The unique design latitude of EPDM.

During the '80s, EPDM rubber has experienced a growth rate well above 5%. In the '90s, although at a lower rate, further growth is expected in all applications. In the automotive segment most of the increase will involve body sealing, which today already accounts for half of the automotive segment.

By meeting the new automotive requirements and allowing high productivity, tight dimensional control and excellent consistency, EPDM is projected to reach 160,000 tons by the year 2000 in body sealings. This challenge requires a thorough understanding of the numerous factors affecting the production of increasingly complex components such as multiple extrusions including rigid supports and low friction coating. Within the increasingly generalized R&D constraints, effective component design will require deep cooperation among auto producers, rubber transformers and polymer suppliers.

The increasing desire of the automotive industry to improve aesthetics, aerodynamics and comfort has prompted the need for body seals of greater size and complexity. Because of its well known advantages (ref. 1) for handling complex shapes and its operational flexibility, ultra high frequency (UHF) has become the vulcanization process of choice for the production of coextruded body sealing profiles. Its extensive use has however highlighted some limitations. The optimum balance between functional properties such as noise insulation, water/air tightness, easy door closing and aesthetic requirements such as surface smoothness can only be achieved through time and cost consuming iterative trials on a production scale. UHF production rates are limited because of compound polarity differences between the solid component and the sponge seal. Furthermore, the presence of a metal insert in the solid component is a source of interference in the microwave heating and reduces vulcanization efficiency.

Objective and approach

In a previous work (ref. 2) we introduced empirical correlations to predict performance attributes of UHF cured sponge seals and we outlined the Exxon Chemical approach to the rational design of a new grade for sponge applications - Vistalon 8600. It is the intent of this article to provide further rationalization for the design of body sealing. Operational set up parameters were explored together with compounding variables affecting microwave heating. Results provide an understanding of the operating parameters which are key to reaching high standards of quality and productivity for the manufacturing of coextruded body seals. The potential of molecular weight tailoring technology to widen the design and operational latitude of the solid component of coextruded profiles is highlighted through the superiority of polymers with a tailored molecular weight structure versus traditional grades.

Experimental work was conducted as per following sequence:

* Monoextrusion to assess the influence of infrared preheating on sponge production;

* Coextrusion of a tubular sponge seal onto a hard solid support to identify optimum line setup for consistent operations;

* Bench scale testing to optimize formulations for balanced dielectric heating of sponge and solid components for increased productivity;

* Preliminary definition of grade structure offering a superior balance of processability and vulcanization rate for the production of the solid component.

Influence of IR preheating on extrusion of a sponge profile

Tests were run on the Exxon Chemical semi-industrial UHF line to assess the effect of sponge profile preheating via an infrared tunnel located prior to the UHF tunnel, with a view to obtaining a smoother surface. Extruder and UHF operating conditions were kept constant while IR power was progressively increased. Profile temperatures at UHF inlet and exit we,re recorded and sponge properties measured. The profile temperature was found to linearly increase with IR power. However, profile temperature at UHF exit levels off when 120 [degree] C are reached at the IR exit. As expected the temperature increase across the UHF tunnel is related to the compound microwave receptivity.

The cellular structure of the profile portion exposed to IR radiation is shown to be changed with formation of larger cells (figure 1). The introduction of an IR tunnel modifies sectional profile temperature gradient between the extruder exit and the UHF exit; as a result, higher temperature of the exposed portions has affected the expansion phase through an earlier skin consolidation and a higher average blow rate.

This structure can be associated to a lower load deflection at no significant surface aspect penalty (figure 2). Therefore, IR preheating appears beneficial as it increases blowing agent efficiency and allows to reduce the levels of azodicarbonamide (figure 3).

To summarize, for the production of monoextruded sponge seals, IR preheating is a means to decrease load deflection and to improve process economics (lower blowing agent levels, hence higher mixing consistency).

Coextrusion: Effect of operating parameters

Coextrusions of a tubular sponge seal onto a hard solid foot (without metal reinforcement) were run on our semi-industrial UHF line. For this purpose, the UHF extrusion line was equipped with co-extrusion facilities including two 45 mm Troester extruders and a coextrusion head. A full factorial experimental design was selected to study the effects of extrusion line speed, extruder die temperature and IR preheating.

Because of the polarity difference between the sponge compound and the solid component, die UH power had to be increased to 2.4 kW (versus 1.2 kW in the monoextrusion study); higher powers were disregarded to avoid fire risks resulting from excessive heating of the solid component. Under these conditions, the maximum temperatures measured during the experiments were 130 [degree] C for the sponge and 200 [degree] C for the solid.

As expected, load deflection correlates reasonably well with specific gravity and was taken as the key sponge quality/consistency indicator. Within the limits of our tests, surface aspect was almost unaffected by the operating conditions. Sponge extrudate mass temperature positively correlates with die temperature and line speed (figure 4); the effect of line speed can be explained by increased shear generation in the extruder. At UHF inlet, sponge temperature becomes almost insensitive to extruded mass temperature when IR preheating is on. At UHF exit, sponge temperature increases with heat history of the profile before the UHF. Load deflection can be controlled by modifying the operating production parameters. At low production rates, load deflection can indeed be lowered by increasing UHF inlet and exit temperatures. However, at higher line speeds, low load deflections cannot be obtained by modifying operating conditions only, but require modifications of the curing/blowing package.

To summarize, load deflection consistency requires a tight control of all temperatures along the production line and specially at UHF inlet and exit. IR preheating and die temperature increase are shown to be valuable means to increase these temperatures whereas UHF power and line speed have to be kept constant to avoid solid component overheating and to have sufficient UHF residence time. With these limitations, it was not possible to raise sponge component temperature in the UHF above 130 [degree] C. It is, therefore, essential to assess whether sponge dielectric heating can be improved through compounding.

Compounding study to reduce temperature differences

between sponge and solid

Sponge and solid compounds contain different amounts of carbon black and hence exhibit dissimilar UHF receptivities. The objective of this section is to describe some basic compounding studies aimed at reducing the difference in UHF energy pick up between sponge and solid components. The choice of the carbon black type and the level in the compound are of course key elements for dielectric heating in the UHF unit. It is well known that particle size, (the smaller@the better) and structure (the higher the better) increase compounds receptivity (ref. 3). Conventional highly reinforcing blacks improve microwave heating, however, their processing drawbacks have prevented their use in EPDM extrusion compounds.

The addition of polar ingredients (ref. 3) such as polar polymers, zinc oxide, factice, diethylene glycol and triethanol amine can also improve compound receptivity. For example, chloroprene has been added in sponge compounds for the manufacturing of coextruded body seals; unfortunately, the sponge component extrusion consistency was affected because of non-uniforrn chloroprene dispersion in the compound.

Experimental carbon blacks (CRX 1416B from Cabot and EB 501 from Degussa) have recently been developed to improve UHF productivity. It was, therefore, appropriate to assess their effectiveness. Addition of non-crystalline polar polymers such as polypropylene oxide (GPO) or epichlorohydrin (ECO) to increase microwave receptivity was also studied. On the other side, the use of less receptive carbon blacks such as Durex-O or of white fillers was tested to reduce solid component temperature build up.

For the preliminary screening phase, the dielecmeter (ref. 4) was used to predict the microwave receptivity of our modified compounds. The principle of this equipment consists of exposing an clastomer compound to a microwave field and of measuring the effects of the disturbance of this field by the material. The dielecmeter measures, calculates and displays the temperature and the dielectric properties of the tested material as a function of the test time. Figure 5 shows the temperature variation of a solid and a sponge compound.

In all cases, the solid temperature linearly increases with time, whereas the sponge temperature exhibits several slope changes. The first one is attributed to the exothermic blowing agent decomposition, while the second one is associated to a loss of physical contact between the temperature probe and the material, as a consequence of blow initiation.

We selected the time for temperature to increase from 50 [degree] C to 150 [degree] C (which we call "dielec time") as an indicator of microwave receptivity. Figure 6 shows this time to decrease when polar polymers were added to our reference sponge compound; at 10 phr level, GPO and ECO appear to behave similarly and show a substantial reduction in dielec time. At 20 phr level, UHF receptivity appears to level off. However, when a high receptive UHF black is substituted for FEF, the effects of these polar polymers become less pronounced.

For the solid component, several combinations of less reactive carbon black and white fillers were studied. The longer dielec time was observed with a (40/60/40 phr) blend Durex O/FEF/sillitin. Some of the compounds tested in the dielecmeter were coextruded under constant conditions on our UHF line to confirm predicted results. As shown in figure 7, the measured UHF exit temperature correlates well with the dielec time on a log log scale (UHF exit temperature decreasing with increasing dielec time). Under experimental conditions such as those used in phase 2 experiments (2.4 kW, 3 m/min) temperatures at UHF exit reached 160 [degree] C for the sponge and 180 [degree] C for the solid. This confirms the validity of the dielecmeter for predicting microwave receptivity. In addition, carefull selection of compounding ingredients is confirmed to be a valid tool to minimize solid and sponge temperature differences at UHF exit, thus allowing higher operational latitude.

     Table 1 - polymer characteristics
                         Tailored grades   Traditional grades
                           A          B      C      D     E
ML (1+4),125 [degree] C   73          65     54    61   50/50
C2 (wt%)                  55          54     51    51   Blend
ENB (wt%)                          Very high: 9-10     of C&D
Mw/Mn (1)                3.2         3.5    3.4   2.7
Mz/Mw (2)                  8           7      5    11
Branching                 ++         +++           ++
(1) GPC-DRI
(2) GPC-LALLS

Suitability of MW tailoring for the production of solid

components

Work to this point has indicated that soft sponge seals can be consistently produced providing a global approach including optimization of compounding and extrusion line set ups is followed. Limited processability of the solid component can, however, reduce overall production rates. It is, therefore, appropriate to assess the potential of MWD tailoring for combining high extrusion speeds and vulcanization rates and tight dimensional tolerances. Previous work has indicated that a controlled amount of ultra high molecular weight chains, although increasing polymer Mooney viscosity, can result into a lower compound viscosity and consistent extrusion (ref. 5).

Bench scale extrusions and vulcanizations were conducted on a Haake rheocord equipped with a 20 L/D, 1 inch extruder screw followed by a lab size hot air tunnel (3 x 1m long) for grade structure definition on small scale synthesized polymer samples. Three extrusion speeds were selected. For each condition, the hot air conveyor belt speed was kept constant so that screw speed had to be adjusted for each polymer. During extrusion, screw speed and measured torque were recorded. From these data, specific energy (energy input from the extruder by gram of extruded compound) could be computed (ref. 6).

Two tailored EPDM structures (differing only in molecular weight and branching level) were compared to reference commercial grades (including a narrow and a broader polymer as well as a blend of them). Basic polymer characteristics are described in table 1 (all polymers have a comparable diene content).

The main indication of this lab screening test is that the two experimental polymers show higher cure rates at potential processability benefits. Specifically, at high rates, their extrusion requires lower specific energy while higher consistency as measured on swell) becomes possible. At low hot air tunnel residence times (1.5 min.), lower compression set values (cure rate/state indicator when measured on extruded profile) are obtained with these polymers (figure 8).

Molecular weight tailoring technology appears, therefore, to be a valid tool for the production of EPDM grades specifically suitable for the grip portion of body seals. Because of their improved extrudability and increased vulcanization efficiency, tailored MW distribution grades show potential for setting new standards of consistency while increasing productivity (lower energy consumption and higher production rates).

Conclusion

This study has confirmed that optimum results can be obtained when systematic effort is put on all the elements which contribute to the production of the final components. Specifically, in the case of body sealings we have shown that measurable improvements can be obtained through optimization of the polymer, the compound and the operating conditions. The molecular weight tailoring technology confirms to be an invaluable help to meet the productivity and consistency challenge of the 90's.

     Appendices
  The ethylene propylene rubber compounds utilized for the
two first sections of the work have the following generalized
formula:
Phase 1: sponge alone    Phase 2: sponge + solid
Sponge                   Solid
Vistalon 8600   100      Vistalon 9500    100
Durex O          65      SRF N-762        100
Omya BSH         30      MT 990            40
Flexon 815       60      Escomer H201       8
ZnO               8      Flexon 876        15
Stearic acid      1      ZnO                8
PEG               1      Stearic acid       2
CaO               2      PEG 3350           3
Sulfur          1.5      Struktol WB 16     3
MBTS              1
TMTDS           0.5      Rhenosorb C        4
DPTTS             1      Sulfur             1
ZDEDC             1      MBT              0.8
TDEDC           0.3      TMTDS            0.4
ZDBDC             1      DPTTS              1
PVI             0.1      ZDBDC            0.8
Genitron AC-4     2
Celogen OT        2
After the compound optimization study, best recommended
formulations fro production of a coextruded profile
are described hereafter.
Recommended formulations after phase 3
Sponge                     Solid
Vistalon 8600      100     Vistalon 9500    100
Experimental black  65     Durex-O           40
MT                  30     FEF N-550         60
Flexon 876          70     Sillitin Z        40
Zinc oxide           8     Flexon 815        45
Stearic acid       0.5     Zinc oxide         8
PEG 3350             2     Stearic acid       2
PVI                0.1     PEG 3350           2
Rhenosorb C          2     Silane A172        1
Sulfur             1.5     Struktol WB16      3
MBTS                 1     Rhenosorb C        4
TMTDS 80%          0.6     Sulfur             1
ZDEDC                1     MBT 80%            1
DPTTS                1     MBTS             0.5
ZDBDC                1     TMTDS 80%        0.8
TDEDC              0.3     Sulfasan R         1
Genitron AC4         2     ZDBDC            0.5
Celogen OT           2

All EPDM compounds were prepared utilizing a two-stage procedure. The masterbatch of polymer,filters, plasticizer, zinc oxide and stearic acid was mixed in a 19.4.1. chamber volume mixer using an upside-down mixing cycle. In the second step, curing agents and other chemicals associated with vulcanization were added to the masterbatch in the same mixer. in the compounding study where smaller amounts could be used, the same procedure was applied in a 1.1 chamber volume mixer. For the last section, a 4.5.1. Intermesting mixer was employed. Profile temperature along the line were measured with a contact thermometer pyroterm 9300 whereas they were measured at UHF exit with an IR thermometer pyroterm IR 55.

Load deflections were measured in accordance with ASTM D 3585-74. Sponge surface was characterized with a Surfcom,surface roughness analyzer (ref.7). For all the coextrusion work the section of the coextruded profile is described in figure 9.

References [1.] H. Focht, International Polymer Science and Technology, 7, T/6-5 (1980). [2.] G. Stella and N.P. Cleremisinoff "Design technology advancements for EPDM sponge seals." presented at he international Rubber Conference, June 1990. [3.] a) C.W. Otterstedt and R.S. Auda "Factors influencing the temperature rise of EPDM in continuous vulcanization media," presented at he 115th Rubber Division meeting. b) V.L. Lite, Rubber World, 26 (1980). c) N. Probst, J. Iker, Kautschuk Gummi Kunststoffe, 37 (1984). [4.] B. Le Rossignol "Nouvelle approche de la vulcanisation UHF au moyen du S.T.E.M," presented at the AFICEP meeting, December 1986. [5.] G. Stella and N.P. Cheremisinoff "Designing EPDM for production efficiency, " presented at the International Rubber Conference, October 1988. [6.] N.P. Cheremisinoff, Polymer Mixing and Extrusion Technology, Dekker, New York, 1987. [7.] Surfcom Surface Analyzer user's manual, Brown & Sharpe Manufacturing Co., 1980.