Nitric Oxide, NO

Mayow in 1674 seems to have first prepared nitric oxide by the action of nitric acid on metals, such as iron, but it was Priestley who, in 1772, first isolated and described the gas. He found that many metals with nitric acid produced a colourless gas which was only slightly soluble in water and produced no change with lime-water. The name of nitrous air was given to this gas by Priestley; it extinguished a taper, was noxious to animals, and was very soluble in a solution of green vitriol, which turned a dark colour. Priestley observed the formation of brown fumes when nitric oxide was mixed with air, and the fact that these fumes immediately dissolved in water.

The composition of nitric oxide was first investigated by Priestley in 1786, who heated iron in nitric oxide by means of a burning glass. He noted an approximate diminution of one-half. Davy in 1800 carried out a similar experiment, but used carbon instead of iron, and, assuming that the carbon dioxide formed contained its own volume of oxygen, concluded that nitric oxide contained rather more than one-half its volume of oxygen and rather less than its volume of nitrogen.

The exact volumetric composition of nitric oxide was demonstrated by Gay-Lussac in 1809, who burnt metallic potassium in the gas and showed that the residual nitrogen occupied just one-half of the original volume. He rightly concluded that "this gas is composed of equal parts by volume of nitrogen and oxygen."

One of the most accurate determinations of the gravimetric composition of nitric oxide was carried out by decomposing a known weight of the gas with heated nickel. The nitrogen liberated was condensed on charcoal and weighed, while the oxygen content was found by determining the increase in the weight of the nickel.

1. The action of nitric acid (sp. gr. 1.2) upon copper turnings or foil gives a ready stream of nitric oxide together with varying amounts of nitrous oxide and nitrogen. Ackworth showed that accumulation of copper nitrate increased the amount of nitrous oxide, so that a purer gas is obtained by removing this solution at the bottom of a tower containing the copper:
   
   3Cu + 8HNO₃= 3Cu(NO₃)₂ + 2NO + 4H₂O.
   
   Purification is best effected by washing the gas with water, absorbing in ferrous sulphate solution, and then expelling it from the latter solution by means of heat.
2. A purer form of nitric oxide is obtained by heating ferrous sulphate with nitric acid, or a mixture of ferrous sulphate and potassium nitrate with dilute sulphuric acid:
   
   2KNO₃ + 6FeSO₄ + 4H₂SO₄ = K₂SO₄ + 3Fe₂(SO₄)₃ + 2NO + 4H₂O.
3. The action of potassium nitrate on a solution of ferrous chloride in hydrochloric acid also yields a fairly pure form of gas:
   
   2KNO₃ + 6FeCl₂ + 8HCl = 2KCl + 6FeCl₃ + 2NO + 4H₂O.
4. A convenient method of preparation is by dropping a concentrated solution of sodium nitrite into a solution of ferrous sulphate in sulphuric acid, or into a solution of ferrous chloride in hydrochloric acid:
   
   HNO₂ + FeCl₂ + HCl = NO + FeCl₃ + H₂O.
5. A slow stream of nitric oxide is obtained by dropping a solution of a mixture of potassium ferrocyanide and potassium nitrite into dilute acetic acid:
   
   K₄Fe(CN)₆ + HNO₂ + CH₃COOH = K₃Fe(CN)₆ + CH₃COOK + NO + H₂O.
6. A very pure form of gas may be prepared by dropping sulphuric acid (50 per cent.) on to a concentrated solution of potassium nitrite (2 parts) and potassium iodide (1 part):
   
   2HNO₂ + 2HI = 2NO + I₂ + 2H₂O.
7. Pure nitric oxide is formed in the Lunge nitrometer method of estimating nitrates, namely, by shaking nitric acid with excess of sulphuric acid and mercury.
8. Nitric oxide is evolved when sulphur dioxide is passed through warm nitric acid (sp. gr. 1.15):
   
   3SO₂ + 2HNO₃ + 2H₂O = 3H₂SO₄ + 2NO.
9. Electrolysis of a mixture of a 20 to 30 per cent, solution of nitric acid, containing 1 to 2 per cent, of nitrous acid, below 50° C. yields nitric oxide.
10. Pure nitric oxide is produced when nitroso-diphenylamine is heated, either in vacuo (40° to 75° C.) or in air (180° to 190° C.).

Nitric oxide is a colourless gas with a low boiling-point, heavier than air and sparingly soluble in water, giving (at first) a neutral solution. Its density relative to air has been determined as 1.0372 and 1.0387. The weight of 1 litre at N.T.P. is 1.3402 grams.

Nitric oxide shows but little deviation from Boyle's Law. Jaquerod and Scheuer have determined the value of the coefficient α₀ in the equation



between pressures of 400 mm. and 800 mm., and found it to be -0.00117.

The coefficient of expansion between -140° and 0° C. is 0.0037074.

Regnault's value for the specific heat at constant pressure (C_p) is 0.23175, and the ratio C_p/C_v 1.40. The value of C_p/C_v, according to Heuse, is 1.38 at both 15° and -80° C. It diminishes steadily at higher temperatures; thus:

Temperature, ° C.	300	500	1000	1500	2000
Cp/Cv	1.399	1.393	1.375	1.347	1.317

The difference, C_p-C_v (=R for a perfect gas), also diminishes slightly at higher temperatures:

Temperature, °C.	300	500	1000
Cp-Cv	1.998	1.990	1.988

The variation of the molar heat at constant volume with the temperature is best expressed by the equation

C_v = 5.102 – 0.0₃564T + 0.0₆9554T² - 0.0₉1934T³.

Nitric oxide is an endothermic compound; the heat of formation is given as -21.575 or -21.600 Cals. at ordinary temperatures. That calculated for 1841° C. is -23.000 Cals.

The viscosity of nitric oxide at 0° C. is 1794×10^-7.

The refractive index of nitric oxide for the sodium D line is 1.-0002939.

The following table gives the values of the absorption coefficient (β) of nitric oxide at various temperatures: -

Solubility of nitric oxide on water

Temperature, °C.	Absorption Coefficient.
0	0.07381
5	0.06461
10	0.05709
15	0.05147
20	0.04706
25	0.04323
30	0.04004
40	0.03507
50	0.03152
60	0.02954
70	0.02810
80	0.02700
90	0.02648
100	0.02628

Nitric oxide is much more soluble in ethyl alcohol than in water, as is shown in the following table:-

Temperature, ° C.	0	5	10	15	20	24
Absorption coefficient (β)	0.31606	0.29985	0.28609	0.27478	0.26592	0.26060

The formula expressing the value of the absorption coefficient between 0° and 24° C. is:

β = 0.31606 – 0.003487t + 0.000049t².

The solubility of nitric oxide in sulphuric acid has been determined by Tower. The following table gives his results at 18° C. and 760 mm. in terms of the solubility coefficient (β'), which includes the vapour pressure of the solvent:-

Solubility of nitric oxide in aqueous sulphuric acid

Per cent. H2SO4	Sp. Gr. of Acid at 15° C	Tension of Water Vapour (mm).	Solubility Coefficient (β')
90	1.820	0.1	0.0193
80	1.733	0.4	0.0117
70	1.616	1.5	0.0113
60	1.503	3.1	0.0118
50	1.399	6.2	0.0120

The earlier work of Lunge gave the values of 0.035 for 98 per cent. H₂SO₄ and 0.017 for 60 per cent. H₂SO₄. Tower points out that for 98 per cent. H₂SO₄ no constant results could be obtained owing to the solubility of the mercury in sulphuric acid in the presence of nitric oxide. Hence for quantitative work with the Lunge nitrometer concentrated acid should not be used. The most accurate results for the estimation of nitrates, nitrites, and oxides of nitrogen are obtained by using sulphuric acid of 70 per cent, concentration, in which both nitric oxide and air show a minimum solubility.

The absorption of nitric oxide by aqueous solutions of various salts has been the subject of much investigation. In the case of ferrous salt solutions the solubility of the gas increases with the concentration of the solution. The limit is reached when the proportions of iron to nitric oxide are in the ratio 1:1, both in aqueous and alcoholic solutions. It is assumed that unstable chemical compounds are formed of the type FeSO₄.NO, FeCl₂.NO, etc., but the ready dissociation of such compounds under the influence of heat indicates only a feeble combination. Usher has investigated the freezing-point of such solutions, and finds that neither the freezing-point nor the pressure of the nitric oxide remained constant, and hence no conclusion can be drawn as to the nature of the compound FeSO₄.NO. The absorption of nitric oxide by bivalent salt solutions of nickel, cobalt, and manganese is of a similar nature. Ferric salts also absorb nitric oxide readily, as also do many metallic and non-metallic halides. Nitric oxide dissolves in solutions of copper sulphate, producing a violet unstable compound, CuSO₄.NO. The solubility of the gas in cuprous halides in various solvents has been determined.

According to Villard, nitric oxide forms an unstable hydrate at O° C. under a pressure of 10 atmospheres, or at 12° C. under 40 atmospheres, but not above 12° C. under any conditions.

From conductivity determinations of water through which nitric oxide is passed, Zimmermann concludes that the gas is neither a weak nor a strong acid.

A purple solid is formed by nitric oxide and hydrogen chloride at -180° C. This melts at -150° C. to a purple liquid, and is assumed to be an unstable complex of the type [NOH]⁺Cl^-.

Faraday unsuccessfully attempted to liquefy nitric oxide at -110° C. and under a pressure of 50 atmospheres, but the liquefaction was accomplished by Cailletet, who allowed the gas to expand from a pressure of 104 atmospheres at -11° C. Liquid nitric oxide is colourless in thin layers but slightly blue when examined in thick layers, which colour is probably due to traces of nitrogen trioxide. The critical constants are shown in the following table: -

Critical constants of nitric oxide

Critical Temperature, °C.	Critical Pressure, Atm.	Critical Density.
-93.5	71.2	0.524
-92.9	64.4	. . .

The variation of the boiling-point of liquid nitric oxide with the pressure is shown in the following table:-

Variation of the boiling-point of nitric oxide with the pressure

Pressure.	Boiling-point, ° C.
356 (mm)	-157.3
760 (mm)	-150.2
9.8 atm.	-125.9
27.7 atm.	-109.2
55.6 atm.	-96.3

Adventowski assumes that liquid nitric oxide at low temperatures is polymerised on account of the anomalous vapour-pressure curve, and the high density at the boiling-point, 1.269, supports this view. It would appear, however, that complete dissociation has occurred at atmospheric pressure on account of the normal vapour density at this pressure.

The mean coefficient of expansion of liquid nitric oxide between -140° and 0° C. at 760 mm. Is 0.0037074. The refractive index for sodium light, is 1.330.

Olszewski obtained a snow-like mass by cooling liquid nitric oxide to -167° C. under a pressure of 138 mm. According to Adventowski, the melting-point of the solid is -160.6° C. at 168 mm. Liquid oxygen and solid nitric oxide combine explosively unless thoroughly cooled by complete immersion in liquid oxygen.

Nitric oxide is much more stable than nitrous oxide, and its decomposition by heat, which begins at 500° C. and is only slight at 900° C., is not complete until the very high temperature of 1775° C. is reached. The formation of nitric oxide from its elements has been studied by Nernst, and its decomposition by Jellinek. A candle, burning sulphur, and feebly burning phosphorus are extinguished by the gas, because the temperatures are too low to bring about its decomposition. Brightly burning phosphorus, however, continues to burn with increased brilliancy, producing phosphoric oxide and leaving nitrogen. Burning carbon similarly removes the oxygen, while burning boron produces a mixture of boric acid and boron nitride.

A number of metals are oxidised at a high temperature in nitric oxide, but if in a finely divided condition will reduce the gas at lower temperatures.

Electric sparks passed through a mixture of sulphur vapour and nitric oxide give sulphurous and nitrous acids, which react further to give nitrosyl-sulphuric acid.

Generally speaking, nitric oxide can be reduced in stages right down to ammonia. Thus, when the gas is mixed with hydrogen and brought into contact with platinum black, or finely divided nickel or copper, tin, iron, or zinc, ammonia is produced.

Stannous chloride reduces nitric oxide to hydroxylamine and ammonia, and also to hyponitrite if the solution is alkaline.

Chromous salts produce ammonia in neutral solution, and hydroxylamine if the solution is acid.

Hydriodic acid reduces nitric oxide to ammonia. Reduction to nitrous oxide is brought about by alkaline pyrogallol, sulphurous acid, phosphine (also some nitrogen), and sulphuretted hydrogen and alkaline sulphides (with some ammonium sulphide).

Many oxidising agents react with nitric oxide, producing chiefly nitric acid. Potassium permanganate, iodine, and hypochlorous acid yield nitric acid. Hydrogen peroxide gives nitrous and nitric acids. Silver oxide, manganese dioxide, lead dioxide, and red lead produce nitric acid. Chlorine peroxide gives nitrogen peroxide, while potassium chlorate and potassium iodate, heated in the gas, form nitrate and nitrogen peroxide with the liberation of halogens.

Nitric oxide reacts with water when preserved in contact for any length of time. According to Moser, both dissolved oxygen and hydrions are responsible for the chemical changes which take place, producing, firstly, nitrous and hyponitrous acids:

4NO + 2H₂O = 2HNO₂ + H₂N₂O₂.

The hyponitrous acid breaks down, giving nitrous anhydride and also ammonia:

3H₂N₂O₂ = 2N₂O₂ + 2NH₃.

This ammonia forms ammonium nitrite with the nitrous acid, and this breaks down with the liberation of nitrogen:

NH₄NO₂ = N₂ + 2H₂O.

The amount of nitrogen increases with the length of time the nitric oxide is kept in contact with the water.

Potassium hydroxide in contact with nitric oxide for some months at the ordinary temperature produces a diminution of 75 per cent, in the volume. The residual gas is nitrous oxide, and potassium nitrite is found in the solution. The reaction at 100° C. leaves a residue which chiefly consists of nitrogen with a small amount of nitrous oxide. At a higher temperature still, namely, 125° C., there is a diminution of 83.3 per cent., and the residual gas is entirely nitrogen:

6NO + 4KOH = N₂ + 4KNO₂ + 2H₂O.

Nitric oxide combines with chlorine and bromine to form nitrosyl chloride, NOCl, and nitrosyl bromide, NOBr (with NOBr₃), respectively, but with fluorine yields nitryl fluoride, NO₂F:

4NO + F₂ = N₂ + 2NO₂F.

The Oxidation of Nitric Oxide

The general explanation of the formation of ruddy fumes when nitric oxide is brought into contact with air or oxygen is that nitrogen peroxide is produced:

2NO + O₂ = 2NO₂.

There seems to be little doubt that nitrogen peroxide is the final product, but it is by no means decided whether the above equation truly represents the mechanism of the oxidation.

In the first place, it would seem that nitrogen trioxide is the sole product when the oxidation is carried out below - 110°, even with excess of oxygen:

4NO + O₂ = 2N₂O₃,

and the production of nitrogen tetroxide only occurs above -100°:

2N₂O₃ + O₂ = 2N₂O₄.

Raschig maintains that at ordinary temperatures a similar intermediate formation of the trioxide occurs, the second oxidation to the peroxide taking a much longer time. According to Lunge, however, the primary product of oxidation is the peroxide, the reaction being of the third order.

Further evidence in favour of nitrogen trioxide being the first oxidation product of nitric oxide, is the instantaneous formation of N₂O₃ when nitric oxide and oxygen are mixed in the ratio of 4 to 1 at ordinary temperatures, the product remaining stable. When the proportions of nitric oxide and oxygen are as 2:1, the N₂O₃ stage is reached very rapidly, then further oxidation to N₂O₄ occurs, 34 per cent, in 20 seconds, and completely in 100 seconds.

Sanfourche also maintains that the first stage in the oxidation of nitric oxide by dry air is nitrogen trioxide, which occurs instantaneously between -50° C. and 525° C. The second stage in the oxidation, which results in the formation of nitrogen peroxide, proceeds according to the equation

2N₂O₃ + O ⇔ 4NO₂, and is governed by the temperature.

The first oxidation product, even in presence of water, is stated to be the trioxide and not the peroxide. The nitrogen trioxide is then decomposed by water to form nitric acid and nitric oxide.

In the presence of nitric acid, Sanfourche considers that nitrogen trioxide is oxidised to form the peroxide and water:

N₂O₃ + 2HNO₃ = 2N₂O₄ + H₂O.

It cannot be said that the mechanism of the oxidation has been definitely settled. One of the chief difficulties encountered is the fact that nitrogen trioxide behaves chemically as if its formula were N₂O₃, whereas its physical properties indicate that it is an equimolecular mixture of nitric oxide and nitrogen tetroxide. Thus it is probable that Lunge's theory as to the primary formation of nitrogen peroxide is correct, as this oxide would then combine with unoxidised nitric oxide to form nitrogen trioxide:

N₂O₄ + 2NO = 2N₂O₃.

It has been shown that under ordinary conditions nitrogen peroxide, nitric oxide, and nitrogen trioxide can exist in equilibrium, which means that oxidation of nitric oxide in the presence of an absorbent results in the removal of nitrogen tetroxide, together with an equivalent of nitric oxide in the form of nitrogen trioxide.

The use of suitable absorbents is of fundamental importance in determining the products of oxidation of nitric oxide. In the case of concentrated sulphuric acid, both Raschig and Lunge agree that a mixture of nitric oxide and nitrogen peroxide is absorbed as nitrogen trioxide to produce nitrosyl-sulphuric acid:

N₂O₃ + 2H₂SO₄ = 2HNOSO₄ + H₂O.

Nitrogen tetroxide is also absorbed by concentrated sulphuric acid, and a rise in temperature causes the reaction to move in the reverse direction:

N₂O₄ + H₂SO₄ ⇔ HNOSO₄ + HNO₃.

Aqueous alkalies do not completely absorb either nitrogen trioxide or nitrogen tetroxide, since secondary reactions occur which are due to the water present. These will be discussed in detail under the respective oxides.

The kinetics of the oxidation of nitric oxide can be dealt with in two ways, according as to whether the final product is NO₂ or N₂O₄. If it be assumed that the reaction is represented by

2NO + O₂ ⇔ 2NO₂,

i.e. that the nitrogen tetroxide is completely dissociated, then the reaction velocity is expressed by the usual trimolecular equation

dx/dt = K(a-x)²(c-x),

where 2a = initial concentration of NO,
c = initial concentration of O₂,
x = amount transformed in time t.

On integration the velocity constant is given by the equation



The value of K can be calculated from the experimental work of Lunge and Berl on the oxidation of mixtures of nitric oxide and air.

The above expression ignores the change in volume in the system, although this error may be minimised by calculating K for successive small intervals of time. An expression is given by Wegscheider to correct for this volume error both in the equation -

2NO + O₂ ⇔ 2NO₂ (1)

and also in the reaction which assumes that one-half of the N₂O₄ is dissociated:

2NO + O₂ ⇔ 0.5N₂O₄ + 0.5(2NO₂) (2)

This expression is -



where M₁ = initial concentration of NO,
M₂ = initial concentration of O₂
μ = volume of oxygen changed in time t in 100 volumes of mixture,

V = total initial volume of gas at temperature T and pressure P, these two latter being constant under the conditions of the experiment,

b = 1 in equation (1) and 1.5 in equation (2).

Further work on the oxidation of nitric oxide has been carried out by Bodenstein; he investigated the variation of the velocity constant with temperature; Todd investigated this from the standpoint of constant volume and constant pressure reactions.

Nitric oxide can easily be detected by the formation of brown fumes when brought into contact with air or oxygen. This oxidation can also be utilised as a means of estimation by absorbing the nitrogen trioxide by dry potassium hydroxide,

4KOH + 4NO + O₂ = 4KNO₂ + 2H₂O,

as four-fifths of the total contraction observed by the addition of a known volume of air or oxygen is nitric oxide. Monoethyl aniline may be used instead of caustic potash with advantage, as it does not absorb N₂O, N₂, CO₂, or CO. Another method of estimation is to observe the contraction which occurs when a gas containing nitric oxide is brought into contact with acidified potassium bichromate. In this case complete oxidation to nitric acid takes place.

Potassium permanganate will also bring about a similar oxidation, but in this case an acidified standard solution is used, and the amount of permanganate used up is determined by ordinary volumetric methods.

Reduction of nitric oxide to nitrogen may be brought about by passing the gas mixed with hydrogen over heated platinum black.

None of the foregoing methods are applicable if a mixture of gases containing nitric oxide, nitrogen trioxide, and nitrogen peroxide is to be analysed, owing to the interaction of nitric oxide and nitrogen peroxide:

NO + NO₂ ⇔ N₂O₃.

An approximate estimation may be made by absorbing the gases in an aqueous solution of sodium hydroxide and determining the amounts of nitrite and nitrate present, formed by the following reactions: -

N₂O₃ + 2NaOH = 2NaNO₂ + H₂O;
N₂O₄ + 2NaOH = NaNO₃ + NaNO₂ + H₂O.

A mixture of gases which contains the nitric oxide in smaller amount than that required by the tetroxide to form N₂O₃ may be analysed by absorption in concentrated sulphuric acid (85 to 95 per cent.):

N₂O₃ + 2H₂SO₄ = 2OH.SO₂.ONO + H₂O;
N₂O₄ + H₂SO₄ = OH.SO₂.ONO + HNO₃.

This nitrosyl-sulphuric acid can be estimated by means of potassium permanganate, and the total nitrogen by the nitrometer, and the relative proportions of the gases can thus be calculated.

When the nitric oxide is in excess of the amount required by the peroxide to form N₂O₃, a combination of two methods may be used. The absorption is first effected in concentrated sulphuric acid, and the gases are then passed through an absorption tube containing acidified potassium permanganate.

Historical

Since the times of Priestley and Cavendish in the latter part of the eighteenth century, it has been known that the passing of electric sparks through air gives " nitric acid," i.e. oxides of nitrogen absorbable by potash with the production of nitre. It was noticed by Crookes that the combination occurred in a high-tension arc. He was awake to the necessity of securing a sufficient supply of combined nitrogen for agricultural purposes. The nitric oxide synthesis was from this time largely investigated with a view to technical use. Rayleigh passed sparks from an induction coil through an oxygen-nitrogen mixture containing 36 per cent, of the latter gas, in the presence of excess of alkali which was circulated continuously. The yield was at the rate of 56 grams HNO₃ per kilowatt hour.5 A number of arcs drawn out by rotation was used to increase the concentration of electrical energy, but it was found that this soon reached a useful limit, beyond which the yield of NO no longer increased. About the end of the century the equilibrium conditions began to be studied, and especially the degrees of combination which could be effected by heat alone.

Although some combination can be effected when the gases are passed over platinum black at 250° C., yet high temperatures are generally required for the synthesis. The oxides of nitrogen are formed in many combustions, such as that of carbon in highly compressed air, or of hydrogen, or in the presence of a platinum wire raised to a white heat by an electric current. The combustion of hydrogen and nitrogen under lime-water seems to afford a possible method of " fixation."

The Thermal Equilibria

Nitric oxide is formed from its elements with absorption of 21.600 Cals. per mol. of NO. The gas is in a state of false equilibrium at the ordinary temperatures, although some observers had found a considerable stability towards heat. When heated to between 500° and 1000° C., however, in a tube it begins to decompose, and the decomposition is considerable at 1200° C. At high temperatures small equilibrium amounts of NO are formed from the elements.

The thermal equilibria in mixtures heated by arcs showed an increase in the amount of NO with diminution in the size of the arc, and therefore presumably at higher temperatures.

On account of the endothermic character of the reaction

N₂ + O₂ ⇔ 2NO,

it shifts more to the right at the highest temperatures.

The equilibrium percentages of NO have been found experimentally:

Temperature, ° abs.	1811	1877
Per cent, of NO (x)	0.37	0.42

From these a constant can be calculated, and from the reaction isochor,



the variation of the constant with temperature is determined. Over the range of temperatures given, the heat of reaction " q " does not vary much with the temperature:

T. ° abs.	1500	1800	1811	2000	2033	2195	2500	2580	2675	2900	3200
x (found)	. . .	. . .	0.37	. . .	0.64	0.97	. . .	2.05	2.23	. . .	5
103K	2.48	8.51	9.16	15.3	15.9	24.2	45.5	52.1	56.8	82.9	117.9
x1	0.10	0.34	0.35	0.61	0.67	0.98	1.79	2.04	2.37	3.20	4.43

x₁ is calculated in the case of air by means of the equation -



When these equilibria are obtained by heating the gases to the highest available temperatures, they must be " fixed " by rapid cooling. In this connection the velocities of the direct and inverse reactions are important. The following values refer to air at atmospheric pressure:-

1. Time required to half decompose the equilibrium amount of NO -
   
   

   Temperature, ° abs	900	1500	1900
   Time in secs.	7350	3.3 mins	1 sec

2. Time required to form half the theoretical amount of NO -
   
   

   Temperature, ° abs	1500	1900
   Time in secs	1810 mins	2.08 mins

The decomposition of nitric oxide at the surface of a heated platinum wire has been studied, and the reaction 2NO = N₂ + O₂ found to be unimolecular with respect to nitric oxide. The reaction is retarded by oxygen but is not influenced by nitrogen.

The Effects of Electrical Discharge

Although the silent discharge has long been known as a powerful agent in chemical synthesis, the spark or arc discharge had, until the last quarter of the nineteenth century, been regarded chiefly, if not solely, as a convenient means for localising a high temperature inside a closed vessel. The elucidation of the mechanism of conduction through gases has shown that the molecules are profoundly affected, and that this may lead to specific chemical effects, apart from those due to the temperatures of the spark, etc.

The conducting particles in a gas are free electrons and molecules or atoms which have been positively or negatively charged by the gain or loss of one or more electrons. These gaseous ions may combine, with mutual neutralisation of charges, thus accounting for such chemical effects as the synthesis of (CN)₂, HCN, etc., by the silent discharge, of ozone perhaps by the union O₂⁺ and O^-, and of NO perhaps by the union of N^- and O⁺. These compounds are formed in appreciable quantities at ordinary temperatures by the silent discharge, and correspond to electrical equilibria, which are different from the ordinary equilibria.

When a gradually increasing voltage is applied to two electrodes separated by a gas or mixture of gases, a small current or non-luminous discharge passes at first which produces no chemical effects. It is due to the ions and electrons naturally present in the gas. At a certain voltage, which naturally varies with the material, dimensions, and separation of the electrodes, a glow or brush discharge is seen. This discharge is stable, for an increase of current necessitates an increased voltage, therefore at a given voltage the current reaches a maximum. This kind of discharge produces ions by collision between electrons and gas molecules, and hence initiates chemical ehanges. If the current "I" is increased still further, the voltage "E" falls, the conductivity rapidly increases, and the brush discharge is succeeded by the high- tension arc. In this the charged particles are produced in large quantities by impact, the resistance of the gap decreases more rapidly than the current increases, or an increase of current is accompanied by a decrease in the voltage. Such arcs are therefore electrically unstable. The temperature rises continuously and the resistance decreases until the arc passes, with a slight discontinuity in the "E, I" graph, into the ordinary low-tension or lighting arc, in which the current is mainly carried by electrons from the kathode and positively charged ions from the material of the anode. On this account, in spite of its high temperature, such an arc is useless for the synthesis of NO. It is necessary to maintain the high-tension arc by electrical devices. A resistance of the metallic kind (which obeys Ohm's Law) is placed in series with the direct current arc. An increase of current increases the voltage drop across this, and therefore decreases that available for the arc. In alternating current arcs the resistance is replaced by an inductance.

In such arcs an electrical equilibrium is set up. between the nitrogen and oxygen which is not the same as the thermal equilibrium, and which only exists when and where the gases are under the influence of the discharge, or perhaps for a short distance beyond (or time after) the discharge. This kind of arc is probably due to the formation of atomic nitrogen and oxygen.

In a high-tension arc at 2000° C., NO can be produced in amounts which would be in thermal equilibrium with N₂ and O₂ at 4500° to 5000° C. Allmand and Ellingham consider that the law of mass or concentration action does not apply to these electrical equilibria.

The theory that the electrical discharge merely supplies a localised high temperature does not account for the facts.

It was shown on the experimental scale that the gases sucked from the middle of a 3 cm. steady vertical arc through a water-cooled platinum capillary tube contain 5 per cent, of NO, corresponding to the thermal equilibrium at 3000° C.

Haber and his co-workers also used a vertical arc between electrodes of platinum or iron. Using air, and with iron electrodes, they obtained nearly 10 per cent, of NO, and with a mixture of equal volumes of N₂ and O₂ passed at a slow rate (0.8 litre per hour), 14.4 per cent, of NO. The voltages employed were from 2050 to 4800, and the current from 0.205 to 0.306 amp. The temperatures corresponding to these yields in the purely thermal equilibrium were certainly not reached. As the arc was an alternating one and was extinguished at every half-period, there must have been great variations of temperature, which would still further reduce the mean value of this. A short direct current arc at 2700° C. with a cooled anode gave up to 9 per cent, of NO from air. If all the air had been heated to this temperature (which can never occur), the maximum yield should have been not more than 4 per cent.

Oxides of nitrogen may also be obtained from the air by silent discharge at the ordinary temperatures, in which case the combination must take place through the ions produced by this discharge.

It has been further discovered that the conditions of maximum yield per hour in any given apparatus (space-time yield) do not correspond to a very slow passage through the high-tension arc (with maximum percentage conversion), but to a moderate speed. The presence of water-vapour diminishes the yield.