renary aseton

8/8/2019 renary aseton

http://slidepdf.com/reader/full/renary-aseton 1/7

250.

Tar Acid Extraction i n a Rotating Disc Contector

JohnV. Fisher, Johnstone S. Elackay, Rudolph J. Gleiss.

Pi tt sb ur gh Chemical Co., Pit tsb urg h 25, Pennsylvania

The ex tr ac ti on of c oa l tar acids f r m tar acid o i l s by means of caus t i csolut ions w a s invest igated i n a b in. I.D. Rotating Disc Contactor e xt ra ct io n column.This invest igat ion w a s one phase of a program directed towards the modernization oft h e c o a l

tarpocessing

plant at Pittsburgh Coke and Chemical Co. Subsequent changesi n the economic pictur e f o r co al tar der ivat ives l ed t o a cance l l a t ion of the projected

moderni z a t on.

Two feed stccks fram in de pn de nt manufacturing sources each cont ainin g

lL-16 wt.$ tar acids i n neu t r a l o i l tiere employed i n t he test series.so lu t ion was f r esh 9 Wt.% caus t i c i n water.opzratixq conditions near cap aci ty opzration.chemically indeterminate.w l e n o l s etc. Ueutral o i l is a mixdure of methyl namthalene, naphthalene, alkyl

benzene etc . The acid-oil r a t i o and the individual const i tuen t ra t i os arz functions

of th e co al and the coking conditions.lccal feed s t o c k s vhich var ia t icns were expected t o be canpounded %hen out sid e sources

were employed as c ap ac i ty o p r a t i o n was approached.th e maxinun likely t o b e encountered over aqy extended period.

The ext ract ing

i l e s e r e v e s e n t ed the projected plantThe tar a ci d i n n e u t ra l o i l system is

T a r acid is a generic name f o r a mixture of phenol, cresols,

Wide var ia ti ons were normally encountered using

The 14-1675 aci d content represented

Thz study a l s o covered the benzene-acetone-water system i n order t o provide

a referen ce base fc r t h e column pzrformance.

l?--e prformances of the FUX colunn are reported f o r bo th t e rnary systems.C er ta in a s p c t s of the r e s u l t s m u s t be vietred u i t n caution.

changes i n c o l w prfornance wi th changes i n t h e levels of the opzrat ing variableswe believed to be accurate.equi libri um phase da ta and i t s in te rp re ta ti on coupLed with th e proximity of the

o p r a t i n g l ine t o the equil ibr ium lire i n these studi es make the absolute values ofthe t ransfer s tage height f o r this system rather doubtful.

The measured relative

However, uncert aint ie s with respz ct t o the tar acid

Ternmy phase ezpilibrium d a t a have been published f o r the benzene-acetom-water s y s t e d . No da ta concerning th e tar aci d oil-aqueous ca us ti c system have been

published. The d e t a il e d developnent of the ternary gas e d i q r m f o r the tar xi& i l -

caus t i c system is not reported i n th is papzr. Holrever, sane discussion of the a p O 2 c h

and the resul t s is e s s e n t i a l t o t h e evaluation of the results of the calm t e s t s .

Exmrirnental

s p c i e s d is t r ibut ion i n the feed s tock, have not been developd t o the p i n t where theyare prac t i ca l on a semi-routine basis.

Absolute analytical techniques f o r both the tar acid content of, aril t h e

An in fr ar ed technique f o r measurirg the

(1) Briggs, Conings, Ind. Engr . Chem. 22 bll (19b3).



251.

concentraLion of the 0%g r o u g w a s developd which is bel ieved t o be f d r l y re l i ab l e .3.c tc.;aL !.;eig@ of % i d is tinen calculated by using an wbitrwf l l loleculw TJeightvhi-h q r e e s with long term plant e:tpriencz.

2

T q acids i n the caus t ic @?as@ >?ere determined by pot ent ime tr ic t i t r a t io n.:it:? X ? t o a gff of 9 in ,a lcoh ol ic solut ion.

?he $ m e boundaries i n par t and the t i e l in es were obtained by mi xi q known

i.:eiShts of tar acid oi l and 9 mt.2 caus t ic and analysins the phases for acid . . '

dis t r ibut ion. The phase bowdaries f o r a number o f feed stocks cbtained over a t v oncr,th pxiod were determined by cloud point t i t ra t ions according t o the method ofW-nar3. The p\ase bra-idary locations f o r sane f o u r feed stocks anci a synthetic feedcf $zmi i n na$kha?ene were very similar over t he t enp erat me r ans e between 30" and

>> _. Xo posi t ive t rend in s l o p or loca t ion as2 r e s u l t of COmpoSitiCfl or tem pr atu re differ enc es could be seen. These da ta suggest51?:ncrnal ir?-plant variations of feed comFosition should not affect the extraction

cs im n Frlormance s igni f icant ly .

a l l these data.

-+a-Tie l ine data scat ter ra ther badly.

A germalized phase diagram 'rias syntiiiesizzd fr.m

q.ne coliuln t e s t s '..?ere c ar ri ed out a-L a la ter date using Lwo feed stocks...xLL -̂i::ere n o t a part- of the above study. However, the gen era lize d d i q r a x was assmedts a??ly t o the ne?..! feeds.of $eRo? aid cr es ol s with re sp zct t o their reac t ions with caustic.n2;r n o t be saf e i f the concentratio n of the ca ust ic changed appreciably.

i h i s aproximatior; i s reasonable i n view of the s imi la r i ty .This approximation

The ba si c un it of th e Xotating Disc Contactor i s a c y l i d r i c a l c e l l b af fle d

The c e l l s of t he p i l o tt each end a d s t i r r e d by a cent ra l ly loca ted sol id di sc .c o l u l n had the following dimersions:

I D ir in.

3e i ch t 2 ir..-4. j r y t c f l e s b in . OD , 3 in. ID, 1/6 in . th ick

2otatir.g .disc 2 I/!! in . 5, l,% in. thick

3 e a~zrly emer.e-acetone-r;ater studies and tne tar acid nxtract ion studies verez ? z r i z d ' o i t i n a glass i;:a!lzd ,cclimn containins 2!4 ce l l s .:mxr stiidios "ere executed i n a s t a in l e s s s t ee l co lum coa ta in ing 36 c e l l s .

The later Sertene-acttone-

FezB 2.d eiflqjitnt rates m r z cont rol lzd bj r o t m z k r s . T'ne zfZ1uzF.t r a ~ e s

Thz s31 idsf3 r

-re-,i?:tatcd , d w i ~ she acid e::traction pi ci cl y m,w'ered the r c t a m t z r s m4"1;' 3ppqu;uz

2 ~ 3~ haTjc ca11s~doEe sticking.-' ,

p.a nte r f . c e x . 1 ~ Znsc.2 ad ccntrollcd. by Eeans of capacitame .prcSes.Snlil is deposition on the pobes during the tar %id ext rac t io n forced s0v.e nodi f i sa t i cn~ . c : !IZ c~ tc : t o r s:lstenr.

ireap:;n3 of t>Lznside of the gl Gs es every z - 3 ~ iwtes as f a m d t s be satisixcto.;.

~ 3 ; ~ ~bz conp:etion of t , ; ~ o1w.n strlriies a f l m s y s t e m in te r fze de tac ioord e . J , 2 ~ c p ~ ,,fLii2?..;= no tsib j e c t 5 0 the F & i z n s xsoc i a t ed wi th so l i ds Szps i t i cn .

5% z c i d stu.i ies w e j a s zd 29 conti?.uous might, xeas.mment.

As a r e s u l t t h e use o f ueigh tan<s ms adop ted .

. . -p.io sid e ncuntcd si gh t glass chambers wera in s ta ll e d with tine

l.;irc: x.r?-ppdwoad th e outsi de of th e 9 1 % ~ . t -Lhc desired levzl . A l z s r n a t z

,". coluan d Z u c i l i a q equipen*; -:rere s t a m traced th*ougi.,out.



252.

Discussion

T:he i n i t i a l d i s p r s i o n of t he a a s e s i s generally achieved witn t h aid of

a nixi ng nozzle. The in te ns iq r of in te r fa c ia l turbulerce i s pr&&Ly a m a x i s u m att n i s point .

tu r h le n ce dies out very rapidly and constant relative veloc i t i es w e approzhedwi thin a few inches up the colunn.Lhe e x t r z t i o n column a s i g n i f i c a n t f r a c t i o n of th e t ransfe r can occw over t h i s

stage.

The mean droplet size may be a mininun. The excess sechanically iniiuced

there onby a few transfer uni ts are at ta inabSe in

Coalescence of the dropleLs begirs a lmo s t inmediately after leaving the nozzle.

Performance of t he mixing system i s one ma s o n rzhy a decrease i n t h e heightof a t r ans fe r s t age i s cibtained with increasin g t o t a l t:hroughput whzn the number oftotal s tages involved i s small.

The coeffic ients of neat o r mass t ransfer pzr uni t a rea of interface havebeen shmn t o depnd ,primarily on t he s e t t l i r g r a t e of the droplets4.

V s t e m of this nature the accelerations imposed by t he s t i r r e r We a small f r a c t i o nof that due t o gravity. The rate of t he s t i r r e r i n a npzked tover , is the refor e t he gaintenance of a mininun droplet sizz spctrum. Stirring.breaks dovn the droplets t o a size spzctrun which a p pa r s t o depx-d on the power inputper u n i t ~ 0 1 ~ ~ 5 .

ra t e t han the surface area inscreases r r i t i l i x r e a s i w s t i r r e r s p ed 4 .quantity tra-nsfzrred across the interface p r unit voluTe of s t i r r e d ves sel incrzases

wi th s t i r r z r s p e d .

i n s t i r re d

c o l m n o r packit% i n a

The coefzicient of mass t r a n s f e r pzr c n i t a r ea f a l l s off a t a s1o:rerTnat is, the

The importame of inLe rnal mixing with in, a d ne'.*' urface gemra t ion on, th edrople t s as a result of coa1esce;lce- ard breakdown has not been resolved.

In mav tu0 phase systems, mixing of the &ases a t condi t ions othzr than

equilibrium re su lt s i n spontaneous in te rf ac ia l turbulence. Spontaneous emulsi ficzti onhas been &served i n a few s'jstens.in te r fzce prduce viscos i ty, densi ty, and surface tansion gradients .

expan6s and contracts 13cal ly.curren ts due t o densi ty gradien ts t o produce convect ion ce l l s a t the interface6, ?.

Tlardom concentration f l u c t u a t i o n s along thz

This movement couples with gravity inducea comectionThe surface

Spontaneous in tz rf ac ia l turbu lerce could be detected undzr ce rt ai n comii t i3nswith bo th of the systems under study.

turbulerce o n v lihen t i e conditions were such t h a t the local solution density could begreate r th an the l iquor densi t y below i t .,phenol in naph thal ene- caus tic system is not certain.

betireen 0.03 and 0.1 in. diameter.

Tine zcetone-benzene-vater s j s t e n eyhi bi ted

Yhether th i s res t r i c t i on appl ies t o theThe ce l l s izes appeared to l i e

The mean drop size observed with th e be,nzene-acetone-water system i n th ecolman rras not more than 3 tililzs the si ze of the ce l l s observed above at :kitA t s&h lox drop s i ze t o zonvec ti on c e l l size ratios, dzvslopeni; 02nterfaces.

these convection cells mizht not be possible.

Tnz preserce of s urf ace active agents a t t he i n t e r f x e does not i n h ib i tn m a l s o le cu l ar d if fu si on . The in tr in si c mass tran sfe r coef fici ent involves both

(b) ,;alderbark, P.H., I'loo-Young, X.B., Clem. Engr. Sci. 14 34 (1961).( 5 ) Reman, G.H., OlneLr, i'La0,Chem. Engr. e o g . 2 h1 fiss).(6) S t e r l i r g , C.V., Scriven, L.E., A. I . Ch. E , Journal 2 l k (1959).

(7) Orel l , A,, k;estk=ter, J. Id., Chem. 9nSr. Sci. &, 137 (1961) .



253.

t u rbu l en t a d molecular'diffusion.

t h e subclass of surfsce active agents vhich rigidize the intt2rface4fB.

Tnis coeff ic iant has been found t o be reduced by

?!ot generally realizec! i s tk, a b i l i t y o f t races of sol id s to . r ig idiz e aninte rface , . Dis t i l led r at er exposeti t o tne atmos&,ere for a few minutes develops anarked surface r igidi ty as a r e s u l t of dust depositio$.

the sane phenomena r ii th ?. nmber o f orrjanic COmFOUndS.

O m of th e author s has noted

The benzene-Qetcne-water sy st en i s bel ieved t o be fre e of s . i rface act iveHowever, the ex te nt of the- fc rn at i on sr" sol ide n t s ard Should be free of sol ids.

prec ipi ta tes at the interface during extract ion of ta r acids i s such tha t majorrecluctiom i n the mass transfer coefficient could nave occurred.

The ex,rerinental vaLues of the tr an sf er stag e he igh t are shown on Figwe 1

and 3 f o r the benzene-ketone-uater system and on Figures 2 and b f o r t h e tar acid i noil-caustic system.

Tlie o d y s i sn i f i zan t p rocess va r i h l e appa -s t o be the r o t o r s p e d b oth f o r

the De?~ene-~e;one-~:aterand %he ta r as id i n oil-caustic system.

neit her ;he ci re cc io n of t r ans fe r 3f Lhe ace tom , th e $=e t.hich was d i s p r s e d , n c rthe cnrcughpt had 5iGnificant effects on the nmber of t ransfer s tages .point did horrever depr?d on the pnase xhich uas dispersed, Fic;ae 5.':as wserved with t he tar acid i n oi l -czus tic sys ten c v e r the l imi ted r w e s studied.

Fo r t h e f i r s t s ys ten

The f l o d i qThe sane pa t t e rn

These data suc,gest that such phenonena x n t e r f x i a l t l l r s u l e r c e, wfacer i gi di ty induced by scl ids xc*Jmulat icna t th e in te r face , and the di r ec t i on of %asstr an sf er zre not normally inp ort znt variables.a p p a r s t o de,pzzd p i m a x i 1 : y an t h e drop l e t s ize s p c t r m a tt ai na bl e i n th e sysr;em.Eoi.ever, capacicy eoes depend on t;le particular phase which i s dispersed.

The volumetric m a s s t rans fer coeff i c ie nt

The reader i,s again cautioned against using th e HTS valces ebtained for theta r ac id i n oi l -caus tic system f o r sondi t ions o t h e r than repx ted . The operating lines

tend t o pinch th e equilibrium curve, Figure Bo. L, and th e exact l oc at io n of t he

equilibrium curvs is uncertair.. Depndif ig on the individua l in te rp e t a t io n o-C the

dzta, a 507; difference in Yiie numder o f transfer stages m q be obtairod.

l m a t i o n of th e curve, while i t changes the numerical values, does not cnange thz

observed dependence of the nunbers of t'ne opzr atin g var iab les .

However, th e

of tne surp- isin g featlmes of t he R E o l u w pr fornance I s the absence

of so l i ds bui ld -up x i t n i n t he s t i r r ed z one s ,s lu r r i e s . I n General, th e RotatinG Disc Co nta cto r de7;elops a high t ransfer rate per

up.it volme , ~ z d s insensikive t o shall u p s e t s i n th e feed system, a& can t o l e ra t e

s ign i f i can t quan t i t i e s of sol id s wi thin the CO'LWLI? p o p r .

The nwhine m i & t be ,capable of h ad li -n g



254.

R D C EXTRACTION COLUMN PERFORMANCE24-36 CELLS 4 inl .D. 2 i n . H . ST IRRER D ISC 294 in.D

Q)

J

3wv T R A N S F E R OF A C ET O N E B E T W E E N

F I G URE No.1

B E N Z E N E A N D W A T E R

1 I400. 200 600

0 S T IR R E R S P EE D ( r p m l0 LEGENDz

1

D A C E T O N E F R O M W A T E R ( Z Q C E L L S )

A A C E T O N E F RO M B E N Z E N E (24 C E L L S )0 A C E T O NE F R O M W A T E R (36 E L L S )

FIGURE No,3B E N Z E N E - A C ET O N E - W A T E R

E Q U l L l E R l U M D A T A-s3 503

-g4 0

W

w

2

wZ

w0

a

2 30

m

-20

g I O

IO 20 30 40 50 60

f A C E TO N E I N W A T E R (WT.Vo)u

L

x FIGURE N0.5

F L O O D I N G L I M I T S

4,-

iIGURE No.2

T R A N S F E R OF T A R A C I D S I N T O

C A U S T I C)

-1J

S T I R R E RO0 S P E E D ( r p m )00 3002

L E G E N D

a 6000 Ih/hr, sq. tt.

o 4500 lb./nr, 8q.tt.

A 3000 Ib. /hr, rq. f t.

F lG URE NoA

NEUTRAL OIL-TAR ACIDS- gut . % C4USTIC

S O L U T I O N E Q U I L I B R I U M D AT A

TA R ACIDS IN N E U T R A L O IL ( w t yo)

L E G E N D

a D T A R A C I U S D I S P E R S E D P H A S E

Q A B E N Z E N E D I S P E R S E D P H A S E

0 W A T E R D I S P ERS ED P H A S E

0 200 400 600

I- S T I R R E R SPEED ( r p m )



TAaLE I

Extraction of Tar Acids with 9 wt,:z NaCH i n i fa te r.r a t i o 1.9 t o 2.1. T a r Acid C i l d i sp x se d phase .

Feed t o Solvent

C o l u m n RotorTempratwe S p e d

rFn,.L

86 0

67 2587 50

a5 ?586 123at3 15592 0

Tota l TarThoughput Feed

Lt&r., sq.ft. I&.,%

19.t20.721.c2 1

20.2

21.32c.5

22.c

33.022.;

2319 9

19.>'13.851:3

20.9

20.320.922.2

9.7

9.19.79.79.29.2

9 .2?.2

9 .s9.99 .6a .98.67.1:9.23 . :

4.19 .2

9.5

Nc. ofTransferStages

2.c2.32.52.e

3.53 .C

5.L'b.52.72.13.22.2

2,ir.72.1

2.92.5

4.0

b.3

c,sL;nthetic cf 5 i l t u n S t i l l T a r Acid O i l and recovered T a r Acid f rom plant.



256.

.Cor>tinucusPhase

3enzene

(24 e l l s )

Phase Throughput

Be mene Vater-Acetone

Ib.,kc., sq. l t D

ketone ContentFeed %tract;i.ster- gemere-.keto?-e

Acetone!;;t,5 . 2.Tt.S

. .

hl.941.3 u.9

39.6 Ll.8L0.3 LC.341.1 kl-3

aemene- !.'&er-

Acetone Accztom

37.8 A. $36.5 9 .>35.5 12

35.5 15

Vater Benzene-Acetcne -Acetone

30.0 33

3 l . C 33.029.0 35.031.0 35.0

30.0 36.529.0 34.529.0 3b.5

30.0 33.7

Raffinate7:a'ater

.ker;cne!-L C

15.1

19.511.5lC.310.0

3enzene-kzfone

0.11.32.72.7

.. .,>

!iater

-Acetone

12.3

12.513.510.7

12.0

12.010.0

8.4

rransler

Stages

X O .

21.5

32.5

2.9

2.32

2.53

2.1

2.52 .0

3.52.8

3 .23 .s3.7

renary aseton

Documents