_Corea~J. Chin.

E ~ , 19~ 1 h :-:7-9203021

Absorption Characteristics of Ammonia-Water System in the Cylindrical Tube Absorber

K i - B o n g Lee, B y u n g - H e e Chun, Jae-Cheol Lee, Chan~Iin Park* and S u n g - H y u n Kim t

Depatlment of Chemical Engineering, Korea University, Seoul 136-701, Korea *Department of Environmental Engineering, Junior College of Inchon, Inchon 402-749, Korea
(Rec~zved 5 _~S~rch 2 c ~ l 9 ~cc~pt~d 2 7 Jzdy 2001 )

Abstraot-Experimental analysis was perfomled in a cylindiicsi tube absorber which is considered to be suitable for the lml:hle mode. Characteri~cs such as concentration, temperature, and pressure were meamred and they reflected the condition of absorber wall. The variation of characteristics was conspicuons near the inlet region of the ammonia gas. The ammonia gas and the solution flowed cocurrenfly and countercurrenfly and the results were compared. Key words: Absorption System. Ammor~a-water, Cylinc~ical Tube Absorber, Mass Transfer, Gas Holdup INTRODUCTION Due to the ozone depletion problem associated with the use of CFC and HCFC refiigemnts, absorption heat pumps and refiigeration systems have gained increasing interest in recent years More and more, they are regarded not only as environmen|ally fi'iendly altmaatives to CFC based systems, but also as energy efficient heathag and cooling technology [Herald et at, 1996]. The absorber, uahieh is a major component in absorption refiigeration systems, greatly affects the overall system perfonnance. Bubble mode has been recommended to enhance heat and mass Wansfer perfonnance in anmonia-w.ater absorption systems [Chtistensen et al,, 1996]. Falling film modes provide relatively high heat transfer coefficients and are stable during operation. However, falling film modes have wettability problems and need good liquid distributors at the inlet of the liquid flow. Bubble modes provide not only high heat transfer coefficients but also good wetmbility and mixing bm,eeen the liquid and the vapor. However, the bubble modes require vapor dislribution. Generally, vapor distribution is e~sier than liquid dg~lmtion. Recently, bubble modes were recommended strongly for ammonia-water absorption systems becmse the low wettability in the falling film modes is critical to the perfonnance of the system [Christensen et al., 1996: Han et al., 1985]. Over the last ten year~ ammonia-water bubble mode has been extensively investigated both numerically and analytically [Herbine and Perez-Blanco, 1995: Kang et al., 1997: "& et al., 2001: Sung, et al., 2000: Tsutsumi et al., 1999: Jang et al., 1999: Hwang et al., 1999: Lira et al., 1999: Noda et al., 2000]. However, few papers have been found concerning the heat and mass lransfer analysis of the bubble absorber. Merrill et al, tested three compact bubble absorbers developed for generator-absorber hen exchange absorption cycles (GAX) [Mewill et al, 1998]. Their resalls show that enhancement techniques are effective in reducing absorber length and inTTowhom correspondence should be addressed E-mall: kimsh@korea.ae.kr *Presented atthe Inf/Syrup. on Chem. Eng. (Cheju, Feb. 8-10, 2001) dedicated to Pros I-I. S. Chun on the occasion ofhls retirement from Korea Ur~versity. 87 creased tube diameters may increase absorber performance. In the present study, the characteristics of the bubble mode in the cylinckical tube absorber were studied by experknent. The results can be used as fimdumental datato design the optimal cylindrical tube absorber. EXPERIMENT The schematic diagram of the cylindrical absorber for the bubble mode studies is given in Fig. 1. The cylinckical cohann is 1 m in height and 0.03 m in diameter. Five sample ports are installed at
?
,. ,

"

.~

J, ,~,,

,I

.,L,

,'3'

:)
,f N tI . . . . . . . . " ! '

,', ~1.~ ..... ~'~.~

Hg. 1. Experhnental abserpfion system for c~Mdrical tube absorbic.

1. Absorber 2. Manometer 3. NH 3 bomb 4. Sampling port 5. Thermocouple 6. Valve 7. Mass flow controller

8. Cartridge heater 9. Data acquisition system 10. Heating tank 11. Solution pump 12. How meter 13. Neutralization tank

88

K.-B. Lee et al.

0.2 rn intm,als axially and twelve thermocouples are set to measure the behavior cs Iransfer and the thermal state of re.her experimental column. Also, seven mananeters are equipped to measure the presmre profile The cylindrical column made o f a ~ f i c resin is Iransparent, so the state of bubble flow can be observed In the experiments, flow rate of @aamoniagas was 1-9 L/rain and the inlet liquid concenlration was selected as 20-30 weight% Ammonia gas flowed upward, utfile anm onia solution flowed bah upward and downwarc[ ANALYSIS The absorption rate is expressed as the following equations by using the overall mass Iranffer coetfidents, K [McCabe et at, 1993: Kim et aL 2001-" Chok 2001: Tsutsumi et al., 1999: Park, 1999].
n~ =KpA~sAx~,a

in the binary mixture, is an imporlant paraneter for charact~izing the absorption performance of the absorber. The gas holdup influences the interracial area between the liquid and the gas and thus detelmines the mass transfer rates during the absorption process [CI~ et at, 1978: Fukuji, 1999: Yarnashita et al., 1999: Choi et al., 1992: Mok et at, 1990: Ide et al., 1999: Kim et al., 1990: Park et at, 2000]. The gas holdup can be obtained in(~'ectly from the pressure variation:

Po- Pl=plg~=p~gZ1
Po- e~= ~ g ~ =e,p,gZe.

(3)

(4)

(i)
(2)

where subscript 0 is standard position, 1 is initial x~Jue when only solution flowed, and 2 is final value when the gas flowed ~ the solution. Dividing Eq. (3) by Eq. (4) and arranging yields
AP eg-pl_:Po (5)

Ax~,s -

[x;g-x,j -[x;~-xo~,l
ln[(x~ -x,~),"(x~ -xo.O]

]'he physical Wop~ies needed to determine the eoetfidents were referred to literature [Reid et al., 1986: KDB]. Gas holdup, "which is defined as the volume ratio of gas phase

RESULTS AND DISCUSSION 1, Effect of Gas Flow Rate on Concentration Profile

,%: rl ,, -, ........... '& 2 i "~'-

~,,~, ,,,-,- , ,,,,, ,, ............. '"~:'"'~ ';'+"" ~,3 k" ,

',,, tg2 ~,,, ,, f& 13 Y~7 2"I, 7 fi I ,'= ~ 7", ,~ ~ " %" rim, '\ '" -" -:,,~4
i

. ,_,,-~,_.

.~rjI,L "
I',r '$

,,, Cr ,b

~etghr .m l 3 ~ , a , ~ o r l T a ~ot ;,(a~Tt~e#t ft~te~

:1~- ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . .....

rN~,. ~i,'
'~,,~ ;~,[ ri .Y,I t~

~,
\~ ~1~,~,,,.,,,>

.~,<~"~' .~ 4r
'"4 C .,'1 ,, ,,

~ f,,ii._.~ '........

~ ........... ~

.......

......

~,

~. ~
7" i

"-.,

,,,.-

~,'t~

~ ',!

::,1

2 4

: 'd

:, J

: ]'

II I'r

,t~

' ,~

I ',

~"~'<~,/Hin"..',',
Fig, 2, The effect of gas flow rate on eoncenWafion profile (coowrent). Jalmary, 2002

Fig, 3. The effect of ~ts flow rate on concmtration profile (conntermrr~).

Absorption Characteristics of Ammonia-Water Systemin the Cylindrical Tube Absorber Fig. 2 shows the effect o f the gas flow rate on the concentration profile vvhen the gas and the solution flows are cocutrent At low gas flow rate, the concentration of solution is aknost same above 012 ml At high gas flow rate, the concenlration of solution increases w~h increasing height of absorber Most of the ammoniagas was absorbed immediately after injection at low gas flow rate. However. as gas flow rme increased, the remaining gas which vcas not absorbed increased Fig. 3 shows the effect o f the gas flow rate on the concentration profile W e n the gas md the solution flows are counte~cmlvnt Contrmy to cocurrent flow, the concentration of solution increases with decreasing height of ~ ~ l Similar to cocurrent flow, the concentration o f ammonia is aknost same above 0.2 m at low gas flow rate. At high gas flow rate, compared to low gas flow rate, the unabsorbed gas increased so that the length of absorber which was needed to absorb gas was increased It is becanse as the amount of input gas increased, the capacity of solution that can absorb the gas decreased. The un-absorbed gas was gradually absorbed as it flowed upward, As can be seen in Fig. 3, the unlabsorbed gas was more in the 3 ~ o ammonia solution than in the 2~.o ammonia solution. 2. Effect of Gas Flow R m e on Temperature lh'olile Fig. 4 shows the effect of the gas flow rate on the t e m p e r m ~

89

i ,,~;,

t~

'~-,,,,~d: qf'JrTr

,a ~,;

~,~,,
,',~ ,

*'~.,

,.',-

o,

r,r r~

~, ~7

I:t g

f~ 5

2d;!% mm~ni=

~oI (,mJ~t*er~eu~ent flow)

: .~ ;~ ~',~ ',~r~- rL~,,,l~l

................................

;",a,?, ~.#q ,.~',, <,'. v - v, [

.~.

-,,,~. ~:~, , ~ ~,,i -~ ,~, :=:

'~lf

~l~, a

5
~l~r ~{flr ~P~HWP~ 1 ~l ll 1 ~ l~lllfl { l~

Fig, 5. The dTect of gas flow rate on temperature profile (counteroarrent).

:2'

I~ l

11' 1

iI~ ~

i ~

~ ~

~ ,~

h';

.............................................

7. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
' ,,@,

:~07~ ~ ' ~

~:

'~ ,* ' :.

:: -'i~'3~

':".'~,=~--'~ ",'~='~,"~"

3 ',~

#~ ,.g~,~,.,,~ ~ "~'=="~,,,.~...~,,~"~

lllg

l~llll

ff

~
l~
~ll Jl l

~ll~

I" " "~l

I$~IIH~

'l~lll- l,,~ ,

l ~

"I"

,:"

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . .

l l ~

~ ~

~] ;

~. 1~

~,' ~'~" I~-r Fl~ 4. The dTect of gas flow rate on temperahwe profile (cooarrent).

profile when the gas and the solution flows are cocurvent The temperature rose rapidly as soon as the gas flowed in, and then it increased a little or remained constant. Fig. 5 shows the effect of the gas flow rate on the temp~nture profile ~Mlen the gas and the solution flows arc countercunent. The temperature of solution was practically constant due to the solution flow: near the top of absorber. It increased ~ n n the middle of the absorber to the bott(in of the absorber where the gas flowed in. The region where the temperature variation occurred was larger at the higher gas flow rate and in the solution of higher concenWation. It means that the ammoniagas is not absorbed well on the condition of high gas flowrate and high concentration solution. 3. Effect of Gas Flow Rate on Pressure Vm~imion Profile and Gas Holdup Fi~ 6 and Fig. 7 showthe effect of the gas flowrate on the profile of gas holdup, z~. ~ e n the gas and the solution flows are cocurrent and countercma~t respectively. The gas holdup was obtained fi~m the measured pressure profile. The gas holdup decreased the increasing length of the absorber, and had the similw decreasing trend ~ the pressure variation. The gas holdup was higher on the condition of high gas flow rate and high concenWation solution. The absorber can be divided into two regions by the slope of Korean J. Chem. En~(Vd. 19, No. 1)

90

K.-B. Ice a al.

.....................................

:. . . . . . . . . . . . :: '77 ..7=:-7 : ._: - y '

,,,~,~,~-

i~,~

S~,=J~ =.,~,, r k'~.r~ ~'.;~" "~

"~~

"=, ,~=.,r~ "

'3L ~ ' ,*~

_~._..,;_~_r "2_ ,':.r, '.. ,,;~

,A ,,.r

...~, -~@

.........~I~..~,~,~.~,,~
. . . . . . . . . . . ~ .............
12 [.'P[ '~ -

~'~
U 3

f~:~
30% a,mm~ma ,~r ( c a u n ~ r e u ~ t
I? [,~ '~ .;~ ' ~ ; w "~.'~ ~ , : v r ~ "

fto~

0 g

~'t I'llY H'.'.

\
~,'~":'~ ........ ~ . . . . . . . ,~ .... ~~"::,:,~:T.~,'.,:,:,:',:,:.7.~

2 #
~5

'r.~,

i, I:1~

9:',

" -,

,;::

:,zl

_--?,-

'L

'

,"(,

FL f

:t~,

,) :)

~,,~

i'r t'~

':~

r ;

,4~, ',3 ~ [r,~]

Fig. 6. The effect of ga~ flow rate on ga~ h~dup profile (cocurrent). gas holdup. The slope is steep in the lower part of the abs&oer and it is gentle in the higher pat. It is because the ammonia gas remained in the lower pat and the pressure varied sharply as the gas was absorbed. After the gas was absorbed, the presa~re variation became fiat. Pressure variation means the difference between the pressure when only the solution flowed and the pressure when the gas flowed together. The region where the ammonia gas was mainly
Table 1. Results of mass traasfer coefficient

Fig. 7. The effect r gas flow rate on gas holdup profile (counteroarr~t).

ab~ogoed c ~ l d be l~own by the t:ressure variation orthe gas holdulx Optimum height of absorber can be obtained by analyzing concenlration, temperature and gas holdup. It is approximately 0.4 m and 0.6 m, when the gas flow rate is 9 L/rain md the concentration of input solution is 20?,5 and 3~,o, respectively.
4. Comparison of Mass Tr~asfer Coefficients

Mass transfer coeflident at 209'0 input solution [m3/min]

C~s flow rate [IJmin]

Plate type absorber Cylinda:ical tube absorber Falling film mode Bubble mode Cocurrent flow Countercurrent flow
1.185 7.131

I
9 1 0 -3 9 1 0 -3 --~ --~ 2.885

5
1.942.10 -'~ 3.815 9 1 0 -'~ -'~

9
- 1 0 -'~ --~ --~ 1 0 --~ 5.760.10 8.728.10 5.910-

1.627.10 1.564.10

3.018.10 3.275

9 1 0 --~

Mass transfer coeffident at 30~ input solution [ m ~ / m i n ]

G-as flow rate [L/min]

Plate type absorber Cylinda:ical tube absorber Falling film mode Bubble mode Cocurrent flow Countercurrent flow

1
9.257.10 4.110.10 1.839.10 7.468.10 -3 -'~ -1 --~

5
4.712.10 8.664.10 2.234.10 1.546.10 -'~ -'~ -1 -1 9.3341.655 2.581 2.228-

9
1 0 -'~ - 1 0 -1 - 1 0 -1 1 0 -1

Januaxy, 2002

Absorption Characteristics &Ammonia-Water System in the Cylindrical Tube Absorber Table 1 stwnmarizes the results of mass transfer coefficient when tile concenb-atic~l of input solution was 2(~.0 and 3(~.0. The results of plate type absorber are from the former study [Lee et al., 21))0]. The mode of cylindrical tube absorber was the bubble mode both in the cocmrent and in the counterctarent flow. This table shows that the mass transfer coefficients in the bubble mode were greater fflszl those in the falling film mode. In particulm, the mass b-ansfer performance in the cylindrical tube absorber was excellent The results both in the co,n-rent and in the countemurrent flow were much the same compared with the results in the plate type absorbel: CONCLUSIONS For the fiarther understanding of bubble mode, the characteristics of bubble mode in the cylin&ical tube were studied. The absorption and complex flow behavior in multiphase flow systems were interpreted by analyzing state variables such as concentration, terape~atu~e, and gas holdup. The following conclusions were drawn from the present experimental studies: 1. As the gas flow rate increased, the concentration, temperature and gas holdup of solution increased 2. When the gas and the solution were cocurrei~ the concentration and temperature of solution increased with increasing height of absorber. However, when counterctwrent~ the concentration and terape~atu~e of solution increo~sedwith decreasing height of absorbel: 3. For the gas holdup, in the lower part of the absorber the slope was steep and in the higher part the slope was gentle. 4. Optimum height of absorber can be obtained to analyze concentratiorl~ temperature and gas holdup in the absorber ACKNOWLEDGEMENT This study was supported by research grants from the Korea Science and Engineering Foundation (KOSEF)through the Applied Rheology Center (ARC), an official ERC, at Korea Uifiversity, Seoul, Korea. NOMENCLATURE A g K ra P x Z : area [ra~'] : gravitational acceleration [m/s 2] : overall mass transfer coefficient [m/min] : mass flow rate [kg/rain] : pressure [mmHg] : concentration [weight %] : height [m] g in l lm out :gas : input :liquid : log mean :output REFERENCES

91

Greek Letters p : density [kg/m ~] e :holdup Superscript eq :equilibrium Subscripts abs :absorption

Choi, K. H. and Lee, W. I-2, "'Behavior of Gas Bubbles in a Concentric Cylindrical Airlift Column~"Konean J Chem. Eng., 9, 66 (1992). Choi, K. H., "'Hydrodyrkamic and Mass Transfer Characteristics of Exte~llal-Loop Airlift Reactors witilout an Fxtension Tube above tile Downcomeff" Ko,ean J. Chem. Eng., 18, 240 (2001). Choi, K. H., "'Hydrodynamic and Mass Transfer Characteristics of Exte~llal-Loop Airlift Reactors without an Fxtension Tube above tile Downcomeff" Ko,ean J. Chem. Eng., 18, 240 (2001). Ct~istense~, R. N., Kang, Y Z, G~imella S. and Pliedemm~, D., "'Generator Absorber Heat Exchanger (GAX) Cycle Modeling, Component Design and Tests;" Final Report to LG Electronics, Contract number: 862805-01/062805 (1996). Clif~ R., Grace, J. R. and Weber, M. E., "'Bubbles, Drops, and Partides~" Academic Press, New York (1978). Fuknaji,Y., "'Effect of Geometrical Pan~neters of I~-aft Tubes and Clear Liquid Height on Gas Holdup in a Bubble Column for Gas Dispersion into Tubes~"Korean ~ Chem. Eng., 16, 789 (1999). H~1, M. W, Choi, D. K. and Lee, W IC, "'Efl}ct of Shape and Wettability of Packing Materials on the Efficiency of Packed ColumlC Korean ~ Chem. Eng., l , 25 (1985). Heroine, G. S. and Perez-Blanco, H., "'Modd of an Almnonia-Water Bubble Absorber~'ASHRAE D'ansaetions, 101 (1995). Hei~ld, K. E., Rademladlel, R. and Klein, S. A., "'Absorplion Clfillers and Heat Pumps~" CRC Press, USA (1996). Hwang, Y G. and Chin1, H. S., "'Enhancement of Mass Tr,~lsfer in the Fluidized Bed Electrode Reactors:" Korean J Chem. Eng., 16, 843 (1999). Ide, M., Uchiyama, H. and Islfika~-a,Z, "'Flow Characteristics of GasLiquid Two Phase Plmlgmg Jet Absorber-Gas Holdup and Bubble Penetration Depth-;" Korean ~ Chem. Eng., 16, 698 (1999). Jang, S. H., Aim, W S. and Ha, J. M., ~Mathe~nalicalModel of a Monolith Catalytic IncineratorT"Lbrean J Chem. Eng., 16, 778 (1999). Kozlg, Y T., Cluistense11, R. N. and Kashiwagi, T., "'Almnonia-Water Bubble Absorber with a Plate Heat Exdlanger:" ASHRAE D'ansactions, 104, 1 (1997). KJllk J. W., Chang, J. H. and Lee, W. IC, "'Intfibition of Bubble Coalescence by the Electrolytes:"L=o,~an~ Chem. Eng., 7, 100 (1990 ). Kiln, S. H., Bidkm; A. and Ngo, H. H., "'Adsoiplionand Mass Transfer Cl~aractelislicsof Mets+ulfumn-Methylon Activated Carboff" Ko#ean J. Chem. Eng., 18, 163 (2001). Lee, IC B., Chin1, B. H., Lee, J. C., Park, C. J. and Kim, S. H., "fftle Comparison of Heat & Mass Transfer in Absorber between Falling Film and Bubble Columnin Heat Pump System:"Proceedings of the Seventh Asian Collf~-~lce oi1 Fluidized-Bedand Tt~ee-Pliase Reactor~, 515 (2000). Linl, J. H., Yi, G. B. and Suh, K. H., "'A Simulalion of Electrodlenfical Kinetics for Gas-Liquid-Solid Phase of MCFC Anode:" Kotean J Chem. Eng., 16, 856 (1999). McCabe, W. L., Smiti1,J. C. and Harriott, P, "'Unit Ope1-alionsof ChinKol~an J. Chem. Eng~(VoL 19, No. 1)

92

K.-B. Lee et al. Sung, W. M., Huh, D. O. and Ryu, B. J., "'Devdopment and Application of Gas Hydrate Reservoir Simulator Based on Depressurizing Mechafflsm~'Ko~an J Chem..Eng., 17, 344 (2000). Thermodyrkarnics and Properties Lab. of Korea University: Korean Themlopliysical Properties Data BaiLk. Tsuts~rni, A., Clie~],W and Kin], Y. H., "'Classificalion and Cli~-acterization &Hydrodynamic and Transport Behaviors of Three-Phase Reactors~"Kotvan ~ Chem..Eng., 16, 709 (1999). Tsuts~rni, A., Clie~],W. and Kin], Y H., "'Classification and Cha-acterizalion of Hy&~dynanic and Tr,~]sport Behaviors of Tt~-ee-Phase Reactors~"Kof~an,~ Chem..Eng., 16, 709 (1999). Yarn~hita F., "'Effect of Geometrical Parameters of Draft Tubes and Clear Liquid Height on Gas Holdup m a Bubble Colunm for Gas Dispersion into Tubes~"Korean ,~ Chem..Eng., 16, 789 (1999). Yu, Y. H. and Sosrua, M. H., "'Modeling for Industrial Heat Excl~qnger Type Steam Refomle~-~" Lb~an ~ Chem. Eng., 18, 127 (2001 ).

ical Engineefing~" McGiaw-Hi11, Singapore (1993). Merrill, T. L., Setoguchi, T. and Perez-Blanco, H., "'Compact Bubble Absorber Design~',EnhancedHeat Yransfet, 5, 249 (1998). Mo!~ "K S., Kim, Y. H. and Kim, S. X, "'Bubble and Gas Holdup Charactefistics m a Bubble Colunm ofCMC Solutiols Kofean,~ Chem. .Eng., 7, 31 (1990). Noda H., Takam~su, T. and Aso, I-2, "'Development on a Coaxial Heat Integrated Dislillalion Coha~m (HIDiC):" Kofvan J Chem..Eng, 17, 593 (2000). P~-k, C. J., "'Hydrodynamic Charactefislics of Air-Lift Activated Carbon Shaly Column with External Loopmg~"_Kof~anJ Chem..Eng., 16, 694 (1999). Park, I. S., Kwak, C. and Hwang, Y G., "'Frequency Response of Conlmuous-Flow Adsorber for Multicomponellt Systens Korean 3: Chem. Eng., 17, 704 (2000). Reid, R. C., t~-ausifitz~J. M. and Poling, B. E., "'The Properties of Gases & Liquids:" McGraw-Hill, Singapore (1986).

January, 2002