Julv. 1930
INDUSTRIAL AND ENGINEERING CHEhfISTRY
759
Structural Changes during the Processing of Rubber' Ernst A. Grenquist THE FISKRUBBERCOMPANY, CHICOPEE FALLS, MASS.
The effect of mastication, heating in air, oxygen, steam, and carbon dioxide gas, and the influence of light upon the plasticity of crude rubber have been investigated. Cold milling causes a permanent change in the consistency of crude rubber, whereas heating in air or steam produces a temporary effect. In absence of air the rubber exhibits a maximum amount of disaggregation at each given temperature. At low temperatures the increase in plasticity proceeds at practically the same rate whether the rubber is heated in oxygen or in carbon dioxide. Ultra-violet light affects concurrently both the plastic and elastic properties of the rubber. The correlation between elasticity and double refrac-
tion of crude rubber has been shown. The anisotropic properties of rubber parallel the recovery values. In this investigation no agency has been found which causes such far-reaching and profound destruction of the elastic properties of the rubber as prolonged physical mastication. The interrelation between plasticity and structure of crude rubber has been studied and the fundamental difference between actual mastication and heat breakdown has been emphasized. Prolonged milling, heating in air, or in oxygen destroy the protein framework of the rubber. Heating in steam or in an inert gas leaves it more or less intact.
Structural Nature of Crude Rubber
(20, 21) on the isolation of two separate fractions of rubber hydrocarbon, have been further steps toward the interpretation of the elastic and plastic properties of the rubber hydrocarbon. According to Bary and Hauser (2) rubber, on this basis, consists of a network of rubber hydrocarbon in a high state of aggregation and containing particles of a lower state of aggregation. Rubber, therefore, may be visualized as a jelly-i. e., as a system where the phases differ only as to their state of aggregation.
H E structure of the individual latex globule has been investigated in detail by Freundlich and Hauser ( G ) , whose work, together with that of Sebrell, Park, and Martin ( 2 4 , van Rossem (as), and others, indicates that ordinary crude unmasticated rubber is a very clojely packed mass of discrete latex globules. De Vries (29) also assumes that during operations such as coagulation and drying in air the rubber globules remain intact. In recent excellent researches von Weimarn (SO) has found that if rubber latex is treated with solvents or dispersing agents for proteins, such as lithium salts, the rubber will solidify during coagulation into a microscopically homogeneous coagulum. Microscopical work described in a previous paper ( 9 ) indicated that crude, unbroken smoked sheet has its globular structure at least partly destroyed a t the plantation during coagulation and sheeting. On mastication the rubber showed a gradual destruction of the globules, and finally, in dead-milled rubber a matrix was observed containing embedded small globules, which seemed capable of more resistance towards disintegration during milling than the larger ones. The proteins and natural resins meanwhile showed a gradual aggregation. The globular structure in acetone-extracted rubber had disappeared completely. It is important, however, t o differentiate distinctly between a globular structure and the fiber structure of rubber under stretch. The globular structure is associated with the secretion by the Hevea tree of pear-shaped and spherical rubber globules. These, to a greater or less degree, retain their identity in the crude rubber during processes following the coagulation. The fiber structure refers to the aggregation of the rubber hydrocarbon itself and is a logical interpretation of x-ray studies of rubber by Katz ( I d ) , Hauser and Mark ( 1 1 ) , Clark (S), and others. Hoch and Siedler ( I S ) have shown that it is possible to break up both cured and uncured rubber into fine fiber bundles by percussion of rubber that has been frozen in liquid air under stretch. Investigations on the double refraction of stretched rubber in po1arizc.d light by Zocher and von Fischer (SS), Kroeger ( I C ) , van Gee1 and Eymers ( 8 ) ,and others also substantiate the conception of a fiber structure. Chemical work by Staudinger (X)on molecular chains of rubber, and by Pummerer and collaborators
T
i Received April 16, 1930. Presented before the Division of Rubber Chemistry at the 79th Meeting of the American Chemical So( iety, .4tlanta, Ga , April 7 to 11, 1930
Theories as to Changes during Processing
The chief agencies which, after coagulation, affect the structure and consistency of crude rubber during processing are mechanical work and heat. The effects of solvents and chemical agencies have not been considered in this investigation. A number of theories are to be found in the literature explaining the gradual plastic change taking place in rubber during mastication. Pickles (19) points out that the flat portion of the plasticity curve begins at a point where the original cell structure has been broken down and represents the consistency of a homogeneous mixture of the various constituents of the rubber. Practically the same opinion is held by van Rossem (24). Garner (7') maintains that during the mastication of rubber globules are partially burst, and that depolymerized hydrocarbon is also formed. The tackiness is due to the action of oxygen and cannot appear until the latter is available. Klein and Stamberger (15), working with benzene solutions of crude masticated rubber, found that the number of ultra microscopical particles increased continuously with the milling. The original crude rubber formed a homogeneous suspension, whereas the completely milled rubber formed a heterogeneous suspension. This, they claim, indicates a breaking up of the protein framework during the milling. During mastication the rubber is influenced, not only by mechanical work, but also by heat, by oxvgen of the air which becomes mixed with rubber, by light, etc. Temperatures below 100" C. seem to have only a slight effect upon rubber according to van Heurn (12) a:id de Vries (28). The effect of higher temperatures is more profound, especially in the presence of air. Staudinger and Geiger (27) found that the melting, or rather softening, point, where the rubber forms a soft mass, varies with impurities, with the mastication, with the time of heating, and with the presence or absence of oxy-
Microsraph 1-Unmasticafed 800
II* = 5.M
P
=
34.5
x
Rubber.
H, = 8.32
Q
i0.364
Micrograph 2-Rubber of Millin*. rr, = 2.22 I' = 82
after 40 Minuter 800 x 111 = 2.33 i) = 0 . 0 1 0
Mlcroaraph 3-Rubbez after 90 Minufes of Mllling. 800 X Ha = 1.43 IIj = 1.46 P = 88.7 Q = 0.003
Experimental Procedure
I'r.nsi~rciml'XSTs~--'h charrges i n plnsticity of rubber taking place under the influence of various ageiicies\+we determined according to Williams ( S I ) . Tlie met,hod was selected in preference to viscosity determinations in order to obtain flgorcs for 1'0th tiie flow and tlir recovery of the oriidc rubber. A modification of the ~7'illiams plastometer w a s used, supplied with roller guides t,oobtain a parallel movernent of tho upper platen in t,he pross. T h e 2-ce. rubber riellet of cvlindrical shaue was nrelieated for ')o For 5 minuks a t 100" C. . The cokpressing Mifroltraph 5--Rubher ~ i n u t ~ a Polarized in Llaht at 300Per force was 5000 grams. The test piece Genf Stretch. 200 x was placed in the press with the plates I I * = 5.80 A i = 8.32 H i = 1.46 P -345 V = 0.364 0 = 0.008 held apart by a small block of steel, and t,he readings were timed froin the moincnt geii. Fry arid Porrit ( 6 ) claim that nrcclianical work, heat the steel block was withdrawn which let the weight fall on and atrnosplierie osygeii are all responsible for tlie physical the sample. The original height of the rubber pellet, H,, is a changes accompanying milling. At 150' C. tho curve of vis. constant and equals 12.70 mm., corresponding to tlie edge cosity against i d l i n g closely approximated the curve of vis- of the steel block which keeps the plates apart. The f l r w cosity against heating. nnmber, Ht, is the height of the pellet in rnillirnot,ers after 3 The plastic changes taking place when rubber is heated in minutes of compression. Tlic recoverg%-alut:,Ifa.i8 tho height air are without doubt to a great esl,ent caused and regirlaterl in millimeters 1 minute after the load is removed and tlie by the atmospheric oxygen. Van Rossem (B) studying , the pellet has been allowed to recover under iio load. oxidation of rubber by heat, concluded that tlie oxidation IWX The plasticity of the crude rubber was expressed in form a secondary state preceded by depolymerization and that the of flow number If2,recovery value Ii, and per cent permantnit depolymerizat.ion by heat is catalytically accelerated bg H , - )% oxygeii. It is evident that there are two distinctly different set, which equals H , . In order to oht,airr an expression effects produced by heat alone or by heat in combination embodying hot,li flow and recovery of the rubber the followiiig with oxygen or with air. This quotient, which we will quotient %vasused: Ha H , - H,' In order to ~ t i i i l ythe effect of lieat alone, inrestigations call Q in the iollow-iiig discussion, represeiit~s the ratio between have beell made in whieti rubber was heated in steam or in an inert gas. Wnrthington and Hyde (32) heated robber in flnw and recoirerg in per cent arid approaclies zero the softer the rubber and approaches one the greater the elasticity. a11atmosphere of live stearn free from air in order to produce a product of uniform softness. Park, Carson, and Sebrell The expression should not be geiiersliseil and is used in this (23)heated raw rubber at 158' C. for 24 hours. They show investigation for relative comparison of tiie eBect of various a n increase in plasticity corresponding approximately to agencies upoii the plasticity of crude rubber. Per cimt perniaiient set, will be designated xritti 1'. five passes through a fairly tight mill. Finally, theeffect of light upon thestructure and coiisisteiicy Gi,onur,~ii STRl!trctcli was selected as a qualitative indication of the fiber structure a i d state of aggregation of the rubber hydrocarbon. Strips of crude rubber 10 X 1 x 1 mm. were stretched 300 per ceiit crosswise over the slide and examined under crossed nicols in a polarizing microscop(>. The intensity of double refraction was noted and correlated with globular structure and plasticity figures. Photomicrographs were taken using a powerful arc lanip as a light source, light filters, and panchromatic plates. P R E P A R h T I O N O F RUBBERSAMPLES--\ batch Of 90 kg. Of crude rubber obtained by blending selected parts of crude unmasticated smoked sheet was used for the experiments. The rubber was vashed for 12 minutes on corrugated washing mills to remove impurities, dried, and homogenized for 6 minutes in a 13anbury mixer. Thirty days were allowed to elapse before starting the various experiments to let the rubber a s u m e constant conditions. Mastication
Four thousand grams of rubber were broken down on a 20-inch (50.8-cm.) mill with cooling water on full a t an average roll temperature of 56" C., and 200 grams of rubber were taken off every 5 minutes for examination. The brokendown rubber was allowed to recover for 24 hours in a constanttemperature room a t 21.1" C. and 65 per cent humidity, and the plasticity figures mere determined. The plasticity determinations were repeated after 42 days of standing in constant temperature to establish the permanence of the changes in consistency brought about during milling. Table I shows the changes that crude rubber underwent when milled for 90 minutes under standard conditions. The flow numbers and recovery values represent the average of three detprminations. Column I gives the figures 24 hours after milling and column I1 those after 42 days of standing. Table I-Effect
TIMEOF1 MILLI\G
I
I
resins. Progressive milling caused a loosening up of the protein framework, the individual constituents assumed more distinct contours, and a matrix was formed in connection with destruction and fusion of globules. After 40 minutes of breakdown (Micrograph 2) an optically empty matrix is 01)served as a background, in which individual globules are pilibedded in great number. The proteins and resins which in unmilled rubber form the border lilies between the rubher globules are beginning to aggregate. After 90 minutes of inilling (RIicrograph 3) the globules have largely disappeared and oiily the very small ones remain. A hoinogeneous matrix has been formed in which, with exception of intact rubber globules, the aggregates of proteins and resins are diatrihuted. The unmasticated rubber showed an intense double refraction a t 300 per cent stretch and the whole microscopical field was lighted u p under crossed nicols. (Micrograph 4) Bright interference colors and indication of fiber alignment appeared parallel with the stretch. With progressive milling the double refraction decreased in intensity and ceased almost completely after 90 minutes. (Nicrograph 5 ) The microscopical field remained practically dark and only the resin crystals lighted up the field. Of interest was the observation that the intensity of the double refraction fairly closely followed the recovery values of the rubber as obtained by plasticity determinations. The higher the recovery, the more intense n a s the double refraction. With destruction of the elastic prvperties of the rubber the anisotropic properties disappeared almost completely.
of Mastication on Plasticity of Rubber
I
ff2
it31
I1
I-
P
H3
I
Q I
I1
0.364 0.070 0.039 0.025 0.014 0.010 0 006 0.003 0 003
0:074 0.033 0.022 0.009 0.006 0.005 0.000 0.000
\I. in ut e 10 20 30 40 60
90
1
5.80 4.26 3.57 2.9% 2 51 2 21 1.78 1.65 1.43
3'97 3.36 2.90 2.45 a 34 1.88 1.56 1.56
8 4 3 3 2 2 1 1 1 1
32 85 93 16 65 33 84 68 46
4:62 3.66 3 12 2 54 2 40 1 94 1 56 1 56
34.5 61.8 69.0 75.1 79.1
63:6 71.2 75 0 79 8
88.7
87.6
~
The effect of mastication is expressed graphically in Chart 1. From the decrease of the quotient Q it can be seen that by far the greatest changes have taken place during the first 10 minutes of milling. The experiment also shows that cold mastication causes a permanent alteration of the consistency of the rubber. Practically the same results were obtained 24 hours after milling and after 42 days of standing. The immediate recovery which takes place during the first 8 hours after milling during hot storage according to Griffiths (IO)has not been considered in this investigation, as the samples were given a rest period of 24 hours before the plasticity figures were determined. Micrograph 1 shows the globular structure of unmasticated rubber. The rubber exhibits a very close network of globules, interspersed, however, with casual aggregates of proteins and
Chart 1-Effect
of Mastication on Plasticity of Rubber
Heating in Air
The rubber was sheeted out to uniform thickness of 5 inin. on the 20-inch (50.8-cm.) mill, by means of three passes; 200-gram samples were heated in a n electrical oven a t 100" C. The plasticity figures were determined after 48 hours' recovery in a constant-temperature room, and the deterininatioiis repeated after 30 days of standing under constant conditions, to establish the permanency of changes observed. The changes taking place during increasing length of heating are presented in Table 11. Column I gives the figures 48 hours after heating and column I1 those after 30 days of standing. It can be seen that the heat chiefly influenced the recovery values of the rubber during the first hour of heating and that consequently a disaggregation of the rubber hydrocarbon took place. There were no indications of tackiness u p to 150 minutes. During prolonged heating the disaggregation was
INDUSTRIAL AND ENGINEERING CHEMISTRY
762
followed by oxidation and after 18 hours the rubber was tacky and had lost its elasticity to a considerable extent. The globular structure and double refraction showed only slight changes during 150 minutes of heating. I n order to study the effect of heating a t higher temperatures the experiment was repeated a t 150" C. These results are also presented in Table 11. After 60 minutes of heating the rubber appeared tacky, but still retained a considerable elasticity. A marked oxidation had taken place, which was first noticeable on the surface but which gradually spread through the whole rubber mass.
H2 Minutes 1000 c. 10 60 90 120 150 18 hours 150' C. 10 30 60 90 120
KO.
7
latex to 150-200° C . The globules assumed a spherical shape and the sire increased up t o 6-10 microns.
The globular structure showed only slight changes up to 30 minutes. A considerable aggregation of proteins and resins took place during the heating in connection with bursting of globules, which checks with previous observations published in a study of the vulcanization process under the microscope (9).
of Heating i n Air on Plasticity of Rubber
Table 11-Effect
TIME
Vol. 22,
I1
5.60 5.03 4 SO 4.72 4.43 2.78
5.25 5.13 4.87 4.92 4 70 3.00
8 6 6 6 5 3
24 62 27 03 58 01
I1
7 6 6 6 5 3
Q
P
H3
I
I
I
58 97 48 35 98 18
1
35 47 50 52 56 76
2 9 8 3 2 0
I1
40 45 49 50 55 75
5 2 1 0 0 0
I
11
0 372 0.211 0.186 0.164 0.139 0.023
0.363 0.243 0.205 0.184 0.152 0.018
5
I
1
I 0
40
24
HOURS
5 4 2 2 2
43 68 73 65 47
5 4 3 2 2
37 65 03 67 56
7.95 5.91 2.92 2.75 2.56
7.53 5.85 3.51 2.94 2.80
41.2 53.3 76.9 78.2 80.6
40.7 54.0 72.1 76.9 78.0
0.346 0 162 0.019 0.010 0.009
0.294 0.146 0.050 0.027 0.024
Chart 2 shows the effect of heating in air a t 100" and 150" C. upon the plasticity of the rubber. It can be seen that a t 150" C. the changes proceed rather slowly up to about 60 minutes, as compared with mastication, while beyond this point the curve becomes fairly constant. Heating for 2 hours a t 150"C. does not bring about so great a change in plasticity as 90 minutes of milling. The increase in plasticity probably is due partly to disaggregation of the rubber hydrocarbon, partly to oxidation, and partly to destruction of globular structure. There was a decided reaggregation during 30 days of standing, contrary to the case with masticated rubber (17). The rubber also had become less tacky to the feel.
Chart 3-Effect of Heating i n Oxygen a n d Carbon Dioxide on Plasticity of Rubber
The double refraction of crude rubber heated for 10 minutes in air a t 150" C. did not differ essentially from that of unheated rubber. After 2 hours of heating the anisotropic properties of the rubber were destroyed to a great extent and faint interference colors appeared in grayish blue under crossed nicols. Heating in Oxygen and in Carbon Dioxide
The rubber was sheeted out as in the previous experiment and heated in a Bierer-Davis oxygen bomb a t a temperature of 70" C. and an oxygen pressure of 300 pounds per square inch (21.09 kg. per sq. cm.). It would have been advisable t o use the same temperature as in the previous experiment for a complete comparison, but the experimental possibilities were not a t hand. Samples of rubber were heated for increasing lengths of time up to 72 hours. The experiment was repeated, replacing oxygen with carbon dioxide, to investigate to what extent the changes in plasticity were caused by oxidation and to what extent by heat, assuming that carbon dioxide would not affect the rubber. The results obtained are presented in Table 111. of Heating i n Oxygen a n d in Carbon Dioxide upon Plasticitv of Rubber
Table 111-Effect
TIME
I
I I 5.33 0 2
Minutes 30 60 120 300 Houvs
5,13 4.98
1
.., .
Chart 2-Effect
of Heating i n Air on Plasticity of Rubber
Micrograph 6 shows the globular structure of the rubber after hours of heating a t 150" C. It can be seen that a destruction of globules has taken place, although it is less pronounced than that during prolonged mastication. A matrix has been formed probably owing to the bursting and fusion of globules. Note-In a private communication Hauser has described unpublished observations on fusion of latex globules when heating protected liquid
P
Q
coz 5.27
....
....
5.15 4 72
MINUTES
1
H3
H2
0 2
7.78 7.54 6.98
7.82
.., . ....
....
7.47
. . 6.35
6.46 6.03 5.87
,
,
3.30
4.72
,
...
38.8 40.6 45.1
38.5
... ....
41.3
..,
49.3 52.5 53.9
,
0. ..3.4.3.
,
....
50.0 56.2 62.8
0 . 3 13 83
coz
....
0.259
, . . . .
.... .
0.307
.....
0.21s 0.187 0.171
0.199 0.138 0.078
. ....
Chart 3 shows the effect upon plasticity of heating rubber a t 70" C. in oxygen and carbon dioxide. The decrease in elastic properties after 24 hours of heating was practically the same in oxygen and carbon dioxide. There was a more rapid initial decrease in case of oxygen. After 24 hours of heating the carbon dioxide curve became fairly constant, whereas the elastic properties continued to decrease when heating in oxygen.
July, 1930
INDUSTRIAL A N D ENGINEERING CHEMISTRY
Microgrs h 6-Rubber Heated In Air for 2 h r 8 at 150' C. 800 X Hs = 2.66 Ha = 2.47 1' = 80.6 0 = 0.009
M i r r a r a p h ?-Rubber Heated in Oxygen for 72 Hour8 at 70- C. 800 x H* = 3.95 IIs = 4.72 P = 62.8 a = 0.078
Micrograplr 7 shows tlie effect oiloxygen on the globular structure of the rubber. There were practically no changes u p to 1 lrour o i heating. Aiter longer periods of heating a destruction of globules was noted which was a.ccoinparried by aggregation o i proteins aid resins. The globules were retained almost completely in carbon dioxide during the whole period of Iicating. (Micrograpli 8) The interference colors faded both in oxygen a.nd in carbon dioxide. Heating in Steam
Twn hundred gram samples o i 5 mm. tliick slieets were heated in a pot press in live steam at 54 pounds per sqnare inch (d.796 kg. per sq. cni.) corresponding to a tempratme of 150' C. Steam was blouw through the pot press five times aiter raising tlie pressure to 54 pounds each time to assure that no air rcmainwi. The pot was allnxed to cool for 10 minutes hefore being opened and the rnhber immediately after being t,akeri out was iinniersed in cold water to prevent air oxidation. During the steam treat,rnerit the rubber absorbed approximately 10 per cent of Tt-ater and sitowed a milky \%-bite color. The water was rensored by drying the slieets in an electric oven a t 5.5" C. witli circnlatirig air until complete transparency was attained. The nibber ri:covered somewbat and lost its pliability during drying. Tlre plasticity figures were det,ermiiied 48 hours after heating and aiter 30 days of standing in a consta,it-temperature room a t 21.1' C. and 65 pcr ceiit humidity. (Table IT')
763
Microwaph 8-Rubber Heated in Carbon Dioxide for 48 Hours a t 70' C. 800 X 11, = 4.45 H a = 5.87 Q = 0.171 P = 63.9
The increase in plasticity when h e a h g in steam a t 150" C. is considerably less than when heating in air a t the same temperature, A marked recovery of the rubber took place during 30 days of standing. In order to obtain maximum breakdown with the equipincirt at hand, the experiment was repeated a t 85 pounds pressore (5.976 kg. per sq. em.) corresponding to 164" C. The rubber was allowed to dry a t a constant teniporature of '21.2" C . and 65 per cent Irumidity. The sheets did not become transparent until aiter l week oi drying. The plasticit,y figures were determined 10 days d t e r lioating arid the determination repeated aiter 48 additional days of standing. l'lre results are presented in Table 1V.
-
0
6
I* &"%*
4.TLI ms.T,"*
*/rtr e
*-. ov>r-*n,*r I
u
e.
,/
HOUSS
Chart 4-Effect of Ef
