Industrial and Engineering Chemistry : analytical edition, Vol. 10, No. 4

IN D U S T R IA L

^ENGINEERING C H E M IS T R Y

V o l. 30, C o n s e c u tiv e N o . 15

ANALYTICAL EDITION

20,600 Copies of This Issue Printed

April 15, 1938

ITarrison E. Howe, Editor Vol. 10, N o. 4

De t e r m i n a t i o no f Ni c k e li n Al l o t St e e l s

... W. J. Boyer 175

Sy m b o l Dzt o Si g n i f y Di t h i z o n k . . . P. L. Hibbard 179

Pr e c i s e Me t h o d f o r Si e v i n g An a l y s e s...

...M. Weber, Jr., and Raymond F. Moran 180

Ob s e r v a t i o n s o nt h e Ra r e Ea r t h s...

C. N. McCarty, L. R. Scribner, and Margaret Lawrenz, with B. S. Hopkins 184

Fl o w i n As p h a l t s Sh o w n b y Me t h o d o f Su c c e s s i v e Pe n e t r a t i o n s . . . R. N. T r a x l e r a n d L. R. M o f f a t t 188

Ce r a t e Ox i d i m e t r y...

G. Frederick Smith and C. A. Getz 191

Ap p l i c a t i o n o f Gr i g n a r d Re a g e n t t o St u d y o f Mi n­ e r a l Oi l s...Robert G . Larsen 195

Pr e s e r v a t i o n o f Ol e u m Sa m p l e s . . . John R. Smith 198

Ex u d a t i o n Te s tf o r “ Bl e e d i n g” i n Bi t u m i n o u s Ro o f­ i n g...G . L. Oliensis 199

In t e r n a l El e c t r o l y s i s w i t h o u t Di a p h r a g m s . . . .

... J. J. Lurie and L. B . Ginsburg 201

De t e r m i n a t i o n o f Su l f u r i n Oi l...

... Robert T. Sheen and H. Lewis Kahler 206

De t e r m i n a t i o n o f In o r g a n i c Sa l t s i n Cr u d e Oi l s . .

...Charles M . Blair, Jr. 207

Ap p r o x i m a t e Sp e c i f i c Gr a v it y De t e r m i n a t i o n . . .

... John G . Waugh 209

Qu a n t i t a t i v e Sp e c t r o g r a p h i c An a l y s i s...

...Harley A. Wilhelm 211

Co p r e c i p i t a t i o na n dp H Va l u ei n Pr e c i p i t a t i o n sw i t h 8 - Hy d r o x y q u i n o l i n e...

... Harvey V. Moyer and Ward J. Remington 212

Sm a l l Po r t a b l e Ai r Co m p r e s s o r Su i t a b l e f o r La b o­ r a t o r y Us e...G . F. Flemons 214

Re f l u x Ba t h Ag i t a t o r f o r Lo w- Te m p b r a t u r e Fr a c­ t i o n a l Di s t i l l a t i o n An a l y s i s Co l u m n s . J. W. Tooke 214

Mo d e r n La b o r a t o r i e s:

Co l l o i d a l Ca r b o n Re s e a r c h La b o r a t o r yo f Co l u m­ b i a n Ca r b o n Co m p a n y...215

Mi c r o c h e m i s t r y :

St a b i l i z a t i o n a n d St a b i l i t y Te s t s o f Ce l l u l o s e Ni t r a t e s . . E. Berl, G . Rueff, and C h . Carpenter 219

Pr o c e d u r e f o r Mi c r o f u s i o n s. . Charles Van Brunt 224

Mi c r o r e f r a c t o m e t e r o f Si m p l e De s i g n...

... A. E. Edwards with C. E. Otto 225

Mi c r o d e t e r m i n a t i o n o f Ar s e n i c . . Alfred E. How 226

T h e A m erican C hem ical Society assum es no responsibility for th e s ta te m e n ts a n d opinions ad v an ced b y c o n trib u to rs to its p u b licatio n s.

P u b l i c a t i o n O f f ic e : E a s t o n , P a .

E d i t o r i a l O f f i c e : R o o m 7 0 6 , M i l l s B u i l d i n g , W a s h i n g t o n , D . C . A d r e r t i s i n g D e p a r t m e n t * 3 3 2 W e s t 4 2 n d S t r e e t , N e w Y o r k , N . Y .

T e l e p h o n e : N a t i o n a l 0 8 4 8 - C a b l e : J i e c b e m ( W a s h i n g t o n ) T e l e p h o n e : B r y a n t 9 -4 -430

P u b lish ed b y th e A m erican C hem ical Society, Pub licatio n 0fficc 20th &

N o rth a m p to n Sts., E a s to n , P a . E n te re d as second-class m a tte r a t 1;he P o et Office a t E a s to n , K . , u n d e r th e A c t of M arch 3, 1879. “ <8 „ „ th e In d u s tria l E d itio n m o n th ly on th e 1st; A nalytical E d itio n m o n th ly on th e 15th; N ew s E d itio n on th e 10th a n d 20th . A cceptance for

ra te of p o stag e p ro v id e d fo r in Section 1103, A c t of O ctober 3, 1917, A nnual s u b sc rip tio n ra te s: In d u s t b i a i, a n d ® N O l N E E H r a a Ch e m i s t e t

com plete, S6.00; ( a ) In d u s t r i a l Ed i t i o n $3.00; ( 6 ) An a l t t i c a i. Ed i t i o n

*2.50; (e) Ne w s Ed i t i o n 51.50; (a) an d (i>) to g eth er, *5.00. F oreign p ostage to co u n tries n o t in th e P a n A m erican U nion *2.40, (o) *1.20;

(b) *0.60; (c) *0.60; C an a d ia n po stag e o n e-th ird th ese ra te s. Single copies:

(a) $0.75; (6) $0.50; (c) $0.10. Special rate* to m em bers.

C laim s fo r copies lo st in m ails to be ho n o red m u st be received w ith in 60 dayB of d a te of issue a n d based on reasons o th e r th a n "m issin g from files.”

T en d a y s ’ ad v an ce n o tice of change of ad d ress is req u ired . A ddress C harles L. P arso n s, B usiness M an a g e r, M ills B u ild in g , W ash in g to n , D . C ., U. S. A.

4 INDUSTRIAL AND ENGINEERING CHEMISTRY VOL. 10, NO. 4

J u st as insurance actuaries can tell you the life expectancy o f any average group o f m en, so the heads o f large laboratories and the purchas­

in g departm ents o f m a n | sc h o o ls, hospitals and industries have a table o f m ortality for their tech ­ nical glassw are. T hey can estim ate the life o f a gross o f test tubes alm ost to the m onth.

T here is no gu essw ork in their purchasing.

Experience and com parison have proven there is a difference in test tubes. That on e lot w ill last just so lo n g , and another lot so many w eek s lo n g er.

T herefore, they figure tube costs not by the dozen, but by the year.

Such departm ents have proven that Pyrex brand test tubes have a lo n g er life expectancy and, consequently, a lo w er replacem ent cost.

T here are three definite reasons for this supe­

riority— resistance to thermal sh o c k — resistance to m echanical sh o ck and resistance to chem ical attack and clou d in g. Pyrex test tubes have the low est co-efficient o f expansion o f any test tubes o f co m ­ mercially m ade glass— .0 0 0 0 0 3 2. T h ey are prac­

tically exem pt to breakage from heat. T est tubes o f “ Pyrex” brand glass are heavier, have stronger w alls that resist the w ear and tear o f classroom and laboratory use.

T h ese are the reasons Pyrex brand test tubes have a lo n g er life expectancy. T h ese are the rea­

sons laboratories prefer them and purchasing de­

partments specify them. For a ll test tubes— for a ll uses— it pays to standardize on Pyrex Laboratory G lassware.

C

o r n i n g

--- m eans---

R e s e a rc h in G lass

“P Y R E X * ’ is a registered trade-m ark a n d indicates m anufacture by

C O R N I N G G L A S S W O R K S • C O R N I N G , N . Y.

APRIL 15, 1938 ANALYTICAL EDITION 5

T O G E T H E R

For accuracy, your C h ro m el-A lu m el C ou ples s h o u ld be co n n ected on ly to C h ro m el-A lu m el Leads. T h ey b elo n g to g eth er . . . . If so -c a lle d “ c o m p e n sa tin g ” lead s are used, th e ir ju n c tio n w ith th e co u p le (in th e cou p le h an d le) b eco m es a n o th er th e rm o -co u p le . T h is ju n c tio n often b eco m es q u ite h o t, and in t h a t case m a y in tro d u ce an error up to around ± 20°. In o th er w ords, th e se

“ co m p en sa tin g ” lead s d o n ’t c o m p en sa te , excep t a t a lower tem p era tu re th a n th a t u su a lly reach ed b y th e couple h a n d le . . . . W hen th e lead s have th e sa m e c o m ­ po sitio n as th e couple, th e above source o f error o b ­ viou sly does n o t exist. So, for accuracy, u se C h rom el- A lu m el C ouples w ith C h ro m el-A lu m el Leads. For a fu ll exp osition of th e se fa cts, ask for F older-G Y . . . . H osk ins M a n u fa ctu rin g Co., D etro it, M ich.

C H R O M E L - A L U M E L

L E A D S A N D C O U P L E S

6 INDUSTRIAL AND ENGINEERING CHEMISTRY VOL. 10. NO. 4

.7/,,’ C H E M I S T R Y

JtAe A U T O M O B I L E

Sw ift and silent—responding to every need for surging power to flatten h ills and turn minutes into m iles—the modem car takes in its stride a hundred demands w hich w ould have taxed the top performance o f only a few years ago.

Yet, for the secret of its mechanical perfection, w e must go far beyond the most carefully coordinated assembly line. Virtually every part o f today’s automobile is made o f materials as carefully designed for a specific job as the part itself. This ''designing" o f special materials to exacting specifica­

tions is the actual starting point o f your car’s performance. Whether the material be an alloy o f steel or other metal, a rubber or synthetic product, chemical reagents in the hands o f skilled technicians played a vital part in its development.

For work o f this exacting nature, research chemists rely on Merck Reagent Chemicals. Experience has show n that their purity and dependability are essential for uniform results, both in the laboratory and under the condi­

tions of actual production. A catalog w ill be mailed on request.

M E R C K & CO. I N C . ^M anufacturing C&Aemi6l4 R A H W A Y , N. J .

APRIL 15, 1938 ANALYTICAL EDITION 9

V À N C O U V t!

• W I N N I P E G

M O N T R E A L *

BOSTON M I N N E A P O L I S

t/^ NEW VO RK N EW ARK :m i cAg o

SALT aKttlW

C IN CI N N A T I

°fNV£R •

FRANCISCO

K A N S A S

C I T Y ST.LOUIS R I C H M O N D *

T U L S A •

MEMPHIS

D A L L A S »

NEW ORLEANS

H O U S T O N •

Call th e N e a r e s t Coleman Consultant

Burrell T e c h n ica l S u p p ly C o.

Pittsburgh

H o w e & F re n ch , Inc.

B o s t o n

C in cin n a ti S cien tific C o . Cincinnati

C o le m a n E le c t r ic C o ., In c . C h ic a g o

N . J . L a b o r a to ry S u p p ly C o.

N e w a r k & N e w Yo rk

E b e rb a ch & S o n C o . Ann A r b o r & Detroit

P h ip p s & B ird, Inc.

Richmond

C a n a d ia n L a b o r a to ry S u p p lie s, Ltd.

T oronto, M o n treal & W in n ip eg

M a r s h a ll-W e lls B. C . Ltd.

V a n c o u v e r

I. L. L y o n s & C o.

N e w O rlea n s

T e c h n ica l P ro d u cts C o . M em phis

R e fin e ry S u p p ly C o . Tulsa

G r e e n e B ro th ers, Inc.

D allas

W . H. Curtin C o m p a n y Houston

D e n v e r Fire C l a y C o m p a n y D en ver, El P aso & S a lt La k e C ity

R e d m a n S cie n tific C o m p a n y S a n Francisco

W . A . T a y lo r & C o ., Inc.

Baltim ore

G e o . T. W a lk e r & C o . M in n eapo lis

...EVERYWHERE

TwENTY-FOUR C o le m a n c o n su lta n ts, trained un d e r f a c t o r y s u p e r v i­

sio n, a r e at y o u r se r v ic e . T h o r o u g h ly v e r s e d in pH practice a n d f u l l y f a m i l i a r with the c o m plete C o le m a n line, th e s e m en can a b l y a s s i s t y o u in the intelligent selection o f pH equipm ent. T h e y ’ re loc a ted a t strategic points th ro u g h o u t this c o u n try a n d C a n a d a , q u ic k ly re a c h e d w h e r e v e r y o u r p la n t m a y be.

C o le m a n o ffers the m ost c o m p le te line o f G l a s s Electrode E qu ipm e n t f o r m e a s ­ uring pH. S e v e n p orta ble m o d e l s — a u to m a t ic f a c t o r y e q u i p m e n t — a w i d e a s s o r t ­ ment o f a c c e s s o rie s .

A ll C o le m a n pH Electrometers a r e true p o tentio m eters. W e d o not o f f e r d eflection t y p e instruments fo r precision d e te rm in a tio n s. P io n e e r in v e s t ig a to r s a n d d e s i g n e r s o f precision instruments in this n e w f ie ld , C o le m a n le a d s w ith the la te st d e v e l o p ­ m ents, a n d m ost c o m plete se r v ic e . Let a C o le m a n in strum ent p r o v e its sup e riority on the jo b in y o u r o w n la b o r a t o r y . W e a r e o ffe r in g a w e e k ’ s f re e trial o f a C o le m a n installation. No ob liga tion on y o u r part. Write f o r d e t a ils to d a y .

To Control— Is First O f All To Know

Send for Coleman Manual o f pH and Practice. First edition exhausted. But can still sup­

ply a limited number o f copies.

W rite on your letterhead TODAY.

COLEMAN ELECTRIC COMPANY, Inc.

10 INDUSTRIAL AND ENGINEERING CHEMISTRY VOL. 10, NO. 4

W EBER E L E C T R IC DRYING O VEN S

IN A NEW, B R I G H T L Y P O L I S H E D A L U M I N U M AND M O N E L M E T A L E X E C U T I O N

WEBER ELECTRIC DRYING OVENS. The front and door frame, heretofore made of dull finished cast aluminum, are now brightly polished, which greatly im­

proves their appearance. Other parts are of polished Monel metal.

These Ovens continue to enjoy a wide sale because of their dependability as to performance and uniformity of tem perature distribution, as welj as rugged construction.

Weber Ovens are provided with autom atic tem pera­

ture control and pointer scale for approxim ate setting at any desired tem perature within the range of the Oven selected.

7805. Electric Drying Oven, Weber Type A, as above de- Code scribed, inside dimensions 10 x 10 x 10 inches, Word with two shelves but without glass door. For operation to 150°C. With thermometer 200°C, cord and plug for attachment to lamp socket. For 110 volts, a.c... 75.00 Lucfn 7807. Ditto, Type B, inside dimensions 12 x 12 x 12 inches,

with three shelves and double pane Pyrex glass door. For 110 volts, a.c : ...

100.00

Ludaw 7809. Ditto, Type BHT, identical with No. 7807 but for

operation to 260°C. Maximum power con­

sumption, 1000 watts. For 110volts,a.c. 150.00 Ludot

W E B E R E L E C T R I C VACUUM O V E N S

OF ST A IN LESS ST E EL

WEBER ELECTRIC VACUUM OVENS are now offered with front panel, sides, top and back of Stainless steel, in addition to the new style door clamp previously announced. These Ovens are also pro­

vided with automatic control and pointer scale for approximate setting at any desired temperature up to 150°C.

The new door clamp permits internal pressures up to 5 lbs. per sq.

in. and consequent rapid and complete interchange of gases.

A vacuum can be held without continued use of the pump. A test in our laboratory showed an evacuation to 0.5 mm of mercury during a continuous run of seven hours with a pump operating. The mano­

meter reading was the same 16 hours after evacuation ceased.

7886. W eb e r E lectric V acuum O ven, S m all M o d el, as a b o v e d escrib ed , Code o u tsid e dim en sio n s x 1 2 ^ x 1 0 ^ inches, w ith v a cu u m W ord c h am b er 8 inches d eep x 7 % inches d ia m e te r, w ith th re e shelves. M ax im u m c u rre n t c o n su m p tio n 500 w a tts. W ith th e rm o m e te r. F o r 110 v o l t s ...2 0 0 .0 0 Lydha 7888. D itto, L arg e M o d el, o u tsid e dim en sio n s 16^6 x 1 5 H x 12*4

in ch es, w ith v a cu u m ch am b er 9 H inches d eep x 9y£ inches d ia m e te r. M ax im u m c u rre n t c o n su m p tio n 800 w a tts.

F o r 110 v o lts ... 2 7 5 .0 0 Lycan 7890. D itto , L arg e M o d el, D o u b le D e p th , id e n tic a l w ith 7888 b u t w ith

v a cu u m c h am b er 19 in ch es d eep x 9 H in ch es d ia m e te r.

M ax im u m c u rre n t co n su m p tio n 1800 w a tts . F o r 110 v o lts ...3 5 0 .0 0 L y e p i

W e b e r E lectric V acuum O ven, L arge M o d el, D o u b le D epth

Copy o f pam phlet EE-106 giving detailed description o f above Ovens, w ith larger sizes, etc., sent upon request.

ARTHUR H. THO M AS COMPANY

R E T A IL — W H O LE SA LE— E X P O R T

L A B O R A T O R Y A P P A R A T U S AND R E A G E N T S

W EST WASHINGTON SQUARE PHILADELPHIA, U. S. A.

Cable Address, “ B alan ce,” Philadelphia

W eb e r Type B E lec tric D ry in g Oven

INDUSTRIAL and ENGINEERING CHEMISTRY

ANALYTICAL EDITION H a rriso n E. H o w e , E d ito r

D eterm ination o f N ickel in A lloy Steels

A P h o to m etric T itrim eter

W . J . B O Y E R , R e s e a r c h L a b o r a to r y , T h e C a r p e n te r S t e e l C o m p a n y , R e a d in g , P a .

T

H E volumetric determination of nickel in steel as pro­

posed by Moore (6) has been investigated so often that further studies in this field may appear to be unnecessary However, the d ata are not very consistent, and apparently no effort has been made to find what conditions are necessary to yield results on a basis of IN i = 2Ag.

Johnson (3) has reported extensively on the subject, al­

though his d ata do not include a study of such conditions as the permissible limits of nickel concentration, the amount of excess ammonia, and the range of temperature. The pro­

cedures recommended by Lundell, Hoffman, and Bright (5) and by the American Society for Testing Materials (1) are almost identical and are presumably the average conditions used by a large number of analysts in the field. Recently, Peters (8) has applied the cyanide titration for nickel to nickel-chromium alloys, b u t fails to specify the temperature a t which the titration should be made. The temperature is just as im portant as the amount of excess ammonia, as is shown below. Kolthoff (4) considered the optimum condi­

tions necessary for the determination of alkali cyanides by titration w ith silver nitrate in ammoniacal solution, using silver iodide as an indicator. He recommended 4 to

6

ml. of 7.5 N ammonia in excess, and 0.2 gram of potassium iodide per

100

ml. of solution.

Electrolytic depo­

s i t i o n is r e c o m ­ mended for umpire nickel determinations in high-nickel steels (5), b u t this method cannot be used for control in a steel- making process be­

cause of the time re- q u i r e d . T h e di- m e t h y l g l y o x i m e method (1, 5) con­

sumes much less time but has the disad­

vantage of requiring small samples (aliquot portions) for high- nickel alloys. An additional factor, too often neglected, is the solubility of the nickel dimethylglyoxime.

The cyanide method, while rapid, is not entirely satisfac­

tory if the end point is judged by visual means. Partridge (7) has demonstrated the use of the photoelectric cell as an indicator in precise titrations. In the present study, it was decided to employ the photronic cell for the determination of the end point in the cyanide titration method for nickel.

Apparatus

A modified arrangement of Partridge’s apparatus was as­

sembled (Figures 1 and 2), using a Leeds & N orthrup students’

type of potentiometer to measure the e. m. f. of the photronic cell. The potentiometer was used only to obtain the curves shown in Figures 3 and 4. All other titrations in this inves­

tigation were made with ordinary radio potentiometers th at were made an integral p art of the completed apparatus.

The galvanometer used was a Leeds & N orthrup instru­

ment of the enclosed lamp and scale type with a sensitivity of

0.02

microampere per mm. division. Figure

1

^{shows a} schematic diagram of the apparatus along with the electrical circuits used. The lamp housing has a small 110-volt 15- watt concentrated filament lamp with a reflector mounted rigidly in one end of a brass tube. The lamp voltage is kept constant with a Raytheon voltage regulator of 60 w atts’

capacity. The manu- f a c t u r e r claims an output of 115 volts =*=

1

per cent. These regulators were found much more conven­

i e n t t h a n s t o r a g e batteries a n d h a v e p r o v e d sufficiently constant for this type of work. The other end of the lamp hous­

ing contains a small lens, 3.5 cm. in di­

ameter, to render the light rays parallel.

The compartment for the glass cell is made of fiber board, 6.3 mm. (0.25 inch) thick, and of such in­

side d i m e n s i o n s as just to accommodate the g la ss c e l l . Its

176 INDUSTRIAL AND ENGINEERING CHEMISTRY VOL. 10. NO. 4 inside height is 14 cm., which is sufficient to shield the solu­

tion from stray light. The glass cells are of the museum jar variety with a total capacity of about 0.9 liter. While these cells are not optically perfect, they are nevertheless satisfac­

tory since the potentiometer setting for each titration has its own initial zero. The photronic cell, manufactured by the Weston Electrical Instrum ent Corporation, is mounted directly opposite the lamp housing with a 3.5-cm. hole for the light to impinge on the cell. This area allows the activation of about 80 per cent of the photronic cell’s surface. The stirring apparatus, marketed by the Arthur H. Thomas Company, is of the worm-gear drive type with a hollow spindle through which the glass stirring rod is fastened by a spring clamp.

This has the additional advantage th a t it can be raised and lowered conveniently when changing the glass cells. All burets used were calibrated by the National Bureau of Standards.

R eagents and Standard Solutions

Ci t r i c Ac i d So l u t i o n. Dissolve 200 grams of citric acid (U. S. P. grade) in 1000 ml. of water.

So d i u m Io d i d e So l u t i o n. Dissolve 10 grams of sodium iodide in

100

ml. of water.

Co p p e r Su l f a t e So l u t i o n. This solution was prepared from a c. p. grade of CuSOcSHjO, and was standardized by precipi­

tating the copper as the sulfide and igniting to cupric oxide.

Its strength was adjusted to equal 0.0002 gram of copper per ml.

C o b a lt S u l f a t e S o lu tio n . The cobalt in a c. p. grade of CoSO< • 7H20 was precipitated once as potassium cobaitinitrite

(3) in order to remove nickel. The potassium cobaitinitrite w as decomposed with nitric acid and the solution fumed with sulfuric acid. I t was standardized by precipitating th e cobalt with a-nitroso-/3-naphthol and igniting to Cos

04

. Its strength was adjusted to equal 0.0002 gram of cobalt per ml.

Pu r e Me t a l l i c Ni c k e l. The nickel metal was a special metal prepared by W. A . Wesley of the International Nickel

Fi g u r e 2 . As s e m b l y o f Ph o t o t i t r i m e t e r

Company. The purity indicated, by difference, after chemical and spectroscopic examination was 99.98 per cent nickel.

St a n d a r d Si l v e r Ni t r a t e So l u t i o n. Dissolve 5.789 grams of silver nitrate in water and dilute to exactly

1000

ml. One milliliter of this solution is theoretically equivalent to

0.0010

^gram of nickel.

St a n d a r d So d i u m Cy a n i d e So l u t i o n. I t is convenient to have two solutions on hand: For low-nickel alloys, dissolve 3.4 grams of sodium cyanide in

1000

ml. of water containing

1.0

gram of sodium hydroxide. For high-nickel alloys, dissolve 28.0 grams of sodium cyanide in

1000

ml. of water containing

1.0

gram of sodium hydroxide.

Standardize by applying the method described in the section under recommended procedure. Cyanide solutions change with age and must be checked daily.

R ecom m ended Procedure

Weigh accurately a 1-gram sample of alloys containing from 5 to 35 per cent of nickel. For alloys containing more nickel, use a weight equivalent to about 0.3 gram of nickel.

Transfer the sample to a 600-ml. beaker, treat with 20 ml. of diluted nitric acid

(1

^to

1

), and heat until solution is complete.

Dilute to 300 ml. with cold water and add 60 ml. of citric acid solution. Add diluted ammonium hydroxide (1 to 1) until just alkaline to litmus, and then 5 ml. in excess. Transfer the solu­

tion to the glass titrating cell, and add

2

ml. of standard silver nitrate solution and 10 ml. of sodium iodide solution. Dilute to 500 ml. and adjust the temperature to 30° C. Add standard sodium cyanide solution until the solution clears and then about

1

ml. in excess. Adjust the galvanometer to zero. Titrate the excess cyanide with standard silver solution until a permanent deflection of 25 mm. is obtained. A correction for the total amount of silver nitrate used must be made.

Standardize the sodium cyanide solution by titrating an iron-nickel alloy of known nickel content. The calculation of the nickel titer of the cyanide solution is made according to Formula 1. The calculation for the per cent of nickel found is made according to Formula 2.

... , , ... grams of nickel + (ml. of AgNO» X 0.0010) Nickel titer = ---ml. of NaCN--- (1) Per cent of nickel =

ml. of NaCN X Ni titer — (ml. of AgNO; X 0.0010) inn weight of sample

Although copper and cobalt consume cyanide, a correction can be made when the amounts are definitely known. The reaction with copper approximates the ratio, 2Cu = 7CN, and the nickel equivalent can be calculated by multiplying by the factor 0.807. While the reaction w ith cobalt is not exactly ICo = 4CN, it may be assumed to be so if the amount present does not exceed 5 mg.

Chromium has no effect on the nickel titer if less than 0.1 gram is present. W ith high-chromium alloys, a sample th a t contains less than

0 .1

gram of chromium m ust be taken or else an equal amount m ust be added to the standard used for determining the nickel titer of the cyanide solution.

Experimental

The experimental work described below was done to de­

termine the optimum conditions necessary for the develop­

ment of the recommended procedure. A brief study of the effect of other elements is included.

C y a n i d e vs. S i l v e r E n d P o i n t . The curves shown in Figures 3 and 4 present comparisons between the nickel- cyanide end point and the excess cyanide-silver end point.

The titrations were made on weighed portions of 0.3 gram of pure nickel, 1.0 gram of 30 per cent nickel steel, and 0.4 gram of nickel-chromium (80/20) alloy. The conditions for the titrations were as follows:

The samples were dissolved in 10 ml. each of hydrochloric and nitric acids, diluted to 300 ml. with cold water, and treated with 120 ml. of the citric acid solution. The solutions were neutralized

APRIL 15, 1938 ANALYTICAL EDITION 177 with 7.5 N ammonia, treated with 5 ml. in excess, and diluted

to 500 ml., and the temperature was adjusted to 30° C. The solutions were titrated with standard sodium cyanide solution until the maximum point of inflection was reached. A small amount of cyanide was added in excess to show the effect of diluting the background color (iron and chromium citrates).

The curves shown in Figure 3 were obtained by plotting the volume of standard sodium cyanide solution against the e. m. f.

of the photronic cell.

After titrating with the cyanide, the solutions were treated with

10

ml. of the sodium iodide solution and the excess cyanide was titrated with standard silver nitrate solution. Figure 4 shows the curves obtained by plotting the volume of standard silver nitrate solution against the e. m. f. of the photronic cell.

Ta b l e I . Ef f e c t o f Am m o n i a Co n c e n t r a t i o n o n t h e Re a c t i o n b e t w e e n Ni c k e l a n d Cy a n i d e Io n s (1-gram sam ples of iro n -n ick el alloy, 29.90 p er c en t nickel) E xcess 7 .5

N A m m onia

per 500 M l. N a C N A gN O i

U sed Nickel®

of Solution A dded F o u n d E rro r

M l. M l. M l. Gram Mg.

J u s t alk alin e 39 .4 1 4 .6 0 .3 0 0 0 + 0 . 4

3 9 .6 2 6 .2 0 .3 0 0 0 + 0 . 4

5 .0 3 9 .7 0 7 .0 b

3 9 .6 5 7 .0

1 0 .0 3 9 .5 9 6 .6 0.2 9 9 4 - 0 . 2

3 9 .6 4 7 .4 0 .2 9 8 9 - 0 . 7

2 0 .0 3 9 .5 7 8 .1 0 .2 9 7 7 - 1 . 9

3 9 .6 2 8 .2 0 .2 9 8 0 - 1 . 6

2 5 .0 3 9 .6 5 9 .2 0.2 9 7 2 - 2 . 4

3 9 .6 0 8 .3 0.2 9 7 7 - 1 . 9

° All titra tio n s m ad e a t 30° C.

b N a C N so lu tio n ■» 0.0 0 7 7 2 8 g ram of N i p er ml.

The end point obtained with the silver solution is sharp and suitable for deflection end points. In view of the data just presented, all the results in this investigation were ob­

tained by using the cloud point (Agl formation) for the end point, as indicated by the photronic cell.

So d i u m Cy a n i d e- Si l v e r Ni t r a t e Ra t i o. In sodium cyanide-silver nitrate titrations the question naturally arises as to what conditions m ust prevail when the relationship IN i = 2Ag holds. I t was found th a t the sodium cyanide- silver nitrate ratio is affected by alkalinity, temperature, the presence of citrates, and the concentration of silver. How­

ever, under the conditions of the recommended procedure, the nickel equivalent of the standard silver nitrate solution can be taken as IN i = 2Ag, since the amount of silver solu­

tion used in a determination is not enough to cause an ap­

preciable error.

Ef f e c t o f Ex c e s s Am m o n i a Co n c e n t r a t i o n. Table I shows the results obtained by the recommended procedure when the amount of 7.5 N ammonia was varied from just alkaline to 25 ml. in excess. The consumption of cyanide decreases with an increase in alkalinity.

Ta b l e I I . Ef f e c t o f Te m p e r a t u r e o n t h e Re a c t i o n b e t w e e n Ni c k e l a n d Cy a n i d e Io n s

(1-gram sam p les of iro n -n ick el alloy, 29.96 per c en t nickel)

T e m p e ra tu re N a C N A gN O , Nickel®

of Solution A dded Used F o u n d E rro r

° C . M l. M l. Gram M 0.

10 3 9 .9 2 8 .3 0 .2 9 7 7 - 1 . 9

39.91 9 .9 0.2961 - 3 . 5

20 39 .9 1 6 .8 0 .2 9 9 2 - 0 . 4

3 9 .8 8 7 .0 0.2987 - 0 . 9

30 3 9 .8 8 6 .3 b

3 9 .9 0 6 .1

40 „ 3 9 .8 9 5 .9 0 .2 9 9 9 + 0 . 3

39.91 5 .9 0.3001 + 0 . 5

50 3 9 .8 8 5 .0 0.3 0 0 3 + 0 . 7

3 9 .9 2 5 .6 0.3 0 0 4 + 0 . 8

a A n excess of 5 ml. of 7.5 Ar am m o n ia used in all titra tio n s.

b N a C N so lu tio n « 0.007666 g ram of N i p er m l.

Ef f e c t o f Te m p e r a t u r e. Table II shows the results obtained by the recommended procedure when the tempera­

Fi g u r e 3. Ph o t o m e t r i c Ti t r a t i o n o f Ni(NH,)6++ w i t h

NaCN

ture was varied from 10° to 50° C. The consumption of cyanide increases with an increase in temperature.

Ef f e c t o f Ir o n a n d Ni c k e l Co n c e n t r a t i o n. The data presented in Table I I I show th a t varying amounts of iron have only a slight effect on the reaction between nickel and cyanide ions. A nickel concentration of 0.05 to 0.4 gram per 500 ml. of solution does not affect the accuracy of the results obtained for nickel. I t appears th a t the concentration of nickel could be considerably increased and still yield values which are correct to within

0.1

per cent.

Ta b l e I I I . Ef f e c t o f Ir o n a n d Ni c k e l Co n c e n t r a t i o n o n Re a c t i o n b e t w e e n Ni c k e l a n d Cy a n i d e Io n s N ickel0

Alloy B. of S.«

55a N ickel N a C N ArNOj Nickel&

T ak en T ak en P rese n t A dded Used F o u n d E rro r

Gram Gram Gram M l. M l. Gram Mg.

1.0000 0 .4 1 3 4 48 .9 5 7 .1 e

1.0000 0 .4 1 3 4 4 9 .0 0 6 .5

0.8 8 0 0 0 .1 2 0 0 0 .3 6 3 8 4 3 .2 2 6 .9 0 .3 6 3 9 + 0 . 1

0.8 8 0 0 0 .3 6 3 8 4 3 .0 5 5 .8 0 .3 6 3 6 - 0 . 2

0.7 3 0 0 0 .2 7 0 0 0 .3 0 1 8 3 5 .9 7 7 .0 0 .3 0 1 5 - 0 . 3

0.7 3 0 0 0 .3 0 1 8 3 6 .5 7 1 1 .5 0 .3 0 2 3 + 0 . 5

0.5 9 0 0 0 .4 1 0 0 0 .2 4 4 0 2 9 .3 6 7 .5 0.2 4 4 3 + 0 . 3

0.5 9 0 0 0 .2 4 3 9 2 9 .6 7 1 0 .6 0 .2 4 4 0 + 0 . 1

0.4 4 0 0 0 .5 6 0 0 0 .1 8 2 0 2 1 .7 9 4 .9 0 .1 8 1 9 - 0 . 1

0.4 4 0 0 0 .1 8 1 9 2 2 .2 9 9 .2 0.1 8 2 1 + 0 . 2

0.2 5 0 0 0 .7 5 0 0 0 .1 0 3 5 13.01 7 .6 0 .1 0 3 7 + 0 . 2

0 .2 5 0 0 0 .1 0 3 4 13.16 9 .6 0.1 0 3 3 - 0 . 1

0.1 2 0 0 0 .8 8 0 0 0 .0 4 9 8 6 .5 8 6 .2 0.0 4 9 9 + 0 . 1

0.1 2 0 0 0 .0 4 9 6 6 .7 7 8 .6 0 .0 4 9 5 - 0 . 1

° N ickel alloy co n tain s 41.34% N i, 0.07% C u, a n d 0.02% Co. N atio n al B u reau of S ta n d a rd s open h e a rth iro n 55a co n tain s 0.019% N i, 0.046% C u, an d 0 .0 0 8 % Co.

b C o rrected fo r copper a n d c o b alt p resen t.

« N a C N so lu tio n = 0.008596 g ra m of N i p er m l., c alcu la te d fro m N i + Co + (0.8 X C u ).

Ef f e c t o f Co p p e r a n d Co b a l t. During melting opera­

tions, nickel limits m ust be held within a very small range to control the final magnetic properties of the alloy. By cor­

recting for small amounts of copper and cobalt known to be present in the mix, it is possible to control the nickel content within a range of 0.2 per cent. The copper corrections in

INDUSTRIAL AND ENGINEERING CHEMISTRY Table IV were made by

using the factor 0.807 to c o n v e r t i t t o its nickel equivalent.

The theoretical rela­

tion of cobalt and cya­

nide is usually given as ICo = 5CN. In e x p e r i m e n t s wh e re c o b a l t a l o n e w a s titrated according to the recommended pro­

cedure, it behaved ac­

cording to the indi­

cated ratio. When 0.7 gram of iron was added, the reaction was sup­

pressed to the extent th at a relation of ap­

p r o x i m a t e l y ICo = 4CN was indicated.

When large amounts of n i c k e l , s u c h a s 0.3 gram, were added r a t h . the iron, the values for t h e n i c k e l f o u n d showed good recoveries when corrected by us­

in g t h e r e l a t i o n of ICo = 4CN.

Ef f e c t o f Ch r o­ m i u m. A solution of nickel sulfate was standardized by electrolytic deposition.

An average of 0.3017 gram of nickel per 50 ml. of solution' was obtained, including the small am ount of nickel remain­

ing after electrolysis. The chromium was added by using a c. p. grade of sodium dichromate. I t was reduced to CrUI with a small excess of sulfurous acid and boiled. The solu­

tion was treated with bromine water and again boiled to expel the excess bromine.

Large amounts of chromium are exceedingly hard to retain in ammoniacal solution, regardless of the amount of citric acid solution used. I t was found th a t 0.3 gram of chromium could be held in solution with reasonable amounts of citrate if considerable ammonium chloride was also used. In the experiments shown in Table V, 25 grams of ammonium chloride were used in conjunction with the recommended procedure.

Both Johnson (3) and Peters (S) claim th at the end point cannot be determined accurately unless the chromium is first oxidized to CrVI. W ith the phototitrimeter, a good end point is obtained regardless of the valence or amount of chromium, b u t when more than

0.1

gram of chromium is present, the nickel titer, as obtained on pure nickel, is slightly low. For high-chromium alloys, therefore, a weight of sample should be taken which does not contain more than

0 .1

gram of chromium.

Comparison o f R esults by Several M ethods

El e c t r o l y t i c Me t h o d. The 30 per cent nickel steel previously referred to was standardized electrolytically. The procedure used was as follows:

_Accuratcly weighed 1-gram samples were dissolved in 500-ml.

Kjeldahl flasks with diluted hydrochloric acid

(1

to

1

) and oxidized with nitric acid. The solutions were evaporated to low volume and double ether separations made to remove most of the iron. The acid layers were evaporated to fumes with sulfuric acid, cooled, and diluted with water. The solutions

were saturated with hydrogen sulfide to precipitate the copper, and filtered on small papers. The filtrates were boiled to expel hydrogen sulfide, oxidized with bromine, and again boiled to expel the excess bromine. The residual iron was twice precipi­

tated with diluted ammonium hydroxide and the filtrates were evaporated to a volume of 150 ml. An excess of 35 ml. of am­

monium hydroxide was added for the electrolysis. All deter­

minations were corrected for the small amounts of nickel remain­

ing after electrolysis.

An average of four determinations gave a value of 29.96 per cent nickel. The sample was found to contain 0.10 per cent copper. The cobalt was determined by using Hoffman’s procedure (“2) to separate the iron, nickel, etc. When 25- gram samples were used, less than 0.005 per cent of cobalt was indicated.

Cy a n i d e Ti t r a t i o n Me t h o d. Three 1-gram samples of the 30 per cent nickel steel were analyzed for nickel by the recommended procedure. An average value of 29.96 per cent nickel was obtained. The cyanide solution was stand­

ardized on a synthetic mixture of 0.3000 gram of pure nickel and 0.7000 gram of National Bureau of Standards open hearth iron 55a.

Di m e t h y l g l y o x i m e Me t h o d. The procedure used for the dimethylglyoxime method was as follows:

Accurately weighed 1-gram samples were dissolved in 20 ml.

of diluted nitric acid

(1

^to

1

) and diluted to

1

liter in a calibrated flask. One hundred-milliliter portions (0.1-gram samples) were transferred to 400-ml. beakers using a calibrated pipet. Ten milliliters of hydrochloric acid and

20

ml. of citric acid solution were added, and the solution was diluted to about

200

ml. and neutralized with diluted ammonium hydroxide (1 to 1). Seven milliliters of glacial acetic acid were added and the solution was heated to boiling. Forty milliliters of dimethylglyoxime solu­

tion (10 grams of dimethylglyoxime and 9 grams of sodium hydroxide dissolved in

1000

ml. of water) were added, the solu­

tion was neutralized with diluted ammonium hydroxide

(1

^to

1

^), and then 5 ml. were added in excess. All solutions were then

VOL. 10, NO. 4

Ta b l e IV. Be h a v i o r o f Co p p e r a n d Co b a l t i n Ni c k e l Ti t r a t i o n s

(1-gram sam p les of iro n -n ick el allo y , 29.96 p e r c e n t nickel) C o p p er

A dded C o b alt

A dded N a C N

A dded A gN O i

Used Nickel®

F o u n d E r ro r

M g. Mg. M l. M L Gram M g.

4 1 .9 1 8 .5 b

4 1 .9 4 8 .7

2 .0 4 2 .2 2 9 .1 0 .2 9 9 7 + 0 . 1

2 .0 42 .2 1 8 .8 0 .2 9 9 9 + 0 . 3

5 .0 4 2 .4 0 7 .1 0 .2 9 9 6 ± 0 . 0

5 .0 4 2 .4 8 8 .1 0.2 9 9 2 - 0 . 4

2 .0 . 5 . 0 4 3 .0 1 9 .9 0 .2 9 9 7 + 0 . 1

2 .0 5 .0 4 3 .0 0 1 0 .0 0 .2 9 9 5 - 0 . 1

° C o rrected for co p p er a n d c o b alt a d d ed . b N a C N so lu tio n = 0.007351 gram of N i p er ml.

Ta b l e V. Ef f e c t o f Ch r o m i u m o n Re a c t i o n b e t w e e n Ni c k e l a n d Cy a n i d e Io n s

(50 m l. of N iSO i so lu tio n , e q u iv a le n t to 0.3017 g ra m of nickel) C hrom ium

A dded N a C N A gN O , Nickel®

(C rIII) A dded Used F o u n d E rro r

Gram M L M L Gram M g.

3 6 .8 4 1 4.4 b

3 6 .8 2 1 4 .4

0 .1 3 6 .7 6 1 3 .9 0 .3 0 1 6 - 0 . 1

0 .1 3 6 .7 8 14 .7 0 .3 0 1 0 - 0 . 7

0 .2 3 6 .8 2 16 .9 0.2991 - 2 . 6

0 .2 3 6 .8 4 1 7 .4 0 .2 9 8 8 - 2 . 9

0 .3 3 6 .8 7 1 8 .7 0 .2 9 7 8 - 3 . 9

0 .3 3 6 .8 3 1 8 .8 0.2 9 7 3 - 4 . 4

° R ecom m ended p ro ced u re + 25 gram s of NH<C1.

b N a C N so lu tio n ==» 0.008583 gram of N i p er ml.

I l l

melee*« chromium alloy 7?. I per c e n t n i c k e l

19 .5 per c e n t cbromiam

_ 0 ^ j-q {3 j o I s

Volume in w llilite ra

Fi g u r e 4 . Ph o t o m e t r i c Ti­ t r a t i o n o f Ex c e s s C N ” w i t h

A g N O s

APRIL 15, 1938 ANALYTICAL EDITION 179 allowed to digest for 30 minutes at about 75° C. The pre­

cipitates were collected in weighed platinum Gooch crucibles with asbestos beds, washed five to six times with hot (70° C.) distilled water, and dried at 110° C. for

1

hour. The factor 0.2032 was used to convert the weight of nickel dimethylglyoxime to nickel.

The average of five closely agreeing results gave a value of 29.82 per cent of nickel in the 30 per cent nickel steel.

When an alcoholic solution of the dimethylglyoxime was substituted for the sodium hydroxide solution in the method just described, an average of two determinations gave 29.70 per cent of nickel. According to Lundell, Hoffman, and Bright (5), alcohol has a greater solubility effect than am­

monium hydroxide, ammonium salts, or alkali acetate.

The difference between the electrolytic and the dimethyl­

glyoxime value is large compared to the close agreement ob­

tained by the cyanide method based on pure nickel. The following experiments were made to determine whether the difference could be due entirely to the solubility of the nickel dimethylglyoxime.

One-gram samples of the 30 per cent nickel steel were dis­

solved in nitric acid, diluted to

1

liter, and divided into ten por­

tions. The nickel in all ten portions was precipitated with dimethylglyoxime (sodium hydroxide reagent), filtered, washed, and run in all respects under the same conditions as the original dimethylglyoxime standardization. The precipitates were dis­

carded and the filtrates evaporated to dryness. The citrio acid, the excess reagent, and the ammonium salts were destroyed by oxidation with nitric and perchloric acids. The portions were combined and the remaining nickel was again precipitated in a small volume with dimethylglyoxime.

An average of two such runs gave a result of 0.12 per ccnt of nickel, to which

0.01

per cent should be added to allow for the final solubility. This corrected value (29.82 + 0.13 = 29.95) is in very good agreement with both the electrolytic and the cyanide values.

A number of other cases seem worthy of reporting. A sample of Invar was found to contain 35.90 per cent of nickel by cyanide titration, based on pure nickel. The dimethylglyoxime method, using

0

.

1

-gram samples (aliquoted from

1

-gram samples), showed 35.62 per cent of nickel.

In another case, an iron-nickel alloy was found to contain 41.34 per cent of nickel by cyanide titration, based on pure nickel. The dimethylglyoxime method, using 0.08-gram samples (aliquoted from 1.6-gram samples), showed 40.99 per cent of nickel.

An alloy of the 80 per cent nickel-20 per ccnt chromium type was analyzed for nickel by cyanide titration, using accurately weighed 0.4-gram samples, the cyanide solution being stand­

ardized against pure nickel. An average of 77.2 per cent of nickel was found. An average of 77.1 per cent of nickel was found by standardizing against 1-gram portions of the 30 per cent nickel steel (29.96 per cent of nickel). The dimethyl­

glyoxime method, using 0.04-gram samples (aliquoted from 0.4- gram samples), gave a value of 76.7 per cent of nickel.

A determination of nickel by the recommended procedure in National Bureau of Standards 18 Cr

-8

Ni steel 101 gave an average value of 8.49 per cent of nickel, after correcting for copper and cobalt. (Cobalt is not listed on the certificate, but this laboratory obtained a value of 0.058 per cent. The certifi­

cate value for coppcr is 0.055 per cent.) The cyanide solution was standardized on a synthetic mixture of 0.2800 gram of the nickel steel (Ni = 29.96), 0.175 gram of chromium (CrIU), and 0.5 gram of National Bureau of Standards open hearth iron 55a. The certifi­

cate value of 8.44 per cent of nickel for this standard may be slightly low, since the dimethylglyoxime method was used by most of the cooperators listed on the certificate of analysis.

Conclusions

A photometric apparatus lias been developed to detect the end point in the cyanide titration method for nickel and found to be greatly superior to the eye, particularly in analyses of high-chromium alloys.

I t is possible to run duplicate determinations in 35 to 40 minutes on iron-nickel alloys. High-chromium alloys neces­

sarily require more time because of the difficulty of dissolving such samples. The accuracy th a t can be expected is of the order of

0.1

per cent.

The solubility of nickel dimethylglyoxime is demonstrated and it is shown th a t less dependence should be placed on this method when applied to high-nickel steels where the highest order of accuracy is desired.

Acknowledgm ent

The author wishes to thank C. Sterling and W. A. Wesley of the International Nickel Company for the pure nickel used in this work. He wishes also to thank the officers of the Carpenter Steel Company for their encouragement during the progress of this work.

L ite ra tu r e C ite d

(1) Am. Soc. T esting M aterials, “ M ethods of Chemical Analyses of M e tals,” p. 34, 193G.

(2) Hoffman, J. I., Bur. Standards J . Research, 8, 659 (1932).

(3) Johnson, C. M., ‘'C hem ical Analysis of Special Steels," 4 th ed., pp. 221-43, New Y ork, Jo h n W iley & Sons, 1930.

(4) Kolthoff, I. M., and Furm an, N . II., “ V olum etric A nalysis,”

Vol. II, p. 239, New York, Jo h n W iley & Sons, 1929.

(5) Lundell, G. E. F., Hoffman, J . I., a n d B right, H . A., "C hem ical Analysis of Iro n and Steel,” pp. 277-87, New Y ork, Jo h n W iley

& Sons, 1931.

(6) Moore, T., Chem. News, 72, 92 (1895).

(7) P artridge, H . M., In d. En’G. Ch e m., Anal. E d., 4, 315 (1932).

(8) Peters, F. P., Chemist-Analyst, 26, 76 (1937).

Re c e i v e d J a n u a ry 4, 1938.

T h e S y m b o l D z to S ig n ify D ith iz o n e

P . L . H IB B A R D

U n iv e r s ity o f C a lifo r n ia , D iv is io n o f P la n t N u t r it io n , B e r k e le y , C a lif.

C

ONSIDERABLE literature has accumulated in regard to use of diphenylthiocarbazone (phenylazothiono- formic acid phenylhydrazide) as an analytical reagent.

Users of this substance soon contracted the name to dithizone, but so far as is known to the writer the only symbol proposed to represent the name is D. Since this symbol is now commonly used to designate deuterium, the heavy isotope of hydrogen, it should not be used to represent anything else.

The writer proposes the symbol Dz to represent dithizone.

This will avoid confusion with anything else and will facilitate writing formulas or equations in which the radical dithizone occurs. When the word is used as the name of the rea­

gent, diphenylthiocarbazone, it should be spelled out

“dithizone.”

Re c e i v e d F e b ru a ry 15, 1938.