INTERNATIONAL JOURNAL OF ENERGY RESEARCH
Int. J. Energy Res. (2012)
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/er.2913
Use of adsorbents for thermal energy storage of solar or
excess heat: improvement of energy density
*,
Dan Dicaire and F. Handan Tezel
Department of Chemical and Biological Engineering, University of Ottawa, Ottawa, Ontario, Canada
SUMMARY
The current paper describes the design of a prototype system to explore the feasibility of the adsorption thermal energy storage.
Water was chosen as the adsorbate, and three different adsorbents were tested. Zeolite 13X, NaLSX zeolite, and an activated
alumina (AA)/zeolite 13X composite adsorbent were used as adsorbents. •xperiments were performed at varying flow
rates and different relative humidities to determine the optimal operating conditions for the system. The regeneration of the
adsorbents also was explored by performing repeated runs on the same adsorbent sample. The results indicate that complete
regeneration was achieved. A maximum energy density of 160 kWh/m3 has been achieved with the AA/13X adsorbent, and
this adsorbent was chosen for further studies. After this adsorbent screening, the system was modified to improve the data
recording and system performance. Tests were performed on AA/13X, and a maximum energy density of 200 kWh/m3 was
achieved, which was much higher than the maximum energy density reported in the literature for adsorption thermal energy
storage systems (165 kWh/m3). Copyright © 2012 John Wiley & Sons, Ltd.
KEY WORDS
Thermal Energy Storage; Adsorption Heat Storage; Zeolite 13X; Water Adsorption; NaLSX Zeolite; Activated Alumina/13X zeolite
hybrid adsorbent
Correspondence
*F. Handan Tezel, Department of Chemical and Biological Engineering, University of Ottawa, 161 Louis Pasteur, Ottawa, Ontario K1N
6N5, Canada.

E-mail: [e-mail]
Received 28 March 2010; Revised 31 January 2012; Accepted 7 February 2012
1. lNTRODUCTlON AND In sensible heat storage systems, a material, like rock or
LlTERATURE SEARCH water, is heated inside an insulated container that slows
thermal leaking. The performance of this type of storage is
Conservation and sustainability are integral parts of our measured by the temperature difference between the material
society. It drives us to explore new sources of energy and and the ambient temperature. •xamples of this type of
find value in what used to be considered waste. As resources storage systems are as follows: Aquifer Thermal •nergy
become depleted and prices for standard commodities, like Storage [4], Borehole Thermal •nergy Storage [5], or water
oil, keep rising, clean sustainable technologies are getting tank storage [6]. This method of storing energy is extremely
more attention and are economically feasible. The  Clean cheap and is currently the dominant form of thermal energy
Tech movement has arrived. storage. However, sensible heat storage systems have low
Thermal energy storage is one of these resulting technol- energy densities, in the range of 40 60 kWh/m3 [7], which
ogies. In Canada, energy is a basic requirement of all requires large volumes to store sufficient energy for heating
industrial, commercial, and residential operations. Instead applications. The thermal energy is also sensibly stored and
of obtaining thermal energy from conventional sources, like is constantly diffusing to the environment. As a result, large
coal or oil, it can prove more profitable to collect it from amounts of insulation are required to slow the heat loss to
unconventional sources like solar heat, when it is abundant the surroundings. Most systems keep the energy for less
and store it until it is required. There is a great demand for than a week, and even the state-of-the-art systems do not
utilization of low-mid grade waste heat and thermal energy last more than a few months, even with volumes above
storage system to provide space heating. 30 000 m3 [8]. These sensible heat storage systems have
There are three groups of thermal storage systems: become widespread as short-term thermal energy storage
sensible heat storage, latent heat storage and thermochemical with small volumes. The systems can provide a limited
heat storage [1 3]. amount of energy depending on the climate, which means
Copyright © 2012 John Wiley & Sons, Ltd.
D. Dicaire and F. H. Tezel Improvement of energy density for adsorption thermal energy storage
that auxiliary heating systems are usually required, espe- terms of technology. There are some solutions that can cater
cially in northern regions. to niche markets, but widespread thermal applications, like
Latent heat storage systems rely on the energy released or residential heating applications, are out of range for currently
absorbed during the phase change of a material to store available technology. The biggest hurdle is the lack of
energy, which is why they are typically referred to as phase permanent long-term thermal energy storage with charging
change materials. These materials are either free flowing or temperatures below 250 C. The second hurdle is the need
encapsulated for easy handling and placed in large heat for more compact thermal energy storage systems that will
transfer containers, or they are infused into building not take up large volumes and could be retrofitted into
materials like dry wall. These materials are mainly designed existing homes. The commercialization target set by the
to maintain a constant temperature around the fusion international thermal energy storage community is a thermal
temperature and are subject to continuous heat loss to the storage system that has 8 to 10 times the energy density of
environment. Therefore, these systems also require a great water, around 480 kWh/m3.
deal of insulation and cannot be used for long-term thermal •nergy storage through adsorption is a physical process
energy storage. Typical phase change materials like paraffin and is a possible solution as the heat quality does not degrade
waxes have energy densities around 55 kWh/m3 [9] and are with time and stays locked away until the two interacting
very good for cooling applications. Some phase change components are brought together. Adsorption is an
materials, like molten salts, have been developed for high exothermic process, which releases thermal energy as the
temperature (300 C and up) applications and can have adsorbent adsorbs the gas it is being exposed to, into its
energy densities as high as 300 kWh/m3 [10]. The latter crystalline structure [7]. The reverse of adsorption is a
materials are best suited for steam production and would desorption process, which is endothermic. After an adsorbent
not be feasible in residential settings. Several prototypes of bed is saturated with gas, it needs to be regenerated to be
latent heat storage systems are being developed, and the used again as an energy source. Using a heat source (like
paraffin waxes are starting to be commercialized, but these solar radiation), the necessary energy for the endothermic
systems are not widespread. desorption process is provided, which releases the gas from
Thermo-chemical heat storage utilizes reversible exo- the adsorbent and restores the adsorbent to its original state,
thermic/endothermic reactions or processes to store heat. ready to dispense more energy. Both of these processes are
•xcess heat is used to perform the endothermic reaction, explained in Figure 1. The temperature and amount of
which usually separates a product into reactants. Once thermal energy used in the regeneration determines the
the reactants are separated, the energy is stored as chemical amount of gas that can be released. Although, typically,
potential. When the energy needs to be released, the reac- temperatures higher than 200 C are necessary to desorb all
tants are brought together for the exothermic synthesis the gas and attain maximum available energy [12], it is
reaction, and as the product is made, the energy is released. sometimes more practical to operate the system between
Because these systems store energy as chemical potential, the 110 and 150 C, sacrificing some potential adsorption
they do not require insulation, and the stored energy does capacity for practicability.
not degrade with time. An example of this kind of thermal The heat released from adsorption is not constant and
energy storage is NH3 dissociation into H2 and N2 being varies as a function of the amount of gas already adsorbed
developed by the Australian National University [11]. into the system as illustrated by Gopal et al. [12]. The
However, most reactions require very high temperatures amount of gas, which can be adsorbed, and therefore the
(400 C and up) for both endothermic and exothermic reac- amount of energy that can be released, also is a function of
tions, which makes them well suited for steam production the gas partial pressure [7] because it acts as the driving force
but not for typical heating applications [11]. Only a few for adsorption. It is good to have a high heat of adsorption as
experimental systems exist worldwide, and the technology more energy is available to the system, but it should be
is still being developed. noted that a high heat of adsorption requires a high drying
As can be seen from the previous paragraphs, although temperature for regeneration and therefore can be a
thermal energy storage is an old concept, it is still young in disadvantage [13]. Many factors must be considered to
Figure 1. Description of adsorption and desorption.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
Improvement of energy density for adsorption thermal energy storage D. Dicaire and F. H. Tezel
obtain the maximum performance from an adsorbent thermal leased, and the consumer would only be paying for the
energy storage system. Because the energy stored suffers no energy used.
degradation with time, they are ideal candidates for thermal Working adsorbent systems have been reported in the
energy storage. A basic design, which outlines a possible literature. Gantenbein et al. [13] produced a working zeo-
residential thermal energy storage system, can be seen in lite system, which had an energy density of 106 kWh/m3.
Figure 2. Both desorption and adsorption cycles are shown The Institute for Sustainable Technologies in Austria
in this figure. investigated thermo-chemical energy storage using 200 kg
When humid air is passed through the adsorbent bed of silica gel as the adsorbent [15]. Hauer [16] reported a
(shown as  Water Vapour In in Figure 2), water vapour gets successful full scale 7000-kg zeolite 13X storage system,
adsorbed by the adsorbent, which releases heat (because which heated a school and was charged by district heating
adsorption is an exothermic process). This increases the over night to offset the peak energy demands and performed
temperature of the air leaving, going into the house and used at 124 kWh/m3 of energy density. Jaenchen et al. [17]
as space heating (shown as  •nergy Out in Figure 2). studied a zeolite system which could measure energy densities
Because the adsorbent bed will get saturated eventually, it of 160 kWh/m3. Dawoud et al. [18] reported their working
will need to be regenerated. This regeneration will require zeolite 13X system performing at 165 kWh/m3. •vennatural
heat, which will be provided by the solar panel (shown zeolite, which has a lack in performance when compared
as  •nergy In in Figure 2). This heat causes the water with its synthetic counter parts, has been identified as a
vapour to be desorbed from the adsorbent bed (shown as potential candidate for use in an adsorption thermal energy
 Water Vapour Out in Figure 2), preparing it to be ready storage system [19].
to adsorb more. In this study, an adsorbent thermal energy storage system
The ultimate goal of this technology would be to store has been designed and built. The prototype presented here is
heat from solar panels or any other source of heat (like waste the first step in the design process and is a useful tool for
heat from aluminum plants or nuclear reactors) when it is in determining the maximum performance of an adsorbent
excess and release it later when it is necessary. Adsorbent system and for screening different types of adsorbents for
beds are ideally suited for these applications because the thermal energy storage applications.
physical storage of the energy does not degrade with time.
This means that with an on-site system, excess energy stored
during the summer can be used during the winter. •nergy 2. MATERlALS AND METHODS
stored during the day can be used at night or that left over
heat from a process can be recycled to be used at a later time. A literature search provided ample suggestions for
Adsorbent bed systems also could be mobile, allowing adsorbates (NH3, methane, water, ethane and CO2) for a
distribution of the energy away from its generation similar variety of different adsorbents (silica gel, ZSM-5, activated
to gas distribution through propane tanks. Dedicated solar carbon, Mordenite and Na-A) [20 22].
farms or aluminum refineries could act as regeneration Suitable adsorbent and adsorbate pairs were identified
stations. The excess thermal energy would be stored and then based on cost, availability and non-toxicity. Water was
be distributed in mobile adsorbent beds to surrounding identified as the most feasible adsorbate and could be
commercial or residential sectors, which may not have the coupled with a number of adsorbents. Zeolite 13X, Na
physical or financial resources to install solar systems [14]. LSX and a hybrid of activated alumina and zeolite 13X
Spent adsorbent beds would be picked up, and charged (AA/13X) were retained. Details of these adsorbents are
beds would be dropped off; the equipment would be given in Table I.
Figure 2. Basic thermal energy storage system.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
D. Dicaire and F. H. Tezel Improvement of energy density for adsorption thermal energy storage
TabIe I. Details of the adsorbents used in this study.
Adsorbent Company Chemical name Mesh
Zeolite 13X Ceca, France Ceca G5CO2 8 12
NaLSX Air Products, Allentown PA USA UOP APG III 8 12
AA/13X Axens (Formerly RioTinto Alcan), Brockville, ON Canada ACTIGUARD 650PCAP 8 14
The primary goal of these experiments was to validate 8 hours, the column was sealed by closing the inlet and
adsorbent beds as thermal energy storage systems by outlet valves, and the column was allowed to cool to room
measuring how much energy could be released from such temperature.
a system. More specific tests also were performed to The experimental phase had two parts. The first was to set
understand how the operating conditions, mainly flow rate the relative humidity of the system by using the by-pass
and relative humidity, affect the release of energy from the line that avoids the column and goes to the hygrometer.
bed. Finally, the system was used to screen different types During this time, the inlet relative humidity was adjusted
of adsorbents and test the regeneration of the adsorbents to the desired value (between 30% and 100%) using the
by repeating the experiments with the same sample of hygrometer measurements at the end of the system as well
adsorbent. Figure 3 shows the schematic diagram of the as controlling the flows through the wet and dry rotameters.
prototype thermal energy storage system that was built. The wet rotameter produces air with 100% RH, whereas the
The column is composed of 10 cm of ½ inch stainless dry air produces air with 0% RH. By combining different
steel tube, with a volume of 9.04 ml and weight of 100 g. It flows from both, the desired relative humidity can be
was loaded with adsorbent by blocking one end with quartz reached. The total flow rate was measured with a precision
wool, loading the adsorbent pellets and shaking the column flow meter. After the relative humidity was set, the data
to ensure that the packing was tight and uniform. The acquisition system was started, the inlet and outlet valves
other end was then blocked with quartz wool as well to to the column were opened and the humid air was passed
prevent the adsorbent particles to leave the column during through the column. The thermocouples monitored the
experiments. All the runs presented in this study were done temperature of the air entering and leaving the column, the
with fresh adsorbent (adsorbent, which had not been through temperature of the outside of the column and the temperature
an adsorption cycle yet), except those which were tested for of the stream when it reached the hygrometer. The RH at the
the regeneration of the adsorbents. The column is insulated exit of the column was measured with the hygrometer. The
in a cylinder of fibreglass insulation 1 think and has two column was considered to have reached its saturation when
valves at either ends to seal it from the rest of the system, the outlet temperature of the column had returned to that of
which is connected with 1/4 Stainless Steel tubing. The the inlet and the outlet relative humidity was equal to that
heater is a 200-W electric heater insulated and attached to a of the inlet. At this point, the system was shut down.
voltage controller. The air comes from a compressed air When the column is being regenerated, dry air is fed into
supply and goes through the wet and dry rotameters. If the the system, and the heater is activated. The heater heats the
 dry rotameter is used, the air is fed directly into the system; air to the desired temperature (between 130 C and
if the  wet rotameter is used, the air is fed through a bubbler 250 C), which is measured by a thermocouple at the exit
where it is saturated with water vapour and then fed into the of the heater. The hot air is fed into the column, where it
system. regenerates the adsorbent, releasing the water and drying
Once the column was packed, it was installed into the zeolite. The now warm wet air exits the column and
the system and regenerated by blowing hot dry air at passes through the hygrometer where its relative humidity
250 C through it for 8 hours to remove any impurities is measured to monitor the water leaving the column.
it may have accumulated during storage or loading. After The end of the regeneration cycle was marked when no
Figure 3. Schematic diagram of the experimental setup.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
Improvement of energy density for adsorption thermal energy storage D. Dicaire and F. H. Tezel
moisture was detected at the outlet of the column. This 2.2. Analysis
indicated that the adsorbent bed was ready to adsorb
The experimental system used in this study was designed to
moisture again.
determine the potential of different adsorbents to be used for
To start the adsorption cycle (to start releasing the stored
energy storage applications. With its strategically placed
energy), humidified air (with the use of the bubbler) is fed
thermocouple, it can measure heat released and, therefore,
into the column at room temperature (the heater is kept
the energy stored by any adsorbent and therefore determine
off). The water vapour is adsorbed as the humid air
its potential as an energy source. The hygrometer can be used
passes through the adsorbent bed. Because adsorption is an
to determine how much water is entering and leaving the
exothermic process, water adsorption releases heat, and
system, which indicates how much water the system has
the energy released heats up the air leaving the column.
adsorbed. The system can accommodate a wide range of
The temperature of this warm air is measured by a thermo-
different flow rates and relative humidity and can be used
couple as it leaves the column. The air stream then passes
to determine the optimal operating conditions as well as the
through the hygrometer to monitor the water leaving the
optimal performance of any adsorbent.
column.
The total amount of water adsorbed was calculated
Before the adsorbent was discarded, a second regenera-
using •quations (1) (4):
tion run was performed with it to measure the amount of
water it had adsorbed.
N
X
Controlled variables: TotalH2OAdsorbed ź dðmH2OAdsorbedÞ (1)
nź1
" Relative humidity at the inlet of the column
where
" Flow-rate
" Regeneration temperature RHinlet
dðmH2OAdsorbedÞź dma gair dma (2)
" Amount of adsorbent
100
RHoutlet
gair
Dependant variables:
100
is the amount of water adsorbed between two time steps
" Temperature at the outlet of the column
(tn and tn-1) and
" Relative humidity at the outlet of the column
gair ź :0044e0:0607T; T in C (3)
Measured and assumed to be constant:
is the maximum amount of water the air can hold at a given
" Inlet temperature, it is measured at the beginning and
temperature at 100% relative humidity. •quation (3) was
then assumed to be constant for the rest of the
extrapolated from the Ä„erry s Chemical •ngineering
experiment
Handbook [23], by fitting an exponential curve to the
" Flow-rate and RH,
psychometric chart for temperatures between 15 C and
" Ambient temperature
35 C. The mass of air that has passed through the column
" Ä„ressure of the system
between two time stamps (i.e., between tn and tn-1), which
is denoted by dFlow, is calculated as follows:
1m3
2.1. Assumptions
dma ź ðtn tn 1Þ ðQÞ rair (4)
1000l
Following assumptions were made to simplify calculations
The total energy released from the system is calculated
or simplify complex concepts:
using •quations (4) (7):
" The relations for heat capacity, water capacity and dðEnergyReleasedÞ Åº dma ðTn Tn 1Þ Cp;a
density of humid air, which are presented in the analysis ź dE (5)
section, are true in the temperature range used for the
is the energy released between two time stamps, tn and tn-1.
experiments.
To calculate the total energy released from the start until
" The inlet relative humidity remains constant throughout
the end of the experiment, these d(Energyreleased) values
the experiment. The air coming out of the air source and
need to be added up as follows:
into the bubbler is very dry, which means that there is
XN
some evaporative cooling occurring in the bubbler,
TotalEnergyReleased ź dðEnergyReleasedÞn (6)
nź1
which gradually lowers the temperature of the water,
which lowers the temperature of the air and its water
capacity. It is assumed that because the experiments Cp;a ź 1:005 þ 1:88 H (7)
are relatively short, the air entering the system is at its
initial measured value and does not change as the where H = RH/100. •quation (7) was taken from Ä„erry s
experiment progresses. Chemical •ngineering Handbook [23], and it corrects the
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
D. Dicaire and F. H. Tezel Improvement of energy density for adsorption thermal energy storage
heat capacity of air with respect to the amount of moisture it
contains.
3. RESULTS AND DlSCUSSlON
Incorporating a hot air purge into the system has shown
great results compared with using an oven for regeneration.
Ä„reliminary results suggest that all the water adsorbed into
the system can be desorbed through this method. The use
of insulation also is critical to the success of these experi-
ments as the system is made of stainless steel, which is a
very good heat conductor.
Three different parameters were of interest for the current
Figure 5. Temperature as a function of time for Zeolite 13X with
adsorbent screening process: flow rate, relative humidity and
100% relative humidity and fresh adsorbent sample for each run
adsorbent type. Flow rate affects the amount of usable
at atmospheric pressure and ambient temperature at the inlet.
energy that can be released from the column. The smaller
the flow rate, the longer the contact time between the
Relative humidity was hypothesized to do the
adsorbent and the adsorbate, which is good for diffusion,
following:
but it also allows for more heat to dissipate. A faster flow rate
has less contact time but also has less energy lost from the
1) Control the rate of energy release from the adsorbent
system. Figure 4 shows the amount of energy released from
bed.
the column as a function of the flow rate. Although these
2) Affect the overall energy available from the
runs were only performed on zeolite 13X, similar trends
adsorbent column.
are to be expected from other adsorbents because the
geometry of the system is the same. Higher flow rates
At any given temperature and pressure, the relative
reported higher release of energy.
humidity is the percentage of the total amount of water the
Figure 5 shows the temperature change as a function of
air could possibly hold. It represents a percentage of the
time for the same runs as Figure 4. •ach curve is very similar
maximum possible partial pressure of water in the air.
in shape, yet the higher flow rates have a sharper and slimmer
Ä„ressure has an effect on adsorption in terms of the
peak than the lower flow rate ones. It should be noted that,
adsorption equilibrium capacity. As partial pressure of water
although higher flow rates release more energy, Figure 5
in the air increases, the adsorption capacity of the adsorbent
shows that, in general, lower flow rates are able to obtain
will increase at equilibrium. When the humid air gets in
higher peak temperatures. It also was observed that the
contact with the dry adsorbent, the system will move towards
energy release times increased with decreasing flow rates.
equilibrium by water molecules diffusing into the adsorbent
If necessary, the system can sacrifice performance for a
because of adsorption capacity of the adsorbent. The amount
higher peak temperature or vice versa by controlling the
of water that will diffuse to reach equilibrium depends on the
flow rate.
amount available in the vapour phase and, therefore, on the
partial pressure of the water in the air. When there is a higher
RH, there is a higher partial pressure, and there is a larger
amount of water adsorbed. As the RH increases, so will the
amount of water adsorbed, which means that more energy
will be released because the energy released relies on the
amount of water adsorbed.
Rate of heat release also will be affected by RH because
as water moves from the air to the adsorbent, the driving
force is going to be the difference in concentrations
between the two phases. Therefore, a lower RH will have
a lower driving force than a higher RH. Therefore, as RH
increases, the driving force increases, and the time it
takes to saturate the column and release all the energy
decreases.
The results confirm these comments: by varying the RH,
it is possible to control the rate at which the energy is
Figure 4. Energy released as a function of flow rate for Zeolite
released and, therefore, the total time it takes to saturate the
13X with 100% relative humidity and fresh adsorbent sample
column and the maximum temperature reached during the
for each run at atmospheric pressure and ambient temperature
at the inlet. experiment as shown in Figures 6 and 7. The amount of
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
Improvement of energy density for adsorption thermal energy storage D. Dicaire and F. H. Tezel
released were obtained using 98% humidity and that the
rate of energy released decreases with decreasing relative hu-
midity. This trend also is reflected in Figure 7, which shows
the temperature lift caused by the adsorbent system. The
98% run reach much higher temperatures than the other runs
but was much shorter. Controlling the relative humidity can
tailor the energy release in terms of rate of release as well as
temperature lift produced.
Once it was determined that the most energy could be
extracted from the system at 100% relative humidity and
8l/min flow rate, three different adsorbents were tested
under these conditions. The runs consisted of loading the
adsorbent, regenerating it, running an adsorption cycle and
then regenerating the same adsorbent sample to perform
another adsorption run. This was done four times with each
Figure 6. Energy release rate as a function of the normalized
adsorbent. The goal was to identify the adsorbent with the
water content in the adsorbent for Zeolite 13X with 8 l/min flow
highest energy release and to ensure that with a complete
rate and fresh adsorbent samples for each run at atmospheric
regeneration at 250 C for 8 hours, the adsorbent does not
pressure and ambient temperature at the inlet.
loose performance as it is recycled. The resulting data are
shown in Figure 8.
All three adsorbents used show similar trends; the energy
released values vary within a certain error range but can be
considered to be constant overall. The zeolite 13X and Na
LSX released approximately the same amount of energy,
but the best performing adsorbent was found to be the hybrid
activated alumina and zeolite 13X (AA/13X). The maximum
amount of energy released from the system was around 5 kJ.
Heat loss has a great impact on the performance of the
system. Insulation was placed on the outside of the column
to minimize the loss of the heat to the environment. To
approximate the amount of heat loss, a thermocouple was
placed on the column, in the middle of its length. It measured
the temperature that the outside of the column reached. This
temperature increased as the adsorption front moved through
the column and then decreased after it had passed. The
weight of the column was 91 g, and it was assumed that each
length of that tube was warmed to the maximum temperature
Figure 7. Temperature as a function of time for Zeolite 13X with
measured by the thermocouple as the adsorption front
8 l/min and fresh adsorbent for each run at atmospheric pressure
moved through the column and then cooled down to room
and ambient temperature at the inlet.
temperature. The tube is connected to the large stainless
steel connectors, which link it to the rest of the system,
water in the column has been normalized for Figure 6
because the capacity of the adsorbent varies with the relative
humidity. Therefore, the x-axis for this figure shows the
percentage of saturation of the adsorbent as a fraction of
the maximum capacity of the adsorbent at that relative
humidity.
Figure 6 displays the amount of energy released per
minute (obtained from •quation (8)) as a function of the
amount of water in the adsorption column (as % of saturation
at the corresponding RH).
dE dðEnergyReleasedÞ
ź (8)
dt ðtn tn 1Þ
Figure 8. Energy released as a function of number of adsorption
runs performed with the same adsorbent sample for Zeolite
Ninety eight percent RH reaches a much higher peak than 13X, NaLSX and AA/13X at 8 l/min and 100% relative humidity
the 60% or 34% RH. This means that higher rates of energy at atmospheric pressure and ambient temperature at the inlet.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
D. Dicaire and F. H. Tezel Improvement of energy density for adsorption thermal energy storage
which are much heavier than the column but reach much The performance of the three adsorbents is similar, but
lower temperatures. Nine grams of weight was added to the activated alumina/zeolite 13X hybrid adsorbent seems
the column weight to account for losses through that addi- to be the most promising one.
tional equipment for a total of 100 g. The heat loss was Hot water tanks are the widest spread thermal energy
approximated by calculating the energy required to heat storage systems in use today. Their energy densities are
100 g of SS from room temperature to the maximum usually around 50 to 60 kWh/m3 depending on the
temperature measured by the thermocouple during each run. temperature difference being employed. Because they store
The calculated heat losses are shown in Figure 9 along sensible energy, no matter how much insulation is used, their
with the usable energy released from each run. As expected, energy is lost within few days because of temperature driving
the lower flow rates have slightly higher heat loss, indicating force between the system and the environment. However,
that a higher residence time results in more energy adsorbents can provide the same performance with less than
dissipation. The total energy curve, calculated by adding heat one third of the volume, and the energy stored in adsorbents
loss and useable energy, follows the same trend as the does not degrade with time as it does in hot water tanks.
useable energy curve. It also is evident from the total energy Because the energy stored in the adsorbents is in the form
curve in Figure 9 that if the heat loss were harnessed as of heat of adsorption, and not the sensible heat, it cannot be
usable energy, the performance of the system would be released from it, until moist air is passed through the
greatly improved. adsorbent to start adsorption. Therefore, there would not be
any premature loss of energy from the adsorption system
while it is not in use.
3.1. Performance
The performance of the system can be calculated on an
3.2. System modifications for improved
energy density basis by dividing the total energy released
performance
from the system by the volume of the column packed with
adsorbent (which is constant at 9.04•-6 m3). The values were
After the completion of the adsorbent screening, it was
converted to kWh for easy comparison with other heating
determined that the AA/13X hybrid adsorbent had the
systems from the literature and are displayed in Table II.
most promising performance. The decision was made
to modify the system and perform further tests with that
adsorbent. The following modifications were made to
the system:
" A new column was built with the following dimensions:
length: 69.5 mm, internal diameter: 33.91 mm, weight:
315 g and capacity: 53 g of adsorbent. The new column
had an L/D ratio of 2.04 and had a much smaller exter-
nal surface area per volume ratio to decrease the heat
loss to the stainless steel column and the environment.
" A new hygrometer chamber was designed to obtain
better humidity readings.
" An Ultrasonic Humidifier was added to the bubbler to
guarantee 100% RH during each run.
" The location of the thermocouple was changed and
Figure 9. Heat loss, useable energy and total energy released
made closer to the exit of the column to be more
as a function of the flow rate for Zeolite 13X with 100% relative
representative of the temperature of the gas exiting the
humidity and fresh adsorbent for each run at atmospheric
column.
pressure and ambient temperature at the inlet.
" The adsorbent was not discarded after each run; it was
cycled for many experiments because it had been
shown that there was no significant loss of performance
TabIe II. Performance of system with different adsorbents
from regenerating previously used adsorbent.
studied given as energy densities.
" Higher flow rates of 16 and 24 l/min were used.
Energy density
After these modifications, the maximum energy den-
Adsorbent (kWh/m3)
sity performance achieved by the system with AA/13X
Zeolite 13X 153.9
at 24 l/min in the new column was 200 kWh/m3. This
Na LSX 146.6
value is much higher than the highest energy density
Activated alumina/Zeolite 13X hybrid 160.6
values given in the literature for an adsorption thermal
Activated alumina/Zeolite 13X hybrid (after 200.0
energy storage system (165 kWh/m3 reported by Dawoud
modifications to the system)
et al. [18]).
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
Improvement of energy density for adsorption thermal energy storage D. Dicaire and F. H. Tezel
Greek letters
4. CONCLUSlONS
The use of adsorbent beds for thermal energy storage has
g =The amount of water the air can hold
been proven. •nergy has been released from dry adsorbents
at a given temperature at saturation
by passing humid air through packed beds of adsorbents,
(at 100 % RH) (kg H2O/kg air)
and the different parameters affecting the system have been
r = Density (kg/m3)
studied on a small scale. Different adsorbents have been
tested, and the maximum performance of the system has
been found with an activated Alumina/Zeolite 13X hybrid
adsorbent, which had an energy density of 200 kWh/m3. It
also has been proven that the system can be completely ACKNOWLEDGEMENTS
regenerated and that there is no loss of performance with
regeneration. Authors would like to acknowledge the financial support
received from the Natural Science and •ngineering
Research Council of Canada for this study.
NOMENCLATURE
REFERENCES
Cp,a = Specific heat of air at constant
1. Demirbas F. Thermal •nergy Storage and Ä„hase
pressure (kJ/kg C)
dE
= Amount of energy released in one Change Material: An Overview. Energy Sources,
dt
minute (kJ/min)
Part B 2006; 1:85 95.
d(EnergyReleased) = Amount of energy transfer to the
2. •rvin G. Solar heat storage using chemical reactions.
air between two measurements,
Journal of Solid State Chemistry 1977; 22:51 61.
that is, between tn and tn 1. It is
3. International •nergy Agency. •nergy Conservation
assumed that this amount of
through •nergy Storage Ä„rogramme, March 2006.
energy corresponds to the
Downloaded from the following website on December
amount of energy released by the
16th, 2009: http://www.iea-eces.org/files/brochure06.pdf
zeolite bed (kJ)
4. Rosen MA. Second law-analysis of aquifer thermal en-
dma = The mass of air that has passed
ergy storage systems, energy. Permagon Press 1999;
through the column between two
24:167 182.
time stamps, that is, between
tnand tn 1 (kg) 5. Ä„ahud D, Matthey B. Comparison of the thermal
dðmH2OAdsorbedÞ = Indicates how much water was
performance of double u-pipe borehole heat exchan-
adsorbed in the column by
gers measured in situ. Energy and Buildings 2001;
comparing the inlet and the
33:503 507.
outlet (kg)
6. Nelson J•B, Balakrishnan AR, Murthy SS. Transient
E = •nergy transferred (kJ)
analysis of energy storage in a thermally stratified
H = Ä„arameter in •quation (7) that is
Water Tank. International Journal of Energy Research
related to RH (dimensionless)
1998; 22:867 883.
mads = Mass of the adsorbent (kg)
7. Shigeishi RA, Langford CH, Hollenbone BR. Solar
n =Counter for time stamp, tn, at which
energy storage using chemical potential changes
data are taken (dimensionless)
associated with drying of zeolites. Solar Energy
N = Number of total time stamps (i.e.,
data points) (dimensionless) 1979; 23:489 495.
RH = Relative humidity (%)
8. Drake Landing Solar Heating Community. Retrieved
RHinlet = Relative humidity measured at the
from the following website on December 16th, 2009:
inlet of the column (%)
http://www.dlsc.ca/
RHoutlet = Relative humidity measured at the
9. Akgun M, Aydun O, Kaygusuz K. •xperimental study
outlet of the column (%)
on melting/solidification characteristics of a paraffin as
Tn = Temperature at given time stamp
Ä„CM. Energy Conversion & Management 2007;
tn ( C)
48:669 678.
tn = Time stamp when a measurement
10. Hoshi A, Mills DR, Bittar A, Saitoh TS. Screening of
was taken (min)
high melting point phase change materials in solar
Q = Total volumetric flow rate of
thermal concentrating technology based on CLFR.
air passing through the system
(litres/min) Solar Energy 2005; 79:332 339.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er
D. Dicaire and F. H. Tezel Improvement of energy density for adsorption thermal energy storage
11. Kreetz H, Lovegrove K. •xergy analysis of an 17. Jaenchen J, Ackermann D, Stach H, Broesicke W.
ammonia synthesis reactor in a solar thermochemical Studies of the water adsorption on zeolites and modified
power system. Solar Energy 2002; 73(3):187 194. mesoporous materials for seasonal storage of solar heat.
12. Gopal R, Hollenbone BR, Langford CH, Shigeishi RA. Solar Energy 2004; 76:339 344.
The rate of solar energy storage and retrieval in a 18. Dawoud B, Amer •H, Gross DM. •xperimental investi-
zeolite water system. Solar Energy 1981; 28(5): gation of an adsorptive thermal energy storage. Interna-
421 424. tional Journal of Energy Research 2007; 31:135 147.
13. Gantenbein Ä„, Brunold S, Flueckiger F, Frei U. Sorbtion 19. Ozaydin S, Kocer G, Hepbasli A. Natural Zeolites in
materials for application in solar heat energy storage. •nergy Application. Energy Sources, Part A: Recovery,
University of Applied Sciences Rapperswil: Switzerland, Utilization, and Environmental Effects 2006; 28(15):
2001. 1425 1431.
14. Storch G, Hauer A, Cost-•ffectiveness of a Heat 20. Tsutsumi K, Mizoe K. Heats of adsorption of water on
•nergy Distribution System Based on Mobile Storage hydrophobic zeolites. Colloids and Surfaces 1988;
Units: Two Case Studies, Ä„roceeding: •costock, 37:29 38.
Stockton USA, 2006, p73. 21. Mishin IV, Brueva TR, Kapustin GI. Heats of adsorp-
15. Jaehnig D, Hausner R, Wagner W, Isaksson C. tion of ammonia and correlation of activity and acid-
Thermo-chemical storage for solar space heating in a ity in heterogeneous catalysis. Adsorption 2005;
single-family house, Ä„roceeding: •costock, Stockton 11:415 424.
USA, 2006, p100. 22. He Y, Seaton NA. Heats of adsorption and adsorption
16. Hauer A. Thermal energy storage with zeolite for heterogeneity for methane, ethane and carbon dioxide
heating and cooling. In: Ä„roceedings of the 7th in MCM-41. Langmuir 2006; 22:1150 1155.
International Sorption Heat Ä„ump Conference ISHÄ„C, 23. Ä„erry RH, Green DW. Perry s Chemical Engineers
pp. 385 390, Shanghai, 2002. Handbook (7th •dn). McGraw-Hill, Tokyo, 1997.
Int. J. Energy Res. (2012) © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/er