Heat transfer engineering an international journal, tập 31, số 11, 2010

CopyrightTaylor and Francis Group, LLCISSN: 0145-7632 print / 1521-0537 online DOI: 10.1080/01457631003603816 e d i t o r i a l Application of Adsorption Technologies for Energy Efficien

Trang 2

CopyrightTaylor and Francis Group, LLC

ISSN: 0145-7632 print / 1521-0537 online

DOI: 10.1080/01457631003603816

e d i t o r i a l

Application of Adsorption

Technologies for Energy Efficiency

KIM CHOON NG1and BIDYUT BARAN SAHA2

1Mechanical Engineering Department, National University of Singapore, Singapore

2Mechanical Engineering Department, Kyushu University, Japan

Adsorption cycles are proven to be a practical and energy-efficient method for converting low-temperature waste heat into

useful effects such as cooling, refrigeration, and heating Such cycles have direct environmental benefits in terms of global

warming, emissions, etc This special issue covers the state-of-the art of adsorption research and technologies for relevant

applications with the objectives of energy efficiency and sustainability.

The efficient utilization of energy at low heat source

tem-perature is a key issue for both industry and academia In

re-cent years, the heat-activated adsorption cycles are found to

be most suited for converting low-temperature waste heat into

useful effects, and this is reflected by the abundance of

litera-ture publications such as the study of adsorbent characteristics

[1, 2], adsorption cooling [3, 4], adsorption heating [5, 6],

ad-sorption refrigeration [7, 8], adad-sorption hybrid process [9], and

adsorption process for automobile air conditioning applications

[10]; the numbers of journal publications in these fields during

the last three decades are shown in Figure 1 In our opinion,

the adsorption technology could be steered toward the key

de-velopment of heat-activated machines for the energy industry,

amounting to several billion dollars per annum, and these could

offer cost-effective solutions for achieving energy efficiency and

environment sustainability

In cooling, refrigeration and heat pumping applications, new

adsorption cycles have been developed This special issue of

Heat Transfer Engineering was commissioned to focus on the

recent developments of adsorption technologies for energy

effi-ciency and environmental sustainability Several experts in many

countries have been invited to address these adsorption

tech-nologies This special issue consists of six papers covering the

aspects of adsorption fundamentals, chemical reaction kinetics

Address correspondence to Professor Kim Choon Ng, Mechanical

Engi-neering Department, National University of Singapore, 9 EngiEngi-neering Drive 1,

Singapore 117576 E-mail: mpengkc@nus.edu.sg

of consolidated adsorbents, a solar-powered adsorption coolingsystem, and an adsorption system for automobile applications.The first three papers describe the adsorption fundamentals.The first paper presents the adsorption isotherms of HFC-134aand highly porous activated carbon (Maxsorb III) measuredusing the constant-volume–variable-pressure (CVVP) methodwith temperatures ranging from 293 to 338 K and vapor pres-sures up to 0.7 MPa The isotherm data are presented withthe Dubinin–Astakhov (DA) equation and correspondingly theheat of adsorption has also been determined, which is essen-tial in designing the pressurized-bed adsorption cooling system.The second paper examines the thermodynamic property sur-

faces for the adsorption of R507A, R134a, and n-butane onto

the pitch-based activated carbons The property field tion leads to the derivation of the entropy, enthalpy, internalenergy, and heat of adsorption as a function of pressure, tem-perature, and the amount of adsorbate or P-T-q state With suchgoverning equations, the entropy and enthalpy property mapsare employed for analysis of adsorption cooling cycles and gasstorage systems The comparison between the ideal and realadsorption cycles is highlighted using the thermodynamic ap-proach The third paper studied the effect of air on kinetics ofwater adsorption on three promising adsorbents, namely, SWS-1L (silica KSK modified by calcium chloride), silica gel oftype RD, and FAM-Z02 (zeolites-based functional adsorbentmaterial) For cooling applications, the experimental conditionsselected are similar to those of real adsorption chillers The ef-ficacy of adsorption cooling cycles is expressed in terms of the

formula-907

Trang 3

908 K C NG AND B B SAHA

Figure 1 Number of journal publications in assorted adsorption fields over

three periods (courtesy of Scopus search engine).

specific cooling power at assorted partial pressures of residual

air

The fourth paper deals with the chemical reaction

kinet-ics of consolidated composite adsorbent–ammonia working

pairs using an iso-volumetric test unit The reaction

kinet-ics, the heat transfer performances of the consolidated

adsor-bents, and the influences of the refrigerant condensation and

the activation energy are analyzed, while the thermal

conduc-tivity of the consolidated adsorbents is determined by a

tran-sient plane-source method The study is useful for the

selec-tion of composite adsorbents having good heat and mass

trans-fer characteristics for the adsorption cooling and heat pump

applications The fifth paper presents an experimental

solar-powered adsorption icemaker with double-stage mass recovery

cycle with a regeneration temperature below 75◦C and

work-ing at pressures just above atmospheric pressure The icemaker

is found to be suited for low-temperature heat sources with

continuous cold production The final paper presents the

de-sign and performance of an activated-carbon ammonia-based

compact sorption generator for an automobile air-conditioning

application It is reported that a pair of unit mass

sorp-tion generators could produce an average cooling rate of

1.6 kW when powered by radiator heat at a temperature of

90◦C

We thank all of the authors who have contributed to the

suc-cess of this special issue and the valuable comments of reviewers

that assisted in improving the quality of articles Lastly, we thank

cordially Dr Afshin J Ghajar, editor-in-chief of Heat

Trans-fer Engineering, for his invitation to commission the special

issue

REFERENCES

[1] Aristov, Y I., Restuccia, G., Cacciola, G., and Parmon, V N.,

A Family of New Working Materials for Solid Sorption Air

Conditioning Systems, Applied Thermal Engineering, vol 22,

Stage, Multi-Bed Regenerative Adsorption System,

Interna-tional Journal of Refrigeration, vol 26, no 7, pp 749–757,

2003

[4] Saha, B B., El-Sharkawy, I I., Koyama, S., Lee, J B., andKuwahara, K., Waste Heat Driven Multi-Bed Adsorption Chiller:Heat Exchangers Overall Thermal Conductance on Chiller Per-

formance, Heat Transfer Engineering, vol 27, no 5, pp 80–87,

2006

[5] Li, T X., Wang, R Z., Wang, L W., and Lu, Z S., mental Study on an Innovative Multifunction Heat Pipe Type

Experi-Heat Recovery Two-Stage Sorption Refrigeration System, Energy

Conversion and Management, vol 49, no 10, pp 2505–2512,

2008

[6] Akahira, A., Alam, K C A., Hamamoto, Y., Akisawa, A., andKashiwagi, T., Experimental Investigation of Mass Recovery Ad-

sorption Refrigeration Cycle, International Journal of

Refrigera-tion, vol 28, no 4, pp 565–572, 2005.

[7] Amar, N B., Sun, L M., and Meunier, F., Numerical sis of Adsorptive Temperature Wave Regenerative Heat Pump,

Analy-Applied Thermal Engineering, vol 16, no 5, pp 405–418,

1996

[8] Tchernev, D I., and Emerson, D T., High Efficiency Regenerative

Zeolite Heat Pump, ASHRAE Transactions, vol 94, no 2, pp.

2024–2032, 1988

[9] Dawoud, B., A Hybrid Solar-Assisted Adsorption Cooling Unit

for Vaccine Storage, Renewable Energy, vol 32, no 6, pp 947–

964, 2007

[10] Lu, Y Z., Wang, R Z., Jianzhou, S., Xu, Y X., and Wu,

J Y., Practical Experiments on an Adsorption Air tioner Powered by Exhausted Heat From a Diesel Locomo-

Condi-tive, Applied Thermal Engineering, vol 24, pp 1051–1059,

2004

Kim Choon Ng obtained his B.Sc (Hons.) and

Ph.D from Strathclyde University in Glasgow, UK,

in 1975 and 1980, respectively He worked briefly at the Babcock Power Ltd in Renfrew prior to joining

in 1981the Department of Mechanical Engineering

of the National University of Singapore, where he

is now a tenured professor His areas of research are two-phase flow, chiller testing and modeling, electro-adsorption chillers, and renewable energy To date, he has written more than 200 peer-reviewed journal and international conference articles, six patents, and co-authored a

book, Cool Thermodynamics, printed by CISP (UK) He is a member of the

IMechE (UK) and the Institution of Engineers Singapore, a chartered neer (UK) and a registered professional engineer (S), and an associate ed-

engi-itor of Heat Transfer Engineering and of Proceedings of the Institution of

Mechanical Engineers, Part E, Journal of Process Mechanical Engineering.

He also serves as an editorial board member for Advances in Mechanical

Engineering.

heat transfer engineering vol 31 no 11 2010

Trang 4

Bidyut Baran Saha obtained his B.Sc (Hons.) and

M.Sc degrees from Dhaka University of Bangladesh

in 1987 and 1990, respectively He received his

Ph.D in 1997 from the Tokyo University of

Agri-culture and Technology, Japan He worked as

a senior research fellow at the Mechanical

En-gineering Department of National University of

Singapore prior to joining the Mechanical

Engi-neering Department of Kyushu University, Japan,

in 2010 as a professor His main research interests are thermally powered tion systems, adsorption desalination, heat and mass transfer analysis, and energy efficiency assessment He has published more than 200 articles in peer- reviewed journals and international conference proceedings He has edited three books and holds seven patents He serves as an editorial board member of

sorp-Advances in Mechanical Engineering, Open Mechanical Engineering Journal

(OMEJ), and Open Thermodynamics Journal (OTherJ).

Trang 5

CopyrightC Taylor and Francis Group, LLC

DOI: 10.1080/01457631003603949

Adsorption Parameter and Heat

of Adsorption of Activated

Carbon/HFC-134a Pair

WAI SOONG LOH1, M KUMJA,1KAZI AFZALUR RAHMAN,1

KIM CHOON NG,1BIDYUT BARAN SAHA,2SHIGERU KOYAMA,3and

IBRAHIM I EL-SHARKAWY4

3Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Japan

4Mechanical Power Engineering Department, Mansoura University, El-Mansoura, Egypt

This article presents the adsorption isotherms of HFC-134a and activated carbon Maxsorb III measured using the

constant-volume–variable-pressure method The adsorption isotherms cover temperature ranges from 293 to 338 K and pressures up

to 0.7 MPa The trends of the experimental isotherms for activated carbon are found to be identical in all cases with previous

studies except that the vapor uptake is slightly higher The adsorption characteristic of the Dubinin–Ashtakov equation has

been regressed from the experimental isotherms data and the maximum specific uptake is 2.15 kg of adsorbate adsorbed

per kilogram of activated carbon The heat of adsorption, which is concentration and temperature dependent, has also been

extracted from the experiments.

INTRODUCTION

Basic research on the adsorption characteristics of adsorbent–

adsorbate pairs, such as HFC-134a and activated carbon

(Max-sorb III), is motivated by the needs to switch to ozone-friendly

working fluids, particularly for the heat-driven type of

refriger-ation cycles Such cycles could help in reducing the emissions

of greenhouse gases, which is in line with the recommendations

outlined in the Kyoto Protocol In recent years, there has been

a substantial increase in interest in using adsorbent–adsorbate

refrigeration cycles, as the activation temperatures are relatively

lower, enabling waste heat recovery to be incorporated in plant

designs

Adsorption data, for the range of operational conditions of

chillers, are unavailable from the manufacturers of adsorbents

In general, only the data on surface area and pore volumes of the

adsorbent are provided, and they are insufficient for engineers

and designers to size the components of heat driven chillers

For example, new adsorbents such as activated carbons may

Address correspondence to Professor Kim Choon Ng, Mechanical

Engi-neering Department, National University of Singapore, 9 EngiEngi-neering Drive 1,

Singapore 117576 E-mail: mpengkc@nus.edu.sg

have characteristics vastly different from those of conventionaltypes The nexus of an adsorption cycle design is the determi-nation of adsorption isotherms of adsorbent–adsorbate pair aswell as the isosteric heat of adsorption, where extensive studiesmay be needed at various combinations of adsorbent–adsorbatepairs For example, Pons and Guilleminot [1] studied activatedcarbon/methanol system for ice production by using renewableenergy source, while the applications of an AC135/methanolpair on refrigeration systems were investigated by Douss andMeunier [2] Miles and Shelton [3] tested experimentally a two-bed system using activated carbon/ammonia as the adsorbent–adsorbate pair, and Wang et al [4] studied the adsorption charac-teristics of ACF/methanol and activated carbon (AC)/methanolpairs Vasiliev et al [5] developed a solar-gas solid sorp-tion ACFs/ammonia-based heat pump Preliminary experimen-tal data for hydrogen, helium, neon, and nitrogen on activatedcharcoal have been presented by Chan et al [6] Lin and Lin [7]studied the adsorption characteristics of HFC-134a on two types

of activated carbon, namely, GAC and EAC (which are in ule and pellet form, respectively) for use in refrigeration andHFC recovery Riffat et al [8] determined the adsorption char-acteristics of refrigerants R407a, R407b, and HFC-32 with AX-

gran-21 Akkimaradi et al [9] studied the adsorption isotherms for

910

Trang 6

Figure 1 SEM picture of the activated carbon specimen Maxsorb III (200,000

magnification).

HFC-134a on activated carbons Chemviron, Fluka, and

Maxsorb II

The present article reports on the ongoing effort toward

gen-erating experimental adsorption isotherm data for samples of

activated carbon Adsorption isotherms of specimen Maxsorb

III over a temperature range of 293–338 K and for pressures up

to 0.7 MPa are measured using the

constant-volume–variable-pressure (CVVP) method The Dubinin–Ashtakov (DA)

equa-tion has been employed to evaluate the adsorpequa-tion isotherm

characteristics In addition, the isosteric heat of adsorption is

extracted from the experimental isotherms data Based upon the

adsorption parameters, the performance analysis of an idealized

adsorption cooling cycle is presented

Figure 2 Schematic arrangement of adsorption isotherm experimental setup.

EXPERIMENTAL SECTION Materials

High-purity grade R134a samples, supplied by Kansai CokeCompany, are used in the studies described in this article Allproperties of R134a are evaluated using the generalized equation

of state proposed by Tillner-Roth and Baehr [10] The form activated carbon of the type Maxsorb III is used in theexperiments for the determination of adsorption characteristicswith the refrigerant HFC-134a A detailed photo of the pitch-type adsorbent, i.e., the Maxsorb III, is captured by the scanningelectron microscope (SEM), as shown in Figure 1, while the BET(Brunauer–Emmet–Teller) surface area and micropore volumeare respectively equal to 3.06× 106m2kg−1and 17.0× 10−4

powder-m3 kg−1, respectively The skeletal density of the sample isabout 2200 kg m−3[11]

Apparatus and Procedure

The experimental apparatus consists mainly of a stainlesssteel (SS 304) adsorption cell and a charging cell with internalvolumes of 59.69 ml and 1026.15 ml, respectively Maxsorb IIIactivated carbon, 2.9283± 0.0001 g, is charged into the adsorp-tion cell, simulating the bed design of an adsorption coolingsystem, and this low amount of adsorbent gives monolayer be-havior during vapor uptake or regeneration

Figure 2 shows a schematic diagram of the experimentalapparatus, where the dosing and charging cells are connectedthrough a capillary tube The cells are immersed in a constant-temperature water bath (HAAKE F8-C35) and are controlled

to preselected temperatures of 5 to 95◦C with an accuracy of

±0.01◦C The pressure readings are measured using a 0–1 MPa

Trang 7

912 W S LOH ET AL.

range KYOWA pressure transducer (PGS-10KA) with an

uncer-tainty of 0.1% of full scale in measurement A redundant

sec-ondary indicator of a 0–1.6 MPa range Bourdon pressure gauge

is also used A class-A type Pt 100- resistance temperature

detector (RTD)±0.1 K is used for temperature measurement

At the adsorption cell, the RTD is in contact with the activated

carbon to enable direct temperature measurement All

temper-ature and pressure readings are monitored by an Agilent data

logger A thermostat-controlled tape heater is wrapped around

the connection tubes to prevent condensation Due to the

con-tinuous circulation of the bath fluid, and the adequate time for

thermal stabilization, it is assumed that no temperature gradient

would occur within the cells

Prior to the start of measurement, the entire assembly is

evac-uated for 24 h using a BOC Edwards direct drive vane vacuum

pump to a vacuum level of 0.05 mbar During the evacuation,

the adsorption cell was maintained at 120–130◦C to desorb any

residue gas in the cell Helium gas is injected into the system

during desorption to improve the evacuation After evacuation,

the charging cell is pressurized with R134a from its source (with

ball valve 2 closed), and the initial pressure and temperature are

then recorded The R134a vapor is released into the adsorption

cell and left to reach an equilibrium state Subsequently, the

temperature of the bath is changed and the next data point along

an isochore is obtained The measurements are at 20, 25, 45, and

65◦C The same procedures are repeated with a different initial

quantity of gas being charged into the charging cell For a low

initial charge, it is possible to bring a system pressure within the

cell to below 1 bar The upper pressure was limited to 0.7 MPa

in the experiment This precaution was necessary to avoid the

possibility of condensation of R134a in the capillary

RESULTS AND DISCUSSION

Data Reduction

The quantitative amount of adsorbed adsorbate is determined

for the generation of isotherms of the single-component

adsor-bate and adsorbent system (R134a–Maxsorb III) For a given

amount of adsorbate contained in a system, the temperature,

volume, and pressure are dependent variables and are related by

the general form:

f (P , V , T )= 0 (1)

or simply known as the equation of state (EOS) With a closed

but adsorbent filled vessel, i.e., constant volume, the variable

V can be treated as a constant (thus ignored) and the modified

equation of state of the adsorbent–adsorbate pair would

incor-porate the concentration of vapor uptake that resides within the

pores of adsorbent, i.e.,

f (P , T , C)= 0 (2)

At an initial stage of the experiment, adsorbate (R134a vapor) isintroduced into the charging cell of a known volume In the ab-sence of the adsorbent, the initial mass of adsorbate is calculatedfrom the following equation:

where ρab i is the initial adsorbate density at charging

temper-ature, T charging and V charging cellis the charging cell volume.When the dosing and charging cells are connected, adsorption

occurs in the pores of the adsorbent and the void volume V void

in the adsorption cell is given by

vol-a chvol-arge of 2.9283 g of Mvol-axsorb III (mevol-asured by vol-a trac Max 5000 moisture analyzer with resolution of 0.1 mg) ispacked into the cell of 60.0 cm3, the void volume of adsorptioncell is equal to 53.4 cm3 This volume is inclusive of the vaporspace of the adsorption cell, and the adsorbate mass is calculatedusing the generalized equation of state proposed by Tillner-Roth

Compu-and Baehr [10] at respective adsorption temperatures, T ads, andpressures, i.e.,

m void= ρab (P , T ads ) V void at T ads (5)where ρab is the density of gas (adsorbate) in the adsorptioncell The remaining amount of adsorbate at each isotherm iscalculated from

m f = ρab f (P , T ads ) V total (6)

where V total = V charging cell + V void, ρab f is the adsorbate sity at the respective isotherm temperature The total mass ofthe control volume at initial and final stage are respectively

den-m total,i (P , T , C) = m ab,i (P , T ) + m ads,i (P , T , C)

+ m ac,i (P , T ) (7)

m total,f (P , T , C) = m ab,f (P , T ) + m ads,f (P , T , C)

+ m ac,f (P , T ) (8)From conservation of mass,

m ab,i (P , T ) + m ads,i (P , T , C) + m ac,i (P , T )

= m ab,f (P , T ) + m ads,f (P , T , C) + m ac,f (P , T ) (9) Since the mass of activated carbon, m acis constant, and there is

no adsorption at the initial stage, the preceding equation reducesto

m ab,i (P , T ) = m ab,f (P , T ) + m ads,f (P , T , C) (10)heat transfer engineering vol 31 no 11 2010

Trang 8

Hence, the amount of refrigerant adsorbed, m ads, can then be

estimated from

m ads (P , T , C) = m i (P , T ) − m f (P , T ) (11)

which is dependent on all three pressure, temperature, and

spe-cific uptake factors

Finally, the specific uptake value or the loading, C, is

deter-mined as

C = m ads

Adsorption Parameters Using DA Equation

The Dubinin–Astakhov (DA) equation [12] is found to

pro-vide the best representation of the adsorption data In the

anal-ysis, the DA equation (13) is used to evaluate the

temperature-independent characteristic curve:

ln(W ) = ln(W o)−

A E

n

(13)

where A is the adsorption potential and W is the adsorbed

vol-ume, W o is the limiting volume of adsorption space of the

ad-sorbent, E is the characteristic energy of the adsorption system,

and n is the structural heterogeneity parameter The adsorption

potential is the work done in the isothermal compression of 1

kg of vapor from the equilibrium pressure, P , to the saturation

vapor pressure, P s, and is given by

A = RT ln

P s P

(14)Similarly, Eq (13) can be expressed as:

ln

P s P

n

(15)where the logarithmic form of Eq (14) is:

ln (W ) = ln (W o)−

RT E

ln

P s P

n

(16)Also in the current study, the experimental isotherms are cross-

checked through the following methods:

i) With no adsorbed volume correction, whereby Eq (15)

re-duced to

ln (C) = ln (C o)−

RT E

ln

P s P

n

(17)ii) With adsorbed phase volume correction, where the adsorbed

volume W in Eq (15) is expressed as:

where v ab is the specific volume of the adsorbed phase of the

adsorbate As the volume of adsorbed phase cannot be

mea-sured directly, it is considered equivalent to the corresponding

liquid specific volume Above the saturation temperature, ferent approximations need to be used The method suggested

dif-by Dubinin [12] was the most appropriate one to represent theadsorption data for HFC-134a Consequently,

v ab = v boil exp [α (T − T boil)] (19)

in the adsorbed state and at its normal boiling point, respectively;

T c is the critical temperature of the adsorbate, and b is the van der Waals volume For HFC-134a, T boil = 246.78 K, T c = 374.21

K, v boil = 7.2643 × 10−4 m3kg−1, and b = 9.39 × 10−4 m3

kg−1 The numerical values for the adsorption parameters can

be regressed from the experimental data; i.e., W o , E, and n are

equal to 0.001711 m3 kg−1, 79.7 kJ kg−1, and 1.37, tively Figure 3 shows the uptake of Maxsorb III–HFC 134aplotted against the pressure–concentration–temperature plane.The shapes of the isotherms obtained from the current exper-imental are similar in all cases and comparable with previousstudy [9], which is shown in Figure 4 The uptake values of HFC134a onto Maxsorb III/R134a are generally higher than for theMaxsorb II–HFC-134a system This spread of specific uptakecan be ascribed to the variability in the surface area and poredistribution of the adsorbent It is therefore important for design-ers to understand the adsorption characteristics of the specimenthey anticipate using This is because it is rarely possible toextrapolate the isotherm characteristics are from one specimen

respec-to another

DA equations with and without volume correction are fittedonto the experimental data in Figure 5 The DA equation withvolume correction has higher uptake value compared to the DAequation without volume correction at lower pressures and tem-peratures, and vice versa From Figure 5, both the DA equationshave good approximation onto the experimental uptake values.Figure 6 makes a comparison of both the isotherm equations,

Figure 3 Isotherms of Maxsorb III/HFC-134a at respective temperature.

Trang 9

914 W S LOH ET AL.

Figure 4 Comparison of isotherm data for Maxsorb III/HFC-134a and

Max-sorb II/HFC-134a systems at 293 K.

Figure 5 Isotherm data for Maxsorb III/HFC-134a for experimental data

points ( ); for DA equation without volume correction (—); and for DA equation

with volume correction (- - -).

Figure 6 Comparison of uptake deviations between calculated and

experi-mental uptake: (i) at 20 ◦C,•without volume correction,◦with volume

cor-rection; (ii) at 25 ◦C, without volume correction,with volume correction;

(iii) at 45 ◦C, without volume correction, with volume correction; and (iv)

at 65 ◦C, without volume correction, with volume correction.

wherein the error between experimental and calculated uptakebetween fitted (i.e., both with and without volume correction)equations are plotted From Figure 6, the difference betweenexperimental data and fitted results fall within±0.13 kg kg−1.

Heat of Adsorption

Heat of adsorption, which is a function of concentration,has a weak dependence on temperature [9, 13] The Clausius–Clapeyron equation is commonly used to estimate heat of ad-sorption at constant concentration, as

H ads=−R∂ ln P

∂ (1/T ) (21)

H ads denotes the isosteric heat of adsorption (kJ kg−1), R is

the gas constant (kJ kg−1 K−1), P is the equilibrium pressure (kPa), and T is the adsorbent temperature (K) Critoph [14]

proposed a relation that is integrated from Eq (20) to estimatethe isosteric heat of adsorption

In the current study, the heat of adsorption, H adsas a tion of vapour uptake is calculated from the measured adsorptionisotherm by a correlation proposed by Chakraborty et al [15],i.e.,

dT (P , T ) (22)

where the first term of the right-hand side indicates the ventional form of the isosteric heat of adsorption derived fromthe Clausius–Clayperon equation and the second term definesthe behavior of adsorbed mass with respect to both the pressureand the temperature changes during an adsorbate uptake, whichoccur due to the nonideality of the gaseous phase Using the DAequation, Eq (21) can be written as:

con-H ads = h f g + E ln

C o C

1/n

+ T v g dP

dT (P , T ) (23)

where ν g is the specific volume of the vapor phase, and dP /dT

represents the gradient of the pressure with the temperature ofthe adsorbate The isosteric heat of adsorption for R134a onMaxsorb III as a function of surface loading and temperature isshown in Figure 7

From Figure 7, one observes that H ads decreases with creasing vapor uptake Since Maxsorb III is highly porous withmicropores of different widths, HFC-134a is adsorbed rapidlyonto the sites of high energy During adsorption, moleculesare adsorbed onto sites of decreasing energy The HFC-134amolecules first penetrate into narrower pores, resulting in astronger interaction between the adsorbate and the adsorbent

in-This implies a higher value of H ads at lower loadings ter completely filling the smaller pores, HFC-134a moleculesare accommodated gradually in larger pores, in which the ad-sorption affinity becomes weaker A monotonic decrease in the

Af-H adsas a function of loading is therefore observed

Trang 10

Figure 7 Heat of adsorption using corrected correlation (Eq 24) at: (i) 20 ◦C,

, (ii) 25 ◦C,♦, (iii) 45 ◦C, , and (iv) 65◦,.

CONCLUSIONS

In this study, the experimental isotherms have been derived

for the activated carbon Maxsorb III/HFC-134a pair, which is

useful for the design of adsorption chiller cycles We have found

that the Dubinin–Ashtakov equation is most suitable to represent

the isotherms both with and without volume correction, and the

maximum uptake is found to be around 2.15 kg kg−1 Similarly,

heat of adsorption is evaluated from the proposed transformation

(Eq (22)) and it varies from 180 to 420 kJ kg−1, depending

on the adsorbate loading or uptake Such basic experimental

isotherms and isosteric heat of adsorption of Maxsorb III/R134a

system can be used in the design and analysis of adsorption

processes in cooling applications

h fg specific enthalpy difference between saturated

vapor and saturated liquid, kJ kg−1

m ac mass of activated carbon, kg

m ad mass of vapor adsorbed, kg

m i initial mass of adsorbate, kg

m f final mass of adsorbate, kg

m void mass of adsorbate in void volume, kg

n exponential parameter describes isotherm,

V void void volume, m3

V charging cell charging cell volume, m3

V total total internal volume of experimental setup, m3

W adsorbed volume, m3kg−1

Greek Symbols

ρ density, kg m−3

ρab density of adsorbate in void volume, kg m−3

ρab i initial density of adsorbate in charging cell, kg

electrical resistance, ohm

∂ partial derivative operator

[1] Pons, M., and Guilleminot, J J., Design of an Experimental

Solar Powered, Solid Adsorption Ice Maker, Journal of Solar

Engineering-Transactions of the ASME, vol 103, no 4, pp 332–

337, 1986

[2] Douss, N., and Meunier, F., Effect of Operating Temperatures

on the Coefficient of Performance of Active Carbon–Methanol

Systems, Heat Recovery Systems and CHP, vol 8, no 5, pp.

149–169, 1988

[3] Miles, D J., and Shelton, S V., Design and Testing of a

Solid-Sorption Heat-Pump System, Applied Thermal

Engineer-ing, vol 16, no 5, pp 389–394, 1996.

[4] Wang, R Z., Jia, J P, Zhu, Y H, Teng, Y., Wu, J Y., Cheng, J.,and Wang, Q B., Study on a New Solid Absorption Refrigeration

Pair: Active Carbon Fiber–Ethanol Pair, Transactions of ASME,

vol 119, pp 214–218, 1997

Trang 11

916 W S LOH ET AL.

[5] Vasiliev, L L., Mishkinis, D A., Antukh, A A., and Vasiliev, L

L., Jr., Solar–Gas Solid Sorption Heat Pump, Applied Thermal

Engineering, vol 21, no 5, pp 573–583, 2001.

[6] Chan, C K., Tward, E., and Boudale, K I., Adsorption Isotherms

and Heat of Adsorption of Hydrogen, Helium, Neon and Nitrogen

on Activated Charcoal, Cryogenics, vol 24, pp 451–459, 1984.

[7] Lin, S H., and Lin, R C., Prediction and Experimental

Verifi-cation of HFC-134a Adsorption By Activated Carbons, Journal

of Environmental Science Health, vol A34, no 1, pp 183–200,

1999

[8] Riffat, S B., Williams, M D., and Corr, S., Adsorption Heat

Pump Using HFC Refrigerants, International Journal of Energy

Res vol 21, pp 481–495, 1997.

[9] Akkimaradi, B S., Prasad, M., Dutta, P and Srinivasan, K.,

Adsorption of 1,1,1,2-Tetrafluoroethane on Activated Charcoal,

Journal of Chemical Engineering Data, vol 46, pp 417–422,

2001

[10] Tillner-Roth, R., and Baehr, H D., An International

Stan-dard Formulation for the Thermodynamic Properties of

1,1,1,2-Tetrafluoroethane (HCF-134a) for Temperatures From 170 K to

455 K and Pressures up to 70 MPa, Journal of Physical Chemical

Reference Data, vol 23, no 5, pp 657–729, 1994.

[11] Saha, B B., Koyama, S., El-Sharkawy, I I., Habib, K.,

Srinivasan, K., and Dutta, P., Evaluation of Adsorption

Parame-ters and Heats of Adsorption Through Desorption Measurements,

Journal of Chem Eng Data, vol 52, pp 2419–2424, 2007.

[12] Dubinin, M M., Progress in Membrane and Surface Science, ed.

D A Cadenhead, Academic Press, New York, vol 9, pp 1–70,

1975

[13] Prasad, M., Akkimaradi, B S., Rastogi, S C., Rao, R R., and

Srinivasan, K., Heat of Adsorption for Charcoal–Nitrogen

Sys-tems, Carbon, vol 37, pp 1641–1642, 1999.

[14] Critoph, R E., Performance Limitations of Adsorption Cycles for

Solar Cooling, Solar Energy, vol 41, no.1, pp 21–31, 1988.

[15] Chakraborty, A., Saha, B B., Koyama, S., and Ng, K C., On the

Thermodynamic Modeling of the Isosteric Heat of Adsorption

and Comparison With Experiments, Appl Phys Lett vol 89, p.

171901, 2006

Wai Soong Loh is a Ph.D candidate, part-time

stu-dent at the Department of Mechanical Engineering, National University of Singapore Currently he works

as a research engineer at the Centre for Offshore Research and Engineering (CORE), National Uni- versity of Singapore He received his M.Sc degree

in mechanical engineering in 2006 from the tional University of Singapore and his B.Eng degree in mechanical and production engineering from the Nanyang Technological University, Singapore in

Na-2003 His areas of research are testing and modeling of adsorption chillers and

natural gas storage system on activated carbon using sorption method.

M Kumja is a Ph.D candidate, part-time student,

in the Mechanical Engineering Department at the National University of Singapore He received his master’s degree at the Mandalay Technological Uni- versity, Myanmar, in 2002 Currently he works as a research fellow at the Building Department, National University of Singapore.

Kazi Afzalur Rahman is a Ph.D student in the

De-partment of Mechanical Engineering, National versity of Singapore He received his B.Sc degree

Uni-in mechanical engUni-ineerUni-ing from Bangladesh sity of Engineering and Technology, Dhaka, in 2005.

Univer-He is currently working on adsorption of gases with activated carbons.

Kim Choon Ng obtained the B.Sc (Hons.) and Ph.D.

from Strathclyde University in Glasgow (UK) in 1975 and 1980, respectively He worked briefly at the Bab- cock Power Ltd., in Renfrew prior to joining the De- partment of Mechanical Engineering of the National University of Singapore in 1981, and he is now a tenured full professor His areas of research are two- phase flows, chiller testing and modeling, electro- adsorption chillers, adsorption desalination, and renewable energy systems He has written more than

90 peer-reviewed journals, holds six patents, and co-authored a book, entitled,

Cool Thermodynamics, printed by CISP (UK) in 2000 He is a member of the

IMechE (UK) and the Institution of Engineer Singapore, a chartered engineer (UK), a registered professional engineer (S), and is associate editor for two international journals.

in 1987 and 1990, respectively He received his Ph.D.

in 1997 from the Tokyo University of Agriculture and Technology, Japan He worked as a senior research fellow at the Mechanical Engineering Department of National University of Singapore prior to joining the Mechanical Engineering Department of Kyushu Uni- versity, Japan, in 2010 as a professor His main research interests are thermally powered sorption systems, adsorption desalination, heat and mass transfer analysis, and energy efficiency assessment He has published more than 200 articles in peer-reviewed journals and international conference proceedings He has edited three books

and holds seven patents He serves as an editorial board member of Advances in

Mechanical Engineering, Open Mechanical Engineering Journal (OMEJ), and Open Thermodynamics Journal (OTherJ).

Shigeru Koyama is a professor and head of the

Department of Energy and Environmental ing, Interdisciplinary Graduate School of Engineer- ing Sciences, Kyushu University, Japan He is now the vice-president of Commission B1 of the Inter- national Institute of Refrigeration (IIR) He received his Ph.D in 1980 from Kyushu University, Japan His main research interests are vapor compression systems, sorption systems, compact heat exchanger design, and heat and mass transfer analysis He has published more than 200 articles in well-recognized journals, books, and pro-

Engineer-ceedings He is an editor of Thermal Science and Engineering, Japan.

Ibrahim I El-Sharkawy received his B.Sc (Hons)

and M.Sc degrees in mechanical power engineering from Mansoura University, Egypt He obtained his Ph.D from Kyushu University, Japan, in 2006 He

is a faculty member at Mansoura University rently, he is joining Kyushu University as a JSPS (Japan Society for the Promotion of Science) post- doctoral research fellow His main research interests are thermally powered adsorption cooling/heat pump systems and heat and mass transfer analysis He has published more than 40 articles in peer-reviewed journals and proceedings He has won four conference paper awards in 2004, 2005, 2006, and 2007.

Cur-heat transfer engineering vol 31 no 11 2010

Trang 12

DOI: 10.1080/01457631003604152

Thermodynamic Property Surfaces

for Adsorption of R507A, R134a,

Carbonaceous Porous Materials

ANUTOSH CHAKRABORTY,1BIDYUT BARAN SAHA,2KIM CHOON NG,1

IBRAHIM I EL-SHARKAWY,3and SHIGERU KOYAMA4

3Mechanical Power Engineering Department, Mansoura University, El-Mansoura, Egypt

4Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, Japan

The thermodynamic property surfaces of R507A, R134a, and n-butane on pitch-based carbonaceous porous material

(Maxsorb III) are developed from rigorous classical thermodynamics and experimentally measured adsorption isotherm

data These property fields enable us to compute the entropy, enthalpy, internal energy, and heat of adsorption as a function

of pressure, temperature, and the amount of adsorbate The entropy and enthalpy maps are necessary for the analysis of

adsorption cooling cycle and gas storage We have shown here that it is possible to plot an adsorption cooling cycle on the

temperature-entropy (T–s) and enthalpy-uptake (h–x) maps.

INTRODUCTION

When gas molecules come into contact with a porous

adsor-bent surface, the gas molecules are captured by the surfaces’

pores in a nonideal manner Some molecules are retained by

the van der Waals force field, while the uncaptured molecules

may bounce off from the surface The efficacy of an

adsor-bent+ adsorbate system, especially in the context of

single-component adsorption, is determined inter alia by the

parame-ters of adsorbed-phase thermodynamics, namely, the adsorption

isotherms, thermodynamic property surfaces of energy and

en-tropy, heat of adsorption, specific heat capacity, and adsorption

kinetics [1] For example, the knowledge of thermodynamic

properties of the adsorbed phase is important and necessary for

the adsorption process to be analyzed, which has wide

appli-cations in the fields of gas separation, purification, adsorption

The authors thank King Abdullah University of Science & Technology

(KAUST) for the generous financial support through the project (WBS

R265-000-286-597).

Address correspondence to Dr Bidyut Baran Saha, Mechanical Engineering

Department, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395,

Japan E-mail: saha@mech.kyushu-u.ac.jp

chillers, cryocoolers, and energy storage systems At any stance, the thermodynamic property of the adsorbed phase is

in-described by the pressure (P ), temperature (T ), and amount

of adsorbate (x), as well as the types of adsorption behavior,

and early theoretical models for physical adsorption systemsare attributed largely to the works of Hill [2] and Everett [3] Amore general approach to adsorption with an exchange of heatand work has been developed by Guggenheim [4], Myers [5],and Myers and Monson [6], who developed the thermodynamicfunctions such as Gibbs free energy, enthalpy, and entropy onthe basis of the isothermal condition

As the porous properties of adsorbent and the adsorptioncharacteristics of the adsorbent–refrigerant pair have significantinfluence on the operating behaviors of adsorption processes,

in this article we study the microporous surfaces of some mon adsorbent such as pitch-based activated carbon (AC) with

com-refrigerants such as R507A, R134a, and n-butane AC has a

much higher surface area compared to that of zeolite or silicagel and occupies relatively higher adsorption capacity for thementioned refrigerants Pitch-based AC (type Maxsorb III) hashigher surface area and good thermal conductivity [7] and issuitable for adsorbate storage and cooling applications On theother hand, refrigerant (R) 507A is a near azeotropic mixture

917

Trang 13

918 A CHAKRABORTY ET AL.

of equal mass fractions of pentafluoroethane (R-125) and

1,1,1-trifluoroethane (R-143A), and it is being used as refrigerant for

low- and medium-temperature refrigeration The nonpolar

hy-drocarbon butane consists of four carbons and 10 hydrogens and

is generally classified as an alkane, or paraffin, where hydrogen

saturates the carbon atoms via covalent single bonds

Using the framework of the derived thermodynamic property

fields of adsorbent–adsorbate surfaces [8–10] and the

experi-mentally measured adsorption isotherms data [7, 11, 12], we

are motivated (i) to calculate and plot the temperature entropy

(T–s) and pressure enthalpy (P–h) maps of R507A, R134a, and

n-butane on Maxsorb III as a function of pressure,

tempera-ture, and the amount of adsorbate, (ii) to describe the adsorption

cooling cycle on a T–s map, and (iii) to evaluate accurately the

isosteric heat of adsorption for these three different systems

THERMODYNAMIC PROPERTY SURFACES OF

PHYSICAL ADSORPTION

In physical adsorption, where an adsorbed gas is taken as

the combination of adsorbed gas plus adsorbent enclosed by a

surface, the thermodynamic properties such as the entropy (s),

enthalpy (h), and internal energy (u) are described in terms of

the measured variables, namely, P , T , and x.The effects of the

accompanying isosteric heat of adsorption (Q st) and the specific

heat capacity of adsorbed phase (c p,a) have to be accounted for,

for accurate determination of s, h, and u Being path

indepen-dent, the change of extensive thermodynamic quantity can be

tracked by integration from:

1 An initial reference pressure P oto nonequilibrium pressure

P at constant T and x, followed by an initial reference

tem-perature, T o , to temperature, T , at zero adsorbate (x = 0),

and finally from zero amount of adsorbate to any adsorbate

uptake, x, at constant T and P

2 Or first an initial reference amount of adsorbate uptake (here

x = 0) to an adsorbate uptake, x, at constant T and P , then

from an initial reference temperature, T o, to temperature,

T , at constant P and x, and finally from zero adsorbate to

adsorbate mass, x, at constant T and P

The integration of the thermodynamic properties can also be

depicted schematically in Figure 1 In the following sections, we

approach the properties of adsorbed phase by first considering

the entropy as a function of P , T , and x The Gibbs law, dh=

T ds +vdP , is then invoked to calculate enthalpy and the internal

energy, u, is obtained with the definition of h, i.e., u = h – Pv.

Entropy

We define the quantity of entropy s (J/mol-K) as the

summation of solid-phase and adsorbed-phase entropies, i.e.,

Path 1

Path 2 Path 3

tively Expressed in terms of the process paths, the entropy of theadsorbent–adsorbate system can be expanded mathematically as

(1)where the cumulative changes are integrated along the paths of

(i) constant pressure P and the amount of adsorbate uptake x, (ii) constant T and x, and finally (iii) constant P and T For

simplification of Eq (1), we introduce the properties of a pureand incompressible solid adsorbent,

where c p,s is the specific heat capacity of the solid adsorbent

In the adsorbed phase,

c p,a = c p,g + Q st

1

v g

dv g dT

(2)heat transfer engineering vol 31 no 11 2010

Trang 14

where Q st indicates the isosteric heat of adsorption [14, 15], v

is the specific volume, and the subscript g indicates the gaseous

phase

Neglecting the change of solid-phase entropy with respect

to pressure and invoking the Maxwell relations for the pressure

Also, the change of entropy with respect to the amount of

adsorbate uptake (or integration) gives

Invoking the relationships between the total and partial

derivatives as well as introducing Maxwell relationships, the

entropy change with respect to the mass of adsorbate is given

Substituting the simplifications just mentioned, the extensive

entropy quantity is reduced to

T2 − 1

T v g

dv g dT

Applying the Gibbs free energy equation, dh = T ds + vdP ,

the thermodynamic property surface of enthalpy for

adsorbent-adsorbate system becomes

v g

dv g dT

By definition, the internal energy (u = h – Pv) of the

adsorbent–adsorbate system is given by

v g

dv g dT

The microporous adsorbent Maxsorb III manufactured byKansai Coke and Chemicals Co Ltd., Osaka, Japan, is used inthe present study Maxsorb III is a highly microporous materialand its pore size distribution is wide, ranging from 0.4 to 3 nm.Therefore, Maxsorb III is considered to be a heterogeneous ad-sorbent For the adsorption on carbon-based heterogeneous ad-sorbents, generally the Dubinin–Astakhov (DA) isotherm equa-tion [16] is found suitable and hence the DA equation is used

to correlate the experimentally measured isotherm data for the

adsorption of R507A, R134a, and n-butane on Maxsorb III The

DA equation can be written as

Trang 15

Table 1 Isotherm parameters for the adsorption of R507A, R134a, and

n-butane on Maxsorb III

System x m(kg/kg) E(J/mol) n References

Maxsorb III + R507A 1.400 5740 1.470 [12]

Maxsorb III + R134a 2.168 7489 1.225 [11]

Maxsorb III+n-butane 0.800 17436 1.050 [7]

and a single-component adsorbed adsorbate is taken as an

ad-sorbent + adsorbate (or adsorbate + adsorbent) system, and

thermodynamic equilibrium prevails between this system and

the single-component unadsorbed gas phase We use

experi-mentally measured isotherm data to show the present findings

Entropy Maps

The entropy is an increasing function of pressure,

tempera-ture, and uptakes The T ưs maps for the adsorption

characteris-tics of (i) R507A+ Maxsorb III, (ii) R134a + Maxsorb III, and

(iii) n-butane+ Maxsorb III systems are shown in Figures 2, 3,

and 4, where a–b–c–d–a represents the adsorption cooling cycle

in terms of entropy as a function of P , T , and x During

regen-eration phase (lines a–b–c), the pressure in the adsorber rises

from P evap to P condby heating the adsorber, and desorption of

re-frigerant from the adsorbent (Maxsorb III) occurs by connecting

the adsorber with the condenser The amount of adsorbed uptake

falls from x ads to x des , and the entropy rises from s ads to s des

On the other hand, during the adsorption phase (lines c–d–a),

the adsorber is cooled and the pressure falls from P cond to P evap

Then the refrigerant vapor is adsorbed on the adsorbent and the

amount adsorbed increases up to x ads It is also found from the

present analysis that the entropy flow, s ( = s ads ư s des), of the

Maxsorb III+n-butane system is higher than that of the other

two systems (Maxsorb III+ R507A and Maxsorb III + R134a)

Figure 2 Temperature–entropy (T ưs) diagram for Maxsorb III and R507A

system Here the subscript ads indicates adsorption, des indicates desorption,

cond represents the condenser, and evap defines the evaporator.

0 10 20 30 40 50 60 70 80 90

2 2.1

(T ads) (T des)

This means that a relatively higher energy is required to drive

the adsorption cooling system for n-butane as adsorbate and

Maxsorb III as adsorbent

Energy Maps

The enthalpy and internal energy of a single component sorbate+ adsorbent system are an increasing function of pres-

ad-sure and uptake The xưh and xưu maps of Maxsorb III +

R507A for three different temperatures are shown in Figure 5.The enthalpy and internal energy increase linearly with temper-ature On the other hand, a graph of enthalpy against uptake

of Maxsorb III+ R134a is plotted in Figure 6, where one can

observe that h increases when the pressure is increased A plot

of enthalpy versus uptakes for Maxsorb III+n-butane is shown

in Figure 7, where the processes of adsorption (C–D–A) and

20 30 40 50 60 70 80 90

Trang 16

0 10 20 30 40 50 60 70

desorption (A–B–C) as a function of P , T , x, and h are

ob-served It should be noted here that the enthalpy approximately

approaches a constant value when the filling of R507A, R134a,

and n-butane adsorbates on Maxsorb III becomes weaker This

indicates that Maxsorb III is highly microporous with

heteroge-neous surfaces Such curves also represent the nature of

adsor-bents

Isosteric Heat of Adsorption

The isosteric heat of adsorption (Q st) for R507A, R134a, and

n-butane adsorbates on Maxsorb III as a function of the amount

of adsorbate uptake are shown in Figure 8 From Figure 8, one

may observe that Q st decreases with increasing uptake, and Q st

is found to be very high at lower loading or Henry’s region

compared to Q st at higher uptakes The Maxsorb III consists

mainly of micropores with different widths; at first R507A or

R134a or n-butane adsorbs rapidly onto sites of high energy,

and molecules then adsorb onto sites of decreasing energy as

adsorption progresses The adsorbate molecules first penetrate

1 bar

0.5 bar 0.75 bar

Maxsorb III + R134a

Figure 6 Enthalpy versus concentration for Maxsorb III + R134a as a

func-tion of pressure and temperature.

30 35 40 45 50 55 60

ing from P evap to P cond, (ii) B–C defines the desorption process, (iii) C–D

indicates switching from P cond to P evap, and (iv) D–A is the adsorption cess.

pro-0 10000 20000 30000 40000 50000 60000

T = 303 K

Figure 8 Isosteric heat of adsorption of R507A, R134a, and n-butane on Maxsorb III as a function of surface coverage x/x mat 303 K.

into narrower pores and provide stronger interaction between the

adsorbate and the adsorbent This implies a higher value of Q st

at lower loadings After completely filling the smaller pores,

n-butane molecules are gradually accommodated in larger pores,

in which the adsorption affinity becomes weaker

CONCLUSIONS

The thermodynamic property surfaces, namely, entropy, ternal energy, enthalpy, and isosteric heat of adsorption, of asingle-component adsorbent+ adsorbate system are the basicfoundations of any adsorbate–adsorbent system Such key ther-modynamic quantities are useful in the design and analysis of(i) solid–gas sorption in the cooling sector such as the adsorp-tion cooling cycle [17, 18], and (ii) gas storage [19] It has beenshown that Maxsorb III+ R507 and Maxsorb III + R134a sys-tems are suitable for cooling applications, whereas Maxsorb III

in-is appropriate for storing n-butane at room temperature The

Trang 17

T − s and h − x diagrams for any adsorbent–adsorbate system

are useful to calculate the performances of adsorption cooling

cycle

NOMENCLATURE

c pa specific heat capacity of adsorbed phase, J/mol K

c pg specific heat capacity of gaseous phase, J/mol K

c ps specific heat capacity of solid adsorbent, J/mol K

u internal energy, J/mol

v a specific volume of adsorbed phase, m3/kg

v g specific volume of gaseous phase, m3/kg

x amount of adsorbate uptake, kg/kg

x m limiting amount of adsorbate uptake, kg/kg

[2] Hill, T L., Statistical Mechanics of Adsorption V

Thermody-namics and Heat of Adsorption, Journal of Chemical Physics,

vol 17, no 6, pp 520–535, 1949

[3] Everett, D H., The Thermodynamics of Adsorption: Part II—

Thermodynamics of Monolayers on Solids, Transactions of the

Faraday Society, vol 46, pp 943–957, 1950.

[4] Guggenheim, E A., The Thermodynamics of Interfaces in

Sys-tems of Several Components, Transactions of the Faraday Society,

vol 35, pp 397–411, 1940

[5] Myers, A L., Thermodynamics of Adsorption in Porous

Materi-als, AIChE Journal, vol 48, no 1, pp 145–160, 2002.

[6] Myers, A L., and Monson, P A., Adsorption in Porous Materials

at High Pressure: Theory and Experiment, Langmuir, vol 18, no.

26, pp 10261–10273, 2002

[7] Saha, B B., Chakraborty, A., Koyama, S., Yoon, S H., Kumja,M., Yap, C R., and Ng, K C., Isotherms and Thermodynamics for

the Adsorption of n-Butane on Maxsorb III, International Journal

of Heat and Mass Transfer, vol 51, pp 1582–1589, 2008.

[8] Chakraborty, A., Thermoelectric Cooling Device:

Thermody-namic Modeling and Their Applications in Adsorption Cooling Cycle, Ph.D Thesis, National University of Singapore, Singa-

pore, November 2005

[9] Steele, W A., The Interaction of Gases With Solid Surfaces, vol.

3, Pergamon Press, New York, 1974

[10] Chakraborty, A., Saha, B B., Ng, K C., Koyama, S., and vasan, K., Theoretical Insight of Physical Adsorption for a Single-Component Adsorbent+ Adsorbate System: I Thermodynamic

Srini-Property Surfaces, Langmuir, vol 25, no 4, pp 2204–2211, 2009.

[11] Akkimaradi, B S., Prasad, M., Dutta, P., and Srinivasan, K.,Adsorption of 1,1,1,2-Tetrafluoroethane on Activated Charcoal,

Journal of Chemical and Engineering Data, vol 46, pp 417–

422, 2002

[12] Saha, B B., El-Sharkawy, I I., Habib, K., Koyama, S., and vasan, K., Adsorption of Equal Mass Fraction Near AzeotropicMixture of Pentafluoroethane and 1,1,1-Trifluoroethane on Acti-

Srini-vated Carbon, Journal of Chemical and Engineering Data, vol.

53, pp 1872–1876, 2008

[13] Chakraborty, A., Saha, B B., Koyama, S., and Ng, K C., cific Heat Capacity of a Single Component Adsorbent–Adsorbate

Spe-System, Applied Physics Letters, vol 90, 171902, 2007.

[14] Chakraborty, A., Saha, B B., Koyama, S., and Ng, K C., On theThermodynamic Modeling of the Isosteric Heat of Absorption

and Comparison With Experiments, Applied Physics Letters, vol.

[16] Ruthven, D M., Principles of Adsorption and Adsorption

Pro-cesses, John Wiley and Sons, New York, 1984.

[17] Ng, K C., Sai, M A., Chakraborty, A., Saha, B B., and Koyama,S., The Electro-Adsorption Chiller: Performance Rating of a

Novel Miniaturized Cooling Cycle for Electronics Cooling, ASME

Journal of Heat Transfer, vol 128, no 9, pp 889–896, 2006.

[18] Saha, B B., Chakraborty, A., Koyama, S., Srinivasan, K., Ng, K.C., Kashiwagi, T., and Dutta, P., Thermodynamic Formalism ofMinimum Heat Source Temperature for Driving Advanced Ad-

sorption Cooling Device, Applied Physics Letters, vol 91, 111902

(1–3), 2007

[19] Saha, B B., Koyama, S., El-Sharkawy, I I., Habib, K., vasan, K., and Dutta, P., Evaluation of Adsorption Parameters and

Srini-Heat of Adsorption Through Desorption Measurements,

Jour-nal of Chemical & Engineering Data, vol 52, pp 2419–2424,

2007

Anutosh Chakraborty received his B.Sc Eng from

BUET, Bangladesh, in 1997 He obtained his M.Eng and Ph.D degrees from the National University of Singapore in 2001 and 2005, respectively At present,

he is working as a research fellow (A) at the partment of Mechanical Engineering, National Uni- versity of Singapore His research interests focus

De-on micro-/nanoscale transport phenomena, thin-film thermoelectric devices, adsorption thermodynamics, adsorption cooling, gas storage, and desalination He

Trang 18

has published 65 articles in peer-reviewed journals and international conference

proceedings and holds five patents.

in 1987 and 1990, respectively He received his Ph.D.

in 1997 from the Tokyo University of Agriculture and Technology, Japan He worked as a senior research fellow at the Mechanical Engineering Department of National University of Singapore prior to joining the Mechanical Engineering Department of Kyushu Uni- versity, Japan, in 2010 as a professor His main research interests are thermally powered sorption systems, adsorption desalination, heat and mass transfer analysis, and energy ef-

ficiency assessment He has published more than 200 articles in peer-reviewed

journals and international conference proceedings He has edited three books

and holds seven patents He serves as an editorial board member of Advances in

Mechanical Engineering, Open Mechanical Engineering Journal (OMEJ), and

Open Thermodynamics Journal (OTherJ).

Kim Choon Ng obtained the B.Sc (Hons.) and Ph.D.

from Strathclyde University in Glasgow (UK) in 1975 and 1980, respectively He worked briefly at Babcock Power Ltd., in Renfrew, prior to joining the Depart- ment of Mechanical Engineering of the National Uni- versity of Singapore in 1981, and he is now a tenured full professor His areas of research are two-phase flow, chiller testing and modeling, electro-adsorption chillers, adsorption desalination, and renewable energy systems He has written more than 90 articles

for peer-reviewed journals, holds six patents, and co-authored a book, Cool

Thermodynamics, printed by CISP (UK) in 2000 He is a member of the IMechE

(UK) and the Institution of Engineer Singapore, a chartered engineer (UK), a registered professional engineer (S), and associate editor for two international journals.

Ibrahim I El-Sharkawy received his B.Sc (Hons.)

and M.Sc degrees in mechanical power engineering from Mansoura University, Egypt He obtained his Ph.D from Kyushu University, Japan, in 2006 He

is a faculty member at Mansoura University rently, he is joining Kyushu University as a JSPS (Japan Society for the Promotion of Science) post- doctoral research fellow His main research interests are thermally powered adsorption cooling/heat pump systems and heat and mass transfer analysis.

Cur-He has published more than 40 articles in peer-reviewed journals and ings He has won four conferences paper awards, in 2004, 2005, 2006, and 2007.

proceed-Shigeru Koyama is a professor and head of the

Department of Energy and Environmental ing, Interdisciplinary Graduate School of Engineer- ing Sciences, Kyushu University, Japan He is now the vice-president of Commission B1 of the Inter- national Institute of Refrigeration (IIR) He received his Ph.D in 1980 from Kyushu University, Japan His main research interests are vapor compression systems, sorption systems, compact heat exchanger design, and heat and mass transfer analysis He has published more than 200 articles in well-recognized journals, books, and pro-

Engineer-ceedings He is an editor of Thermal Science and Engineering, Japan.

Trang 19

CopyrightC Taylor and Francis Group, LLC

DOI: 10.1080/01457631003604335

Effect of Residual Gas on Water

Adsorption Dynamics Under Typical

Conditions of an Adsorption Chiller

IVAN GLAZNEV,1DANIIL OVOSHCHNIKOV,2and YURIY ARISTOV1

1Boreskov Institute of Catalysis, Novosibirsk, Russia

2Novosibirsk State University, Novosibirsk, Russia

In this article the effect of a non-adsorbable gas (air) on kinetics of water adsorption on loose grains was studied for three

adsorbents promising for adsorption chilling: SWS-1L (silica KSK modified by calcium chloride), silica Fuji type RD, and

FAM-Z02 The experimental conditions were fixed similar to real operating conditions during an isobaric adsorption stage

of the basic cycle of an adsorption chiller (AC) Reduction of the adsorption rate was revealed even at a partial pressure of

residual air as low as 0.06 mbar Dependence of the characteristic adsorption time on air partial pressure was found to be

linear for partial pressures greater than 0.4 mbar, with the slope depending on the adsorbent nature Desorption stage was

less affected by the residual air Specific power released in an AC evaporator during the adsorption process was estimated

as a function of the partial pressure of residual air, and recommendations how to improve cooling performance are made.

INTRODUCTION

Water vapor adsorption is an important part of any

adsorp-tion unit for transformaadsorp-tion of low-temperature heat, such as

heat pumps, chillers, ice makers, etc Adsorption of pure vapor

on a single adsorbent grain is generally controlled by surface

or/and intraparticle heat and mass transfer resistance The

wa-ter adsorption from binary mixtures produces an extra heat and

mass transfer resistances from the gas-phase side, which may

become very large if one of the components is a non-adsorbable

gas In practice, such a gas can be present inside evacuated

equipment: air due to leakage or desorption from the adsorbent

bed (layer), hydrogen due to corrosion, etc The slowing down

of the adsorption process is due to the residual gas sweeping to

the surface, where it accumulates as a “gas-rich” layer [1, 2] as

shown in Figure 1 The transfer of vapor to the surface may then

become controlled by the process of diffusion through this layer,

which is relatively slow compared with the adsorption process

controlled by surface and intraparticle resistances

The authors thank the Russian Foundation for Basic Researches (projects

08-08-00808 and 09-08-90405) for partial financial support, and Dr M Ito for

providing us with a Fuji silica type RD.

Address correspondence to Professor Yuriy Aristov, Boreskov

Insti-tute of Catalysis, Pr Lavrentieva, 5, Novosibirsk, 630090, Russia E-mail:

aristov@catalysis.ru

The influence of a non-adsorbable gas on the rate of wateradsorption has not been purely studied so far In reference [3] theauthors revealed the reduction of heat transfer in a granulatedadsorbent layer in the presence of residual air It was supposedthat this effect can significantly reduce the performance of anadsorption heat pump (AHP) In reference [4] it was stated that

“as little as 1–2% non-condensing gases such as air can ‘poison’two-phase heat transfer, reducing evaporator and condenser heatrate by 50%,” although no justification was presented

The aim of this article is detailed experimental investigation

of the water adsorption dynamics during an isobaric adsorptionstage of an AHP cycle, which is the most sensitive to the pres-ence of non-adsorbable gas We used a new dynamic approachthat was proposed in reference [5] It allows obtaining the uptakecurve for an adsorbent located on a metal holder subjected to afast temperature drop that initiates the water adsorption much

as in adsorption heat pumps This approach was first applied tostudy the effect of residual air on dynamics of water adsorption

on loose grains of SWS-1L (mesoporous silica KSK modified

by calcium chloride) [6] in reference [7] Here we perform acomparative study of this effect on three well-known adsorbentsthat are considered promising for adsorption chilling, namely,SWS-1L [8], silica Fuji type RD [9–11], and FAM-Z02[12] Thethermophysical properties of these adsorbents are presented inTable 1

924

Trang 20

Figure 1 Radial distribution of the air and vapor concentration (a) and their

temporal evolution (b) on the grain surface (r= R grain ).

SWS-1L is a composite adsorbent that consists of a

meso-porous silica gel of KSK type and calcium chloride impregnated

inside its pores (33.7 wt%) Its water sorption capacity is as large

as 0.6 g/g [6, 13] This adsorbent has recently been suggested

to increase the water sorption capacity of common mesoporous

silica for application in an adsorption heat pump [13] and was

successfully tested in a prototype of an adsorption chiller [8,

14]

The microporous silica gel type RD manufactured by the Fuji

Davison Ltd was indicated as the adsorbent of water used in

commercialized adsorption chillers that utilize low-temperature

heat (50–80◦C) [15–17] Experimental equilibrium data for

wa-ter adsorption on this silica were reported [15, 18] and

math-ematical equations for approximation of these data were

pre-sented [15, 19]

FAM-Z02 has recently been suggested by the Mitsubishi

Chemicals Ltd as a promising adsorbent to be used for

adsorp-tion air condiadsorp-tioning [12], and now is under intensive testing in

several laboratories

EXPERIMENTAL APPARATUS, PROCEDURE,

AND MATERIALS

Apparatus

The basic diagram of the experimental setup is shown in

Figure 2 It contains three main compartments: the measuring

Table 1 Thermophysical properties of SWS-1L, silica gel type RD, and

FAM-Z02.

Property SWS-1L Fuji RD FAM-Z02

Specific surface area (m2/g) 230 820 450

Porous volume (cm3/g) 0.6 0.4 0.2

Average pore diameter (A) 150 22 100

Adsorption capacity (weight %)

Figure 2 Schematic diagram of the kinetic experimental setup.

cell (volume VMC = (0.14 ± 0.01) × 10−3 m3), the vaporvessel (VVV = (30.5 ± 0.6) × 10−3 m3), and the evaporatorwith liquid water, as shown in Figure 2 Loose adsorbent grainswere placed on an isothermal surface of the metal holder Itstemperature can be adjusted with an accuracy of±0.1◦C using

a heat carrier circuit coupled by a three-way valve (3WV) toeither circulating thermal bath 1 or 2

The temperature of the constant volume vapor vessel as well

as all connecting pipelines was maintained at 60 ± 0.5◦C by

using the air bath oven Water vapor was generated by the orator with a cooling jacket The evaporator temperature wasmanaged by circulating thermal bath 2 through the valve VEV.This temperature unambiguously set the vapor pressure, whichwas measured by an absolute pressure transducer MKS Baratrontype 626A (accuracy±0.01 mbar)

evap-Three-way valves were installed into the heat carrier loop

to leap the metal plate temperature By rotating the 3WV tothe reverse position, the heat carrier flows from bath 1 andbath 2 could be switched, thus initiating either a desorption oradsorption process

The Tested Sorbent

The tested adsorbent SWS-1L was prepared by the dry pregnation of a silica KSK with a saturated aqueous solution

im-of CaCl2 at T= 25◦C as described in reference [6] Grain size

was between 0.8 and 0.9 mm For SWS-1L and Fuji RD, biggergrains of 1.4–1.6 mm size were also tested

Experimental Procedure

The dry sorbent was heated to the starting temperature of the

isobaric adsorption stage T0 = 60◦C and evacuated up to 10−2

mbar for 2 h using a vacuum pump (valves VMC, V2, and V3areopened, VAand V1are closed) Then the vapor vessel was filledwith water vapor by its connecting to the evaporator and the

starting pressure for the adsorption process PEV(T = 10◦C)=12.4 mbar was set (V1 was open, V2 and VMC were closed).heat transfer engineering vol 31 no 11 2010

Trang 21

926 I GLAZNEV ET AL.

Then, the measuring cell was connected with the vapor vessel

(VMCand V3were open, V1and V2were closed) and the sample

was equilibrated with water vapor for 2 h The air was let inside

the setup as a non-adsorbable gas through a fine control needle

valve VA The valves VMC and V1were closed, so that the air

additive could be uniquely identified by the pressure increment

P Before the start of the kinetic experiments, an hour was

allowed for equilibrium conditions to set in

To initiate the adsorption process the metal holder was cooled

down to T f = 35◦C by turning the 3WV (VMC was open).

This resulted in reducing the vapor pressure over the adsorbent,

which did not exceed 2.0 to 2.5 mbar, so that the process can

be considered as quasi-isobaric Small reduction of the vapor

pressure during the adsorption stage can be the case for real

AHPs with undersized evaporators, which can not maintain the

vapor pressure constant After reaching the adsorption

equilib-rium, the reverse temperature leap was performed to initiate the

desorption run

Data on the pressure evolution P (t) required for calculating

the water uptake m H 2O (t) were recorded each 1 s by a data

acquisition system The uptake was calculated as [5]:

m H 2O = M · P (VMC+ VVV)

A detailed analysis for the accumulated error in

estimat-ing the absolute water loadestimat-ing showed a maximum value of

±10−3kg/kg that leads to the accuracy of the differential water

loading equal to±1.5%

Typical evolution of the temperature of the metal plate, the

vapor pressure over the sample, and the water uptake are shown

in Figure 3 After initiating the holder cooling, its

tempera-ture reached the final temperatempera-ture Tf within approximately

1 min (Figure 3a), while the pressure reduction was slower

and completed within about 10 min (Figure 3b) As a first

ap-proximation, the adsorption took place at constant temperature

of the holder equal to its final temperature

Pure Water Vapor Adsorption

The kinetic curves of pure water adsorption are presented

in Figure 4 for the SWS-1L, Fuji silica RD, and FAM-Z02

(the pressure of residual air P A ≤ 10−2 mbar) As mentioned

in the introduction, these adsorbents have different pore

struc-tures (pore size and pore volume) and different mechanisms

of intrinsic adsorption In spite of this, the absolute rate of

adsorption is identical in the initial region (at low uptakes

w < 0.05 g/g for loose grains 0.8–0.9 mm) This may

indi-cate a significant contribution of the heat transfer between the

grain surface and the metal plate to the total adsorption rate

Indeed, at t = 0 the driving force for the vapor transport is

Figure 3 Temperature evolution of the metal holder (a), the pressure of pure water vapor over the SWS-1L sample (b), and the water uptake (c) for cooling (1) and heating (2) runs for grains 0.8–0.9 mm size.

zero, while the driving force for the heat transport is maximum,because of the maximum temperature difference between thegrains and the metal support This driving force gives a maincontribution to the adsorption dynamics at low uptakes, and islikely to result in similar adsorption rates at short adsorptiontimes

Final uptakes for SWS-1L and FAM-Z02 were close, whilefor Fuji RD it was approximately two times smaller As a resultthis silica reached equilibrium in a much shorter time of about

300 s (Figure 4) For SWS it took approximately 900 s because of

a tail at long time, although for reaching 80% of equilibrium take less than 300 s was sufficient For this adsorbent it is worth-while to stop the adsorption process at the dimensionless wateruptake 0.8 For FAM-Z02 the shape of kinetics curve is differentfrom that of the SWS and Fuji RD because of a very long tail

up-at long time and the uptake w > 0.12 g/g (Figure 4) This

indi-cates a slow process with a characteristic time600 s This may

be attributed to complex structural transformation or chemicalreaction between water and this adsorbent Just to have a prelim-inary idea about the reversibility of this process at large conver-sions, we performed a set of runs of “adsorption/desorption”initiated by periodical temperature drops and jumps be-tween 60 and 35◦C as shown in Figure 5 The durationheat transfer engineering vol 31 no 11 2010

Trang 22

Figure 4 The kinetics curves of water adsorption initiated by temperature

drop 60 to 35 ◦C for SWS-1L (◦), Fuji type RD ( ) and FAM-Z02 ().

P 0 (H 2 O) = 10.3 mbar, grains 0.8–0.9 mm size.

of adsorption and desorption stages was fixed at 12 and 8

min A reduction of adsorption capacity by 6% after 9

ad-sorption/desorption runs was revealed that might be attributed

to partial irreversibility of the mentioned long time process

(Figure 5) More detailed study is required to learn this

ca-pacity loss as well as the nature of the slow transformation at

large conversions

Figure 5 The pressure P H2O (top) change by adsorption/desorption process

initiated by periodical temperature drops and jumps 60 to 35 ◦C (bottom) of metal

plate with the FAM-Z02 grains 0.8–0.9 mm size Adsorption stage continued

12 min, desorption 8 min.

Figure 6 The inhibition of adsorption rate for SWS-1L loose grains 0.8–0.9

mm (1) and 1.4–1.6 mm (2) in the presence of different partial pressure of air (in order PA= 0, 0.4, 1, and 4.7 mbar from left to right) Symbols represent experiment; lines show approximation by the function mt= m 0 + (mf– m 0 )[1

– exp(–t/τexp )].

Adsorption in the Presence of Residual Air

We found that the presence of air can visibly decrease therate of adsorption For SWS-1L the reduction of the adsorptionrate was revealed even at a partial pressure of residual air PA

as low as 0.06 mbar According to [1] and [2], the reduction ofthe adsorption rate can be caused by the Stephan flux, whicheffectively sweeps air to the grain surface where it accumulates

as a gas-rich layer (Figure 1) The effect was especially

sig-nificant at the intermediate dimensionless uptakes 0.3 < w < 0.9, while at w < 0.2 and w > 0.95 no influence of residual

air on the shape of kinetic curves was revealed (Figure 6) This

was probably because at w < 0.2 the time for forming the

air-rich layer was not sufficient and its influence was negligible At

w >0.95 the adsorption rate reduced to such a low value that

it was not sufficient to generate a sufficiently large Stephan fluxnecessary to form and stabilize the air-rich layer (Figure 1) At

PA ≥ 0.4 mbar the kinetic curves were near-exponential overthe whole range of uptake so that mt = m0+ (mf – m0)[1 –

exp(-t/τexp)] The characteristic time of adsorption τ 0.8,

corre-sponding to 80% of the final uptake, increased as τ 0.8 = τ0+

KP A, where K= 1200 ± 85 s/mbar

The rate of adsorption by the loose grains of Fuji silica RDwas less sensitive to the presence of residual air than for SWS-1L (Figure 7) This happened probably because a lesser amount

of air was swept to the grain external surface because the finaluptake was two times smaller than that for SWS-1L As a result,the adsorption time was much shorter, and the layer was less

thick The characteristic time τ 0.8 linearly increased with therise in PAwith slope K= 250 ± 6 s/mbar

For FAM-Z02 grains the fast process was slowed down by

residual air (Figure 8), and near-linear dependence τ 0.8 (P A) wasobserved with K= 590 ± 45 s/mbar (Figure 9) The slow pro-

cess was independent of P A This possibly indicated that itsnature is not linked with the vapor diffusion but is most likelyassociated with a complex structural transformation of theheat transfer engineering vol 31 no 11 2010

Trang 23

Figure 7 The inhibition of adsorption rate for Fuji RD loose grains 0.8–0.9

mm (1) and 1.4–1.6 mm (2) in the presence of different partial pressure of air (in

order PA= 0, 0.4, 1, 4.7, and 14.3 mbar from left to right) Symbols represent

experiment; lines show approximation by the function mt= m 0 + (mf– m 0 )[1

– exp(–t/τexp )].

adsorbent or chemical reaction between water and the

alu-minophosphate phase

At P A ≥ 1.0 mbar the kinetic curves are coincident for the

grains of 0.8–0.9 and 1.4–1.6 mm size for SWS-1L (Figure 6)

The Fuji RD grains have the same tendency and the kinetic

curves are very close for 0.8–0.9 and 1.4–1.6 mm at P A >4.7

mbar (Figure 7) Hence, the adsorption process is likely to be

controlled by the external mass transfer resistance caused by the

slow diffusion of water vapor through the air-rich layer

accumu-lated near the external surface of the grains This nonstationary

layer can also affect the heat transfer from the external grain

sur-face Moreover, air can also accumulate inside the grain pores

and affect the adsorption dynamics Indeed, the residual air more

strongly affects adsorption on the mesoporous SWS than on the

microporous FAM and silica RD (Figure 7) If this is so, the

nature of this influence is not clear, because the water transport

inside the grain can be described by the Knudsen diffusion [11,

20], so that water molecules do not collide either with each other

Figure 8 The inhibition of adsorption rate for FAM-Z02 grains 0.8–0.9 mm

in the presence of air (in order PA= 0, 0.4, 1, and 4.7 mbar from left to right).

Figure 9 Characteristic adsorption time τ 0.8as a function of air pressure for SWS-1L (•1.4–1.6 mm and◦0.8–0.9 mm), silica Fuji ( 1.4–1.6 mm and

− 0.8–0.9 mm) and FAM-Z02 (− 0.8–0.9 mm) Dotted line is the linear

approximation τ [s]= A + KP [mbar].

or with air molecules Thus, the process involved is very plex and combined effects of manifold impacts on the adsorptionrate should be a subject of detailed mathematical modeling andexperimental study

com-The desorption stage was less affected by the residual air forall the three adsorbents because the air-rich layer did not form

on this case The desorption rate was limited by the intraparticlevapor diffusion or the dissipation of adsorption heat from thegrain external surface

Average Specific Cooling Power

The average specific power W (t) consumed in the evaporator

during the water adsorption stage is a key parameter of AC as itdefines the necessary amount of adsorbent and size of the unit.Measuring the characteristic adsorption times allows estimation

of the cooling power (Table 2) For the simple configuration

of one layer of loose grains the specific cooling power reached1.8 kW/kg at the beginning of the adsorption process, which

is of high practical interest This power significantly decreased

in the presence of residual air and could reduce down to 0.Probably, in real AC units the effect of the residual air will beless strong than that in the experiments performed because ofthe restricted number of air molecules in the dead volume ofthe unit Nevertheless, careful degassing of an adsorbent bed aswell as water in an evaporator (condenser) is strictly obligatorybefore starting up an AC operation

Table 2 The specific cooling power (W/kg) corresponding to the

characteristic adsorption time τ 0.8for loose grains 0.8–0.9 mm

Trang 24

Detailed measurement of uptake curves under typical

condi-tions of AC was carried out for loose grains of SWS-1L, silica

Fuji RD, and FAM-Z02 The rate of water adsorption during the

isobaric stage of AC cycle and the cooling power generated in

the evaporator were very sensitive to the presence of residual

air Linear dependence of the characteristic adsorption time on

the pressure of residual air was observed at P A >0.4 mbar with

the slope dependent on the adsorbent nature The initial

cool-ing power reached 1.8 kW/kg and significantly decreased in the

presence of residual air, so that careful air removal before startup

is highly recommended Similar precautions should be taken in

adsorption units for vacuum drying of thermo-liable materials

M molar weight (0.018 kg/mol)

m weight of adsorbed water (kg)

mads weight of dry adsorbent (kg)

P pressure (mbar)

R gas constant (8.31 kJ/kg K−1), grain radius (m)

r distance to center of grain (m)

0 initial value (at t= 0 or at PA= 0 mbar)

0.8 value corresponding to 80% of the final uptake

plate metal holder

t instantaneous value at time t

VV vapor vessel

REFERENCES

[1] Nusselt, W Z., Surface Condensation of Water Vapor, Z Ver.

Deut Ing vol 60, pp 541–546, 1916.

[2] Frank-Kamenetskiy, D A., Diffusion and Heat Transfer in

Chem-ical Kinetics, Nauka, Moscow, 1967.

[3] Heifets, L I., Predtechenskaya, D M., Pavlov, Y V., and Okunev,

B N., Modeling of the Dynamic Effects in the Adsorbent Beds 1.Simple Method of Estimation of Thermal Conductivity of theComposite Adsorbent Bed (CaCl2 Impregnated Into Pores of

Silica Gel Lattice), Vestnik MGU Ser2, vol 47, no 4, pp 274–277,

2006

[4] Lambert, M A., Design of Solar Powered Adsorption Heat Pump

With Ice Storage, Applied Thermal Engineering, vol 27, pp.

1612–1628, 2007

[5] Aristov, Y I., Dawoud, B., Glaznev, I S., and Elyas, A., ANew Methodology of Studying the Dynamics of Water VaporSorption/Desorption Under Real Operating Conditions of Ad-

sorption Heat Pumps: Experiment, International Journal of Heat

and Mass Transfer, vol 51, no 19–20, pp 4966–4972, 2008.

[6] Aristov, Y I., Tokarev, M M., Cacciola, G., and Restuccia, G.,Selective Water Sorbents for Multiple Applications: 1 CaCl2

Confined in Mesopores of the Silica Gel: Sorption Properties,

Reaction Kinetics Catalysis Letters, vol 59, pp 325–334, 1996.

[7] Glaznev, I S., and Aristov, Y I., Kinetics of Water Adsorption onLoose Grains of SWS-1L Under Isobaric Stages of Adsorption

Heat Pumps: The Effect of Residual Air, International Journal of

Heat and Mass Transfer, vol 51, pp 5823–5827, 2008.

[8] Restuccia, G., Freni, A., Vasta, S., and Aristov, Y I., SelectiveWater Sorbents for Solid Sorption Chiller: Experimental Results

and Modelling, International Journal of Refrigeration, vol 27,

Silica Gel–Water Adsorption Isotherm Characteristics, Applied

Thermal Engineering, vol 21, pp 1631–1642, 2001.

[11] Aristov, Y I., Tokarev, M M., Freni, A, Glaznev, I S., and cia, G., Kinetics of Water Adsorption on Silica Fuji Davison RD,

Restuc-Microporous and Mesoporous Materials, vol 96, pp 65–71, 2006.

[12] Kakiuchi, H., Iwade, M., Shimooka, S., Ooshima, K., Yamazaki,M., and Takewaki, T., Novel Zeolite Adsorbents and Their Appli-

cation for AHP and Desiccant System, Kagaku Kagaku

[15] Chua, H T., Ng, K C., Chakraborty, A., Oo, N M., and Othman,

M A., Adsorption Characteristics of Silica Gel–Water System,

Journal of Chemical Engineering Data, vol 47, pp 1177–1181,

2002

Trang 25

[16] Matsushita, M., Adsorption Chiller Using Low-Temperature Heat

Sources, Energy Conservation, vol 39, no 10, p 96, 1987.

[17] Yonezawa, Y., Matsushita, M., Oku, K., Nakano, H., Okumura,

S., Yoshihara, M., Sakai, A., and Morikawa, A., Adsorption

Re-frigeration System, U.S Patent number 4881376, 1989

[18] Wang, X., Zimmermann, W., Ng, K S., Chakraboty, A., and

Keller, J U., Investigation of the Isotherms of Silica Gel+ Water

System: TG and Volumetric Methods, Journal of Thermal

Analy-sis and Calorimetry, vol 76, pp 659–669, 2004.

[19] Saha, B B., Boelman, E C., and Kashiwagi, T., Computer

Sim-ulation of a Silica Gel-Water Adsorption Refrigeration Cycle-the

Influence of Operating Conditions on Cooling Output and COP,

ASHRAE Trans.: Research, vol 101 (Part 2), pp 348–357, 1995.

[20] Aristov, Y I., Glaznev, I S., Freni, A., and Restuccia, G., Kinetics

of Water Sorption on SWS-1L (Calcium Chloride Confined to

Mesoporous Silica Gel): Influence of Grain Size and Temperature,

Chemical Engineering Science, vol 61, pp 1453–1458, 2006.

Ivan Glaznev is a researcher at the Laboratory of

Energy Accumulating Materials and Processes at the Boreskov Institute of Catalysis (BIC), Novosibirsk, Russia He received his master’s degree from the Novosibirsk State University (2003), and his Ph.D.

degree from the BIC (2006) His research activity covers the field of gas adsorption by porous media, heat and mass transfer in pores, and physical chemistry of superdispersed hygroscopic salts.

Daniil Ovoshchnikov is a Ph.D student at the

Lab-oratory of Energy Accumulating Materials and cesses at the Boreskov Institute of Catalysis, Novosi- birsk, Russia, under the supervision of Prof Yuriy Aristov His research activity covers the field of gas adsorption by porous media, heat and mass transfer

Pro-in pores, and physical chemistry of superdispersed hygroscopic salts.

Yuriy Aristov is a professor of physical chemistry

and the head of the Laboratory of Energy lating Materials and Processes at the Boreskov In- stitute of Catalysis (BIC), Novosibirsk, Russia He received his M.Sc degree from the Moscow Physico- Technical Institute, and his doctoral degree from the BIC His research contributions have been in the fields of radiation chemistry, low-temperature elec- tron tunneling, fractal analysis of porous solids, and thermochemical transformation of heat He is currently working on novel composite adsorbents for adsorptive chilling, gas drying, and maintaining relative humidity in museums.

Accumu-heat transfer engineering vol 31 no 11 2010

Trang 26

SU LING LI, JING YI WU, ZAI ZHONG XIA, and RU ZHU WANG

Institute of Refrigeration and Cryogenics, Shanghai Jiao Tong University, Shanghai, China

An iso-volumetric test unit was built to test the adsorption and desorption performances of a consolidated composite

adsorbent (mixture of CaCl 2 and expanded graphite) with ammonia as the working fluid The reaction kinetics and heat

transfer performance of the consolidated adsorbent were studied and two reaction orders [8/4 and 4/2] were considered in

this study The influences of the refrigerant condensation and activation energy on adsorption performance were analyzed.

The thermal conductivity of the consolidate adsorbent was tested by a transient plane source method Experiment results

showed that the liquid condensation phenomenon in the adsorption and desorption process was mainly produced by the

condensation of ammonia, which would lead to wrong results of the chemical kinetics; THE maximum error of the kinetic

models was about 9.6%, and could meet well the engineering application requirements The thermal conductivity of the

consolidated adsorbent tested in this article was very useful for the selection of a composite adsorbent with good mass and

heat transfer performance employed in the adsorption system.

INTRODUCTION

Adsorption refrigeration systems have attracted many

re-searchers because they can use natural refrigerants such as

ammonia, water, methanol, etc., which are environmentally

friendly, with zero ozone depletion potential (ODP) and global

warming potential (GWP) Furthermore, adsorption systems can

contribute to energy conservation as they can efficiently use

so-lar energy and low-grade waste heat Adsorption systems also

have the advantages of less vibration, simple control, low

run-ning costs, and less noise compared to vapor compression and

absorption systems [1] Therefore, more attention has been paid

to adsorption refrigeration systems

Many researchers in adsorption refrigeration fields have

fo-cused more on the reaction mediums Fin tubes and block

ad-sorbents used in the adsorbers can enhance the heat and mass

This work was supported by National Science Fund of China under contract

number 50736004.

Address correspondence to Professor Jing Yi Wu, Institute of

Refrig-eration and Cryogenics, School of Mechanical Engineering, Shanghai Jiao

Tong University, 800 Dongchuan Road, Shanghai - 200240, China E-mail:

jywu@sjtu.edu.cn

transfer performance of the adsorber, which shortens the cycletimes of the adsorption system from more than 10 h [2] to afew minutes, and led to the adsorption/desorption process be-come a nonequilibrium from an equilibrium process As a result,the equation of adsorption rate in an equilibrium process is notsuitable for the nonequilibrium process

For a nonequilibrium process, the equation of the

adsorp-tion/desorption quantity x and the pressure P , temperature T

does not meet the static equation,

x = f (P, T )

Actually, the adsorption rate should be expressed by

dx

dt = f (P, T ) where dx/dt is the adsorption rate In the nonequilibrium pro-

cess, the adsorption rate is the core of the adsorption principleand adsorption performance

As is well known, the equation of adsorption rate for physicaladsorption has been studied thoroughly [3–5] and is governed

by the linear driving force kinetic equation when the adsorber

931

Trang 27

932 S L LI ET AL.

is controlled by macroscopic diffusion into the adsorber:

dx

dt = K sap (xmax− x) where K sapis the overall mass transfer coefficient for adsorption

and desorption, x max is the maximum value of the adsorption

quantity, and t is the time.

Many researchers have already shown the chemical kinetic

equations for adsorption rate However, the adsorption kinetic

equations given are different for different working pairs and even

for the same working pairs [6–8] Furthermore, all the

chem-ical kinetic equations for CaCl2−NH3 adsorption rates were

all presented at the constant adsorption pressure and adopted

only one equation The reaction processes of CaCl· 2· 8NH3⇔

CaCl2· 4NH3 and CaCl2· 4NH3 ⇔ CaCl2· 2NH3 are not

dif-ferentiated

But in the practical application, the desorption/adsorption

process is not an isobaric process but an iso-volumetric

pro-cess [9, 10]; the pressure ranges from 1 bar to 20 bars

Additionally, in the desorption/adsorption process, the

reac-tions of CaCl2· 8NH3 ⇔ CaCl2· 4NH3 and CaCl2· 4NH3 ⇔

CaCl2· 2NH3do not take place simultaneously

As a result, an iso-volumetric measurement method was used

to test the adsorption performance of the composite adsorbent

and ammonia working pairs in this article The composite

ad-sorbent used in this article is treated with a water solution [11],

which has better heat and mass transfer performances compared

with the simple mixed composite adsorbent [6, 8]

The iso-volumetric measurement method can

distin-guish the two reactions CaCl2· 8NH3 ⇔ CaCl2· 4NH3

and CaCl2· 4NH3 ⇔ CaCl2· 2NH3 Therefore, the

differ-ent kinetic equations of the adsorption rates for the

reac-tions CaCl2· 8NH3 ⇔ CaCl2· 4NH3 and CaCl2· 4NH3 ⇔

CaCl2· 2NH3were obtained in this study

EXPERIMENT STUDY

Composite Adsorbent Preparation

The preparation process for the composite adsorbent is shown

pictorially in Figure 1 Raw expandable graphite powders were

heated in an electric oven at the temperature of 700◦C for 14

min; expansion and exfoliation occurred during the heating

pro-cess Calcium chloride powders were dried at a temperature

of 350◦C to remove the crystallization water and get the pure

calcium chloride powders (CaCl2) The composite adsorbent

mixed from CaCl2and expanded graphite presented in this

arti-cle is based on our previous works of Wang et al [12, 13], who

deeply studied the adsorption performances of the composite

adsorbents with different mass ratios of CaCl2 and expanded

graphite The results they obtained indicated that the

compos-ite adsorbent with mass ratio 4:1 has the best adsorption

per-formances, so the composite adsorbent with mass ratio 4:1 of

anhydrous CaCl2and expanded graphite is used in this article

Figure 1 The steps involved in the preparation of composite adsorbent: (a) expanded graphite at 700 ◦C; (b) dried calcium chloride at 350◦C; (c) simple

mixture of the dried calcium chloride and expanded graphite with mass ratio 4:1; (d) solution of the mixture; (e) dried at 350 ◦C; (f) the composite adsorbent

powder; (g) the composite adsorbent block used in this study; (h) magnified picture of (c); (i) magnified picture of (f).

The black material visible in the sample (Figure 1c), is expandedgraphite, whereas the white ones are small particles of calciumchloride Purified water was added into the sample (Figure 1c)

in order to dissolve the CaCl2and ensure uniform distribution ofthe CaCl2in the expanded graphite The mixture of CaCl2solu-tion and expanded graphite was then heated at 350◦C in order toremove the free water and allow for the impregnation of CaCl2into the expanded graphite matrix A homogeneous mixture ofCaCl2and expanded graphite powder was obtained ComparingFigure 1f with Figure 1c, it is apparent that the homogeneity

of Figure 1f is superior to Figure 1c, which is necessary to sure complete adsorption/desorption in the composite adsorbent[10] Finally, the dry mixture of expanded graphite and CaCl2

en-was compressed to form the composite block, with enhancedheat transfer performance

Test Unit of Thermal Conductivity

The transient plane heat source method was used to measurethe thermal conductivities of the composite adsorbent blocks.The thermal conductivity test unit consisted of two thermal in-sulation plates, two composite adsorbent blocks, and a test planeprobe, which were set up as shown in Figure 2 The diameters ofthe composite adsorbent blocks were 53 mm, whereas the radius

of the testing plane probe was 6.4 mm, so a one-dimensionalthermal conductivity model in an infinite plane can be adopted

to calculate the thermal conductivity

The heating resistance wire of the plane probe coils was ranged in double helical lines The heating resistance wire mate-rial was nickel, which has a uniform temperature resistance Theheat transfer engineering vol 31 no 11 2010

Trang 28

ar-Figure 2 Test sketch map of thermal conductivity.

voltage and current input at point A (Figure 2) were measured

Then the resistance of the wire was calculated by the value of the

voltage and current Finally, the temperature of the plane probe

wire was gotten by the wire resistance The detailed calculation

method has been presented by Li et al [14]

The Test Unit of Adsorption Performance

A test unit was designed as shown in Figure 3 to test the

ad-sorption performance of the composite adsorbent with ammonia

as the working fluid The system mainly consisted of two parts,

i.e., adsorber and gas container The adsorber was placed in an

insulated tank and was heated/cooled by circulating hot/cold

water from thermostatic baths

The detailed adsorber design is shown in Figure 4 It is

com-prised of two parts, which were sealed by the aid of a flange The

distance between the fins was 5 mm The height of the fins was

12 mm The composite adsorbent blocks were packed into the

spaces between the fins The mass of the composite adsorbent

packed in the adsorber was 53.6 g

This test unit has two working processes: heating process

and cooling process of the adsorber In the heating process, the

valve is open, the adsorber is heated by the heating water from

thermostatic water tank, and when the adsorber has a higher

temperature, it desorbs ammonia to the gas container In the

cooling process, the valve is also open, and the adsorber is

cooled by the cooling water from another thermostatic water

tank When the temperature of the adsorber reaches a lower

value, the ammonia in the gas container is adsorbed by the

Figure 3 Adsorption test unit.

Figure 4 Structure of adsorber.

adsorber If the pressure in the gas container does not change in

60 min, the desorption/adsorption process was considered to becompletely finished

The parameters measured were:

• The temperature T c and pressure P cof the gas container

• The temperature, T ad and pressure, P adof the adsorber

• The temperature of the composite adsorbent Ta

• The reaction conversionX(t) and reaction rate dX/dt [10].

All the temperature probes used in this work had the sameaccuracy of 0.1◦C, whereas the pressure sensors had a measure-ment error of 0.1%

SELECTION OF THE KINETIC MODEL

The reaction formulas for the complex reaction betweenCaCl2and NH3are described as follows [10]:

CaCl2 · 8NH3+ H84⇔ CaCl2×4NH3+4NH3

at temperatureT eq84 (1)CaCl2 · 4NH3+ H42 ⇔ CaCl2 · 2NH3+ 2NH3

at temperatureT eq42 (2)

where H84 and H42 are reaction enthalpy, and T eq84 and

T eq42are the equivalent reaction temperatures

The general form of adsorption rate is a function of the local

state variables X, P , and T , as given in Eq (3):

dX

dt = f (X) · K(P, T ) (3)Equation (3) is based on a large amount of experiment datausing the iso-volumetric measurement method in this article,

and f (X) and K(P , T ) are the vacant sites presented by Wang

and Wang [6] as Eq (4) :

f (X) = (1 − X) n

(4)

where n is the reaction order; X is the reaction conversion, the

value of which is between 0 and 1 and can be expressed in aheat transfer engineering vol 31 no 11 2010

Định dạng
Số trang	57
Dung lượng	4,28 MB