A new EPDM sponge grade for high performance and consistency.

The commercial production of EPDM is more than 30 years old. In those 30+ years, EPDM found its use in an ever-increasing number of applications. Now, it is the most widely used technical elastomer in non-tire applications, e.g., automotive sealing systems and hoses, window gaskets, roofing sheets, and wire and cable. Even today, a number of EPDM producers still invest serious efforts to grow the market for EPDM even further. Recent successes are potable and wastewater seals, engine mounts and window wipers. Also, one should not forget the increasing use of TPVs that contain significant amounts of EPDM. At present, the global demand for EPDM exceeds 900 metric tons, and is expected to grow to more than 1,000 metric tons in 2005. One could ask the question whether this business is (fully) matured, and whether the product is actually a commodity. Although most professionals active in the field do not regard EPDM as an exotic species anymore, we still see many opportunities for improving the production process and the product itself. Especially because of the aforementioned new EPDM applications that are being developed, it is justified to claim that the market is constantly rejuvenating. As a consequence, our research and development efforts are well above those of commodities like polyethylene.

One of the driving forces for further product improvements has certainly been the automotive industry and their suppliers. At the moment, more than 50% of the total EPDM demand finds its use in automotive applications, with sealing systems being the most important sector. The global body seal market based on EPDM is approximately 180 mt, of which 35-40 mt is used for the production of sponge seals. Our customers active in this segment are imposing more and more stringent demands on consistency, performance, aesthetics and costs, all at the same time. This differs markedly from other application areas which do have a (more) differentiating priority setting of demands. So, to be successful in the automotive sealing market, an EPDM supplier needs to excel in many aspects simultaneously.

For many years, EPDM suppliers and customers struggled with the balance between processing and physical properties. At first, there was a focus on molecular weight and composition, e.g., the higher the molecular weight, the better the physical properties. By tuning the ethylene content and unsaturation, behavior at various temperatures and vulcanization characteristics could be manipulated. Presently, other polymer parameters such as molecular weight distribution (mwd) and, more recently, long chain branching play very crucial roles. In contrast to molecular weight (or the Mooney viscosity) and composition, mwd and long chain branching are far more difficult to control. To a large extent, the latter two parameters are the outcome of the selected production technology and polymerization catalyst used in the EPDM plant. However, both mwd and long chain branching have major impacts on a wide range of processing and physical parameters. In general, a wide mwd assures the converter of good processing, as relatively broad EPDM polymers show good flowability. On the other hand, an EPDM polymer with a narrow mwd offers good physical properties, usually at a cost of poor processing behavior (especially during extrusion). Therefore, it proved necessary to re-visit both process technology and catalyst technology to support the required product development. The main question to be answered was: "Is it possible to merge the advantages of polymers showing good processing behavior and good cure/physical characteristics?" Fortunately, the answer is yes. Extensive research efforts have resulted in an advanced catalyst system able to produce polymers having a very narrow mwd. In combination with a specific process technology, a desired level of long chain branching can be "added" to generate high performance EPDMs, still having a narrow mwd. The new catalyst system effectively suppresses cationic reactions that are responsible for significant broadening of the mwd when applying conventional Ziegler Natta catalysts. In addition, the suppression of these cationic reactions strongly contributes to a higher level of product consistency. In essence, the new technology disconnects long chain branching and mwd and is referred to as controlled long chain branching (CLCB) (ref. 2). The relation between mwd and branching for CLCB EPDMs and conventional EPDMs is shown in figure 1. Now, CLCB polymers exhibit good processing behavior and good curing characteristics and physical properties. Based on this technology, a new EPDM grade, K7341A, has been developed, specifically targeting automotive sponge applications. This grade is the third in a series of CLCB polymers produced by DSM Elastomers (ref. 3), satisfying requirements for a wide range of automotive applications.

[FIGURE 1 OMITTED]

Product development strategy

The development strategy selected for the new sponge seal grade, as partly discussed in this article, is outlined in figure 2. The customer participation (from around the globe) proved to be very fruitful to co-guide the development. The customer input was essential:

* to determine key success factors for the new polymer; and

* to evaluate the performance of various versions of the development grade.

[FIGURE 2 OMITTED]

The generic focal points for automotive body seals are well known:

* Sealing (water, noise and airtight);

* comfort (easy door closing and non-freezing seal);

* aesthetics (smooth and shiny [North America] vs. dull [South America, Europe and Asia]); and

* economics (reduction of total system costs).

It is important, however, to identify those items that need further improvement. Based on extensive market research, the following items were considered most important. The order does not reflect its importance:

* Higher EPDM consistency a) to reduce the (very) high scrap rates and b) to develop seals with better surface finish;

* improved mixing behavior by, e.g., elimination of carbon black scorch;

* reduction of compression deflection to minimize door closing efforts; and

* cost down efforts.

Experimental

To support product development, the level of long chain branching is determined by using a dynamic mechanical spectrometer and expressed by the so-called delta-delta ([DELTA][delta]) value, which is used as a measure of the non-Newtonian viscoelastic behavior of EPDM polymers (ref. 1). The [DELTA][delta] parameter is defined as the difference between the phase angle [delta] at [10.sup.-1] rad/s and the phase angle [delta] at [10.sup.2] rad/s, as derived from frequency sweep plots obtained in dynamic mechanical spectrometry. The [DELTA][delta] value decreases with increasing degree of branching. The presence of branched molecules will decrease the [delta] value specifically at low frequencies, due to (extensive) polymer entanglement. The [delta] value at high frequencies is governed by the average molecular weight of the polymer.

The data shown in this article are based on large-scale compound evaluations. All mixing took place in a 70 liter internal mixer, resulting in compound batches of about 60 kg. Such a quantity proved to be sufficient for extrusion trials, and additional compound evaluations. For the extrusion trials a 90 mm, 16D extruder was employed with a UHF-HA vulcanization line.

The aim of the extrusion trials was to produce sponge profiles having similar dimensions and densities, to ensure meaningful comparative evaluations. To increase sensitivity for, e.g., compression set measurements and collapse resistance, a special D-shaped die was prepared, producing a bulb with a wall thickness of approximately 2 mm. The belt speed of the vulcanization line was kept constant for all experiments; 10 meters/minute. It proved possible to keep dimensions under control by manipulating a) screw speed of the extruder and/or b) pressure of the support air. Dimensions were checked on the spot by using the profile projector P500. The UHF part consists of 2 x 6 kW microwave units. The hot air line following the UHF unit was operated at a temperature of 270 [degrees] C for all extrusion trials. The following evaluation items received special attention:

Processing

* Mixing (sensitivity to carbon black scorch);

* extrusion (collapse resistance of profile leaving the extruder and surface finish).

Vulcanization

* Profile temperature leaving the UHF line ([degrees] C) (UHF out);

* Mooney scorch at 125 [degrees] C (ISO 667) (min.) (T2);

* rheometer, ODR (first compound study, table 2) and MDR (second compound study, table 3) at 180 [degrees] C (ISO 6502) (min.) (T90).

NB cure rate is defined as (MH-ML)/(t90-ts2) (CR).

Physical properties

* Density (g/c[m.sup.3]) (D);

* tensile strength (MPa) (TS), and elongation at break (%) (EB) using dumbbell 2 (ISO 37);

* tensile strength (MPa) (TSag) and elongation at break (%) (EBag) after aging (42 days x 100 [degrees] C [DIN 53508 6.2]);

* compression deflection at 10, 20, 30 and 50% (N, length of test piece: 10 cm); (CD10%)(CD20%)(CD30%) (CD50%).

* Compression set on sponge profile (CS).

Results and discussion

Various catalyst systems and process technologies have been screened for this product development to select the optimum combination. To assure a high level of sponge seal consistency, seen as the most important key success factor by many customers, the EPDM production process needs to be highly robust. Certain disturbances in the polymerization process (e.g., fluctuations in production rate) should not affect the polymer structure significantly. It should be noted that the production of EPDMs combining a high ENB content with a high mw, as usually required for the production of automotive sponge seals, is relatively sensitive towards process and subsequent product fouling. Again, the CLCB technology showed the best overall performance, resulting in the most consistent production process and product properties. Compared to traditional Ziegler Natta catalysts (VO[Cl.sub.3] - SEAC (or e.g., DEAC, resulting in ethylene sequencing and wide mwd), the CLCB catalyst is much more homogeneous, resulting in very narrow mwds, typically in the range of 2-3. Also, the suppression of cationic reactions contributes to the higher product consistency.

To validate the CLCB structure and composition of the new sponge polymer, it was decided to compare K7341A with two EPDMs exhibiting significantly different polymer characteristics. For details see table 1. Polymer A and B represent current state-of-the-art sponge polymer grades. The most important differences between the three polymers are related to branching, molecular weight and molecular weight distribution. The presence (or need) for extender oil is related to the high mw of K7341A and polymer B. Without oil, such EPDMs would be difficult to process and finish in the EPDM plant. In addition, it would be extremely difficult (and for that reason uneconomical) for convertors to mix such EPDMs.

The EPDM polymers have been compared in several conventional and nitrosamine (NA) safe systems. Nitrosamine safe compounds are mainly used in Germany, but in several other European countries the commercial use of such compounds has started. In table 2, the compound formulations are shown; recipes 1 and 2 have a conventional curing system (confidential information), whereas recipes 3 and 4 are NA safe. The discussion with respect to the first compound evaluation (based on compounds 1-4, as shown in table 2) is split up in three parts; processing (mixing and extrusion), vulcanization (rheometer data) and cured properties.

Processing

The mixing procedure (70 L internal mixer and two roll mill) was kept constant for each compound; two minutes of polymer crumbling was followed by four minutes of actual mixing, and subsequent dumping of the compound on the mill. The power consumption during polymer crumbling was highest with both oil extended EPDMs; e.g., in recipe 4, polymer A consumed only a fraction of the energy consumed by K7341A and polymer B. Taking the whole mixing cycle into account, polymer B consumed the most energy, which resulted in the highest dump temperatures, closely followed by the K7341A compounds. In general, the dump temperature obtained with the compounds based on polymer A was significantly lower. Most likely, the dump temperature is influenced by the molecular weight of the polymers, with polymer A having the lowest molecular weight. It is important to mention that the three polymers showed rather different sensitivities towards carbon black scorch. This study showed once again the relation between ENB content and the occurrence of black scorch. Under similar compound and mixing conditions, an increase in the ENB content of the EPDM results in a higher sensitivity towards carbon black scorch. This (reversible) phenomenon (ref. 4) usually occurs during mixing and results in a drastic increase in power consumption towards the end of the mixing cycle (figure 3). During extrusion of profiles, a related problem can occur, resulting in surface imperfections, thus producing high and expensive scrap rates. Similar to the black scorch experienced during mixing, these surface imperfections can sometimes be suppressed by adding sulfur during the mixing stage. In this respect, K7341A with its relatively low ENB content of 7.5 wt% showed the most robust mixing and extrusion results. The relatively high branching level of K7341A easily satisfies the requirements on collapse resistance of the non-vulcanized profile. This high level of branching hardly affects the mwd and, consequently, physical properties. In addition, the level of CLCB improves the surface finish of extruded profiles.

[FIGURE 3 OMITTED]

Vulcanization

The structure and composition of the three tested EPDMs show significant differences. Consequently, the vulcanization behavior is rather different for each polymer. Let us first focus on the conventional curing systems (recipes 1 and 2). Differences in scorch time (at 125 [degrees] C) are relatively small, with polymer A having the shortest scorch time. Also, the differences in network density (MH-ML) are relatively small. When comparing the test results, it should be kept in mind that K7341A compounds are loaded higher (+20 phr oil) and K7341A contains significantly less ENB compared to polymers A and B. One would expect that the combination of a very high ENB content and the very high mw (and Mz) of polymer B should result in the fastest cure rate and highest network density. This, however, is not the case. Probably the wider mwd "spoils" the potential advantages of a high ENB content and a high mw with respect to the curing behavior. The rheometer curves obtained with recipe 1 are shown in figure 4.

[FIGURE 4 OMITTED]

When applying the selected nitrosamine safe recipe, the vulcanization behavior of polymer B appears significantly different from either polymer A or K7341A. Clearly, cure rates are faster and network densities are higher with polymer B. Note that recipes 1 and 2 are much more optimized (i.e., very efficient), when compared to the cure packages used in recipes 3 and 4; in other words, putting less "weight" on the ENB content of the polymer. Differences in vulcanization behavior between polymer A and K7341A appear not to be significant for recipes 3 and 4. The rheometer curves obtained with recipe 3 are shown in figure 5.

[FIGURE 5 OMITTED]

Cured properties

In order to compare the cured properties of sponge polymers, it is important to assure uniform densities and dimensions of profiles. Fortunately, differences in sponge densities are very small due to, e.g., minor differences in the compound Mooney viscosities (table 2). Also, differences in the shape of profiles proved to be minimal.

With respect to the cured properties, attention was focused on tensile properties, compression deflection and compression set. Again, please note that the recipes based on K7341A contain an additional 20 phr of oil. For all four compound recipes, K7341A proved to have the best tensile properties; at similar tensile strengths, higher elongation at break could be achieved when compared to polymers A and B (figure 6). The same observation applies after hot air aging (42 days x 100 [degrees] C). A relatively high level of elasticity is often wanted to avoid rupture of the sponge profile (e.g., by abrasion or when opening a car door which is frozen to the seal). Another advantage of higher elasticity is the reduction of "wild blowing," which can be experienced with highly loaded compounds.

[FIGURE 6 OMITTED]

It is well known that the compression deflection is a significant parameter to control door closing comfort. High levels of compression deflection are unwanted, as it impedes door closing comfort. With all tested recipes, the use of K7341A resulted in a reduction of compression deflection (see table 2 and figure 7).

[FIGURE 7 OMITTED]

The compression set results indicated no significant differences between the EPDMs when applying the conventional curing package. Using the (less efficient) NA safe curing package (as shown in table 2) resulted in relatively low compression sets with polymer B, as can be expected from the rheometer data indicating its higher network density. K7341A and polymer A showed similar compression sets, both higher than observed with polymer B.

As indicated previously, the majority of EPDM converters has most experience with conventional vulcanization systems. The importance of nitrosamine safe recipes, however, is growing. Therefore, producers of vulcanization ingredients spend more and more efforts to develop efficient NA safe cure packages. From the results obtained with recipes 1-4, it was concluded that the NA safe system of recipes 3 and 4 was not very efficient. In table 3, the effect is shown of a higher performing, i.e., more efficient, NA safe curing package on the vulcanization behavior and physical properties of the same EPDMs as used in the previous study (table 2).

It can be concluded from the data presented in table 3 that by applying an efficient NA safe vulcanization system, the need for high ENB contents has disappeared. The MDR data (table 3) and curves, as presented in figure 8, do not show significant differences between the cure characteristics (e.g., cure rate [CR]) of K7341A and polymers A and B. Please note again that the compound formulation with K7341A contains 20 phr of extra oil to synchronize compound Mooney viscosities. Due to the similar crosslink densities (as concluded from the rheometer curves in figure 8), differences in compression set values are minimal. In fact, K7341A gives the best performance despite the higher loading level of the compound (figure 9). The high molecular weight of K7341A (again) ensures the higher elongation at break.

[FIGURES 8-9 OMITTED]

Conclusions

The robustness of the CLCB production technology offers a product with a high level of consistency, which is required for the production of automotive sponge seals.

The branching level of CLCB polymers can be tuned to optimize processing behavior without disturbing physical properties.

The CLCB structure of EPDMs eliminates the need for high levels of unsaturation. Reduction of the ENB content in EPDM reduces the sensitivity towards carbon black scorch.

Keltan 7341A embodies high consistency, high performance, excellent aesthetics, thus creating high market value.

Table 1 - structure and composition of EDPM
polymers used in this study (typical data)

Polymer                    K7341A   Polymer A   Polymer B
Production technology      CLCB     Convent.    Convent.
ML (1+4) 125 [degrees] C               90          70
ML (1+8) 150 [degrees] C     53
C2 wt%                       53        45          48
ENB wt%                     7.5         9          11
Oil (phr)                    20        --          20
Mw                          360        280         330
Mw/Mn                       2.9        3.6          4
Delta-d                      16        24          12
Table 2 -  compound formulations and (vulcanized) properties
obtained with K7341A, polymer A and polymer B
(abbreviations are explained in the experimental section)

Recipe                           #1               #2

EPDM grade *                 K7341A/A/B       K7341A/A/B
EPDM *                      120/100/120      120/100/120
ZnO                              8          [right arrow]
St. acid                         2          [right arrow]
PEG                              1          [right arrow]
CaO-80                          3.5         [right arrow]
CB N-550                         90               90
Whiting                          20               55
Sunpar 2280 *                 85/85/65         90/90/70
Flexon 876 *
ADC                             2.8         [right arrow]
OBSH                            1.60        [right arrow]
Curing system               Confidential    [right arrow]
                             (6.11 phr)
MBT-80
ZDBP-50
ZBEC-70
DTDC-80
S-80
Total phr *                 340/320/320      380/360/360

T2 (min.)                   2.2/1.9/2.1      2.6/2.4/2.5
T90 (min.)(ODR)             3.8/3.5/4.1      3.6/3.2/4.2
CR (ODR)                    1.5/1.8/1.4      1.6/2.0/1.4
CD (MDR)                   0.53/0.62/0.64   0.52/0.57/0.60
D (g/[cm.sup.3])           0.51/0.53/0.50   0.54/0.55/0.53
TS (MPa)                    3.7/3.9/3.4      2.9/2.8/2.6
EB (%)                      286/267/248      270/238/220
TSag (MPa)                  3.7/4.3/3.5      2.5/2.6/2.2
EBag (%)                    180/153/142      171/141/132
CD10% (N)                   2.7/4.9/3.1      2.9/3.7/2.6
CD20% (N)                   5.0/8.4/6.0      5.0/7.1/4.8
CD30% (N)                    7.6/11/8.5      7.1/9.4/7.8
CD50% (N)                     14/24/17         14/16/15
CS; 24h x 70 [degrees] C      13/14/13         12/14/16

Recipe                           #3               #4

EPDM grade *                 K7341A/A/B       K7341A/A/B
EPDM *                      120/100/120      120/100/120
ZnO                              5          [right arrow]
St. acid                         1          [right arrow]
PEG                              2          [right arrow]
CaO-80                          2.14        [right arrow]
CB N-550                         65               70
Whiting                                           35
Sunpar 2280 *
Flexon 876 *                  65/65/45         70/70/50
ADC
OBSH                            3.00        [right arrow]
Curing system

MBT-80                           2          [right arrow]
ZDBP-50                          4          [right arrow]
ZBEC-70                          2          [right arrow]
DTDC-80                          1          [right arrow]
S-80                            1.9         [right arrow]
Total phr *                 275/255/255      320/300/300

T2 (min.)                   2.6/2.6/3.2      2.6/2.4/2.6
T90 (min.)(ODR)             6.8/7.8/5.8      6.7/7.0/4.4
CR (ODR)                    0.6/0.5/1.1      0.7/0.7/1.3
CD (MDR)                   0.18/0.15/0.34   0.19/0.17/0.39
D (g/[cm.sup.3])           0.48/0.46/0.50   0.50/0.49/0.49
TS (MPa)                    3.1/2.7/2.8      3.0/2.6/2.6
EB (%)                      305/307/244      372/318/263
TSag (MPa)                  2.6/2.7/2.3      3.4/2.9/2.7
EBag (%)                    139/133/118      137/110/107
CD10% (N)                   2.7/3.3/2.5      2.4/3.2/3.5
CD20% (N)                   4.7/5.6/4.8      4.3/5.7/6.2
CD30% (N)                   6.6/7.4/7.5      6.3/7.6/8.6
CD50% (N)                     14/16/21         13/17/18
CS; 24h x 70 [degrees] C       9/7/8           16/13/10

* Important information: In order to keep compound Mooney
viscosities as similar as possible (for each different
compound), the total loading of formulations based on K7341A
is always 20 phr (oil) higher than the compounds based on
polymers A and B. This is the result of the characteristic
combination of high molecular weight and narrow MWD of
K7341A.
Table 3 - efficient NA safe compound formulations
and (vulcanized) properties obtained with
K7341A, polymer A and polymer B (abbreviations
are explained in the experimental section)

EPDM grade                   K7341A     Polymer A       Polymer B

EPDM phr                      120          100             120
ZnO                            8      [right arrow]   [right arrow]
St. acid                       2      [right arrow]   [right arrow]
PEG                            1      [right arrow]   [right arrow]
CaO-80                        3.5     [right arrow]   [right arrow]
CB N-550                       90     [right arrow]   [right arrow]
Whiting                        20     [right arrow]   [right arrow]
Par. oil                     90 (!)        90              70
OBSH                           3      [right arrow]   [right arrow]
Geniplex A-80                 1.5     [right arrow]   [right arrow]
ZBEC-70                       1.5     [right arrow]   [right arrow]
ZAT-70                        1.3     [right arrow]   [right arrow]
Rhenocure AP7                 4.3     [right arrow]   [right arrow]
Rhenocure SDT                  2      [right arrow]   [right arrow]
S-80                          1.9     [right arrow]   [right arrow]
Total phr                     350          330             330

T2 (min)                      2.05        1.78            1.78
T90 (min)(MDR)                2.37        2.44            2.24
MH-ML (MDR)                   0.95        1.01            0.99
CR (MDR)                      0.45        0.47            0.51
D (g/[cm.sup.3])              0.45        0.46            0.44
TS ag (MPa)                   2.6          2.7             2.5
EB (%)                        255          214             205
TS ag (MPa)                   2.4          2.8             2.7
EB ag (%)                     252          236             223
CD10% (N)                     3.0          2.7             2.9
CD20% (N)                     5.6          5.3             5.7
CD30% (N)                     8.0          8.0             8.3
CD50% (N)                      16          16              18
CS1; 24h x 70 [degrees] C      8           10              12
CS2; 200h x 70 [degrees] C     30          34              32
CS3; 72h x 85 [degrees] C      27          36              32
CS4; 22h x 90 [degrees] C      33          37              36
CS5; 70h x 100 [degrees] C     42          51              46

References

(1.) H.C. Booij, Kautsch. Gummi Kunstst., 44, 128 (1991).

(2.) H.J.H. Beelen, Kautsch. Gummi Kunstst., 52, 406 (1999).

(3.) M.J. Dees, Presentation at the 156th ACS Rubber Division Meeting in Orlando, paper #104, (1999).

(4.) X. Zhang and R. Whitehouse, Presentation at the 152nd ACS Rubber Division Meeting in Cleveland (1997).