Reclaiming Agents for Synthetic Rubber

rubbers during and after World War II have presented an unprecedented series of problems to the reclaimed rubber industry. The ordinary reclaiming pro...
0 downloads 0 Views 1MB Size
Reclaiming Agents for Synthetic Rubber J

W. S. COOK, H. E. ALBERT, F. L. KILBOURNE, JR., AND G. E. P. SMITH, JR. The Firestone Tire & Rubber Company, Akron, Ohio

T h e large scale manufacture and use of synthetic rubbers during and after World War I1 have presented an unprecedented series of problems to the reclaimed rubber industry. The ordinary reclaiming processes used for years in the production of natural rubber reclaims have failed to produce acceptable products when applied to vulcanized synthetic rubber scraps. Results are reported which indicate that certain polyalkyl phenol sulfides are effective in small quantities as “reclaiming agents” or catalysts for the reclaiming of various vulcanized synthetic rubber scraps. Their use in the correct proportions, together with the proper swelling agents and tackifiers, has resulted in the production of reclaims from

ITH the advent of synthetic rubbers during the recent war emergency, the problem of reclaiming these new products became of vital interest to the reclaimer. After preliminary experiments were undertaken, it became evident that methods and materials used for manufacturing natural rubber reclaim would not be immediately applicable for reclaiming synthetic rubbers. It was therefore necessary t o develop new procedures, plasticizers, and chemicals in order that synthetic reclaim should be practical when the supply of natural rubber scrap dkappeared. Since the major portion of the production of synthetic rubbers was t o be of the butadiene-styrene type-that is, GR-S-active research was undertaken t o develop suitable means for reclaiming vulcanizate scrap obtained from this polymer. The first laboratory tests on reclaiming GR-S, using methods applied to natural rubber, resulted in a dry, hard, and brittle product which could scarcely be considered reclaimed rubber. Further work indicated that large amounts of oil plasticizers and tackifiers were necessary t o prepare a product possessing workability properties similar t o natural rubber reclaim. Even with such large amounts of oils and tackifiers the quality of the reclaim was poor. This product had a low hydrocarbon content and a high acetone extract because of the added oils. The problem then facing the reclaimer was the discovery of new and more active reclaiming agents or peptizers for GR-S scrap. A “reclaiming agent,” as the term is used in the present work, may be defined as a chemical which acts as a catalyst in hastening the process of reclaiming vulcanized natural or synthetic rubber scrap, These reclaiming agents should be distinguished from the oils and resins which are used as swelling agents, plasticizers, and tackifiers in the various reclaiming processes. AS catalysts or promoters, the reclaiming agents are most often employed together with the usual plasticizers and tackifiers. The effective reclaiming agents are required in much smaller quantities than the usual swelling oils and tack-producing resins. The use of organic chemicals as catalysts for the reclaiming processes for natural rubber began as early as 1910 when Lutz (19) described the use of aniline. More recently, the use of the reaction products of monoarylhydrazines with aldehydes and ketones has been reported ( 1 1 ) . The aromatic mercaptans as reclaiming agents for natural rubber have been covered by Garvey

.

GR-S and other synthetic elastomers which compare favorably with natural rubber reclaim in processability and compatibility, and in both cured and uncured physical properties. A study of the fate of sulfur during the reclaiming of GR-S vulcanizates indicated that the polyalkyl phenol sulfide reclaimink agent does not act by removing sulfur from the vulcanizate. O n the contrary, some of the sulfide sulfur appeared to combine with the vulcanizate during the reclaiming operation. It is postulated that these phenol sulfide reclaiming agents may act by catalyzing oxidative breakdown of the polymer vulcanizate, while at the same time they may inhibit further crosslinking reactions.

(91,Neal and Schaffer ( 2 2 ) ,and Gumlich and Ecker ( 1 2 ) . Dashei used various hydroxylamines, aliphatic amines, aliphatic poly amines, and their mercapto derivatives (6). As the result of wartime activity in the reclaiming of synthetic rubber in general and of GR-S in particular, the use of both long chain aliphatic amines and cyclohexylamine ( 2 8 ) has been reported. The aromatic mercaptans, as covered in the Garvey patent (9),also were found to have some activity in GR-S (68). These reclaiming agents, however, were later superseded by the more active cresol-sulfur chloride reaction products of Kirby and Steinle (18, 34). The use of aromatic mercaptans for reclaiming the various types of synthetic rubber scraps or vulcanizates has had a parallel development in England ( 2 , 3) and Germany (3, 1 4 ) . The English materials are known as the RRA’s and the German materials were sold under the name of Renacit. The chemical compositions of Renacit I, 11,and 111were, respectively, @-naphthyl,trichlorophenyl, and 9-anthryl mercaptans. Each succeeding material. was reported to have possessed a reduced irritating effect on the skin with approximately equivalent activity as a reclaiming agent for Buna S. As a result of a large number of tests carried out in this laboratory during the past several years, it has been found that the sulfides prepared from certain highly alkylated phenols are extremely active catalysts for reclaiming GR-S and other synthetic rubbers. The object of this paper is to report the preparation, evaluation, and structural correlation of these new materials with respect to their activity as reclaiming agents. PREPARATION OF RECLAIMING AGENTS

The phenol sulfides used in this investigation were prepared in a three-necked flask fitted with a reflux condenser, a mercury seal stirrer with electric motor drive, and a dropping funnel. The reactions were carried out a t the boiling points of the solvents used, either carbon tetrachloride or ethylene dichloride, and heat was applied by an electric heating mantle controlled by a variable voltage transformer. I n all cases, except where otherwise indicated, the ratio of 2 moles of alkylphenol to 1 mole of sulfur monochloride was used in the sulfide preparations. After the addition of the sulfur monochloride was complete, the mixture was allowed to stir for half an hour and the solvent was then distilled in vacuo, care being taken to remove the last traces without overheating the product. The crude mixture of sulfides of the alkylphenol formed by the above procedure was used without further purification unless

1194

1195

INDUSTRIAL AND ENGINEERING CHEMISTRY

July 1948

otherwise indicated. I n the following paragraphs this reaction mixture is referred t o as the alkylphenol sulfide. Many of these sulfides have not been recorded in the literature and it is expected that some of them will constitute a communication from this laboratory in the near future. RECLAIMING PROCEDURE

*

A pan heater or open steam type reclaim, made from 100% GR-S vulcanized tread stock, was chosen t o compare the various reclaiming agents t o be tested. The vulcanized stock was obtained by securing a factory lot of 100% GR-S tread compounds, curing in open steam, and then grinding t o pass a 5-mesh screen. No natural rubber contamination was permitted in the preparation of this scrap. A reclaiming recipe for thia scrap was developed which contained the following ingredients: 200 parts Dipentene fraction* (b.p. 173O to 201O C.) 12 parts Coumarone indene resina 12 parts Realaiming agent 3 parts Solvenol supplied by Hercules Powder Company. b Cumar i l / z MH, supplied by Barrett Division, Allied Chemical & Dye Corporation. GR-S tread scrap (5-mesh)

-The swelling agent, tackifier, and reclaiming agent were 'weighed out, heated together gently until solution was obtained and then incorporated into the scrap by use of a laboratory Baker-Perkins mixer. After mixing, the scraps were aged for a t least 6 hours before the reclaiming operation. The heating process was carried out in a laboratory pan heater; the scrap was subjected t o open steam of 175 pounds per square inch for 4 hours. After completion of the heating period, factory vacuum was applied to the stock in the heater for a period of 1 hour to ensure dryness. The softened scrap mass in biscuit form was then removed from the heater and allowed to cool to room temperature. Up to fifteen or more individual reclaims could be made in one cooking period. The softened biscuit from the heating operation was removed from the pan, blended, and massed by four passes through an open laboratory mill, to ensure uniformity, Approximately 50-gram portions of each biscuit in a group of reclaims were then refined three passes on a cool (110O F.) laboratory mill. This mill was equipped with a refiner knife and was tightened to the proper setting, so that a thickness of 0.005 inch was obtained when a piece of lead or solder was passed between the rolls. Each grou of reclaims was refined in consecutive series each pass throu the mill, in order that each reclaim sample 'in the group migtt receive the same refining treatment. EVALUATION OF RECLAIM SHEET

The evaluation of a reclaim sheet has been a problem peculiar to the reclaimer. This evaluation has been largely an art, based on the manual observation of the individual technician. I n this laboratory three essential properties have been observed and recorded in the evaluation of a refined reclaim: sheet thickness, body, and tack, the last two properties being estimated by hand evaluation. The over-all hand evaluation is often referred to in the industry as the "workability" of the reclaim. In addition t o the above properties, it has been shown during the course of the present work that valuable quantitative data may be obtained on the refined reclaim by the use of the Mooney plastometer ($1) and also by the determination of the tensile strength and elongation of the refined sheet. I n the latter case a 0.25-inch restricted area dumbbell was died out from the three-pass refined reclaim sheet and pulled on a small Scott tensile machine (Model X-5), with a range of 2 pounds a t a rate of 20 inches per minute. The data so obtained agree qualitatively with the previously used evaluation tests of thickness, body, and tack and with the over-all hand evaluation or workability of, the reclaim. These types of data, as exemplified by thickness, plasticity, and tensile properties of the uncured; refined reclaim sheet, tend to tie down the differences between reclaims to measurable quantitative amounts. However, these methods have not been used long enough t o establish completely their validity in commercial practice.

The thickness of the reclaim sheet was measured immediately after refining with a Randall-Stickney thickness gage. Measurements were made across the entire sheet at a point approximately 5 inches from the end of the sheet as it was cut from the roll by the refmer knife. The average value of sheet thickness, in inches, was considered t o be a measure of the softness of the reclaim, and consequently of the activity of the reclaiming agent. The body rating of the sheet was estimated by observing the stretch or elongation of the refined sheet, stretched by hand, and by the appearance and uniformity of this stretched sheet. The body was rated as very good (V-G), good (G), fair (F),poor (P), very poor (V-P), tough (T),and lacy (L), or as some combination of these. A good sheet must have satisfactory elongation without tearing, and a smooth uniform appearance. A poor sheet ha? poor stretch or elongation, tears, and may be nonuniform, tough, and lacy. I n a number of cases these normal body evaluations have been given numerical ratings as follows: V-G G G-

6 5

Very good Good Gqod Fair

F+

5-

These numerical ratings facilitate the formation of a quality index, by addition of the tack ratings (below), which should be approximately equivalent t o the over-all workability of the reclaim as referred t o in the reclaiming industry. The third property of the refined reclaim sheet was that of tack-defined as the tendency of the refined reclaim sheet t o adhere to itself. It was estimated by laying a portion of the refined reclaim sheet across the hand and then pressing the thumb and the f i s t finger of the hand together. When the thumb and first finger were spread apart, a small but definite force WM required t o separate the two adhering surfaces, A rating of 5 was given t o the force required t o separate a sheet of typical natural rubber whole tire reclaim. Milled crude rubber was given a rating of 10 and crude GR-S with no tack was given a rating of zero. This is essentially the same method and the same scale used previously (80)in this laboratory for the estimation of the tack in milled sheets of unvulcanized GR-S with and without added tack-producing compounds. Values between those assigned t o the controls mentioned above were estimated by the observer and could be duplicated easily by different independent observers with an accuracy of * 1 unit. ALIPHATIC AND AROMATIC MERCAPTANS AS RECLAIMING AGENTS

A comparison of the aliphatic and aromatic mercaptans t l reclaiming agents was made desirable by the activity reported for

TABLE I. MERCAPTANS Compound Control ~4,6-di-tert-butyl-3methylphenol sulfide) Phenol Phenyl mercaptan (thiophen01) o-Tolyl mercaptan (o-thiocresol) m-Tolyl mercaotan (m-thiocresbl) ' p-Tolyl mercaptan (p-thiocresol) n-Naphthyl mercaptan @-Naphthyl mercaptan Octyl mercaptan Decyl mercaptan Dodecyl mercaptan Tetradecyl mercaptan Hexadecyl mercaptan

Thickness 0.008 0.020

Body

Qual-

Mooney Value,

1ty

MS/4/ 212

Tack Index 4+

2+

1

41 69 56

0.012 P-F (3'/z)

2+

52

0.013 P-F (3l/d

2'

54

0.011

2'

G- (5)

P-L (2) 0.013 P-F- (3)

0.010 0.010

0.025 0.020 0.020 0.025

0.020

P-F (z1/z) F (4) F (4) P-L (2) P-L (2) P-L 2 ) P-L 12) P-L (2)

4-

3

1+ I+

I+

I+ 1+

52

47 52

70 66 68

67 64

~

1196 certain aromatic mercapLans (88). The controls contained 4,6-di-tert-butyl-3m e t h y l p h e n o l sulfide, the reaction product of 4,6-di-tertbutyl-3-methylphenol and sulfur monochloride. The first part of the data in Table I covers the aromatic mercaptans. Phenyl mercaptan (thiophenol) as the first member of this series was found t o be somewhat more active than phenol, its oxygen c o u n t e r p a r t . The isomeric tolyl mercaptans (thiocresols) showed no advantages over phenyl mercaptan, but the naphthyl mercaptans were somewhat more active. Compounds containing the naphthalene nucleus were usually slightly more active than the corresponding benzene compounds. The aliphatic mercaptans were very poor r e c l a i m i n g agents in this process, and much less active than the aromatie mercaptans. It wa6 apparent that the variation in the chain lengths of the aliphatic compounds had very little effect on their activity in this process. ALKYL PHENOL SULFIDES A S

RECLAIMING AGENTS

INDUSTRIAL AND ENGINEERING CHEMISTRY

Vol. 40, No. 7

TABLE11. ALKYLPHENOL SULFIDES AS RECLAIMING AGENTS Compound Control Phenol SzClz SaCln o-Cresol m-Cresol S&Iv

++ +

Thickness Body Methyl and Dimethyl Phenol Sul6deh 0,012 0.020

54

Methyl, Ethyl, and Trimethyl Phenol SulfideR 0.005 0 (51/2) 6+ 0.0091/, 0.0081/2 2+

36

0,020

4-

p' til +

+ ++

0.011 0,010

P+-T (11/z) P-L ( 2 )

0.008'/? 0.0081/z

P4)31/2)

O.O041/z

G - (4l/z)

Amyl, Diamyl, and Triamyl Phenol Sulfide6 Control 0.007 Phenol SzClz 0.018 p-tert-Amylphenol SzClz 0.020 P (1) 2,4-Di-tert-amylphenol SzCIz 0,010 F - (3l/z) 2,4,6-Tri-see-amylphenol SzCll 0,014 P-F (3l/z) 2,4-Di-sec-amylphenol C SzCb 0.010 F (7)

+

35 53 48 59 45 46 48

0.020

0.015 0.015 0.015 0.015

++ ++

5

Mooney Value, MS/4/212

1+ 2+ 2 I' 3+ 4 4-

0.015

Control SzClz Phenol rn-Cresol SzClz rn-Ethylphenol SzCln , p-Ethylphenol SzClz 3-Methyl-5-ethylphenol SzCL SzCL 3.5-Dimethylphenol 2,3,5-TrimethylphenoI SzCh

Tack

a?!

+ + +

Amyl and Dinrnsl Sulfides . Cyolohexanol . O.O061/z G (5) 0.011. P (3) 0.014 P-T (1) 0.0121/z P-F-T (31,'s) 0.0151/2 P-T (1) 0.009 F - (z1/z) 0.009 F- (3l/z)

21-

312+ 3 +-4 3 --4 6+

5+ 1'

I+ 4 2 4-

6 -2

58

53 53 59 51 49 36 38 61

63 44 47 59

40 66 84

242 5

64 75 46

41

io

62 3* 3-

66 66 64

p-Monoalkyl Phenol Sulfides Control SzClz Phenol p-Cresol SzClz p-Ethylphenol SzClz p-tert-Butylphenol SzC1, SzCL p-tert-Amylphenol p-Hexylphenol SzClz p-Octylphenol SzClz p-Nonylphenol SzCll

++ + + + +++

443

;:

40

67 64

64 61 60

Butvl and Dibutvl Methvl Phenol Siilfideb Control Phenol p-Cresol

++SzClz SnClz 4-tert-Butyl-3-methylphenol 4- fizC12 2-Butyl-4-methylphenol + SzClz 2,4-Dimethyl-6-butylphenol+ SeClz 2.6-Di-tert-butyl-4-methylphenol + Sac12 4,6-Di-tert-butyl-3-methylphenol + SzCL

6

215

37 58 56 45 43 59

The reclaims made from the 5alkylphenol sulfides were corn1 59 2pared in groups in vulcanized 37 6 GR-S scrap, as shown in Table Variation of Alkyl Phenol-Sulfur Monoohloride Ratio 11. Each group included a 33 Control 0.0031/a 7reclaim sample containing the 54 0.0101/z 2 Phenol + SzClz 57 1 0.010 4,6-di-tert-butyl-3-methyl41 5 0.0051/2 phenol sulfide as a control. 40 0,005 6 41 0.005 5This remained fairly conatant 46 30.007 throughout the study. The Mooney plasticity values of the control varied from 34 t o A series of mono-p-alkyl substituted phenol sulfides with alkyl 41 (average 36.5) in ten different rum. The relative ratinge groups containing one to nine carbon atoms, produced reclaims for body and tack of the reclaim sheet also remained the same within reasonable limits, in these runs. uniformly poor in quality, which made it difficult to report any significant differences in their properties. Variations in chain Only minor differences were observed betm-een the several isomeric cresol sulfides and between the several isomeric xylenol length in the p-alkyl group made little difference in the activity of the reclaiming agents. sulfides. The xylenol sulfides were of slightly higher activity The unsubstituted and the butylated cresols were compared. than the cresol sulfides, as indicated by the hlooney plasticity The sulfide prepared from 2-butyl-4-methylflhenol was much values, while the unsymmetrically substituted trimethylphenol more active than that prepared from p-cresol. The most active sulfide was much more active than either. sulfide in this series was that prepared from 4,6-di-tert-butyl-3The p-tert-amylphenol sulfide showed approximately the same methylphenol. The 2,4,6-substituted compounds closely related activity as the reaction product of phenol and sulfur monochloride. However, a large increase in activity was observed to the latter compound-that is, the sulfides prepared from 2,4dimethyl-6-butylphenol and 2,6-di-tert-butyl-4-methylphenolbetween the p-amyl- and the 2,4-diamylphenol sulfides. The 2,4,6-triamylphenol sulfide dropped off considerably in activity. were found to be very low in activity. The cyclohexanol-sulfur monochloride reaction products preThe pblyhydric phenol sulfides were found to produce reclaims pared from the unsubstituted, the monoamyl, and the diamylwith inferior properties. Reaction products prepared from catecyclohexanols all were somewhat less active than the correchol, resorcinol, hydroquinone, pyrogallol, and phloroglucinol were all poorer than the reaction product of phenol and sulfur sponding benzene derivatives. The diamylcyclohexanol sulfide was fairly rlosr in activity to 2,4-diamylphenol sulfide. monochloride. Various alkyl derivatives of these compounds -'

INDUSTRIAL AND ENGINEERING CHEMISTRY

July 1948

1197

monochloride reaction product or sulfide mixture was the most effective, being slightly Compound Thickness Body Tack more active than the sulfur dichloride reaction product Sulfide, Sulfoxide, Sulfone, and Sulfur Chloriae Reaction Products which, in turn, was slightly Control O,OO51/a 6 11 32 more active than the purified 254 Phenol + SzCh 0.010 31/a 445 4-tert-Butyl-3-methylphenol sulfoxide 0 010'/2 7 monosulfide. A surprising re0'.013 l/e 2 3 59 4-tert-Butyl-3-methylphenol sulfone 3 46 4-tert-Butyl-3-methylphenol monosulfide 0.01 11/a 6'/2 sult was that the sulfoxide 39 O.O071/z 5 9 4-tert-Butyl-3-methylphenol+ SzClz was slightly more effective, 443 8 4-tart-Butyl-3-methylphenol + SCh 0.009 1/2 weight for weight, than the Miscellaneous Phenol Sulfides monosulfide. The poorest reControl 0.004 6 lll/Z 36 56 claim was obtained from the 2 0,008 3 Koppers coal-tar cresylic acid SDa + SzClz sulfone. Both sulfoxide and 47 53 0.0081/2 Koppers coal-tar cresylic acid 1-Bb 4- SzClz 41 81/a 50 0.007 Koppers coal-tar cresylic acid X-1C + SaCh sulfone appeared t o be rela7 48 0.008 Koppers coal-tar cresylic acjd X-2d + Sac12 0.007 50 61/a Koppers coal-tar cres lic acid No. 28 + SaClz tively insoluble in the scraps 3+ 0.007 61/2 42 48 Shell petroleum cresyzc acid 2000f + SaCla before heating. 5 --5 Shell petroleum cresylic acid 4020 A8 + SzClz O.O051/a 9l/a 4Shell petroleum cresylic acid 9035 Ah + SzCh 0.0061/a 7 48 above comparison of 5 --5 O.O051/a Oronite petroleum cresylic acid F F i + SzClr 9 42 . theThephenol 0.0051/* 59 45 Oronite petroleum oresylic acid H i + SzClz sulfides of differ2 0.0071/e 3 57 Phenol + SaClz ent structures makes possible 0.010 4,4'-Dihydroxydiphenyl sulfide 3l/a 66 0.017 4,4'-Dihydroxydiphenyl sulfone 3 1/z 68 the following generalizations Control 0.0031/r 11i/a 33 concerning their reclaiming 41/2 57 54 Phenol + SnCla O.O101/a 2 3 p-Phenylphenol + SzCla O.Oll~/lr activity. 4 56 2 p-Cyclohexylphenol + SzCla 0.010 RECLAIMING ACTIVITY. ln3+ 6 57 p-Hexylphenol + SzClz O,0081/z 2+ 5 56 0.007 1/z m-Cresol + SaCh creasing the length of the 2+ 41/2 52 0.010 47 Distilled Cardolite 5679k + SzCIz 3+ 0.007 1/z 6 Shell Dutrex 441 + SzCle chain in the monoalkylphenol 3 +-4 6 44 0.007 Octyl petroleum cresylic acid 2000m + Sack sulfides was not effective in 9 43 6 Shell petroleum cresylio acid 4020Ag + SaClr O.O041/z 21 Phenol + SnCla 0.008'/2 31/a 58 increasing their activity. The 1 6- and p-Mixed benzyl henols + SaCln 0.011 3 57 3 52 2 Dibenzylphenol + Szcfl, 0,010~/2 introduction of a second and a third alkyl group into Polyhydric Phenol Sulfide* the phenol nucleus produced 6 11 37 Control considerable enhancement of 258 Phenol + SzClz ?i/Z 78 iCatechol + SaCh softening ability. 1 91 Resorcinol + SzClz The position of the vari1 78 Hydroquinone + SrCh 67 1' Pyrogallol + SaClz ous alkyl groups on the phenol 78 Phloroglucinol + SaCle 77 4-tert-Butylcatechol StClz nucleus was not as impor75 Di-tert-but lcatechol + SzCL 75 2,5-Di-tert-~~tylhydroclulnone SZCb tant to activity as the num74 1 Octylcatechol + SzClz ber of alkyl groups, as long as one of the ortho or para a Distillation, 5 0 7 not above 206O C., dry below 225O C. positions was left open for b Distillation, 5 0 9 not above 213O C., dry below 230" C. Distillation, 20 not above 210° C., dr below 217' C. the introduction of the suld Distillation, 5 2 0 , between 214 and 21T0 C., 95% off between 220 and 225O C. e Distillation, 50% above 213O C., dry above 230° C. fide or sulfoxide groups. Phef Boiline ranne 200 to 228O C. nols with alkyl groups in the Boiling range 221 to 245O C. h Boiling range 248 to 300' C. 2,4; 3,4; 2,3,5; and 3,4,6, i Boiling range 210 to 235-240' C. max i Boiling range 235 to 260-271O C. max. positions formed sulfides in k Boiling range 194 to 210° C. a t 1 mm. 1 Boiling range 135O C. a t 10 mm. to 190" C. a t 4.5 mm. the 2,4, or 6 positions which m Boiling range 140° C. a t 10 mm. to 165O C. a t 3.5 mm. were found t o be active as reclaiming agents. The 2,4,6-trialkyl substituted compounds, with the ortho and para positions filled, reacted with sulfur did s o t produce any significant increase in softening activity chloride, but the products were relatively low in reclaiming compared with the parent polyhydric phenol. activity. The comparison was made between 4-tert-butyl-3-methylphenol The reaction products prepared from the various commercial and a number of sulfide reaction products prepared by the variation of the phenol-sulfur monochloride ratio. Very poor softencresylic acid mixtures indicated that the higher boiling petroleum acids were more active than the lower or higher boiling coal-tar ing activity was obtained with unreacted 4-tert-butyl-3-methylacids. This indicates that the greater number of alkyl groups phenol. The normal reaction product (1.0 mole of 4-tert-butylaround the phenol nucleus, which characterizes the higher boiling 3-methylphenol to 0.5 mole of sulfur monochloride) was found to petroleum acids, contributed greater activity than the larger be very effective as a softener. The products obtained by renumber of fused ring phenolic compounds found in the higher acting 0.75 or 1.0 mole of sulfur monochloride with 1 mole in of boiling coal-tar acids. 4-tert-butyl-3-methylphenol were approximately as active as The reduced ring compounds, the cyclohexanols, like the the normal preparation. However, a higher ratio of sulfur monophenols, are ineffective without substituent alkyl groups. It was chloride (1.0 mole of 4-tert-butyl-3-methylphenol to 1.25 moles of interesting t o find that the diamylcyclohexanol reaction product sulfur monochloride) produced a product of slightly lower activity. was active, almost equal to the corresponding diamylphenol T o determine the importance of the sulfide bridge between derivative. phenol rings, pure 4-tert-butyl-3-methylphenol monosulfide was The polyhydric phenols and their alkyl derivatives formed sulcompared with the corresponding sulfoxide and sulfone. I n the same group were compared the crude reaction products prepared fur chloride reaction products which were found to be ineffective as reclaiming agents. The introduction of a second and a third from sulfur monochloride and sulfur dichloride. The crude sulfur

TABLE I€. (Continued)

Mooney Quality Value Index Ms/4/i12

33:

::

2';

+

C

+

;:

::

1198

INDUSTRIAL AND ENGINEERING CHEMISTRY TABLE 111.

GR-S tread scrap (5-mesh) Dipentene Coumarone-indene resin 4,6-Di-tert-butyl-3-methylphenol sulfide a-Naphthyl mercaptan @-Naphthylmercaptan Xylyl mercaptana Zinc salt of xylyl mercaptan Cresol sulfide mixture

COWPARIsON O F

A 100 6

6

B 100 6 6

RECLAIMS

C 100 6 6

D 100 6

E 100

F 100

6 6

6

6

6

1.6 1.6

1.5

3.0

1.5 1.6

Properties of Uncured Refined Sheet evaluation Thickness, inch 0.006 0.008 Body FC-G F+ Tack 5+ 4+ Mooney plasticity values, MS/4/212 41 50 Physical tests Elongation 300 250 Tensile 105 135 Chemical analvsis Acetone extract 16.65 1.6.76 Ash 3.80 3.80 IJncured chloroform extract 14.85 12.68 Carbon black 24.33 24.33 Total sulfur 1.41 1.50 Cured chloroform extract 5.85 6.03 Total rubber hydrocarbon 48.78 47.77

Reclaim Sheet 0,011 0.010 F F 3+ 3-

0,010 0 ,013 F P-F 3+ 2+

56

55

58

67

230 121

220 114

230 123

190 106

16.47 16.57 16.02 3.80 3.80 3.80

15.95 3.80

11.75 1 0 , 5 0 11.67 8.27 24.33 24.33 24.33 24.33 1.54 1.26 1.37 1.52 4.86 3.66 3.63 4.93 50.20 49.18 49.45 50.77

Properties of Cured 811-Reclaim Stocks Test Formula Reclaim 200.0 Zinc oxide 5.0 Stearic acid 2.0 Sulfur 3.0 Captax 0.5 Diphenylguanidine 0.2 Physical Tests A B

C D E Cure 20 min. a t 287O F, 300% strem 350 350 376 400 450 Elongation 460 470 480 460 460 Tensile lb./sq. inch 775 875 900 900 925 Shore hardness D 43 43 43 44 46 Cure 25 min. a t 287" F. 300% stress 350 375 400 400 500 Elongation 480 480 480 470 440 Tensile lb./sq. inch 875 9.50 950 950 950 Ghore hardness D 43 45 44 46 46 Cure 30 min. at _. 287O F. 300% stress 376 400 400 425 500 Elongation 440 480 470 460 420 Tensile Ib./sq. inch 800 975 525 1000 975 Shore hardness D 44 45 45 46 46 50% solution of xylyl mercaptan in inert hydrocarbon solvent.

F 675 380 1000 49 725 380 1125 49

~

800 370 1125 49

hydroxyl group produced compounds with much lower activity than the monohydroxy. compounds. The comparison of the sulfides with the corresponding sulfoxide and sulfone prepared from 4-tert-butyl-3-methylphenol provided some very interesting results. The sulfoxide linkage proved to be equivalent to or slightly more active than the monosulfide linkage, whereas the sulfone was practically inert. The crude preparations from sulfur dichloride and from sulfur monochloride, which included a certain proportion of polysulfides in the reaction mixture, were slightly more active than the purified monosulfide. PROPERTIES OF RECLAIMS

.

In order t o obtain a comparison between reclaims produced with the aid of different reclaiming agents, the sulfide mixture prepared from the reaction of sulfur monochloride with 4,6di-tert-butyl-3-methylphenol was compared with a commercial cresol sulfide mixture, a-naphthyl mercaptan, @-naphthylmercaptan, xylyl mercaptan, and the zinc salt of xylyl mercaptan (names of suppliers of commercial products furnished on request). These materials were chosen as active reclaiming agents which have been used in recent years both in this country and abroad. The data obtained on the uncured refined reclaim sheet (Table 111)included the complete chemical analysis, as well as the tensile and elongation as dpterminpd on small dumbbell strips using the

Vol. 40, No. 7

small Scott tensile machine, Mooney plasticity values, and the usual thickness, body, and tack ratings. These all agreed in showing that the 4,6-di-tert-butyl-3-methylphenolsulfide was the most active reclaiming agent of the group, and produced a thinner reclaim sheet, with the best body and tack, the lowest Mooney plasticity value (RfS/4/212),the greatest elongation and lowest tensile, and the highest uncured chloroform extract. The cresol sulfide mixture was the least active of the series, with the other materials intermediate in activity. The various physical tests on the uncured reclaim sheet agree remarkably well in rating the different reclaims. The physical properties of the vulcanized reclaims do not vai\ as widely as those of the uncured reclaim sheets. The poor, rather nonuniform reclaim sheet prepared using the cresol sulfide mixture had both low elongation and tensile; the vulcanized reclaim showed both high modulus and tensile coupled with lox$ elongation. The vulcanized reclaim prepared from 4,6-di-tertbutyl-3-methylphenol sulfide exhibited consistently lower modulus and tensile together with equivalent elongation as comparcd to the reclaim vulcanisates from the aryl mercaptans. This i i taken to be an indication of a greater degree of break-down produced by the 4,6-di-tert-butyl-3-methplphenolsulfide. Table IV shows some effects on the reclaim sheet, produced b\ increasing the concentration of reclaiming agent. A total content of 12 parts of softener plus reclaiming agent per hundred of scrap was maintained during the study. The reclaim made from the addition of swelling agent alone had the lowest cured and uncured chloroform extracts. With the reclaiming agent present in concentrations of 0.5, 1.0, and 2.0 parts, the qualit> of the reclaim improved considerably and both chloroform 4.0 and 8.0 extracts increased. Higher concentrations-e.g., parts of reclaiming agent-produred reclaims which were vrr! soft and therefore of less desirable workability. The chloroform extract of these reclaims again showed a significant increase. The acetone extracts did not follow the chloroform extracts, actually decreasing slightly Kith increasing reclaiming agent However, the largest amount of reclaiming agent used, 8 parti, did cause an increase in the acetone extract of the product. The increase in the chloroform extracts with increasing reclaiming agent concentration is very interesting. I t indicates that an increase in the amount of lower molecular weight fragments from the GR-S vulcanizate is produced with an increase in the amount of reclaiming agent used. The cured chloroform extract is generally thought to represent segments of the vulcanized rubber structure which have been degraded to such a n extent that they do not readily revulcanize (94). PHENOL SULFIDES AS RECLAIMING AGENTS FOR OTHER SYNTHETIC RUBBERS 1

The sulfides of 4,6-di-tert-butyl-3-methylphenolwere found to be active for reclaiming several other types of synthetic rubbers such as the butadiene-acrylonitrile copolymers (N types)

TABLE IV. EFFECT OF COXCENTRATION OF RECLAIMING AGENT A

B

C

GR-S tread scrap (5-

mesh) 100 100 100 Swelling agent and tackifier 12.0 11.6 11.0 4,6-Di-te+buty1-3methylphenol sulfide 0.5 1.0 Sheet evaluation Thickness, inch 0.013 0.011 0.0081/2 Body P-T, F Tack 2 3+ Chemical analysis Acetone extract 19.45 18.15 18.63 Uncured chloroform extract 7.35 8.68 10.40 Cured chloroform extract 1.45 1.71 1.66

$

D

E

F

100

100

100

10.0

8.0

4 0

2.0

4.0

8.0

0.008 G

0.007 0.006 VG Oversoft

7+

8+

1 7 . 9 3 ' 18.45

22.76

11.33

11.48

14.70

2.13

2.71

3.42

6+

July 1948

INDUSTRIAL AND ENGINEERING CHEMISTRY

and neoprene. Table V reports the comparison of this highly alkylated phenol sulfide with the mercaptan type of softeners. The degree of activity of the phenol sulfide softener in these rubbers is not as great as in GR-S type scraps. This may be due in part t o the fact that the large amounts of oils and tackifiers, which were necessarily used to swell and plasticize the vulcanized scrap, masked some of the action of the chemical softener.

TABLEV.

PHENOL SULFIDES AS RECLAIMING AGENTSFOR OTHERSYNTHETIC RUBBERS A

N type synthetic rubber 8crap (&mesh) Oils and tackifiers 4,6-Di-tert-butyl-3-methylphenol sulfide a-Naphthyl mercaptans Zinc salt of xylyl meroaptanb Sheet evaluation Thickness, inch Body Tack Neoprene scrap (5-mesh) Oils and tackifiers 4,6-Di-tert-butyl-3-methylphenolsulfide a-Naphthyl mercaptans Zinc salt of xylyl mercaptanb Sheet evaluation Thickness, inch

100 35 2.0

0.017

8

100 36 3 .O

2.0

0.020

G 100 35 5.0

C 100 35 2.0

0.030

G

2+ 100 35 6.0

0.023 G 5+ 65 0 30Tc solution of ?-naphthyl mercapran in inert hydrocarbon solvent. b SOYc solution of zinc salt o i xylyl mercaptan in inert hydrocarbon solvent.

$3

0.021

33 100 35

V-2

0.023

6

The use of highly alkylated phenol sulfides in reclaiming these oil-resistant polymers is very beneficial. Without them, even greater amounts of oil would be necessary t o produce a satisfactory reclaim. The alkylphenol sulfides, therefore, produce from these scraps reclaims which have higher rubber contents and lower acetone extracts than it has heretofore been possible t o obtain. The alkylated phenol sulfides are even more active for the reclaiming of natural rubber than for the synthetic rubbers. FATE OF COMBINED SULFUR AND RECLAIMING AGENT DURING THE RECLAIMING PROCESS

The vulcanization of rubber by sulfur is only partly understood; nevertheless, it certainly involves the chemical reaction of rubber hydrocarbon chains with sulfur and the joining of these chains to form a cross-linked structure. Soft vulcanized rubber, even in the presence of antioxidants, is more or less readily attacked by the oxygen of the air, being only slightly more resistant to attack than the corresponding unvulcanized rubber. Aged vulcanized scrap, the usual raw material of the reclaim industry, therefore consists of a cross-linked hydrocarbon chain structure containing both chemically combined sulfur and oxygen. The reclaiming process is frequently referred to as “devulcanization.” I n the sense that a stiff, hard, high modulus, nonprocessible vulcanizate is converted during the reclaiming process to a soft, plastic, more tacky, low modulus, processible, and vulcanizable product having many of the properties of the original rubber, the term devulcanization is appropriate. However, in the chemical sense it has been generally recognized that the reclaiming process does not reverse the chemical reactions taking place during vulcanization. Chemically speaking, therefore, the so-called devulcanization process, involving the separation of combined sulfur from the rubber hydrocarbon, does not take place in the usual reclaiming processes (IO,20). I n this connection it was of interest t o examine the effect of the most active reclaiming agents with respect to the fate of the sulfur, particularly the combined sulfur in the vulcanized scrap, during the reclaiming process. I n fact, the hypothesis may be advanced that the phenol sulfides, particularly those prepared from sulfur monochloride, might act through a monosulfide-

1199

polysulfide equilibrium. At temperatures below 150O C., it has been shown that these materials may furnish polysulfide sulfur to the rubber hydrocarbon (36). At temperatures above 175’ C., it might be postulated that these sulfides could remove sulfur from the vulcanized rubber. As a result of this speculation, the authors have determined the combined sulfur, before and after reclaiming, of four carefully controlled and reclaimed samp!es : A, a nonsulfur-cured GR-S, and B, C, and D, three samples of a typical sulfur-cured GR-S (Table VI). One of these samples, D, was acetone-extracted before analysis and reclaiming. A small amount of combined sulfur was found in the nonsulfur-cured stock, and this must have been introduced with the GR-S polymer, the carbon black or the zinc oxide, for it was common to all the stocks. The nonsulfur-cured GR-S, A, and two of the sulfur-cured samples, including that which was acetone-extracted, C and D, were reclaimed by the usual procedure in the presence of 1.50 parts per 100 parts of scrap of pure 4,6di-tert-butyl-3-methylphenol monosulfide. Sample B was treated in exactly the same manner, but without the addition of the alkyl phenol sulfide reclaiming agent. I n each vulcanizate the difference between the total sulfur in the sample and the total sulfur in the acetone extract was taken as the sulfur actually combined with the GR-S polymer. These differences were calculated to the basis of the 100 parts of GR-S used in the preparation of the original vulcanizate. They show that the three samples reclaimed in the presence of the sulfide reclaiming agent actually gained about 0.1% of sulfur combined per 100 parts of polymer during the reclaiming process, whereas a small amount of sulfur was actually removed from the stock treated in the absence of reclaiming agent. These differences are small, but they are larger than the probable errors involved in their calculation, and the excellent internal agreement shown in this series of analyses lends weight t o the conclusion that during this reclaiming process, a vulcanizate contafning the alkylphenol sulfide reclaiming agent shows an increase in combined sulfur relative to a vulcanizate treated without the reclaiming agent. I n fact, in the stocks which contained no free sulfur a t the start of t h reclaiming process, it may be postulated that increase in combined sulfur must represent reclaiming agent (or molecular fragments from its reactions) which has combined with the polymer vulcanizate during the process. Such a calculation was made for stocks A and D (Table VI) and showed 47 and 6670, respectively, of the reclaiming agent sulfur cpmbined with the vulcanizate during the reclaiming process. If some of the rechiming agent combines with the polymer during reclaiming, then the remainder should be recoverable in the acetone extract of the reclaimed samples. This calculation also was made, after correcting the total sulfur in the acetone extract for free sulfur (determined by the sulfite method, W), and for sulfur added with the reclaiming oils and tackifiers. The results show a recovery of 38 and 34%, respectively, of the reclaiming agent sulfur in the acetone extracts from reclaims A and D. The sum of the sulfur combined and the sulfur recovered in the acetone extract accounts very satisfactorily for the sulfur introduced into the process as reclaiming agent. It is obvious that these figures are not by any means absolute. However, the values obtained appear t o be significant, and certainly the trends are significant. It may then be concluded that in vulcanizates reclaimed with the aid of the alkylphenol sulfide reclaiming agent, part of the sulfur introduced as reclaiming agent combines with the vulcanizate during the reclaiming process, and the remainder is recoverable in the acetone extract of the finished reclaim. The preceding analyses make untenable the “sulfur equilibrium” hypothesis, and another mechanism for the activity of these reclaiming agents must be sought. The evaluation of the reclaims produced for the analyses of Table VI is shown in Table VII. This illustrates the importance of the acetone-soluble oils and plasticizers which were lacking in sample D, and also shows that the alkylphenol sulfide reclaiming

1200

INDUSTRIAL AND ENGINEERING CHEMISTRY

Vol. 40, No. 7

c

m P o n

m rn

0

0

9 9 $

c

00

el.

W N

?? 00 i t -

2g 00

2rsm CQ

tt-

(D (D ri

00

0

+

0

-?

m

J d

-2

4

r

-2 a

.

g f 0

d

9

0

?

* " 5

:

0

INDUSTRIAL AND ENGINEERING CHEMISTRY

July 1948

1201

years experts in different countries have come to the conclusion

AND SULFUR-CURED RECLAIMS that, on reclaiming vulcanized rubber, as well as on plasticizing TABLE VII. NONSULFUR-CURED crude rubber, the decisive part is acted by oxgyen, whereas softPRODUCED FOR ANALYSIS (TABLEVI)

A

-NO"-.-

GR-S tread scrap Oils and taokifiers 4,6-Di-terl-butyl-3-methylphenol monosulfide Sheet evaluation Thickness (one pass), inch Body Tack

sulfurCured GR-€3 tOO.0 12.0

1.6 0.021

P

2-

B

C

SulfurCured GR-8 100.0 12.0

SulfurCured GR-S 100.0 12.0

D SulfurCured GR-8 Acetone Extracted 100.0 12.0

1.5

1.6

0.014 G

0.024

... 0.031 P-T 3-

6

P-F 3+

agent was effective in the nonsulfur-cured vulcanizate. This last observation was extended to a higher concentration of reclaiming agent in GR-S, and also to a nonsulfur-cured butyl rubber vulcanizate in Tables VI11 and IX. From the data cited in Tables VI to I X , inclusive, it may be concluded that carbon-carbon bonds, rather than carbon-sulfur bonds are mainly affected in the vulcanizate during this reclaiming process.

TABLE VIIL

NONSULFUR-CURED GR-S RECLAIMS

A Dibenzo G M F a cured GR-S tread scrap 100.0 Oils and tackifiers 12.0 4,6-Di-tert-butyl-3-methylphenol monosulfide Sheet evaluation 0.036 Thickness, inch P-L-T Body 0 Tack 5 Dibenzoyl quinone dioxime.

. ..

€4

c

100 0 12.0 1.5

100 0 12.0 3.0

0.018

0.013 F-G 6-

P-L-T 3-

~~

TABLE XX. NONSULFUR-CURED BUTYLRECLAIMS

Body Tack

Formula without Alkyl Phenol Sulfide V-P-T

Formula with 2 Parts 4,6-Di-tert-butyI-3methylphenol Sulfide

2-

CONCLUSIONS CONCERNING MECHANISM OF RECLAIMING PROCESS

A study of the literature on reclaiming leads to the almost inevitable conclusion that the oxidative scission of polymer vulcanizates plays a predominant role in most of the reclaiming processes, The reclaiming temperatures used in the present experiments were below those at which appreciable thermal depolymerization or cracking (natural rubber) occurs in the absence of oxygen ( 7 ) . Cyclization, except in so far as it is an adjunct of the oxygen and reclaiming agent attack, probably can be ruled out as the main reaction by the known resistance of vulcanized rubber to cyclizing agents, even at elevated temperatures, and by the lack of properties generally associated with cyclized rubber products&g., hardness, thermoplasticity, low elongation, and loss of chemical activity. The effect of the oxygen concentration on the speed and extent of the reclaiming processes for both natural and synthetic rubbers is well known (4, 6, 13, 15, 17, 20, 26). Probably every reclaim manufacturer has demonstrated that oxygen introduced under pressure will hasten the reclaiming of natural and synthetic rubbers. An appropriate quotation may be taken from the first page of'a German report on the Renacit reclaiming agents (14).

Especially, there has proved to be untenable the former view to the effect that the reclaiming would be due to a desulfurization of the vulcanizates under the influence of alkali. I n the past ten

eners, heating, and mechanical working exert only an accelerating effect. Acids and alkalies merely serve for destroying the fabric. The process for the reclaiming of the Buna synthetic rubbers, as developed by I. G. Farbenindustrie A. G., included the use of the aromatic mercaptans to accelerate the oxidizing process and the use of air at 2 to 3 atmospheres' pressure to increase the oxygen concentration (14). Other evidence that oxygen arid oxidation are fundamentally involved in the reclaiming process may be deduced from the well known fact that well-aged, oxidized vulcanized scrap is much more susceptible to reclaiming than fresh unoxidized rubber vulcanizate (20); also, that as the amount and effectiveness of antioxidants (and of accelerators showing strong antioxidant effects) are increased, the vulcanizate becomes increasingly difficult to reclaim successfully (IS). If we can assume that the fundamental reaction of the usual reclaiming processes is oxidative in nature, it is natural to conclude that the alkylphenol sulfide reclaiming agents are catalysts for the oxidative scission reactions of rubber hydrocarbons and vulcanizates, while remaining inert to, or actually inhibiting, the oxidative aggregation reactions which are known to take place sirriultaneously during the oxidation of both natural and synthetic hydrocarbon polymers (1, 7, 8, 27, 29,31, 32, 33). The reactions of hydrocarbon chains with oxygen are generally believed to proceed through radical chain mechanisms, and these reclaiming agents, themselves, may act by the formation of radicals or fragments which may aid in the oxidative breakdown, perhaps by aiding the dehydrogenation of the polymer chains and also by reaction with the fragments formed. The alkylphenol sulfides may help to prevent Qxygen-catalyzed aggregation in the same manner by which similarly alkylated phenols protect synthetic polymers and vulcanizates (N and GR-S types) from aggregation or stiffening during natural and accelerated aging (29). I n connection with the above ideas the following facts may be considered significant: Both mercaptans and disulfides are known to catalyze oxidation

or dehydrogenation reactions, presumably through free radical mechanisms (26, 36, 37).

An increase in oxygen pressure is reported to increase the plasticity and quality of a Buna S reclaim in the presence of an aryl mercaptan reclaiming agent (14). The alkylphenol sulfoxides and sulfides, which are relatively less stable under reclaiming conditions, are much more active than the corresponding sulfones which contain more oxygen but are very stable compounds. The increase in the chloroform extracts of the uncured reclaims with increase in the amount of reclaiming agent used suggests the formation of polymer vulcanizate fragments of low mo. lecular weight. The activity .of the alkylphenol sulfide reclaiming agents in several different Dolvmer vulcanizates (GR-S. butadiene-acrvlonitrile copolymers, nGoprene, Butyl), cured with or without suifur, suggests that the reactivity must center on the only common functional groups in these materials-that is, the hydrocarbon chains of the original polymers. Further work is indicated in order to prove the relationship of the alkylphenol sulfides and of oxygen in the usual reclaiming processes. SUMMARY

Some highly alkylated phenol sulfides have proved to be among the most active reclaiming agents for GR-S and other synthetir rubbers. A study has been made of the activities of these materials as related to molecular structure and to the fate of sulfur during the reclaiming process. The nature and mechanism of the chemical reactions taking place during the reclaiming of natural and synthetic rubbers have been discussed in the light of the results obtained-

INDUSTRIAL AND ENGINEERING CHEMISTRY

1202

ACKNOWLEDGMENT

The authors wish to acknowledge the assistance of H. G. Dawson of the Firestone Research Laboratory for the analytical values of Table VI and to H. H. Miller of the Xylos Rubber Company for various reclaim evaluations and other data. Appreciation is expressed for the encouragement and assistance of F. W. Stavely and R. F. Dunbrook, and for the permission t o publish this manuscript by the managements of the Firestone Tire & Rubber Company and the Xylos Rubber Company.

Vol. 40, No. 7

Ibid., 2,359,122 (1944) ; 2,363,873 (1944) ; 2,372,584 (1945). Lutz, G~ntmi-Ztg.,25, 120-1 (1910). Miller, G. R., in “Chemistry and Technology of Rubber,” bi Davis, C. C. and Blake, J. T., pp. 720-38, New York, Rein-

hold Publishing Gorp., 1937. Mooney, M., IND. ENG.CHEM.,ANAL.ED.,6 , 147 (1934). Neal, A. M., and Schaffer, J. R., Jr., U. 5. Patent 2,333,810 (1943).

Oldham, E. W., Baker, L. M., and Craytor, M. W., IND.ENG. CHEM., ANAL.ED.,8 , 4 1 (1936). Palmer, H. F., and Kilbourne, F. L., Jr., IND.ENG.CHEM.,32, 512 (1940).

Ritter, J. J., and Sharp, E. D., J. Am. Chem. SOL, 59, 2351

LITERATURE CITED

(1937).

Andrews, R. D., Tobolsky, A. V., and Hanson, E. E., J . Applied Phys., 17, 352-61 (1946),

British Ministry of Supply, private communication (June 1943). British Ministry of Supply, private communication (May 1946). Carpenter, A. S., IND. ENG.CHEM.,39, 187 (1947). Dasher, P., U. S. Patents 2,304,548, 2,304,549, 2,304,550, 2,304,551 (1942).

Essex, W. G., U. S. Patent 2,154,894 (1939). Farmer, E. H., in “Advancesin Colloid Science,” Vol. 11,p. 303, New York. Interscience Publishers. 1946. Farmer, E. H,, Trans. Faraday SOC.,42,228-36 (1946). Garvey, B. S., U. S. Patent 2,193,624 (1940). Gillman, H. H., Rubber Aoe (New York), 58, 709-14 (19461, Gumlich, W., U. S. Patent 2,280,484 (1942). Gumlich, W., and Ecker, R., U. S. Patent 2,338,427 (1944). Hughes, A. J., and Amphlett, P. H., Trans. Inst. Rubber Ind., 19, 165 (1944).

I. G. Farbenindustrie, A. G., Kautschuk-Zentrallaboratorium, Leverkusen, “Report on Reclaiming of Buna Vulcanizates with the Aid of Renacit” (Aug. 28, 1945). Ioannue, J. P., U. S. Patent 2,069,151 (1937). Jones, F. A., Owen, E. W. B., Tidmus, J. J., andFraser, E. E., Trans. Inst. Rubber Ind., 19, 190 (1944). Kirby, W. G., and Steinle, L. E., U. 9. Patent 2,279,047 (1942).

Sebrell, L. B., Canadian Patent 289,290 (1929) ; Chem. Zentr.. 103, I, 2392 (1932). Shelton, J. R., and Winn, H., IND. ENG.CHEM.,39, 1133 (1947). Simmons, H. E., War Production Board, Office of the Rubber Director, private communication, Feb. 23 and June 3, 1943. Smith, G. E. P., Jr., “Antioxidant Effects in Natural and Synthetic Rubbers,” Symposium on Degradation and Aging of High Polymers, Polytechnic Institute of Brooklyn, Nov. 30. 1946.

Smith, G. E. P., Jr., Ambelang, J. C., and Gottschalk, G. W., IND. ENG.CHEM.,38, 1166 (1946). Spence, P., and Ferry, J. B., J. Am. Chem. SOC., 59, 1648 (1937). Taylor, H. S., and Tobolsky, A. V., Ibid., 67,2063 (1945). Tobolsky, A. V., Prettyman. I. B.. and Dillon, J. H., J . Applied __ Phys.,-l5, 380 (1944).

U. S. Rubber Co., British Patents 575,545, 575,546, 575,545 (1946).

Waters, W. A., Trans. Faraday Soc., 42,189 (1946), Wolf, G. M.,Deger, T. E., Cramer, H. I., and DeHilster, C. C.. IND.ENG.CHEM.,38, 1157 (1946). Ziegler, K., and Ganicke, K., Ann., 551, 213 (1942). RECEIVED June 10, 1947. Presented at the meeting of the Division of Rubber Chemistry of the AMERICANCHEMICAL SOCIETY, Cleveland, Ohio May 26 t o 28, 1947.

CITRIC ACID Production by Submerged Fermentation with Aspergillus niger PIKG SHU AND MARVIN J. JOHNSON University of Wisconsin,Madison, Wis. Average yields of 72 grams of anhydrous citric acid per 100 grams of added sucrose were obtained by submerged culture of Aspergillus niger in shake flasks on a synthetic medium at an initial sucrose concentration of 140 grams per liter. The fermentation required 9 days. A 70% yield was obtained i n 12 days at a sucrose concentration of 260 grams per liter. Data on the effect of changes i n composition of the medium are presented. The optimal conditions for shake flask fermentations include potassium dihydrogen phosphate above 1 gram per liter, magnesium sulfate heptahydrate above 0.25 gram per liter, iron concentration of 1 mg. per liter, 2.5 grams per liter of nnimonium nitrate, and an initial pH between 2.2 and 4.2.

ONSIDERABLE research effort has been expended on attempts t o develop a submerged fermentation process for citric acid production (16). Amelung (1) reported citric acid production from sucrose by aerating a submerged culture of Aspergillus japonicus. The yield of citric acid was very low. According to a recent patent of Szucs (fd),citric acid has been successfully produced in submerged cultures of Aspergillus niger. Szucs’ preferred procedure involves transfer of preformed mycelium from a growth medium t o a fermentation medium, and

the use of oxygen or air-oxygen mixture for aeration. Similarly, Karow and Waksman (2, 13) reported the production of citric acid in submerged cuItures of Aspergillus wentii. The maximum citric acid yield is obtained when oxygen is used for aeration, and after the desired growth is obtained the growth medium is replaced by a fermentation medium. Using shake flask technique, Perquin (8)systematically studied the effect of variation of the environmental conditions (both gaseous and liquid phases) on the production of citric acid in submerged cultures of Aspergillus niger. He concluded that the presence of zinc sulfate, potassium chloride, and increased concentration of magnesium sulfate in the liquid phase favored the production of citric acid. On the other hand, the presence of a high concentration of potassium dihydrogen phosphate in the medium was unfavorable. The use of oxygen or oxygen-air mixture for aeration resulted in a higher citric acid yield. Under all conditions, only a small amount of citric acid was produced. Karow and Waksman ( 2 ) demonstrated the requirement of manganese sulfate for maximum citric acid production by submerged culture of Aspergillus wentii. The substitution of urea for other nitrogenous salts proved satisfactory. They also found that the presence of magnesium sulfate and high concentrations of potassium dihydrogen phosphate in replacement or fermentation medium were unfavorable for citric acid production.