A nalytical Edition Vol. 9, No. 2
IN D U S T R IA L
aiE N tiM E E R IN G
C H E M I S T R Y
Vol. 29, Consecutive No. 7
Published b y th e Am erican Chemical Society H arrison E . Howe, E ditor
F ebruary 15, 1937
Publication Office: Easton, Pa. . Editorial Office: Room 706, Mills Building, Washington, D. C. . Telephone: National 0848 Cable: Jiechem (Washington) . Advertising D epartm ent: 332 West 42nd Street, New York, N. Y. . Telephone: B ryant 9-4430
C O N T E N T S
I9/i00 Copies of This Issue Printed
The Role of Spectrophotometry in C o lo rim etry ...
... M . G. Mellon 51
Determination of Diarylamines in Rubber Compositions . ... David Craig 56
Determination of Quartz in the Presence of Silicates . . ...Willard B. Line and Paul W. Aradine 60
Examination of Rubber Latex and Latex Compounds. II.
Chemical Testing M e th o d s ... D. E. Fowler 63
Determination of Iron. Colorimetric o-Phenanthroline Method . . . . L. G. Saywell and B. D. Cunningham 67
Determination of Sulfur in Rubber. Use of Tetrahydroxy- quinone as T itration I n d ic a to r ...
Robert T. Sheen, II. Lewis Kahler, and Delberl C. Cline 69
Determination of Carotene in Forage. Modification of the Guilbert M ethod...
Walter J . Peterson, J . S. Hughes, and II. F. Freeman 71
Determination of Sodium. Removal of Phosphorus be
fore Determining Sodium by the U ranyl Zinc Acetate M e th o d ... 0. B. Overman and 0 . F. Garrett 72
Determination of Small Quantities of 2-Naplithylamine, Aniline, and o-Toluidine in Certain Food Dyes . . . .
... 0 . L. Evenson, J . A . Kime, and S. S. Forrest 74
Volumetric Determination of Bromide. After Oxidation to Bromate in Presence of Much C h lo r id e ...
... I. M . Kolthoff and II. Yutzy 75
Determination of Sulfute, Calcium, and Magnesium in Salt Samples of High P urity . A . C. Shuman and N .E . Berry 77
Applications of Confined Spot Tests in Analytical Chem
istry. Preliminary P a p e r ... Herman Yagoda 79
Rapid Cleaning of M e r c u r y ... . II. F. Easly 82
The Iron Content of Grapes and W i n e ...
. . . . John Byrne, L. G. Saywell, and W . V. Cruess 83
The Principle of the Suspended Level. Applications to the Measurement of V iscosity... Leo Ubbelohde 85
Determination of Higher Alcohols in Distilled Liquors . . IK. B. D. Penniman, Dudley C. Smith, and E. I. Lawshe 91
Support for Perforated Platinum C ru c ib le s...
. . . William M . Thornton, Jr. and Joseph G. Smith 95
Precise Determination of Carbon Dioxide in Air. H and- Operated A p p a r a tu s ... J . G. Waugh 96
Rate of Dehydration of E thyl Alcohol Using Metallic Cal
cium ... G. Frederick Smith and C. A . Getz 100
T h e A m eric a n C h e m ica l S o c ie ty a ssu m e s n o r e sp o n sib ility for th e s t a t e m e n t s an d o p in io n s a d v a n c e d b y c o n tr ib u to rs to its p u b lic a tio n s .
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 S t s ., E a s to n , P a . E n ter e d as seco n d -cla ss m a tter a t th e P o s t- O ffice a t E a s to n , P a ., u n d er th e A c t of M arch 3 , 1879, as 4 8 tim es a y ea r.
I n d u stria l E d itio n m o n th ly on th e 1st; A n a ly tic a l E d itio n m o n th ly on th e 15th ; N e w s E d itio n on th e 10th and 2 0 th . A cc ep ta n ce for m a ilin g a t sp ecia l ra te of p o sta g e p r o v id ed for in S e c tio n 1 1 0 3 , A c t o f O ctober 3 , 1 917, a u th o r
iz ed J u ly 13. 1918
A n n u a l su b sc rip tio n r a tes: (a ) In d u s t r i a l Ed i t i o n$ 5 .0 0 ; (b) An a l y t i c a l Ed i t i o n $ 2 .0 0 ; (c) Ne w s Ed i t i o n $ 1 .5 0 ; (a) a n d {b) to g e th er, $ 6 .0 0 ;
(a ), (6 ), a n d (c) c o m p le te , $ 7 .5 0 . F o reig n p o sta g e to co u n trie s n o t in th e P a n A m erica n U n io n , (a) $ 1 .2 0 ; (6) $ 0 .3 0 ; (c) $ 0 .6 0 ; to C a n a d a o n e -th ird th e se ra tes. S in g le co p ie s: (a) $ 0 .7 5 ; (6) $ 0 .5 0 ; (c) $ 0 .1 0 , S p e c ia l ra tes to m em b ers.
C la im s fo r c o p ie s lo s t in m a ils t o b e h o n o red m u st b e r e ce iv e d w ith in 6 0 d a y s o f d a te o f issu e a n d b a sed on rea so n s o th er th a n “ m issin g fro m files."
T en d a y s ’ a d v a n c e n o tic e o f ch a n g e o f ad dress is req u ired . A d d ress 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 sh in g to n , D , C ., U . S . A j
4 IND U STR IA L AND E N G IN E E R IN G CHEM ISTRY VOL. 9, NO. 2
R o c k efeller C e n te r— a c ity w i t h i n 'a c ity ; a m o d e rn m a ste rp ie c e o f a rc h ite c tu ra l s k ill a n d s tr u c tu ra l e n g in e e rin g . B u t even before th e a rc h ite c ts c o u ld p re p a re th e ir p la n s fo r th is stu p e n d o u s g ro u p o f b u ild in g s , re se a rc h c h e m is ts in v a rio u s fields su c h as m e ta llu rg y , g lass, p a in t, p la stic s, ce ram ic s, a n d c e m e n t, h a d to te s t a n d p ro v e th e fitness o f th e m a te ria ls used in c o n s tru c tio n .
T h e in d u s tria l c h e m is t e sta b lish e s d efin ite sta n d a rd s fo r b u ild in g m a te ria ls ju s t as M e rc k e s ta b lish e s— an d m eets— defin ite s ta n d a rd s fo r R e a g e n t C h e m ic als w h ic h are used b y e x a c tin g ch e m ists w h o d e m an d d efin ite re su lts.
'■fa Merck Reagent Chemicals, A c id s, A lk a lie s , O rg a n ic a n d In o rg a n ic Salts, a n d S o lv e n ts , a re d e s c r ib e d in a handy catalog which is a v a ila b le on re q u est.
M E R C K & C O . I N C . ^ J la n itfa ctu n m p C€/em iA U R A H W A Y , N . J .
FEBRUARY 15, 1937 ANALYTICAL E D IT IO N 5
for the MOST EXACTING Laboratory
R equirem ents
PRODUCERS O F FINE CH EM ICA LS FOR O V E R SIXTY-NINE Y E A R S
W H E N laboratory results of extreme accuracy are required the pu rity of the Analytical Reagent Chemicals employed is a m atter of para
m ount im portance.
M allinckrodt A nalytical R eagent Chemicals, m ade under the m ost stringent laboratory methods, are habitually employed in laboratories of well known universities and clinics.
In all cases where specifications have been established by the American Chemical Society Com m ittee, M allinckrodt m eets the requirem ents fully.
Every M allinckrodt R eagent Chemical bears on its label th e lim it of im purities present.
For unusually exacting work special M allinckrodt Reagents of an even higher grade of purity th an specified by the A. C. S. are offered.
1 2 th A NN UAL T R A D E D IN N E R th is y e a r is to b e h e ld a t t h e W a ld o rf- A s to ria o n T h u r s d a y , M a r c h 4, u n d e r th e a u s p ic e s o f t h e D ru g , C h e m ic a l a n d A llied T ra d e s S e c tio n o f t h e N ew Y o rk B o a rd o f T r a d e . B e s u r e to c o m e .
M A L L I N C K R O D T A n a l y t i c a l R e a g e n t C h e m i c a l s
A F ew o f th e A c c u ra te M a llin c k r o d t A n a ly tic a l
R e a g e n ts
A lcohol, M eth y l A bso lu te (A cetone Free)
B enzol (T h io p h e n e F ree) B e n z y l C hloride
L itm u s P ap er, R ed a n d B lue Lead C h ro m a te
M ercury
P o ta ss iu m N itra te
S o d iu m C h ro m a te , N e u tra l S o d iu m C yanide (G ra n u la te d ) S o d iu m O xalate, N e u tra l
S tro n tiu m C a rb o n a te (Pow dered) Z inc M etal, M ossy (Low A rsenic) Z inc C hloride (S tick s)
ST. L O U IS C H IC A G O P H IL A D E L P H IA
N E W Y O R K T O R O N T O M O N T R E A L
6 IND U STR IA L AND E N G IN E E R IN G CHEM ISTRY VOL. 9, NO. 2
S T O R M E R V I S C O S I M E T E R
TW O NEW V A R IA B L E W E IG H T O U T F IT S
7 6 4 9 - M . 7 6 4 9 - 0 -
S T O R M E R V IS C O S IM E T E R , V a ria b le W e ig h t O u tfits . F o r use w ith m aterials requiring adju stm en t of w eight to produce a predeterm ined ra te of shear as, for example, in the control of th e con
sistency of drilling fluids w here the stan d ard procedure is the determ ination of the w eight necessary to produce 600 revolutions of th e cylindrical ro to r per m inute; see H . A. Ambrose and A. G. Loomis, “ Some Colloidal P roperties of B entonite Suspensions,” Physics, Vol. 1 (A ugust, 1931), p . 129. Also chosen for use w ith v a t printing pastes in th e textile ind u stry as it perm its th e ra te of shear to be varied as m uch as a thousand fold; see S ivert N . G larum , “A F undam ental S tudy of V at P rin tin g P astes,” American D yestujf
Reporter, Vol. X X V , N o. 6 (March 23, 1936), p. 150. Suitable for use w ith m any sim ilar m aterials of colloidal character where a change in th e ra te of shear changes th e ir stru ctu re.
W hile th e stan d ard w eight box filled w ith lead shot perm its ad ju stm en t of the operating w eight from approxim ately 28 grams, th e w eight of the box em pty, to approxim ately 153 grams, th e w eight filled w ith lead shot, we offer as m ore convenient o utfit No. 7649-M w ith special w eight carrier sim ilar to th e pans and bows of an analytical balance on which stan d ard analytical w eights w ith a to ta l w eight of 266 gram s can be used. F o r use w ith m aterials of high viscosity requiring heavier weights, we offer outfit N o. 7649-Q w ith 1 kilo set of slotted weights. C om binations of separate S lotted W eights N o. 7653-G offer a wide range of selection for w eights w hich m ay be required for specific m aterials.
7649-M. V isco sim cter, S to rm e r, V ariab le W eight O u tfit, as described, with cylindrical rotor and test cup provided with two side vanes, central baffle and thermometer holder as in No. 7649 b u t w ith aluminum W eight Carrier, taking analytical balance weights up to 250 grams (total weight including carrier, 266 grams), in place of the usual weight box. This outfit is offered as suitable for use with drilling fluids and similar materials requiring weights within this range. Complete in case, with thermometer, but without balance w eights... 110.00 C odeW ord... Leuga 7649-Q. V isco sim eter, S to rm e r, V ariab le W e ig h t O u tfit, for materials of high viscosity. W ith cylindrical rotor and test
cup provided with two side vanes, central baffle and therm ometer holder as in standard outfit; with set of Slotted Weights, 1 kilo, in place of the usual weight box. This outfit is offered as suitable for use with v at printing pastes and other materials requiring weights within this range. Complete in c a se ... 123.75
Code W ord... Leuks
7653-G. S lo tte d W eights, only, of polished brass, as supplied w ith 7649-Q Outfit, lished by the U. S. Rureau of Standards for Class A and R Weights.
Size, gram s... 25________50 E a c h ... 1.75 2.00 Code W ord... Levqr Levqs
Adjusted within the tolerances estab-
100 200 500 1000
2.25 Levqu
2.50 Levqw
3.50 Levqx
4.75 Levqy C opy o f neic 10-pp p a m p h le t EE-96, illu s tr a tin g a n d d escrib in g th e various m o d els o f th e
S to r m e r V isc o sim eter w i th so m e new accessories, s e n t u p o n r e q u e st.
A R T H U R H. T H O M A S CO M PAN Y
R E T A I L — W H O L E S A L E — 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 E S T W A SH INGTO N S Q U A R E P H IL A D E L P H IA , U.S.A.
C a b le A d d re ss, “ B a la n c e ,” P h ila d e lp h ia
IN D U S T R IA L
andENGINEERING
C H E M I S T R Y
H a r r iso n E . H o w e , E d ito r ANALYTICAL EDITION
T he R ole o f Spectrophotom etry in C olorim etry
M . G. M ELLO N , P u rd u e U niversity, L afay e tte , In d .
O
U T S ID E of th e w hite polar regions, th e black night, and th e gray desert wastes, color plays a m ore im portant esthetic and practical p a rt in our daily lives th a n is generally realized. T he conference on color a t the M assachusetts In stitu te of Technology in July, 1936, dem onstrated the present physical and chemical in terest in the subject. A t this sym posium m ore th a n 25 papers, m ostly by industrial men, were presented on various subjects relating to th e measurem ent of color and the use of colorimetric d a ta in form u
lating useful specifications.
In th e work of th e analytical chem ist colorimetric phe
nom ena have a t least th e following im p o rtan t applications (83): (a) th e m arking of th e course of various chemical re
actions incident to th e determ ination of constituents; (b) the q u alita tiv e detection of various constituents by m eans of specific colorimetric reactions; (c) th e q u an tita tiv e determ i
nation of constituents b y ap propriate m eans; and (d) the actual “ color analysis” of an object. T he present paper is concerned w ith both q u alita tiv e and q u an tita tiv e aspects, in so far as the spectrophotom eter m ay be utilized in evalu
ating th e color analytically.
The word colorimetry implies measurement of color.
Inchemistry the practice of colorimetry involves almost entirely a comparison of an unknown system with a known system to determine whether they have the same intensity of color.
The process of comparison is carried out in a number of dif
ferent ways. One m e th o d is to use what the c h e m ist calls a colorimeter.
A s H a r d y a n d P errin have pointed o u t (2-2), such in s t r u m e n t s should really be called color com parators, s i n c e there is no m easure
m ent of color in the sense th a t physicists use the term color
im etry.
In physics color
im eters are in stru m e n t s in w h ic h s t i m u l i a r e com
b i n e d in s u c h a m anner th a t th e re
s u lta n t c o l o r s e n sation is th e same
as th a t of th e object being m easured. T his resu lt m ay be achieved in several ways b y a num ber of different in stru m ents which serve to synthesize stim uli. In general, such ap p a ratu s is of th e trichrom atic or m onochrom atic type. In the form er the color to be m easured is m atched by a m ixture of th e three prim aries, red, green, and violet, produced in some suitable m anner. In the la tte r th e color under te s t is m atched b y a m ixture of w hite light w ith m ono
chrom atic lig h t produced by a dispersing system . Such colorimeters have n o t found extensive use in chem istry. The chief difficulties are to m a in ta in th e desired q u ality of the light source and to o btain observers having norm al vision.
A lthough spectrophotom eters have been know n for m any years, it is only in the la st decade or tw o th a t th eir value has begun to be generally appreciated in chemical work (50), and there are still few laboratories equipped w ith such instrum ents.
W ith increased production, decreased prices should resu lt in a m ore widespread d istribution of spectrophotom etric equip
m ent.
P r in c ip le o f S p e c tr o p h o to m e tr y
Essentially, a spectrophotom eter is a device for determ ining, wave length by wave length, th e proportion of ra d ia n t energy incident upon a body th a t is reflected or tran sm itte d b y it.
In order to m easure th e transm ission of a colored solution, for example, the arrangem ent of a num ber of in stru m e n ts is
Fi g u r e 1. Iÿ ô n ig- Ma k t e n s Sp e c t r o p h o t o m e t e r Fi t t e d w i t h Lo n g Tr a n s m i s s i o n Ce l l s
51
52 IND U STR IA L AND E N G IN E E R IN G CHEM ISTRY VOL. 9, NO. 2
Fi g u r e 2 . Hi l g e r Sp e c t r o p h o t o m e t e r
F o r w ork in th e visual region th e type of in stru m e n t largely used consists of a lig h t source, a m onochrom ator, a photom e
ter, an eyepiece for observing th e photom etric field, an d a holder for the sample. T he holder is a cell for transm ission m easurem ents of a liquid, or a device for supporting an opaque object on which reflection m easurem ents are to be m ade. Such instrum ents are illustrated in Figures 1 to 3.
Figure 1 shows th e K önig-M artens type, m anufactured by S chm idt an d H aensch, which is th e visual in stru m en t used by th e U . S. B ureau of S tandards. A nother foreign instrum ent is illu strated b y th e H ilger form shown in Figure 2, arranged for transm ission m easurem ents. I t has a unique construction for m aking reflection m easurem ents on opaque samples.
Figure 3 illustrates one of th e best know n in stru m en ts m ade in this country, as m anufactured by B ausch and Lom b for m easuring in th e same cell different thicknesses of a liquid.
All th e b etter instrum ents have variable, calibrated slits for controlling th e w idth of th e spectral b and of light adm itted.
Various details of operation and a thorough discussion of sources of error for th e K önig-M artens in stru m en t are con
tained in a publication b y M cNicholas {28).
D uring th e la st 15 years m any efforts have been m ade to su b stitu te a photoelectric cell for the hum an eye as th e “ seeing”
m echanism of such instrum ents. T he eye has two definite disadvantages in being easily fatigued, and in having a very low sensitivity a t 400 and 700 mju, w here the norm al visibility curve is practically zero. E arly difficulties w ith cells have been overcome u n til certain instrum ents equipped w ith this device are now very satisfactory. Zscheile, Hogness, and Y oung have assembled an a p p a ratu s (52) which gives very precise results although it is som ew hat tedious to use. Less sensitive instrum ents of this gen
eral ty p e are available com mercially. Razek a n d M u l d e r ( 8 8 ) devised an instrum ent (m anufactured by th e Thw ing-A lbert In stru m en t Co. of Philadelphia) th a t provides a spectrophotom etric curve, 8.75 X 11.25 cm.
(3.5 X 4.5 inches), for th e sam ple as soon as the photographic film is developed. Figure 4 illustrates th e instrum ent. A th ird ty p e of photoelectric in stru m en t is th a t devised by H ard y (20) and m anufactured b y th e General E lectric Com pany (83A ). Figure 5 illustrates a late model.
A modified G eneral E lectric instrum ent is used in the a u th o r’s laboratory for all tra n s
m i s s i o n and reflection m easurem ents. Any ordinarily sm ooth curve is p lotted au to m ati-
Tr a n s m i s- cally in 3 to 5 m inutes on 20 X 27.5 cm. (8.5 X 11 inch) paper.
such th a t th e light from a suitable source is divided into two beam s of equal intensity. As these pass through a mono
chrom ator, a narrow spectral b and of th e w idth and wave length desired m ay be selected. T he solution, in a suitable cell, is placed in one of the two beam s; and th e solvent only, in an identical cell, is placed in th e other beam . Since the solution will absorb m ore light th a n th e solvent, th e em ergent beam s are of unequal intensity. B y m eans of a photom eter in th e light p a th th e m ore intense beam m ay be reduced to m atch th e other, as determ ined b y bringing th e tw o halves of an optical field to equal inten sity . T he graduation of th e in stru m e n t enables one to determ ine th e proportion of the light incident on th e solution th a t is tra n sm itte d b y it. This value is know n as th e per cent tran sm itta n cy (48). B y placing the sam ple in a different position, reflection values can be obtained for such m aterials as paper, textiles, and p ain t panels.
S p e c t r o p h o t o m e t r ic A p p a r a tu s
In co n tra st to th e chem ist’s “ colorim etric” determ inations, spectrophotom etric m easurem ents are n o t lim ited to colored system s. F o r m a n y years absorption spectra have been determ ined, chiefly b y photographic m eans, for th e u ltra violet an d infrared regions of th e spectrum . T he present discussion is lim ited to instrum ents used for th e visual region, which are th e m ost useful for m easuring colored system s, b o th opaque and tran sp aren t. A general review of such a p p a ratu s was presented some years ago by a com m ittee on spectrophotom etry of the Optical Society of America (18), and more recently Gibson has discussed th e subject (17).
Fi g u r e 3 . Ba u s c i i a n d Lo m b Sp e c t r o p h o t o m e t e r Fi t t e d w i t h s io n’ Ce l l s f o r Va r i a b l e De p t h
FEBRUARY 15, 1937 ANALYTICAL E D ITIO N 53
Fi g u r e 5. Ge n e r a l El e c t r i c Re c o r d i n g Ph o t o e l e c t r i c Sp e c t r o p h o t o m e t e r
T h e N a t u r e o f S p e c tr o p h o to m e tr ic D a ta
T he fundam ental d a ta provided by a spec
trophotom eter show the proportion of th e light incident upon a sam ple which is reflected or tran sm itte d by it. T here m ay be a single value for a given wave length, or values m ay be determ ined to cover th e whole visible range.
In the la tte r case th e results are generally plotted in the form of a curve w ith transm ission or reflec
tion as ordinates and wave length as abscissas.
W hen a curve is to be m ade covering p a rt or all of th e visible range, th e question arises as to the w ave-length interv al a t which to de
term ine th e individual points, and th e w idth of spectral b and to use for the light source.
If the curve is steep and contains small sharp absorption bands, th e points m ay need to be
tak en a t every millimicron w ith a spectral b and as narrow as possible. Zscheile found such precautions necessary in determ ining chlorophyll (51). If the curve is sm ooth and not steep, points a t each 10 mp. are adequate, an d spectral bands of 3 to 10 nifj are sufficiently narrow . R ecording instrum ents provide a continuous curve.
If instrum ents are g raduated to read directly, the values will be in one of several possible term s, such as percentage, optical density, or extinction coefficient. T he percentage system is probably th e m ost com monly used a t present, although th e others have m erit in special cases. Benford has discussed th e p lo ttin g of spectrophotom etric d a ta (5), and T w ym an and Allsopp include definitions of term s (45). In instrum ents such as Zscheile’s, operating on a substitution basis, th e ratio of two deflections of a galvanom eter or elec- tro m e te rm u st be calculated for each wave-length setting. For m any purposes curves of th e percentage-w ave-length type are sufficient in them selves. Figures G and 7 show such curves for several different systems.
F or certain purposes it is desirable or necessary to convert the d a ta of th e percentage-w ave-length curve to the trichro
m atic or m onochrom atic basis, th u s providing a real an a
lytical m easurem ent of th e color. In this sense th e spectro
photom eter becomes a colorimeter, yielding values on the subtractive or additive system . T he trichrom atic, or tri- stim ulus, values are in term s of the percentages of th e three
F i g u r e 4. R a z e k - M u l d b r R e c o r d i n g P h o t o e l e c t r i c S p e c t r o p h o t o m e t e r
primaries, red, green, and violet, required to m atch the ob
ject. T he m onochrom atic values are in term s of d o m inant wave length and of percentages of brightness and pu rity . Reference is m ade in th e section on color analysis to the m ethod of m aking such calculations.
A p p lic a tio n s
A lthough th e analytical applications of spectrophotom etry are m entioned only occasionally in books on general analytical m ethods, or even in works dealing w ith colorim etric m ethods of analysis, a survey of th e literatu re reveals m any papers.
These cover a v ariety of specific applications, an d th e y are to be found in a num ber of different fields. T he outline below is intended to be representative of th e kinds of applications th a t are being m ade.
Q u a l i t a t i v e U s e s . T he transm issive or reflective char
acteristics of a system m ay be very specific. T hus, there are probably no o ther solutions having th e sam e spectral tran s
mission curve as those of neodym ium salts which show a num ber of sharply defined large and sm all bands in th e visual region. In such a case as this th e spectrophotom etric evidence is quite definite. One m u st always keep in m ind, of course, th a t th e shape of the curve will be changed b y th e presence of an y o ther substance showing a different selective absorption or reflection.
M an y solutions, glasses, and substances such as paper and paints yield curves which are relatively flat and regular.
U sually th e farth e r they go in this direction th e greater is the.
possibility of finding several system s having ab o u t th e same curve. T he curves for m an y solutions have been determ ined, particularly for organic com pounds in th e ultrav io let region., T his qualitative use of curves is equally applicable in th e ultraviolet, visual, and infrared regions.
As a specific example, the recent paper of R uehle m ay be cited (89). In th e course of a stu d y of vitam in B he obtain ed a com pound w hich appeared to. be a thiazole derivative.’
Solutions of th e unknow n and known com pounds gave practically identical curves. As confirm atory evidence various derivatives of th e know n and unknow n compound^
were prepared an d m easured sim ilarly. On obtaining curves which again agreed closely w ith each o ther for th e tw o pro d ucts the conclusion was reached th a t th e tw o compound^
were th e same.
As a second exam ple we m a y consider p ea n u t oil an d m us
ta rd oil. Two samples of such oils which h ad identical coloi1 values on the Lovibond scale yielded curves sig n ific an t^
different (14).
Such examples as those ju st given illu strate th e desirability in certain cases of plotting th e ordinates for a curve on some special basis. Transm ission curves of th e percentage-w ave- length ty p e often differ considerably in shape for different
54 IND U STR IA L AND E N G IN E E R IN G CH EM ISTR Y VOL. 9, NO. 2
Fi g u r e C. Sp e c t r a l Tr a n s m i s s i o n Cu r v e s
F o r sta n d a r d red, a m b er, a n d b lu e g la sse s, a n d for a s o lu tio n co n ta in in g m od ified m e th y l o ran ge a t th e e n d p o in t. T h e lo c a tio n is sh o w n for th e
cu rv es for a p er fec t b la ck , w h ite , a n d g r a y (5 0 per c e n t ) .
thicknesses of th e system m easured. If one plots the loga
rith m of th e optical density (density = 1/transm ission), the curves will all have th e sam e shape regardless of thickness.
Of course one can readily convert transm issions for a n y given thickness to any o ther by m eans of a special slide rule, m en
tioned below, which has a scale based on B ouguer’s law.
In a general w ay it m ay be said th a t this use of spectro- photom etric curves applies to a v arie ty of products for the establishm ent of sta n d ard s an d for th e com parison of subse
qu en t products w ith th e stan d ard s. In this sense th e d a ta are com parable to m easurem ents such as refractive indices.
In a p ain t or dye laboratory, for example, curves would be obtained for th e best obtainable or sta n d a rd grades of the i'uaterials to be tested. T hese d a ta are perm an en t and can be filed for reference a t an y fu tu re tim e. T h ey have all th e advantages inherent in an y num erical specification. The effects of im purities, m ethods of m anufacture, an d o th e r sig
nificant factors can be determ ined in so far as th e y influence th e curves of th e standards. A djustm ent can th e n be m ade un til th e permissible color tolerance is attain ed , as m easured by conform ity to th e sta n d a rd curves. T here is a wide ap
plication possible in this direction. Two exam ples of such uses are for dyes and textiles (9 ,3 5 ,3 6 ,4 0 ), an d for p ain t (1,16,25).
Q u a n t i t a t i v e U s e s . T he possibility of applying spectro
photom etry for m aking q u a n tita tiv e determ inations b y m eans of absorption has been recognized since th e tim e of V ierordt (48). F or some years such m ethods have been recognized by physicists (15, 27, 41). In m ore recent years chem ists have begun to apply such m ethods to a v ariety of systems.
R epresentative examples include dyes (2, 3, 6 ,1 3 ,1 9 ,2 3 ,2 4 , 49), hydrogen-ion concentration (7 ,1 2 ), nitrogen (29), hemo
globin (10), bilirubin (11), chlorophyll (51), vitam ins and horm ones (34), and general applications (44, 46).
T h e common m ethod of m easuring a co nstituent spectro-
photom etrically has been applied recently b y M ehlig for the determ ination of m anganese in steel (30). T he procedure used depends upon the fac t th a t th e tran sm itta n cy of light a t a given w ave length is a function of th e concentration for a solution such as perm anganate. If a reference curve is constructed, plotting th e tran sm ittan cy , a t th e given wave length, of a series of sta n d ard perm anganate solutions against th e know n m anganese concentrations, it is possible to convert th e tran sm itta n cy of an unknow n perm anganate solution into concentration of m anganese by use of th e curve.
Before constructing such a reference curve, one should de
term ine th e transm ission curve from 400 to 700 m/x for the kind of solution to be m easured in order to select a suitable region for m aking th e readings. T he preferable wave length is where th e re is least change in tra n sm itta n c y for a given change in w ave length— th a t is, where there is a m axim um or m inim um in th e transm ission curve. On steep portions of a curve a sm all error in wave length results in too great an error in th e tran sm itta n cy . F o r visual instrum ents th e wave length selected should be as near as possible to th e peak of the rela
tive visibility curve to tak e adv an tag e of th e m axim um sensi
tiv ity of th e eye. Likewise, an inspection of the curve coordinating concentration an d tra n sm itta n c y shows th a t certain portions will yield results of g reater accuracy th a n others. R eadings in th e optim um range can be obtained by suitable dilution of th e sample.
In a la te r paper M ehlig modified th e scheme for calculating th e results in applying th e m ethod for the determ ination of copper in ores (31). F o r solutions conforming to th e Bouguer-Beer equation
I = Jo X i o - *
in which I„ represents th e in ten sity of th e incident light of given w ave length entering th e solution, I th e intensity on leaving th e solution, I th e length of th e cell in centim eters, c th e moles of absorbing substance per liter of solution, and e th e m olecular extinction coefficient. Solving for c, we have
log j
c = —r - moles per liter el
T he values for Jo, I , an d I are know n for an y given determ ina
tion. I t is necessary to p repare a solution containing a known concentration, c, in order to calculate th e value of e for the desired w ave lengths for use in subsequent work. As in the use of a co ncentration-transm ittancy curve, th e w ave lengths selected should be those of th e optim um portion of th e tra n s m ission curve.
Such q u a n tita tiv e m ethods have been designated b y B ar
nard an d M cM ichael (4) as “ analysis by m onochrom atic tran s
mission.” These authors have developed a derivation of Bouguer’s law w hich enables one to calculate th e proportions of th e com ponents in a b in a ry m ixture from m easurem ents of th e transm ittancies of each com ponent and of th e m ixture a t some one w ave length. T w ym an and Allsopp discuss such system s also (46,47) ■ T ern ary m ixtures having certain types of curves for th e separate com ponents m ay be handled in m uch the sam e way. E xcept under th e best of conditions, the accuracy of the results obtained in this w ay is n o t all one would wish.
In connection w ith his work on copper ores, M ehlig reached th e conclusion th a t th e spectrophotom etric m ethod gave results as good as the titrim etric iodide m ethod and th a t the d a ta were obtained m ore rapidly and m ore conveniently.
A t present such m ethods seem of value chiefly for colored system s, since photographic m ethods have hard ly sufficient accuracy. As reliable a p p a ratu s is expensive for th e ordinary laboratory, th e present outlook is th a t th is kind of m ethod will be of value chiefly in situations where some o th e r m ethod is n o t available, as in th e work of Zscheile on chlorophyll.
FEBRUARY 15,1937 ANALYTICAL E D IT IO N 55 i
/~ 6 ra y
¿ - R e d
— 3 - 0 / /r e '1 -Brown S -K d lo ^ v
7-Da r k b /uc
/
s /
/
2
---
& i
\ /
.
__■/i
" t :7 /-f-
is -7 r
&
(Y a re l e n g t h
F i g u r e 7 . S p e c t r a l R e f l e c t i o n C u r v e s F o r p o rcela in en a m el p a in t sta n d a r d s
C o l o r A n a l y s i s . C u rren t chemical publications are alm ost devoid of suggestions for using th e results of “ color analysis.” As generally understood, th e expression refers to th e com bination of stim uli found to be equivalent to th e color being m easured or analyzed. T he results are expressed num erically and represent a unique specification of th e color.
T he results for th e trichrom atic system are expressed in per
centages of th e th ree prim ary stim uli, red, green, and violet, while those for the m onochrom atic system are expressed in dom inant w ave length and in percentages of brightness and purity. T he la tte r values m ay be taken, respectively, as m easurem ents of th e three a ttrib u te s of color—hue, relative brilliance, an d satu ratio n .
T he ordinary spectrophotom eter yields directly only d ata upon th e basis of which one m ay plot graphs such as trans- m ittancy-w ave-length curves. In such cases th e values m entioned above m u st be calculated. T he m eans for ac
complishing th e calculation have been readily available since th e appearance in 1922 of th e report of the Colorim etry C om m ittee (43). F or use w ith the d a ta in this publication a special Keuffel an d E sser color slide rule was of advantage.
An example of its application was given by the author (32).
In 1931 th e In tern atio n al Commission on Illum ination adopted more reliable d a ta (21, 26) which have now superseded those used earlier. Using th e selected ordinate m ethod, outlined by H ard y (20, 21), calculations on an adding m achine are accom
plished fairly rapidly. A device developed in the au th o r’s laboratory (42) facilitates th e operation still more. H ard y designed a calculating m achine (20) for operation in conjunc
tion w ith his recording photoelectric spectrophotom eter, and Pineo has recom m ended its use for testing textile dyeing (37).
So little has been accom plished to bring these num erical evaluations of colorim etric characteristics into everyday use th a t one can h ard ly predict to w h at extent they will become a p a rt of our thinking ab o u t colors in th e near future. U n
doubtedly the invention of rapid, reliable spectrophotom eters and the sim plification of th e tristim ulus calculations will do m uch to extend th e application of such data.
T he num erical values represent a kind of color language which is m ore definite th a n a system based upon samples of
colored m aterials. T hey provide a specification for a color so th a t one can form from them an idea of th e color w ith o u t seeing it. A few examples will illustrate th e possibilities.
Pineo (37) has pointed o u t th e value of trichrom atic character
istics in m easuring color fading, in color m atching, an d in dyeing control. Shelton and Em erson com m ent on th e pos
sibility of using such d a ta for dyed fabrics (40). H a rd y (21) gives th e brightness, dom inant w ave length, and p u rity of some colors th a t are widely used. Since Ja n u ary , 1935, the colors of th e U nited S tate s flag have been specified in tri
chrom atic values by th e B ureau of S tandards (5).
R e s e a r c h W o r k , I n m any research problem s, b o th in analytical and o ther fields, one encounters color as a signifi
can t an d variable property in th e system u nder stu d y . In such cases chem ists who have such equipm ent available are coming m ore and m ore to rely upon spectrophotom eters as an im p o rta n t m eans for obtaining valuable inform ation. In some instances it represents th e m o st im p o rta n t use of such instrum ents. O bjective d a ta are provided showing a t once th e effect on the color of variables such as pH value, concen
tratio n and p u rity of reagents, constituents in th e system affecting th e color, th e order of adding reagents, th e stab ility of th e color, and other factors.
A photoelectric instrum ent, on account of th e sensitivity of the cell in th e violet and red regions, is p articu larly valuable for detecting sm all differences in th e spectrophotom etric curves. N u ttin g has reported (35, 36) recent observations on this point. A recording in stru m e n t enables one to obtain a series of curves quickly on th e sam e sheet of paper. By using pens containing different colored inks, especially in te r
esting com parisons can be b rought out.
L ite r a tu r e C ite d
(1) Am. Soc. T esting M aterials, D esignation D 307-30, “ S tan d ard M ethods of Analysis,” 1930.
(2) Appel and Brodo, I n d . E n g . Ch em., 16, 797 (1924).
(3) Appel, Brode, an d W elch, Ibid., 18, 627 (1926).
(4) B arn ard an d M cM ichael, I n d . E n g . Ciie.m., Anal. E d ., 2 , 363 (1930).
(5) Benford, Trans. Iliu m . Eng. Soc., 18, 67 (1923).
(6) B rode, I n d . E n q . C h e m ., 18, 708 (1926).
(7) Brode, J. A m . Chem. Soc., 46, 581 (1924).
(8) Bur. S tandards, Federal Specification TT-C-591 (1935).
(9) Cuneliffe, J. Soc. Dyers Colourists, 45, 305 (1929).
(10) D avis an d Sheard, Arch. Internal M ed., 40, 226 (1927).
(11) D avis and Sheard, J. Lab. Clin. M ed., 19, 593 (1924).
(12) Deforge, Halle aux Cuirs, Suppl. tech., 1927, 229.
(13) D raves, J. Optical Soc. A m ., 21, 336 (1931).
(14) Ellsw orth, B arn ard , and M cM ichael, Oil & Fał In d ., 7 ,1 5 (1930).
(15) F erry, "P hysics M easurem ents,” Vol. I, pp. 150, 225, New Y ork, Jo h n W iley & Sons, 1926.
(16) G ardner, “ P hysical a n d Chem ical E x am in atio n of P a in ts ,”
C hap. IV , W ashington, D. C., In s titu te of P a in t and V arnish In stitu te , 1935.
(17) Gibson, J. Optical Soc. A m ., 21, 564 (1931).
(18) Gibson e t al., Ibid., 10, 169 (1925).
(19) Gibson et al., U. S. Bur. S tan d ard s, Sci. P aper 440 (1922).
(20) H ard y , J . Optical Soc., 25, 305 (1935).
(21) H a rd y e t al., “ H andbook of C olorim etry,” Boston, Technology Press, 1936.
(22) H a rd y and Perrin, “ T he Principles of O ptics,” p. 310, New Y ork, M cG raw -H ill Book Co., 1932.
(23) Holmes, I n d . E n g . Chem., 15, 833 (1923).
(24) Holmes, Scanlon, and Peterson, U. S- D ept. Agr., Tech. Bull. 310 (1932).
(25) Ingalls, Proc. A m . Soc. Testing M aterials, 26, 347 (1926).
(26) Ju d d , J. Optical Soc., 23, 359 (1933).
(27) Kriiss, “ K olorim etrie u n d q u a n tita tiv e S p ectralanalyse,” p.
291, 1891.
(28) M cN icholas, U. S. B ur. S tan d ard s, Research Paper 30 (1928).
(29) M ain and Loekc, J. Biol. Chem., 64, 75 (1925).
(30) M ehlig, I n d . E n g . Ch em., Anal. E d . , 7, 27 (1935).
(31) Ibid., 7, 387 (1935).
(32) M ellon, J . Phys. Chem., 33, 1931 (1929).
(33) M ellon, Ferner, an d Mehlig, J . Chem. Education, 10, 691 (1933).
(33A) M ichaelson and Liebhafsky, Gen. Elec. Ren., 39, 445 (1936).
(34) M orton, "T h e A pplication of A bsorption Spectra to the Stu d y of V itam ins and H orm ones,” London, A dam H ilger, 1935,
56 IND U STR IA L AND E N G IN E E R IN G CHEM ISTRY VOL. 9, NO. 2 (35) N u ttin g , A m . D yestuff Iieptr., 23, 251, 275 (1934); Textile Re
search, 4, 323 (1934).
(36) N u ttin g , J . Optical Soc., 24, 135 (1935).
(37) Pineo, A m . D yestuff Reptr., 22, 470 (1933).
(38) R azek and M ulder, B ritish P a te n t 349,565 (1931); J . Optical Soc. A m ., 20, 155 (1930).
(39) Ruehle, J . A m . Chem. Soc., 57, 1887 (1935).
(40) Shelton and E m erson, Ind. Eng. C h e m ., Anal. E d., 4 ,248 (1932).
(41) Spitzer, thesis, P urdue U niversity, 1910.
(42) Sw ank, unpublished paper.
(43) T ro lan d e t al., J . Optical Soc. A m ., 6, 527 (1922).
(44) T w ytnan, Chemistry and Industry, 49, 535, 578 (1930).
(45) T w ym an an d Allsopp, “ T he Practice of A bsorption S pectro
p h o to m etry ," p. 18, London, A dam Hilger, 1934.
(46) Ibid., p. 26.
(47) Ibid., p. 41.
(48) V ierordt, “ Die A nw endung des S p ectralap p arates zur P h o to m etric der A bsorptionspectren und zur q u a n tita tiv e chem ischen A nalyse,” T ubingen, 1873.
(49) W ales, A m . D yestuff Reptr., 12, 751, 791, 855, 863 (1923).
(50) W eigert, “ Optische M ethoden der C hem ie,” C hap. V II, Leipzig, A kadem ische Verlagsgesellschaft, 1927.
(51) Zscheile, J . P hys. Chem., 38, 95 (1934).
(52) Zscheile, Hogncss, and Young, Ibid., 38, 1 (1934).
Re c e i v e dS e p tem b e r 18, 1936. P r e se n te d b efo re th e D iv is io n of P h y s ic a l a n d In o r g a n ic C h e m istr y , S y m p o s iu m on C o lo r im etry , a t th e 9 2 n d M e etin g of th e A m eric a n C h e m ica l S o c ie ty , P ittsb u r g h , P a ., S e p te m b e r 7 to 11, 193S.
D eterm in ation o f D iarylam ines in R ubber C om positions
DAVID CRAIG, T h e B. F . G o o d rich C o m p an y , A kron, O hio
T
H IS paper deals w ith sem i-quantitative m ethods for the determ ination of certain typical diarylam ine age resisters— nam ely, diphenylam ine, phenyl-/3-naphthylamine, an d A7,Ar'-diphenyl-p-phenylenediam ine. I t serves as an essential feature of a stu d y of th e general behavior of these m aterials in rubber to be discussed in a subsequent comm unication.
Previous m ethods for th e detection and estim ation of rubber age resisters have been of lim ited though of useful application. T h u s E nd6 (3) described th e colors produced by th e action of sulfuric acid on a large num ber of commercial age resisters unm ixed w ith rubber. L ater E ndo (4) published papers dealing w ith th e action of concentrated nitric acid and E rd m an reagent (solution of 0.5 ml. of concentrated nitric acid in 100 ml. of concentrated sulfuric acid) as well as of M andelin reagent (solution of 1 gram of am m onium v an a
d ate in 200 gram s of concentrated sulfuric acid) an d of con
centrated sulfuric acid containing 1 per cent of arsenic acid.
Tests similar to those of Ends have been used in this labora
tory. For example, a good test for the presence of N,N'-di-0- naphthyl-p-phenylenediamine in a rubber mixture involves cutting a gram of the stock into fine pieces and covering them with concentrated sulfuric acid. After several minutes the acid layer, which is usually straw-colored, is decanted and treated with one drop of a 10 per cent mixture of concentrated nitric acid in concentrated sulfuric acid. If the diamine is .present a blue color appears immediately. The presence of N ,N '-di- phenyl-p-phenylenediamine is attended by the formation of a red color under these conditions. This-test is not vitiated by the presence of the usual quantities of diphenylamine or the phenyl- naphthylamines which may be present in the rubber. If neither a red nor a purple color is produced it is probably safe to say th a t less than 0.1 per cent of either diamine is present. Interfering substances sometimes produce colors with sulfuric acid alone—for example, 2,4-diaminodiphenylamine produces a deep red which changes somewhat when nitric acid is added.
U sually age resisters cause staining of th e surface of light- colored stocks during cure or during aging. T he colors pro
duced, especially those arising from exposure to light, are m ore o r less characteristic of th e age resister present. K ir- chof (6) has suggested tl*e use of m etallic salts w ith or w ith
o u t exposure to light for th e purpose of distinguishing certain antioxidants, b u t this suggestion seems n o t to have been adopted. M orris (7) has published preliminary' d a ta on th e fluorescent colors of several antioxidants.
Sometimes an age resister will bloom in sufficient am ounts so th a t it can be removed, purified, if necessary, and identified
by th e usual qualitative m ethods, such as th e determ ination of th e m elting poin t a n d mixed m elting p o in t w ith an au th en tic specimen. T hus Ar,Ar/-diphenyl-p-phenylenedi- am ine, phenyl-j0-naphthylam ine, p-phenylphenol, p-hydroxy- iY-phenylmorpholine, p-triphenylm ethyldiphenylam ine, and others have been detected an d identified. In a sim ilar m an
ner antioxidants have been recognized occasionally in the sedim ent from rubber cements. No one seems to have sub
jected rubber to very low tem peratures for th e specific purpose of causing age-resister bloom, although icebox tem peratures have been tried occasionally. A microscope for th e ex
am ination of rubber surfaces is of great help in identifying th e presence of certain antioxidants.
T he isolation of age resisters from rubber b y extraction w ith solvents (1, 6) has been infrequently reported in the literature, including those which are rem ovable from crude ru b b er by extraction w ith acetone.
A t th e beginning of th e present work it was found possible to rem ove phenyl-/3-naphthylam ine from rubber b y extrac
tion w ith acetone or w ith constant-boiling m ethanol-benzene m ixture. T he am ine could th e n be steam -distilled from the dried ex tra ct or it could be p recipitated from hexane or ben
zene solution as phenyl-jS-naphthylam m onium chloride. T he extracts, however, were ra th e r intractable, an d the recovery of am ine was low in all experim ents.
A m uch b e tte r m ethod involved th e steam -distillation of diarylam ines from rubber. T he prelim inary results were successful and seem to be of sufficient in te re st to w arra n t publication. T h u s it is possible to isolate such am ines as phenyl-/3-naphthylamine, iVjA^-diphenyl-p-phenylenediamine, an d diphenylam ine from rubber as th e pure com pounds or in the form of suitable derivatives.
P r o c e d u r e s
De t e r m i n a t i o n o p Pi i e n y l-|3 -n a p h t h y l a m i n te. Aside from th e rem oval of sulfur a n d o ther alkali-soluble m aterials by reaction w ith sodium hydroxide, th e only reactions in
volved in this determ ination are th e form ation of th e hydro
chloride of th e am ine and the subsequent hydrolysis of the hydrochloride.
The stock is preferably either ground to pass a 14-mesh screen or is sheeted on a tight mill. For batches containing between 1 and 2 per cent of antioxidant a 30-gram sample is used, but for higher concentrations the weight can be reduced correspondingly.
The 30-gram sample is refluxed with 600 ml. of water and 7.5
FEBRUARY 15, 1937 ANALYTICAL E D IT IO N 57 grams of sodium hydroxide for 2 hours. The hot mother liquor
is then decanted or removed by filtering. After washing the stock with a small amount of hot water it is refluxed with 600 ml.
of water and 1 gram of sodium hydroxide. The extracted stock is then transferred to the distillation tube, A, shown in Figure 1, in the neck of which is a filter plug of glass wool. The stock is finally washed with a small amount of hot water. The tube is then submerged in the hot oil bath, D, and steam is introduced a t 170° to 180° a t such a rate th a t 4 to 4.5 liters of distillate are collected during 2.5 to 3 hours. I t passes through an air con
denser into a receiver cooled with water.
The d istillate has a characteristic rubber-like odor. I t is neutral to litm us. T he suspended m a tte r th a t is present is nearly colorless an d contains, in addition to th e anti- oxidant, a q u a n tity of stearic acid an d o th e r m aterial soluble in alkali, presum ably f a tty acids. T he suspended m a tte r is easily filtered off, especially afte r it has stood overnight, which is th e usual practice.
For this operation a 60-mm. Büchner funnel with a No. 2 Whatman filter paper has been found convenient. When as much as possible of the water has been removed by suction, the Büchner funnel is transferred to a tared 50-ml. suction flask and the solid m atter, including th a t deposited in the air condenser and the receiving flask, is rinsed through the funnel into the suc
tion flask with 75 ml. of ether from a 10-ml. pipet. Much of the ether is lost through evaporation during the rinsing operation. If 0.5 cc. or.more of water is collected in the suction flask, it is re
moved by means of a pipet and washed with a little ether, using the pipet also for this washing operation. The wash ether is then added to the suction flask and all the ether is evaporated on the steam plate. Small amounts of water can be removed by adding 10 ml. of benzene and evaporating until the benzene is removed. The last traces of water are removed by heating for 20 minutes in an oven a t 120° to 130°.
The solid remaining, to be referred to as the “residue from the ether evaporation,” is weighed to 0.01 gram, and dissolved in 10 to 20 ml. of warm hexane. A stream of hydrogen chloride is al
lowed to play for a few minutes over the surface of the solution while it cools to room temperature. The hydrogen chloride causes phenyI-/S-napbthylamine hydrochloride to precipitate as a clot th a t clings firmly to the bottom and walls of the flask.
The precipitation is complete when the mother liquor becomes clear or has stood for 2 hours. The m other liquor is then poured into another tared flask and the hydrochloride washed with a few milliliters of hexane. The washings are united with the mother liquor and evaporated to dryness, finally, in the oven at 120° to 130° for 20 minutes. The residue, known as the “residue from the hexane evaporation,” is then weighed to 0.01 gram. It contains nonbasic materials.
The flask containing the hydrochloride is quickly evacuated in a vacuum desiccator and weighed to 0.01 gram. The hydro
chloride is not pure, but seems to contain traces of free hydro
chloric acid and possibly traces of other difficultly volatile mate
rial. I t is hygroscopic. In order to determine the phenyl-/3- naphthylamine content, the hydrochloride in hydrolyzed. This is easily done by boiling with 10 ml. of benzene and 10 ml. of water until the solid has dissolved. The water layer is then separated and washed with ether by means of a pipet. The ether washings are added to the benzene layer, which is then evaporated on the steam plate until the solvent is removed.
The last traces of water are removed by heating for 20 minutes in an oven a t 120° to 130°. The residue is phenyl-/S-naphthvl- amine, and is weighed to 0.01 gram.
A description in m ore detail of th e ap p aratu s shown in Figure 1 follows.
The distillation tube, A, is made by bending a piece of Pyrex tubing 30 cm. (12 inches) long and 4.4 cm. (1.75 inches) in outside diameter, so th a t it can be submerged in the oil bath, D. A neck made from a piece of 1.25 cm. (0.5 inch) outside diameter tubing is attached and bent so th a t it can be connected to the short air condenser, E. The thermometer well in A is inserted by means of a ring seal. The brass superheater, B, of the type sold by the Fisher Scientific Company, is fitted with a thermometer well and 0.6-cm. (0.25 inch) copper tubing connections. The steam generator, C, is a 20-liter (5-gallon) steel drum. Service line steam is unsatisfactory because of admixed oxygen. I h e still head on C has a bulb of 300-ml. volume. The steam generator, C, the superheater, B, and the oil bath, D, are heated with gas burners. The receiver, F, is a 5-liter balloon flask cooled with
tap water. The various parts of the apparatus are connected by means of rubber tubing or rubber stoppers.
I I t is probable th a t th e precision an d speed of conducting th e determ inations would be im proved by th e use of small all-glass apparatus, th u s m aking it possible to use small samples.
D e t e r m i n a t i o n o f ¿Y ,A r '- D iP H E N Y ir - p - P H E N T L E N E D i- a m i n e . The procedure is the sam e as th a t for phenyl-/?- naphthylam ine up to th e poin t of th e hydrochloride pre
cipitation. H ere benzene m u st be su b stitu te d for hexane, and benzene m ust be used instead of eth er to w ash th e p roduct into th e precipitation flask.
Since the hydrochloride does not adhere well to the precipita
tion flask, filtration is necessary in order to separate the mother liquor. The hydrochloride Ls collected in the precipitation flask and hydrolyzed by boiling for several minutes with a mixture of 5 ml. of water, 3 ml. of concentrated ammonium hydroxide, and 5 ml. of benzene. The water layer is separated by means of a pipet and washed with a few milliliters of benzene. The ben
zene layers are united and evaporated to dryness. Finally the residue is heated for 20 minutes a t 120° to 130° C. and weighed.
I t should melt a t approximately 145°.
D e t e r m i n a t i o n o f D i p i i e n y l a m i n e . T he procedure given for phenyl-jS-naphthyl&mine can also be used for di- phenylam ine if due allowance is m ade for vo latility and for solubility in th e extracting solutions. I t is m ore convenient, however, since diphenylam ine is volatile in steam a t 100°, to use a modification of Cook’s m ethod (2) for diphenyl
am ine in smokeless powder.
A 30-gram sample of the stock, 5 grams of sodium hydroxide, and 3 liters of water are distilled in a 5-liter distilling flask until 4 liters of distillate have been collected. W ater is added to the distillation flask from time to tim e as necessary. The diphenyl
amine is extracted from the distillate with three 75-ml. portions of petroleum ether. The extracts are collected in a tared 150-ml.
flask and evaporated to a volume of about 5 ml. Five milliliters of ether and 8 ml. of a 10 per cent solution of bromine in carbon tetrachloride are added. The solvents are boiled off and the residue is heated in a current of air a t about 80° during 20 minutes. I t is then dissolved in 5 to 10 ml. of hot benzene.
The benzene is removed in a current of air a t about 80° and heated to constant weight a t this temperature. The melting point of the residue, which is 2,2',4,4'-tetrabromodiphenylamine, should be between 178° and 185°. A low melting point usually indi
cates incomplete bromination, although aged rubber usually gives a product of somewhat lower melting point.
N o te s o n P r o c e d u r e s
Several solvents were used for extracting small am ounts of th e diarylam ines from w ater. E th e r was objectionable be
cause of its high solubility in w ater. H exane and petroleum eth er (b. p. 40° to 50°) were satisfactory for phenyl- /3-naphthvlamine and diphenylam ine. Ar,Ar'-diphenyl-p-
