Content uploaded by A. Elsayed
Author content
All content in this area was uploaded by A. Elsayed on Nov 29, 2015
Content may be subject to copyright.
Content uploaded by A. Elsayed
Author content
All content in this area was uploaded by A. Elsayed on Nov 29, 2015
Content may be subject to copyright.
International Conference on Applied Energy
ICAE 2012, Jul 5-8, 2012, Suzhou, China
Paper ID: ICAE2012-10630
ETHANOL ADSORPTION CHARACTERISTICS ON METAL ORGANIC
FRAMEWORKS FOR LOW TEMPERATURE COOLING APPLICATIONS
Ahmed Rezk, Raya AL-Dadah
1
, Saad Mahmoud, Baoshing Shi and Ahmed Elsayed
1 School of Mechanical Engineering, University of Birmingham, Birmingham, United Kingdom
r.k.al-dadah@bham.ac.uk
0044-12141435
ABSTRACT
Adsorption cooling is a promising technology that can
effectively utilize waste heat from many industrial
processes for refrigeration and air conditioning.
Commercially available adsorption systems are based on
silica gel / water, zeolites/water and activated
carbon/ammonia pairs. These suffer from various
limitations including achieving cooling below zero for
refrigeration applications and highly poisonous refrigerant
like the ammonia. MOFs are new micro-porous materials
with exceptionally high porosity and large surface area that
can be used as adsorbents. Ethanol is a natural refrigerant
with low environmental impact and non toxic that can
operate at temperatures below zero. This paper
experimentally investigates the characteristics of two MOFs
namely Cu-BTC (HKUST-1) and MIL-100-Fe in terms of
ethanol adsorptivity in order to achieve low temperature
refrigeration. Results revealed that HKUST-1 outperformed
MIL-100-Fe in terms of the ethanol adsorption uptake. At
20
o
C, HKUST-1 adsorbed up to 0.5 kg
ethanol
/kg
ads
of ethanol
while the MIL-100-FE adsorbed 0.3 kg
ethanol
/kg
ads
. Also,
results from theoretical modeling have shown that using
HKUST-1/ Ethanol pair can achieve evaporator temperature
close to -10
o
C and brine outlet temperature of -3.5
o
C which
highlights the potential of HKUST-1/ethanol in such
application.
Keywords: metal organic framework, ethanol,
adsorption, low temperature cooling.
NONMENCLATURE
Symbols
D
so
Pre-exponential constant [m
2
/s]
d/dt Change rate [1/s]
∆H
ads
Isosteric heat of adsorption [J/kg]
E
a
Activation energy [J/mol]
F
o
Adsorption rate constant
h Specific enthalpy [kJ/kg K]
LMTD Log mean temp difference [K]
M Mass [kg]
N
bed
Total number of increment
N Increment number
P Pressure [kPa]
R Universal gas constant [J/mol K]
R
p
Particle radius [m]
T Temperature [K]
t Time [s]
UA Overall conductance [kW/K]
W Uptake value [kg
water
/kg
silica
]
w* Equilibrium uptake [kg
water
/kg
silica
]
Greek symbol
∆ Difference
Σ Summation
Φ Flag
δ Flag
ξ Flag
γ Flag
Subscript
ads adsorbent
cond Condenser
evap evaporator
g gas
Hex Heat exchanger
i inside
k Increment index
l iquid
ref Refrigerant
s saturation
w water
1. INTRODUCTION
Commercially available adsorption systems capable of
producing low temperature cooling for refrigeration
applications are limited to ammonia based refrigerant that
has the disadvantage of being toxic. Therefore, there is a
need to use a nontoxic refrigerant in a low temperature
Paper ID: ICAE2012-10630
2
Copyright © 2012 by ICAE2012
adsorption system. Ethanol refrigerant has the advantages
of high thermal stability, low solidification temperature,
high degree of non-toxicity and compatible with wide range
of materials with good latent heat of evaporation [1].
Metal organic frameworks (MOFs) are new micro-porous
materials with exceptional high porosity, uniform pore size,
well-defined molecular adsorption sites and large surface
area (up to 5500m2/g). MOFs have two main components:
the organic linkers considered as organic secondary
building unit, act as struts that bridge metal centers known
as inorganic primary building units and act as joints in the
resulting MOF architecture. The two main components are
connected to each other by coordination bonds, together
with other intermolecular interactions, to form a network
with defined topology [2, 3]. MOFs have been originally
designed and investigated for gaseous fuel storage such as
hydrogen and methane [4, 5].
Based on an earlier work by the authors regarding the
water adsorptivity of MOF materials, HKUST-1 and MIL-100
have shown the highest water adsorption characteristics
due to their large porosity and surface area (2100m
2
/g and
1600m
2
/g) [6]. Similar conclusion was made with regard to
the superiority of HKUST-1 and MIL-100 to other MOF
materials [9]. This paper experimentally investigates the
adsorptivity behavior of HKUST-1 and MIL-100 MOF
towards ethanol adsorption. Results have shown that
HKUST-1 outperformed MIL-100 in terms of ethanol
adsorption. Therefore, the investigation has focused on
further investigating the adsorption characteristics of
HKUST-1 / ethanol. The HKUST-1 / ethanol adsorption
results were then used to investigate the capability of this
pair to generate low temperature cooling using a validated
empirical lumped analytical model of an existing two bed
adsorption system. The investigation includes adsorption
isotherms and kinetics modeling based on Langmuir and
LDF models respectively for HKUST-1 / ethanol pair. To the
author’s knowledge, there is no published work on
characterization of ethanol adsorption on HKUST-1 MOF.
2. EXPERIMENTAL SETUP
There are two methods for measuring adsorption
characteristics namely; volumetric method and gravimetric
method [7, 8]. In this investigation a dynamic vapour
sorption (DVS) gravimetric analyser has been used to study
the ethanol adsorption characteristics (adsorption
isotherms, kinetics and isosteric heat of adsorption) of the
HKUST-1 MOF, Figure 1. The DVS analyser has the
advantages of directly measuring the adsorbent mass and
preventing adsorbate vapour condensation on the moving
balance parts. The adsorbent mass is measured using
sensitive recording microbalance (Cahn D200) which has
high long-term stability in microgram level as it adsorbs
controlled concentrations of water or organic vapours. Dry
Nitrogen is used to purge the balance head and reaction
chamber prior to sample loading. The purge flow is
automatically controlled to prevent vapour condensation in
the balance head and hence accurate uptake measurement
is achieved. Mass flow controller is used to control the
vapour pressure with a mixture of dry and saturated vapour
gas. The test conditions were verified using optical vapour
pressure sensor and RTD temperature probe very close to
sample pan. The DVS analyser is automatically controlled by
a dedicated PC microcomputer, which is interfaced with the
microbalance. The accuracy of the DVS analyser
microbalance is verified by using 100 mg standard
calibration mass, where the expected accuracy of the
tested sample is ±0.05 mg.
Samples of 10 mg each has been placed in the reaction
chamber and were locally dried at fixed temperature of
55˚C until the condition of no change of mass, then allowed
to undergo through adsorption / desorption tests at various
partial pressures. The sample mass is recorded every 4
seconds at different vapour pressure values to determine
the adsorption kinetics. The samples isotherms are
measured at each value of vapour pressure at the point of
no change in adsorbent mass during adsorption /
desorption to determine the isothermal hysteresis.
Dual vapour generator
Vapour-1
Vapour-2
Mass flow controller
Dry gas
Gas for
vapour-1
Gas for
vapour-2
Dry gas
Temperature controlled
enclosure
Microbalance
Reaction
chamber
Counter weight
chamber
a
Figure 1, DVS analyser schematic and pictorial diagram
Paper ID: ICAE2012-10630
3
Copyright © 2012 by ICAE2012
3. EXPERIMENTAL RESULTS
Figure 2 shows SEM image of the tested MOF materials
and table 1 presents the granular size, BET surface area and
bulk density for this material. Figure 3 shows the
adsorption isotherms for HKUST-1 and MIL-100 at 20
o
C
which clearly indicates that HKUST-1 outperforms MIL-100
in terms of the ethanol uptake. HKUST-1 can adsorb up to
0.48kg
ref
/kg
ads
while MIL-100 adsorbs only up to 0.3
kg
ref
/kg
ads
of ethanol.
Previous work [9] have shown that HKUST-1 suffered
from stability issue in terms of deteriorated adsorptivity of
water. Figure 4 shows repeated tests for the adsorptivity of
ethanol at 20˚C. It is clear from this figure that HKUST-1 has
exhibited a stable performance with ethanol after the first
test. Figure 5 shows the adsorption and desorption
isotherms of ethanol at 20 and 30˚C. The results show
insignificant difference between the adsorption and
desorption curves indicating no hysteresis.
a
b
Figure 2, SEM image of HKUST-1 and MIL-100
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0 1 2 3 4 5 6
Equilibrium Uptake [ kg
ref
/ kg
ads
]
Vapour Pressure [ kPa ]
HKUST
-
1 Adsorption
MIL
-
100 Adsorption
Figure 3, Adsorption isotherms of
MIL-100 & HKUST-1/Ethanol
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0 1 2 3 4 5 6
Equilibrium Uptake [ kg
ref
/ kg
ads
]
Vapour Pressure [ kPa ]
Adsorption cycle #1
Adsorption cycle #2
Adsorption cycle #3
Adsorption cycle #4
Adsorption cycle #5
Figure 4, Stability of HKUST-1/Ethanol
Table 1 Physical properties of HKUST-1
Property HKUST-1 MIL-100
Granules size 16 µm 5 µm
BET Area 1500-2100 m
2
/g 1300-1600 m2/g
Bulk density 0.35 g/cm
3
0.16-0.35 g/cm3
0
0.1
0.2
0.3
0.4
0.5
0.6
0 2 4 6 8 10
Equilibrium Uptake [ kg
ref
/ kg
ads
]
Vapour Pressure [ kPa ]
HKUST-1 Adsorption 20˚C
HKUST-1 Desorption 20˚C
HKUST
-
1 Adsorption 30˚C
HKUST-1 Desorption 30˚C
Figure 5, Adsorption isotherms of HKUST-1/Ethanol
4. HKUST-1 / ETHANOL ADSORPTION
CHARACTERICS MODELLING
4.1 Isothermal modeling
There are many adsorption isotherms models such as
Dubinin-Astakhov (D-A), Sips, Tóth, Freundlich, Langmuir,
Temkin and Hill-de Boer [10-13]. The selection of the
suitable isotherm model depends on its profile. HKUST-1 /
ethanol isotherm shows a hyperbolic profile that is
compatible with Langmuir model, equation 1. Experimental
results from figure 4 have been used to conduct hyperbolic
regression for two isotherms at different temperatures 20
and 30°C during adsorption / desorption. Figure 6 shows
that the Langmuir model with W
∞
and b of 0.479 kg
e
kg
ads
-1
and 121.95 respectively has predicted the experimental
results with 95% confidence.
(
)
(
)
[
]
ss
*
PPb1PPbWW ⋅+⋅=
∞
[1]
Paper ID: ICAE2012-10630
4
Copyright © 2012 by ICAE2012
Partial Pressure [ - ]
0.0 0.2 0.4 0.6 0.8 1.0
Equilibrium Uptake [ kg/jg ]
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Langmuir Model
Equilibrium Uptake
95% Confidence Band
95% Prediction Band
Figure 6, Adsorption isotherms of HKUST-1/Ethanol
(Langmuir model)
4.2 Kinetic modeling
Figure 7 presents the experimental isothermal temporal
of ethanol vapour uptake for HKUST-1 at different vapour
pressure steps and 20˚C. Data from this figure can be used
to determine the kinetic performance of HKUST-1 ethanol
pair. Theoretically, the linear driving force (LDF) model is
used for simulating the adsorption kinetics of different
pairs [14-18] as given by equation 2.
(
)
w)(wRTEExpCdtdw
*
a
−⋅−⋅=
[2]
2
pso
RDFoC =
[3]
Where E
a
is the activation energy and C is a constant that
is a function of adsorbent granule size R
p
, surface diffusion
coefficient D
so
and pre-exponential constant Fo as given by
equation 3. Using data from two isothermal temporal
uptake curves at 20 and 30
o
C, values of E
a
and C were
found to be 445.9 J mol
-1
and 0.01149 s
-1
respectively. This
developed model has shown good agreement with the
experimental results as illustrated in figure 7.
0 20 40 60 80 100 120 140
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
Ethanol Vapour Uptake [ kg
ref
/kg
ads
]
Time [ min ]
Atcual Uptake
Predicted Uptake
0.00
0.75
1.50
2.25
3.00
3.75
4.50
5.25
Vapour Pressure
Vapour Pressure [ kPa ]
Figure 7, Adsorption kinetics of HKUST-1/Ethanol
(LDF model)
5. SYSTEM MODELING
5.1 System description
Figure 8 shows a schematic diagram of a commercially
available two-bed silica gel / water adsorption system. This
system was modelled, tested and the model was validated
with maximum absolute deviation of 17.8% [19]. The model
of this system was modified to accommodate the HKUST-
1/ethanol working pair in order to produce low
temperature cooling. Each adsorbent bed is connected to
the evaporator or condenser by flap valves operated by the
effect of pressure difference between heat exchangers
during adsorbing or desorbing respectively. On the other
hand, the flow of cooling and heating water (secondary
fluids) to the adsorber and desorber respectively, flow of
the chilled glycol solution through the evaporator and flow
of cooling water to the condenser are controlled by 12
pneumatic valves.
Physically, the adsorbent bed heat exchanger is
constructed from plain copper tubes with aluminium
rectangular fins and the adsorbent particles are packed to
fill the gaps between fins. The adsorbent bed is covered by
a metal mesh to prevent the falling of adsorbent particles.
These adsorbent beds are installed in two-bed system
incorporating mass and heat recovery schemes. The
operation modes and secondary flow valving control
method were presented in details in [19].
Figure 8, Schematic diagram of the system
5.2 Model description
The simulation model of the adsorption system was
constructed from four sub-models describing the heat and
mass transfer performance of evaporator, condenser,
adsorber and desorber. The four sub-models were linked
together taking into account the various operating modes.
Equations 4-6 present the energy balance equations for
adsorbent bed, evaporator and condenser respectively,
where the adsorbent, adsorbate and heat exchanger metal
are assumed to be individually momentarily at the same
temperature. Equation 7 presents the refrigerant mass
Paper ID: ICAE2012-10630
5
Copyright © 2012 by ICAE2012
balance in the evaporator taking into account no flow
condition in case of heat and mass recovery.
(
)
( )
( ) ( )
{ }
( ) ( ) ( )
{ }
( )
∑
=
=
−
−−
−
×−++
−−+
−
∂•+=
++
+
bed
Nn
1n
bednbedsgsgsg
sgsg
bedbedgHexHexg
bedHexgHexg
bed
metbedmetbedsgsg
bedrefsgsgbedwwbed
LMTDdUAζ1dtdw∆HφM
dtdwM
T,PhT,Phγ1
T,PhThγ
φdtdT
CMCM
)(TCpwMTCζM
[4]
(
)
[
]
[ ]
dtdELMTDUA
dtdwMhhφ
dtdTMCMTCp
pumpevapevap
sgsgoutevap,ref,inevap,ref,
evapmetevapmetevapevapref,evapfref,
+×
+−
=+
−−
[5]
(
)
[
]
( )
( )
[ ]
condcond
sgsgbedcondrefgcond,ref,lcond,ref,
condmetcondmetcondcondref,condlref,
LMTDUA
dtdwMTTCphhφ
dtdTMCMTCp
×
+−+−
=
+
−−
[6]
(
)
dtdwdtdwMφdtdM
adsdessgevapf,ref,
+⋅−=
[7]
Where M, C, Cp, T, LMTD, P, h, w, ∆H
ads
and t are the
mass, specific heat, specific heat at constant pressure,
temperature, log mean temperature difference, pressure,
specific enthalpy, uptake value, isosteric heat of
adsorption, and time respectively. The subscripts bed, evap
and cond refer to adsorbent bed (ads/des), evaporator and
condenser condition respectively and w, sg, ref and met
refer to water, silica gel, refrigerant and metal respectively.
Subscripts g, f, refer to fluid vapour and liquid condition
respectively and Hex refers to the heat exchanger that
interconnect with the adsorbent bed. A group of flags
(
, γ, ζ,
∂
φ
) were used to enable or disable some of the
equation terms based on the operating mode as shown in
table 2. More details about the model can be found in [19]
Table 2, Flags
Mode
ζ
φ
∂
γ
Ads-evaporation 0 1 1 1
Des-condensation 0 1 0 0
Mass recovery 1 0 1 0
Heat recovery 0 0 1 0
The rate of adsorption and adsorption isotherms of
HKUST-1/ethanol were calculated using LDF and Langmuir
models respectively as developed in section 4 based on
experimental results. The thermal properties of HKUST-1
were obtained from [20] and those of ethanol and glycol
were obtained from Refprob®. Figure 9 shows the
predicted evaporation temperature and the glycol outlet
temperature for the operating conditions listed in table 3. It
is clear from this figure that evaporation at -10
o
C was
achieved and a glycol outlet temperature of -3.5
o
C. These
results highlight the potential of using HKUST-1/ethanol for
low temperature cooling.
Table 3, Operating conditions
Conditions Value Condition Value
m
cw
66.6 L/s T
cw,i
29˚C
m
hw
18.3 L/s T
hw,i
88.5˚C
m
gl
2 L/s T
gl,in
2˚
0 200 400 600 800 1000
-12
-10
-8
-6
-4
-2
0
2
4
6
Temperature [
o
C ]
Cycle Time [ s ]
Refregerant Temp
Brine Inlet Temp
Brine Outlet Temp
Figure 9, Adsorption cooler cyclic operatin
6. CONCLUSIONS
MOFs are new micro-porous materials with exceptionally
high porosity and large surface area that can be used as
adsorbents. Ethanol is a natural refrigerant with low
environmental impact and non toxic that can operate at
temperatures below zero. This paper investigates the use of
HKUST-1/ethanol pair for low temperature adsorption
cooling. Experimental results have shown that HKUST-1
MOF can adsorb ethanol and exhibit a stable performance.
The adsorption characteristics of HKUST-1/ethanol can be
modeled using the Langmuir model for isotherms and the
linear driving force (LDF) for adsorption kinetics. Results
from theoretical modeling have shown that using this pair
can achieve evaporator temperature close to -10
o
C and
brine outlet temperature of -3.5
o
C which highlights the
potential of MOF/ethanol in such application.
REFERENCE
[1] Aldadah RK, Rezk ARM. Empirical Simulation Model of
Silica Gel/Water Adsorption Chiller. ASME-ATI-UIT, Thermal
and Enviromental Issues in Energy Systems. 2010. Sorrento,
Italy.
[2] Küsgens P, Rose M, Senkovska I, Characterization of
metal-organic frameworks by water adsorption.
Microporous Mesoporous Mater. 120 (2009) 325-330.
[3] Qiu S, Zhu G, Molecular engineering for synthesizing
novel structures of metal-organic frameworks with
multifunctional properties. Coord Chem Rev. 2009;
253:2891-2911.
[4] Deng S, Sorbent tecjnology. Encyclopedia of Chemical
Processing. DOI: 10.1081/E-ECHP-120007963 2006; 2825-
2845.
[5] Grant Glover T, Peterson GW, Schindler BJ, MOF-74
building unit has a direct impact on toxic gas adsorption.
Chem Eng Sci. In Press, Accepted Manuscript (2010).
[6] Rezk A, Al-Dadah R, Mahmoud S, Experimental
Investigation of Metal Organic Frameworks Characteristics
for Water Adsorption Chillers. Under Review for publication
in J Mech Eng Sc, Proc. of IMechE Part C 2012.
[7] Wang XL, Chua HT, Gao LZ, A thermogravimetric
analyzer for condensable gas adsorption under
Paper ID: ICAE2012-10630
6
Copyright © 2012 by ICAE2012
subatospheric conditions. J Therm Anal Calorim 2007; 90:
935-940.
[8] Wang X, Zimmermann W, Ng KC, Investigation on the
isotherm of silica gel+water systems TG and volumetric
methods. J Therm Anal Calorim 2004; 76: 659-669.
[9] Henninger SK, Schmidt FP, Henning HM, Water
adsorption characteristics of novel materials for heat
transformation applications. Appl Therm Eng. 2010; 30:
1692-1702.
[10] Lin C-I, Wang L-H, Rate equations and isotherms for
two adsorption models. J Chin Inst Chem Eng 2008; 39: 579-
585.
[11] Llano-Restrepo M, Mosquera MnA, Accurate
correlation, thermochemistry, and structural interpretation
of equilibrium adsorption isotherms of water vapor in
zeolite 3A by means of a generalized statistical
thermodynamic adsorption model. Fluid Phase Equilib
2009; 283: 73-88.
[12] Akpa OM, Unuabonah EI, Small-Sample Corrected
Akaike Information Criterion: An appropriate statistical tool
for ranking of adsorption isotherm models. Desalination
2011; 272: 20-26.
[13] Hamamoto Y, Alam KCA, Saha BB, Study on adsorption
refrigeration cycle utilizing activated carbon fibers. Part 2.
Cycle performance evaluation. Int J Ref. 2006; 29: 315-327.
[14] Chua HT, Ng KC, A Malek, Modeling the performance
of two bed, silica gel-water adsorption chiller. Int J Ref.
1999; 22: 194-204.
[15] Saha BB, Koyama S, Kashiwagi T, waste heat driven
dual-mode, multi-stage, multi-bed regenerative adsorption
system. Int J Ref. 2003; 26: 749-757.
[16] Akahira A, Alam KCA, Hamamoto Y, Mass recovery
adsorption refrigeration cycle--improving cooling capacity.
Int J Ref. 2004; 27: 225-234.
[17] Saha BB, El-Sharkawy II, Chakraborty A, Study on an
activated carbon fiber-ethanol adsorption chiller: Part I -
system description and modelling. Int J Ref. 2007; 30: 86-
95.
[18] Saha BB, Chakaborty A, Koyama S, A new generation
cooling device employing CaCl2-in-silica gel-water system.
Int J Heat Mass Transfer 2009; 52: 516-524.
[19] Rezk ARM, Al-Dadah RK, Physical and operating
conditions effects on silica gel/water adsorption chiller
performance. App Energy 2012; 89: 142-149.
[20] Mu B, Walton KS, Thermal Analysis and Heat Capacity
Study of Metal Organic Frameworks. Journal of Physical
Chem 2011; 115: 22748–22754.
