Determination of structural, thermodynamic and phase properties in the Na2S-H2O system for application in a chemical heat pump

Article abstract of DOI:10.1016/S0040-6031(02)00158-2

Structural, thermodynamic and phase properties in the Na₂S-H₂O system for application in a chemical heat pump have been investigated using XRD, TG/DTA and melting point and vapour pressure determinations. Apart from the known crystalline phases Na₂S·9H₂O, Na₂S·5H₂O and Na₂S a new phase Na₂S·2H₂O has been proven to exist. Na₂S·1/2H₂O is not a phase but a 3:1 mixture of Na₂S and Na₂S·2H₂O, presumably stabilised by very slow dehydration kinetics. The vapour pressure-temperature equilibria of the sodium sulphide hydrates have been determined and a consistent set of thermodynamic functions for these compounds has been derived. XRD measurements indicate the topotactic character of the transitions between the hydration states.

Full text of DOI:10.1016/S0040-6031(02)00158-2

Thermochimica Acta 395 (2003) 3–19
Determination of structural, thermodynamic and phase
properties in the Na2S–H2O system for application
in a chemical heat pump
R. de Boer, W.G. Haije*, J.B.J. Veldhuis
Energy Research Centre of the Netherlands, P.O. Box 1, 1755 ZG Petten, The Netherlands
Received 1 March 2002; accepted 6 March 2002
Abstract
Structural, thermodynamic and phase properties in the Na2S–H2O system for application in a chemical heat pump have been
investigated using XRD, TG/DTA and melting point and vapour pressure determinations. Apart from the known crystalline phases
1
2
Na2Sꢀ9H2O, Na2Sꢀ5H2O and Na2S a new phase Na2Sꢀ2H2O has been proven to exist. Na2Sꢀ H2O is not a phase but a 3:1 mixture
of Na2S and Na2Sꢀ2H2O, presumably stabilised by very slow dehydration kinetics. The vapour pressure–temperature equilibria of
the sodium sulphide hydrates have been determined and a consistent set of thermodynamic functions for these compounds has
been derived. XRD measurements indicate the topotactic character of the transitions between the hydration states.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Sodium sulphide hydrate; Vapour pressure; Phase diagram; Thermodynamic properties; Heat pump
1. Introduction
cooled and dissipated into the environment. On an
European scale this amounts to approximately 10
times this number [1]. In recent years much effort
has been put into the development of heat driven solid-
sorption chemical heat pumps and heat transformers
[2] for heating as well as cooling purposes. The
working principle of this apparatus is based on the
exothermic absorption and endothermic desorption of
vapour in solids, e.g. water in silica or zeolites [3],
ammonia in metal salts [4] or in carbon [5]. The
thermal effects and thus the attainable energy densities
are much larger, about one order of magnitude, in the
case of chemisorption, e.g. hydration of metal salts,
than in the case of physisorption, e.g. water in silica.
The present paper concerns the structural, thermody-
namic and phase properties of the Na2S–H2O system
that is to be used in a waste heat driven solid–vapour
The present concern about themes like greenhouse
effect, pollution, renewable energy and durable
society has incited researchers to find better ways
of using the limited energy resources at our disposal.
This resulted in the development of fuel cells, solar
cells and heat pumps, to mention only a few. At the
same time energy efficiency has a huge potential for
reduction of energy consumption and CO2 reduction.
In the Netherlands alone about 100 PJ yrꢁ1 (1 PJ
ðpetajouleÞ ¼ 1015 J) worth of industrial waste heat
in the temperature range of 50–150 8C is actively
* Corresponding author. Tel.: þ31-224-564-790;
fax: þ31-224-5682-2615.
E-mail address: haije@ecn.nl (W.G. Haije).
0040-6031/02/$ – see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 0 - 6 0 3 1 ( 0 2 ) 0 0 1 5 8 - 2

4
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
absorption heat pump for cooling purposes in build-
ings and industrial processes. The heat pump opera-
tion is based on the following equilibrium reaction:
the experiments the starting material was dehydrated
until the desired water content was reached. For
dehydration a weighed amount of the starting material
was placed in a glass container connected to a cold
trap in liquid nitrogen and a vacuum pump. After
evacuation of the system, water vapour from the
starting material started to condense in the cold
trap. The glass container with the starting material
is heated to approximately 50 8C to accelerate the
dehydration process. The colour of the sample chan-
ged during the dehydration from white/colourless to
yellow when the amount of Na2Sꢀ5H2O in the sample
increased. The resulting water content of the samples
was calculated from the weight loss of the starting
material. All handling of sodium sulphide was con-
ducted as much as possible under protective nitrogen
atmosphere, to prevent any reaction with oxygen and
carbon dioxide in the air.
The composition of the samples used for the experi-
ments was determined by weighing the remaining
sodium sulphide hydrate after partial dehydration
and then calculating the water content, assuming
Na2Sꢀ9H2O to be the starting compound. Most sam-
ples were checked by X-ray powder diffraction for
their composition. The measured diffraction patterns
were critically compared to the known diffraction
patterns of Na2SꢀxH2O, where x ¼ 0, 2, 5 and 9
[9–13]. In addition the diffraction patterns of Na2SO3
and Na2S2 were sometimes present at very low inten-
sity ðI=I1 0 0 < 0:05Þ.
1
Na2S ꢀ 5H2O ðsÞ $ Na2S ꢀ H2O ðsÞ þ 4 12 H2O ðgÞ;
2
DrH ꢂ 300 kJ molꢁ1
(1)
Preliminary tests with lab and bench scale heat pump
units based on this solid–vapour pair yielded results
that could not entirely be understood on the basis of
the known literature on this system [6–9]. The most
important feature being the fact that the amount of
water after charging with a temperature difference
between condenser and accumulator of about 60 8C
was only about 66% of the expected value. When this
temperature difference was raised to about 70 8C the
rest of the water condensed. This suggested there
should be another phase with a composition close
to Na2Sꢀ2H2O present in the phase diagram, which
had already been proposed by Andersson et al. [8,9]
but was never proven as a fact. Therefore, experiments
were performed to prove the existence of this new
phase, to determine its p–T equilibrium line and to
re-establish data already published in older literature
[6–9] on the phase diagram and solid–vapour equili-
bria. Temperature dependent XRD was used to estab-
lish the number of crystallographic phases present
in the system. DTA/TG was used to determine the
amount of water released at each dehydration step
and vapour pressure measurements were performed in
order to establish the operational window for the heat
pump and to derive the reaction enthalpy and entropy
The samples used in the determination of the vapour
pressure of sodium sulphide hydrate were also che-
mically analysed by titration for the presence of
sulphide, sulphite and sulphate. All samples contained
sulphide as the main component, a very small amount
of sulphite and even less sulphate. The presence of
these substances is considered not to influence the
vapour pressure measurements, because mixing of the
sulphide and the sulphite or sulphate does not occur
[14].
1
for the Na2Sꢀ2H2O $ Na2Sꢀ H2O transition.
2
It should be borne in mind, as will become clear in
1
2
the following, that the hydration numbers 2 and
should be interpreted as ‘approximately 2’ and
1
‘approximately ’.
2
2. Materials and methods
2.1. Sample preparation
2.2. Thermal analysis
Na2SꢀxH2O (Merck, extra pure, about 35% Na2S)
was used as starting material for all of the experi-
ments. The water content corresponds to approxi-
mately sodium sulphide nanohydrate, Na2Sꢀ9H2O.
For the preparation of the samples that were used in
Thermal analyses on samples of sodium sulphide
hydrate were done to obtain information on the dehy-
dration behaviour of Na2SꢀxH2O. The goal was to
determine temperatures at which dehydration reac-
tions occur and to determine the heat of the reactions.

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
5
Table 1
Thermal properties of the standard materials used for calibration
Standard material
Mole weight (g molꢁ1
)
Ttransition (8C)
DHtransition (J molꢁ1
)
References
KNO3
101.11
118.96
299.5
128
230
299
166
ꢁ5106
ꢁ7062
ꢁ13772
ꢁ3923
[15,16]
[15,16]
[15,16]
[16]
Sn
KClO4
RbNO3 (Alfa, puratronic)
147.47
The measurements were done using a Netzsch STA
409 TG/DTA apparatus. The system was calibrated
prior to the measurement, using KNO3, RbNO3,
KClO4 and Sn, in order to determine the calorimetric
sensitivity. A temperature calibration is part of the
standard measurement procedure and was checked
before the measurements using the same materials
as for the calorimetric calibration (Table 1).
and the influence of slow mass transfer processes on
the measurements is minimised. Lower heating rates
resulted in poor determination of the accompanying
heat effects. These measurements were performed in
pans without a lid.
2.3. X-ray powder diffraction
The standard measurement conditions were as
given in Table 2. Sample pans were used either with
or without a lid depending on the required mass
transport of the water vapour out of the sample.
The calorimetric sensitivity was determined at
0:31 Æ 0:02 (Æ6%) mV s mJꢁ1. The influence of
changes in heating rate, sample weight, and helium
flow rate on the calorimetric sensitivity were negli-
gible.
For conducting controlled dehydration and rehydra-
tion reactions of sodium sulphide, the helium flow to
the TG/DTA system was saturated with water vapour
at 15 8C, resulting in a water vapour pressure of
17 mbar in helium. For these measurements the sam-
ple mass used was about 10 mg and apart from using
1.6 K minꢁ1 the heating rate was also lowered to
0.4 K minꢁ1. In this way the measurements are per-
formed under thermal equilibrium as much as possible
For the identification of the hydrates of sodium
sulphide X-ray powder diffraction was applied, using
a Guinier type camera. The samples were put in small
polyethylene bags to prevent reactions with air or
(de)hydration reactions during the measurement.
The measured X-ray diffraction patterns were com-
pared to the known patterns [10]1, [11–13] of the
sodium sulphide–water system.
For a crystallographic study of the dehydration
´
reaction of sodium sulphide hydrate a Guinier–Lenne
type camera was used with Cu Ka1, 2 radiation,
˚
l ¼ 1:5418 A. The sodium sulphide sample of about
50 mg was placed in the camera on a platinum grid in
an atmosphere of helium and water vapour. The
temperature of the sample or the water vapour pressure
was varied, by changing the temperature of the satura-
tor, in order to induce the desired dehydration or
rehydration reactions. The diffraction pattern was
recorded continuously while either changing the water
vapour pressure in the camera or the sample tempera-
ture, thus inducing phase changes in the sample.
Table 2
Standard measurement conditions for the analysis of the thermal
dehydration of Na2S–H2O samples
2.4. Phase diagram determination
Crucible material
Reference material
Average sample weight
Heating rate (high)
Heating rate (low)
Atmosphere
Al2O3
DTA measurements were done for the determina-
tion of the melting temperatures in the part of the
Na2S–H2O phase diagram relevant for practical heat
pump operation. Samples of about 100 mg of
Al2O3
5 mg
1.6 K minꢁ1
0.4 K minꢁ1
Helium, 99.999% at a flow
rate of 200 ml minꢁ1
1 [10] Tel.:þ610-325-9814; fax: þ610-325-9823, info@icdd.com

6
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Na2SꢀxH2O ð0 < x < 5Þ were placed in small evac-
uated and sealed glass capsules (height 10 mm, dia-
meter 8 mm) with a very thin polished bottom surface.
The samples were heated three times at 1 8C minꢁ1 to
130 8C and cooled with 1 8C minꢁ1 to 20 8C and kept
constant at that temperature for 2 h. An empty glass
capsule was used as reference.
Only after stabilisation of the temperature of the
sample a pressure reading was taken. Data were
acquired both on stepwise heating and on cooling.
3. Results and discussion
3.1. Thermal analysis: measurements
in a dry helium atmosphere
In addition visual melting point determinations
were done in a temperature controlled water bath.
Crystalline sodium sulphide hydrate was put in a
tubular glass capsule (height 70 mm, diameter
4 mm) that was evacuated and sealed. The glass
capsule was connected to a calibrated Pt100 tempera-
ture sensor and placed in the water bath that was
heated at a rate of 3 8C hꢁ1. The melting temperature
was determined visually at the point where crystallites
started to soften and melt.
On heating a sample of Na2Sꢀ9H2O in a flow of
helium the first dehydration step to Na2Sꢀ5H2O occurs
readily in all samples, provided the heating rate is low,
and there is no lid on the sample pan, so the mass
transfer of water vapour is not limited. Already at
room temperature dehydration to Na2Sꢀ5H2O occurs
just by passing a dry helium flow over the sample. At
high heating rates, or with a lid on the sample pan the
melting temperature of Na2Sꢀ9H2O (49 8C) is reached
before the first dehydration reaction has finished and a
melting peak of the remaining Na2Sꢀ9H2O was seen in
the DTA signal. In that case limited mass transfer of
water vapour prevented the dehydration reaction to
proceed fast enough. On further heating of the samples
again a distinction could be made between measure-
ments with water vapour transport limitation and
without this limitation. When water vapour transport
is limited the dehydration reaction of Na2Sꢀ5H2O to
lower water content starts at 60 8C and stops between
90 and 100 8C, when (partial) melting of the sample
occurs. The residual water content is then evaporated
at approximately 180 8C. In the case water vapour
transport is not limited the dehydration occurs under
almost equilibrium conditions, Na2Sꢀ5H2O then first
loses its water to Na2Sꢀ2H2O and then to Na2S. Further
heating to 300 8C does not give any further weight
loss, indicating that the final composition is comple-
tely dehydrated solid Na2S. The average enthalpy of
the complete dehydration of Na2Sꢀ5H2O to Na2S is
ð3:1 Æ 0:2Þ Â 102 kJ molꢁ1 of Na2S.
2.5. Vapour pressure measurements
The water vapour pressure of sodium sulphide
hydrate as a function of its temperature was deter-
mined using a calibrated MKS Baratron 690A dia-
phragm type pressure sensor connected to a glass
container (diameter 24 mm, height 50 mm) filled with
the salt hydrate. The container with the salt was placed
in a temperature controlled bath (Julabo, type HP 25,
Germany), and its temperature was determined using a
calibrated Pt100 temperature sensor connected to a
Keithley multimeter. Calibration of the temperature
sensor was performed using the same sample holder
filled with Al2O3 powder in order to simulate the
measurement conditions: estimated accuracy
Æ0.2 8C. The temperature sensor was positioned in
an insert in the centre of the sample container. The
pressure sensor was maintained at constant tempera-
ture of 45 8C to avoid temperature drift of the sensor
during measurement and to avoid condensation of
the water vapour in the sensor. The tubing of the
measurement set-up was heated to 60 8C to prevent
condensation of the water vapour. A high vacuum
pump was connected to the system in order to evacuate
the system prior to the measurement.
3.2. Thermal analysis: measurements
in a helium/water vapour atmosphere
The temperature of the sample was changed step-
wise approximately 10 8C after each measurement.
When the temperature of the bath reached it pre-set
temperature, it took approximately 15 min for the
sample temperature to stabilise at this temperature.
The results of the TG/DTA experiments carried out
in a flow of helium with 17 mbar water vapour pres-
sure are shown in Fig. 1, where the DTA curve and the
weight loss curve as a function of temperature for

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
7
Fig. 1. DTA and weight change curves for the dehydration reaction of Na2SꢀxH2O in an atmosphere of helium with 17 mbar water vapour.
Heating rate curve a: 1.6 K minꢁ1; curves b and c: 0.4 K minꢁ1
.
three dehydration measurements are plotted. The
measurements differ in heating rate and in their maxi-
mum temperature. It is found that at the highest
heating rate (curve a) the first dehydration reaction
Na2Sꢀ9H2O ! Na2Sꢀ5H2O is not completed at the
time the sample reaches the melting temperature,
49 8C, of the Na2Sꢀ9H2O phase. Partial melting of
the sample occurs, which gives a melting peak in DTA
curve a. The same is observed in the next dehydration
step from Na2Sꢀ5H2O ! Na2Sꢀ2H2O where a distinct
change in DTA curve a is observed at 83 8C, caused
by the melting of some of the Na2Sꢀ5H2O phase.
Despite the partial melting of the sample the dehy-
dration reaction still follows the Na2Sꢀ5H2O ! Na2Sꢀ
illustrates that a large temperature gradient is present
in the sample due to its poor heat conductivity. A lower
heating rate of 0.4 K minꢁ1 (curves b and c) does not
results in melting of the sample. In the weight change
curves b and c a small plateau is observed at 80 8C
corresponding with the Na2Sꢀ2H2O phase. In one
measurement the heating rate was set at 0 K minꢁ1
,
i.e. kept constant, at T ¼ 82 8C (curve c). At this point
the dehydration of the Na2Sꢀ2H2O phase started and
continued while the temperature was kept constant
1
2
until the Na2Sꢀ H2O composition was reached. There-
fore, DTA curve c in Fig. 1 does not show the last
dehydration peak.
Fig. 2 shows the TG curves recorded for the dehy-
dration–rehydration sequence in a flow of helium with
17 mbar water vapour pressure. The weight change as
1
2
2H2O ! Na2Sꢀ H2O sequence, thus a significant part
of the sample still is present as a solid phase. This

8
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Fig. 2. Rescaled TG curves for the dehydration–rehydration reactions of the Na2SꢀxH2O system in an atmosphere of helium wit 17 mbar water
vapour. Curve a, heating rate: 1.6 K minꢁ1, cooling rate: 1.6 K minꢁ1; curves b and c, heating and cooling rate: 0.4 K minꢁ1. Rehydration data
for curve b were not recorded. Arrows in the curves indicate the direction of change.
function of temperature is scaled to the hydrate num-
ber of the sample. The dehydration part is the same as
in Fig. 1. Starting compositions of the samples varied
due to differences in the equilibration time before the
start of a measurement. The measurements at the low
heating rate (curves b and c) show a strong overlap and
result in the formation of Na2Sꢀ2H2O with a water
content slightly higher than that formed with the high
heating rate (curve a). The actual water content of
Na2Sꢀ2H2O apparently depends on the conditions
under which it is formed. It is also obvious from this
figure that at high heating and cooling rates the limited
heat and mass transfer during the dehydration and
rehydration reactions give rise to increased hysteresis.
Also the (partial) melting of the sample leads to a more
compacted structure which can drastically slow down
water vapour uptake. The measurements in an atmo-
sphere of helium with water vapour confirmed the
stepwise dehydration from Na2Sꢀ9H2O to Na2Sꢀ5H2O
1
2
to Na2Sꢀ2H2O and then to Na2Sꢀ H2O. Contrary to the
experiments in dry helium, where at a low heating rate
complete dehydration to anhydrous Na2S occurs, the
presence of some water vapour in the surrounding
atmosphere leads to an incompletely dehydrated
sodium sulphide. In the rehydration reaction the for-
mation of Na2Sꢀ2H2O was not observed, probably due
to kinetic effects of the reactions.
The extrapolated onset temperatures, Te, of the
dehydration and rehydration reactions are given in
Table 3 together with the measured enthalpy of reac-
tion. The enthalpy of the dehydration reaction and the
rehydration according to reaction (1) was calculated to

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
9
Table 3
Extrapolated onset temperatures, Te, and enthalpies of reaction per mole of dry Na2S for dehydration and rehydration reactions at
pH O ¼ 17 mbar
2
Reaction
Te (8C)
DrH (kJ molꢁ1
)
Na2Sꢀ9H2O ðsÞ ! Na2Sꢀ5H2O ðsÞ þ 4H2O ðgÞ
Na2Sꢀ5H2O ðsÞ ! Na2Sꢀ2H2O ðsÞ þ 3H2O ðgÞ
32.5 Æ 1.5
72.0 Æ 1.0
82.0 Æ 0.5
71 Æ 1
215 Æ 20
176 Æ 12
94 Æ 8
1
2
Na2Sꢀ2H2O ðsÞ ! Na2Sꢀ H2O ðsÞ þ 1 12 H2O ðgÞ
1
Na2Sꢀ H2O ðsÞ þ 4 12 H2O ðgÞ ! Na2Sꢀ5H2O ðsÞ
ꢁ308 Æ 20
2
be ð2:9 Æ 0:2Þ Â 102 kJ molꢁ1 of Na2S. This supports
the earlier value of ð3:1 Æ 0:2Þ Â 102 kJ molꢁ1 of
Na2S determined for the complete dehydration in
the atmosphere of dry helium. The hysteresis of the
dehydration–rehydration reaction at a heating and
cooling rate of 0.4 K minꢁ1 is 11 8C, which is larger
than that found by Andersson et al. [8], who did the
measurements at a heating and cooling rate of
structure occurs as can be seen in the diffractogram in
Fig. 3. The diffraction pattern of Na2Sꢀ9H2O changes
into that of Na2Sꢀ5H2O.
In a second experiment, using the same sample, the
diffraction pattern was recorded while increasing the
sample temperature at a rate of 1 8C hꢁ1 from 55 to
75 8C. The water vapour pressure was kept constant at
9 mbar. The diffraction pattern of Na2Sꢀ5H2O, present
at the top of the diffractogram (see Fig. 4) disappears
gradually at a sample temperature around 63 8C. A
new pattern then arises that corresponds to the phase
Na2Sꢀ2H2O. On further heating of the sample some
small changes are seen in the diffraction pattern of
Na2Sꢀ2H2O. At a sample temperature of 67 8C the
diffraction pattern of Na2S appears while the diffrac-
tion pattern of Na2Sꢀ2H2O gradually fades. The dif-
fraction pattern of Na2Sꢀ2H2O corresponds well with
the reported pattern by Andersson and Azoulay [9].
There is clearly no pattern corresponding to an inter-
0.05 K minꢁ1
.
In the work of Andersson et al. [8,9] expressions for
the vapour pressure of Na2Sꢀ5H2O and Na2Sꢀ2H2O are
given from which a total heat of reaction is calculated
for reaction (1) of 285 kJ molꢁ1 Na2S, which corro-
borates the value determined in our work. The dehy-
dration and rehydration reactions of Na2Sꢀ5H2O are
described thoroughly in the same work by Andersson
et al. It was found that Na2Sꢀ2H2O exists in a tem-
perature range between 60 and 85 8C with a composi-
tion varying between Na2Sꢀ2.3H2O and Na2Sꢀ1.6H2O.
1
2
A
somewhat disordered crystal structure of
mediate Na2Sꢀ H2O phase.
Na2Sꢀ2H2O can lead to these variations in its water
content. This will be discussed in greater detail in
Section 3.3.
From the development of the diffraction pattern of
Na2SꢀxH2O as a function of the water content a
number of interesting observations can be made. All
transitions in this so-called crystolysis process [17],
show a coexistence region of the higher and lower
hydrate; the higher the hydration number the smaller
this region. There is no gradual position shift of the
diffraction lines that is indicative for the occurrence of
a solid solution. Several diffraction lines belonging to
one phase appear in the other phase with a different
intensity, being indicative for a topotactic transition
[17]. The line-width in X-ray diffraction is a measure
for the size of the diffracting grains: the smaller the
grains the broader the line. No line broadening is
observable with this set-up, suggesting no gradual
diminishing of the grain size of all grains, during
the transition from one hydration state to the other,
but a gradual reduction of the number of grains in a
3.3. X-ray powder diffraction
The X-ray powder diffraction technique was used
primarily for identification of the crystalline phases in
the samples during the subsequent decomposition
steps. The reference patterns for the known phases
were taken from [10–12]. For the study of the dehy-
dration reaction of sodium sulphide hydrate its dif-
fraction pattern was measured while decreasing the
water vapour pressure in the helium flow. The sample
was kept at 34 8C while the water vapour pressure was
slowly reduced from 23 to 9 mbar by slowly reducing
the saturator temperature from 20 to 5 8C with
1 8C hꢁ1. At 16 mbar a distinct change in the crystal

10
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Fig. 3. X-ray diffractogram of Na2SꢀxH2O at 34 8C as a function of water vapour pressure in the surrounding atmosphere. At the top of the
picture the diffraction pattern of Na2Sꢀ9H2O is present, the bottom part shows the diffraction pattern of Na2Sꢀ5H2O. From top to bottom the
water vapour pressure decreases from 23 to 9 mbar. Pt indicates the lines arising from the platinum grid used.
specific dehydration state. This is true for all transi-
tions in this system.
in which 3 out of 5 water molecules can be removed
from the crystal, followed by a shift in the crystal-
lographic position of some of the sodium atoms and
the remaining water molecules. This gives a new
structure, the Na2Sꢀ2H2O phase. Dehydration of this
compound to the anhydrous Na2S phase was also
assumed to occur in a topotactic process.
Andersson and Azoulay [9], and Mereiter et al. [13]
considered the structure of Na2Sꢀ5H2O to consist
of alternating layers of sodium, water and sulphur,
represented by ꢀ ꢀ ꢀ ꢁSꢁH2OꢁNaꢁH2OꢁSꢁ ꢀ ꢀ ꢀ. This
structure leads to a high mobility of 4 out of 5 water
molecules. The transformation of Na2Sꢀ5H2O to
Na2Sꢀ2H2O was assumed to be a topotactic process,
1
There is no Na2Sꢀ H2O phase. It is a more or less
2
‘stable’ mixture that only exists under mild conditions,

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
11
Fig. 4. X-ray diffractogram of Na2SꢀxH2O as a function of sample temperature. The water vapour pressure in the surrounding atmosphere was
9 mbar. At the top of the picture the diffraction pattern of Na2Sꢀ5H2O in present, the bottom part shows the diffraction pattern of anhydrous
Na2S. From top to bottom the sample temperature increases from 55 to 75 8C. Pt indicate the diffraction lines arising from the platinum grid
used.
presumably due to slow kinetics of the dihydrate
particles in the anhydrous matrix. Dehydration at
t > 350 8C under high vacuum yields pure Na2S that
does no longer react with water vapour [18].
3.4. Phase diagram
The results of the melting point determinations
using DTA on samples of sodium sulphide hydrate

12
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Table 4
Results of the DTA measurements and melting point determinations
Sample composition
Mole fraction Na2S
DTA transition temperatures (8C)
Tmelt (8C)
Na2Sꢀ3.24H2O
Na2Sꢀ3.0H2O
Na2Sꢀ1.8H2O
Na2Sꢀ0.93H2O
0.24
0.25
0.36
0.52
75 Æ 1
74 Æ 1
75 Æ 1
74 Æ 1
89 Æ 1
80 Æ 1
80 Æ 1
80 Æ 1
83.6 Æ 0.2
83.7 Æ 0.2
82.9 Æ 0.2
83.4 Æ 0.2
92 Æ 1
92 Æ 1
92 Æ 1
are given in Table 4. The transition temperatures
however, were only determined on heating the sample.
On cooling the sample, effects of supercooling pre-
vented a reproducible measurement. The transition
temperatures were determined using extrapolated
onset temperatures. Due to reactions of the molten
sample with the glass capsule, transition temperatures
were taken only at the first heating. In subsequent
heating runs the transition temperatures were shifted
and the enthalpy effect had decreased. Already before
the first measurement the sample was molten shortly
during sealing of the capsule. A small part of the
sample could have been contaminated prior to the
actual measurement, leading to additional heat effects
(pre-melting). To avoid effects of contamination of the
sample more accurate melting point determinations
were done using tubular glass capsules that were much
larger in order to prevent the melting of the sample on
sealing the capsule. The results of these measurements
are given in the last column of Table 4.
The results of the TG/DTA experiments (see Sec-
tions 3.1 and 3.2), the DTA measurements and the
melting point determinations are plotted together into
the phase diagram of the Na2S–H2O system
(Fig. 5).The first description of the Na2S–H2O system
was given by Kopylov [6] and recently extended with
the high temperature section by Kopylov and
Kaminski [7]. The results of our melting point mea-
surements correspond to the determinations of Kopy-
lov and Kaminski. The main difference as deduced
from our X-ray determinations is that the phase
1
2
Na2Sꢀ H2O reported by Kopylov and Kaminski [7]
Fig. 5. Phase diagram of the Na2S–H2O system (based on [6,7]) and measured transition temperatures: ~, represent results of the DTA
measurement; *, are the results of the melting point determinations; *, represent the transition temperatures measured by the TG/DTA
experiments presented in Section 3.1. Phase denominations: V, vapour; L, H2O based solution; C1, C2 and C3 are the crystalline phases
Na2Sꢀ9H2O, Na2Sꢀ5H2O, and Na2Sꢀ2H2O, respectively.

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
13
Table 5
Results of the vapour pressure measurements of water
T (8C)
pH O (102 Pa)
Difference (102 Pa)
Difference (%)
2
Experimental
Reference [19]
9.99
14.95
14.92
14.95
15.00
24.89
24.89
34.73
34.80
39.89
39.93
43.59
43.56
12.87
17.35
17.70
17.17
17.24
31.46
31.54
54.70
56.11
73.21
73.04
88.12
87.76
12.27
17.01
16.98
17.01
17.07
31.48
31.48
55.43
55.64
73.37
73.56
89.18
89.04
0.61
0.33
4.95
1.96
0.72
4.25
0.15
0.17
0.90
1.01
ꢁ0.02
0.07
ꢁ0.06
0.21
ꢁ0.73
0.47
ꢁ1.32
0.84
ꢁ0.15
ꢁ0.52
ꢁ1.07
ꢁ1.28
ꢁ0.21
ꢁ0.71
ꢁ1.19
ꢁ1.44
is now replaced by Na2Sꢀ2H2O. Furthermore it is seen
that the presence of the Na2Sꢀ2H2O phase does not lead
to significant changes in the melting temperature for
samples with compositions between Na2Sꢀ5H2O and
For the vapour pressure measurements on
Na2Sꢀ5H2O two different samples were used in order
to check the reproducibility of the measurement. The
results of both measurement are put together in
Table 7. The vapour pressure of Na2Sꢀ2H2O was also
determined on two samples in the same way as with
Na2Sꢀ5H2O. The results of these measurements are
compiled in Table 8.
In Fig. 6 the results of the measurements are dis-
played graphically together with the calculated func-
tions for the water vapour pressure of the three sodium
sulphide hydrates. This function is based on the Van’t
Hoff equation for the vapour pressure of a substance.
The limited accuracy of the measurement does not
allow for a more detailed description of the vapour
1
2
Na2Sꢀ H2O. At 83 8C melting of sodium sulphide
hydrate for the aforementioned compositions occurs.
This means that for a chemical heat pump, using Na2S–
H2O as working pair, regeneration temperatures cannot
exceed 83 8C without melting part of the salt hydrate.
The present study does not confirm the existence of
1
2
the compound Na2Sꢀ H2O: the TG/DTA measure-
ments yield varying compositions for the ‘‘dehy-
drated’’ sodium sulphide ranging from Na2Sꢀ0.9H2O
to Na2S, dependent on the reaction conditions.
Andersson and Azoulay [9] assumed that the residual
water content was caused by the stabilisation of a
small amount of the Na2Sꢀ2H2O phase in the dehy-
drated Na2S phase.
Table 6
Results of the vapour pressure measurements of Na2Sꢀ9H2O
T (8C)
pH O (102 Pa)
2
3.5. Vapour pressure measurements
4.99
5.06
2.19
2.28
The results of the vapour pressure measurements
are summarised in Table 5 through Table 8. First the
vapour pressure of pure water was measured to check
the performance of the experimental set-up. The result
of these measurements (Table 5) shows that vapour
pressures can be determined with an accuracy of
approximately 2%. The vapour pressure measure-
ments on a sample of Na2Sꢀ9H2O gave the results
as shown in Table 6.
14.92
14.97
16.99
24.92
24.92
34.83
34.83
43.61
43.61
5.10
5.15
6.11
11.11
11.04
22.91
22.77
41.41
41.39

14
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Table 7
Results of the vapour pressure measurements of Na2Sꢀ5H2O
Table 8
Results of the vapour pressure measurements of Na2Sꢀ2H2O
T (8C)
pH O (102 Pa)
T (8C)
pH O (102 Pa)
2
2
14.97
14.97
14.97
24.92
24.92
24.92
34.80
34.88
34.90
44.77
54.73
54.78
54.81
54.83
64.78
69.68
69.81
74.71
82.59
0.32
0.25
0.27
0.60
0.53
0.46
1.28
0.99
1.29
2.85
6.32
5.88
6.26
4.73
12.44
17.28
14.30
23.80
38.48
5.04
14.97
24.92
29.82
34.88
34.88
39.86
43.64
43.64
49.75
49.75
59.74
59.77
69.60
69.76
79.75
0.02
0.08
0.22
0.23
0.82
0.59
0.74
1.30
1.21
2.45
1.66
4.68
4.02
8.99
8.18
15.66
Fig. 6. Results of the vapour pressure measurements on water and sodium sulphide hydrates. (&) H2O; (^) Na2Sꢀ9H2O; (~) Na2Sꢀ5H2O;
(*) Na2Sꢀ2H2O. The solid lines represent the calculated Van’t Hoff equation for the vapour pressure of the indicated salt hydrates.

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
15
Table 9
Calculated enthalpy and entropy of the dehydration reactions of sodium sulphide hydrates
Reaction
DrH (kJ molꢁ1
)
DrH (kJ molꢁ1 H2O)
DrS (J molꢁ1 Kꢁ1
)
DrS (J molꢁ1 H2O Kꢁ1
)
Na2Sꢀ2H2O ðsÞ ! Na2S ðsÞ þ 2H2O ðgÞ
148 Æ 10
189 Æ 8
222 Æ 2
74 Æ 5
63 Æ 3
353 Æ 33
447 Æ 26
592 Æ 4
177 Æ 16
149 Æ 9
148 Æ 1
Na2Sꢀ5H2O ðsÞ ! Na2Sꢀ2H2O ðsÞ þ 3H2O ðgÞ
Na2Sꢀ9H2O ðsÞ ! Na2Sꢀ5H2O ðsÞ þ 4H2O ðgÞ
55.4 Æ 0.4
pressure, e.g. by means of the method described by
Oonk et al. [20] that takes into account the change in
heat capacity of the reactants and the products.
The results of the vapour pressure measurement are
used to calculate the change in enthalpy and entropy of
the dehydration reactions by means of the Van’t Hoff
equation, expressed per mole of water:
ect determinations of vapour pressure equilibria in
TG/DTA and XRD experiments. The latter are derived
from the transition temperatures observed while heat-
ing a Na2SꢀxH2O sample in a flow of helium with a
constant known water vapour pressure. A good agree-
ment is found between the different types of measure-
ments. The TG/DTA and XRD results suggest a
slightly lower vapour pressure for the respective
compounds, but the dynamic character of these mea-
suring methods can give rise to slightly higher transi-
tion temperatures.
The vapour pressure data of Andersson and Azoulay
[9] are also plotted in Fig. 7, and illustrate that their
TG/DTA results are in close agreement with our data.
The vapour pressure functions determined by Anders-
son, in the figure indicated by the dashed lines, show a
deviation of a few degrees Celsius from the functions
determined in this study. The reason in the case of
Na2Sꢀ2H2O is that Andersson took the average value
from dynamic measurements of dehydration and
rehydration. Our data are based purely on static equi-
librium measurements, as are the experimental points
from [9] indicated by the solid symbols. The equili-
brium data are to be preferred.
p
pꢃ
DrS DrH
ln
¼
ꢁ
R
R ꢀ T
with pꢃ ¼ 1 bar and R ¼ 8:3145 J molꢁ1 Kꢁ1
.
The calculated enthalpy and entropy of the reac-
tions are given in Table 9. The reported errors are
derived from the least squares fit of the measurements
to the Van’t Hoff equation and represent 95% con-
fidence limits.
The values of DrH and DrS increase with decreasing
water content as can be expected: the lower the
hydration number the stronger the water molecule is
bound. In Table 10 a comparison is made between the
derived enthalpies and entropies of reaction with the
literature values from [9]. A difference of less than 7%
is found for the enthalpy of the reactions but a much
larger difference is observed for the entropy of the
reactions. The values of 200 J molꢁ1 Kꢁ1 for the
reactions is given by Andersson and Azoulay [9] seem
to be too high considering that the standard entropy of
3.6. Calculation of the thermodynamic functions
of sodium sulphide hydrates
pure water vapour is 188 J molꢁ1 Kꢁ1
.
In Table 11 the calculated enthalpy of formation of
the three crystalline sodium sulphide hydrates are
In Fig. 7 the results of the direct vapour pressure
measurements are compared with the results of indir-
Table 10
Comparison between enthalpy and entropy of reaction in this study with literature values [9]
Reaction
DrH (kJ molꢁ1 H2O)
DrS (J molꢁ1 H2O Kꢁ1
)
This study
[9]
Difference (%)
This study
[9]
Difference (%)
Na2Sꢀ2H2O ðsÞ ! Na2S ðsÞ þ 2H2O ðgÞ
ꢁ73.9
ꢁ62.9
ꢁ69.0
ꢁ60.6
6.6
3.6
177
149
220
199
24
34
Na2Sꢀ5H2O ðsÞ ! Na2Sꢀ2H2O ðsÞ þ 3H2O ðgÞ

16
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Fig. 7. A comparison between the calculated vapour pressure functions and the results of TG/DTA measurements, XRD measurements, and
data from [9]. Solid lines represent the calculated vapour pressure functions of the indicated sodium sulphide hydrates. (*) TG/DTA; (~)
XRD; solid symbols are TG/DTA results of [9] and the dashed lines represent the vapour pressure functions from [9].
compared to literature values. The enthalpy of forma-
tion of Na2Sꢀ2H2O is calculated using the enthalpy of
formation of the reaction products, Na2S and H2O and
the measured enthalpy of the reaction
The calculated and earlier reported literature values
agree very well.
The enthalpy and entropy of the equilibrium reactions
given in Table 9 are also used to calculate the thermo-
dynamic functions of the sodium sulphide hydrates. The
well-known thermodynamic functions of H2O (g) [22]
and Na2S [23] are also used for this calculation.
The enthalpy of Na2Sꢀ2H2O at each temperature
is calculated from the enthalpies of the reaction
DfHꢃ ðNa2S ꢀ 2H2O ðsÞÞ
¼ 2DfHꢃ ðH2O ðgÞÞ þ DfHꢃ ðNa2S ðsÞÞ þ DrH
The enthalpies of formation of the other sodium
sulphide hydrates are calculated in a similar way.
Table 11
Calculated enthalpies of formation of the sodium sulphide hydrates and a comparison with literature values
Compound
DfH8 (298.15)
DfH8 (298.15)
Difference
kJ molꢁ1
(kJ molꢁ1), this study
(kJ molꢁ1), [21]
%
Na2Sꢀ2H2O
Na2Sꢀ5H2O
Na2Sꢀ9H2O
ꢁ998
ꢁ1912
ꢁ3101
–
–
ꢁ25
ꢁ27
–
ꢁ1886.6
ꢁ3074.0
1.3
0.9

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
17
Table 12
Heat capacity, Cp, entropy, S8, and Gibbs energy of formation, DfG8, of the sodium sulphide hydrates
Compound
Cp ¼ A þ B ꢀ T
A (J molꢁ1 Kꢁ1
S8 (298.15)
DfG8(298.15)
(J molꢁ1 Kꢁ1
)
(kJ molꢁ1
)
)
B (J molꢁ1 Kꢁ1
)
Na2Sꢀ2H2O
Na2Sꢀ5H2O
Na2Sꢀ9H2O
142.4
238.3
370.7
2.562 Â 10ꢁ2
4.197 Â 10ꢁ2
4.875 Â 10ꢁ2
120.3
239.4
400.5
ꢁ855
ꢁ1596
ꢁ2555
products, Na2S and H2O and the measured enthalpy of
the reaction. For the entropy of Na2Sꢀ2H2O the same
procedure is applied, using the entropies of the reac-
tion products and the measured entropy of the reac-
tion. The heat capacity, Cp, of the sodium sulphide
hydrates is calculated from the sum of the heat capa-
cities of water vapour and Na2S and fitted with the
relation
where A and B are constants and T the temperature in
Kelvin. In Table 12 the values for A and B are given
together with the standard entropy and the Gibbs
energy of formation.
The Gibbs energy of the sodium sulphide hydrates
is simply calculated from the enthalpy and the entropy.
The thermodynamic functions of Na2Sꢀ2H2O and
similarly derived for Na2Sꢀ5H2O and Na2Sꢀ9H2O,
are given in Tables 13–15. The equilibrium reactions
given in Table 9 are used for the calculation.
Cp ¼ A þ B ꢀ T
Table 13
Thermodynamic functions of Na2Sꢀ2H2O
Temperature T /K
Heat capacity
Cp, m/J molꢁ1 Kꢁ1
Entropy
Sꢃm/J molꢁ1 Kꢁ1
Enthalpy
Hmꢃ /kJ molꢁ1
Gibbs enerꢁg1y
Gꢃm/kJ mol
273.15
278.15
283.15
288.15
293.15
298.15
303.15
308.15
313.15
318.15
323.15
328.15
333.15
338.15
343.15
348.15
353.15
358.15
363.15
368.15
373.15
149.4a
149.5
149.6
149.7
149.9
150.0
150.1
150.3
150.4
150.5
150.6
150.8
150.9
151.0
151.2
151.3
151.4
151.5
151.7
151.8
151.9
107.2
109.9
112.6
115.2
117.8
120.3
122.8
125.3
127.7
130.1
132.4
134.7
137.0
139.2
141.5
143.7
145.8
147.9
150.0
152.1
154.2
ꢁ1001.2
ꢁ1000.5
ꢁ999.7
ꢁ999.0
ꢁ998.2
ꢁ997.5
ꢁ996.7
ꢁ996.0
ꢁ995.2
ꢁ994.5
ꢁ993.7
ꢁ993.0
ꢁ992.2
ꢁ991.5
ꢁ990.7
ꢁ990.0
ꢁ989.2
ꢁ988.4
ꢁ987.7
ꢁ986.9
ꢁ986.2
ꢁ1031
ꢁ1031
ꢁ1032
ꢁ1032
ꢁ1033
ꢁ1033
ꢁ1034
ꢁ1035
ꢁ1035
ꢁ1036
ꢁ1037
ꢁ1037
ꢁ1038
ꢁ1039
ꢁ1039
ꢁ1040
ꢁ1041
ꢁ1041
ꢁ1042
ꢁ1043
ꢁ1044
RT
CP;m
T
Nomenclature in accordance with SGTE (Scientific Group Thermodata Europe): Sꢃm ¼ D0T Smꢃ
¼
dT; Hmꢃ ¼ DfHmꢃ ð298:15 KÞ þ
0
RT
298:15CP;mdT; Gꢃm ¼ Hmꢃ ꢁ TSꢃm
a Figures in italic are extrapolated.

18
R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
Table 14
Thermodynamic functions of Na2Sꢀ5H2O
Temperature T /K
Heat capacity
Cp, m/J molꢁ1 Kꢁ1
Entropy
Sꢃm/J molꢁ1 Kꢁ1
Enthalpy
Hmꢃ /kJ molꢁ1
Gibbs enerꢁgy1
Gꢃm/kJ mol
273.15
278.15
283.15
288.15
293.15
298.15
303.15
308.15
313.15
318.15
323.15
328.15
333.15
338.15
343.15
348.15
353.15
358.15
363.15
368.15
373.15
249.7a
250.0
250.2
250.4
250.6
250.8
251.0
251.2
251.4
251.6
251.8
252.1
252.3
252.5
252.7
252.9
253.1
253.3
253.5
253.7
253.9
217.4
222.0
226.4
230.8
235.1
239.4
243.6
247.7
251.7
255.7
259.6
263.5
267.3
271.0
274.8
278.4
282.0
285.6
289.1
292.6
296.0
ꢁ1918
ꢁ1917
ꢁ1915
ꢁ1914
ꢁ1913
ꢁ1912
ꢁ1910
ꢁ1909
ꢁ1908
ꢁ1907
ꢁ1905
ꢁ1904
ꢁ1903
ꢁ1902
ꢁ1900
ꢁ1899
ꢁ1898
ꢁ1897
ꢁ1895
ꢁ1894
ꢁ1893
ꢁ1977
ꢁ1978
ꢁ1980
ꢁ1981
ꢁ1982
ꢁ1983
ꢁ1984
ꢁ1985
ꢁ1987
ꢁ1988
ꢁ1989
ꢁ1991
ꢁ1992
ꢁ1993
ꢁ1995
ꢁ1996
ꢁ1997
ꢁ1999
ꢁ2000
ꢁ2002
ꢁ2003
R
0
T CP;m
T
Nomenclature in accordance with SGTE (Scientific Group Thermodata Europe): Sꢃm ¼ D0T Smꢃ
¼
dT; Hmꢃ ¼ DfHmꢃ ð298:15 KÞ þ
R T
298:15CP;mdT; Gꢃm ¼ Hmꢃ ꢁ TSꢃm
a Figures in italic are extrapolated.
Table 15
Thermodynamic functions of Na2Sꢀ9H2O
Temperature T /K
Heat capacity
Cp, m/J molꢁ1 Kꢁ1
Entropy
Sꢃm/J molꢁ1 Kꢁ1
Enthalpy
Hmꢃ /kJ molꢁ1
Gibbs enerꢁgy1
Gꢃm/kJ mol
273.15
278.15
283.15
288.15
293.15
298.15
303.15
308.15
313.15
318.15
323.15
384.0a
384.3
384.5
384.8
385.0
385.2
385.5
385.7
386.0
386.2
386.5
366.8
373.8
380.6
387.4
394.0
400.5
406.9
413.2
419.4
425.5
431.6
ꢁ3110
ꢁ3108
ꢁ3106
ꢁ3104
ꢁ3103
ꢁ3101
ꢁ3099
ꢁ3097
ꢁ3095
ꢁ3093
ꢁ3091
ꢁ3210
ꢁ3212
ꢁ3214
ꢁ3216
ꢁ3218
ꢁ3220
ꢁ3222
ꢁ3224
ꢁ3226
ꢁ3228
ꢁ3230
R
0
T CP;m
T
Nomenclature in accordance with SGTE (Scientific Group Thermodata Europe): Sꢃm ¼ D0T Smꢃ
¼
dT; Hmꢃ ¼ DfHmꢃ ð298:15 KÞ þ
R T
298:15CP;mdT; Gꢃm ¼ Hmꢃ ꢁ TSꢃm
a Figures in italic are extrapolated.
4. Conclusion
confirmed by TG/DTA and XRD techniques. In addi-
tion the existence of a new phase of composition
Na2Sꢀ2H2O has been proven as a fact. Furthermore
In the Na2S–H2O system the existence of the com-
pounds Na2Sꢀ9H2O, Na2Sꢀ5H2O and Na2S has been
1
2
the compound with composition Na2Sꢀ H2O is not a

R. de Boer et al. / Thermochimica Acta 395 (2003) 3–19
19
genuine phase but is a 1:3 mixture of Na2S and
Na2Sꢀ2H2O (Fig. 5). The kinetics of the reaction of
Na2Sꢀ2H2O to Na2S in small domains of the dihydrate
is so slow that under mild conditions this composition
appears to be stable. The reaction enthalpy and
entropy of the pentahydrate to the dihydrate as well
as of the dihydrate to dry sodium sulphide transitions
have been established. The water vapour pressure as
a function of temperature of all three hydrates has
been determined and based on these measurements a
consistent set of thermodynamic functions for the
sodium sulphide hydrates is calculated. The melting
temperature of Na2Sꢀ9H2O is 49 8C and mixtures of
Na2Sꢀ5H2O with Na2Sꢀ2H2O and of Na2Sꢀ2H2O with
Na2S melt at 83 8C.
Environment (NOVEM) are gratefully acknowledged
for financial support.
References
[1] J.A. Carp, P.W. Bach, Energy Research Centre of the
Netherlands, P.O. Box 1, 1755 ZG Petten, The Netherlands,
personal communication on European Waste Heat Potential,
(2000).
[2] C. Schweigler, S. Summerer, H.-M. Hellmann, F. Ziegler
(Eds.), Proceedings of the International Sorption Heat Pump
Conference, Munich, Germany, March 24–26, 1999.
[3] G. Restuccia, J.A. Arnoud, R. Quagliata, Int. J. Energy Res.
12 (1) (1988) 101.
[4] V. Goetz, A. Marty, Calorimetrie et Analyse Thermique 23
(1992) 431.
[5] Z. Tamainot-Telto, R.E. Critoph, Appl. Therm. Eng. 21
(2001) 37.
The results of the experiments are in good agreement
with earlier work on the crystallographic, physical and
thermochemical properties of the Na2S–H2O system
[6–13]. As to the exact composition of the ‘dihydrate’ it
is recommended to perform neutron or X-ray diffrac-
tion experiments to elucidate the crystal structure.
For utilisation of the Na2S–H2O system as working
pair in a solid-sorption type chemical heat pump with
integrated heat or cold storage function, the operating
window has been established. The melting tempera-
ture of the salt hydrate in the composition range
[6] N.I. Kopylov, Russ. J. Inorg. Chem. 13 (2) (1968) 276.
[7] N.I. Kopylov, Yu.V. Kaminski, Russ. J. Inorg. Chem. 44 (2)
(1999) 261.
[8] J.Y. Andersson, J. de Pablo, M. Azoulay, Thermochim. Acta
91 (1985) 223.
[9] J.Y. Andersson, M. Azoulay, J. Chem. Soc. Dalton Trans. 3
(1986) 469.
[10] JCPDS-ICDD Powder Diffraction File, International Centre
for Diffraction Data, 12 Campus Boulevard, Newtown
Square, PA 19073-3273 USA, 1997.
[11] D. Bedlivy, A. Preisinger, Zeitschrift fu¨r Kristallografie 121
(1965) 114.
1
2
between Na2Sꢀ5H2O and Na2Sꢀ H2O is 83 8C. The
[12] D. Bedlivy, A. Preisinger, Zeitschrift fu¨r Kristallografie 121
(1965) 135.
heat storage capacity based on the dehydration reac-
tion of the salt is
[13] K. Mereiter, A. Preisinger, A. Zellner, W. Mikenda, H. Steidl,
J. Chem. Soc. Dalton Trans. 7 (1984) 1275.
[14] S. Andersson, Chem. Scripta 20 (1982) 164.
[15] Reference materials for Thermal analysis GM 758, Certified
by the International Confederation for Thermal Analysis,
distributed by the US National Bureau of Standards (today
NIST, USA), 1971.
1
Na2S ꢀ 5H2O ðsÞ ! Na2S ꢀ H2O ðsÞ þ 4 12 H2O ðgÞ;
2
DrH ð298:15 K; 1 barÞ ¼ 300 Æ 15 kJ molꢁ1
(2)
This equals 3.84 MJ kgꢁ1 Na2S or approximately
1 kWh kgꢁ1 Na2S. One mole of Na2S can absorb
4 12 moles of water, which is equal to the absorption
of 1 kg of water in 1 kg of dry salt. The theoretical cold
storage capacity of this working pair, based on the heat
of evaporation of water is 2.54 MJ kgꢁ1 Na2S.
[16] R. Riesen, G. Wiedmann, Thermoanalyse, Dr Alfred Hu¨thing
Verlag GmbH, Heidelberg, 1984.
[17] A.K. Galwey, Thermochim. Acta 355 (2000) 181.
[18] W.G. Haije, R. de Boer, Unpublished results of failed
chemisorption experiments on dry Na2S.
[19] D.R. Lide (Ed.), Handbook of Chemistry and Physics, 78th
Edition, CRC Press, New York, 1997.
[20] H.A.J. Oonk, P.R. van der Linde, J. Huinink, J.G. Blok, J.
Chem. Thermodyn. 30 (1998) 897.
Acknowledgements
[21] D.D. Wagman, W.H. Evans et al., J. Phys. Chem. Ref. Data
11 (1982) 306.
Mr. R.G. Nyqvist (Thermal Analysis) and Mr. P. van
Vlaanderen (X-ray diffraction) are acknowledged for
painstakingly performing these measurements. SWEAT
BV and the Netherlands Agency for Energy and the
[22] J.D. Cox, D.D. Wagman, V.A. Medvedev, CODATA Key
Values for Thermodynamics, Hemisphere, New York, 1989.
[23] M.W. Chase Jr., C.A. Davies et al., J. Phys. Chem. Ref. Data
14 (1985) 1595.