ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Final Report CURING SYSTEMS AND CURING PROCEDURES WHICH IMPROVE THE AGE RESISTANCE OF RUBBER Lo Mo Hobbs Supervisor and Associate Professor of Materials Engineering R. G. Craig Principal Investigator and Research Associate Co W.o BurkhartResearch Assistant Project 2376 THE ROCK ISLAND ARSENAL ORDNANCE CORPS, DEPARTMENT OF THE ARMY CONTRACT NO. DA-20-018-508-ORD-(P)-42 ORDNANCE PROJECT NO. TB 4-521 RAD ORDER NO. 57180100-99-00604 November 1956

The University of Michigan * Engineering Research Institute TABLE OF CONTENTS List of Illustrations iii Tables iii Figures iv Acknowledgment s v Object vi Summary vii Introduct ion 1 Experimental Part 3 Reference Compositions 3 Compounding and Curing 3 Scorch Measurements 4 Physical Property Measurements 4 Aging Experiments 5 Reproducibility of Compounding, Curing and Aging Procedures 5 Presentation of Results 7 Sulfur-Bearing Accelerator Vulcanizing Systems 7 Dicumyl Peroxide Vulcanizing System, 10 Effect of Oxygen-Containing Inhibitors on the Dicumyl 12 Peroxide Curing System Dicumyl Peroxide-Sulfur-Bearing Accelerator Vulcanizing Systems 15 Effect of Curing Time on Various Dicumyl Peroxide Systems 20 Scorch Properties of Selected Vulcanizing Systems 21 Aging of Selected Vulcanizates at Various Temperatures 24 Aging Studies at 212~F, 24 Aging Studies at 250~F 29 Aging Studies at 300~F 50 Dynamic Properties of Selected Vulcanizates 31 DiscuSsion of Results 33 References 38 _______________________ ii

The University of Michigan * Engineering Research Institute LIST OF ILLUSTRATIONS Tables I Reproducibility of Experimental Procedures 6 II Properties of GR-S Vulcanizates Cured with Sulfur-Thiazole and Thiuram-Thiazole. Combinations. 8 III Properties of GR-S Vulcanizates Cured with Dicumnyl Peroxide 11 IV Effect of Oxidation Inhibitors on the Dicumyl Peroxide Curing System 13 V Properties of GR-S' Vulcanizates Cured with Dicumyl PeroxideThiuram Combinat ions 17 VI Properties of GR-S Vulcanizates Cured with Dicumyl PeroxideDithiocarbamate Combinations 18 VII Properties of GR-S Vulcanizates Cured with Dicumyl PeroxideAccelerator Combinations 19 VIII Scorch Resistance of Selected Compositions 22 IX Stability of Dicumyl Peroxide Compositions to Heating in Air 23 X Effect of Aging at 212~F on Selected Vulcanizates 25 XI Effect of Aging at 250~F on Selected Vulcanizates 26 XII Effect of Aging at 300~F on Selected Vulcanizates 27 XIII Dynamic Properties of Selected Vulcanizates at 77~F 32 XIV Summary of Aging, Scorch and Cost Data for Selected Vulcanizates 34 XV Effect of Aging —-on Tensile and Tear Strength of Selected Vulcanizates 35 L____________________________________ iii _

The University of Michigan * Engineering Research Institute Figures 1. Effect of concentration of sulfur-bearing accelerators on tensile propertieD of vulcanizates 9 2. Physical properties of GR-S 1500 vulcanizates cured with dicumyl peroxide 12 3. Tensile properties of dicumyl peroxide - sulfur-bearing accelerator cured vulcanizates 15 4. Effect of curing time on tensile properties of dicumyl peroxide compositions 20 5. Effect of extended air-oven aging at 212~F on the elongation of GR-S vulcanizates 24 6. Effect of extended air-oven aging at 212~F on the tensile strength of GR-S vulcanizates 28 7. Effect of extended air-oven aging at 2120F on the 300% modulus of GR-S vulcanizates 28 8. Effect of aging at 250~F on elongation of selected vulcanizates 50 iV _

The University of Michigan * Engineering Research Institute ACKNOWLEDGMENTS We express our appreciation to A. C. Hanson, R. Shaw, and Z.'T. Ossefort of the Rock Island Arsenal for their helpful discussions; to S,LD. Gehman of The Goodyear Tire and Rubber Company for his assistance with the measurement of the dynamic properties and to R. N, Wetterholt and J. E. Marberry, graduate students in the Department of Chemical and Metallurgical Engineering, for their help in compounding and physical testing of rubber samples. _____________________________________ V _______________________________v

The University of Michigan * Engineering Research Institute OBJECT The purpose of this investigation was to study non:'ree sulfur curing systems for GR-S synthetic rubber which would yield vulcanizates of improved age resistance and which could be handled satisfactorily with present day processing machinery. vi

The University of Michigan * Engineering Research Institute SUMMARY The resistance to air-oven aging of GR-S vulcanizates cured with sulfur-bearing accelerators, with dicumyl peroxide and with dicumyl peroxide in combination with sulfur-bearing accelerators, has been studied. Some attention was given to the processing stability of the uncured compositions,, and limited work was devoted to the supression of blooming tendencies of certain vulcanizates. It was found that vulcanizates cured'with the peroxide and Combinations of the peroxide with sulfur-bearing accelerators possessed excellent resistance to air-oven aging and no blooming tendency. The uncured compositions showed moderate resistance to precuring or decomposition at processing temperatures up to 250~F. As a convenient basis for appraising the properties of the experimental compositions, two systems were selected as reference standards: I, a conventional formulation containing free sulfur and N-cyclohexyl-2-benzothiazyl sulfenamide, and II, a non-free sulfur system containing dipentamethylenethiuram-tetrasulfide and 2-mercaptobenzothiazole. Specific properties of five experimental systems were compared with I and II; namely: III, tetramethyl-thiuram-disulfide with dibenzothiazyldimethylthiol urea; IV, tetramethyl-thiuram-disulfide with 2-mercaptobenzothiazole; V, dicumyl peroxide; VI, dicumyl peroxide with tetramethyl-thiuramdisulfide; and VII, dicumyl peroxide with zinc dibutyl-dithiocarbamate. The age resistance of the vulcanizates of the peroxide compositions, V, VI, and VII, was substantially greater than the vulcanizate of reference composition I and somewhat greater than that of vulcanizate II. The resistance to scorch, that is the procesaing.. stability as measured in minutes at 250~F, of the Uncured compositions V, VI, and VII was about one-half that of composition I but several times longer than that of composition II. The values of the tensile strength of the unaged vulcanizates V, VI, and VII were slightly lower and the values of the per cent elongation were somewhat greater than those of vulcanizate I. The tensile properties of vulcanizate VII were the most nearly equivalent to those of I. The vulcanizates of the sulfur-bearing accelerator compositions III and IV showed about the same tensile properties aS'Vuicanizate II, but better resistance to aging than the latter product. The Scorch resistance of the uncured compositions III and IV was aboUt twici that of II. ___________________________ vii _____________

The University of Michigan * Engineering Research Institute Vulcanizates III and IV had a tendency to bloom. Results of preliminary experiments indicated that the blooming in vulcanizate III could be supressed by the presence of a small amount of low molecular weight polyethylene wax. On the basis of current market prices, the costs of the raw materials for experimental compositions III, IV, V, VI and VII would amount to some 2 - 6X more than for the conventional sulfur-accelerator system, I. It was concluded that the excellent age resistance of the GR-S vulcanizates V - VII can justify the increase in material costs, and some limitation in processing conditions in the production of rubber items for military and mechanical goods applications where good age resistance is necessary. viii

The University of Michigan * Engineering Research Institute INTRODUCTION The need for improvement of the age resistance of rubber vulcanizates is of general importance, and that need is especially critical in the storage of military rubber itemso The age resistance of rubber vulcanizates is known to be influenced by the nature of the vulcanizing system, the products of the curing reactions, and the presence of inadvertantiimpurities such as metals. The commonly used curing system consisting of free sulfur in combination with small amounts of one or more accelerators has the advantages of being relatively inexpensive, and being relatively resistant to scorch or precuring at the moderately high temperatures used in processing before the final vulcanization stage. The conventional free sulfur based curing system has the limitation, however, of yielding products that fall short in resistance to aging. The possibility that there may be other vulcanizing systems which have acceptable stability during processing and yield products having improved aging properties is of high interest. For example, use of combinations of the sulfur-bearing accelerators normally employed in the free'sulfur system, but in the absence of added free sulfur, has given vulcanizates possessing improved age resistance. However, these combinations are more sensitive to scorch, and have a further disadvantage in cost since the use of the relatively larger amounts of these complex synthetic organic chemicals as the primary vulcanizing agent is more expensive. Unfortunately, it appears that most of the curing systems that have been found to yield vulcanizates with improved aging qualities are not only relatively expensive but are also relatively more sensitive to precuring or decomposition in processing before the final vulcanization. The problem is made more difficult by the continuing demand of the rubber industry for higher production rates,. which in turn, with the present equipment and procedures, requires higher processing temperatures before the final vulcanization stage. The Rock Island Arsenal has been following these trends and has been contributing in the developments over a number of years. The present study was undertaken to examine certain new vulcanizing systems which offered promise of yielding products of improved age resistance, and at the same time might reet the condition of, at least, a working compromise in cost and in stability during processing. __________________________1 __________________________________

The University of Michigan * Engineering Research Institute In the experimental work described below GR-S 1500, a copolymer of styrene and butadiene prepared at 41~F, was employed as the base rubber. Two compositions were selected as reference standards: I. a conventional sulfur vulcanization system, containing free sulfur in combination with a sulfur-bearing accelerator, and IIo a non-free sulfur system comprising a combination of sulfur-bearing accelerators. The age resistance of experimental vulcanizates has been compared with the age resistance of the vulcanizates of these two systems. Similarly, the tendencies of the experimental systems to scorch or decompose under processing conditions have been compared with the scorching tendencies of these reference compounds. The experimental systems have comprised mainly, a. various combinations of accelerators commonly used with free sulfur but employed here in the absence of added free sulfur (as in reference compound II), b. systems containing dicumyl peroxide as the principal curing agent, and c, systems containing combinations of dicumyl peroxide with certain of the accelerators used in (a). In brief, it has been found that several of these experimental compositions, particularly certain of those listed under (b) and (c) above, yielded vulcanizates possessing substantially improved resistance to air-oven aging compared to reference compositions I and IIo They showed a moderate improvement in resistance to scorch or decomposition during processing, as compared with refene mpsitin but werene composition Ill less resistant to scorch than the free sulfur reference composition IL ____________________________. 2 ___________________________

The University of Michigan * Engineering Research Institute EXPERIMENTAL PART Reference Compositions GR-S 1500 with 50 PHR of medium abrasion.furnace black (MAF) was used throughout the study. This mixture was further compounded with various vulcanizing agents, and in the subsequent sections of this report a particular composition shall be specified simply by identifying the nature and proportions of vulcanizing agents, In the production of GR-S 1500, 1.25 PHR of phenylP-naphthylamine (PBNA) are added by the manufacturer. In identifying the formulations PBNA will be specified only when an additional quantity of the material was used. The components of the two compositions selected as standards for reference, mentioned in the Introduction, are as follows: Reference Composition I Reference Composition II Free Sulfur + Accelerators Accelerators Ingredient Parts by weight Ingredient Parts by weight GR-S 100 GR-S 100 MAF black 50 MAF black 50 Zinc oxide 3 Zinc oxide 3 Stearic acid 1 Stearic acid 1 PBNA 1 PBNA 1 Sulfur 1.75 DPMTTSb 1 BTSCHAa 1 MBTc 1 aBenzothiazole-2-sulfenecylclohexylamide Or N-Cyclohexyl-2-benzothiazyl b su Itrenamide Dipentamethylene -thiuram-tetrasulfide C2-Mercaptobenzothiazole Compounding and Curing All batches were milled in a 3" x 8" two-roll mill using the following mixing procedure: (1) the GR-S 1500 was passed three times through a cold, tight mill (1 min.); (2) the mill was opened to 0.055" and the rubber masticated, a 3/4 cut being made each half minute (10 min.); (3) the ZnO was added (2 min.); (4) one-half of the carbon black was added, one 3/4 cut being made from each side and then the remainder of the black was added (10 min,); (5) the stearic acid and antioxidant were added (2 min.); (6) the accelerator and vulcanizer were added (2 min.); (7) three 3/4 cuts were made from each side (2 min.); (8) the batch was cut from the mill and passed endwise through the mill six times at a mill setting of 0O030" (2 minm.) and finally, (9) the mill ________________________. 5 ------------------------

The University of Michigan * Engineering Research Institute was opened and the batch withdrawn to give two uncured pads of approximately 6" x 6" x 0.1". An electrically heated hydraulic press and a 6" x 6" x 0.080" chromium plated steel mold were used to cure the samples. The uncured pads were placed in the hot mold at 307~F, the pressure was raised to 2200 psi, the samples cured for 30 minutes at 307~F, and the cured pads were quenched in cold water. Scorch Measurements The scorch characteristics of the uncured compositions were determined by use of a shearing (scored) disk viscometer at 250~F according to ASTM Method D 1077-55To Duplicate experiments were made on all compounds evaluated The scorch test, ASTM D 1077-55T, as not entirely suitable for indicating the processing characteristics of the compositions containing dicumul peroxide (DCP-40), since the peroxide tends to decompose without forming cross-links between the rubber molecules at 250~F. Another procedure was devised to indicate the behavior of uncured DCP-40 compositions in the temperature range of 230-250~F. The uncured compound was sheeted on the'mill at about 0.080" thickness and 4" x 4" pads were cut out of the sheets. The pads were placed in a circulating air oven at a temperature in the range, 230250~F, for times up to three hours, After aging~ the pads were cured by the procedure described above for thirty minutes at 307~F and 2200 psi0 The tensile properties of the cured sheets could then be used as the basis for estimating the extent of the decomposition of the peroxide due to heating the uncured compositions. Physical Property Measurements Dumbbell specimens were cut from the 6" x 6" x 0,080" cured sheets by use of ASTM tensile die C (ASTM D 412-51T). Specimens for the tear measurements were prepared by use of ASTM Tear die C (ASTM D 624-54), The ultimate tensile strength, modulus, per cent elongation and tear resistance were measured by use of an automatic recording tensile testing machine The values for the physical properties of the vulcanizates reported in succeeding tables are average values of four or more determinations. Hardness was determined according to ASTM D-676-55T. Dynamic properties yere determined by use of a forced vibrator of the type designed by S. D. Gehman. The samples, 1 cm. diameter and 1 cm. high were formed in a special mold under the same conditions of curing as described above for the 6" x 6" x 0.080" sheets0 The measurements were made at a frequency of approximately 60 cycles per second. The dynamic modulus E (kg/cm2), the internal friction rn(kilopoise)^ the heat buildup at constant amplitude Hx (calories/cycle), and the resilience R (par cent), are reported. ______________._________ 4

The University of Michigan * Engineering Research Institute Aging Experiments A mechanical convection oven was used for the aging experiments. The temperature variation of the oven at 212~F was within + 0o70F. The conditions for most of the experiments were 72 hours at 212~F, However, the aging properties of selected compositions were determined for periods of 32 days at 212~F, 7 days at 250~F and for 12 hours at 3000F. Reproducibility of Compounding, Curing and Aging Procedures Since a maximum of only 200 g. of compounded rubber could be prepared at one time by use of the 3" x 8" mill it was desirable to establish the reproducibility of procedures in order that a large number of compositions could be studied. In Table I the tensile properties of duplicate vulcanizates of the twc standard compositions are recorded. The measurements were made with specimens from the cured sheets before and after aging in the-mechanical convection oven at 212~F for 72 hours. The data for duplicate mixes listed in- Table I show satisfactory agreement. Numerous other satisfactory checks on other duplicate compositions have been obtained. __________________________________ 5 ----

TABLE I REPRODUCIBILITY OF EXPERIMENTAL PROCEDURES -i ~*.... I. _..,_..... C.urin Ss. a Tensile, psi 5 30 Modulus psi Elongation, O Hardness o OCurn SystemAged Orig. Aged Orig. Aged Orig. AgeI 1.75 Sulfur + 2680 2920 2140 -- 400 250 68 74 1 BTSCHA 175 Sulfur + 2870 2980 2170 410 250 68 7 3 1 BTSCHA 1 DPMTTS + 2210 2380 1330 1830 470 380 66 1 MBT r 1 DPMTTS + 2230 2460 1340 1850 46 90 65 67 1 MBT In addition to the curatives listedthe compositions contained 1 part of stearic acid, 3 parts of ZnO, 1 part of PBNA and 50 parts of MAF black per 100 parts of GR-S 1500.; b~pecimens vere agedrin air at 212' for 72 hours.o Specimens were aged in air at 212~F for 72 hours.

The University of Michigan * Engineering Research Institute PRESENTATION OF RESULTS Sulfur-Bearing Accelerator Vulcanizing Systems As mentioned in the Introduction, combinations of the sulfur-bearing accelerators (which are frequently used with free sulfur) were employed as the total vulcanizing system in the absence of added free sulfur, Three systems were (a) 1. PIER Dipentamethylene-thiuram-tetrasulfide (DPMTTS) 1 PHR 2-Mercaptobenzothiazole (MBT) (b) 1 PER Dibenzothiazyl-dimethylthiol urea (DBTDMTU) 1-3 PHR Tetramethyl-thiuram-d.isulfide (TMTD) (c) 1 PHR Mercaptobenzothiazole 1-2 PER Tetramethyl-thiuram-disulfide Also, as previously mentioned, the first composition (a) was selected as a reference standard (reference composition II) for non-free sulfur vulcanizates since the Rock Island Arsenal had already developed considerable information regarding its properties The tensile and tear properties of these vulcanizates, measured before and after aging, are summarized in Table II. Table II also contains similar data for the free sulfur reference vulcanizate (reference composition I). In Figures la and lb the tensile properties of the vulcanizates from the second and third compositions, (b) and (c), are recorded. It may be seen from these curves that the optimum compositions are approximately: (III) 1 PHR DBTDMTU 1.75 PER TMTD, and (IV) 1 PHR MBT 1.75 TMTD By comparing the tensile properties of vulcanizates from these two optimum compositions, III and IV, with those of the free sulfur containing reference composition I, it may be seen that they possess lower values of tensile strength, 300 modulus and hardness, about the same tear resistance and considerably higher elongations. By making similar comparisons of vulcanizates of the optimum compositions, III and IV, with the non-free sulfur reference standard II, it may be seen that the physical properties are generally comparable, with the per cent elongations of vulcanizates III and IV being somewhat the higher. When the physical properties of the aged samples in Table II are compared, however, it is apparent that vulcanizates from III and IV were somewhat more resistant to aging than reference composition II and they were substantially more resistant than reference composition I.

TABLE II PROPERTIES OF GR-S VULCANIZATES CURED WITH SULFUR-THIAZOLE AND THIURAM-THIAZOLE COMBINATIONS - =r ro Curin 5 a Tensile, psi Stress at 500bEpsi Elongation,% Hardness |Tear,lb/in. at 77~F ^ Curing System ysem Orig. Aged Orig. Aged Orig. Aged Orig. Aged (Orig. Aged. 175 Sulfur + (I) 2780 2950 2150 -- 410 250 68 74 260 170 1 BTSCHA 1 DPMTTS + (II) 2220 2420 1340 1890 470 390 65 67 280 210 1 MBT 0 1.5 TMTD + 1970 2680 800 1270 650 610 65 65 r 1 DBTDMTU 1.75 TMTD + (III) 2510 2650 950 1290 65o 580 65 65 290 280 1 DBTDMTU co 2 TIMTD + 2620 2850 1200 1620 610 520 64 69 1 DBTDMTU I 5 TMTID + 2970 2890 1750 2070 500 270 67 72 1 DBTDMTU 1 TMTD + 970 2080 440 950 64o0 6oo 61 64 1 MBT 1.75 TMTD + (Iv) 2550 2750 1020 1440 610 550 66 66 280 270 1 MBT 2 TMTD + 2580 2850 1200 1610 580 490 66 66 1 MBT aIn addition all systems contained 1 part of stearic acid, 5 parts of ZnO and 50 parts of MAF black per 100 parts of GR-S 1500. b i Samples were aged in air for 72 hours at 212~F.

The University of Michigan * Engineering Research Institute 3500 — 3500 Ultimote Tensile(Original) Ultimate Tensile (Aged 72 hrs(! 2120F) - -- Ultimate Elongation (Original) Ultimate Elongation (Aged 72 hrs(2 12~F) ---- 3000- o- 3000 0 2500 700- 2500 1~~ 22o oo60 20 00/ W 2000- % T "-2000 - / 500- \ 1500 ( / 1500 Fig 1 Effect of concentration of sul/ur-bearing 1000 \ o- 1000 a later section. ____Vulcanizates from composn IV, tt is 2 3 i 2 PHR TMTD(+l PHR DBTDMTU) PHR TMTD (t I PHR MBT) (a) (b) Fig. 1. Effect of concentration of sulfur-bearing accelerators on tensile properties of vulcanizates The scorch resistance of these compositions shall be discussed in a later section, Vulcanizates from compositions I11 and IV, that is tetramethylthiuram-disulfide in combination with dibenzothiazyl-dimethylthiol urea or 2-mercaptobenzothiazole exhibited a tendency to bloom, the blooming properties of the former being the more pronounced. It was learned through private communication2 with a company which produces low molecular weight polyethylene waxes that the addition of such wax to rubber compositions might hinder or completely prevent blooming. A few vulcanizates of the composition III (TMTD with DBTMTU) were made with 2 PHR of the polyethylene wax. The products showed slightly lower tensile strength and slightly higher per cent elongation and no bloom. While this study was limited to a few mixes of only one type of composition, the results do offer encouragement for the possibility of better control of bloom. In aVmmary, the properties of these non-free sulfur cured vulcanizates Showed considerably greater resistance to air-oven aging, somewhat lower values of tensile strength and higher per cent elongation than the free sulfur-BT'fSA- rwcAi z-ates.- Certain of. the non-free sulfur vulcanizates showed a tendency to bloom. The blooming was controlled in one composition by the incorporation of 2 PHR of low molecular weight polyethylene wax, _______________________ 9 _____..________________

The University of Michigan * Engineering Research Institute Dicumyl Peroxide Vulcanizing System The improved aging qualities of vulcanizates produced in the absence of added free sulfur, such as presented in the preceding section, have increased interest in curing systems containing no sulfur at all, such as systems based on peroxides. Much of the early work on various types of peroxides3 had not indicated much promise for commercial applicationso It happened that at the beginning of this project, however, the commercial availability of dicumyl peroxide was announced and the first information' indicated that it possessed promising qualities for rubber vulcanization. Thus, it was of considerable interest to examine this material. (During the course of this work the successful vulcanizations of nitrile and Hevea rubbers with dicumyl peroxide was reportedlY) Rubber compounds containing dicumyl peroxide were prepared and vulcanized under the same conditions and with the same equipment as used in the preparation of the two standard vulcanizates, The physical properties of vulcanizates cured with various amounts of dicumyl peroxide (DCP-40 -- 40%o dicumyl peroxide, 60o CaC03) are tabulated in Table III, the measurements having been made before and after aging. Figures 2a and 2b show graphs of the tensile and elongation values of the vulcanizates versus the original DCP-40 concentration. The original rubber compounds in Figure 2a contained only the PBNA originally present in the GR-S 1500 (1o25 PHR) while those in Figure 2b contained an additional PER of PBNAMj It should be noted that the concentration of DCP-40 corresponding to the maximum tensile value was shifted from approximater 1.75 to 2,5 PHR of DCP-40 by the addition of 1 PHR of PBNA. Thus, it was apparent that a portion of the PBNA and the DCP reacted. The optimum DCP-40 vulcanizates had somewhat low values of tensile strength, per cent elongation, 300o modulus and tear resistance in comparison with those of the standard BTSCHA-sulfur cured vulcanizate (Table II), while the hardness values were comparable. These DCP-40 products had lower per cent elongation and tear resistance values than the DPMTTS-MBT reference composition II but otherwise the original physical properties were similar. After air-oven aging, however, the physical properties of DCP-40 cured vulcanizates were better than either the BTSCHA-sulfur or the DPMTTS-MBT cured standards. The presence of ZnO and stearic acid with DCP-40 composition appeared to give vulcanizates with slightly improved ultimate tensile and elongation values compared to those obtained when only DCP-40 was used (Table III). Thus, although the use of DCP-40 as a curing agent yields vulcanizates having somewhat lower tensile, elongation, 300*6 modulus and tear resistance values than those of the standard BTSCHA-sulfur cured vulcanizate, the excellent airooven age resistance of these vulcanizates is striking.. ]0~~b

TABLE III PROPERTIES OF GR -S VULCANIZATES CURED WITH DICUMYL PEROXIDE a Tensile, psi[ tress at [/E, psi Elongation, % Hardness Tearlb/l at ('F' Curing System Orig. | Aged. Orig. oE Aged| Orig. Aged Orig Aged Orig 1 DCP-40 1710 1510 750 300 770 6oo 5350 63 65 230 230, 1.5 DCP-40 2260 2230 1420 " 1510 430 420 65 66 220 220 4 2 DCP-40 2280 2160 1650 " 1590 350 360 66 67 180 170 0 2.5 DCP-40 2220 2270 1370 200 1410 280 280 67 67 5 DCP-40 1830 1930 1160 100 1170 130 130 75 76 _, 1.25 DCP-40b 1320 1130 550 300 530 610 550 58 6o 230 230 ^ 2 DCP-40ob 2210 2060 1260 I 1170 500 490 62 65 270 230 = 2.25 DCP-40b 2310 2190 148o " 1430 450 410 65 65 2.5 DCPo40b 2320 2240 1730 " 1680 390 590 65 65 230 260 3.75 DCP-40b 2090 2290 1480 200 1460 280 280 69 70 m 5 DCP-40b 2000 2060 630 i 720 190 200 72 71 1.25 DCP-40c 990 870 390 300 390 640 620 58 6o 200 200 o 2 DCP-40c 1930 2000 1030 " 1010 510 530 61 65 260 260 2. 2.5 DCP.40c 2370 2290 1590 " 1540 430 410 65 66 230 260 ~ 5.75 DCP-40c 2320 2560 2210 " 2180 310 34o 67 67 5 DCP-40 2170 200 1950 190 210 220 72 72 0 aAll systems contained 50 parts of MAF black per 100 parts of GR-S 1500. bThe curing systems contained 1 part of PBNA, The curing systems contained 1 PBNA + 3 ZnO + 1 stearic acid. = Samples Were aged in air for 72 hours at 212~F.

The University of Michigan * Engineering Research Institute Ultimate Tensile (Original) -— O Ultimate Tensile (Aged 72 hrs(a 212~F) - - O- - Ultimate Elongation (Original) ----- Ultimate Elongation (Aged 72hrs ( 21 2~F) —- - 600 - 2500- \ 12500 00 ~~ b -500 - /H -0 C- 4 (+ OF400/oE / 1 1500,500n (a) (b) Fig. 2o Physical properties of GR-S 1500 vulcanizates cured with dicumyl peroxide Effect of Oxygen-Containing Inhibitors on the Dicumyl Peroxide Curing System It was noted above that there was some reaction between phenyl-3naphthylamine and dicumyl peroxide during vulcanization. The surprising fact is, however, that the reaction is not dominant. That is to say, the vulcanization still took place. Therefore, it was of interest to carry out similar vulcanizations in the presence of oxygen-containing inhibitors, such as phenolic antioxidants. The phenolic type materials, resorcinol, P-naphthol and hydroqqinone were used in combination with DCP-40 and the results are summarized in Table IV. In these compositions the regular GR-S 1500 containing 1.25 PHR of PBNA was used with an additional PHR of PBNA. Therefore, the data listed in Table IV are actually for compositions including a combination of the oxygen-and nitrogen-containing inhibitors. It may be seen in Table IV that the physical properties of the vulcanizates containing resorcinol and hydroquinone did not compare favorably with those of vuleanizates with PBNA alone (Table III). However, the properties of the composition containing P-naphthol were about equivalent to those of several of the vulcanizates listed in Table III. This study was very limited, and possibly the P-naphthol compositions should be examined 12

r TABLE IV ~ EFFECT OF OXIDATION INHIBITORS ON THE DICUMYL PEROXIDE CURING SYSTEM S f) Curing System Tensile, psi I Elongation, % j 300 Modulus,psi.l Hardness k3~ 2,5 DCP-40 + 1 PBNAa 1820 530 970 61 + 0.75 Resorcinol 2.5 DCP-40 + 1 PBNAa 1990 440 1410 64. + 0.3 PNaphthol = 2-.85CP-40 + 1 PBNAa 3590500 240 58 + O5 Hydroquinone aThe curing system contained an additional PHR of PBNAo 0 m+ m,

The University of Michigan * Engineering Research Institute further with consideration to the facts that Braden et al noted that certain peroxides catalyze the decomposition of rubber, and Bolland and Ten Have observed that some phenolic substances inhibit the oxidation of olefins. L_______________________l -4 _____I________________

The University of Michigan * Engineering Research Institute Dicumyl Peroxide - Sulfur-Bearing Accelerator Vulcanizing Systems Turning to the third type of vulcanizing system mentioned in the Introduction, several sulfur-bearing accelerators were used in combination with dicumyl peroxide. They included the compounds in the following groups: 1) zinc dibutyl-dithiocarbamate (ZDBDTC) tetramethyl-thiuram-disulfide (TMTD) tetrabutyl-thiuram-monosulfide (TBTM) 2) 2-mercaptobenzothiazole (MBT) diphenylquanidine (DPG), and 5) 2,4-dinitrophenyl-dimethyl-dithiocarbamate (DNPDMDTC) In brief, vulcanizates containing DCP-40 with members of the first group possessed the most desirable combination of physical properties, those from the second group appeared to have some less desirable properties, and those prepared with the product in the third group appeared to have the least desirable properties. I I I t I I I I I Ultimate Tensile (Original) -- Ultimate Tensile (Aged 72hrs (212~F) — 0 — Ultimote Elongation (Original) -0Ultimate Elongation (Aged 72 hrs@2120F) -0 —3C0001 5 "-sNyoo'soo' S! 2500 /~~~~~~ -7 ~00-" /r ^ -~~12500.500-/ / & 700- / 1000, - 2 / I I ~3 4 5 2 3 4 500 PHR DCP- 40 (+0.75TMTD,3ZnO,ISTEARJC ACID) PHR DCP - 40 (+1 ZDBDTC, 3 ZnO, I (a) STEARIC ACID) (b) Fig 35- Tensile properties of dicumyl peroxide sulfrbearing accelerator cured vulcanizates 4015 PHR DCPI - 40 (+0.75ATMTD, 3 Zn O7-I STEAR-C ACID) PHR DCP —-40 (+ ZDBDTC 3 ZnO, I 15

The University of Michigan * Engineering Research Institute Properties of the DCP-40-TMTD and DCP-40-ZDBDTC vulcanizates are shown in Figures 3a and 3b and Tables V and VI while the properties of the other combinations are recorded in Table VIIo The dicumyl peroxide - tetramethyl-thiuram-disulfide (DCP-40-TMTD) combination (Table V and Figure 3a) was improved by the presence of I additional PHR of phenyl-P-naphthylaminei 1 PHR stearic acid and 3 PHR zinc oxide.. From Figure 3a it may be judged that the optimum proportions of the curing agents are 3.75 PHR DCP-40, 0.75 PHR TMTD, 3 PHR zinc oxide plus one additional PHR PBNA. On comparing the properties of this vulcanizate with those obtained from similar compositions containing DCP-40 as the only curing agent (Table III) it is apparent that the presence of the TMTD has caused some increase in tensile strength and a large increase in elongation with little effect on the aging properties and other physical properties. Similar and somewhat more pronounced changes resulted from the presence of zinc dibutyl-dithiocarbamate with the DCP-40 (Figure 3b and Table VI). From Figure 3b, the optimum composition is 5.75 PHR DCP-40, 1 PHR ZDBDTC, 3 PHR ZnO, plus 1 additional PHR PBNA and 1FHR stearic acid. Again, by comparing the properties of this vulcanizate with those of a similar composition with DCP-40 as the only curing agent (Table III) it can be seen that the presence of the zinc dibutyl-dithiocarbamate has given considerably higher tensile and elongation values with little effect on the aging properties and other physical properties. The tensile values for the aged vulcanizates in Figures 3a and 3b are higher than those for the original products, the effect being the more pronounced for DCP-40-TMTD compositions. The reasons for this effect are not clear but must be due at least in part to additional curing during the aging process. This effect shall be discussed further in the following section. In summary, vulcanizates formed with combinations of dicumyl peroxide with tetramethyl-thiuram-diisulfide or with zinc dibutyl-dithiocarbamate, possessed slightly lower tensile and tear values, somewhat higher values of per cent elongation, and markedly improved resistance to aging when compared with the standard free sulfur-BTSCHA (reference composition I) vulcanizate; when compared with the non-free sulfur vulcanizates (such as reference composition II), these vulcanizates showed improved physical and aging properties; and when compared with DCP-40 vulcanizates (with PBNA) they showed improved physical properties and equivalent age resistance. ________________________________ 16 ______________

TABLE V PROPERTIES OF GR-S TULCANIZATES CURED WITH DICIMYL PEROXIDE-T'HIURAM COMBINATIONS -II =r ensile, psi Stress at 500Esi Elongation,, Hardness Tear, lb/in at 77~F Curing System rig. Age. Orig, Aged Orig. Aged Orig. Aged Orig. Aged 2.5 DCP-40 - b 1070 1500 440 510 590 610 60 65 210 210 5 TMTD 2.5 DCP-40 + b 590 660 290 570 66o 640 60 61l6o 8o.75 TMTD 5.75 DCP-40 +b 1120 1250 48o 570 610 610 59 65 220 240.75 TMTD 2,5 DCP-40 + C 1250 1500 510 6oo00 610 6oo 61 65 220 260.4 TMTD 5.75,DCP-40 +C 1900 1990 920 1010 540 550 62 65 210 210.4 TMTD 1.25 DCP-40 +C 1100 1470 420 650 66o 6oo00 60 65 250 260.75 TMTD 2 DCP-40 + 1550 1840 650 830 6oo 590 6o 64 28o 290.75 TMTD 5.75 DCP-40 +C 233550 2570 1170 1290 550 560 62 66 210 210.75 TMTD 5.0 DCP-40 + C 2440 2640 1490 1650 440 45o 65 68 l80 l8o - 0.75 TMTD a All systems contained 50 parts of MAF black per 100 parts of GR-S 1500. bin The curing system contained 1 additional PER of PBNA., c The curing system contained 1 additional PHR of PBNA +- 5 ZnO -+- 1 stearic acid. Samples dwere aged in air for 72 hous at 212F. Samples were aged. in air for 72 hours at 212 ~F.

TABLE VI PROPERTIES OF GR-S VULCANIZATES CURED WITH DICUMYL PEROXIDE-DITHIOCARBAMATE COl1)INATIONS a ensile, psi, Stress at 3005oE psi Elongation, a i Hardness Tear, lb/in, at 77~F Curing Syrigm Aged0 Orig, Ad Ori Aged Orig. 1 Aged Orig. Aged 5.75 DCP-40 +- 2240 2550 1500 1590 470 470 65 66 i8o 200 0.4 ZDBDITC 5.75 DCP-40 + 2170 2240 1150 1220 510 500 65 65 210 200 0.75 ZDBDTC 1.25 DCP-40 + 570 910 240 590 700 650 6i 62 18o 210 1 ZDBDTC H co 2.5 DCP-40 + 1750 1890 700 900 620 570 62 64 250 240 1 ZDBDTC 5.75 DCP-40 + 2540 2650 148o 161o 470 45o 65 66 i8o 200 1 ZDBDTC 5 DCP-40 + 2410 2540 1900 2000 570 58o 67 69 i6o l6o 1 ZDBDTC -I a 0 All curing systems contained 1 part of PBNA, 5 parts of ZnO, 1 part of stearic acid, and 50 parts of MAF black 3 per 100 parts of GR-S 1500. b Samples were aged in air for 72 hours at 212~F.

TABLE VII PROPERTIES OF GR -S VULCANIZATES CURED WITH DICUMIL: PEROXIDE —ACCELERATOR COMBINATIONS Curing System~ Tensile, psi b Stress at 500E~ psi Elongation, o rt Hardness | Tear, lb/inat 77~F.Orig Aged Orig. Aged. Or A Orig. | Aged Orig. Aged 5.75 DCP-40 + 2280 2200 - 2010 310 310 68 69 110 110.4 MBT 3.75 DCP-40 + 2110 2210 1940 1860 310 330 68 68 130 130 s.75 MBT 3.75 DCP-40 + 2360 2250 2060 1980 330 320 68 70 160 160 =.2 DPG 5 3.75 DCP-40 + 2090 2180 -- 270 280 69 71 140 150.4 DPG 3575 DCP-40 + 1360 1720 690 890 570 550 62 66 210 270.75 DNPDMDTC 3.75 DCP-40 + 2460 2480 1480 1500 480 460 65 67 190 210.75 TBTM aAll curing systems contained 1 additional part of PBNA, 3 parts of ZnO, 1 part of stearic acid, and 50 parts' of MAF black per 100 parts of GR-S 1500. s b Samples were aged in air for 72 hours at 212~F.

The University of Michigan * Engineering Research Institute Effect of Curing Time on Various Dicumyl Peroxide Systems Some evidence for post-curing during air-oven aging of dicumyl peroxide vulcanizates was noted above (see Figure 3a). It was of interest to examine the effect of curing time at 307~F on selected DCP-40 compositions. In Figures 4a, 4b and 4c variations of tensile properties with curing time are shown for vulcanizates containing DCP-40 and combinations of DCP-40 with tetramethyl-thiuriam-disulfide (TMTD) and with zinc dibutyl-dithiocarbamate (ZDBDTC). 2.25 DCP-40 + IPBNA 3.75 DCP-40 +0.75 TMTD 3.75 DCP-40 + + IPBNA + 3 ZnO + I Stearic acid I ZDBDTC+ IPBNA + 3ZnO+ 250 - I Steoric acid -600-500o-~- 2500 - 600- -- 600 200.- - - 2 - -. 00-_ 40 - 40 0.....0o- 4 e-i / 20 *-<1... 0- -50 0' / " T ---- - O. O O - S T R E S S % E 15~~~~-400 1500 at00 these3'X)- 3005t a n 300 300% Modulus f % Elongation Tensile -200- 2200- 00 25 30 (a) 35 40 25 30 35 40 20 30 40 50 60 70 ~(a~) ~(b) (C) CURE TIME (MIN @3070F) Figo 4. Effect of during time on tensile properties of dicumyl peroxide compositions In Figures 4a and 4c it may be seen that the tensile properties of these vulcanizates are not greatly changed by curing at 307~F for periods longer than about 30 - 355 minutes. However, the values of the tensile strength and 300 modulus for the DCP-40-TMTD product, Figure 4! appear to be rising at the 40 minutes point, and therefore, it seems likely that post'curing during aging can account for the behavior shown in Figure 5a. 20

The University of Michigan * Engineering Research Institute Scorch Properties of Selected Vulcanizing Systems As indicated in the Introduction, the weight of previous experience with non-free sulfur vulcanizing systems would immediately raise the question of processing stability for almost any non-free sulfur system under consideration Conventional scorch measurements were made according to the ASTM D 1077-55T procedure with a shearing disk type viscometero The results are summarized in Table VIIIo The standard BTSCHA-sulfur compound had a high resistance to scorch as indicated by the time of 61 minutes required for a five unit rise of viscosity over the minimum value (t ). The addition of 1 PHR of the retarder N-nitroso-diphenylamine (NNDPA~ increased the t5 value to 70 minutes. The non-free sulfur standard of DPMMTS + MBT had poor scorch resistance with a t5 value of seven minutes; addition of 1 PHR of NNDPA to the compound resulted in a t5 value of fourteen minutes. The TMTD-MBT and the TMTD-DBTDMTU compounds had scorch values intermediate to the two standards; t was 23 minutes for the former and 19 minutes for the latter. The compSunds containing DCP-40 alone as the curing agent had t values of 38 minutes for 3575 PHR of DCP-40, and. 62 minutes for 2.25 P~R of DCP-40. Compounds containing combinations of DCP-40 and sulfur-bearing accelerators had a wide range of scorch resistance as indicated by the t values; DCP-40 + TMTD or DPG compounds had t values of about 25 minutes while the DCP-40+ZDBDTC compound had a t5 vaTue of 40 minute s As mentioned in the Experimental Part, the conventional scorch test for the peroxide compositions is not entirely meaningful since some peroxide may undergo decomposition without causing vulcanization during the test, Therefore, the supplementary test described in the Experimental Part was used which involved heating sheets of the raw compound (as obtained directly from the mill) in a circulating air oven for periods up to three hours at temperatures in the range 230 - 250~F and then curing at 307~F during 30 minuteso The properties of three vulcanizates from compostions containing originally 2.25 and 2.50 PHR DCP-40 and 1 additional PHR PBNA are summarized in Table IX. It may be seen that the compositions tolerated preheating for about 350 minutes at 2500F and somewhat longer periods at the lower temperatures before changes in the physical properties of the vulcanizates were substantial. As previously indicated, it is difficult to state what would be a satisfactory scorch resistance since the current industrial trend seems to be toward more severe processing conditions. While none of these experimental compositions have the scorch resistance of the conventional free sulfur compound (reference composition I, t = 60 minutes), the several experimental compositions having stabilities of 20 - 40 minutes at 250~F could be used safely if their properties could justify some limitation in processing conditions. 21

TABLE VIII SCORCH RESISTANCE OF SELECTED COMPOSITIONQS...._ _.Scorch Time Curing System Io Retarder 1 PER NNDr,,, 1 PER EDHTMQ 1PR NNDSBPPDAh....... 10 5 10 1 BTSCHAd + 61 67 70 88 1.75 Sulfur -1.5 DPMTTSd + 7 8 14 16 1L MBT 1.75 TMTDd + 23 28 23 27 21 24 14 16' 1 MBT 1.75 TMTDd + 19 22 1 DBTDIMTU -~~~~~1 DBTDMT~ aTest run at 250~F (ASTM Method D 1077-55T). o2,25 DCP-40e 62 95 ro b e_ i t eis the time in minutes for a 5 unit rise over 2 25 DLc-40e 54 70 tte minimum. m 3.75 DCP40e 38 53 e OQ 53~75 DCPi-4~ 58 53 t is the time in minutes for a 10 unit rise over =' 3.75 DCP-40e + 26 33 t minimum 0.75 TMTD d In addition the system contained 1 part of stearic' 5.00 DCP-40e + 25 29 acid, 3 parts of ZnO, and 50 parts of MAF black per 0.75 TMTD 100 parts of GR-S 1500. 3.75 DCP-40e + 40 58 eIn addition the system contained 1 part of PBNA, 1 1 ZDBDTC part of stearic acid, 3 parts of ZnO,,and 50 parts 3.75 DCP0e + 65 88 of MAF black per 100 parts of GR-S 1500. f 3 75 MBT N-nitroso-diphenylamine 5.75 DCP40e + 25 35 g6-ethoxy-l2-dihydro-,24-trimethylquinoline 0.4 DPG N,N' -disecbutyl-p-phenylenediamine

TABLE IX STABILITY OF DICUMYL PEROXIDE COMPOSITIONS TO HEATING IN AIR |Tensile psi E ongation, Stress at 500 E, pssi i Hardness Curinga T Hours iHours Hours H Te. s..0. -. System - "0 1/2 1 2 3' 1/2 2 3 0 1/2 1 2 3 1/2 1 2 3 2,25 DCP=40 250 2380 2300 1770 1010 -- 490 490 610 680 -= 1240 1190 710 380 - 65 65 62 60 + 1 PBNA 2,50 DCP40b 240 2270 2230 2140 1610 1670 400 420 460 510 510 1760 1570 1340 870 950. 67 66 65 65 653 + 1 PBNA Ob.2o50 DCP-40 230 2150 2120 2080 15001510 500 520 560 590 600 1280 1130 950 780 740 65 65 65 63 62 + 1 PBNA m OQ a - In addition all systems contained 50 parts of MAF black per 100 parts of GR-S 1500. 00 b Compositions from a new supply of GR=S 1500. The new supply required 2.50 PHR DCP=40 to yield about the samv state of cure as was obtained with the previous product and 2.25 PHR of DCP-40. =r.m

The University of Michigan * Engineering Research Institute Aging of Selected Vulcanizates at Various Temperatures Perhaps the answer to the question of whether these compositions can be used lies in the question of just how desirable are the aging properties of these products. To gain further information on this point, several of these compositions were subjected to air-oven aging at 212~F for periods up to 32 days at 250~F for 15 days and at 300~F for 12 hours. The results are summarized in Tables X, XI and XII. Aging Studies at 212FThe results at 212~F are shown graphically in Figures 5, 6, and 7, in which the values for per cent elongation, tensile strength and 3050 modulus are plotted against time of aging. Inspection of these curves shows that the per cent elongation of the vulcanizates varied more regularly with time of aging than did the 300%o modulus or tensile strength. Thus for discussion, attention may be concentrated on Figure 5. ~ I 0 — 0 — 1.5 DCP-40 — 0 —-- 2.25DCP-40+ IPBNA I~0~ ~ ~ ~ -— A —- 3.75 DCP-40+-.75 TMTD — 0 — 3.75 DCP-40+ / ZDBDTC C -- ~ 0 I BTSCHA + 1.75 Sulfur 600 ----- I DPMTTS + I MBT ----- 1.75 TMTD + I MBT — |\ —-- 1.75 TMTD + I DBTDMTU 500 —_ _ Days AT 212~F z 0 ioJ 0 2 4 6 8 10 20 30 40 DAYS AT 2120 F Fig. 5. Effect of extended air-oven aging at 212~F on the elongation of GR-S vulcanizates __________ 24

TABLE X EFFECT OF AGING AT 212~F ON SELECTED VULCANIZATES Tensile, psi Stress at 300% E, psi Elongation, %.. Hardness -1 Curing System Orig. DaysC Orig. Days Orig. Days Orig. Dy 1 3 7 14 21 321 3 7 14 21 32 3 7 14 21 2 2 2 a 2.25 DCP-40t 1 PBNA 2280 2170 2140 2120 2020 1760 1320 1300 1390- 1510 1630 1710 450 440 440 430 380 310 64 66 66 67 68 69 b S 5.75 DCP-40-+.75 TMTD + 1 PBNA 2430 2620 2530 2420 2400 2270 1270 1450 1410 1490 1670 1920 520 500 490 480 410 340 65 66 67 69 69 70 b 3.75 DCP-40+ ro 1.0 ZDBDTC + J1 1 PBNA 2310 2440 2400 2440 2580 1810 1330 1430 1510 1620 1720 -- 480 480 470 450 395 280 64 66 67 68 69 70 1.5 DCP-40a 2220 1970 1800 1620 1250 610 1550 1450 1500 1640 - - 420 400 350 290 210 50 65 66 67 68 69 79 1.75 TiMTD + 1 4BT + L PBNa 2360 2720 2890 2700 2570 1950 1020 1440 1740 1910 1950 1950 610 550 470 415 400 300 66 66 66 67 69 70 b 1.75 TMTD-+ 1 DBTDMTU + 1PBNA 2130 2680 2700 2480 2390 1890 900 1330 1470 1590 1750 1860 650 590 510 440 420 310 65 65 66 68 70 71 b 1 DPMTTS +. 1 MBT + 1 PNA 2100 2400 2280 2200 1800 1330 1300 1890 2050 1960 - - 440 380 310 310 280 180 65 67 68 69 70 72 S 1 BTSCHA + 1.75 Sulfur + 1PBNA 2450 2550 2020 1750 1390 800 1430 -- - - - - 480 260 150 100 60 20 66 73 75 80 84 89 n aTn addition systems contained 50 parts of MAF black per 100 parts of GR-S 1500. 2. re bIn addition systems contained 1 part of stearic acid, 3 parts of ZnO, and 50 parts of MAF black per 100 parts of GR-S 1500. 5 C Days aged in air at 2120F.

TABLE XI EFFECT OF AGING AT 250~F ON SELECTED VULCANIZATES |Tensile, psi Elongation, %| Stress at 300% E, psi IardnessL Curing System Orig. Days Orig. Days Orig. Days Orig. Days 1 3 7 11 15 1 3 7 11 1 3 7 _ 1 3 7 11 15,. 1.75 DCP-40a 2130 1890 1320 410 1340 1320 420 370 290 60 20 1500 1330 1190 -- 65 65 66 82 93 97 a f 2.5 DCP-40 +a 1 PBNA 2140 1890 1340 480 1050 1050 410 370 320 70 20 1550 1500 1300 -- 66 66 77 91 95 b 3 4.00 DCP-40 +.75 TMTD + Io 1 PBNA 2150 2200 2120 1820 460 270 510 510 470 270 40 1200 1220 1270 - 63 65 69 71 81 92 ON b m rr 4.00 DCP-40 + m 1 ZDBDTC + 1 PBNA 2350 2450 2380 660 950 280 480 480 460 150 20 1400 1480 1560 - 65 66 69 71 89 92 3 o ~~~~~~~~0 ~~~~~~~~~~~~~~~~~~~~~~~~~1 1 BTSCHA + _ 1.75 Sulfur + 1 PBN 2920 2370 2120 1180 790 790 460 240 180 40 10 1840 --- -- - 69 73 78 83 91 95 Q b f0 1 DPMTTS + 1 MBT + 1 PBNA 2090 2190 2090 1480 690 620 500 410 350 110 30 1260 1420 1710 - 65 66 70 70 84 93 r aew supply of GR-S 1500. Systems contained 50 PHR of MAF black. In addition systems contained 1 part of stearic acid, 3 parts of ZnO, and 50 parts of M AF black per 100 parts of GR-S 1500. CDays aged in air at 2500F.

TABLE XII EFFECT OF AGING AT 500~F ON SELECTED VULCANIZATES C Tensile, psi Stress at 500% E, psi Elongation, % Hardness Curing System o Orig. -Hours Orig. Hours Orig. Hours Orig. Hours 9 12 20 9 12 20 9 12 20 9 12 20 1.75 Sulfur +a r 1 BTSCHA 2950 750 560 490 2590 ---- ---- ---- 400 90 70 40 68 75 77 80 2.25 DCP-40 +b 1 PBNA 2300 1340 990 560 1170 870 ---- ---- 550 410 510 140 64 65 66 75 R)S "^ ~~~~~a 5.75 DCP-40 +.75 TMTD + m 1 PBNA 2200 1910 1690 790 1140 970 1110 ---- 530 520 410o 150 65 65 66 70 3 a 3.75 DCP-4o + I: ZDBDTC + I -1 pDDTC +2450 2260 2070 760 1530 1500 1290 ---- 470 60 450 140 65 5 66 70 00 7r3 a P In addition these systems contained 1 part of stearic acid, 3 parts of ZnO, and 50 parts of M A F black per 100 parts of GR-S 1500. C bThis system also contained 50 parts of M AF black per 100 parts of GR-S 1500. ^ours aged in air at 500~F. qC

The University of Michigan * Engineering Research Institute 3000 1 1 1000 _ \ - 5005 -----— ~ I DPMTTS + I MBT -------- 1.75 TMTD + I MBT DAYS AT 212 ~F Fig. 60 Effect of extended air-oven aging at 212~F on the tensile strength of GR-S vulcanizates ]~~ — ~-0 —-|- -. —- - 1.5 DCP-40 -— 0 —-- 2.25 4+PBNl -0 — 2-25 DCP-40 + PBNA - ---- - 3.75 DCP-40 +.75 TMTD --- -- 3.75 DCP - 40 + ZDBDTC 0I BTSCHA + 1.75 Sulfur 0*- I DPMTTS + I MBT, —. —. 1.75 TMTD +I MBT ----- 1.75 TMTD + I DBTDMTU 0 I I I I I I - I I I I I I J100 _ 11 11 r8{d~~~~~~ ~DAYS AT 212~F Fig. 7. Effect of extended air-oven aging at 212F 21 on the 00tensile st of G-S vulcanizates ____________________________ 28 3.75DCP-4 ___0 4-.75TMTD U I DPMTTS + I MBT -4 —- I.77.5 TMTD- -I I D BTDMTU w I~ ~~i. 7. Efc fetne i-vnaiga

The University of Michigan * Engineering Research Institute In Figure 5 it is interesting to note that the compositions which best resisted the effects Of aging had been cured witho VI 3.75 PER DCP-40 + 075 PHR TMTD + 1 (additional) PHR PBNA VII 3575 PHR DCP-40 + 1 PHR ZDBDTC + 1 (additional) PHR PBNA, and V 2.25 PHR DCP-40 + 1 (additional) PHR PBNAo During the first eight days of aging the vulcanizates formed with: III 1.75 PHR TMTD + 1 PHR DBTDMTU, and IV 1.75 PER TMTD + 1 PHR MBT, lost their very high elongations; but for the remainder of the 32-day aging period their values of per cent elongation were about the same as those for the members of the above group. The values of per cent elongation for the reference composition II (DPMTTS + MBT) were, after an initial decrease, essentially constant for the 8th - 22nd day periodo The product formed with 1.5 PER DCP-40 and without added PBNA showed a uniform and fairly rapid rate of loss of elongation throughout the whole 32 day period. The free sulfur containing reference composition 1 showed relatively poor aging properties. The relatively poor aging properties of the vulcanizate formed with 1.5 PHR DCP-40 without added antioxidant as compared with those of the product formed with 2.25 PER DCP-40 + 1 additional PHR PBNA indicates that sufficient PBNA can be added to give satisfactory antioxidant protection. At the same time, an additional amount f dicumyl peroxide must be employed. From these results it appears that 1 PHR of PBNA reacts with 0,75 PHR DCP-40 or 0.30 PHR of dicumyl peroxide. The remarkable fact is that the remainder of the two agents carry out their functions separately. Aging Studies at 250~F By reference to Figure 8 and Table XI it may be seen that air-oven aging for several days at 250~F represents severe conditions, The products prepared from: 4.00 PHR DCP-40 + 0.75 PHR TMTD + 1 (additional) PHR PBNA, and 4.00 PHR DCP-40 + 1,00 PHR ZDBDTC + 1 (additional) PHR PBNA, showed the greatest resistance to aging. The reference composition II showed somewhat less resistance and the free sulfur reference composition I exhibited the least resistance, The per cent elongation for all of the compositions was essentially zero after 11 days at 2504F. 29

The University of Michigan * Engineering Research Institute 600 |1-1-1-1 1 1' 1- 1 1 1 I I' I I, 0 1.75 DCP-40 A 2.5 DCP - 40 +IPB NA ~50 C]-3u 4 DCP-40-.75 TMTD+ I PBNA ^ 4 DCP-40 I ZDBDTC+IPBNA,^\ ~~~ ~ ~~~~~ ""^^"^S X I BTSCHA + 1.75 Sulfur 1* DPMTTS + IMBT z 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 DAYS AT 250~F Fig, 80 Effect of aging at 250~fFon elongation of selected. vulcanizates Ain Studies at 300OF In Table XII the properties of the aged, vulcanizates from three DCP-40 compositions and. the reference composition I are recorded. Again, the dicumyl peroxide - sulfur-bear ing accelerator formulations exhibited improved properties. The two vulcanizates prepared from, 3575 PIR DCP-40 + 1 FHR ZDBDTC + 1 (additional) PER PBNA, and 3.75 FHR DCP-40 + 0,75 FHR TMTD + 1 (additional)PHR PBNA, showed high resistance to aging at 3000F for somrn hours. Thus, the resistance of these non-free sulfur vulcanizates, particularly those from compositions containing DCP-40 in combination with TMTD or ZDBDTC, is very striking and would justify some limitation in the processing temperatures used for BTSCHA-free sulfur compositions. 30

The University of Michigan * Engineering Research Institute Dynamic Properties of Selected Vulcanizates In order to more fully compare the properties of the various experimental vulcanizates with those of the reference compositions I and II, the dynamic properties were determined by use of a forced resonanice vibratorlo The quantities measured were E, the dynamic modulus in compression (kilograms/cm ),, the internal friction (kilopoises), Hx, the heat build up (calories/cycle at the constant amplitude of 0.0164 cm.), and R, the resilience (per cent). The values obtained for the vulcanizates at 77~F are reported in Table XIII along with the estimated errors in the determination For purposes of ~rientation and comparison, values obtained by S.Do Gehman and colleagues for the dynamic properties of GR-S and Hevea gum and tread vulcanizates are listed below. These vulcanizates were cured with a sulfur plus accelerator system and contained 50 PHR of carbon ".laeck and thus the dynamic properties can be compared directly with those i.is ed in Table XIII. E R kg/cm2 kilopoises GR-S Tread 130 55 31 Hevea Tread 80 24 45 GR-S Gum 6 3 40-45 Hevea Gum 3 1 60 As may be seen, the dynamic properties listed in Table XIII agree rather closely with those of a conventional GR-S tread stock, There are, however, little differences in the dynamic properties among the experimental vulcanizates. This similarity of the dynamic properties of the experimental products is interpreted to mean that with the same elastomer5 and the same type and amount of reinforcing carbon black, the several experimental vulcanizing systems have been cured to approximately the same levelo 351

TABLE XIII DYNAMIC PROPERTIES OF SELECTED VULCANIZATES AT 77~F r' _ _E _^ _ Hy (x = o.0164 cm.) R 3 Curing System kgc 2 kilopoises cal/cycle x 104 I - - 1 BTSCHA + 1.75 Sulfur 123+ 2 48.8 + 1.0 7.4 +.2 39.1 + 1.0' 1 DPMTTS + 1 MBTa 123 52.0 7.9 36.1, 1.5 DCP-40 123 50.1 7-6 37.5 b 2.25 DCP-40 + 1 PBNA 125 51.0 7.7 37.5 3575 DCP-40 + 0.75 TMTD a 122 52,2 7.9 36.2 + 1 PBNA 5.0 DCP-40 + 0.75 TMTD a 123 53.9 8.2 35.0 + 1 PBNA 3.75 DCP-40 + 1 ZDBDTC a 125 51.2 7.8 37.3 5 + 1 PBNA 3.75 DCP-40 + 0.75 TBTMa 123 50.1 7.6 37.5 +1 PBNA 3.75 DCP-40 + 0.75 MBT' 127 50.1 7.6 38.7 + 1 PBNA -I a In addition these systems contained 1 part of stearic acid. 3 parts of ZnO, and 50 parts of MAF black per 100 parts of GR-S 1500. b e In addition these systems contained 50 parts of MAF black per 100 parts of GR-S 1500.

The University of Michigan * Engineering Research Institute DISCUSSION OF RESULTS As a convenient basis for appraising the results, aging data, scorch times and cost information for selected compositions are condensed in Table XIV. Tl.us. composition VII (3.75 PHR DCP-40 and 1 PHR ZDBDTC) is the oest of those listed; the unvulcanized compound has a 40 minute scorch valule (t) and tne vulcanizate has the best resistance to aging. The raw materials cost for this composition is approximately 5.6% greater than that for the free sulfur —BTSCHA compound. Product V (2,25 PHR DCP-40 + 1 additional PHR PBNA) may be considered in second position; the aging properties of the vulcanizage are good, the raw compound may be safely heated in air at 250~ for 30 minutes before serious decomposition occurs, and the cost of the raw materials is approximately 1.8% greater than that of the free sulfur-BTSCHA compound. It may be possible that the results of the aging experiments at 212~F more nearly reflect the conditions of military storage than do the results of the aging experiments carried out at higher temperatures. Then compositions III - VIIL inclusive, should be placed in position of approximately equivalent preference. The age resistance of the vulcanizates is good, and the scorch resistance values range from 20 to 40 minutes. Some tensile and tear values for the vulcanizates I - VII are relisted in Table XV, The tensile values of the unaged non-free sulfur vulcanizates are a little lower than that of the free sulfur-BTSCHA product, The tear value of the unaged vulcanizate VII is somewhat lower than the tear values of the other compositions. If, however, the comparisons are confined to the properties of the aged samples the non-free sulfur and the non-sulfur vulcanizates are superior in both tensile strength and tear resistance. Thus, in view of the excellent aging properties of these vulcanizates; it is reasonable to suggest that they could be safely used with justification of the 2 - 6% additional materials costs for many rubber items, particularly in mechanical goods and military applications were good age resistance is necessary. In completing this discussion brief attention may be given to each of several miscellaneous items of interest which should be considered in fur.ther work. -______________________55 —----------

TABLE XIV SUMARY OF AGING, SCORCH AND COST DATA FOR SELECTED VULCANIZATES Composition Curing System Elongation Scorch Materials Per Cent Number t5 min. Cost, $, Cost ~ Unaged Aged 100 lbso Increase, 52 days 7 days 12 hours Compounded Over I..|..______ _______ 1O J212~F 250~F 300F __ Rubber o I 1 BTSCHA + 400 20 40 70 61 19.05 -- 1.75 Sulfur II 1 DPMTTS + 1 MBT 460 180 110 130 8 20.05 535 III 1.75 TMTD + 1 DBTDiMTU IV 15TMD+ Ba 630 300 110 270 21 20.10 5.5 IV 1.75 TMTD + 1 MBTa V 2.25 DCP-40 + 450 310 70 310 30b 19.40 1o 8 1, 1 PBNA VI 3.75 DCP-40 + 0.75 TMTD + 26 1 PBNAa 510 310 210 420 20.10 56 6 VII 3.75 DCP-40 + 1 ZDBDTC + 40 1 PBNAa s aA Average values are listed. bpreheated before curing in air oven at 250~F, see Experimental Part. Estimated from market prices listed in Ruabber Age 79, 870 (1956).

The University of Michigan * Engineering Research Institute TABLE XV EFFECT OF AGING ON TENSILE AND TEAR STRENGTH OF SELECTED VULCANIZATES Composition Tensile Strength psi Tear Strength lbs/in, Number UnagedAged at 21F U'naged Agged at 212,F 7 days 32 days 3 days I 2650 2020 800 260 170 II 2100 2280 133550 280 210 a III a 2270 2790 1920 290 280 IV V 2280 2140 1760 230 260 VIa 2370 2460 2040 190 200 VIIa a Average values are listed. 35

The University of Michigan * Engineering Research Institute The question of the stability of DCP-40 compositions during processing still remains to be settled. The half-life of the product (in the absence of rubber) is listed by the manufacturers as 6 hours at 2480F. It was noted above that the DCP-40 raw rubber compound (which had been milled in air) could be heated at 250F in air for 30 minutes without producing serious changes in the properties of the product after regular vulcanization (30 minutes, 3070~F, essential absence of air). It is assumed that these data define the limitations of retaining the unvulcanized.composition on a hot mill but do not bear directly on Banbury and extrusion operations carried out in the essen tial absence of air, Mention has been made of the problem of bloom in vulcanizates of cfompositions III and IV and that the results of preliminary experiments showed promise for the use of low molecular weight polyethylene wax as a-means of solving the problem. Further work might well include attention to the -blooming tendency of these vulcanizates since the resistance to aging and scorch of the products represent an improvement over those of refereice product II (DPMTTS + MBT). It is interesting that vulcanizates of compositions VI and VII containing DCP-40 showed no bloom..The general results of this study are in accord with the viewl1 that vulcanizates formed from systems containing non-free sulfur and no sulfur are likely to have improved aging properties0 The results may also be qualitatively related to estimates of the strengths of the different kinds of bonds consitituting the cross-links0 For example., Dogadkin and Torasova list strengths of the probable cross-links formed in sulfur vulcanization as followss Bond Strength of Bond Possible Sources of Bond kcal -C-C- 62.7 free radicals from peroxides, heat, light, etc. -C-S-C 54-5 disulfides -C - -S -C 275 tetrasulfides -C-Sn-C- free sulfur Possible sources of the different kinds of cross-links are also listed. From this point of view then, the weaker the cross-link the more likely it may be broken to give active centers (presumably free radicals) which cause scission of the main chain as well as cross-linking and branching reactions. The overall results include shortening the trunk, chains, more branching and crosslinking which, in turn, leads to decreasing tensile strength and elongation and increasing hardness and brittleness. Thus, according to this view, the vulcanizates from free sulfur and tetrasulfide combinations would be expected to be generally less stable than those from non-sulfur, and mono- and di-sulfide compositons. _____________________ 6 _____________

The University of Michigan * Engineering Research Institute The fact that phenyl — naphthylamine, a free radical inhibitor, can be used in combination with dicumnyl peroxide, a free radical generator, has been emphasized, While there is some reaction between the two, a -small increase in the proportion of each in the rubber composition leads to satisfactory vulcanization and a product with excellent aging qualities, Possibly related to the above is the fact that certain nitrogensulfur-bearing vulcanization accelerators (Table XIV) in combination with dicumyl peroxide yield a vulcanizate with improved physical properties and equally good aging properties. Between the Rock Island Laboratory~2 and the members of the project, three suggestions have been made which may lead to an understanding of the situation, namely: 1) in each case, the unreacted'portions of the nitrogen or sulfur-nitrogen compounds (remaining in the product after vulcanization) may act as inhibitors for aging in air; 2) in each case, the nitrogen- or the sulfur-nitrogencompounds may induce the decomposition of the peroxide to yield free radicals by a process of lower activation energy than the thermal decomposition, resulting in a faster decomposition of the peroxide,, but, at the same time, some wastage of peroxide; and 3) particularly in the latter case., the nitrogen-sulfurbearing compounds might act as dispersing agents for the DCP-40o (The presence of ZnO and stearic had a beneficial effect on the DCP-40 vulcanizate.) While the results summarized in this study do not afford a decision regarding the relative merits of these suggestions, certainly each suggestion is straightforward.and should be considered in further study of the subject. In conclusion, several peroxide and peroxidex-sulfur-bearing accelerator combinations have been found to yield vulcanizates possessing excellent aging qualities, and good tensile properties, The processing stabilites of the unvulcanized compositions, while not equivalent to those of conventional free sulfur compositions, appear to be suSficient for practical application. 57

The University of Michigan * Engineering Research Institute REFERENCES 1. So Do Gehman, Do E. Woodford, and Ro Bo Stambaugh, Ind. Eng. Chem., 33, 1032 (1941). 2. Private Communication from Semet-Sovay Petrochemical Division, Allied Chemical and Dye Corp., New York, N. Y. 3. Ao van Rossem, Po Dekker, and R. S, Brawirodipoero, Rubber Chem. and Technol., 5, 97(1932); Eo H. Farmer and C, G. Moore, J. Chem. Soc., 142(1951). 4, Hercules Powder Company, New Product Data Sheet No. 4-5-55. 5. M. Braden and W. P. Fletcher, Trans. IRI, 31, 155(1955); C. Ho Lufter, Rubber World, 1335 511(1956). 6. Mo Braden, W. P. Fletcher, and Go Po McSweeney, Trans. IRI, 30, 44(1954) 7. Jo Lo Bolland and P. Ten Have, Disc, Faraday. Soco, 2(1947). 80 Jo H. Dillon and S. D. Gehman, India Rubber World, 115, 2(1946); R, S, Enabnit and S. Do Gehman, Ind. Eng. Chemo, 43, 346(1951); Ko:. Gui, C. S. Wilkinson, Jr., and So D, Gehman, ibid., 44, 720(1952) 9. Hercules Powder Company, New Product Data Sheet No. 5-2-55. 10, A. Lo Pedersen, "Aging Properties of Low Sulfur Rubber Vulcanizates," a communication from Northern Cable and Wire Works, Ltd,, Copenhagen (1954); A. Haehl, Rubber Chem. and Technol., 2., 147(1954); Ao S. Kuzmenskii and S. Io Bass, ibid., 2, 793(1955)o 11o B. Ao Dogadkin and Z. N. Tarasova, ibido, 27, 883(1954); B. A. Dogadkin and Reznikovskii, Kolloid Zhur, 11, 314(1950). 12. Letter from Z. T. Ossefort, Rock Island Arsenal, August 27, 1956. --------------------- 8 —----------