N=32.8 per cent., whereas CHON, requires N=32.18 per cent. The end result of these changes may thus be represented:

2CO(NH⚫CH, OH)2 → H2O + CH2O+C5H10O3N4

Expt. 13.-Approximately pure methylolcarbamide, when condensed by means of dilute hydrochloric acid, yielded a product containing only 35.5 per cent. of nitrogen ("methylenecarbamide" requires N=38.89 per cent.), the deficit presumably being due, as earlier suggested, to the attack of the methylolcarbamide, when in the act of vielding its normal condensation product, by the formaldehyde which is liberated before that act is complete.

The experiment was repeated, 1 molecular proportion of formaldehyde being first added to the solution of methylolcarbamide; in these circumstances, the solid product was found to consist solely of Goldschmidt's compound. (Found, N=32.25. CH10O3N4 requires N=32.18 per cent.)

Expt. 14.-In order to learn whether the last-named compound might result through an interaction between the two methylolcarbamides, somewhat as follows:

NH, CO NH•CH,•OH+CO(NH•CH, OH),

2

a molecular mixture of the components, in aqueous solution, was acidified, the mixture (which began to condense in forty seconds) being allowed to remain overnight. Here, if the two jointly condensed, the sole product must be the material containing N=32.18 per cent., whereas, if they condensed independently, the first precipitate would be the mixture referred to in Expt. 13 as containing N=35.5 per cent., to which later the condensate of dimethylolcarbamide (N=32.18 per cent.) would add itself; so that, ultimately, the product must give a figure somewhere between the two. The percentage actually found was 33.35, and when the experiment was repeated, but with 0.75 mol. of dimethylolcarbamide, the figure was 34-6; hence, it is concluded that, in a mixture of methylol- and dimethylol-carbamide, acidified with hydrochloric acid, the two substances condense independently.

Although in the acid condensation of the former, the elimination of formaldehyde prevents the production of pure "methylenecarbamide," yet fused methylolcarbamide soon decomposes, water being evolved, but little or no formaldehyde. In this case, the residue is doubtless free from much contamination with Goldschmidt's compound; no analysis, however, was made.

Expt. 15.-Solutions were prepared (a) of methylolcarbamide, (b) and (c) of carbamide and formaldehyde in molecular proportion. All had the same "carbamide-concentration" (1 gram in 6 c.c.); equal volumes of each, at 13°, were acidified with one-twelfth the volume of concentrated hydrochloric acid. Prior to acidification, however, (b) was kept at 13° for an hour and a half. The times elapsing before precipitation were, for (a) less than three seconds, for (b) twenty-eight seconds, for (c) forty-five seconds. Ready-made methylolcarbamide, therefore, condensed at least fifteen times as rapidly as a mixture of its components, which had remained for some five or six minutes, whilst a like mixture, kept beforehand for an hour and a-half, condensed in nearly half the time.

Evidently, the union of formaldehyde with carbamide takes place only by degrees; in fact, a solution, kept for twenty-four hours in a vacuum desiccator, still had a strong odour of formaldehyde. In the presence of barium hydroxide, too, at the concentration recommended by Einhorn and Hamburger, the same phenomenon was noticed.

CHEMISTRY DEPARTMENT,

UNIVERSITY COLLEGE,

CORK.

[Received, January 1st, 1918.]

XXIX.-The Sub-bromide and Sub-chloride of Lead. By HENRY GEORGE DENHAM.

IN a recent paper (T., 1917, 111, 29), the author has described a method for the preparation of lead sub-iodide by the action of methyl iodide vapour on lead sub-oxide. As indicated in that paper, the method appears of a somewhat general nature, applicable at least to the preparation of other sub-salts of lead. In the present paper, the method has been used to prepare lead subbromide and sub-chloride, the only radical alteration being in the replacement of the methyl haloid by the ethyl compound, owing to the more convenient boiling point of the higher homologue.

Lead Sub-bromide.

The apparatus already described in the previous paper (loc. cit.) has, with a few minor alterations, proved quite suitable. The lead sub-oxide was prepared from lead oxalate (Found: Pb-70.17. Calc.: Pb 70·18 per cent.) precipitated from an acid solution of lead acetate by the addition of oxalic acid, the actual decomposition of the oxalate into the sub-oxide being carried out precisely as described in the former paper. The alterations in the apparatus to be noted were the replacement of phosphoric oxide as a drying agent by anhydrous calcium chloride, and the introduction of a contriction in the glass tubing leading from the oven to the condenser. The former was a matter of necessity, owing to a temporary shortage of the phosphoric oxide, the latter was devised to facilitate the control of the distillation. Shortly after the ethyl bromide* had been introduced into the apparatus, a little distillate collected at the constriction and formed a most convenient gauge for noting the velocity of distillation. This is a very

* Purification of the Ethyl Bromide. The ethyl bromide was prepared from chemically pure potassium bromide, ethyl alcohol, and sulphuric acid. The distillate was treated with dilute sodium carbonate solution, several times with water, then shaken at least four or five times with concentrated sulphuric acid, and finally with water (owing to a tendency to emulsify, it was often found necessary to wash several times with dilute sodium hydroxide solution before the final washing with water). The bromide was dried over calcium chloride, and fractionated from phosphoric oxide, the fraction of constant boiling point alone being used. This method of purification had to be rigidly followed, otherwise samples of ethyl bromide were obtained that contained traces of some reactive impurity in sufficient quantity to cause appreciable errors when a relatively large amount of the bromide was distilled through the sub-oxide.

essential point, for the ethyl bromide appears to react much less vigorously with lead sub-oxide than does methyl iodide. Thus, in the preparation of lead sub-iodide, about fifty minutes' distillation sufficed to convert the sub-oxide into the sub-haloid, whilst in the present case at least 150 minutes' distillation was required to secure complete reaction. The need for a preliminary heating of the vapour before it enters the reaction bulb is even more necessary than in the case of lead sub-iodide. This was again obtained by passing the vapour through a capillary spiral. The extreme slowness of the reaction naturally leads one to search closely for the temperature at which the ethyl bromide just escaped

decomposition. By the process of "bracketing," it was found

that the distillation can be safely carried through at 261°, but not higher. At temperatures much lower than this, the last traces of sub-oxide react very sluggishly; for example. at 258°; analvsis indicated the presence of 74.1 per cent. of lead (PbBr requires Pb-72.16 per cent.), although, by the prolonged absence of the evolution of gas, the reaction appeared quite complete. At 262° gas was evolved until the end of the experiment, the product was much lighter than that obtained in experiments carried out at 261°, and analysis gave Pb=70.2 per cent. Undoubtedly there had been incipient decomposition of the bromide vapour, with liberation of bromine and subsequent oxidation of the lead subbromide to the normal bromide. Indeed, it may be safely asserted that whenever this evolution of gas continues throughout the experiment, one invariably obtains a product contaminated by more or less lead bromide. After the distillation was completed, the receiver containing the distillate was cooled with liquid ammonia, the apparatus partly exhausted, the receiver sealed off, and the exhaustion continued until the pressure had fallen to about 1 mm. It was found to be unnecessary to continue the exhaustion longer as several experiments showed no trace of a volatile product, such as occurred in the preparation of lead subiodide. On the other hand, the amount of non-volatile carbonaceous matter was sufficiently high to necessitate estimation, generally amounting to from 0.3 to 0.5 per cent. of the weight of the lead subbromide.

Method of Analysis. The mixture of lead sub-bromide, carbonaceous matter, and silica was digested with concentrated acetic acid. then extracted with water; after four extractions the residue was treated with a little hot, very dilute nitric acid, as it was found that the carbonaceous matter otherwise tended to coagulate and retain traces of lead bromide. Two washings with the dilute nitric acid sufficed to remove the whole of the dissolved bromide. The