V O L U M E 27, N O . 10, O C T O B E R 1 9 5 5 The magnitude of electron donation by the same group substitutions in the ferroine chelation, is greater than that for the cuproine reaction. I n the latter case the percentage increase is phenoxy, 13.5; < ethyl, 20; < phenyl, 67.5. This decrease in effective modulation in the case of cuproine chelation is possibly due to steric hindrance. These conclusions have been substantiated ( I ) . The influence of substitutions by these newly described phenanthrolines does not promote alterations in the wave length of maximum absorption, A(,, ), which are accurately additive, as was found in the case of corresponding methyl group substitutions (2). Assuming that synthesis of 4,7-diethyl-lJ10-phenanthroline may be provided with equal facilitv to the 4,7-dimethyl substitutions, the former is much to be preferred ( E = 15020 as compared to 14000), for the ferroine reaction. The preferred cuproine reacting ligand of those newly described is 4,7-diethyl-l,IO-phenanthroline, ( E = 8675). This sensitivity does not approach the sensitivity of bathophenanthroline, [4,7-diphenyl-l,lO-phenanthroline, ( E = 12140)], or that of bathocuproine, [2,9-dimethyl4,7-diphenyl-l,10-phenanthrolineJ ( E = 14200)l. The results obtained in modulation of physical constants of the chelates of the modified 1,lO-phenanthrolines are not in accord
1575 with the known electron donor properties of the modifying substituent groups employed. An explanation of these facts a t present is not capable of interpretation. SUMMARY
Eleven newly synthesized substituted 1.10-phenanthrolines have been studied rrith determination of the wave length of maximum absorption and molecular extinction coefficients of their iron(I1) and copper(1) complex formulations. The new substitutions for replaceable hydrogens in the parent ligand are ethyl, methoxy, phenoxy, pyrido, and cpclohexeno groups. The hydrogen replacements are in the 2,9 position, the 4 and 5 positions, the 4,7 positions, and the 5,6 substitutions. LITERATURE CITED
(1) Gillis, J., Anal. Chim.Acta, 8 , 113 (1953). (2) McCurdy, W.,and Smith, G . Frederick, A n a l ~ s t77, , 418 (1953). (3) Smith, G.Frederick, ANAL.CHEM.,26, 1534 (1954). (4) Smith, G.Frederick, and Brandt, W. W., Zbid.,21, 948 (1949). (5) Smith, G. Frederick, and Wilkins, D. H., Anal. Chim. Acta, 9. 418 (1953). RECEIVED for review January 17, 1955. Accepted June 20, 1955. Presented before the Division of Analytical Chemistry at the 125th meeting of the AMERICAN CHEMICAL SOCIETY, Kansas City, Kan., March 1954.
Identification of Accelerators and Antioxidants in Compounded Rubber Products M. J. BROCK and GEORGE D. LOUTH Chemical and Physical Research Laboratories, The Firestone Tire & Rubber Co., Akron 7T, O h i o
A new, simple, and comprehensive approach to the identification of the accelerators and antioxidants used in rubber products is described. This procedure is unusual in that it utilizes the tendency of accelerators to decompose during mixing or vulcanization, or upon extraction from compounded stocks. The accelerator fragments are isolated using distillation and liquidliquid extraction principles. The identification of the fragments is simple, rapid, and unambiguous. The accelerators used are then determined from a knowledge of the decomposition behavior of known compounds. The antioxidants are recovered unchanged and can be identified by their ultraviolet absorption cbaracteristics and color tests. This method is applicable to all types of rubber products, including cements, and is particularly valuable in the identification of accelerators which decompose upon mixing, vulcanization, or extraction. Most of the commonly used accelerators and amine antioxidants can be classified or identified by the analytical scheme described. The procedure will be useful in identifying many accelerators and antioxidants which may be developed in the future.
T
H E identification of rubber compounding ingredients in the past has been accomplished mainly through the use of color reactions (5, 7, 10, 14, 19, 21). Recently these identifications have been made more positive through the use of chromatographic separation techniques ( 2 ) and characterization by ultraviolet (8, 12, 13) and infrared ( 1 5 ) spectrophotometry. Parker and Berriman ( 1 7 ) have recently published an excellent comprehensive fitudy on the chromatographic determination of accelerators and antioxidants in vulcanized rubber, including data on many of the acceleratorq and antioxidants which are widely used
today. A complete characterization of the composition of an unknown vulcanizate necessitates a detailed know ledge of the chromatographic behavior, not only of all the commercially used accelerators and antioxidants, but also of their decomposition products which are produced during vulcanization. Therefore, considerable background experience is necessary before the chromatographic technique can be used successfully. The tendency of accelerators to decompose or react during mixing, vulcanization, or extraction has plagued many analysts in the past and has resulted in loss of accelerators, or their conversion to other chemical compounds. Examples of some of the difficulties n-hich have been encountered are given below. Parker and Berriman ( 1 7 ) have shown that stearic acid interferes Xvith the detection of bis(dimethylthiocarbamoy1) disulfide (Methyl Tuads) in vulcanized rubber. Similarly, bis(diethy1thiocarbamoyl) disulfide (Ethyl Tuads) and bis(dimethylthiocarbamoy1) sulfide (Rlonex) could not be detected by these investigators after vulcanization. The conversion of his( dimethylthiocarbamoyl) disulfide to zinc dimethyldithiocarbamate (Methyl Zimate) in the presence of zinc oxide during vulcanization has been observed by Du Fraisae and Jarrijon ( 9 ) , Mann (15), and others (17, $1). Mann (15)states that he n m unable to detect either the piperidine salt of 1-piperidinecarbodithioic acid (Pip-Pip) or zinc pentamethylenedithiocarbamate (Pipazate) in vulcanizates where the former was used as an accelerator. Mann ( 1 5 ) has also observed (Santocure) is conthat N-cyclohexyl-2-benzothiazolesulfenamide verted to 2-mercaptobenzothiazole (XIBT) during vulcanization. The cyclohexylamine fragment of the molecule could not be found after vulcanization. Du Fraisse and Houpillart (8)identified both 2-mercaptobenzothiazole and 2,2'-dithiobisbenzothiazole (MBTS) in vulcanized rubber compounds which originally contained only one of these accelerators. They failed to detect diphenylguanidine (DPG) in
A N A L Y T I C A L CHEMISTRY
1576 acetone extracts of vulcanizcd rubber cornpounds accelerated with diphenylguanidine (8). The procedure proposed here simplifies these prohlems by using an extraction technique nhich promotes decomposition of the accelerators and collects the fragments obtained.
Table 11. Antioxidants Studied lbbreriation or Trade Name 'IgoRite Stalite RLE
4bbrevia-
,,
tion or Trade Name MBT UBTS zenite Special santooure (CBS)
Chelnicirl Nanrc Source R. T. Vandcihilt Mixture of mono- &d diheptyldiphenylamines Aoeto?e-diphenylhmino condensation Naugatuol
y".%)
'antoRex 1 I V
SentoRex B
szntofiex BX pbilito rilermonex TlierrnaHer 4
NOB8
6~Eth~r~-il,%diB~dr~-2,2.4-
Lrimethyliluinoline G-Phenyl-I ,Z-dihydru-2.2,4trimethylwinoline santonox B p ~ DPPD ~ > ~ N,Y'-diphenvlethylenedi&mine Di-;o-methoxydiphenylamine Thermoffex plus PBN.4 and D P P D
Monssnto Monsnnto Monsanto C. P. Hall Du Pant D u l'ont
______._
SW&l
Vulkhoit
4z IBS CPBS
*PP
CDETS
Firestone ( I , 901
TauEstud
Nauqatuok
124
Nmgatuol;
hIaner Methyl
R. T. Vanderbilt
Ethyl Tusds Tetrone A
R. T. Vanderhilt
Tuads
ZIetlryl zitnate Ethyl zimstc
D u Pont
Zino dimethyldithiooarbamate
R. T. T'anderbilt
Zino diethyiditbioes*l.bhmate
R. T . Vs*ndeihilt
pipazate
Nawatuek I\ronsa.nto
Philoure
Ethylac
Phillips Petrolcull, hronshnto D u Pant D u Font Shardes
SR.4 No. 2
Du Pont
pip-pip
113 DPG DOTG
TPG
.4 perusal of 1 pounding shows UI ~ L L L C U I I ~ ~ L LwL Ic "1 iwir UI uic W L L Y W U L ~ components: a tliiazale (or related compound), an amine, a guanidine, and carbon disulfide. The proposed procedure isolates and identifies each of these components along with the amine antioxidant used. The accelerator which was present in the rubber compound can then be determined, in most instances, from a knowledge of the fragments obtained from known eccelerntors. Consequently, many of the difficulties encountered by previous investigators are eliminated. Easily decomposed clcceleratars can be classified or identified, as the snslyticsl scheme is designed primarily to isolate and identify the decomposition product,? and not the original accelerators. The extraction medium used must not only promote the accelerator decomposition hut it must also be efficient in extracting the decomposed fragments or unchanged oompounding ingredionts. Humphrey (fl)has shown that a mixture of benzene and aqueous hydrochloric acid faoilitates the removal of the ertmctionresistant guanidine-type acoelerators. I n the present "-ark a niixture of ethyl alcohol and 1N aqueous hydrochloric acid vas found to be an excellent estraction medium, in t h a t it successfully decomposes unstahle accelerators and recovers the guanidine and ant,ioxidantsunchanged. The extracted components are subsequently separated using distillation and liquid-liquid estrsction techniques. The sepa, rations involved are complete and little interference is encountered in the identification of the products.
Sorelca x-ray diffraction equipment was used to iaentity tile crystalline products isolated. The photographic technique was employed, using copper X-dpha radiation. A Beokman Model DU ultraviolet spectrophotometer was used in the identification of Some of the uroducts. Absolute ethyl alrohol u.as the solvent used. The apparatus used for extracting rubber compounds under rcflur is shown m Figure 1. It. consists of a 1-liter, single-necked, round-bottomed flask. fitted with a Chisen-tvne adaDter nhich connects i t to an air'inlet tuhe. and to a H&kikins h e reflux
~
~
~~~~
ing air. T h e lower end of this "tube-dips below the liquid level in the flask. Connected to the outlct t,ithe of the reflux condenser is another ga;Rs-aashing tuhe which contains 50 ml. of 10% aqueous copper sulfate to remove any hydrogen sulfide generated during ihe refluxing period. A third gas-washing tube containing 0.2V sleoholic sodium hydroxide is used t o trap the carbon disulfide t h a t may he liberated from the ruhhrr campound. This tuhe is connected to a vacuum line.
F i g u r e 1. A p p a r a t u s for extractive decomposition of rubber c o m p o u n d i n g m a t e r i a l s
V O L U M E 27, NO. 10, O C T O B E R 1 9 5 5
1571
Except for gum rubber tubing connecting the gas-washing tubes, mound-Elass joints are used throughout this assembly rand the distillationdpparatus described below. For distillation of the amines and their absolption in acid, a conventional tvue of rtumratus mav be used. such as t,hat, illustrated in Fig&; 2. A'kodified cdindrical Kjeldahl spray trap ir inserted between the distillation flask and the condenser.
Extraction of Antioxidant and Accelerator Fragments from Rubber Products. The rubber products are repared in the usual m y by mixing and sheeting out on tl ruEber mill. Vulcaniaates can a180 be ground in a Tl'iley mill. Cements are prepared by evaporation to the solids, referably in a vacuum oven. The solids can then be mixed a n l s h e e t e d out. Compounded latex and latex products may he prepared in the same way.
MATERIALS
.411 reagents and solvents used were analytical grade. The accelerators and antioxidants used in this study were all commercial grade. They are listed in Tables I and 11, with the trade names, chemical names, and suppliers. These accelerators and antioxidants were cornpounded in the test formula given in Table 111. N-phenyl-Znapbtbylamine (PBNA) NZS used as an antioxidant in some of the stocks compounded for accelerator identification. N-oyclobeayl-2-benzothiazolesulfenamide was the accelerator used for the stocks containing different antioxidants.
Table 111. C o m p o u n d i n g Formula Material Part. Natural rubber 100.0 Carbon black 5n.n Zino oxide 3.0 Stearic acid 3.0 Paraflu. (softener)* 4.0 Sulfur 2.3 Antioxidant 1 . n to 1 .E .4eee~erator a 0 . 5 t o 1 .o 411 stooks, except TPG, bured'50 minutes at 280' Ii. TPG cured 00 minutes at 298' F. a Supplied by C. P. Hall. 6 Triphenylguanidine oompounded at 1.5 parts.
F i g u r e 2.
A p p a r a t u s foor. distilling volatile amines
Place 15 to 20 grams of tho prepared rubber product in the Miter, sinele-necked, round-bottomed flask. and add 100 ml. of
PROCEDURE
The rubber vulcanizates containing the different antioxidants and accelerators listed in Tables I and I1 were analyzed by the procedure given below. The results of the accelerator analyses m e shown in Table IV. All of the antioxidants studied were iEolated in the neutral or basic fraction and successfully identified.
into a 500-ml. sin&-necked, round-bottomed fl&k. Wash the rubber vesidue with 100 ml. of water and add the washings to the main snhbinn. ~
T a b l e IV. Thiazole Thiszolesulfenamid~ Santoeure (CBS) XOBS S ~ e o i a l X ulkaeit A 2 l-augatuek 124 IBS CPBB Thiooarbamvlsulfenhmid~ CDETS
"..~ ...."
XBT. \IBT 3IBT XBT
... ?JBT es
.. .
...
...
. .. ... ... ...
...
...
... ...
MET MBT ... .~'.h,.+..,".,. :, 16.. I
mDr*snt*n r.ll
Results < co~nponentsIdentified Carb Amine disulf
{
Diethylamine Dioyoiohexylamine Isopropylamine Piperidine
?PRS
Cyclohexylamine Diethylamine
JDETS
..... .... .... Dimethylaniine Dimethylamine Dimethyllapine Diethylamine Diethylamine Pioeridine Piberidine l'iwridine
.....
..... .....
Diethylamine Dlmeth&Ae 4...... t h _.
rI._...._ li.tilld n .lol I. ~
.... 1'"s. 1'08.
PO%. POS.
....
....
r Methyl Zimate mate mate *one A mne A mne A
POS.
PO$. POS.
PO%.
Ne& Seg.
Nee.
....
.... DPG TPG DOTG
. .. . ....
DFG or DPG salt TPG 01 TPG salt DOTG or DOTG salt
Etlrylho or Vulkaeit AZ SR.4 No. 2 or benrothisrole-DPG mixture Icrt-Butylsulfenyldimethyldithioearbamat~ amine and prowdore modified t o isolate and identify mercaptan 88 memury d t . Xeg.
Neg. PO&
DFG
ANALYTICAL CHEMISTRY
1578 Table V.
Separation of Nonvolatile, Neutral, Basic, and Acidic Materials Alkaline Aqueous Solution, Free of Alcohol and Volatile Amine
I
Dilute with water, cool, and extract with chloroform
I
I
I I
Chloroform Solution
Aqueous Basic Solution
I
Make acid with HCI soln., extract with chloroform
Extract with 1N HC1 I Chloroform Solution
I Aqueous Acid Solution
Dry with KazSOd, evaporate solvent
Make alkaline with NaOH s o h , extract with chloroform
I
I
Neutral Compounds (Antioxidants, etc.)
I
I 1 Chloroform Solution I Dry with NazSO4. evaporate solvent
I
I
rlqueous Acid Solution
1
I
Discard
Aqueous Alkaline Solution
I
Chloroform Solution
I
Dry with NazSOa,
evaporate solvent
I
Acidic Compounds (Thiazoles, etc.)
I
Discard
Basic Compounds (Guanidines, etc.) Table VI.
Ultraviolet Absorption Characteristics of Amine Antioxidants and MBT
The basic fraction will contain the guanidines, which can he identified by their x-ray diffraction diagrams. The 253 .. principal diffraction maxima together 2'8 258 .. 288(11) .. .. 281 with the relative intensities of di288 3lO(IIIs) 225 .... 340(IIIS) .. 236 253(II) 2'88 phenylguanidine, triph e n y l g u a ni d i n e 3IO(III) 272(II) 232 .. 283 (TPG), and di-o- t ol y 1- g u a n i d i n e 350(IIs) 300 .. .. .... .. .... 2X(I) 278 310(II) (DOTG) are listed in Table VII. I n 310(III) 219 208(I) 256(II) 278 294(IIIs) 212(I) 250(II) 218 272 order to eliminate interferences due to 288(11) 253 208(I) 245(IIIs) 231 polymorphism, all samples were recrps282 300(11)' '31O(II) 232 272(III) 220(I) 325 ( I ) 2'34 230(II) 238(III) .. .... 275 tallized, by evaporation to dryness from For each antioxidant maxima are designated I , 11, and 111 in descending order of intensitb. chloroform, in a 110' C. oven before the a s denotes slight maximum. x-ray data mere obtained. The acidic fraction will contain 2mercaptobenzothiazole, if a benzothiazole-type accelerator is used. 2-MerDistillation, Recovery, and Identification of Amine. l\Zake the captobenzothiazole is identified by its ultraviolet absorption charcooled alcoholic-acid solution alkaline with approximately 2 5 5 acteristics given in Table VI. Other acidic thiazoles and related aqueous sodium hydroxide solution, connect the flask to the diacidic materials will he isolated in this fraction. They can usually tillation apparatus, turn on the heating mantle, and distill the be identified by their ultraviolet absorption characteristics or amine into 35 ml. of 0.5.V aqueous hydrochloric acid. Continue the distillation until about 150 ml. of distillate are collected or x-ray diffraction diagrams. until most of the alcohol has been distilled from the mixture. Test for Carbon Disulfide. Transfer 10 ml. of the alcoholic The appearance of foam in the boiling, alkaline solution indicates that the hulk of the alcohol has been distilled. Concentrate the sodium hydroxide solution from the carbon disulfide absorption tube to a test tube and make the solution weakly acid by the distillate by boiling, and finally evaporate to dryness in a 110" C. addition of glacial acetic acid. Add 5 ml. of 1% aqueous ropper oven. Boil the dried residue gently with 2 or 3 ml. of chloroform sulfate and shake the tube thoroughly. If carbon disulfide is and filter to separate the more soluble amine hydrochloride from the sodium and ammonium chlorides. Evaporate the chloroform absent, a blue solution or a blue precipitate results. The blue precipitate will dissolve upon the addition of several drops of a t 70" C. and identify the dried amine hydrochloride by the concentrated nitric acid. When carbon disulfide is present, the x-ray diffraction method of Brock and Hannum (4). precipitate is yellowish green in color, and the addition of nitric Separation and Identification of Acid, Basic, and Neutral acid and reshaking leave a yellow precipitate of the copper Materials. Dilute the alkaline residue from which the amine xanthate. was distilled with 50 ml. of water, allow it to cool, and then subColor Tests for Antioxidants (6). TESTSOLUTIOKS. Stannic ject it to an extraction procedure for separating the nonvolatile Chloride Solution. Dissolve 14.7 ml. of fuming stannic chloride neutral, basic, and acidic materials. in anhydrous analytical reagent grade benzene and make the volume up to 250 ml. The liquid-liquid extraction procedure used is similar to Amy1 Sitrite Solution. Dilute 5 ml. of amyl nitrite to 100 ml. that employed by Braus, Middleton, and Ruchhoft (3)for sepawith benzene. rating the constituents of organic industrial wastes. A schematic TESTI. Stannic Chl0ride-~4my1 Kitrite Reaction. Dissolve approximately 1 mg. of the basic fraction in 5 ml. of benzene. diagram of the separation procedure is given in Table V. Add 1 ml. of stannic chloride solution and 3 drops of amyl nitrite The neutral fraction thus obtained normally contains the solution. A red precipitate indicates the presence of Stabilite. amine antioxidants, which can he identified by their ultraviolet TEST11. Stannic Chloride-Benzotrichloride Reaction. Disabsorption characteristics. Table VI lists the principal maxima solve approximately 1 mg. of the neutral fraction in 5 ml. of pure ethylene dichloride. Add 1 ml. of stannic chloride solution and and minima of the antioxidants included in this study. Color 2 drops of benzotrichloride. BLE (an acetone-diphenylamine tests are also helpful in identifying these antioxidants. The condensation product) gives a violet color with this test, differtests of Burchfield and Judy ( 5 ) as given below have been exentiating it from AgeRite Stalite (a mixture of mono- and ditremely useful. heptyldiphenylamines), which gives an insignificant yellow color. hfin., rnp
Maxima, 200-280 m p
Min., rnp
Afaxima, 250-300 m p 288(II)
Min., rnp
Maxima, 300-350 r n p
V O L U M E 27, NO. 10, O C T O B E R 1 9 5 5
1579
INTERPRETATION OF RESULTS
2,2’-dithiobisbenzothiazole, and 2-mercaptobenzothiazole, zinc
The accelerators used in the rubber product are determined by the fragments or combination of fragments identified. A knowledge of normal compounding practices will aid in reconstructing the accelerator system used. I n general, the detection of a thiazole, a guanidine or carbon disulfide will classify the accelerator as to type-namely, a thiazole, usually 2-mercaptobenzothiazole or a derivative, a guanidine, or a thiuramsulfide or dithiocarbamate. The specific amine identified, if any, designates the particular amine activator, sulfenamide, thiuram sulfide, or dithiocarbamate used. Table IV lists the accelerators, which are indicated by the components identified. Although the antioxidant is usually identified using ultraviolet absorption characteristics, some antioxidants are indicated by color reactions inherent in the procedure. Antioxidants containing S,iV’-diphenyl-p-phenylenediamine ( D P P D ) and di-pmethoxydiphenylamine (Thermoflex) are indicated by a green or blue-green color in the cooled extraction medium. When Thermoflex is present, the color turns red as the extraction medium is made alkaline with sodium hydroxide. If D P P D is present, the alkaline solution is a yellow-brown color. Other color tests are used to distinguish betaeen antioGidants having similar ultraviolet absorption characteristics. Consequently BLE can be distinguished from AgeRite Stalite using color test I1 (stannic chloride-benzotrichloride reaction). N,N’-Diphenylethylenediamine (Stabilite) is the only one of the antioxidants studied which was recovered in the basic fraction. Stabilite can be identified in the presence of the guanidines using color test I (stannic chloride-amyl nitrite reaction). The xanthate test was found to give a good positive test for carbon disulfide in the case of thiuram and dithiocarbamate type accelerators, and a negative test with other accelerators. If a more sensitive test is used, accelerators other than thiurams and dithiocarbamates may give a slight positive test for carbon disulfide and confuse the identification.
derivative (Zenite Special). Similarly, no distinction can be made between thiuram sulfides and dithiocarbamates containing the same amine. This disadvantage is minimized in view of the fact that these accelerators undergo interconversion or decomposition during vulcanization, and identification of the original accelerator is always difficult. The identification of accelerators and antioxidants in aged vulcanizates introduces the problem of decomposition due to aging. Burchfield and Judy ( 5 )have recognized this difficulty in the identification of antioxidants. S o detailed study of aged samples was made in the present investigation. It was observed, however, that 2-(morpholinothio)benzothiazole (KOBS Special) could not be identified in aged stocks and was difficult to identify in freshly cured vulcanizates. Santocure and DPG, on the other hand, could be identified easily in vulcanizates aged as long as 3 1 ears. Antioxidants isolated from aged stocks gave ultraviolet absoiption characteristics which were generally more difficult to identify. 1\Ii\tures of antioxidants, as v-ould be expected, are more difficult to identify, but they can usually be resolved by careful examination of the ultraviolet absorption data and by using color teqts. Mixtures of accelerators generally do not present too much of a problem, as their decomposition products usually occur in different fractions.
DISCUSSION
One disadvantage of this procedure is that some accelerators give identical decomposition products. As a result, it is impossible to differentiate between 2-mercaptobenzothiazole,
Table VII.
X-Ray Diffraction Data of Guanidines
Diphenylguanidine (DPG) da
I/Iib
Triphenylguanidine (TPG) d 1,’Il 9.15 1.00 8 24 0.38 0.06 7.05 0.03 6.50 0.46 5.37 0.03 5.00 0.25 4.72 0.31 4.52 0.48 4.34 0 38 4.07 0.19 3.79 0 38 3.56 0.03 3.42 0 25 3.08 0 19 2 70
Di-o-tolylguanidine (DOTG) d I/Il 10.33 0.46 6.79 0.25 6.18 1 .oo 0.25 5.22 4.63 1.00 0.46 4.04 3.88 0.17 0.46 3.76 3.37 0.05 3.18 0.11 3.03 0.05 2 50 0.05 2.43 0 05 2.09 n 03
0.82 10.36 0.07 6.96 0.12 6.43 0.07 6.07 0.17 5.06 1.00 4.68 0.24 4.51 0.32 4.19 0.24 3.95 0.17 3.80 0.40 3,63 0.24 3.49 0.04 3.26 0.07 3.09 0.07 2.99 0.04 2.91 0.02 2.64 0.04 2.54 2.44 0.02 2.41 0.02 2.31 0.04 0.04 2.21 0.04 2.14 0.02 2.01 0.02 1.03 1.86 0.04 1 69 0 04 a d. Interplanar spacing in Angstrom units calculated from Bragg’s law x where d = X is wave length of characteristic C u K a radiation, and 0 is one half t h e angle of diffraction. b I/Ii. Relative intensity.
msin8’
To demonstrate the usefulness of the method in identifjing mixed accelerators and antioxidants, a viilcanizate R as prepared Containing Santocure, Ethyl Tuads, TPG, BLE, and Santoflex BX ( a mixture of 6-pheny1-1,2-dihj dro-2,2,4-trimethylquinoline and D P P D ) . ,411 five of the isolated fractions contained a t least one product. The identification of cyclohexylamine and MBT indicated the presence of Santocure. Diethylamine and carbon disulfide indicated the use of Ethyl Tuads or zinc diethyldithiocarbamate (Ethyl Zimate). T P G was identified in the basic fraction. BLE, DPPD, and 6-phenyl-1,2-dihydro-2,2,4-triniethylquinoline (Santofleu B) 1% eie all indicated to be present in the neutral fraction. As Santoflex B and D P P D partially obscure the ultraviolet absorption characteristics of BLE, the presence of the latter is confirmed by color test 11. The extracted portion of the softeners used in compounding m-ill usually be isolated in the neutral fraction. If these materials exhibit characteristic ultraviolet absorption, they could possibly interfere with the identification of the antioxidants. However, in the present study, no undue interference was encountered from either Paraflux or pine tar. This is primarily due to the fact that the aqueous alcohol extraction medium has little solvent power upon softeners that are insoluble in water. The presence of large amounts of oils, in oil-extended polymers, interferes seriously with the identification of antioxidants. I n such cases it may be necessary to resort to chromatographic techniques, before satisfactory results can be obtained. APPLICATIONS
The procedure described here is offered as a new approach to the determination of accelerators and antioxidants in rubber products. It has been applied successfully to both vulcanized and unvulcanized rubber compounds. I t is equally useful in the analysis of latex products and rubber cements. Data on 24 accelerators and 12 antioxidants have been obtained in this study. The method has been used also in the identification of new accelerators. The analytical scheme is such that it will include many of the accelerators which are likely to be developed in the near future-for example, the recently reported accelerator, tert-butylsulfenyldimethyl dithiocarbamate (Philcure 113) ( 18), has been identified using the method. The procedure was modified slightly to include the identification of the mercaptan portion of the molecule (see Table IV). The method will also have use in identifying some of the newer combinations of accelerators, such as the combination of the
1580
ANALYTICAL CHEMISTRY
thiuram and guanidine types which has been recommended for use in the vulcanization of neoprene rubber Type W (16). ACKNOWLEDGMENT
The authors wish to acknowledge the help of RIarjorie Jean Hannum in preparing this paper. Thanks are expressed to the Firestone Tire and Rubber Co. for permission to publish the results of this investigation. LITER4TURE CITED
(1) hlliger. G., and Harrison, S. R., India Rubber World, 123, 181-5 (1950). (2) Bellamy, L. J., Lawrie, J. H., and Press, E. W. S., Trans. I n s f . Rubber Ind.. 23. 15 (19491: 22. 308 (1947). (3) Braus, H., Middieton, F. hI., and Ruchhoft, C. C . , ANLL. CIIEM.,24, 1872 (1952). (4) Brock, M. J., and Hannum, &I.J., Ibid., 27, 1374 (1955). (5) Burchfield, H. P., and Judy, J. X . , Ibid., 19, 786 (1947). (6) Carr, E. L., Smith, G. E. P., J r . , and Alliger, G., J . Oiy. Chern.. 14, 921 (1949). (7) Deal, .4.J. A., Trans. Inst. Rubber Ind.,23, 148 (1947).
(8) Du Fraisse, C., and Houpillart, J.. Rev. gdn. caoutchouc, 19, 207 (1942). (9) Du Fraisse, C., and Jarrijon, A., Rubber Chem. and Technol., 17, 941 (1944). (IO) Endoh, H., J . SOC.Chem. Ind. Japan, Suppl. Binding, 38, 618 (1935). (11) Humphrey, B. .J., ISD. ENG.CHEM.,ANAL. ED., 8, 153 (1936). (12) Jarrijon, A., Rev. gin. caoutchouc, 18, 217 (1941). (13) Koch, H. P., J . Chem. Soc., 1949, 401. (14) Kreps, V., J . Rubber Ind. (U.S.S.R.), 9, 45 (1933). (15) Mann, J., Trans. Inst. Rubber Ind., 27, 232 (1951). and Thompson, D. C., Division of Rubber Chem(16) Murray, R. M., istry, BM.CHEM.SOC.,Louisville, Ky., April 14, 1954; abstract' in India Rubber World, 130, 7 1 (1954). (17) Parker, C. A, and Berriman, J. M., Trans. Inst. Rubber Ind.. 28, 279 (1952). (18) Kailsbach, H. E., Svetlik, J. F., Biard, C. C., and Louthan, R. P., Ind. Eng. Chem., 47, 352 (1955). (19) Shimada, K., J . SOC.Chem. Ind. Japan, Suppl. Binding 36, 82 (1933); 36, 260 (1933). (20) Smith, G. E. P., Jr.. and others, J . Org. Chem., 14, 935 (1949). (21) Vaanderbilt Sews, 13, No. 1 , 24 (1947). RECEIVED for review September 27, 1954. Accepted April 18, 1959. Presented before the Division of Rubber Chemistry at the 126th Neeting of the A h I E R [ r A S C H E Y I C . A L s O C I E T Y . Xew YOrk, N. Y . , 1954.
4-Aminopyridine as Standard in Acidimetry CLAYTON E. VAN HALL and
K. G. STONE
Kedzie Chemical Laboratory, Michigan State University, East Lansing, M i r h .
4-Aminopyridine is a high melting (161" C.) nitrogen base with a dissociation constant of 1.3 X Methyl red indicator may be used with either O.1Nor 0 . W acid. The free base may be purified by recrystallization from toluene or benzene, or by sublimation at reduced pressure, and may be recovered easily after use. Standardization of acids with 4-aminopyridine yields normalities within 1 part per thousand of those found using sodium carhonate.
THE
search for new substances which have the required properties of primary standards is never ending (3). This is particularly true with respect t o bases suitable for the standardization of acids. The most recent suggest,ion, tris(hydros,vmethy1)aniinomethane ( 4 ) , requires a mixed indicat,or which is unsatisfactory in the hands of students ( 8 ) , because inexpeiieric*ed people have trouble with the color change. 4-Aminopyridine satisfies most of the requirements for a primary standard. 4-Aminopyridine was first synthesized by Camps ( 2 ) in 1902. The most widely used synthesis was suggested b!- Koenigs and Greiner (6) and is based on the hydrolytic cleavage of 4-pyridylpyridiniuni dichloride to yield hminopyridine and polynievixed t#arrymat,erial. Techniral grade 4-aminopyridine is also available (Reilly T a r ' and C h e m i d Corp., Indianapolis, Ind.) 4Aminopyridine is a weak, monoprotic base with an equivalent weight of 94.12. Tropsch (R) reported a dissociation constant of 1.3 X 10-5 a t 25" C. from conductivity measurements, and Albert ( 1 ) cited data which lead to a value of 1.6 X l O P . 4Aminopyridine is soluble in water and eth>-lalcohol, and moderately soluble in benzene, t,oluene, and chloroform (1). EXPERIMENTAL
4-Aminopyridine used in this work was prepared by the procedure of Koenigs and Greiner ( 6 ) , and was recrystallized from benzene, ground to a powder, and dried for 2 hours a t 105" C . before use. The melting point, was 161' C. with a heating rate of 2' t o 3" per minute. A 6-liter carboy of approsimately 0 . 1 4 hydrochloric acid solu-
tion was prepared by dilution of reagent grade concentrated arid and was protected against temperature changes. The exact normality was determined by the method of Kolthoff and Sandell ( 7 ) , using Mallinckrodt primary standard grade sodium cwbonate which was treated according to the directions of Kolthoff and Sandell ( 7 ) before use. The acid was also compared against carbonate-free sodium hydroxide solution, which had been standardized against potassium acid phthalate according to the method of Hillebrand, Lundell, Bright, and Hoffman (6). All titrations in this work were made with one volumetric, buret, which was calibrated at 25' C. in the normal wa). 4 1 1 other volumetric equipment used was calibrated if necessary. The concentrations of indicator solutions were those normall! used in analytical work. Indicator corrections were applied for phenolphthalein and bromocresol green used in the carbonate titration, but no correction was necessary with methyl red indicator. The p H titrations were made with a Beckman p H meter, Model H-2, equipped with a glass indicator electrode and :I calomel reference electrode. All distilled water was boiled freshly before use. All samples of 4-aminopyridine were dried a t 105" C. for 2 hours, unless otherwise stated. All weiahings were made with calibrated brass weights and were not corrected to vacuo. STABILITY OF 4-AMINOPYRIDINE
Quantitative experiments to show the stability of 4-aminop7 ridine were carried out in the following manner. Samples of the original recrystallized material (the material after standing 96 days a t room temperature, the material after 56 hours a t 105" C., ana some material kept in a closed bottle for 6 months) were dried 2 hours a t 105' C. and titrated with approximately 0.12Y hydrochloric acid solution. The normality of the acid was calculated assuming that all the samples of 4-aminopyridine were pure, and a value of 0.1020 was found from all samples. It, therefore, must be concluded that 4-aminopyridine is stable under the conditions described. The hygroscopicity of 4-aminopyridine was observed by placing a weighed portion in a weighing bottle, storing the sample and another weighing bottle in a beaker (covered to keep out the dust) a t room temperature, and weighing both the sample and the tare at frequent intervals. The tare was used to determine how much of the change was due to adsorption of moisture on the glass surface. The sample continuously lost weight over a period of 95 days, the total loss being 0.15%. During this time the temperature range was 22" to 31" C., and the relative hu-
