DELA CRUZ, Patricia Bianca B. 4 ChE – B
Date Performed: March 3, 2015 Date Submitted: April 25, 2015 Problem B1
Pressure Drop ad !loodi" i a Pa#$ed Colum
%&
%trodu#tio
Packed towers occur in almost all chemical plants for separation processes such as gas asorption, sol!ent e"traction, distillation or chemical reactions. #he packed column in figure 1 consists of a gas and li$uid inlet and outlet, a distriuting space at the top top and and otto ottom, m, and and impor importa tantl ntl% %, the the packi packings ngs.. #he #he enteri entering ng gas flows flows from from the the distriuting distriuting space elow the packed section to the packing interstices where it contacts the the desc descend ending ing li$uid li$uid.. &t also also opera operates tes in a wa% wher where e two two diff differ erent ent fluid fluid phase phases, s, particularl% gas and li$uid, were allowed to flow countercurrentl% enaling a chemical component, known as solute, to e transferred from one phase to the other phase.
'igure 1 Packed (olumn Meanwhile, the packings pro!ide the large surface area needed for intimate contact etween the li$uid and the gas phase. As shown in figure 2, the most commonl% used commercial packings are raschig rings, lessing rings, erl saddles, and pall rings )1*.
'igure 2 Most commonl% used packings+ a- raschig rings, - lessing rings, c- erl saddles, and d- pall rings #his e"periment mostl% deals with the gas asorption separation process in!ol!ing the airwater s%stem. /ne of the oecti!es of this e"periment is the determination of !oid fractions of the packed eds. &n gasli$uid flow s%stems, !oid fraction is defined as the fraction of the channel crosssectional area that is occupied % the gas phase )2*. &t is one of the most important parameters used to characterie two phase flows and ha!e a fundamental importance in models predicting the pressure drop )3*.
/ther oecti!es for this e"periment are the determination of the effects of li$uid holdups on the pressure drop of the packed column and determination of packing factor e"perimentall% with the use of flooding !elocit% calculations. 'rom a fluid mechanical perspecti!e, the most important issue is that of the pressure drop re$uired for the li$uid or the gas to flow through the column at a specified flow rate. rgun e$uation is one of the man% e$uations to sol!e for the pressure drop across a packed ed length ut with the limitation of onl% ha!ing an a!erage of 0. !oid fraction )*.
2
2
∆ P 150 μ v o ( 1− ε ) 1.75 ρg v o ( 1− ε ) = + 2 2 3 Z ε D p ε D p
Ergun Equation
4here !o superficial gas !elocit%, p is the particle diameter, 6 is gas !iscosit% and 7 represents the !oid fraction. Also, the rgun e$uation descries flow for oth laminar and turulent. 8owe!er, one e$uation that was onl% applicale for a laminar flow was % Blake9oen% which is actuall% the first term of the right side of the rgun e$uation. Another separate e$uation % BurkePlummer was the second term of the rgun e$uation applicale onl% for turulent flows. Meanwhile, 'ahien and :chri!er ga!e a
modified rgun e$uation for computing pressure drop as function of porosit% as shown elow )1*. ϕ L=
ϕ T =
136
;aminar 'low
0.38
( 1 −ε )
0 / 75
29
( 1− ε )
1.45
ϵ 2
+
1.87 ϵ
N ℜ , p
#urulent 'low
(1− ε)
0.26
ϕ I =q ϕ L +( 1−q ) ϕ T
&ntermediate
where+ 2
− ε (1−ε ) N ℜ , p
q =e
12.6
3
ϕ=
2
∆ P ε D P 2
Z v o μ ( 1 −ε )
'or irrigated packed eds, ;e!a 1=5- added a correction factor on the orifice e$uation. /n the other hand,
2
C 4 L f
where+ ∆ Pd =C 3 G f 10
∆ P L= 0.4 ( 0.00005 Lf )
0.1
( ∆ Pd)
4
Gf =986 F s ( 0.05 F pd )
0.5
0.1
0.5
Lf = μ L L ( 0.05 F pd )
( ) 62.4
ρ L
#he !alue of ( 3 was >."10 ? while ( was 2.>"10 5. /n the other hand, the packing factor, ' pd, was defined % ;oo et al. as+
F pd =
%%&
( −ε )
6 1 3
ε D p
'ethodolo"(
'or this e"periment, the Armfield @as;i$uid Asorption (olumn apparatus was used as shown on the figure elow. &olumn
Water Flow #al$e Air Flow #al$e
Pa!ked 'eds
Equipment On-O Swit!" Water Flow Meter Air Flow Meter Water-Dye Manometer Air and Water Knobs Dis!"ar%e Pipe #al$e
Sump Tank
Air Pump Water Pump
'igure 1 Armfield @as;i$uid Asorption (olumn
Before the e"periment was conducted, length of the packed eds and the diameter of the gas column were first measured. All remaining water in the e$uipment was also drained and the sump tank was cleaned. Afterwhich, the sump tank was filled again with water up to >5 of its capacit%. 'urthermore, the onoff switch and knos were turned off as depicted % the figure elow.
'igure 2 #urned off+ e$uipment switch left-, air and water knos right-
#he air and water flow !al!e together with the drainage !al!e found at the ottom of the sump tank was also closed. /n the other hand, the discharge pipe !al!e and all pressure taps were opened. 'or startup, the switch was turned on to run the air pump where the flow rate was set to 150 ;min for 15 minutes for the remo!al of an% water in the column. #he threewa% glass cocks were also adusted such that all the gas flowing were directed to the manometer alread% containing water and a redorange d%e. /n the e"periment proper, the air control !al!e was throttled ack to C0 ;min. #he differential pressure in mm8 2/ was measured and recorded accordingl%. Afterwards, the gas flow rate was increased with an increment of ten 10- up to the 150 ;min flow rate accompanied % the measurement of differential pressure for each inter!al. #he procedure was repeated ut with different water flow rates from 1 ;min up to > ;min with an increment of one e"cept that pressure was also recorded for the water flow rate of C.5 ;min. 'or a proper shutdown of the e$uipment, all water was drained with the gas rate set to 150 ;min for 15 minutes. 'inall%, the pump and the switch were turned off properl%.
%%%&
Results ad Dis#ussio
'rom the raw e"perimental data of pressure difference ased on the manometer fluid height, pressure drop was computed as follows with a specific gra!it% of 1.0.
P =
∆h
ρ g
g C
where densit% D- is for water at 25 o(. #ale 1. "perimental Pressure rop L%)U%D 0.0 1.0 2.0 3.0 .0 5.0 C.0 !L*+ RA,E -L.mi/ A%R !L*+ PRESSURE DR*P -lb.ft 0/ RA,E -L.mi/ 20 30 0 50 C0 >0 ?0 =0 100 110 120 130 10 150
0. 0. 0. 0.? 1.C 2.0 2.= 3.3 3.> .5 5.3 5.> C.1 C.5
0.? 0.? 0.? 0.? 1.C 2.0 3.3 .5 5.3 C.5 >.> ?.2 =. =.?
0.? 1.2 1.2 2.0 2. 3.> 5.> >.3 11. 11.? 13. 1.> 15.5 1C.3
1.C 2.0 2. 3.> 5.3 >.3 10.C 15.= 1>.5 20.0 22.0 2.= 32.C 33.
1.C 2. 2.= .1 5.> >.> 11. 1C.3 22.? 33. 1.2 51. flood
2.0 2. .5 ?.2 13.= 1=.C 2=.? 3?.3 flood
2.0 .1 C.= 10.C 21.2 32.2 flood
C.5
>.0
2.0 .5 =.0 1C.3 23.2 3=.5 flood
2.0 C.1 10.C 1C.> flood
'rom the computed data, pressure drop increases as the air and li$uid water flow rate increases. #he highest pressure drop reading was 51. lft 2 where li$uid and air flow rates were ;min and 130 ;min respecti!el%. 4ith a gas flow rate of 10 ;min and same flow rate for water, li$uid accumulation at the top of the packings was oser!ed signaling flooding. ue to this oser!ation, flooding is therefore defined as the condition where a large pressure drop occurs with a small change in gas !elocit%. Additionall%, the lowest air flow rate that produces flooding was C0 ;min with the corresponding ma"imum allowale li$uid flow rate of > ;min. #hus, flooding could e also oser!ed with lower air flow rate when li$uid flow rate was high.
Eoid fractions were computed using the formula of 'ahien and :chri!er e"pressing pressure drop as function of porosit% which was the modified e$uation of rgun. 'or this e"periment, the following were the !oid fractions otained. #ale 2.
Eoid 'ractions 7- for ifferent Air 'low
L%)U%D !L*+ RA,E -L.mi/ A%R !L*+ RA,E -L.mi/
20 30 0 50 C0 >0 ?0 =0 100 110 120 130 10 150
0.0
1.0
2.0
3.0
.0
5.0
C.0
C.5
>.0
0.2>2 5 0.251 0.23 > 0.221 0.1=0 > 0.1>C 3
0.2>2 5 0.2 > 0.21> = 0.1=5 3 0.1?5 C 0.1C5 ?
0.2>2 5 0.22 0 0.20> 5 0.1=3 =
flood
flood
*%D !RAC,%*2S -3/
0.?C 5 0.5C 0 0.1 0.3> 1 0.3?C 2 0.3>= C 0.3C0 C 0.35= 2 0.35? 0 0.3? 0.31 3 0.31 ? 0.32 3 0.32 >
0.3? 0.3?C 2 0.1 0.3> 1 0.3?C 2 0.3>= C 0.3? 5 0.330 > 0.325 0.315 ? 0.30? > 0.311 1 0.305 C 0.30> ?
0.3? 0.3? 0.3>5 0 0.3? 5 0.3? 5 0.32C 3 0.300 5 0.2?= ? 0.2C 0 0.2C? 5 0.2C5 0.2C = 0.2CC 3 0.2C> >
0.2?= ? 0.30 5 0.313 2 0.2=? 2 0.2?3 5 0.2>0 0.253 2 0.233 0.233 = 0.231 5 0.230 = 0.22? 1 0.215 5 0.21? 3
0.2?= ? 0.2?= ? 0.300 5 0.2?= ? 0.2>> ? 0.2CC 3 0.2? 0 0.231 > 0.21C > 0.1== 0.1=2 0.1? 5
0.2>2 5 0.2?= ? 0.2C5 0.23? ? 0.21C 0 0.20 3 0.1?> ? 0.1?0
flood
flood
flood
Ealues of computed !oid fractions range from 0.?C5 to 0.1C5?. #he lowest air and li$uid flow rates showed the highest !oid fraction. 4hile, an air flow rate of >0 ;min with C.5 ;min li$uid water flow rate otained the lowest !alue of !oid fraction. Also, !oid fraction !alues decrease with increasing flow rates for oth air and water.
Meanwhile,
N Re =
DP v0 ρ
(1 − ε ) µ where 6 and D is for air.
'or dr% packings where water flow rate was e$ui!alent to 0 ;min, pressure drop !ersus the
'igure 3.
Plot for ∆P !s G
#herefore, increasing !alues for the total pressure drop promotes increasing
#he plot for the logarithm of the ratio of total pressure drop for different li$uid flow rates and column height, H, !ersus the logarithm of gas mass !elocit% @- in l mhrft 2 is shown in figure . (olumn height is e$ui!alent to >0 cm or 2.2=C ft. 4hile, gas mass
!elocit% was computed using the e$uation
G=
v o ρ
( 1 −ε )
.
*)))) ,*0)))
12) 12,
,*))))
12 Log (DP/Z)
12(
)*0)))
12. )*)))) *))))
120 *0)))
12/
-)*0)))
12/*0 123
-,*)))) Log (G)
'igure .
Plot of ;og PH- !s ;og @- for ifferent ;i$uid 'low
#he plot on figure showed an increasing slope with an increase in water flow rate. Also, change in slope for each flow rate has a drastic change due to the occurrence of flooding. Meanwhile, pressure drop was also plotted against the gas loading factor @ f - for different li$uid loading factors ; f -. #he @f was computed through the formula =?C' s0.05 'pd-0.5 where 'sI J tDg0.5 and 'pd I)C17-*7 3p-. /n the other hand, li$uid loading factors was a!eraged for each li$uid flow rate. 'igure 5 shows the plot for
∆P
!s @ f .
)))))*)) ,0))))*)) ,)))))*)) 0))))*)) )*)) 0)))*))
,))))*))
,0)))*)) Gf
))))*))
0)))*))
'igure 5.
Plot of Pressure rop ∆P- !s @as ;oading 'actors @ f -
;oading one is the enhancement of mass transfer ut as rates were increased further, flooding occurs. #his results when gas !elocit% also ecomes the function of the li$uid holdup instead of ust a function of li$uid rate. &t is where the pressure drop increases at an accelerated rate that e!entuall% leads to flooding. #hus, in figure 5, the loading one was descried % the shaded region. Packing factors calculated for different !olumetric flow rates of water where shown on tale 3. A!erage packing factors was also otained from the calculated porosit% !oid fraction- !alues in each li$uid flow rate. 'rom the tale, it was oser!ed that packing factors increases with also an increasing li$uid flow rate.
#ale 2. Packing 'actors of ifferent ;i$uid 'low
!pd
0 1 2 3 5 C C.5 >
33C5.22 C02.3? >3=.10 132C.C0 1?50.?> 1?>>?.?C 20??5.C0 25?2.5 20?==.5=
'or dr% packings, pressure drop calculated from the rgun and
#ale 3. Pressure rop ased on "perimental ata, rgun $uation and
A%R !L*+RA,E ;min20 30 0 50 C0 >0 ?0 =0 100 110 120 130 10 150
Eperimetal lft 20. 0. 0. 0.? 1.C 2.0 2.= 3.3 3.> .5 5.3 5.> C.1 C.5
Er"u E5uatio lft 21.523 3.13?0 C.=3C C.5?0 13.023 1C.=3>3 2.11=C 2>.>?03 31.?=5 3?.=>> C.>2= 50.3>0= 5.31> 5?.3121
Robbis E5uatio in 82/ft1.?? 2.2 3.> 2.? .= .>? 5.> 5.?3 5.?= C.50 C.=? C.= C.=1 C.??
'or nonirrigated s%stems, pressure drops were influenced % man% factors. /ne of these factors is the !oidage or the percent free space in the ed. Pressure drops otained from the rgun e$uation was !er% different with the e"perimental data while pressure drops calculated from the
tale 3, oth the rgun and ;min was shown in tale . #ale .
(omparison of "perimental Pressure rop with
%&
Experimental *) /*, ,)*/ ,/*3
Roin! E"#ation 3.3)*+)3, ,)0).,*344/ ++(//*//(/ 3)3343*+(+
As6ers to )uestios
1. 4hat are the characteristics that a packing should ha!e for it to e emplo%ed in mass transfer operationK Packing characteristics are one of the significant factors in!ol!ed in packed tower design since packings are considered to e the heart of a packed tower. #hus, the properties of packings emplo%ed for mass transfer operations should ha!e+ high surface area per unit !olume, high ratio of effecti!e area to total area, high percentage of free space or !oidage, irregularit% in shape to a!oid pattern like packing, low side thrust L a function of the packing shape on tower walls for structural reasons, fa!orale li$uor distriuting $ualities, low apparent densit% and high unit strength, low cost, low pressure drop, and lastl%, durailit% )C*.
2. "plain the mechanism of gas flow through a packed ed with li$uid flowing countercurrentl%. #hrough the distriuting space elow the packings, the richgas containing the solute was allowed to flow upwards past the interstices of the packings in the packed ed column. #he packings encourage intimate contact etween the li$uid
and gas phase since it pro!ides large contact area etween the two 2- phases. #he fresh li$uid entering from the top of the tower flows countercurrentl% with the gas phase and asored the solute present in the richgas thus, lean gas lea!es the top. #he soluteenriched li$uid flows down where concentrated li$uid lea!es the ottom of the tower through the li$uid outlet.
3. ifferentiate etween static and d%namic or operating holdup. 8ow does this affect the pressure drop through a packed columnK :tatic li$uid holdup is defined as the !olume fraction of li$uid that remains in the ed after complete draining while the d%namic or operating li$uid holdup is a freedraining li$uid not contained in the particles of the packed ed and collects at the ottom of the column after a sudden shutoff of the li$uid feed )>*. ;i$uid holdup is a function of the li$uid rate onl% up to the loading region. 4hen loading region is entered, it also ecomes a function of the gas !elocit%. #he holdup uilds up as the gas flow rate is increased, there%, resulting in the reduction of free space. &n conse$uence, the pressure drop also increases at an accelerated rate and e!entuall% leads to flooding )C*.
. efine loading and channelingK @i!e the rele!ance of these two factors in packed column operation. ;oading is characteried % a mild li$uid uildup on the packing where packed column operation is fre$uentl% most economical in this loading region. #his also gi!es reasonal% high capacit% coefficient since the packing is fairl% well wetted and pressure drops are still comparati!el% low )C*. /n the other hand, channeling occurs when the fluid flowing through the packed ed finds a preferred pathN through the ed. #his effect happens when li$uid films grow thicker in some places of the packing surface while thinner in others, thus, the li$uid collects into small ri!ulets and flows along localied paths through the packing. &n low li$uid rates, the packing surface is most likel% dr% or co!ered % a stagnant film of li$uid resulting in the poor performance of large packed towers especiall% when filled with stacked packings )?*.
5. 8ow does the packing factor otained from the flooding !elocit% differ from the one estimated empiricall% with the use of the correlation of ;oo et alK Packing factor otained from the flooding !elocit% considers the point where flooding occurs. 8owe!er, packing factor estimated empiricall% from the correlation of ;oo et al is onl% dependent on the ed porosit% and does not consider flooding. #hus, packing factor !alue from the ;oo et al correlation is different from the !alue otained from flooding !elocit%. &
Co#lusio
#he !oid fractions in packed eds, pressure drops and packing factor were successfull% determined in the e"periment. A!erage !oid fraction for dr% packings was calculated as 0.3?2. /n the other hand, porosit% near all flooding point showed a smaller !alue. #hus, further decrease in porosit% results ecause li$uid holdups take up space inside the packings which then e!entuall% leads to flooding. (onse$uentl%, pressure drop was also large when li$uid holdup was oser!ed ecause gas flow could not pass through without disturance of the li$uid holdup. Meanwhile, C mm ceramic raschig rings used in calculation ha!e an effecti!e diameter of 0.22 inch with C2 !oid fraction and a dr% packing factor of 5350m. Packing factor otained for dr% packing packings e"perimentall% was 3C=C.>5ft or 1212?.m. ;astl%, it was also concluded that flooding is an important matter in packed tower applications and the appropriate t%pe of packing material was also of importance to calculate the pressure drop and flooding.
%&
Refere#es
1.
@eankoplis, (. O. 1==5-. :tage and (ontinuous @as;i$uid :eparation Processes. &n (. O. @eankoplis, Transport Processes and Unit Operations 3rd ed., pp. 5?C32-. :ingapore+ Prentice 8all &nternational.
2.
8ewitt, @. '. 2011-. Eoid 'raction. Thermopedia. doi+10.1C15AtoH.!.!oidfraction
3.
n.d.-. Eoid 'ractions in #woPhase 'lows. &n Engineering Data Book III pp. 1>1 1>33-.
.
'ahien, <. 1=?3-. Fundamentals of transport Phenomena. Gew Qork+ Mc@raw 8ill, &nc.
5.
Perr%, <., R @reen, . 1==>-. Perrys !hemical Engineers "and#ook >th ed.-. Gew Qork+ Mc@raw8ill Book(o.
C.
;e!a, M. 1=53-. To$er Packings and Packed To$er Design. /hio+ #he Jnited :tates :tone 4are (ompan%. C2=S!iewI1upSse$I>
>.
de 9lerk, A. 2003-. ;i$uid 8oldup in Packed Beds at ;ow Mass 'lu". %I!hE &ournal' ()C-, 15=>2000.
?.
Mc(ae, 4. ;., :mith, O. (., R 8arriott, P. 200C-. @as Asorption. &n 4. ;. Mc(ae, O. (. :mith, R P. 8arriott, Unit Operations of !hemical Engineering pp. 5C5C12-. Gew Qork+ Mc@raw8ill.
%%&
Appedi#es
Appedi A – Ra6 Data of Pressure Differe#e i #etimeter L%)U%D 0.0 1.0 2.0 3.0 .0 5.0 C.0 !L*+ RA,E -L.mi/ A%R !L*+ PRESSURE D%!!ERE2CE -#m/ RA,E -L.mi/
C.5
>.0
20
0.2
0.
0.
0.?
0.?
1.0
1.0
1.0
1.0
30
0.2
0.
0.C
1.0
1.2
1.2
2.0
2.2
3.0
0
0.2
0.
0.C
1.2
1.
2.2
3.
.
5.2
50
0.
0.
1.0
1.?
2.0
.0
5.2
?.0
?.2
C0
0.?
0.?
1.2
2.C
2.?
C.?
10.
11.
flood
>0
1.0
1.0
1.?
3.C
3.?
=.C
15.?
1=.
?0
1.
1.C
2.?
5.2
5.C
1.C
flood
flood
=0
1.C
2.2
3.C
>.?
?.0
1?.?
100 110 120 130 10 150
1.? 2.2 2.C 2.? 3.0 3.2
2.C 3.2 3.? .0 .C .?
5.C 5.? C.C >.2 >.C ?.0
?.C =.? 10.? 12.2 1C.0 1C.
11.2 1C. 20.2 25.2 flood
flood
Appedi B – Properties Used i Cal#ulatios Density water /*,,) lb78t( 61 2 ,3 50& 9ra$ity % 2 (* 8t7s 9ra$ity :lbm*8t;7 (* &onstant %& 2 :lb8 *s; Diameter!olumn D 2 34*0 mm )*)3(/ Densityair 5 0& lbm78t( 6% 2 43 Diameterparti!le Dp 2 / mm #is!osity air 5 )*),+/ <% 2 ,/ !P 0& 1en%t"!olumn 3) =2 !m #is!osity water 5 )*)))4 <1 2 ,/ Pa*s 0& Appedi C – Sample Cal#ulatios ρ g ∆h = 62.11017 (32.2)(0.2 / 30.48) P = g C 32.2 Pressure rop+
= 0.408
lb f ft 2
:uperficial @as Eelocit%+
20 x0.035 airflowrat e 60 = 0.22031 ft vo = = 2 x sec tionalareacolumn 3.14 79.5 s 4 304.8 (1 − ε ) ε
Eoid 'raction 'ahien and :chri!er-+ solve)
2
1.62
=
∆ PD
2
P
136 Zv0 µ
(solved using shift
