View Article Online / Journal Homepage / Table of Contents for this issue

3863
Published on 01 January 1918. Downloaded by Christian Albrechts Universitat zu Kiel on 26/10/2014 01:02:45.

XXXIL- The Synthesis of Ammonia at IrIi,9h Te.lu~pe~*a-

tures. Part -€I.
By EDWARD
BRADFORD
MAXTED.
IN a recent paper (this vol., p. 168), the equilibrium between
nitrogen, hydrogen, and ammonia a t high temperatures was dis-
cussed from a thermodynamical point of view, and evidence was
brought forward showing that in all probability the ammonia con-
tent of such a gas mixture in equilibrium, after decreasing with
increasing temperature, eventually passes through a minimum and
finally rises once more. On the basis of this theoretical indica-
tion of the possibility of a therrnal synthesis of ammonia, it, was
shown experimentally that considerable yields of ammonia may be
obtained by cooling a mixture of nitrogen and hydrogen extremely
rapidly from the temperature of the oxy-hydrogen flame to that
of the room.
I n continuation of the above work, and especially in view of the
known fact that a mixture of nitrogen and hydrogen may be com-
pletely converted into ammonia by sparking in a eudiometer over
dilute acid (Donkin, Proc. R o y . SOC.,1873: 21, ZSl), a n investiga-
tion was begun of thei yields of ammonia obtainable from a rapidly
cooled, high-tension arc, also by sparking, the present paper being
a summary of the results obtaiiied by induction discharge modified
in such a way as to constitute a small, high-tension arc burning
within a capillary tube, through which the mixture of nitrogen
atid hydrogen was passed.
By confining the action to a capillary tube, in the manner
 View Article Online
AMMONTA A T HJ(STT TRMMPBRATTTRRS. PART I T . 387
Published on 01 January 1918. Downloaded by Christian Albrechts Universitat zu Kiel on 26/10/2014 01:02:45.

described, the gas to be treated may be broughtl uniformly into

contact with the discharge, and it was found easily possible t o
obtain a t atmospheric pressure yields of ammonia amounting t o
1.5 per cent. by volume of the gas-mixture taken for treatment.
Tnduction sparks, as such, were found t o exert a comparatively
feeble action on the synthesis, energetic formation of ammonia only
taking place when the electrodes were brought sufficiently close
together to transform the ordinary spark discharge into a small,
high-tension arc, accompanied b y cz visi1)le arid apparently con-
t i 11i~oiisfl ant e of high tern peratnre.

E x P E n I M E N T A I,.
The apparatus employed for the series of exyerimeiits about, to
be described consisted of a capillary glass tube having an internal
diameter of 0.65 mm., an external diameter of 5 mm., and a
length of about 10 cm. Platinum wire electrodes, 0.25 min. thick,
were sealed into the capillary tube in such a way as t o leave a
spark-gap of the size required, whilst the passage of gas through
the t a b s was effected by means of fused-on glass side-tubes.
It was found inadvisable, on account, of frequent fractures, t o
employ a heavy discharge, the most satisfactory conditions for the
investigation being obtained with a “two-inch ” coil, supplied with
a primary current of from 3 to 3.5 amperes from the laboratory
220-volt main. Pure hydrogen and nitrogen for the synthesis
were mixed in the proportion o i three t o one in a large gasholder
and compressed into a cylinder f o r convenience in use. This mixed
gas was passed a t carefully determine8d rates through the capillary
spark-gap, the ammouia formed being absorbed in dilute acid and
estimated by means of Kesslsr’s solution.
The first point to loo investigated was the influence of the size
of the spark-gap on the yield o€ ammonia, this gap being varied
f r o m 10 t o 0.5 mm. while the rate of pzssage of the gas was first
kept constant atl 40 C.C.per hour, and, secondly, varied in such a
way that the time of contact was kept a t 0.0015 second f o r spark-
gaps of various sizes. Table I summarises the results obtained
by the first of these two methods, namely, with a constant, rate of
flow, table I1 bhosel obtained by the second, that is, w i t h a constant
time of contact with the spark-gap, the pi-iinary current being i n
every case 3.5 amperes at 220 volts.
 View Article Online
388 SYNTHESIS OF AMMONIA AT HIGH TEMPERATURES. PART n.
Published on 01 January 1918. Downloaded by Christian Albrechts Universitat zu Kiel on 26/10/2014 01:02:45.

TABLEI.
Rate of Flow of Nydrogeiz-~Vitrogen M i x t u r e 40 C.C. per hour.
Percentage of am-
Length. of spark- Nature of monia by volume
gap in mm. discharge. in issuing gases.
10 Spark ..................... 0.1
6 Incipient flame ......... 0.25
1.5 Arc flame ............... 0.8
0.5 Arc flame ............... 1-1

TABLE 11.
Estinaatcd T i m e of Contnct, 0.0015 see.
Percentage of am-
Length. of spark- Nature of monia by volume
gap in mm. discharge. in issuing gases.
10 Spark ..................... less than 0.01
5 Incipient flame ......... 0-04
1.6 Arc flame ............... 0.5
0.5 Arc flame ............... 1-1

It will be seen t h a t as the size of the spark-gap is decreased, the

flame loses its well-defined spark-like character and becomes a
small but intensely hotl high-tension arc, this flame, probably by
virtue of its high temperature combined with the rapid cooling
effect afforded by the relatively cold walls of the glass capillary
tube, possessing the power of inducing the combination of nitrogen
and hydrogen to an extent, obtainable otherwise only by the action
of a catalyst under a high pressure. The rate of flow of the
nitrogen-hydrogen mixture through the capillary tube was
measured by allowing the gas issuing from the reaction tube to
pass through a small, specially constructed gas-washing bottle, in
which the ammonia was absorbed by very dilute sulphuric acid,
the number of bubbles which formed per minute being counted
and the volume of a bubble being known. The estimation of
ammonia in the resulting solution was carried out by means of
Nessler’s reagent.
It is difficult to estimate the temperature of the small high-
tension arc formed, which, however, for the purpose of obtaining
rough comparative figures for the various times of contact, has
been assumed t o be a t approximately 3000°, b u t t h e effect of
increasing temperature on the yield of ammonia formed is clearly
seen. This rise in yield with increasing temperature a t high
temperatures (in contradistinctJon from the decrease obtained a t
moderate temperatures) agrees with the results already reported.
When working with small spark-gaps, a considerable deposit of
platinum, removed from the electrodes by volatilisation and other
 View Article Online
ALTdEN ATOMIC! AND MOT,ECUTLAR NTJMBFRS. 389
Published on 01 January 1918. Downloaded by Christian Albrechts Universitat zu Kiel on 26/10/2014 01:02:45.

means, was obtained on the sides of the capillary tube, and in some
cases the tube became fractured by the intense local heat, although
in general the tube itself remained moderately cool, and in any
case showed no tendency to soften.
The above results having demonstrated the necessity for employ-
ing as small a spark-gap as possible, in order t o obtain the maxi-
mum local heating effect and consequently the maximum yield of
ammonia, the effect of varying the time of contact with a spark-
gap of constant small length (0.5 mm. and 1-5 mm.) was
investigated. With an exceedingly small arc of the nature
described, sufficiently rapid cooling for the retention of the
ammonia formed is readily obtained by the action of the tube
itself, the difficulty being rather the uniform heating to arc
temperature of the comparatively rapid current of gas.
Table I11 demonstrates this point, it being found t h a t any
increase in the time of contact of the gas with the small arc,
within the limits studied, causes a corresponding rise in the
ammonia-content of the issuing gas.
TABLE111.
Percentage of
Velocity of Estimated time ammonia by
Spark-gap flow in C.C. per of contact volume in
i n mm. hour. in seconds. issuing gases.
(a) 0.5 97.2 O*OOQG 0.7
0.6 40.5 0.0015 1.1
0.5 18.9 0 *0032 1.4
0.5 15.0 0-0041 1.5
(b) 1.6 670.0 0.00027 0.95
1.6 130.0 0.0014 0.5
1- 5 64.8 0.0028 0.6
1.5 33.8 0.0052 0.87
The above results appeared t o offer considerable encouragement
for the investigation of the formation of ammonia by rapidly
cooled high-tension arcs of a larger size. The results of such work
will be communicated in a later paper.
[Received, March 2 1 st, 19 1 8.1