Industrial and Engineering Chemistry : analytical edition, Vol. 8, No. 1

INDUSTRIAL

AND ENGINEERING

CHEMISTRY

Vol. 28, Consecutive No. 3

Published by the American Chemical Society Harrison E. Howe, Editor

Publication Office: Easton, Pa. . Editorial Office: Room 706, Mills Building, Washington, D. G. . Telephone: National 0848 Cable: Jiechem (Washington) . Advertising D epartm ent: 332 West 42nd Street, New York, N. Y. . Telephone: B ryant 9-4430

A nalytical Edition

Vol. 8, N o. 1 January 15, 1936

C O N T E N T S

17,500 Copies of This Issue Printed Solvent Refining of Lubricating Oils with Nitrobenzene . .

. . . 5. 5. Kurlz, Jr., C. E. Headinglon, and B. Zieber ] Determination of Radium in Carnotite and Pitchblende . .

...L. D. Roberts 5 A Method of Analysis for F lu o rid e ...

...W. M. Hoskins and C. A . Ferris 6 lodometric Determination of C o p p e r ...

... William R. Crowell, Thomas E. Hillis, Sidney C. Riltenberg, and Raymond F. Evenson 9 The Determination of Rhenium. I . . . Loren C. Hurd 11 A Color Reaction for Detection of Cyclopentadiene . . . .

...Boris N . Afanasiev 15 Potentiometric Determination of M e rc a p ta n s ...

... Miroslav IV. Tamele and Lloyd B. Ryland 16 Determination of P h o sg e n e ... W. P. Yant, J . C.

Olsen, H. H. Slorch, J . B. Littlefield, and Leopold Scheflan 20 Analyses and Softening Temperatures of Coal Ash . . . .

...David J. Jones and E. L. Buller 25 An Inexpensive Ball M i l l ... Laurence L. Quill 27 Determination of Viscosity of Small Samples of Oil from Oil-

Impregnated P a p e r ...

... II. F. Schneider, Jr., and T. A. McConnell 28 Determination of W ater in G ly c e ro l...

...C. P. Spaeth and G. F. Hutchison 29 Calculating the B l a n k ... Barlholow Park 32 The Reactivity of Coke . . D. A . Reynolds and J . D. Davis 33 Q uantitative Analysis of Mine D u s t s ...

...George L. Clark and Dexter H. Reynolds 36 A Sensitive Check V a lv e ... E. L. Green 40 Determination of Free Sulfur in R u b b e r ...

. . . E. W. Oldham, L. M . Baker, and M . W. Craytor 41 Accurate Separation of Precipitated Mercuric Sulfide and

Sulfur in the Gravimetric Determination of Mercury . . ... Earle R. Caley and M . Gilbert Burford 43 Determination of Formic Acid in Pyroligneous Liquors . .

...Herman D .W eihe and P. Burke Jacobs 44

A Rapid M ethod for the Determination of Titanium . . . . Henry B. Hope, Raymond F. Moran, and Arthur 0 . Ploetz 48 Determination of Organic Sulfur by the Liquid Ammonia-

Sodium M ethod ...

. . . F. J. Souia, V. G. Arcadi, and J, A . Nieuwland 49 A Source of Loss of Ammonia in Kjeldahl Distillations . . .

... Hoke S. Miller 50 A Rapid M ethod for the Volumetric Determination of In ­

dium . . Henry B. Hope, Madeline Ross, and J . F. Skelly 51 Lignin in Douglas F i r ...A. J . Bailey 52 M ethods of W'ine A n aly sis...

... C. II. McCharles and G. A . Pitman 55 Micro-Dumas Generation of Carbon Dioxide . Waller S. Ide 56 Direct Simultaneous Microdetermination of Carbon, H y­

drogen, and Oxygen. I V ... W. R. Kirner 57 A Precision Pyenometer for L iq u id s ...

... S. T. Yuster and L. II. Reyerson 61 An Apparatus for Sugar and Other T itr a t io n s ...

...Edward S. West 62 Detection and Separation of Difficultly Soluble Compounds

by Concentrated Hydriodic A c i d ...

... Earle R. Caley and M. Gilbert Burford 63 N ote to A u t h o r s ...6 7 An Im pact C utting Test for Tire Tread Stocks.. J . H. Dillon 68 A Differential Refractorneter . D. Rau and W. E. Roseveare 72 An Electrically Heated Melting Point A pparatus . . . .

... Edwin Dowzard and Michael J . Russo 74 Prevention of Foaming in Crude-Fiber Determinations . .

... II. W. Gerrilz 75 An Automatic Recording B a la n c e ...

... D. S. Binninglon and W. F. Geddes 76 Semi-Micro-Cottrell Boiling Point A p p a r a tu s ...

... M ary L. Willard and Delcena E. Crabtree 79

P u b lish e d b y th e A m erica n C h e m ica l S o c ie ty , P u b lic a tio n Office, 2 0 th &

N o r th a m p to n Sta., E a s to n , F a . E n te r e d as sec o n d -cla ss m a tte r a t th e P o s t- O ffice a t E a s t o n , P a ., u n d er th e A c t o f M a r ch 3 , 1879, as 4 2 tim e s a y ea r.

I n d u s tr ia l E d itio n m o n th ly on th e 1 st; N e w s E d itio n on th e 1 0 th a n d 2 0 th , A n a ly tic a l E d it io n b im o n th ly on th e 1 5 th . A c ce p ta n ce for m a ilin g a t sp e cia l r a te of p o sta g e p r o v id ed for in S e c tio n 1 1 03, A c t o f O cto b e r 3 , 1 917, a u th o r­

iz e d J u ly 13, 1918. „ ^ ^ , r i .

A n n u a l su b sc r ip tio n r a tes: (a) In d u s t r i a l Ed i t i o n$ 5 .0 0 ; (6) An a l y t i­ c a l Ed i t i o n S 2 .0 0 ; (c) Ne w s Ed i t i o n $ 1 .5 0 ; (a) a n d (6) to g eth er , $ 6 .00;

(a ), (6) , a n d (c) c o m p le te , $ 7 .5 0 . F o r eig n p o sta g e to c o u n tr ie s n o t in th e P a n A m er ic a n U n io n , (a) $ 1 .2 0 ; (6) $ 0 .3 0 ; (c) $ 0 .6 0 ; t o C a n a d a o n e -th ir d th e se r a te s. S in g le c o p ie s: (a) $ 0 .7 5 ; (b) $ 0 .7 5 ; (c) $ 0 .1 0 . S p e c ia l r a te s t o m em b ers.

C la im s fo r c o p ie s lo s t i n m a ils t o b e h o n o red m u s t b e r e ceiv e d w ith in 6 0 d a y s o f d a te o f is s u e a n d b a sed on rea so n s o th e r th a n “ m issin?- fro m files."

T en d a y s' a d v a n c e n o tic e o f c h a n g e of a d dress is re q u ir ed . A d d r e ss C harles L . P a r so n s, B u sin e ss M a n a g er, M ills B u ild in g , W a s h in g to n , D . C .,

4 INDUSTRIAL AND E N G IN E ER IN G CHEMISTRY VOL. 8, NO. 1

inTechical Glassware

.

M m J H

• \ Egg fwBf ^ k k

P io n e e r in g has made Corning Glass Works the largest manufacturers of technical glassware in the world. It was Corning pioneering dial created heat-resistant laboratory glassware in this country.

Pioneering by Corning scientists in other fields has produced the fieat-res/stant lantern globes and other glassware that protect life in railway travel; glass insulators that defy the fury of the power arc;

the gl ass ovenware that is used in a million homes; the tubing that is used in high-grade thermometers; and the largesttelescope disc ever cast.

It is Corning pioneering that makes "P y re x " the symbol of de­

pendability, economy and constant progress in laboratory glassware.

" P Y R £ X " U * r tV n lx r td tradt-jngtk iin d in d icsttt ;' - - t * by

CORNING GLASS WORKS, Co,ninS, N. y.

PYREX LAB ORATORY GL

ANALYTICAL E D IT IO N 5

Retention

Efficiency

Speedy

Filtration

. not 3 G rades

T h e idea surely must appeal— it may seem radical. T im es change. Proc­

esses im p rove— even in filter paper treatm ent. Baker &_ A dam son is out in front.

T h in k o f th e c o n v en ien ce. O ne paper— n o hesitan ce in ch oosin g at th e b en ch , n o m u ltip le stock to w atch in the storeroom . O ne grade

— retentive en ou gh for fine pre­

cipitates, sufficiently strong for all practical require­

m ents and speedy b eyon d an ything o f equal fineness. W ill amaze you on Iron or A lu m in u m precipitates.

T ry it out at our expense. W e’ll quit talking and sen d you an ade­

quate test supply, if y o u ’ll give it a critical trial — Y o u ’ll be surprised and w e ’ll get an order.

TEST SAMPLES OF B.&A. FILTER PAPER

IF YOU HAVE N O T RECEIVED YIO U R C O P Y O F T H E B A K E R & A D A M S O N CATALOG . . . . PLACE A C H E C K M A R K H E R E

BAKER & ADAMSON D IV ISIO N

GENERAL CHEMICAL C O M P A N Y , 4 0 R ector St., N e w Y ork Send me your FREE SAM PLE ENVELOPE o f B & A Filter P aper Circles, w hich I w ill test for a personal observation o f its speed and efficiency.

.

. Ä f f

A ll -purpose Filter Paper

Name Address

6 INDUSTRIAL AND E N G IN EER IN G CHEM ISTRY VOL. 8, NO. 1

HOSKINS

Electric

FURNACES

T he W ir e

E lectrica l H e a t P o ssib le

F r o m th e M ine, to U s - a n d B a c k A g a in

o re th a t s u p p lie s th e 80% n ic k e l c o n ten t, o f C h r o m e l h e a t in g - e le - m c n t s . I n o u r p la n t it is a llo y ed w it h 20% c h r o m iu m a n d d raw n to w ire t h a t is u s e d i n a ll H o sk in s F u r n a c e s. T h e n h a ck to o n e o f th e I n te r n a t io n a l N ic k e l C o. la b o r a to ­ ries w e n t th e fu r n a c e s h o w n b e lo w .

T h is fu r n a c e , F D -203, h a s c o ile d u n i t s o f C h r o m e l, o p e r a te s o n lin e v o lta g e , a n d is g o o d fo r u s e u p to 1800° F . T h e u n i t is a o n e - p ie c e h e lic a l c o il t h a t is w ra p p e d a r o u n d a grooved m u ffle . T h u s i t is very s im p le to r e n e w . S e n d y o u r i n ­ q u iry to y o u r d e a le r . H o s k in s M fg . C o ., D e t r o it.

W e h a v e a h a n d y l i t t l e g a d g e t c a lle d a H e a t in g - U n it C a lc u la ­ to r . T e lls h o w t o d e s ig n c o ile d u n i t s lip l o 1000-W , u s in g C h r o - m c l w ir e . A sk fo r F o r m -K Y .

ANALYTICAL E D IT IO N 7

A S P H A L T S

R O O F IN G S

W EA TH ER -G M ETER F A D E -O M E T E R

C R E A T E

Y O U R O W N

S U N L I G H T a n d W E A T H E R

S o m e o f t h e p r o d u c ts t e s t e d b y th e

F a d e - W e a th e r -

O m e te r : O m e te r :

D yes Paints

Cotton Varnishes

Silk Lacquers

W o o l Enamels

Rayon Japans

Clothing A sbestos

Dress G o o d s A sp h a lt

C ellulose Tar

A u t o topping O il

Knit goods Roofings

Inks Pavings

Printing Paper W ire

W allpaper C able

Lithographing A w n in g s and m any others

J

Test your products for resistance to sun, rain, heat and cold

O IN llie h u n d re d s of m a n u fa c tu re rs , b u y e rs, sellers a n d g o v e rn m e n ta l b u re a u s to w hom th e F a d e -O m e te r a n d W e a th e r-O m e te r a re in d is ­ p en sab le. Q uickly a n d d e p e n d a b ly , th e y fo re te ll th e a c tio n of su n lig h t, m o istu re a n d cooling on th e p ro d u c ts th ey m ak e, b u y , sell o r consum e.

U sers consist of im p o rta n t in stitu tio n s in 32 c o u n trie s all ov er th e w o rld ,

— in E u ro p e , A u s tra lia , th e O rie n t a n d N o rth a n d S o u th A m erica. P ro te c t y o u r re p u ta tio n s ! S a feg u a rd y o u r p u rc h a se s! D e p e n d on th ese ’O m eters.

W rite fo r B ulletins

ATLAS E L E C T R IC D E V IC E S CO.

377 W . S u p erior S t. - C h icago

ADE-OMETE

M EA SU RES W E A T H E R IN G A C T IO N

MEASURES SUNLIGHT FADING

SEROLOGICAL

PIPETTES

VOLUMETRIC AND

OSTWALD PIPETTES

KIMBLE

BLUE LINE EXAX

E im portance of calibration accu racy in a pipette cannot be too highly stressed.

Every drop of the life-giving fluid handled in a serological pipette— the fraction of a m illiliter dispensed from a volumetric pipette — may mean success or failu re of the venture at hand. For this reason, every Kimble Blue Line E X A X pipette bears the ''R E T E S T E D " stamp of extrem e accuracy and assurance. Every pipette carrying the famous

" K " is retem pered (strain - free) against the possibility of breakage — each pipette is m ade of straight, thick-w alled, uniform bore tubing. A ll numbers are d e ep ly acid-etched and filled with a durable, brilliant blue glass, fused in.

V o lu m e t r ic a n d O s t w a l d P ip e tt e s

Size 1 ml.

2 ml.

3 ml, 4 ml.

5 ml.

10 ml.

Tolerance

±0.012 ml.

0.012 ml.

0.02 ml.

0.02 ml.

0.02 ml.

0.04 ml.

Size 15 ml.

20 ml.

25 ml.

50 ml.

10 0 ml.

20 0 ml.

ese rig id tolerances:

P ip e tt e s M e a s u r i n g a n d S e r o lo g ic a l P ip e tt e s

To/erance 1/10 ml. ± 0.005 ml.

X0.06 ml. 2/10 ml. 0.008 ml.

0.06 ml. 1 ml. 0.02 ml.

0.0Ô ml. 2 ml. 0.02 ml.

0.10 ml. 5 ml. 0.04 ml.

0.16 ml. 10 ml. 0.06 ml.

0.20 ml.

•

25 ml. 0.10 ml.

INDUSTRIAL AND E N G IN E ER IN G CHEM ISTRY VOL. 8, NO. 1

KIMBLE GLASS C O M P A N Y • • . v i n e l a n d , n. j.

N E W Y O R K ' C H I C A G O - P H I L A D E L P H I A * D E T R O I T ' B O S T O N

S t a n d a r d iz e on B lue Line E X A X P ip e tte s . . f o r a s s u r a n c e . A fu ll lin e o f K im b le E X A X g la s s w a r e is s to c k e d b y le a d in g L a b o r a ­ to ry S u p p ly H o u se s t h ro u g h o u t th e U n ite d S ta te s a n d C a n a d a .

ANALYTICAL E D IT IO N 9

HOEPPLER VISCOSIMETER

Designed to operate by the principle of the falling ball, and to measure viscosity in absolute units, the Hoeppler Viscosimeter opens new avenues for viscosity measurements with a hitherto unequalled degree of accuracy.

The diameter of the inner glass tube, which is made of highly heat resisting, chemically insoluble glass and those of the calibrated balls are of optical precision, being accurate within 0.0005 mm. The inner tube is filled with the sample, the appropriate ball is inserted and the tube sealed with a gold plated capillary plug and screw cap.

As the instrum ent is leveled a t a 10° angle from vertical, the ball rolls down the side of the tube, thereby eliminating lateral oscilla­

tions and insuring a constant area of fluid passage between the wall of the sample tube and the ball throughout the period of descent.

The time of fall of the ball between two calibrations on the sample tube is accurately recorded and then multiplied by the factor of the ball selected which gives the viscosity of the sample in absolute units with an accuracy of 0.1% to 0.5%.

Duplicate determinations are easily and accurately made by merely inverting the instrum ent, allowing the ball to return and then repeating the measurement. The precision model, being mounted on a special stand maintains a 10° angle when inverted, and a dupli­

cate determination may be made by recording the time of fall of the ball as it returns to its original position. The gold plated capillary plug completely seals the sample tube, permitting inversion without introduction of air into the column, and duplicate determinations may be made without errors resulting from surface tension, evaporation or skin formation since all of these factors are eliminated.

Throughout the entire range of measurable viscosities (0.01 to 1,000,000) high tem peratures are unnecessary except for viscosity tem perature curves. Either precision or industrial models may be supplied with knife heaters for elevated temperatures. For very' precise measurements, a thermostatic water source equipped with a device to circulate water through the jacket of the viscosimeter makes possible an accuracy of 0.05%.

Precision models of the Hoeppler Viscosimeter have a range extending from 0.01 to 1,000,000 centipoises (hydrogen to grease R) and an accuracy of 0.1%. Industrial models have a range extending from 0.6 to 100,000 centipoises and an accuracy of 1.0%.

W rite fo r illustrated bulletin

E. H . Sargent # Co. 155 E . Superior St. Chicago, III.

Viscosity by the Falling B all

Viscosity D irectly in Centipoises

R a n g e 0.01 to 1,0 0 0 ,0 0 0

(in c e n tip o is e s )

R a p id Measurements

Exact Duplications

S R R G E f l T

E . H . S H R G E f l T S- C O . C H I C A G O

L A B O R A T O R Y S U P P L I E S

10 INDUSTRIAL AND E N G IN E ER IN G CHEM ISTRY VOL. 8, NO. 1

A NEW EDITION

S O L U B I L I T Y

of N o n- EIcctro Iytes

by Joel H. Hildebrand, Ph.D

Professor of Chemistry in the University of California

SECOND EDITION— Revised

A. ^C. S. M o n o g r a p h N o . 17

T h e in te r v a l sin ce th e p u b lic a tio n o f th e first e d itio n (1924) h a s se en m a n y ch a n g es in th e s tu d y o f th e p ro b le m s p r e s e n te d b y th e v a rio u s p h e n o m e n a o f s o lu b ility ; th e th e o ry o f d ip o le m o m e n ts h a s g ro w n to m a t u r i t y ; th e fo u n d a tio n s o f a g en e ra l th e o r y o f in te rm o le c u la r forces h a s b ee n la id ; se v e ra l p ro m isin g a tte m p ts h a v e been m a d e to d e a l q u a n t ita tiv e ly w ith th e d e v ia tio n s o f s o lu tio n s from th e id e al s o lu tio n la w s, p a r tic u la rly for t h a t class o f s o lu tio n s t h a t h a s com e to be k n o w n as re g u la r; e x te n siv e a d d itio n s h a v e b ee n m a d e to th e b o d y o f d a t a h a v in g th e o re tic a l sig n ific an c e. I n c o n s e q u e n c e o f th e s e d e v e lo p m e n ts , th is M o n o g ra p h h a s been a lm o s t e n tir e ly r e w r itte n .

C O N T E N T S

1. In trod u ctory

The solvent and the solute. Designation of components.

Units used in stating solubilities.

2. T h e Id ea l S o lu tio n

Raoult’s law. Henry’s law. Van’t Hoff’s law. Statement of Raoult’s law when the vapor cannot be regarded as perfect gas. The fugacity. Other expressions for Raoult’s law. The activity coefficient. Solubility relations based upon Raoult’s law. Partial vapor pressures. Total vapor pressure. Boiling-point composition curve. Solubility of gases. Miscibility of liquids. Solubility of solids. Other solubility relations.

3. D e v ia tio n s from R aou lt’s L aw — G en eral Types of deviation from Raoult’s law. Relation between the activities of the components. Effect of temperature upon deviations from Raoult’s law. Formation of two liquid phases. Effect of deviations upon the solubility of gases. Effect of deviations upon the solubility of solids.

An ideal solution obeys Raoult’s law at all temperatures and pressures. Heat of mixing. Volume changes on mixing iiquids. What substances can obey Raoult’s law at all pressures and temperatures? Regular solutions. An equa­

tion for regular solutions of substances having identical volumes. Components of unequal volumes.

4. P olarity

Dielectric constant and dipole moment. Negativity scales.

Relation of polar groups to the molecular field. Potential between isolated dipoles. Dipole liquids. Langmuir’s treatment of solutions from the standpoint of surface energies.

5. Forces b e tw e e n N o n -P o la r M o le c u le s Intermolecular attractive forces. Repulsive forces.

6. S o lv a tio n , Io n iza tio n a n d A sso cia tio n Chemical combinations between the components. Solva­

tion. _ Solvation and polarity. Difference in positive and negative character of the components. Association of one component. Criticisms of foregoing theories.

7. V a p o r P ressu re o f L iq u id M ix tu re s Aliphatic hydrocarbons. Benzene solutions. Carbon di­

sulfide solutions. Carbon tetrachloride and bromine.

Carbon disulfide and carbon tetrabromide. Hydrogen and deuterium. Constant boiling mixtures—minimum boiling points. Constant boiling mixtures—maximum boiling points. Fractional distillation. Molecular weight from rise in boiling point.

8 . S o lu b ilitie s o f G ases

Gases of low polarity. Chlorine. Ammonia. Carbon dioxide and nitrous oxide. Solubility of gases in solids.

Solubilities of gases in aqueous solutions.

9. S o lu b ilitie s o f L iq u id s in L iq u id s

Stannic iodide. Phosphorus. Molten sulfur. Aniline and hexane. Systems with water as one component. Salting out. Effect of pressure upon the miscibility of liquids.

10. S o lu b ilitie s o f S o lid N o n -E le c tr o ly te s Iodine. Stannic iodide. Sulfur. Phosphorus. Naphtha­

lene, phenanthrene, anthracene, anthraquinone, p-dibromo- benzene and triphenylmethane. Molecular weight by the freezing point method. Mortimer’s method of calculating solubilities. The choice of a solvent for recrystallization.

Diphenyl. Effect of pressure upon the solubility of solids.

11. M e ta llic S o lu tio n s

12. P a rtitio n o f S o lu tio n s b e tw e e n I m m is ­ c ib le L iq u id s

13. S o lu b ility a n d V a rio u s R e la te d P h e ­ n o m e n a

Effect of size of particles upon solubility. Surface tension, adsorption and solubility. Effect of the solvent upon equilibrium in solution. Electromotive force and solubility.

A p p e n d ix A u th o r I n d e x

solubility.

Subject In d ex

203 Pages Illustrated

R E I N H O L D P U B L I S H I N G C O R P O R A T I O N

Price $4.50

330 West 42nd Street New York, U. S. A .

INDUSTRIAL

AND ENGINEERING

CHEMISTRY

Ha r r i s o n E . Ho w e, Ed i t o r

Solvent Refining o f Lubricating Oils with Nitrobenzene

A nalytical M ethods

S. S. K U R T Z , J R ., C. E . IIE A D IN G T O N , A N D B . Z IE B E R , T h e A t l a n t i c R e f in in g Co., P h ila d e lp h ia P a .

T

H IS paper describes the analytical m ethods which have been w orked o u t during the period covering th e developm ent of th e operation and equipm ent for th e nitrobenzene process for solvent refining of lubricating oils as d e s c r i b e d in previous papers (3, 4, 5). T hese m ethods are also useful in th e control of

T h e a n a ly t ic a l m e t h o d s p r e s e n te d c a n b e u s e d to d e te r m in e t h e c o m p o s it io n o f t h e s o lv e n t - o il s o lu t io n p r e s e n t in a n y p a r t o f th e e x tr a c tio n s y s t e m . M e th o d s a re a lso d esc r ib e d fo r d e t e r m in in g t h e c o n c e n t r a ­ t io n o f lo w - b o ilin g fr a c t io n s i n t h e c h a r g e o il a n d t h e c o n c e n t r a t io n o f n itr o b e n z e n e in w a te r .

th e commercial plants now using

this process. T he discussion is divided into five p arts:

source of samples, analysis of nitrobenzene-oil solutions of low nitrobenzene content, analysis of nitrobenzene-oil solutions of high nitrobenzene content, analysis of nitrobenzene-water solutions, and analysis of charge stocks. M ethods for deter­

m ining th e p u rity of fresh nitrobenzene can usually be sup­

plied b y th e producers of this m aterial.

S o u r c e o f S a m p le s

T able I gives a sum m ary of th e p la n t samples on which analyses m ig h t be desirable, the poin t from w hich th e samples are ta k en as shown on th e flow diagram in Figure 1, frequency of sam pling, probable concentration, m ethod used for th e analysis, and th e approxim ate tim e required for th e analysis.

T he w ater samples, instead of being tak en from th e line a t L , are actu ally tak en from the bottom s of a sm all still

inserted a t this poin t and used to recover th e small am o u n t of n i t r o b e n z e n e dissolved in th e w ate r from a w ater separator.

These units also receive th e con­

densed steam from th e vacuum pum ps.

Because of th e w i d e s p r e a d differences in th e nitrobenzene content of s a m p l e s t a k e n a t different points in th e p lant, it was necessary to develop several analytical m ethods giving th e desired accuracy w ith a m inim um expenditure of tim e.

Accordingly, four groups of m ethods are p resen ted : (1) m ethods for determ ining 0 to 5 w eig h t-p e r cent of nitro ­ benzene in oil, (2) 5 to 100 w eight-per cent of nitrobenzene in oil, (3) nitrobenzene in w ater, and (4) low-boiling fractions in charge stocks for the nitrobenzene plant.

D e t e r m in a t io n o f 0 to 5 P e r C e n t o f N it r o ­ b e n z e n e in O il

Two m aterials in th e p la n t contain nitrobenzene in con­

centrations of from 0 to 5 per ce n t: th e ex tra ct and the raffinate. Tw o m ethods are presented for analyzing these m aterials. T he first m ethod involves th e reduction of th e nitrobenzene to aniline by titan o u s sulfate. I t is accurate and reliable b u t requires 1 hour per determ ination and some

Ta b l e I. Su m m a r y

S a m p le T a k e n

R a ffin a te

E x tr a c t

W here T a k en (F ig u r e 1)

AD

C D E

F G II I J

R e co v er ed n itr o b e n z e n e W a ter

C h arge sto ck

K LM

O ne sa m p le p er ta n k 3 to 6 per d a y F o r s p e c ia l p u r p o se o n ly F o r s p e c ia l p u r p o se o n ly F o r s p e c ia l p u r p o se o n ly F o r sp e c ia l p u r p o se o n ly F o r s p e c ia l p u r p o se o n ly F o r s p e c ia l p u r p o se o n ly O ne per d a y

O ne s a m p le per ta n k 2 p er m o n th 3 to 6 p er d a y O ne s a m p le p er ta n k

P r o b a b le C o n c e n tr a tio n A n a ly sis D e s ir e d R a n g e

% N itr o b e n z e n e 0 to 0 . 1 N itr o b e n z e n e 0 to 0 . 1 N itr o b e n z e n e 0 . 1 to 1 . 0 N itro b en ze n e 10 to 3 0 N itro b e n z e n e 10 to 3 0 N itro b e n z e n e 5 0 t o 90 N itro b e n z e n e 5 0 to 9 0 N itro b e n z e n e 0 . 1 to 1 . 0 N itr o b e n z e n e 0 to 0 . 1 N itro b e n z e n e 0 t o 0 . 1 N itro b e n z e n e 9 5 to 100 N itr o b e n z e n e 0 to 0 . 2

L ig h t o il 0 t o 1 . 0

M a x im u m T im e fo r A n a ly tic a l M e th o d U se d A n a ly sis M in .

T ita n o u s s u lfa te 60

P e n s k y M a r te n s flash 3 0

P e n s k y M a r te n s flash 30

S p e c ific g r a v ity or d is tilla tio n CO S p e cific g r a v it y or d is t illa t io n 6 0 S p e cific g r a v it y or d is t illa t io n 60 S p e cific g r a v it y or d is tilla tio n 60 S p e cific g r a v it y or d is t illa t io n 60

P e n s k y M a r te n s flash 3 0

T ita n o u s s u lfa te 60

D is t illa t io n a n d fr ee zin g p o in t 90

C o lo r im e tric 30

F la sh d ifferen ce or d o u b le d is ­

tilla tio n 90

1

1 Extract St Evaporahor 'Condenser c R affinate

Evaporator Condenser

E x tract Strippen.

Raffinate Stripper

,Chilled fO fobenzene-Oil Charqe L Nitrobenzene

lOintinn G

2 INDUSTRIAL AND E N G IN E ER IN G CHEM ISTRY VOL. 8, NO. 1

«Second Sfegp E xtract

Fi g u r e 1

skill in m anipulation; consequently it is used only to deter­

mine the nitrobenzene content of each ta n k of finished oil.

The second m ethod depends upon th e fact th a t th e presence of small am ounts of nitrobenzene in a lubricating oil stock produces an appreciable lowering of the P ensky M artens closed-cup flash of th e stock. This m ethod is simple and rapid, and is used as a periodic check on p la n t operation.

Ti t a n o u s Su l f a t e Me t h o d. In this method, described in a previous paper (1), a small sample of oil admixed with xylene, methyl alcohol, 40 per cent sulfuric acid, and standard titanous sulfate solution is boiled under an atmosphere of carbon dioxide.

The nitrobenzene is reduced and the remaining titanous sulfate back-titrated with ferric alum, the titanous sulfate consumed in the reaction being the measure of the nitrobenzene present.

Because titanous sulfate is oxidized in the presence of air it must be kept in an oxygen-free atmosphere a t all times; hence the necessity for careful technic in this determination. When the proper technic is employed, the absolute error of the method varies from ±0.01 per cent on samples containing up to 0.1 per cent of nitrobenzene to 0 .2 per cent on samples containing 5 per cent of nitrobenzene. All results are calculated to a weight- per cent basis.

Pe n s k y Ma r t e n s Fl a s h M e t h o d. W ith th e aid of Figure 2 the Pensky M artens closed-cup flash point (A. S. T . M .

®7o Nitrobenzen«

Fi g u r e 2

D esignation D93-22) of an oil-nitrobenzene m ixture m ay be com pared w ith the flash point of th e original stock an d thereby used as a m easure of th e nitrobenzene co n ten t of th e oil.

Figure 2 is applicable for oils varying from 149 C. (300° F.) to 316° C. (600° F .) closed- cup flash, b u t is m ore accurate for oils having a flash point above 204° C. (400° F .). T he series of curves shown were interpolated from m any experim ental d ata.

In using Figure 2, th e curve corresponding to the flash poin t of th e original stock, as indi­

cated by th e arrows, is selected. T he point where this curve crosses th e line representing the flash poin t of the m ixture (read on th e y axis) is read on th e x axis as the w eight-per cent of nitrobenzene in th e m ixture. T he ap­

plication of this m ethod is lim ited to operation where th e original charge oil has been properly stripped to elim inate light fractions, as evi­

denced b y a sm all spread between th e open- and closed-cup flashes, described in m ore detail below. If th e closed-cup flash of th e raffinate or ex tract differs from th a t of th e charge, th e closed-cup flash of the product in question should be used in evaluating its nitrobenzene content.

I t m il be observed from Figure 2 th a t this m ethod will not indicate concentrations above 1.0 per cent. I ts accuracy in the working range as based on te st oil-nitrobenzene m ixtures is given in T able II.

Ta b l e II. Ac c u r a c yo f Pe n s k y Ma r t e n s Fl a s h f o r De t e r­ m i n i n g Ni t r o b e n z e n e

N itro b e n zen e P r esen t

P e n s k y M a rte n s F la sh of O rigin al S to c k

P e n s k y M a r te n s F la sh of M ix tu r e

N itr o b e n z e n e

(F ig u r e 2 ) D e v ia tio n

% o

Fm

_{0 F. % %}

0 . 0 1 495 4 85 0 . 0 2 + 0 . 0 1

0 . 3 5 4 95 3 9 5 0 . 3 0 - 0 . 0 5

0 . 0 5 4 40 43 0 0 .0 3 - 0 . 0 2

1 .0 0 4 4 0 3 3 0 0 .8 3 - 0 . 1 9

0 .0 5 4 00 3 9 0 0 . 0 5 0 .0 0

0 . 5 0 • 4 0 0 35 0 0 . 4 0 - 0 . 1 0

0 .0 5 3 65 3 6 0 0 .0 4 - 0 . 0 1

1 .0 0 3 6 5 3 1 0 1 . 0 0 0 00

0 . 0 1 3 05 3 0 5 0 . 0 0 - 0 . 0 1

0 .3 5 3 05 3 0 0 0 . 2 0 - 0 . 1 5

Since this procedure is intended for rapid plant-control work, it has been found desirable to h eat rap id ly to ab o u t 149° C. (300° F .) in th e case of high-flash oils an d then a d ju st to th e sta n d ard ra te of heating prescribed in A. S. T. M . procedure D93-22.

D e t e r m in a t io n o f 5 t o 100 P e r C e n t o f N it r o ­ b e n z e n e i n O il

W hile th e m ethods given above are satisfactory for low concentrations of nitrobenzene in oil, th eir relative errors are high enough to preclude th e ir use on samples coming from various points in the plan t where th e nitrobenzene content ranges from 5 to 1 0 0 per cent.

Two m ethods are presented for th is concentration range.

T he first m erely takes advantage of th e wide difference be­

tween th e specific g rav ity of oil and nitrobenzene. Because the specific grav ity of th e oil stock is usually known in plan t operation, th is m ethod provides a sim ple and rapid control procedure. T he second m ethod m ay be used when th e specific gravity of th e oil is n o t know n or w hen it is desired to isolate a sam ple of th e oil from th e nitrobenzene for physical property determ inations. I t consists of distilling th e nitrobenzene from th e oil under vacuum and m easuring th e w eight of the residual oil. Wlien a check on the oil boiling in th e n itroben­

ANALYTICAL E D IT IO N zene. range is desired, th is m ethod m ay be supplem ented by

a freezing poin t determ ination on th e nitrobenzene distillate.

Sp e c i f i c Gr a v i t y Me t h o d. T he ad d itiv ity of the specific gravities of nitrobenzene and oil on a volum e basis was checked experim entally for a v a rie ty of light and heavy oils. This offered a sim ple m eans of estim ating nitrobenzene concentra­

tion w hen th e oil grav ity was known. Accordingly, a series of charts was prepared relating th e specific g ravity of the nitrobenzene-oil m ixture to th e volum e and w eight-per cent of nitrobenzene. A range of from 0.8 to 1.02 in th e specific grav ity was covered and in order to insure th e hom ogeneity of th e m ixtures all specific gravities are tak en a t 60° C.

(140° F .) or 100° C. (212° F .) depending on th e viscosity of th e m ixtures. T he specific g ravity of plan t nitrobenzene was found to be 1.167 a t 60° C. and 1.127 a t 100° C. B oth values are referred to w ater a t 15.5° C. (60° F.).

Va c u u m Di s t i l l a t i o n a n d Fr e e z i n g Po i n t. T his m ethod for determ ining the nitrobenzene content of an oil- nitrobenzene m ixture consists of a vacuum distillation of a weighed charge of th e m ixture to a specified liquid tem ­ p erature, afte r which th e oil is stripped free of nitrobenzene w ith an in e rt gas, th e residual oil weighed, and an y oil distill­

ing over w ith th e nitrobenzene m ay be estim ated from the freezing p o in t of th e distillate and added to th e residual oil.

In m ost cases this procedure m ay also be used when it is desired to isolate a sam ple of th e oil from th e m ixture for physical pro p erty m easurem ents. E xperim ental work has shown th a t using a properly stripped light stock (400° F.

flash point) a m axim um of 0 .6 per cent of oil based on the m ixture distills w ith th e nitrobenzene. W hen using stocks of 260° to 316° C. (500° to 600° F .) flash point 0.1 per cent or less distills. Physical properties on th e residual oil m ay then be used for p la n t control except where a light stock is present in the m ixture in low concentrations.

The apparatus is illustrated in Figure 3. A is a 2-liter balloon flask fitted with a Claisen head, B. B is electrically heated by wrapping w ith 550 cm. of No. 30 nichrome wire, insulated with asbestos and heated by 110-volt alternating current. C is a capillary tube for adm itting carbon dioxide and preventing bump­

ing. D is 0° to 300° C. thermometer, E a water-cooled condenser, F an adapter provided with vacuum and manometer outlets, and G a 1-liter balloon flask. This size of apparatus permits the distillation of 1-liter samples, which is desirable when physical properties are to be determined on the residual oil.

When only the nitrobenzene content of the mixture is desired, flask II (upper left insert, Figure 3) is preferred. This is a 250- cc. distillation flask with the side arm bent as shown in order to decrease entrainment and is equipped with an additional side neck fitted with a capillary tube for adm itting carbon dioxide gas.The freezing point apparatus (right insert, Figure 3) is made of two test tubes, one placed inside the other to form an air- jacketed freezing point tube. A convenient size for these tubes is 15 and 25 mm. in diameter, respectively. A thermometer calibrated in 0.1° C. over the range 0° to 6° C. is placed through cork in the inside tube and stirring is effected by a wire loop stirrer.

A charge of approximately 1000 grams is weighed to 0.1 gram into the weighed distilling flask, A . The flask is connected to the Claisen head and the system is evacuated to a pressure of 30 mm. A small stream of carbon dioxide bubbles is adm itted to the flask to prevent bumping, and the oil is distilled a t 30 mm.

pressure until the liquid temperature reaches 2 2 0° C. (428° F.).

At this point, the carbon dioxide rate is increased somewhat and heating is continued until a temperature of 250° C. (482° F.) is reached in order to strip the residual oil of nitrobenzene. The tem perature is held a t this point for 3 minutes, after which the oil should be free of nitrobenzene. The flask and contents are then cooled and again weighed. The residual oil in the flask represents the heavy oil in the sample. The same procedure is used in the distillation of 100-gram samples, using flask I I in­

stead of A and Claisen head B.

In order to determine the oil in the nitrobenzene overhead, a portion of this material is poured through the condenser and adapter in order to dissolve any oil remaining from the stripping procedure, the total distillate Is mixed, and the air-jacketed test

tube filled half full of this mixture. Two drops of water are then added to the test tube and the tube is fitted with the thermometer and stirrer. The liquid is stirred vigorously for a few moments in order to saturate the sample with water, then placed in an ice- and-watcr bath, and the freezing point determined.

The weight-per cent of light oil distilling with the nitrobenzene is approximated from the freezing point by the following formula:

W e ig h t-p e r c e n t o f o il =

(5 .1 3 — free zin g p o in t 0 C .) X 165

X w e ig h t - p e r c e n t o f d is tilla te .

5.13° C. is taken as th e freezing point of plant nitrobenzene, 165 is the experimentally determined molecular weight of oil distilling with nitrobenzene, and 70 is the freezing point constant (0) of nitrobenzene. The absolute error of the method is about

±0.5 per cent.

D e t e r m in a t io n o f N itr o b e n z e n e in W a te r W hile the condensed steam from th e vacuum pum ps and th e strippers of th e nitrobenzene p la n t will reta in less th a n 0 .2 per cent of nitrobenzene, even th is sm all am ount can be economically recovered. T h is is accom plished b y vaporizing abo u t 1 0 per cent of th e w ater, which on condensation throw s o u t nitrobenzene while th e w ate r condensate sa tu ra te d w ith nitrobenzene is reprocessed. P ractically all th e nitrobenzene will appear in this distillate and th e rem aining 90 per cent of the w ater can be discarded.

T he nitrobenzene content of th e stripped w ater bottom s is determ ined by a colorimetric m ethod, based on th e reduction of nitrobenzene to aniline w hich forms a colored pro d u ct on reaction w ith sodium hypochlorite. T he color in te n sity is a m easure of th e nitrobenzene content of th e w ater exam ined.

T his m ethod is a m odification of E volve’s te st for aniline (2).

Re a g e n t s a n d Ap p a r a t u s: 1 to 1 hydrochloric acid; 32 per cent hydrochloric acid; 15 per cent potassium hydroxide;

15 per cent solution of cobaltous chloride crystals (CoCl2, 2H20 ) containing 25 cc. of 32 per cent hydrochloric acid p er liter; 7.5 per cent solution of ferric chloride crystals (FeCl3, 6H20 ) contain­

ing 25 cc. of 32 per cent hydrochloric acid per liter; sodium hypochlorite solution containing 0.11 per cent of available chlorine; powdered zinc; solutions of nitrobenzene in water, 0.01, 0.03, 0.04, and 0.05 per cent by weight; one daylight lamp assembly; a test-tube condenser made of a test tube th a t will fit the neck of a 250-cc. Erlenmeyer flask and provided with a cork carrying water inlet and outlet tubes.

INDUSTRIAL AND E N G IN E ER IN G CHEM ISTRY VOL. 8, NO. 1

Pe r m a n e n t Co l o r St a n d a r d s. The permanent color stand­

ards are prepared by taking the quantities of 7.5 per cent ferric chloride solution and 15 per cent cobaltous chloride solution indicated in Table III, and making up to 100 cc. in a Nessler tube with water containing 25 cc. of 32 per cent hydrochloric acid per liter.

Ta b l e III. Pr e p a r a t i o n o f Co l o r St a n d a r d s

Ta b l e V. We i g h t- Pe r Ce n t o f Ni t r o b e n z e n e i n Wa t e r

S tand ard N o .

7 .5 P er C en t F e C li S o lu tio n

Cc.

10.0 7 . 5 7 . 0 11.0 1 2 .0 3 3 .0

15 Per C en t C oC lj S o lu tio n

Cc.

0 . 7 1 . 1 1 .9 4 . 0 5 . 5 6 . 0

The Nessler tubes containing these solutions should be corked whenever n o t in use. They should be checked m onthly against stan d ard nitrobenzene solutions. The color standards are considered satisfactory if they m atch th e colors produced by samples of known nitrobenzene content accord­

ing to Table IV.

Ta b l e IV. Ch e c k i n g o f Co l o r St a n d a r d s S iz e of N itro b e n zen e S ta n d a rd W h ich S h o u ld

Sam ple in S a m p le C h eck C olor P ro d u ce d

Cc. %

10 0 . 0 1 1

20 0 . 0 1 2

10 0 .0 3 3

10 0 .0 4 4

10 0 .0 5 5

20 0 .0 3 6

Pr o c e d u r e. If the sample contains sediment or oil, it should be filtered. The amount of sample taken will depend upon the concentration expected (Table V). The required amount is diluted to 25 cc. with distilled water in a 250-cc. Erlenmeyer flask which is provided with the test-tube water-cooled condenser extending into the flask about 11 cm., and 15 cc. of 1 to 1 hydro­

chloric acid are added. The solution is brought to a boil, approxi­

mately 0.3 gram of zinc dust added, and the condenser replaced as quickly as possible. The solution is allowed to stand for exactly 2 minutes and is then filtered, cooled in ice water, and 5 cc. of sodium hypochlorite (0.11 per cent of available chlorine) are added, followed by 50 cc. of 15 per cent potassium hydroxide solution. The solution is again cooled in ice water, filtered, and made up to 100 cc. in a Nessler tube which is placed in the tube rack. The color of this solution is compared, a t the end of 30 minutes from the time when the sodium hypochlorite was added, with the permanent color standards by looking down through the full length of the tubes, a porcelain plate at the bottom of the rack being illuminated by a daylight lamp. Check determina­

tions should be run.

Table V gives th e w eight-per cent of nitrobenzene for speci­

fied sizes of samples when th e color developed corresponds to one of th e color standards.

: /

/ -■

/ /

/ >

&

&

^'

/ £ ?

V' -

>_ "

/

■ ‘«f t

/

0 0 - °c A

" >i i f ' "

C tiif « Sti

350 400 <50 500

0

Open-Cup FUsh of 1hc Sample °f Fi g u r e 4

V o lu m e o f S a m p le

T ak en 1 2 3 4 5 6

Cc. % % % % % %

20 0 .0 0 5 0 . 0 1

.

0 .0 1 5 0 . 0 2 0 .0 2 5 0 .0 3

10 0 . 0 1 0 . 0 2 0 .0 3 0 .0 4 0 .0 5 0 . 0 6

5 0 . 0 2 0 .0 4 0 .0 6 O.OS 0 . 1 0 0 . 1 2

2 . 5 0 .0 4 0 .0 8 0 . 1 2 0 . 1 6 0 . 2 0 0 . 2 4

T he accuracy of the m ethod is indicated by th e experim ental d a ta in Table V I.

Ta b l e VI. Ac c u r a c y o f Co l o r i m e t r ic Me t h o d f o r Es t i­ m a t in g Ni t r o b e n z e n ei n Wa t e r

S a m p le

N itr o b e n z e n e A dd ed

%

0 .0 0 5 0.020 .0 2 0.02 0 .0 6 0 .0 6 0 .0 6 0 .1 0 0 .1 0

N itro b e n z e n e E s tim a te d

%

0 .0 0 5 0 .0 1 7 0 .0 1 8 0 .0 2 0 0 .0 5 8 0 .0 5 4 0 .0 5 6 0 . 1 0 0 .0 9

C h a r g e S to c k A n a ly s is

In all solvent-refining processes it is necessary either to avoid contam ination of the solvent w ith lighter fractions which m ay be brought into th e system b y th e oil charge or else to resort to special m ethods of separating th e oil and solvent.

If th e oil charged to th e p la n t contains fractions of th e same volatility as nitrobenzene, in stripping solvent from oil some of th e oil will be taken off w ith th e solvent and so contam inate th e la tte r w ith a nonselective diluent.

In th e nitrobenzene process th e introduction of co ntam inat­

ing oil fractions into th e system is preferably and sim ply avoided b y proper steam -stripping of th e oil charge in the initial distillation of th e crude. Experience has shown th a t less th a n 0.05 w eight-per cent of light fractions in the oil charge gives negligible contam ination of th e solvent n itro ­ benzene. C onsequently, oil charge stocks are examined for light oil content for which two te s t m ethods are available.

T he first m ethod, for use on stocks having an open-cup flash point (A. S. T . M . D esignation D92-33) above 232° C.

(450° F .) depends upon th e difference between th e open- atid closed-cup flash points of th e sample, while the second m ethod, for use on stocks having an open-cup flash of less th a n 232° C. (450° F .) consists of a double vacuum distilla­

tion and a d irect m easurem ent of th e light oil present.

Fl a s h Di f f e r e n c e Me t h o d. Owing to th e presence of air currents which carry aw ay sm all am ounts of vaporized oil, th e open-cup flash p oin t of a high-flash oil is only v ery slightly affected b y th e presence of oil boiling in th e nitrobenzene range in concentrations up to 1.0 per cent. On th e other hand, the Pensky M artens closed-cup flash poin t is v ery sensitive to small am ounts of light oil. If no light oil is present th e two flashes will be identical w ithin th e lim its of experim ental error ( ± 5 ° F . for each flash).

T his phenom enon has been utilized in th e determ ination of low boiling oil in charge stocks to the nitrobenzene p lan t. I t is very well suited for this purpose, because both th e am ount of oil th a t will distill w ith th e nitrobenzene and th e effect th a t it will have on th e closed-cup flash poin t of th e stock are roughly proportional to the volatility of th e oil. T h u s the only assum ption necessary regarding the characteristics of the light oil is th a t its initial boiling poin t is n o t far below th a t of nitrobenzene, which has been the case in a v ariety of actual stocks examined. T his assum ption is necessary only because the lower boiling oil will tend to vaporize from the n itro ­ benzene in p lan t operation and be ejected from th e system