SYNTHESIS OF AMMONIA AT HIGH TEMPERATURES. PART 111. 113 XII.-TJLe Synthesis of Ammonia at High Temperatures. Part 111. By EDWARD BRADFORD MAXTED. IN previous communications (T. 1918 113 168 386) some accountl has been given of the formation of ammonia in a rapidly cooled high-tension arc and in water-cooled flames and evidence has been brought forward to show that the percentage of ammonia in equilibrium with nitrogen and hydrogen after decreasing with increasing temperature passes through a minimum value and a t very high temperatures increases with increasing temperature a result which may also be shown thermodynamically. A determination of t,he equilibrium ammonia-content a t the temperature of the high-tension arc ( J . SOC. Chent. Id. 1918 37, 232) gave approximately 1.7 per cent.by volume for the equil-ibrium percentage a t atmospheric pressure under the experimental conditions employed and yields of ammonia up to 1.5 per cent. could be recovered by suitable cooling 114 MAXTED THB SYNTHESIS OF AMMONIA This reversal in the direction of variation of the equilibrium ammonia-percentage with temperature appears t o be sufficiently interesting to justify further study and it is proposed in the p r e sent paper t,o deal wit'h the formation of ammonia in an arc of larger size and more usual character than those hitherto employed. For the production of the arc single-phase 50-cycle alternating current supplied a t a maximum pot'ential of 375 volts was trans-formed to high tension by means af an oil-immersed static trans-former having a step-up factor of 31.5.It was fuund easily possible to obtain an appreciable concen-tration of ammonia by almost any method by which the mixture of nitrogen and hydrogen was brought into contact with the arc flame with subsequent rapid cooling for instance by means of a blown arc similar to that used by McDougall and Howles (Proc. Mamchester Phil. SOC. 1900 44 No. 13) for the synthesis of n i t r i c oxide but in such cases a considerable proportion Qf the gas mix-ture blown into the arc failed to reach the requisite uniform high temperature and for a preliminary study of the maximum per-centage of ammonia formed as distinguished from the maximum quantity formed with a given expenditure of electrical energy it was considered preferable t o allow the arc to burn freely in the reacting gas mixture and to draw off samples of gas by means of a silica tube of small diameter placed in close proximity to the arc.The lat'ter part of the present paper was carried out with a view to bringing additional evidence that the formatign of ammonia a t high t'emperatures really takes place by the direct union of hydrogen and nitrogen and not by $he subsequent reduc-tion of nitric oxide formed from traces of oxygen in the reacting gases. It should further be stated in this connexion that 'the mixture of nitrogen and hydrogen employed in all experiments reported both in this and in the previous papers was as far as pmsible free from oxygen and in no case contained sufficient of this to account for such secondary formation of any appreciable part of the ammonia obtained.EXPERIMENTAL. The apparatus employed is shown in the figure. The electrodes A and B are of platinum and terminate in small spheres slightly more than 1 mm. in diameter. C is a t'hick-walled capillary silioa tube its upper end being opened out and flattened so as t o form a slit approximately 4 mm. long and 1.5 m. wide. The silic AT HIGH TEMPERAT'URES. PART In. 116 wall bounding the ends of this was ground away and in the depressions thus formed a t each end of the silica slit the electrodes rested the lower part of the tube being circular in section and about 1.5 mm. in internal diameter. The arc was enclosed in a large inverted flask provided with a side-tube and three-way tap D for preliminary exhaustion and subsequent admission of the gas to be subjected to the aotion of the arc this gas normally passing into the flask a t D and leaving the syst'em by way of the silica tube already described.For the following measurements the current employed for arc-formation was limited by the interposition oli an adjustable resist-ance on the low-tension side of the transformer. The platinum electrodes became quickly white hot and the arc flame burned steadily across the slit and was, to a certaih degree drawn into the tube by the current of gm. Sufficient cooling for the recovery of the greater portion of the ammonia formed and for the prevenlhn of the fusion of the silica tube was obtained without water-cooling provided that a thick-walled silica capillary was used and that the arc employed was not too large.Samples of gas taken a t various rates were analysed by passage through N/lO-acid and in some cases by allow-'& + ing a small volume to pass through Nessler's solution practically identicall results being obtained from each The quantity of ammonia observed is. influenced necessarily not only by the temperature of the arc, but also by heating and cooling factors. With very slow currents of gas heating to arc temperature occurs satisfactoriIy but the ammonia formed undergoes considerable decomposition owing to the slowness with which it leaves the region of madmum tempera-ture. Passage of the gas too quickly through the arc results in imperfect heating such rapid passage however involving rapid cooling and consequently a more complete retention of the ammonia formed.It will be seen from table I that the concentration of the ammonia recovered a t the ordinary temperature first of all rises rapidly with increasing velocity of passage then passes a maximum method of analysis. &nil 116 MAXTED THE SYNTHESIS OF AMMONIA value the percentage of amionia subsequently falling gradually with etill greater velocities of passage. This form of the concentra-tion-velocity curve is a necessary result of the nature of the heat-ing and cooling factors discussed above. The issuing gas of course had a strong odour of ammonia a t all the rates of flow studied and the percent.ages obtained were of the same order of magnitude as those previously observed with small arcs not produced directly by a high-tension current of the usual sinusoidal wave form.For this series of experiments 0.04 ampere a t a potential of 3250 volts was taken for arc-formation. TABLE I. Gas Mixture ATitrogen 25 per cent.; Iiydroyen 75 p e r cent. Rate of flow of gas in litres per hour. o s . 5 0.57 0.85 1.14 2.1 (2.1 Concentration of ammonia per cent. by volume. 0.49 0.6 1.2 1.04 0.58 0.56) Rate of flow of gas in litros per hour. 3.4 (3.4 4.0 6.1 7.5 (7.5 C Gncentration of ammonia per cent. by volume. 0.49 0.45) 0.43 0.40 0.36 0.32) It appears desirable at this point' t'o discuss the evidence that the formation of ammonia at high temperatures takes place by the direct union of nitrogen and hydrogen and not secondarily by the reduction of nitric oxide formed from nitrogen and traces of oxygen in the reacting gas.The point is of fundamental import-ance in any consideration of the variation of the ammonia equil-ibrium with temperature and for this reason has been examined in such a way as to place beyond doubt t-he directl nature of the synthesis of ammonia a t arc temperatures. It' may easily bs shown from a Consideration of the nitric oxide equilibrium particularly on account of the small partial pressure of the nitrogen in the gas mixture employed such mixture consist-ing uniformly of 25 per cent. of nitrogen and 75 per cent. of hydrogen that a trace of oxygen amounting t o 1 per cent. by volume or less will not accounh for the percentage of ammonia obtained even assuming that' all the nitric oxide which can be formed under the conditions of experiment is quantitatively reduced to ammonia.The amount of nitric oxide that can be formed from nitrogen and oxygen a t partial pressures of the order mentioned is depressed by the ratio of partition of oxygen between hydrogen and nitrogen to a very small fraction of the alread AT HIGH TEMTERATURES. PART 111. 117 small percentage which may be calculated as capable of being formed from a consideration of the nitric oxide equilibrium only. I n spite however of the small order of magnitude of the per-centage of ammonia that might be formed secondarily from the traces of oxygen which are always present in commercial gases it was considered preferable both for the work described in the present paper and for all measurements of the formation of ammonia a t arc temperature previously reported to employ a gas known to1 be free from oxygen within the limits of the ordinary methpds of analysis.This gas was prepared i n a manner similar to that used by Haber and Van Oordtl (Zcitnch. nnorg. Chena. 1905 44 341) for their determinations of the ammonia equilibrium a t lower tempera-tures in the presence of a catalyst by decomposing ammonia by passage through a heated iron tube the mixture of nitrogen and hydrogen produced being carefully and thoroughly freed from ammonia by treatment with sulphuric acid and compressed for convenience into a previously exhausted steel cylinder by means of a tot-ally immersed compressor of such design as to render impossible any penetration of air to the gas during compression.Each cylinder of nitrogen and hydrogen prepared in this way was carefully tested for absence of ammonia before use by bubbling a considerable volume through Nessler's solution. The gas-mixture contained certainly less than 0.1 per cent. of oxygen from which percent age the ammonia capable of being formed secondarily would be negligible and in any case incapable of accounting for a yield of ammonia of 1 per cent. or more. I n a few preliminary measurements a gas was used which had been made by the catalytic removal of traces of oxygen from a mixture of commercial hydrogen and nitrogen by passage over a heated metal but whilst the resulting gas was equally satisfactory from the point of view o€ the yield of ammonia obtained by passage through an arc the preparation of an oxygen-free mixture in this way was more troublesome than by the firstl method.I n addition to employing a gas free from oxygen it was con-sidered interesting t o examine the synthesis a t arc temperature under such conditions that the same volume of nitrogen and hydrogen was repassed a number o€ times through the arc the ammonia formed a t each passage being absorbed and measured. It is obvious that any trace of oxygen would particularly by reason of the hydrogen present be removed during the first few passages through the arc and that the conversion t o ammonia of an approximately constant percentage of the gas-mixture during F 118 MAXTED THE SYNTHESIS OF AMMONIA.PART III. each successive passage would confirm beyond doubt the direct nature of the synthesis. The experimental method employed consisted in confining a known volume of an oxygen-free mixture of nitrogen and hydrogen in a graduated vertical glass capillary tube about 1 metre long. The upper end of this tube was fused on to a second short hori-zontal capillary tube containing platinum wire electrodes 0.5 mm. apart between which a small induction arc as described in a previous communication could be formed this second capillary tube ending in a small absorption pipette filled with dilute sulphuric acid. The lower end of the graduated capillary tube was sealed by means of mercury covered with a small quantity of sulphuric acid and by the regulated motion of this seal up and down the graduated tube the thread of gas could be passed and repassed through the arc as often as desired.An approximately uniform rate of passage was obtained by making the graduated capillary tube one limb of a U-tube and causing the required motion by means of a mercury flow this being normally regulated so that each double passage through the arc occupied about six minutes when 1 C.C. of the gas mixture was taken for experiment. Working as above described the ammonia formed during each upward passage was removed by the small absorption pipette sealed to the other end of the short capillary arc tube whilst that pro-duced during each downward passage was absorbed by sulphuric acid clinging to the side of the graduated capillary.The arc tube was of course not allowed to become wet on account of the danger of fracture and the thread of gas after its introduction passed no joints by means of which penetration of air might occur. Table I1 summarises the results obtained in two experiments of this nature the estimation of ammonia being in this case carried out volumetrically by noting the contraction after each passage. A preliminary small expansion occurred on starting the arc and passage was only begun after the volume had become more or less constant . The yields of ammonia are much the same as those previously found for such arcs by other methods of analysis and with more accurate control over the rate of passage. Probably by reason of the more rapid nature of the cooling the yields are slightly higher than those obtained with the larger arc described in the first part of the present paper.Each cm. of the graduated capillary corre-sponded with 0.01 C.C. of gas so that the volume could be read off with fair accuracy to 0.001 C.C TRE EFFECT OF SOME SIMPLE ELECTROLYTES ETC. 119 TABLE 11. Contrahtion Vol. of gas No. of pas- after doublo passed through sages since be- passage through Percentage of No. of arc. ,ginning of arc. ammonia expt. C.C. formed. expt. C.C. 1 0.74 1 (original volume) 0.715 0.69 0.67 0445 0-627 0.605 0.59 2 0,825 (original volume) } 0-795 0.77 0.75 0.723 0.70 0.68 0.655 0-63 0.605 0.596 0.575 0.655 0.532 0.51 0-49 0.475 3 4 6 8 10 12 14 16 2 4 G 8 10 12 14 16 18 20 22 24 26 28 30 32 34 0.025 0.025 0.02 0.025 0.018 0.022 0.015 0.02 0-03 0.025 0.02 0.027 0.023 0.02 0.025 0.025 0-025 0:Ol 0.02 0.02 0-023 0.022 0.02 0.015 0.015 1.7 1.7 1-5 1.9 1.4 1.8 1-2 1.7 1.8 1.6 1.3 1.8 1.6 1.4 1.8 1.9 2.0 0.5 1.7 1.7 2-0 2.0 1.9 1.6 1.6 The experimental conditions including control of rate of flow and the method of analysis were not suitable for very accurate measurements but the approximate constancy of the yield of ammonia and especially the absence of any indication that a normal amount of ammonia is formed during the first passage and little or none during subsequent passages appears to demonstrate without doubt the direcb nature of t3he synthesis a t high temper a t ures. [Received December 23rd 191 8.