has recently proposed an equation re- terns, i t cannot be stated whether the chelates of

lating AHo for complex formation with the ioniza- ethylenediaminetetraacetate are abnormal in this
tion potentials of the metal atoms and the radii of respect.
the corresponding metal ions. I t was not found Acknowledgment.-The writer is indebted to Dr.
possible, however, to correlate the data in Table I Frank Byrne and his group for the metal ion
in this way. Since there are insufficient data to test analyses.
adequately the equation of Williams with other sys- EASTPITTSBURGH, PA.

[COSTRIBUTIOS FROM ARTHUR D. LITTLE,I N C . ]

The Dissociation Pressure and Free Energy of Formation of Ammonium Nitrate

BY GEORGE
FEICK
RECEIVEDJUXE 14, 1954

The dissociation pressure and enthalpy of liquid ammonium nitrate have been measured for the temperature range from
190 t o 270'. From these data, the free energy and entropy have been calculated for the liquid, and for the solid a t 25".
In conkmation of the results of other investigations, the data indicate that the vapor of ammonium nitrate is completely
dissociated into NH3 and HNOs in the range of temperatures investigated.

The thermodynamic properties of ammonium to the objection that the evolved decomposition products
nitrate are of interest in connection with the study may have had an appreciable effect on the observed pres-
sure. The magnitude of this effect may be estimated with
of its thermal decomposition,' especially in the the aid of the following equation
range of temperatures above its melting point
(169.6'), for which existing data are not adequate.
For this reason, the dissociation pressure has been
measured a t temperatures up to 250' and the en- where P is the observed pressure, p is the true dissociation
pressure, AH, is the molal heat of dissociation of "4x03,
thalpy of liquid ammonium nitrate has been meas- AH, is the molal heat of decomposition into NzO and H20,
ured up to 270'. From these data, A H and AF and Q is the heat added t o the system per mole of NHcN03
have been calculated as functions of temperature decomposed. For the development of this equation, see
for the reaction reference 1.
Since the effect of decomposition is most marked a t high
"ix03(1) ---7t SHa(g) f HS03(g) (1) temperatures, a check was made of the value of 41 mm.
measured a t 249.1'. At this temperature, AH, is 38.3 kcal.
on the assumption that the dissociation is complete (see below), and AH, is -14.0 kcal.'. Q was estimated as
in the vapor phase.2 These results were used to follows: the rate of pressure rise in the apparatus with the
calculate the free energy and thence the entropy of stopcock to the vacuum pump closed was 0.233 mm./sec.
molten ammonium nitrate a t various temperatures. a t a pressure of 41 mm. with 5 g. of "*NOS in the tube.
From the known volume of the apparatus (1550 ml.), the
The resulting entropies were checked against each decomposition rate was found t o be 7.07 X mole/sec.
other by showing that substantially identical re- The heating rate was approximated by noting that 4 ml. of
sults were obtained by reducing each value to 25'. water was completely vaporized in 50 sec. from a similar
The resulting value is in fair agreement with that vessel under the same conditions of heating. This corre-
sponds t o a heating rate of 0.0476 kcal./sec. Q is then
calculated from the activities of its ions in aqueous found by dividing the heating rate by the decomposition rate:
s o l ~ t i o n . ~The agreement of these values may be Q = 0.0476/7.07 X 10-8 = 6740 kcal./mole.
construed as supporting the assumption of complete Substitution of the above values in equation 2 gives:
dissociation of the vapor, as well as the numerical P / p = 1.0085, which indicates that the observed pressure
differs from the true dissociation pressure by less than
values of the vapor pressure and enthalpy. 1%. This difference will be ignored since it is of the
Experimental Procedures.-J. T. Baker C.P.ammonium same order of magnitude as the error in reading the mer-
nitrate was used throughout these experiments with no puri- cury manometer. For pressure readings a t lower tempera-
fication other than drying a t 110" before use. tures, the error due to thermal decomposition will, of course,
Because ammonium nitrate decomposes a t an appreciable be much smaller.
rate even a t temperatures as low as 190", the method for The enthalpy of molten "4x0s was measured by the
measuring vapor pressure must be chosen so that the evolved method of mixtures in which the molten salt was poured
gases ( X 2 0and H?O) interfere as little as possible. For this into water contained in a 500-ml. silvered dewar flask.
reason, a modified boiling point apparatus was used for the The necessity for jacketing the flask was avoided by chooi-
present purpose since the evolved gases are continuously ing the conditions so that the final temperature mas near
swept away from the zone of measurement. This apparatus room temperature. Although it is not commonly consitf-
was similar to that described by Cottrel14 except that an ered good practice to use this method for materials at tem-
iron-constantan thermocouple was used to measure the tem- peratures above the boiling point of water, because of the
perature. The pressure in the system was controlled by a possibility of loss by spattering and vaporization, the
vacuum pump connected through a stopcock and was meas- method was considered the most practical for the present
ured by means of a butyl phthalate manometer (up to 20 purpose because the results are not affected by the doecorn-
m m . ) or a mercury manometer (above 20 mm.). I t was position of the molten salt, which is very rapid a t 270 , and
found necessary to pump continuously on the system during appreciable a t lower temperatures. The danger of spatter-
the measurements in order t o keep the pressure constant. ing was mitigated to some degree by using ice-cold water
Under this latter condition, the measurements are open and by the large negative heat of solution of NHaYOa. It
was found, in fact, that slight spattering occurred only :it
( I ) G. Feick and R >Hainer,
I. THISJOURNAL, 76, 5860 (1954). the two highest temperatures measured.
1 2 ) P. C . Ray and S. C. Jan&, J . Chem. Soc., 103, 1565 (1913). Another source of error in this procedure was the possi-
(3) C. C. Stephenson, private communication. bility that the molten nitrate retained some of the water
( 4 ) 1'. G Cottrell, THISJOURNAL. 41, 721 (1919). formed by its decomposition. At temperatures above
Nov. 20, 1954 DISSOCIATION OF AMMONIUM
PRESSURE NITRATE 5859

220°, this water is probably less than 1 % . 6 A t lower tem- I I I I I I

peratures, greater amounts of water may be retained, but
the reaction is slow and care was taken to avoid keeping
the melt hot for long periods of time.
Dissociation Pressure Data.-The observed
vapor pressures are given in Table I. The average
value of d(ln p)/d(l/T) corresponds t o a heat of
vaporization of 39.8 kcal./mole, which is in approxi-
mate agreement with the values of 38.5 to 39.1 cal-
culated from thermal data for this temperature
range.
TABLE
I
PRESSURE,FREEENERGYAND ENTROPY
DISSOCIATION OF
DISSOCIATIONAND ENTROPY OF AMMONIUMNITRATE
SNH~NOS SNHtNOa
6.15 K CAL/MOLE
AFdiaa., ASdiaa., atl a t 25'
Temp., Pressure, kcal./ cal./deg. cal./dbg. cal./dei.
oc. mm. mole mole mole mole
188.2 3.25 11.28 60.3 59.2 36.3 71 I I I I I
205.1 7.45 10.12 60.1 60.3 36.0 160 160 200 220 240 260 280
215.9 11.55 9.49 59.7 61.1 35.9 TEMPERATURE- ' C .
223.1 15.80 9.01 59.6 61.8 36.1 Fig. 1.-Enthalpy of liquid NH4N03 based on solid ( I V ) a t
236.7 27.0 8.15 59.3 62.7 35.9 25 '.
249.1 41.0 7.47 58.9 63.8 36.1
- nitric acid vapor is given only to 500°K., graphical
Best value 36.0 extrapolation was made for the higher tempera-
Enthalpy Data.-The data leading to the en- tures. The results are given in Table 111.
thalpy of molten NH4N03are summarized in Table TABLE
I11
11. In making the calculations, the heat of
solution data were taken from the National Bu- HEATOF DISSOCIATION OF AMMONIUMNITRATE
reau of Standards Tables6 and the specific heats of Temp., "C. 169.6 200 250' 300' 350'
aqueous NH4N03 solutions were taken from AH, kcal./rnole 39.33 38.92 38.25 37.65 37.12
Gmelin.' Based on extrapolated values for the enthalpy of " 0 3
The resulting values of enthalpy are plotted (PI.
against temperature in Fig. 1. The data appear to
be precise to about *lOO cal. and are represented Free Energy and Entropy Calculations.-The
within their precision by a straight line whose slope free energy of dissociation was calculated from the
corresponds to a heat capacity of 38.5 cal./mole deg. dissociation pressure a t each of the observed tem-
Extrapolation to the melting point gives an en- peratures by means of the relation
thalpy of 8.15 kcal./mole for liquid NH4N03 a t AF = - R T l n R = - R T l n qPZ
(3)
169.6O, based on the solid (IV) a t 25'.
TABLE
I1 where p is the dissociation pressure in atmospheres.
ENTHALPY
OF LIQUID
AMMONIUM
NITRATE From these values, the entropy of dissociation was
Enthalpy found, using the equation
Temperatures, Weights, g. above 2 5 O
NHcNOa Water
OC.
Mixture NHiNOi Water' kcal./mol; A F = AH - TAS (4)
185.5 1.12 29.1 178.4 302.6 8.72
and the previously calculated values of AH.
190.25 1.74 11.72 65.9 302.1 9.00
The absolute entropy of ammonium nitrate a t
211.27 1.76 13.65 65.0 294.5 9.74
each of the above temperatures was calculated, us-
217.51 2.16 15.24 69.8 304.3 9.95
ing the entropy values for ammonia and nitric acid
227.36 1.56 15.52 68.2 300.3 10.49
given by Kelley.8 Values for the latter substance
240.54 1.60 26.85 109.8 297.9 10.91
above 500'K. were found by extrapolation. The
269.14 1.92 14.56 49.1 301.0 11.95
results of the foregoing calculations are summarized
0 The effective heat capacity of the calorimeter was found

by measurement to be equivalent to 19.5 g . of water.

in Table I.
The data may be checked for internal consistency
Heat of Dissociation.-The above data were by reducing the absolute entropy values to a com-
used to calculate AH for reaction 1 a t various tem- mon temperature, which was chosen as 25'. The
peratures. The enthalpies for the reactants a t 25' transition temperatures and heats of transition for
were taken from the Bureau of Standards Tabless the various solid forms of ammonium nitrate were
and the enthalpies of the products a t higher tem- taken from the Bureau of Standards Tables6
peratures from Kelley.8 Since the enthalpy of The heat capacities used in the calculation are
(5) "Gmelins Handbuch der anorgauischen Chemie, 8 Aufl. Sys- summarized in Table IV, together with the sources
tem Nummer 23: Ammonium. Lieferung 1," Verlag Chemie, Berlin, from which they were taken. The value of 45.6
1936, p. 123. kcal./mole for c, I was found by difference from the
(6) "Selected Values of Chemical Thermodynamic Properties," measured enthalpy of the liquid. Although the ac-
Circular of the National Bureau of Standards, 500, 1952.
(7)Reference 5 , p. 124. curacy of this value is undoubtedly poor, its effect
(8) K. K. Kelley, Bureau of Mines Bulletin 476. on the over-all entropy calculation is small.
5Sc;O VBICR
CRORGP AND R. M. HAJNER Vol. 7ti

TABLE
IV obtained by Stephenson from low-temperature calo-
HEATCAPACITY
DATA rimetric data.a
Form Cp, cal /mole, OC. Source From the entropy and heat of formation the
c, I V 33.6 Ref. 8 standard free energy of formation of ammonium
c, I11 28.4 Ref. 9 nitrate is found to be -43.82 kcal./mole.
c, 11 34.1 Ref. 9 The author takes pleasure in acknowledging the
c, I 45.G See text continuing advice and criticism of Dr. Clark C.
Liq. 38.5 This work Stephenson of Massachusetts Institute of Technol-
ogy during the course of this work.
The resulting values for the entropy are given in This work was performed under the auspices of
Table I and are in good agreement except for the
value a t 188.2”. The best value for the absolute the United States Coast Guard and the Advisory
entropy of ammonium nitrate a t 25‘ may be taken Committee on Hazards of Ammonium Nitrate
as 36.0 cal./mole “C. The agreement of these val- Transportation of the National Research Council.
ues confirms the assumption that the vapor of am- Permission to publish this article has been granted
monium nitrate is completely dissociated in the by the United States Coast Guard but neither of
range of temperature investigated. the sponsoring agencies necessarily approves the
The value of entronv obtained by this method technical opinions or conclusions herein expressed.
(86.0 e.u.) may be cort&ired with thevalue of 36.11 MASS.
CAMBRIDGE,

[ C O N T R I B L ~ T I O NFROM ARTHLTRD. LITTLE,

INC.]

On the Thermal Decomposition of Ammonium Nitrate. Steady-state Reaction

Temperatures and Reaction Rate
EY GEORGEFEICK AND R. M. HAINER
RECEIVED JUNE 14, 1954

When ammonium nitrate is heated, it decomposes exothermically into nitrous oxide and water. A t the same time, it
dissociates endothermically into ammonia and nitric acid vapor. As a result of these two concurrent reactions, the molten
mass of nitrate tends to reach a steady-state temperature, which is a function of the ambient pressure and of the heating
rate. At pressures near atmospheric, the reaction is thus limited to rates which are very SIOIV compared to those of explosive
reactions. An equation has been derived relating the temperature to the above variables, which is in good agreement with
experiment over the limited range of conditions studied. The constant temperature reaction has been applied to the
measurement of reaction rate.

Although it is well known that ammonium ni- tures a t which the decomposition occurs with easily
trate decomposes exothermically when heated, measurable velocity. When heat is added, even
giving N 2 0 and HzO, its behavior on heating is not from a very hot source, the temperature of the
typical of an exothermic reaction, in that in small NH4N03is limited by its own dissociation, to m l -
quantities it usually shows little tendency toward ues a t which the decomposition rate is still coni-
self-accelerating decomposition, or even toward paratively moderate. At higher pressures, however,
continued reaction after the source of heat is re- dissociation is repressed and the temperature can
moved.’ Sherrick2 failed to cause an explosion of rise until an explosive rate of decomposition is at-
unconfined “4x03, even by bringing it into con- tained. The application of these results to the ques-
tact with molten iron from the “thermit” reaction. tion of the explosion hazard of ammonium nitrate
On the other hand, Herquet3 has shown that suit- will be discussed in more detail by Hainer.4
ably confined NHlNO8 is capable of rapid, self-ac- Steady-state Reaction Temperature.-The main
celerating decomposition a t initial temperatures reactions of interest are irreversible decomposition
of 260 to 280’.
The present paper offers an interpretation of such
NHaNOail) --+ NiO(g) +
2H?O(g) (1

observations by showing that molten, decomposing and reversible dissociation

ammonium nitrate tends to seek a definite and pre- NH~NOI(I) NHa(g) $- HNOa(g) (2)
dictable limiting temperature which is a function of
the external pressure and of the heat supplied to, or Saunderss has shown that reaction 1 accounts for
removed from the molten mass by its surroundings. 98yo of the irreversible decomposition of NH4N03
This limitation of temperature is the result of the in the temperature range 230 to 260’. The side
endothermic dissociation of NH4N03into gaseous reactions will be considered negligible for the pres-
NH, and “Os, a reaction which, under the steady- ent purpose. The enthalpy change for this reaction
state temperature conditions, absorbs all the heat (4) R. M. Hainer,“The Application of Kinetics to the Hazardolis Rr-
available froin the decomposition. At pressures havior of Ammonium Nitrate.” Paper presented t o the Fifth Inter-
near atmospheric, and in the absence of externally national Symposium on Combustion, August, 1954. R . M. Hainer
and W. C. Lothrup, “Thermal Hazard in Ammonium Nitrate and High-
supplied heat, the reaction is limited to tempera- Percentage Ammonium Nitrate Materials,” presented to the Division
(1) G. Feick and R. M. Hainer, Nature, 173, 1188 (1954). of Industrial and Engineering Chemistry, Section C, American Chemi-
( 2 ) J. L. Sherrick, Army Ordnance, 4 , 237 (1924). cal Society. September. 1954.
(3) M. L. Herquet, Ezplosifs. I,29 (1952). ( 5 ) H. L. Saunders, J . Chcm. Soc., l a ¶ , 698 (1922).