Applied Thermal Engineering 29 (2009) 793–798

Contents lists available at ScienceDirect

Applied Thermal Engineering

journal homepage: www.elsevier.com/locate/apthermeng

Adsorption cooling cycles for alternative adsorbent/adsorbate pairs working

at partial vacuum and pressurized conditions
W.S. Loh a, I.I. El-Sharkawy b,1, K.C. Ng a,*, B.B. Saha b,*
a
Department of Mechanical Engineering, National University of Singapore, 10 Kent Ridge Crescents, Singapore 119260, Singapore
b
Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Kasuga-koen 6-1, Kasuga-shi, Fukuoka 816-8580, Japan

a r t i c l e i n f o a b s t r a c t

Article history: This article presents the performance analysis of both ideal single-stage and single-effect double-lift
Received 14 June 2007 adsorption cooling cycles working at partially evacuated and pressurized conditions. Six specimens of
Accepted 16 April 2008 adsorbents and refrigerant pairs, i.e., ACF (A-15)/ethanol, ACF (A-20)/ethanol, silica gel/water, Chemvi-
Available online 23 April 2008
ron/R134a, Fluka/R134a and MaxsorbII/R134a have been investigated. The relationships between equilib-
rium pressures, adsorbent temperatures and equilibrium adsorption concentrations (Dühring diagram)
Keywords: are presented. Parametric analyses have been carried out with various regeneration (desorption) and
Adsorption
evaporation temperatures. Theoretical analysis for adsorption cycles working in single-stage mode shows
Single-stage
Thermal activation
that ACF (A-20)/ethanol can achieve a specific cooling effect (SCE) of 344 kJ/kg_ads, which is followed by
Single-effect double-lift cycle the silica gel/water pair with 217 kJ/kg_ads at a regeneration temperature of 85 °C. On the other hand,
when the regeneration temperature is below 70 °C, single-effect double-lift cycle has a significant advan-
tage over single-stage cycle, at which the SCE is higher due to the reduction in adsorption bed pressure in
single-effect double-lift cycle.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction between the bed adsorbing vapor from the bed in desorption
mode. They reported that the SCE of a single-effect double-lift cy-
The study of thermally activated adsorption cooling cycle is cle can be improved up to 20% when the mass ratio of upper to
motivated by two factors, namely the need to switch to bottom bed is fixed at 2.5 provided that the cycle is driven by a
ozone-friendly working fluids for refrigeration in the wake of heat source of temperature 80 °C and the delivered chilled water
the ratification of the Montreal Protocol by several countries temperature is achieved at 7 °C [4]. However, since the system is
and environmental concern in reducing the greenhouse gas driven by low-temperature waste heat, Ng [5] has classified that
emissions in accordance with the Kyoto Protocol [1]. Thermally the COP of this thermally activated adsorption systems is nor-
activated adsorption cooling cycles or vapor adsorption refrigera- mally very low.
tion systems are considered to be the alternatives of conventional The primary step for developing vapor adsorption refrigeration
vapor compression cycles as they can be driven by low tempera- systems is the characterization of adsorbent/adsorbate pair, which
ture heat sources [2], i.e., industrial waste heat, solar hot water includes the experimental determination of adsorption parame-
system, etc. Apart from this, the systems are relatively simple ters. Recently, extensive effort has been put in to investigate the
to construct, as it has no major moving parts. Electricity is needed possible adsorption working pairs [6–9]. The performance of zeo-
for the pumping of heat source and coolant flows only. Moreover, lite/water [10], activated carbon/ammonia [11], activated carbon/
the heat source temperature can be further reduced as low as methanol [12], and silica gel/water [13–15] have been analyzed.
50 °C if single-effect multi-lift (multi-stage) regeneration is The adsorption parameters enable the derivation of the relation-
implemented [3]. Recent study shows that the cooling capacity ship between adsorption uptake (the quantity of adsorbate ad-
of the single-effect double-lift silica gel/water refrigeration cycle sorbed per unit mass of adsorbent), adsorption pressure and
can be improved by optimizing the adsorbent mass allocations temperature. From these perspectives, the current study aims to
analyze the performance of sub-atmospheric and pressurized
adsorption cooling cycles without heat recovery at which ACF (A-
* Corresponding authors. 15)/ethanol, ACF (A-20)/ethanol and silica gel/water function as
E-mail addresses: mpengkc@nus.edu.sg (K.C. Ng), bidyutb@cm.kyushu-u.ac.jp
adsorbent/refrigerant pairs for sub-atmospheric cycles, whilst
(B.B. Saha).
1
Permanent address: Mechanical Power Engineering Department, Faculty of
Chemviron/R134a, Fluka/R134a and Maxsorb II/R134a are em-
Engineering, Mansoura University, El-Mansoura, Egypt. ployed for pressurized cycles.

1359-4311/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.applthermaleng.2008.04.014
794 W.S. Loh et al. / Applied Thermal Engineering 29 (2009) 793–798

Nomenclature

a; b constants in Eq. (10) (–) T e;d temperature at the end of desorption process (K)
CE cooling effect per unit volume of adsorbent (MJ/m3) T s;a temperature at the start of adsorption process (K)
Cp specific heat capacity (kJ/kg K) T s;d temperature at the start of desorption process (K)
COP coefficient of performance (–) W adsorption capacity per unit mass of adsorbent at equi-
E characteristic energy of the assorted adsorbent/adsor- librium (kg/kg)
bate pair (J/kg) W0 maximum value of W (kg/kg)
hfg specific enthalpy difference between saturated vapor
and saturated liquid (kJ/kg) Subscripts
LH latent heat (kJ/kg K) ads adsorbent
n exponential parameter describes isotherm, (–) c condenser
P pressure (Pa) cri critical
Ps saturated pressure (Pa) d desorption
Qh heat added or rejected (kJ/kg_ads) e evaporator
qst heat of adsorption (kJ/kg) fg difference in property for saturated vapor and saturated
R gas constant (J/kg K) liquid
SCE specific cooling effect (kJ/kg_ads) g regeneration
T temperature (K) ref refrigerant/adsorbate
T e;a temperature at the end of adsorption process (K) w water

2. Theoretical analysis single-effect double-lift cycle was represented by doted lines. As

can be seen from the idealized Dühring diagram of Fig. 1, the sin-
The Dühring diagram (P–T–W) illustrates the relationship be- gle-effect double-lift cycle can operate in lower heat source tem-
tween the pressure, adsorbent temperature and adsorption con- perature as compared to single-stage cooling cycle for the same
centration at equilibrium state conditions, and from which the uptake capacity. Hence, single-effect double-lift cycle more prefer-
actual adsorption/desorption processes can be estimated. In the able in lieu of single-stage cycle for the utilization of low-temper-
present study, the Dubinin–Radushkevich (D–R) equation (Eq. ature solar or waste heat sources; this offers an attractive
(1)) is used to predict the adsorption characteristics of ACF possibility for improving energy conservation and efficiency in
(A-15)/ethanol and ACF (A-20)/ethanol [16]. Alternatively, for cooling purposes.
Chemviron/R134a, Fluka/R134a and Maxsorb II/R134a pairs, Dubi-
nin–Astakhov equation (Eq. (2)) is employed therein to predict the 3. Simulation of equilibrium model
adsorption isosteres [17]. For silica gel/water pair, the modified
Freundlich (S–B–K) equation (Eq. (3)) is used to predict the adsorp- 3.1. Single-stage adsorption cycle
tion isotherms [14,18].
The detailed analysis on thermodynamic processes of an ideal
W ¼ W 0 expf½RT lnðPs =PÞ=E2 g; ð1Þ
adsorption cycle has been discussed previously by Cacciola and
W ¼ W 0 expf½RT lnðPs =PÞ=En g: ð2Þ
Restuccia [21]. The ideal cycle (shown in Fig. 1 by solid lines) basi-
In the Dubinin equations, E is the characteristic energy of the as- cally consists of four sequential thermodynamic processes, i.e., pre-
sorted adsorbent/adsorbate pair, which can be evaluated experi- heating (c ? d), desorption (d ? a), pre-cooling (a ? b) and
mentally and R denotes the gas constant. The parameter n is an adsorption (b ? c) processes. An equilibrium model is adopted to
exponential constant, which provides the best fittings of the exper- evaluate the cycle performance in terms of specific cooling effect
imental isotherms. The term W denotes the specific mass of adsorp- (SCE) and coefficient of performance (COP). Being thermodynami-
tion or uptake (kg of adsorbate per unit mass of adsorbent) while
W 0 is the maximum uptake. The other parameters for D–R and
D–A equations are given by El-Sharkawy et al. [19]. In Eq. (3),
PðT w Þ and P s ðT s Þ are the saturation vapor pressure at water vapor
temperature T w and silica gel temperature T s , respectively. The
numerical values for A0 —A3 and B0 —B3 are determined by the least
square fit to experimental data [2]. The saturated vapor pressure
and temperature are correlated by Antoine’s equation as shown in
Eq. (4), where T is the absolute temperature [20].

W ¼ AðT s Þ½PðT w Þ=P s ðT s ÞBðT s Þ ; ð3Þ

where

AðT S Þ ¼ A0 þ A1  T s þ A2  T 2s þ A3  T 3s
BðT s Þ ¼ B0 þ B1  T s þ B2  T 2s þ B3  T 3s
 
3820
Ps ¼ 133:32  exp 18:3  : ð4Þ
T  46:1
The single-stage and single-effect double-lift adsorption cooling
cycles were superimposed onto the Dühring diagram in Fig. 1. The Fig. 1. Qualitative Dühring diagrams (P–T–W) for both single-stage and single-
single-stage cycle was represented by the solid lines whereas the effect double-lift cycles.
 W.S. Loh et al. / Applied Thermal Engineering 29 (2009) 793–798 795

cally equilibrium, all the thermal contributions are calculated 3.2. Single-effect double-lift adsorption cycle
based on heat and mass balance from the Dühring diagram. The
COP of the adsorption cooling cycle is defined as the ratio of the The working principle of the single-effect double-lift adsorption
SCE and heat input to the system, Q h , as shown in Eq. (5). cycle has been proposed by Saha et al. [18]. The ideal cycle as
The heat added to the system which is shown in Eq. (6) is simply shown in Fig. 1 by dotted lines, consists of two sequential thermo-
the summation of both the heat added to the adsorbent (Eq. (7)) and dynamic processes (I and II). These sequential thermodynamics
adsorbate or refrigerant (Eq. (8)). In Eq. (7), the specific heat capac- processes employed the same principles as in the single-stage sin-
ities of the assorted adsorbents are 924 J/kg K for silica gel, 900 J/ gle-lift cooling cycle, at which the pre-heating, desorption, pre-
kg K for activated carbon fibers and 930 J/kg K for activated carbons, cooling and adsorption processes for the first and second stages
respectively [19,22,23]. The first term of Eq. (8) is the isosteric heat- are indicated as c ? d0 ? a0 ? b and c00 ? d00 ? a00 ? b00 , respec-
ing of the adsorbate, while the second and third terms represent the tively. In the single-effect double-lift adsorption cycle, there is an
sensible heat added to the adsorbate and heat of adsorption, respec- additional pair of adsorbent beds incorporated in between the
tively. Heat of adsorption is described by Eq. (9) and the function evaporator and condenser of the system. These extra adsorbent
f ðTÞ as given in Eq. (10) is deduced through experimental isotherm beds allow the refrigerant pressure to increase via two consecutive
data for ethanol on activated carbon fibers (A-15 and A-20), HFC- steps from evaporator to condenser. At the same time, this enables
134a on activated carbon powder (Fluka and Maxsorb II) and acti- the single-effect double-lift cycle to reduce the regeneration tem-
vated carbon granule (Chemviron), whereas the heat of adsorption perature of the adsorption cycle by dividing the evaporating pres-
for silica gel/water system is equal to 2800 kJ/kg [20,2]. sure lift (pressure difference between the condenser and
The specific cooling effect (SCE) is defined as the difference be- evaporator) into two smaller steps. In the present simulation, the
tween the evaporator load and heat dissipated to cool the adsor- total mass of adsorbent for single-stage cycle is assumed to be
bate from condenser temperature to evaporator temperature, and equal to single-effect double-lift cycle, meaning that each bed in
is shown by Eq. (11). the single-effect double-lift cycle will carry only half the amount
of adsorbent of the adsorbent bed in single-stage adsorption cycle.
SCE The simulated performances of single-effect double-lift adsorption
COP ¼ : ð5Þ
Qh cycle are compared with the single-stage adsorption cycle in terms
Q h ¼ Q ads þ Q ref : ð6Þ of COP and SCE.
Z Ta
Q ads ¼ C pads dT: ð7Þ
Tc 4. Results and discussion
Z Td Z Ta Z W min
Q ref ¼ W max C pref dT þ WC pref dT þ qst dW : ð8Þ Simulations for the six working pairs [ACF (A-15)/ethanol, ACF
Tc Td W max
1=n (A-20)/ethanol, silica gel/water, Chemviron/R134a, Fluka/R134a
qst ¼ hfg þ E½lnðW 0 =WÞ þ E  f ðTÞ: ð9Þ
and Maxsorb II/R134a] are carried out at partial vacuum and pres-
b
f ðTÞ ¼ aðT=T cri Þ : ð10Þ surized conditions. The working pressures for ACF (A-15)/ethanol,
 Z Tc 
ACF (A-20)/ethanol and silica gel/water are between 1 and 10 kPa.
SCE ¼ ðW max  W min Þ LHT e  C pref dT : ð11Þ The Chemviron/R134a, Fluka/R134a and Maxsorb II/R134a based
Te

Table 1
Adsorbent temperature and uptake capacity for idealized adsorption/desorption cycle simulated at T evap ¼ 6:7 °C and T cond ¼ 29:4 °C in accordance to ARI Standard 560

Adsorbent/refrigerant pairs
ACF (A-15)/ethanol ACF (A-20)/ ethanol Silica gel/water Chemviron/R134a Fluka/R134a Maxsorb II/R134a
Single-stage cycle
Temperature (°C) Ta 80 80 80 80 80 80
Tb 52 52 50 53 53 53
Tc 30 30 30 30 30 30
Td 55 55 55 55 55 55
Uptake (kg/kg) Wmin 0.28 0.25 0.04 0.31 0.36 1.39
Wmax 0.47 0.58 0.12 0.34 0.45 1.75
DW 0.19 0.33 0.08 0.03 0.09 0.36

Single-effect double-lift cycle

1st Cycle
Temperature (°C) T 0a 66 66 65 63 67 67
Tb 52 52 50 50 53 53
Tc 30 30 30 30 30 30
T 0d 42 42 42 42 42 42
Uptake (kg/kg) Wmin 0.28 0.25 0.04 0.31 0.36 1.39
Wmax 0.47 0.58 0.12 0.34 0.45 1.75
DW 0.19 0.33 0.08 0.03 0.09 0.36
2nd Cycle
Temperature (°C) T 00a 66 66 65 63 67 67
T 00b 52 52 50 50 53 53
T 00c 30 30 30 30 30 30
T 00d 42 42 42 42 42 42
Uptake (kg/kg) Wmin 0.39 0.43 0.08 0.33 0.40 1.58
Wmax 0.58 0.76 0.16 0.36 0.49 1.94
DW 0.19 0.33 0.08 0.03 0.09 0.36

The subscripts correspond to thermodynamic state points in Fig. 1.

796 W.S. Loh et al. / Applied Thermal Engineering 29 (2009) 793–798

Table 2
COP and SCE for single and two-stages ideal adsorption cooling cycle, simulated at coolant temperature of Tevap = 6.7 °C and Tcond = 29.4 °C in accordance to ARI Standard 560

Adsorbent/adsorbate pairs Regeneration Temperature (°C)

45 50 55 60 65 70 75 80 85
Single-stage cycle COP (A-15)/ethanol – – 0.05 0.30 0.42 0.49 0.53 0.55 0.56
(A-20)/ethanol – – 0.08 0.42 0.56 0.62 0.65 0.67 0.68
Silica gel/water – – 0.02 0.52 0.62 0.66 0.68 0.68 0.68
Chemviron/R134a – – 0.01 0.02 0.03 0.04 0.05 0.06 0.06
Fluka/R134a – – 0.01 0.06 0.10 0.12 0.14 0.15 0.15
Maxsorb II/R134a – – 0.02 0.09 0.14 0.18 0.20 0.22 0.23
SCE (kJ/kg_ads) (A-15)/ethanol – – 2.92 29.0 63.7 99.7 135.9 171.2 205.0
(A-20)/ethanol – – 5.85 56.8 121.6 184.8 243.8 297.2 344.0
Silica gel/water – – 0.91 53.3 104.4 143.7 174.2 198.2 217.3
Chemviron/R134a – – 0.13 0.83 1.75 2.71 3.72 4.75 5.82
Fluka/R134a – – 0.56 3.28 6.50 9.53 12.4 15.1 17.7
Maxsorb II/R134a – – 1.83 11.2 23.0 34.7 46.4 57.9 69.3
Single-effect double-lift cycle COP (A-15)/ethanol 0.12 0.21 0.25 0.28 0.29 0.30 0.31 0.31 0.32
(A-20)/ethanol 0.21 0.32 0.35 0.36 0.36 0.37 0.37 0.38 0.38
Silica gel/water 0.26 0.34 0.36 0.36 0.36 0.36 0.36 0.36 0.36
Chemviron/R134a 0.01 0.02 0.03 0.03 0.03 0.04 0.04 0.04 0.04
Fluka/R134a 0.03 0.07 0.09 0.10 0.10 0.10 0.10 0.11 0.11
Maxsorb II/R134a 0.05 0.09 0.11 0.12 0.13 0.14 0.14 0.15 0.15
SCE (kJ/kg_ads) (A-15)/ethanol 5.64 17.7 32.2 48.6 66.2 84.5 102.9 120.9 138.1
(A-20)/ethanol 15.3 50.3 85.0 117.8 147.6 173.7 195.8 213.8 228.2
Silica gel/water 15.2 46.5 69.6 87.0 100.3 110.5 118.5 124.8 129.9
Chemviron/R134a 0.20 0.58 1.00 1.45 1.93 2.43 2.95 3.49 4.05
Fluka/R134a 0.94 2.84 4.61 6.28 7.85 9.33 10.7 12.1 13.3
Maxsorb II/R134a 3.16 8.99 15.0 21.0 27.2 33.3 39.4 45.4 51.3

pressurized adsorption cycles are operated for pressure between

300 and 800 kPa. These pressures are calculated from the saturated
pressure at the evaporator and condenser temperatures.
Table 1 shows the temperature data at pre-cooling, adsorption,
pre-heating and desorption processes for both the single-stage and
single-effect double-lift cycles of the six assorted adsorbent/adsor-
bate pairs. These temperatures are simulated at evaporator tem-
perature, T evap of 6.7 °C and condenser temperature, T cond of
29.4 °C in accordance to ARI Standard 560. The uptakes (W) of
the six working pairs are simulated from Eqs. (1)–(3). For the same
amount of uptake capacity (DW = Wmax–Wmin), single-effect dou-
ble-lift cycles require lower desorption or regeneration tempera-
tures, T g compare to those of the single-stage cycle. For instant,
Maxsorb II-R134a working pair at uptake capacity of 0.36 kg/kg,
the T g at single-stage and single-effect double-lift cycles are equal
to 80 °C and 67 °C, respectively. Among these working pairs, Max-
Fig. 2. Effect of regeneration temperature on specific cooling effect and coefficient
sorb II/R134a has the highest uptake capacity which is followed,
of performance at single-stage and single-effect double-lift adsorption cooling cycle
respectively by ACF (A-20)/ethanol, ACF (A-15)/ethanol, Fluka/ for ACF (A-20)/ethanol pair.
R134a, silica gel/water and Chemviron/R134a pairs.
Table 2 summarizes the effect of regeneration temperature on
the COP and SCE of single-stage and single-effect double-lift cycles. fect double-lift cycle for all six assorted working pairs are at least
From Table 2, it is evident that the SCE values increase linearly about 40% higher than that of the single-stage cycles at T g below
with the increase of regeneration temperature for all six assorted 60 °C. This is caused mainly due to the lower desorption in sin-
adsorbent/adsorbate pairs. For both single-stage and single-effect gle-stage cycles.
double-lift cycles, ACF (A-20)/ethanol provides the highest value The COP values of single-stage and single-stage double-lift cy-
for SCE which is followed by silica gel/water, ACF (A-15)/ethanol, cles increase abruptly with increasing T g up to 80 °C and 60 °C,
Maxsorb II/R134a, Fluka/R134a and lastly Chemviron/R134a. It is respectively. However, the improvement in COP values is only
worthy to mention that the sub-atmospheric systems always pos- marginal when T g is greater than 80 °C and 60 °C for single-stage
sess higher SCE and COP compared to those of pressurized systems. and single-stage double-lift cycle, respectively. This happens due
For adsorption pairs with activated carbon as the adsorbent i.e., to the fact that the requirement of heat input becomes significantly
ACF (A-15), ACF (A-20), Chemviron, Fluka and Maxsorb II, the sin- higher when temperature difference between heat source and heat
gle-effect double-lift cycles have higher SCE values compared to sink is larger.
those of the single-stage cycles when T g is below 70 °C. However, Contrary to SCE, the single-effect double-lift cycles have lower
for silica gel/water pair, the SCE of single-effect double-lift cycle COP values compared with those of the single-stage cycles. Since
is higher compared to that of the single-stage cycle when T g is low- in single-effect double-lift cycle comprises with four reactor beds,
er than 65 °C. Fig. 2 shows the SCE of ACF (A-20)/ethanol adsorp- the heat input requirement increases in this cycle as compared to
tion pair for single-stage and single-effect double-lift cycles. The that of the single-stage one resulting a lower COP as compared to
simulated curves for single-stage and single-effect double-lift cycle that of single-stage cycles. COP of single-stage cycle is more than
coincide when T g is about 67 °C. In addition, the SCE of single-ef- 50% compared to single-effect double-lift cycle at the same regen-
 W.S. Loh et al. / Applied Thermal Engineering 29 (2009) 793–798 797

Table 3
Cooling effect for single-stage and single-effect double-lift of ideal adsorption cooling, simulated at coolant temperature of T evap ¼ 6:7 °C and T cond ¼ 29:4 °C in accordance to ARI
Standard 560

Adsorbent/adsorbate pairs Regeneration temperature (°C)

45 50 55 60 65 70 75 80 85
CE (MJ/m3 of adsorbent) Single-stage cycle ACF (A-15)/ethanol – – 0.58 5.79 12.74 19.94 27.17 34.25 41.00
ACF (A-20)/ethanol – – 1.17 11.37 24.31 36.93 48.76 59.45 68.80
Silica gel/water – – 0.64 37.33 73.11 100.57 121.92 138.72 152.10
Chemviron/R134a – – 0.10 0.62 1.32 2.04 2.79 3.57 4.38
Fluka/R134a – – 0.22 1.28 2.54 3.72 4.84 5.90 6.91
Maxsorb II/R134a – – 0.52 3.14 6.45 9.76 13.04 16.28 19.46

Single-effect ACF (A-15)/ethanol 1.13 3.54 6.44 9.72 13.25 16.91 20.58 24.18 27.61
double-lift cycle ACF (A-20)/ethanol 3.06 10.06 17.00 23.56 29.52 34.74 39.15 42.77 45.64
Silica gel/water 10.67 32.54 48.74 60.91 70.18 77.33 82.93 87.37 90.92
Chemviron/R134a 0.15 0.44 0.75 1.09 1.45 1.83 2.22 2.63 3.04
Fluka/R134a 0.37 1.11 1.80 2.45 3.07 3.65 4.20 4.72 5.21
Maxsorb II/R134a 0.91 2.53 4.20 5.91 7.63 9.35 11.06 12.75 14.40

eration temperature. However, COP values of the single-effect dou- (b) Silica gel/water pair gives the highest value of cooling effect
ble-lift adsorption cycles are higher than those of the single-stage per unit volume of adsorbent at the specified packing
cycle at T g ranging from 45–55 °C, as shown in Table 2. This is be- densities.
cause at relatively lower regeneration temperatures, it is impossi- (c) Silica gel/water pair is found to be the most feasible adsor-
ble for a single-stage cooling cycle to operate. The minimum bent/adsorbate pair among the six assorted adsorbent/
desorption temperature for the six working pairs at single-stage refrigerant pairs. However, increasing the packing density
mode is about 55 °C [24], but for the single-effect double-lift cycle, of ACF (A-20) in the reactor beds may lead the ACF (A-20)/
it can operate at temperatures below 55 °C, which means single-ef- ethanol based adsorption cooling systems promising for
fect double-lift cycle is possible to operate at near ambient temper- cooling application.
ature, i.e., between 45 and 55 °C, which can be achieved easily from (d) Partial vacuum adsorption cooling cycles provide better per-
a solar hot water system or industrial waste heat. This shows the formances in term of COP and SCE as compared to those of
major advantage of single-effect double-lift cycle. If higher temper- the pressurized cycles at the same operating temperature
ature heat source is available, single-stage adsorption cycle tends conditions.
to provide relatively higher performance. The effect of T g on COP (e) The main advantage for single-effect double-lift adsorption
for ACF (A-20)/ethanol pair is also shown in Fig. 2. cycle is the ability to operate at relatively lower heat source
From the design point of view, the compactness of reactor beds temperature i.e., about 45–55 °C.
is essential to achieve high cooling load at smaller chiller footprint. (f) At lower regeneration temperature (less than 70 °C), the sin-
As adsorbents are compressible substances, the mass of adsorbent gle-effect double-lift cycle provides higher value of specific
per unit volume (m3) highly depends on the packing density, qeff of cooling effect.
the adsorbents. Packing density is defined as the mass of adsorbent (g) Single-stage cycles attain higher COP values than those of
that can be packed in a unit volume. Thus, cooling effect per unit the single-effect double-lift cycles at relatively higher heat
volume (CE) can be determined by multiplying the specific cooling source temperatures.
effect and the adsorbent packing density. The effects of regenera-
tion temperature on CE per unit volume for the respective adsor-
bent/adsorbate pairs at assorted packing density are illustrated in References
Table 3. It is noticed that the cooling effect for silica gel/water pair
having the highest value which is followed by ACF (A-20)/ethanol, [1] J.T.Houghton,Y.Ding,D.J.Griggs,M.Noguer,P.J.vanderLinden,X.Dai,K.Maskell,C.A.
Johnson, Climate Change 2001: The Scientific Basis, Published for the
ACF (A-15)/ethanol, Maxsorb II/R134a, Fluka/R134a and Chemvi- IntergovernmentalPanelonClimateChange,CambridgeUniversityPress,2001.
ron/R134a. In the present simulation, Chemviron/R134a working [2] B.B. Saha, S. Koyama, T. Kashiwagi, A. Akisawa, K.C. Ng, H.T. Chua, Waste heat
pair has the highest packing density of 752 kg/m3, followed by sil- driven dual-mode, multi-stage, multi-bed regenerative adsorption system,
International Journal of Refrigerant 26 (2003) 749–757.
ica gel/water pair with 700 kg/m3, Fluka/HFC-134a pair with [3] B.B. Saha, A. Akisawa, T. Kashiwagi, Solar/waste heat driven two-stage
391 kg/m3, Maxsorb II/HFC-134a pair with 281 kg/m3 and ACFs/ adsorption chiller: the prototype, Renewable Energy 23 (2001) 93–101.
ethanol with 200 kg/m3 [2,25]. [4] Y. Hamamoto, K.C.A. Alam, A. Akisawa, T. Kashiwagi, Performance evaluation
of a two-stage adsorption refrigeration cycle with different mass ratio,
International Journal of Refrigerant 28 (2005) 344–352.
[5] K.C. Ng, Thermodynamic tools for chiller diagnostics and optimization, Heat
5. Conclusions Transfer Engineering 25 (2004) 1–4.
[6] E. Passos, F. Meunier, J.C. Gianola, Thermodynamic performance improvement
Six specimens of adsorbents and refrigerant pairs, i.e., ACF (A- of an intermittent solar-powered refrigeration cycle using adsorption of
methanol on activated carbon, Journal of Heat Recovery System 6 (1986) 259–
15)/ethanol, ACF (A-20)/ethanol, silica gel/water, Chemviron/
264.
R134a, Fluka/R134a and Maxsorb II/R134a have been analyzed [7] R.Z. Wang, J.P. Jia, Y.H. Zhu, Y. Teng, J.Y. Wu, J. Cheng, Q.B. Wang, Study on a
for the performance based on an ideal thermally driven adsorption new solid absorption refrigeration pair: active carbon fiber–methanol pair,
Transactions of ASME 119 (1997) 214–218.
cooling cycle operating with single-stage single-lift and two-stage
[8] Q. Cui, G. Tao, H. Chen, X. Guo, H. Yao, Environmentally benign working pairs
double-lift cycles at assorted operating pressures. The salient for adsorption refrigeration, Energy 30 (2005) 261–271.
points of the present study are summarized as follows: [9] L.W. Wang, R.Z. Wang, R.G. Oliveira, A review on adsorption working pairs for
refrigeration, Renewable and Sustainable Energy Reviews (2008), doi:10.1016/
j.rser.2007.12.002.
(a) ACF (A-20)/ethanol pair provides the highest value of spe- [10] D.I. Tchernev, D.T. Emerson, High efficiency regenerative zeolite heat pump,
cific cooling effect. ASHREA Transaction 94 (2) (1988) 2024–2032.
798 W.S. Loh et al. / Applied Thermal Engineering 29 (2009) 793–798

[11] R.E. Critoph, R. Vogel, Possible adsorption pairs for use in solar cooling, [19] I.I. El-Sharkawy, B.B. Saha, S. Koyama, K. Srinivasan, Isosteric heats of
International Journal of Ambient Energy 7 (4) (1986) 183–190. adsorption extracted from experiments of ethanol and HFC-134a on carbon
[12] R.E. Critoph, Activated carbon adsorption cycles for refrigeration and heat based adsorbents, International Journal of Heat and Mass Transfer 50 (2007)
pumping, Carbon 27 (1989) 63–70. 902–907.
[13] B.B. Saha, E.C. Boelman, T. Kashiwagi, Computer simulation of a silica [20] M.Z.I. Khan, B.B. Saha, K.C.A. Alam, A. Akisawa, T. Kashiwagi, Study on solar/
gel–water adsorption refrigeration cycle-the influence of operating waste heat driven multi-bed adsorption chiller with mass recovery, Renewable
conditions on cooling output and COP, ASHRAE Transaction 101 (1995) Energy 32 (2007) 365–381.
348–357. [21] G. Cacciola, G. Restuccia, Reversible adsorption heat pump: a thermodynamic
[14] B.B. Saha, E.C. Boelman, T. Kashiwagi, Experimental investigation of a silica model, International Journal of Refrigeration 18 (1995) 100–106.
gel–water adsorption refrigeration cycle-the influence of operating conditions [22] K.H. Yeung, K. Sumathy, Thermodynamic analysis and optimization of a
on cooling output and COP, ASHRAE Transaction 101 (1995) 358–366. combined adsorption heating and cooling system, International Journal of
[15] B.B. Saha, K.C.A. Alam, A. Akisawa, T. Kashiwagi, K.C. Ng, H.T. Chua, Two-Stage Energy Research 27 (2003) 1299–1315.
Non-Regenerative Silica Gel–Water Adsorption Refrigeration Cycle, vol. 40, [23] L.W. Wang, R.Z. Wang, Z.S. Lu, C.J. Chen, K. Wang, J.Y. Wu, The performance of
ASME Advanced Energy Systems Division (ASME-AES), 2000. pp. 65–69. two adsorption ice making test units using activated carbon and a carbon
[16] I.I. El-Sharkawy, K. Kuwahara, B.B. Saha, S. Koyama, K.C. Ng, Experimental composite as adsorbents, Carbon 44 (2006) 2671–2680.
investigation of activated carbon fibers/ethanol pairs for adsorption cooling [24] B.B. Saha, I.I. El-Sharkawy, A. Chakraborty, S. Koyama, N.D. Banker, P. Dutta, M.
system application, Applied Thermal Engineering 26 (2006) 859–865. Prasad, K. Srinivasan, Evaluation of minimum desorption temperatures of
[17] N.D. Banker, K. Srinivasan, M. Prasad, Performance analysis of activated thermal compressors in adsorption refrigeration cycles, International Journal
carbon + HFC-134a adsorption coolers, Carbon 42 (2004) 117–127. of Refrigerant 29 (2006) 1175–1181.
[18] B.B. Saha, S. Koyama, K.C. Ng, Y. Hamamoto, A. Akisawa, T. Kashiwagi, Study on [25] B.S. Akkimaradi, M. Prasad, P. Dutta, K. Srinivasan, Effect of packing density
a dual-mode, multi-stage, multi-bed regenerative adsorption chiller, and adsorption parameters on the throughput of a thermal compressor,
Renewable Energy 31 (2006) 2076–2090. Carbon 40 (2002) 2855–2859.