CHEMICAL
l M i T A l L U R Q I C A l
_____________ ENGINEERING_____________
ESTABLISHED 1902 S. D . KIRKPATRICK. E di to r
JUNE, 1942
S T A N D BY FOR F U R T H E R O R D E R S !
W e f i n d in creasin g uneasiness am ong you n g
chem ical engineers who have difficulty in seeing how th ey a re a c tu a lly h e lp in g in the w a r effort. Some of them feel th e y should th ro w overboard th e ir te c h n i
cal a n d p rofessional tr a in in g in o rd e r to e n te r the arm ed forces. A s till la rg e r g ro u p w ould like to go to w ork fo r th e g o v ern m en t in some chem ical e n g in e e rin g cap acity , o r to tr a n s f e r to a m ore essen
tia l ty p e of in d u s tria l em ploym ent, y et to do so th e y feel th e y m ay be accused of m erely seeking to dodge th e d r a f t. N o th in g could be f a r th e r fro m th e tru th .
C hem ical engineers in a n y k in d of w a r work—
a n d th a t in clu d es a g re a t m an y activ ities beside m ak in g m u n itio n s— a re p ro b ab ly serv in g th e N ation b e tte r in th e ir p re s e n t jobs th a n th ey would in alm ost a n y c a p a c ity w ith th e A rm y or the N avy.
C hem ical engineers w o rk in g in o th er th a n w a r in d u s trie s sh o u ld co n tin u e w here th e y a re in p re p a ra tio n fo r th e tim e w hen g re a te r num b ers of ex p erien ced m en w ill be needed an d cannot be o b tain ed ex cep t b y tr a n s f e r from existing in d u s
tries. J u s t as the d is tillin g in d u s try has been com
p letely converted to w a r p ro d u ctio n , so th ere will be a n in ev itab le sw itch-over to w a r a c tiv ity on the p a r t of m ost of o u r o th e r chem ical an d process in d u strie s. M eanw hile th e m en w orking in these p la n ts fo rm th e o n ly re a l chem ical engineering reserve we have in th is co u n try . T he)' should be k e p t in active tr a in in g in th e ir chosen profession.
T h is m eans, of course, th a t th e in d iv id u a l should n o t on ly w ork as h a rd as he can a t his re g u la r job, b u t should t r y activ ely to im prove him self in other lines o f chem ical en g in eerin g w hich he, him self, can recognize as of in creasin g im p o rtan ce in the w ar effort.
D ean F r a n k C. W h itm o re of P e n n S tate, who has become a p a s t m a ste r in e d u c a tin g local d r a f t b oards, likes to call th e ir a tte n tio n to ju s t two figures, viz., 27 m illions an d o n e-q u a rte r m illion.
T he first re fe rs to the to ta l n u m b er of m en up to 35 y e a rs of age who have re g istered in the Selective
S ervice System . The second n u m b er is slig h tly g re a te r th a n the p re se n t to tal en ro llm e n t o f tlfe N atio n al R oster of Scientific an d Specialized p e r sonnel, w hich includes chem ists, physicists, chem ical engineers, bacteriologists, geologists a n d o th er scientifically tra in e d m en from th e g ra d u a tin g classes of 1942 u p to the age of ab o u t 80 y ears. I f we a re to lose this w ar, he holds th a t it w ill n o t be because of th e 27 m illions b u t because th e q u a rte r m illion is e ith e r too sm all or we have n o t u sed it as effectively as we should.
O f the q u a rte r m illion only ab o u t 60,000 are chem ists a n d chemical engineers. The fa c t th a t even d u rin g the w o rst days of the d epression only ab o u t a th o u san d of these w ere ever unem ployed would seem to prove th a t they a re needed to c a rry on th e n o rm al peace-tim e o p eratio n s of th e co u n try . B u t the trem endous expansion of chem ical a n d m etallu rg ical in d u strie s a n d the in ten siv e a p p lic a tio n of w a r research an d developm ent now call fo r a m uch la rg e r p ro p o rtio n of te ch n ically tra in e d m en. W h en the new sy n th e tic ru b b e r, m agnesium , alum inum an d like m u n itio n s p la n ts get into f u ll sw ing in 1943, th e n u m b er of chem ical engineers needed fo r su p erv isio n a n d co n tro l will be f a r in excess of the available su p p ly , even w ith th e com
plete diversion of those still engaged in fields n o t d ire c tly connected w ith the w a r p ro g ram . So it is v ita lly im p o rta n t th a t we have no m ore v o lu n ta ry desertions or a c tu a l ro bbing of the chem ical e n g in eerin g ran k s, m o tiv ated by m isg u id ed p a tr io tism on th e p a r t of local d r a f t b o a rd s o r th e r e c r u it
in g officex-s of the A rm y a n d N avy.
F o rtu n a te ly , th e re seems to be no such m iscon
ception or confusion in th is m a tte r in th e a ttitu d e of G eneral Ile rs h e y a n d his ra n k in g associates in Selective Service H e a d q u a rte rs in W a sh in g to n . Likew ise th e p u b lic u tte ra n c e s of A d m in istra to r P a u l V. M cN utt, ch a irm a n of th e W a r M anpow er Commission seem to hold prom ise of a f a r b e tte r ap p re c ia tio n o f the key im p o rta n c e of the te c h n i
cally tra in e d m en on whom th e w a r m achine
d ep en d s fo r its design, co n stru c tio n a n d o p eratio n . A n d , finally, th e re is the tin y , b u t g ra d u a lly grow in g id ea o f a technological hig h com m and to m u ste r all o f o u r scientific a n d en g in e e rin g resources fo r w ar roles of technology. So, in the w elter of con
fu sin g counsel, we u rg e th a t chem ical engineers should sta n d by tlie ir gu n s a w hile longer. New o rd ers o f a tta c k a re in the m aking.
PATENTS AND RE-SALE PRICES
Pa t e n t s a re n o t v e t a “ lost a r t ” b u t, u n d e r re c e n t c o u rt decisions, th e y seem to have less value th a n fo rm e rly . C e rta in ly th e S u p rem e C o u rt in. a case decided M ay 11 m akes it clear th a t a p a te n t does no t in a n y w ay ex ten d a rig h t to co n tro l re-sale prices.
The case in q uestion h ad to do w ith m u ltifo cal eye-glass lenses. B u t it m ig h t ju s t as well have been sQme in d u s tria l chem ical. T he p ro d u c e r sold his p a te n te d p ro d u c t to d is trib u to rs u n d e r c o n tracts w hich a tte m p e d to do two th in g s : first, the c o n tra c t gave a license to u tiliz e a p a te n te d m ethod of p ro c essin g : second, th e c o n tra c t p rescrib ed th e p ric e a t w hich th e finished p ro d u c t should be sold to the w e a re r of the eyeglasses.
The S u p rem e C o u rt of th e U n ited S ta te s in th is ease says th a t b y sale of th e a rtic le m ade u n d e r a p a te n t, the ow ner of th e p a te n t c learly h as ex
hau sted his m onopoly. T he c o u rt re p e a ts w h a t it considers a long-tim e po licy fixed by all U n ite d S ta te s co u rts, th a t the r ig h t to co n tro l stops w hen th e ow nership o f th e p a te n te d a rtic le passes to the custom er.
Those who seek to m a in ta in re-sale p rices of th e ir tra d e -m a rk ed an d n a tio n a lly ad v e rtise d goods m ust, th e re fo re , fa ll back on th e M iller-T y d in g s A ct as th e ir sole a u th o rity in fe d e ra l sta tu te s. W h en th e y ca n n o t c learly m eet the conditions of th a t law , th e y w ill be re g a rd e d by th e co u rts as v io la tin g a n ti
t r u s t s ta tu te s w henever th ey u n d e rta k e to co n tro l re-sale prices. T h ere seem s to be no hope fo r excep
tio n fro m th is g e n eralizatio n in the C o u r t’s finding, re g a rd le ss of w h eth er th e p a te n t be 011 a chem ical p ro d u c t, a device, or a process of m ak in g eith er.
COOPERATION OR COLLUSION?
Be t w e e n p a trio tic co operation a n d illeg al collusion, th e lin e is v e ry vague. T h is is u n fo rtu n a te . M any w o rth w h ile efforts th a t w ould save scarce goods or services are n o t being m ade because of th is d o ubt in the m inds of th o u g h tfu l in d u stria lists.
I t was hoped th a t th e P r e s id e n t’s ad m o n itio n to A ssista n t A tto rn e y -G e n e ral A rn o ld w ould re s u lt in th e la y in g aside fo r the d u ra tio n of zealous p ro secu tio n s w hich tr y to m ake crim in al acts o u t of every cooperative g e stu re o f in d u s try . B u t th is a r d e n t tr u s t b u s te r is n o t to be d e te rre d , even by th e P r e s id e n t’s im plied w arn in g . I n d u s tr y m ust, th e re fo re , fre q u e n tly tak e a chance 0 11 a p p a re n t vio latio n o f th e law in o rd e r to get th in g s done.
T h ere re m a in only a few a c tiv ities w hich M r.
A rn o ld does n o t question. The w ork of technical com m ittees, the fo rm u la tio n o f specifications, a n d m ost of th e ty p e s of c o n stru ctiv e trad e-asso ciatio n activ itie s are ra re ly a tta c k e d . N o t so, how ever, w ith m a n y o th e r h ig h ly d esirab le efforts.
R ig h t now th e chem ical in d u s try m u st, d esp ite th is difficulty, tr y to find some w ay in w hich to cooperate in h a n d lin g goods in tra n s p o rt. The sh o rta g e of tr a n s p o rta tio n is so serious th a t th e in d u s tr y m a y have to tak e th e chance of in c u rrin g M r. A r n o ld ’s w ra th in o rd e r to move goods th a t are ab so lu tely v ital to the w ar effort.
S uch cooperation m ay o fte n become p racticab le, a n d leg ally defensible, if care is ta k e n re g a rd in g neg o tiatio n s a n d a rra n g e m en ts. I n p la n n in g tr a n s p o rta tio n , ev ery step should be ta k e n in close coop
e ra tio n w ith an d w ith th e official know ledge of the Office of D efense T ra n s p o rta tio n . O.D.T.-Boss E a stm a n m u st be asked to f u rn is h te c h n ical aid, both to p ro te c t th e in d u strie s a n d to see to i t th a t each in d u s tria l effort harm o n izes w ith th e com pre
hensive n a tio n a l p ro g ram . A n d te n ta tiv e a g ree
m en ts so reached can th e n be su b m itte d w ith some assu ran ce to the D e p a rtm e n t of J u s tic e fo r review a n d criticism .
S im ila r a rra n g e m e n ts w ith resp ect to th e sh a rin g of ra w m a te ria ls m u st be m ade in cooperation w ith th e W .P .B . Some g o v ern m en tal agency m u s t be d ra w n in e a rly in a ll cooperative efforts. E v e n th e n M r. A rn o ld m ay n o t be h a p p y . B u t we s till believe, an d c e rta in ly hope, th a t the F e d e ra l C o u rts w ill not sh are M r. A r n o ld ’s zeal in u n d e rta k in g to m ake every cooperative effort a n illeg al th in g p er se.
MANPOWER CONSERVATION
In t h e n a tio n ’s d riv e fo r w a r p ro d u c tio n , we som etim es th in k of tim e a n d m a te ria l as th e only irrep laceab le fa c to rs th a t lim it th e o u tp u t of o u r m achines. W e lose sig h t of th e fa c t t h a t la s t y e a r a ris in g toll of accid en ts claim ed th e lives of 101,500 persons, p e rm a n e n tly d isab led a n o th e r 350,000 a n d in ju re d 3,750,000. I n term s of p ro d u c tio n , in d u s
tr y la s t y e a r lost, fro m accid en ts alone, 460,000,000 m an -d ay s of work. T h a t is enough tim e a n d en erg y to have b u ilt tw ice as m an y b a ttle sh ip s as a re to be fo u n d in th e A m erican an d B ritis h navies com bined. To c u t th a t toll by 20 p e rc e n t, w hich is w ell w ith in th e experience a n d e x p e ctan cy of a well o rd ered sa fe ty p ro g ram , w ould b u y u s te n b a ttle sh ip s of th e “ N o rth C a ro lin a ” class, 2,100 F ly in g F o rtre ss, or 600 h eav y tan k s.
No w onder D o n ald N elson has g o tte n sq u a re ly b eh in d P re s id e n t R oosevelt’s p ro c la m a tio n callin g on th e officers a n d d ire c to rs of th e N a tio n a l S a fe ty C ouncil “ to m obilize its n atio n -w id e resources in le ad in g a co ncerted a n d intensified cam p aig n a g a in st accid en ts a n d to call u p o n every citizen in p u b lic a n d p riv a te c a p a c ity to e n list in th is cam p a ig n a n d do h is p a r t in p re v e n tin g w astag e of h u m an a n d m a te ria l resources of th e n a tio n th ro u g h a c c id e n ts.” No w onder th a t a g ro u p of th e n a tio n ’s lead in g in d u s tria lis ts , u n d e r th e ch airm an -
74—tf . JUXI-: . CHEMICAL & .METALLURGICAL EN G IN EER IN G
6— 79
S o m e o l th e b e d s of b a u x ite lie c lo s e e n o u g h to th e s u r fa c e to b e m in e d b y th e o p e n p it m eth o d
l a r g e filter p r e s s e s sim ila r to th is o n e w ill b e u s e d for r e m o v in g th e im p u rities from th e ore
In la r g e ro ta tin g k iln s th e a lu m in u m tr ih y d ra te w i l l b e h e a te d w h ite hot to d riv e off c h e m ic a lly c o m b in e d w a te r
T h e a lu m in a w ill b e s h ip p e d to a n oth er p la n t n e a r b y w h e r e it w ill b e
r e d u c e d to m e ta llic a lu m in u m
CHEM ICAL & METALLURGICAL EN G IN EERIN G • J U N E 191,2 •
T h is p ilo t p la n t s o lv e d th e p r o b le m of r e c o v e r in g su lp h u r from so u r n a tu r a l g a s a n d c o n v e r t in g th e c o n ta m in a tin g h y d r o g e n s u lp h id e from a n u is a n c e to a u s e f u l c o m m e r c ia l p ro d u ct
Sulphur from Arkansas Sour Gas
JAMES A. LEE
M a n a g i n g E d ito r o f C h e m . & M e t .--- C h e m . & Met. I N T E R P R E T A T I O N ■
A m ethod h a s b e e n su c c e ssfu lly d e v e lo p e d for the rem oval of sulphur from the sour g a s of Southw est A rk a n sa s w h ich m a y m ak e a v a ila b le for the first tim e in the state a source of ele m en ta l sulphur io r the production of sulphuric a cid a n d other ch e m ic a ls. B ased on e stim a te s of g a s rese rv es the total m a y b e 1,500,000 ton s.— Edi t or
S
e v e r a l g a s f i e l d s have recently been discovered in Southwest A rkansas. They are located west of th e M agnolia oil fields and south of th e fields in Buckner and Louisville, .and almost midway between the Ro- d essa developm ent in Southw est A rk an sa s and N orthw est Louisiana and the Sm ackover fields in A rkansas.
T h e fields contain rich deposits of petroleum condensate and n atu ra l gas.
These enormous reserves of gas give the state a source of cheap fuel which Gov. H om er M. Adkins is using w ith much success in his cam p aig n to a ttra c t w ar industries and to industrialize the state by processing A rk an sas’ n a tu ra l resources in A r
kansas. They also have another in terestin g fea tu re . They contain an unusually large hydrogen sulphide
content which through the efforts of the Texas Gulf S u lp h u r Co. has been dem onstrated m ay be converted into a su p p ly of elemental su lp h u r fo r the chemical industry of the state.
A t present the gas acts as the lif t
ing vehicle in the production o f oil or distillate, b u t it is being wasted when it reaches the ground surface except fo r a small portio n which is used as fuel in well drilling.
The distillate th a t is delivered a t the well mouth along with gas in the B ig Creek and M cKam ie fields is in gaseous form in the producing hori
zon, an d as the p ressure in the p ro ducing horizon is about to decline, there will be progressive condensation and absorption o f the distillate in the producing horizon, the am ount being dependent on the reduction in pressure with a corresponding loss
in ultim ate recovery. I t is ap p a re n t, therefore, th a t reasonably uniform w ithdraw al of distillate and gas is im perative in these fields, and it is p referab le th a t this w ithdraw al be a t a ra te somewhat less than the w ater drive to insure the satisfactory and maximum production of distillate.
These gas fields, because they form a substantial addition to the known gas resources of this section of the country, are o f real interest as the source of cheap fuel. The gas, how
ever, is a byproduct of the produc
tion o f oil or distillate. To prevent the w asting o f this valuable byprod
uct, the state of A rkansas passed an oil and gas conservation law known as Act 105 o f the 52nd G eneral As
sembly under which the A rkansas Oil and Gas Commission has adopted regulations to prevent such waste and to obtain the greatest possible re
covery of its n a tu ra l resources and to conserve gas f o r its m ore w orth
while uses.
W hile petroleum condensate (60 deg. A .P .I.) can best be recovered by m echanical se p arato rs erected a t the individual wells, n a tu ra l gasoline and butane can be recovered through the in stallation o f a central trea tin g p la n t to serve the entire field. The presence o f hydrogen sulphide in the
■ SO— 6 J U N E 191,2 • CHEMICAL & METALLURGICAL E N G IN EER IN G
gas prevents the m arketing of the gas and in tu rn prevents the production o f petroleum condensate, n a tu ra l gas
oline and other hydrocarbons.
The hydrogen sulphide content of the gas varies from 30 to 4,500 grains p e r 100 cu.ft. Such gas to be useful as fu el either f o r the industrial p u r
poses o r f o r domestic uses m ust be purified. Gas containing 360 grains o f hydrogen sulphide p e r 100 cu.ft.
m ay not be objectionable when used directly as boiler fuel gas. H igher concentrations o f hydrogen sulphide are objectionable because they con
stitu te a health hazard and also may cause severe corrosion of equipment.
M ore than 1$ g rains o f hydrogen sulphide p e r 100 cu.ft. of gas is not perm itted in domestic gas because of its corrosive effect.
I t was suggested th a t a recycling operation m ight be set up by means of which pressure could be m ain
tained and the distillate removed which would convert the fields into oil fields. A t a 're c e n t hearing, before the A rkansas Oil and Gas Commis
sion, it was pointed out by operators and independent engineers th a t this program would be too costly as the su lp h u r would first have to be re
moved from the gas which involves reduction to 600 lb. pressure and then recom pressing to 4,000 lb. fo r injec
tion into the sand.
Because o f the extreme “ sourness”
of the gas, m any technical men were skeptical o f the ability of any purifi
cation process to sweeten it. The G irdler Corp. dispelled all doubts on this score by installing a p ilo t plan t in the M cKam ie field and demon
strated th a t the gas could be purified com pletely and economically.
The G irdler process will be used in two large purification plants, one fo r the M cKam ie Gas Cleaning Co.
and the other fo r the A rkansas F uel Oil Co. The form er will be located in the M cKam ie field and the la tte r n ea r K erlin to tre a t gas from the
M acedonia and Dorcheat fields. The combined capacity of these plants will be nearly 100,000,000 cu.ft. of gas p er day, p a r t o f which will be used to provide fuel for the two alu
minum p la n ts under construction else
where in the state.
The M cKamie Gas Cleaning Co.
was incorporated under the laws of A rkansas, on Nov. 12, 1941, fo r the purpose o f constructing a n a tu ra l gas treatin g p la n t having a capacity of 25,000,000 cu.ft. of gas a day. The treatin g p la n t will remove restric
tions upon the production of v alu able petroleum condensate. Unless the treatin g p la n t is erected the p ro duction of condensate will be cur
tailed. The p la n t will produce n a t
ural gas fo r the aluminum and pos
sibly other w ar p la n ts in the state.
I t will also produce n atu ra l gasoline, butanes, and hydrogen sulphide.
Briefly, the gas will be processed as follows : The hydrocarbons exist in the wells in a homogeneous v apor phase a t pressures above 4,000 lb. p er sq.in. The gas is brought to the su r
face and as the pressure is reduced to 800 lb. p er sq.in. gage retrograde continuous condensation of heavier hydrocarbons takes place. I t is then separated from the distillate a t this pressure. Some operators fu rth e r disengage the gas in a second separa
to r a t a lower pressure. The raw gas a t sep arato r pressure o r upon re
compression passes first to the Girbo- tol p la n t where it is purified by a w ater solution of monoethanolamine.
Gas originally containing as high as 5,000 grains of IL S p e r 100 eu. ft.
will be treated and the IL S content reduced to less than 0.1 g rain p e r 100 cu. ft. so th a t it will meet the N ational B ureau of S tandards lead acetate test fo r sweet gas. A fter purification it flows from the Girbotol p lan t to a gasoline recovery p la n t where n atu ral gasoline and butanes are separated. The gas then passes through a dehydration u n it which re-
SULPHUR IN A R K A N S A S SO UR G A S FIELDS
D aily Production Reserves* .---*---»
» *--- % Oil Gas
Nam e Oil Gas (bbl.) (cu. ft.)
A tlanta... 6 10 2 ,8 7 7 4,6 0 3 ,0 0 0 B ig Creek... 3 67 241 6,0 2 5 ,0 0 0 Buckner... 8 1 2,5 0 4 676,000 Dorcheat... 18 144 1,988 2 1,027,000 M agnolia... 161 200 18,948 16,845,000 M acedonia... 12 184 900 12,150,000 M cK am ie... 37 427 3 ,2 0 0 22,0 8 0 ,0 0 0 M t. H o lly ... 5 10 400 1,600,000 Jones P ool... 35 50 13,500 2 4 ,300,000
Reynolds P ool 3 5 2,4 2 4 2 ,1 3 3 ,0 0 0
V illage... 3 10 1,191 5,7 1 7 ,0 0 0
♦Oil in millions of barrels; gas in billions of cubic feet-
Hi S per 100 cu. ft.
grains
Possible S. daily (in gr.
tons)
Total sulphur in gas reserves (in. gr. tons) 1,800
335 167 2,400 895 1,900 4,5 0 0 374 30 865 374
5.2 6 1.28 0.7 5 32.
9.5 5 14.6 6 3 .4 0 .3 8 0 .46 1.2 1.36
11,480 14,310 106 220,400 114,200 223,000 1,225,000 2.385
956 2,758 2.385
moves sufficient w ater v apor to p re vent difficulties in transm ission due to the form ation and accum ulation of hydrocarbon hydrates. The de
hydration unit employs a solution of am ines and glycols which is very ef
fective fo r rem oving w ater vapor.
The effiuent from the Girbotol pilot p la n t which contains on an average o f 58 percent by volume of hydrogen sulphide, 41 percent o f carbon dioxide and 1 percent o f hydrocarbons (m e
thane) on the d ry basis passes to a second pilot p la n t erected by the Susearch Corp., subsidiary o f the Texas G ulf S ulphur Co., fo r the con
version of hydrogen sulphidfi to ele
m ental sulphur. This pilot p la n t was located on the Cornelius T ank F arm of the C arter Oil Co. n ear McKamie, L afayette County. The design and construction of the p la n t are the re- • suits o f experim ents which had been under way fo r some time. In con
nection with the pilot p lan t, there was also a laboratory fo r control p u r poses. The p rim a ry function o f the p la n t was to solve by experim enta
tion and research the problem of re covering sulphur from sour gas and conversion o f the contam inating h y drogen sulphide from a nuisance to a useful commercial product. The diagram of the p la n t shown on the opposite page is based 011 the reac
tion :
2 IL S + SO» = 2 1LO + 3S.
About 25,000 cu.ft. p e r day of gas are received from the Girbotol plant.
One-third of this gas, a fte r being mixed with the p ro p e r prop o rtio n of a ir is burned, the combustion of the hydrogen sulphide taking place in a
Hi*S is r e m o v e d from n a tu r a l g a s in th is G irb otol p ilot p la n t a n d s e n t to S u se a r c h p ilo t p la n t for c o n v e r s io n in to su lp h u r
CHEM ICAL & M ETALLURGICAL EN G IN EER IN G J U N E 19J, 2
specially designed burner. The pro d ucts o f combustion are then cooled to a selected tem perature in a chamber provided w ith an explosion vent, and are next mixed with the rem aining tw o-thirds o f the original hydrogen sulphide gas stream and caused to enter the bottom of the catalyst cham
ber, which is packed w ith a catalyst o f special composition. The gases en ter the catalyst cham ber a t about 450 deg. F . H ere the hydrogen sul
phide reacts with the su lp h u r dioxide to form su lp h u r and w ater. The gases leaving the cataly st cham ber are im
m ediately cooled and forced through the condensing system w here the sul
p h u r is draw n off in a molten state.
As a resu lt o f this small scale o p eration it is proposed when construct
ing a commercial installation to burn about one-third o f the available IL S acid gas under a small w ater tube boiler with induced d r a ft fan to su p ply a ir and give necessary pressure fo r o p erating the system. I t may be necessary to install a cooler o f some type a t outlet o f the fa n to lower the tem perature of the gases fu rth e r than can be done in the boiler, if steam pressure is to be k ep t up, before be
ing introduced into the cataly st since it is possible th a t a fte r a period of operation, the tem perature o f mixed gases may have to be as low as 250 deg. F . in order to m aintain a correct tem perature in the catalyst bed. I t is also possible th a t rad iatio n may take care o f this need—a g rea t deal depending on the final design. I t is also ju s t as true th a t provision m ay have to be made to get some hotter gases from the boiler than would be expected of norm al operation of the boiler. The burned gases would dis
charge from the fan through a 30 or 36-in. duct into a catalyst cham
ber about 15 ft. square and 10 or 12 ft. high. Secondary gas should be introduced into the duct ju st ahead of the catalyst chamber. Duct and
chamber are to be lined and insulated.
Provision is to be m ade fo r draining possible su lp h u r from the catalyst.
A num ber o f vertical fire tube boilers with two and one-half or three inch tubes are to be used as condensers.
The first condenser boiler could be used as a cooler and the steam gen
erated used to m aintain p ro p e r tem p era tu re on the other condenser. A scheme of p ip in g m ay be worked out as condensers. The tail gases can be discharged to atm osphere through a steam coil lined stack some 30 or 36 in. in diam eter and 40 or 50 ft.
high.
Results of the p ilo t p la n t work dem
onstrated th a t more than 80 percent o f the su lp h u r in the hydrogen sul
phide can be easily and efficiently re
covered in the form of brim stone. All o f the d ata collected, however, have not been thoroughly appraised, hence it is difficult to estim ate a t this time the cost of converting hydrogen sul
phide to sulphur. W hether o r not it will be feasible to install a large sulp h u r recovery p la n t in connection w ith each of th e gas cleaning p la n ts now p rojected can only be determ ined by a more careful survey of all the facto rs concerned in the development o f the n atu ra l gas resources o f A r
kansas.
EC O N O M IC S
B yproduct hydrogen sulphide has been available in several localities for m any years. I t has been estim ated th at 35,000 tons o f su lp h u r in the form o f hydrogen sulphide was be
in g removed from fuel gases; how
ever, o f this q u an tity only three- qu arters was converted into saleable form . I n the sm aller p la n ts hydrogen sulphide was burned as fuel since production was too small to ju stify installation of special equipm ent to convert it to useful form . In Ger
many byproduct hydrogen sulphide has been converted to elemental sul
phur, but p rio r to the discovery of
the new sour gas fields in Southwest A rkansas such conversion has always proved uneconomical in the United States.
U tilization o f the hydrogen sul
phide in A rkansas presents peculiar problem s. P roposals to convert it to sulphuric acid find some interest be
cause the increasing need f o r avia
tion gasoline has resulted in the in
stallation o f m any new alkylation p la n ts in which sulphuric acid is re quired. However, acid m ust lie con
sumed w ithin a radius o f about 200 miles to be economical due to high fre ig h t rates. The alternative seems to be conversion o f hydrogen sulphide to some chemical product other than sulphuric acid o r the conversion to sulphur in which form it may be read
ily shipped g rea ter distances.
In d u stry is a t presen t engaged in ap p ra isin g the gas purification situ a
tion in an attem p t to find a practical and economical outlet fo r the by
p roduct hydrogen sulphide. I t should be noted th a t because of the limited quantities o f sulphide available in certain areas only th a t available in the M cKamie, D orcheat and M ace
donia, and possibly the M agnolia fields, can be converted economically.
The total hydrogen sulphide available from these fields, according to p res
ent-day estim ates, m ay be equivalent to more than 100 net tons o f sulphur daily. Based on estim ates of reserves, the total available m ay be 1,500,000 tons w ith the m ajo r portion o f 1,000,- 000 available in the M cKam ie field.
Since it is im practical to convert this q uantity o f hydrogen sulphide to sulphuric acid, other uses m ust be found. The sim plest, o f course, would be the conversion of the hydrogen sulphide to elem ental sulphur. P lan ts fo r such conversion, however, arc ex
pensive, therefore they probably can be economically justified only where the gas is high in hydrogen sulphide
•ind the supply large.
P ro p o rtio n in g e q u ip m e n t for a ir a n d g a s f e e d s to th e c o m b u stio n c h a m b e r o f th e S u s e a r c h p ilo t p la n t
Left to rig h t— C o m b u stio n c h a m b er , c o o le r , m ix in g c h a m b er (H~S a n d S 0 2), c a t a ly s t c h a m b e r , c o n d e n s e r s
C ooling Tower Psychrom etry—III
EDWARD SIMONS
E n g i n e e r , R e d w o o d M a n u t a c t u r e r s Co ., S a n F ra n ci s c o . Cal if.
C h e m . & M e t . I N T E R P R E T A T I O N
C oncluding a se ries of articles on psychrom etry of w ater coolin g tow er d esig n , this installm ent is in three parts. The first p resents eq u a tio n s for counter-flow forced convection tow ers a n d in clu d es an origin al short-cut m ethod of problem solution. S econ d part considers h ea t a s a driving force. The third is a d iscu ssion of h ea t e x ch a n g e an d introduces the n e w concept of the cy c le m od u lu s.—E di to rs.
A
s e c t i o n o f a typical counter-flow cooling tower is represented by P ig. 5. The ra te of cooling in the differential volume clV may be expressed by the equation
dQ = K D adV (43),
where K — overall coefficient o f heat tra n sfe r based upon the mass velocity o f the a ir; D = n et effective driving force difference causing cooling in the section; a = active contact area per u n it volume of the a p p a ra tu s;
V = tower volume, and Q = ra te of heat exchange in the tower.
I f the change in the w ater ra te due to evaporation w ithin the tower is neglected and if the specific heat of the w ate r is taken a t unity, the energy balance fo r the differential volume may be w ritten as:
dQ = Gdh = L J T (44), w here k is the enthalpy of the air.
I f is the m ean n et driving force fo r the volume V, the h eat ra te Q.
the air ra te G, and the w ater ra te L~, and if K is constant, Eq. (43) and Eq. (44) m ay be combined and in tegrated to yield
M 2 * - Ti) = K D u a V (45).
I t may be shown' th a t the value of n a t any p o in t in the tow er is [p r — Pw + Z ( T — f«.)], w here Z =
\sM 0/r„ M L) ( B — p j ) . F ig. 6 in d i
cates values o f Z fo r barom etric p res
sures from 22 in. through 30 in. H g, w ith barom etric intervals o f 1 in.
Table IV gives a step integrated solution o f a very extreme cooling problem . F ig. 7 show's a p lo t o f the n et effective driving force difference versus the w ater tem perature. The d ata fo r F ig. 7 are contained in Table IV . I f the v ariation in the net effective driving force difference were linear with respect to the w ater tem
peratu re, the logarithm ic mean driv
ing force would apply. However, the variation is not linear, and the use of the logarithm ic m ean would result in a driving force which would be too la rg e ; the size of the designed tower would be too small.
The value of A in Table IV (column 12) is the area exposed p e r pound o f d ry a ir fo r cooling in a given step.
Thus, the sum m ation of the step values of K A is equal to K a V /G . I f the basis of the a ir flow is 1 lb. o f air p e r mill., Eq. (45) becomes
A (ft - TO/G =
K D u (Summation A) (4C).
Therefore,
D>, = M f t - Ti)/GKX
(Summation of A) (47).
The heat given up in the cooling, p e r pound of dry air, is (0.670 x
F ig . S— D ia g r a m m a tic r ep re sen ta tio n of a t y p ic a l c o u n ter -flo w c o o lin g to w e r . F ig . 6— -V a lu es o f Z ( s e e ta b le of n o m e n c la ture p . 85) for b a ro m etric p r e s s u r e s from
22 th ro u g h 30 in c h e s of m ercu ry
FIG. 5 js
«
C o n d itio n s 2
E n e r g y E x c h a n g e f o r dV:
d Q * G d h * L d t
C o n d itio n s 1
X X
57.0) o r 38.19 B.t.u. p e r lb. of dry a ir p er min. Du is equal to 38.19 divided by the summ ation of K A , or,
D u = 38.19/102.4 = 0.373 in. Hg.
I t has been possible to compute and a d ju st a method whereby the logarithm ic mean o f the n et driving force differences a t the ends of the tower m ay be used as a basis fo r com putation o f an adjusted mean effective driving force difference, Du. L et Z)2 and D t be the n e t driving force differences a t the conditions 2 and 1 respectively. Then, the logarith
mic mean is
A .». = D, - A /log ( A / A ) (48) The value of the tem perature T , is located as follow s:
r , = f t + { T , - 3\ ) ‘( A . „ . - A ) /
( A - A ) (49).
W hen T , has been found, the value of the net driving force difference a t T , is com puted; the designation of the driving force is D*. The value (It = D ,,mJ D x) is com puted. The curve o f F ig. 8 indicates values of /. fo r various values o f It. W hen the p ro p er value of >. has been ascer-
CHKMTCAL & METALLURGICAL E N G IN EE R IN G . J U N E 19/,2 . ff— 83.
Water T em perature , D eg. F.
tained, the value o f D u is found by use of the equation
Du = R \ \ D , + D|.„. (1 - \) ) /
(R + 0.05) (50).
In the case o f the problem of Table IV , the following values a p p e a r in solution fo r the ad ju sted logarith
mic mean driving force difference:
D, = 1.952; A = 0.1296; D ,.m. = 0.6726; T , = 77.0; D , = 0.2880; I i = 2.34; 1 = 0.811; ¿>* = 0.353.
The value of Du is 0.353 in. H g when calculated by the adjusted m ethod; the value is 0.373 in. H g when calculated by a step in te g ra
tion. The value calculated by the adjusted method is 94.6 percent of the step value; the logarithm ic mean is 180 p ercen t o f th e step value; and the arithm etic m ean is 279 percent of the step value. The size o f a tower com puted from the adjusted value o f Du would be conservative.
HEAT A S A D R IV IN G FO RCE
In the preceding discussion, the driving forees have been in term s of vapor pressure. I t is possible to state the equations in term s o f the enthalpy of the a ir.10”'11“ The net driving force difference is then the difference between the enthalpy of satu rated a ir a t the w ate r tem pera
tu re and the enthalpy o f the a ir in contact w ith the w ater. Let f c be equal to the surface coefficient fo r heat tran sfe r, and let k' be the v a p orization coefficient fo r use with a driving force based up o n the differ
ence in the values of hum idity o f sa t
u rate d a ir a t the bulk w ater tem pera
ture and the hum idity in the a ir stream . The humid specific heat of the a ir stream is
s = 0.24 + 0.45// (51).
The Lewis equation ” * holds rea-
F ig . 7— N e t e f f e c t iv e d r iv in g fo r c e d if
fe r e n c e v s . w a t e r te m p e r a tu r e : d a t a c o n ta in e d in T a b le IV . F ig . 8 — V a lu e s o f X fo r v a r io u s v a l u e s o f fl, u s e d in s o lu tio n
o f E q. (50)
sonably well fo r the evaporation of w ater into air. I t is stated as fo llo w s:
/ . = k’s (52).
The en th alp y o f the a ir is given by the expression*
h = 0.24 ( t - 0 ) + H (1,061 +
0.45J) (53).
The la te n t h eat o f sa tu ra te d w ater v apor in the a ir a t tem p eratu re T is the enthalpy o f the v ap o r m inus the heat o f the liquid above 32 deg. P . I t is given by the expression
rT = 1,061 + 0.45T - T + 32 (54).
The differential increase in the enthalpy o f the air, dh, is the in crease in the sensible heat content of the original air-w ater m ix tu re ; plus the enthalpy of the evaporated w ater, d li, a t the tem p eratu re of th e m ake
up w ater, T ; m inus the sensible heat of the evaporated m ake-up w ater above the tem p eratu re of the final dry-bulb o f the heated and humidified m ixture. Thus,
dh = s d t + d H (rr + T - 32) - 0.45d// [ T - (i + dt)] (55a);
or,
G dh — t i G d H -J- G sd t — 0 A 5 G d H ( T - I - dt) +
G d H ( T - 32) (55b).
The mass tra n sfe r is (see M g. 5)
G d H = k ' a d V ( H r - IT) (56).
The equation fo r h eat tra n sfe r into the a ir is
G sd t = f . a d V ( T - i) =
k ’s a d V ( T - t ) (57).
Eq. (56) is m ultiplied by r T and added to Eq. (5 7 ):
T r G d H 4 - G sdt = k ' a d V [r r (H r - H ) +
(0.24 + 0.45//) (T - i)l (58a).
F rom E q. (55b)
r r G d H + G sd t = Gd h +
0.45 G d H ( T - t - dt) -
G d l l ( T — 32) (58b).
A com bination o f (54), (58a), and (58b) yields
G dh + 0.45 G U I (T - i - dt) -
G d H ( T - 32) = k ' a d V [ { h T - h ) -
( H r - H ) ( T - 32)] (58c) In Eq. (58c), the effect o f om it
ting the expressions “0A 5 G d E ( T — t — dt ) — G dH ( T — 3 2 )” and
“ — k fa d V (H r—3 ) (T — 3 2 )” is small a t the o rd in ary operation tem pera
tures of cooling towers, and Eq. (58c) m ay be simplified to the form
Gd h = k ' a d V ( h r - h) (59).
Eq. (55b) indicates th a t when the reduction in w ater ra te due to evap
o ration w ithin the volume is neglected, the energy balance fo r the differential section o f the tow er volume m ay be w ritten
L d T = G d h (44a).
I f a mean effective value of the driving force is established, and if integration is perform ed between the
T ab le IV— Step in teg ra ted solution of an extrem e coolin g problem Wet-bulb = 55.0 deg. F.: B, = 24.0 i
(1)T (2)2 / (3)U T—t*(4), (5)Z , (6),(4) X(5)
117 63.63 90.7 26.3 O.OOS98 0.2362
114 61.62 89.4 24.6 0.00S96 0.2204
111 59.61 S8.1 22.9 0.00895 0.2050
108 57.60 S6.8 21.2 0.00894 0.1895
105 55.59 S5.4 19.6 0.00S93 0.1750
102 53.58 83.9 18.1 0.00892 0.1615
99 51.57 82.4 16.6 0.00S91 0.1479
96 49.56 80.8 15.2 0.00S90 0.1353
93 47.55 79.2 13.8 O.OOSS9 0.1227
90 45.54 77.5 12.5 0.00SSS 0.1110
8S 44.20 76.3 11.7 0.00887 0.1038
S6 42.86 75.1 10.9 O.OOSS6 0.0966
S4 41.52 73.8 10.2 O.OOSS5 0.0903
82 40.18 72.5 9.5 0.00884 0.0s40
SO 38.84 71.2 8.8 0.00884 0.0778
78 37.50 69.8 8.2 O.OOSS3 0.0724
76 36.16 68.4 7.6 0.00882 0.0670
74 34. S2 66.9 7.1 0.00SS1 0.0625
72 33.48 65.4 6.6 0.008S0 0.0581
70 32.14 63.8 6.2 O.OOS79 0.0545
69 31.47 63.0 6.0 0.00879 0.0527
68 30,SO 62.2 5.8 0.00S7S 0.0509
67 30.13 61.4 5.6 O.OOS7S 0.0492
66 29.46 60.5 5.5 0.00877 0.0482
65 28.79 59.6 5.4 O.OOS77 0.0474
64 28.12 58.7 5.3 0.00S76 0.0464
63 27.45 57.8 5.2 0.00S76 0.0456
62 26.7S 56.9 5.1 0.00S75 0.0446
61 26.11 56.0 5.0 O.OOS75 0.0438
60 25.44 55.0 5.0 0.00S74 0.0437
n. Hg: U / G = 0.670 lb. water per lb. air
(7) (8) (9) (10) (12)
VT Pw (7 )-(8 ) />=(6)+(9) A (j KA**&Q/D
3.169 1.453 1.716 1.952 1.01 0.52
2.911 1.395 1.516 1.736 2.01 1.16
2.672 1.339 1.333 1.538 2.01 1.31
2.449 1.285 1.164 1.354 2.01 1.49
2.243 1.229 1.014 1.189 2.01 1.69
2.052 1.171 0.8S1 1.043 2.01 1.93
1.875 1.116 0.759 0.9069 2.01 2.22
1.712 1.060 0.652 0.7S73 2.01 2.56
1.561 1.006 0.555 0.6777 2.01 2.97
1.422 0.9509 0.471 0.5S20 1.68 2.89
1.335 0.9138 0.421 0.5248 1.34 2.55
1.253 0.8780 0.375 0.4716 1.34 2.85
1.175 0.8406 0.334 0.4243 1.34 3.16
1.102 0.S048 0.297 0.3S10 1.34 3.52
1.032 0.7701 0.262 0.3398 1.34 3.95
0.9666 0.7342 0.232 0.3044 1.34 4.41
0.9046 0.6999 0.205 0.2720 1.34 4.93
0.S462 0.6646 0.182 0.2445 1.34 5.49
0.7912 0.6310 0.160 0.2181 1.34 6.15
0.7392 0.5968 0.142 0.1965 1.01 5.14
0.7144 0.5S02 0.134 0.1S67 0.670 3.59
0.6903 0.5641 0.126 0.1769 0.670 3.79
0.6669 0.54S5 0 .11S 0.1672 0.670 4.01
0.6442 0.5313 0.113 0.1612 0.670 4.16
0.6222 0.5145 0.10S 0.1554 0.670 4.31
0.6009 0.4982 0.103 0.1494 0.670 4.48
0.5S02 0.4S24 0.0978 0.1434 0.670 4.67
0.5601 0.4669 0.0932 0.137S 0.670 4.S7
0.5407 0.4520 0.08S7 0.1325 0.670 5.06
0.5218 0.4359 0.0S59 0.1296 0.335 2.58
Summation of KA » 102.41
84— 6 • J U N E 19ĄS • CHEM ICAL & M ETALLURGICAL EN G IN EER IN G
NO M ENCLATURE
a = Area of active cooling surface per unit of tower volume, sq. ft. per cu. ft.
ai — Area of packing exposed to water flow, per unit of tower volume, sq. ft.
per cu. ft.
A = Area of cooling exposed in an integration step, sq. ft. per lb. dry air.
b, — A function of the cycle modulus, Fig. 10.
6/ = A function of the cycle modulus, Fig. 11.
B = Barometric pressure, in. Hg.
B . — Barometric pressure used in calculations involving comparison with basic data a t 30.0 in. barometer, in. Hg.
c = Height of a packing cycle, ft. per cycle.
D = N et driving force difference, or effective potential, in. Hg.
D , - Diffusivity, sq. ft. per min.
Di. = Logarithmic mean net driving force difference, in. Hg.
Dm — Mean net driving force difference, in. Hg.
D , = N et driving force difference a t water temperature T ,, in. Hg.
/ . = Overall coefficient of heat exchange by conduction and convection, B.t.u.
per sq. ft., deg. F. temp, diff., and min.
G — R ate of air flow through tower, lb. bone dry air per min.
h = Enthalpy of the air-water mixture, B.t.u. per lb. dry air.
hr = Enthalpy of saturated air-water mixture a t the temperature of the bulk of the liquid, B.t.u. per lb. dry air.
II = Air humidity, lb. of water vapor per lb. bone dry air.
H t = Humidity of saturated air at temperature T, lb. per lb.
I = Cycle Modulus, g.p.rn. per sq. ft. per cycle.
k = Vaporization coefficient, or conductance for vapor transfer, lb. water _ vapor per sq. ft. min., and in. Hg diff.
k' = Vaporization coefficient, or conductance for vapor transfer, lb. water vapor per sq. ft., min., and lb. per lb.
ko = Vaporization coefficient, lb. moles per min., sq. ft., and atm.
K — Overall coefficient of heat exchange based upon the mass velocity of the air, B.t.u. per sq. ft., min., and in. Hg diff.
L = W ater rate through cooling tower, lb. per min.
m — Exponent used in calculation of K in Eq. (77), a function of I M o — Molecular weight of dry air
M l = Molecular weight of w’ater
n — Exponent used in calculation of K in Eq. (7S), a function of I.
N i — The rate of diffusion of water vapor, moles per unit area and unit time.
N T U = Number of transfer units.
p m = Log mean partial pressure of inert gas, air, atmospheres.
pa = Partial pressure of water vapor in the main air stream, atm.
pi = Partial pressure water vapor at the interface, considered the pressure of water vapor in saturated air a t T, atm.
p » = Saturated pressure of water vapor a t the wet-bulb temperature, in. Hg.
P . = Total absolute pressure, atm.
Po = Pressure of a gas, in. Hg.
q = One tenth of the apparent free convection value for K a t B =30 in.
Q = Rate of exchange in a cooling tower, B.t.u. per min.
r = Latent heat of vaporization of water %rapor, B.t.u. per lb.
R = D ,.„./D „
R , = The gas law constant.
s = Humid specific heat air-water mixture, B.t.u. per deg. F., and lb. of dry
& = Gross cross sectional area of cooling tower a t right angles to the air flow,air.
all packing removed, sq. ft.
t = Temperature of the air-water mixture, or dry-bulb temperature, deg. F.
I . = Wet-bulb temperature, deg. F.
T = Bulk temperature of the water, deg. F. . T i — Absolute temperature, deg. Rankine or Kelvin.
T , = Temperature of water a t which D . occurs, deg. F.
U. = Average linear velocity of air stream over packing surface, ft. per min.
V h = Humid volume of air entering tower.
Vo — Molecular volume of a i r .IV
V I — Molecular volume of water v ap o r.17/
1 = Effective total film thickness, unit length.
Z = General value of f./k r , in. Hg per deg. F.
a = A factor to compensate for unwetted packing, equal to 0.30 + 1.90/.
X = A function of R used in Eq. (50) for determination of Dm. 2 = The Sigma Function, in main air stream, B.t.u. per lb. dry air.
S = Denotes sensible equivalence.
SUBSCRIPTS
M = A mean effective value.
T = Conditions at water temperature, T.
1 = Conditions a t the air entrance and water exit of a counter-flow tower.
2 = Conditions a t the water entrance and air exit of a counter-flow tower.
p ro p e r lim its associated w ith the tow er volume, Tesultant equation is G (ht — h ) = k'aV (hr - h)u (60a), where (h T— h ) x is the m ean effective driving force potential.
Eq. (60a) m ay be re-arranged as follow s:
( h i — h i ) / ( t i T — h ) u =
k’aV/G = N T U (60b), where N T U is the num ber o f tra n sfe r
units. The height of the packing di
vided by the N T U gives the height of a tra n sfe r unit. The tra n sfe r u n it corresponds to a change in enthalpy equal to the mean driving force d if
ference. The N T U is a m easure o f the difficulty o f the cooling operation.
The use of the enthalpy driving force and the N T U concept offers an a p proach to a design m ethod sim ilar to the method used in the design of extraction towers.
In Eq. (58c), the q u an tity “h T— S T (T — 3 2 )” is the sigm a function fo r sa tu ra te d a ir a t the te m p eratu re T.
Tlie q u a n tity “h—I I ( T — 3 2 )” is the sigm a function o f the a ir stream less the h eat o f liquid above the w et-bulb;
practically, it is the sigm a function of the a ir stream . The h eat added to the a ir stream p e r pound o f dry a ir is, from Eq. (27),
2, - S, S U (T, - T,) '/G (27a).
Therefore, it is possible to set u p a p ractical design equation in term s of the sigm a function o f satu rated air a t the tem perature of the bulk o f the liquid and the sigm a function of the m ain a ir stream . The equation is stated as follows:
G (Z2 - Si) = Lt (T2 - Ti) = k’a V ( 2 r - 2 )„ (60c).
THE COEFFICIENT O F HEAT E X C H A N G E
The coefficient of h eat exchange is in term s of B.t.u. p e r min. p e r sq. ft.
o f contact area p e r in. H g of driving force difference. I t is the product o f the mass tra n sfe r coefficient, or conductance, and the la ten t h eat of the vapor. Coffey and H orne1“ m ade tests to determ ine the value of this coefficient; they proposed a straig h t line form ula fo r definition of K . In 1918, C arrier11 presented straig h t line form ulas fo r K which were for the cases o f transverse and p ara llel flow;
a re-p lo ttin g of the Coffey and H orne d ata indicated a close agreem ent with the resu ltan t p lo t o f C arrier’s points f o r transverse flow. The stra ig h t line form ulas showed the values o f K to be dependent upon the lin e ar velocity o f the a ir stream p assin g over the wetted surface. A t zero linear vel
ocity, the value o f K was not zero;
the effect o f fre e convection was in dicated. These experim ents w ere p e r
form ed upon evaporators w ith wick- covered surfaces.“
In a cooling tower, th e surfaces of the p acking are covered with m oving w ate r films, and the evapora
tive action differs from the action which occurs fro m a wick-covered s u rfa c e ; the surface of a wick, due to its texture, differs from the surface of a w ater film. As the w ater passes
CHEM ICA L & M ETALLURGICAL E N G IN EE R IN G • J U N E 191,2 6— So
