Controlling the iridescent bloom in EPDM body sealing applications.

Ethylene propylene elastomers are used extensively in the production of extruded or molded articles, in particular for the automotive industry. Examples of application are door sealing, sponge and dense, glass run channel and weatherstrips for static windows. Analysis of a recent vehicle shows that more than 60% of the rubber used in a car is based on EPDM and about 50% of this rubber is EPDM body seals. This illustrates the unique position of EPDM in the automotive industry due to its wide processing latitude and its excellent resistance to ozone and weather.

Being a major component in a car, EPDM seals are part of the overall quality that any customer perceives. They pay more attention today to the aspect of the sealing system, in particular to the color of the outdoor seals. It has been observed that some EPDM profiles may change color once mounted on the car and exposed to the direct sunlight. Their surface may turn white, often yellowish or bluish, and this is not acceptable anymore. Therefore, rubber seal suppliers spend a lot of energy and money to control the rubber aspect. For example, different types of coating are sprayed onto the profile to provide a better and consistent aesthetic appearance to a glass run channel or a belt line seal.

This article will address the different issues relative to the color evolution of the sealing profile, understanding the surface transformation after sunlight exposure. It will lead to recommendations for formulating and processing the rubber in order to better control the iridescent bloom.

Experimental part

Bloom test description

To carry out the systematic evaluation of the bloom and make further analytical tests, we used EPDM rubber pads, cured under pressure between two films of Mylar for eight minutes at 180 [degrees]C. One sample was exposed outdoors to the sunlight for one week in Texas. A similar pad was kept inside without direct exposure of any sort, natural or artificial light. We tested different methods to evaluate the color change, like gloss measurement with a glossmeter, or the color intensity breakdown according to the L, a, b parameters. Actually the readings on that equipment were not discriminative enough to show significant changes, so the evaluation of the bloom was quoted visually.

The samples were prepared with a similar composition for their rubber hydrocarbon (RHC) and fillers content. In the first part, we studied the influence of different accelerator combinations. In the next parts, we took the "best cure system to study the influence of the other compounding parameters (table 1).

Table 1 - formulation used in the study

EPDM Vistalon 8609                       100 phr
Carbon black FEF N550                    130 phr
Paraffinic oil type Flexon 815            85 phr
Calcium oxide (80%)                        7 phr
Stearic acid                             1.5 phr
Zinc oxide                                 5 phr
Sulfur                                   1.5 phr
MBT                                      1.5 phr
TMTDS                                    0.5 phr
Other accelerators                 up to 2.5 phr

Surface analysis techniques used in this study

On some of the EPDM rubber parts exposed as described previously, we proceeded with a systematic surface analysis. The objective was to highlight differences between a "good" sample and one turning to a colored surface either after sun exposure or after continuous vulcanization in a hot-air oven.

Within the numerous methods of surface analysis offered to the chemist, we selected the following (table 2):

[TABULAR DATA OMITTED]

* Surface "washing" with a solvent like an alcohol or a ketone. The technique allows recovery of all types of chemical residues present at the surface of the cured rubber. Then we try to identify them with an IR spectrum.

* The second test method we used is the surface thermal desorption. This test is carried out at high temperature (150 [degrees]C) for 10 minutes.

The emitted species are analyzed by gas chromatography coupled to a mass spectrophotometer CG-MS). It identifies the chemicals present at the surface of the rubber at the time of the extraction, the low boiling point coming first.

* By the third method, we wanted to look at a thin surface layer, within the micrometer thickness. It is in the range of the visible light wavelength and covers the spectrum from red (0.8 [mu]m) to blue (0.4 [mu]m). These colors are typical of iridescent bloom. NMR spectroscopy and x-ray photoelectron spectroscopy (XPS) have a too-large analysis depth and cannot be used for the purpose of this study. The time-of-flight secondary-ion mass-spectrometry (TOF-SIMS) provides a selective tool for characterizing polymer surface chemistry like rubber specimens. In the SIMS process, a primary Ga+ ion beam is directed at [10.sup.-9] Torr onto the microtomed sample surface, sputtering both elemental and molecular ions. These secondary ions, characteristic of the analyzed surface, are then extracted under 3000V into a mass spectrometer for analysis and detection. This is a form of static SIMS (SSIMS), where the dose of primary ions is kept small to prevent specimen damage and focused on molecular ions analysis. It has the capability of distinguishing various functional groups with quick response time (10 minutes) and with high lateral resolution (0.3 [mu]).

Results and discussion

Evaluation of a sample giving iridescent bloom

As a result of the first series of tests carried out as previously described, the non-exposed sample kept its original color. It showed no trace of any particular blooming, neither white, generally due to an excess of nonreacted accelerators migrating to the surface, nor colored iridescent bloom.

On the contrary, the sun-exposed part showed an important color change toward the yellow-bronze, typical of the iridescent bloom observed on outdoor profiles mounted on a car and exposed to the weather for several months.

The two samples were submitted for comparative surface analysis, with the objective to detect the chemical composition differences and then understand a possible mechanism for this iridescent bloom.

Results of the analytical tests

Previous studies carried out by our analytical laboratory using the surface washing technique showed that the exposed surface was more rich in curative residues of the vulcanization process containing sulfur, nitrogen and oxygen. A possible explanation is that those chemicals, supported by an oily carrier or low molecular weight fractions of the plasticizer present in the compound, slowly migrate to the surface of the product and react with the fight radiation to produce a color change.

We used the thermal desorption technique to get a comparative "picture" of the surface of those two samples. The results we obtained were unexpected, apparently contradicting the migration model of chemical species described in the previous paragraph.

Figure I shows the mass spectrograph of a surface not exposed to the sun versus an exposed surface.

Interpretation of the results of the thermal desorption

The non-exposed surface is very rich in low molecular weight fractions of polar molecules containing sulfur, nitrogen and oxygen in the form of carbonyl, or hydroxyl. This could be contained either in reacted curatives or at the surface of the carbon black particles. This non-exposed surface is black, without any kind of blooming.

In contrast, the exposed surface did not contain such light molecules, but its color definitely turned to yellow. Therefore, a possible explanation could be a chemical reaction like in the photo oxidation process.

The low molecular weight polar species were present at the surface of the rubber after curing. Some of the speculated structures are shown: Amine and thiazole are likely to come from the accelerators. As well, carbonyl, morpholine and thiocyanate should come from the selected accelerators like mercaptobenzothiazole (MBT), tetrabutylbenzyl sulfenamide (TBBS) and dithiomorpholine (DTDM). Those are simple molecules, easy to extract by desorption and identified as low boiling point species.

Under the action of the fight and the heat, there is an oxidation of the surface, rearrangement of those molecules and other degradation. Two phenomena may occur:

* These molecules crosslink or dimerize, have a higher molecular weight, and therefore cannot be extracted anymore by the analytical desorption technique.

* These polar species are revealed in the visible spectrum by an image that was already chemically present at the surface of the rubber at the time of the extrusion and the vulcanization. Keeping in mind this mechanism, it was interesting to study systematically the different components of the formulation and to observe their surface reactivity after daylight exposure.

Compounding ingredients study

Accelerator combinations

In the first part, we selected the accelerator system since it is likely the one leaving most of the polar chemicals at the surface of the sample. Since the rubber has to be cured properly, a minimum quantity of each has to be present in the compound. In figures 2-4, the relative activity of each curative regarding the cure properties is observed.

DTDM is the most active among the sulfur donors. ZDBC (zinc dibutyl dithiocarbamate) tends to level out beyond 0.5 phr.

DTDM is very active at low level when DPTT  (dipentamethylene tamethylene tetrasulfide) sees its efficiency growing with its concentration.

DTDM is the most scorch sensitive among the sulfur donor accelerators. All tend to level out beyond 1.5 phr, and the scorch time becomes longer. The carbamate, in contrast, continuously shortens the scorch time.

In table 3, we mapped the accelerator type and level according to the color change and white blooming.

[TABULAR DATA OMITTED]

Combining cure performance and color sensitivity, we selected the cure system described in table 4. It is interesting to see that the preferred accelerators are those with a low melting point, in the range of the temperature commonly reached by a rubber batch when dumped at the finalization stage of the mixing operation. We think that the contact with the rubber is maximized, and this helps to solubilize the accelerators in the rubber matrix. Also a high cure state decreases the amount of residues and generates a very tight crosslink density. The migration of residues to the surface will be therefore reduced.

[TABULAR DATA OMITTED]

Fillers and EPDM

Using a similar approach, we studied the other components of the formulation. We will not present all the details here, but will higlight the trend fine results for better formulating:

* Carbon blacks of small particle size, more reinforcing, tend to give iridescent bloom. It could be related to the presence of the oxygen group at the surface and also to a lower relative concentration of polymer surrounding the particles.

Plasticizers like paraffinic oil containing some naphthenic species are better than a pure white paraffinic oil. Surprisingly, a low-viscosity paraffinic oil was better than a high viscosity. We carried out surface desorption analysis on samples after sunlight exposure. The compound containing white oil became yellowish and did not contain any low molecular weight species. The sample with a light paraffinic oil stayed black and did not show polar species on the surface, but more of the higher molecular weight paraffin molecules were present. A possible explanation could be that the presence of those higher MW paraffins on the surface have played a role of protector against the light radiation.

* EPDMs with higher diene content are less prone to iridescent bloom. It can be explained by the creation of a tighter crosslink density and a higher level of accelerator consumption, reducing the migration of residues toward the surface of the sample.

* Narrow-molecular-weight EPDMS are not recommended because of the lack of low molecular-weight species: They limit the optimum dispersion of all the ingredients during the mixing. Slightly crystalline EPDM is not advisable (over 75% ethylene content) since the plasticizer absorption is limited and the excess tends to bleed to the surface along with accelerator residues.

Extrusion and continuous vulcanization

From those results, we developed two compounds: one with a low-volatility plasticizer viscosity of 100 SSU at 100[degrees]C) and the other with a low-viscosity plasticizer (40 SSU at 100[degrees]C), more volatile. They were extruded on a laboratory CV line composed of a 38 mm (1-1/2")-20D extruder followed by a 9 meter (30 ft.) hot-air tunnel. The hot-air tunnel was set at two different temperatures. A low temperature of 230[degrees]C (450[degrees]F) heated the profile to 160[degrees]C (320[degrees]F). The color of the surface did not change. On the contrary, a setting at 300[degrees]C (570[degrees]F) generated a surface temperature of 220[degrees]C (430[degrees]F),turning the surface to a yellow color.

We used the SSIMS analysis to qualify and quantify the type of chemical species present at the surface of the profiles. The chemical structures which appeared at the surface were typical of MBTS mercaptobenzothiazole disulfide) and, less significantly, of zinc combination. The presence of NOTS type can be explained by the decomposition of the TBBS, and the zinc is a residue of the ZDBC present in the compound. Table 5 summarizes the results.

[TABULAR DATA OMITTED]

The chemical analysis of the surface by SIMS is in agreement with the surface desorption technique: A yellowish surface has less chemical species than a black one. A too high curing temperature will favor the presence of chemical species like TBBS residues at the surface, therefore leading to color change by oxidation. This phenomenon is enhanced by the use of a low viscosity plasticizer, increasing the concentration of accelerator at the surface. But the cure temperature has to be high enough to ensure a full reaction of all the curativies. For example, a low oven temperature induced a high ZBDC concentration at the surface, suggesting that the profile was not fully cured.

Conclusion

In this article, I tried to contribute to the improvement of the quality of the EPDM body sealing by understanding the cause and the mechanism of the development of iridescent bloom. We used interesting and powerful surface analysis techniques to identify the change at the surface of EPDM profiles cured in a hot air oven and exposed to the daylight.

We showed that the number and the variety of the chemical ingredients generally used in an EPDM formulation and their sensitivity to oxidation by heat and radiation play an important role.

The chemical species present at the surface at the moment of the extrusion are then oxidized, turning the profile into a yellowish or bluish color. A careful selection of accelerators for their melting point and their solubility parameter will improve their miscibility and the efficiency of the vulcanization.

The filler selection also plays an important role either by their surface activity or by their ability to migrate at the elevated vulcanization temperature.

The EPDM characteristics are also key factors with, for example, a high diene version, giving a higher crosslink density to limit the diffusion of unreacted species.

Finally, in conjunction with the formulation parameters, the processing conditions have to be carefully controlled to provide an optimum cure rate and state in the less oxidizing environment.

References

[1.] R.W. Layer, Practical considerations of acceleration based on theory, " Elastomerics, May 1988. [2.]"Elastomere in V-W Fahrzeugen, " Kautschuk+Gummi Kunststoffe 1, 1991. [3.] V. Gallez, "Composition and process effects upon surface aspect, " Exxon Chemical internal report, 1994. [4.] Alan Galuska, "Surface characterization of elastomers," Exxon Chemical internal report, 1994. [5.] Alan Galuska and Eric Halverson, "Surface characterization using XPS and TOF-SIMS," Exxon Chemical Company internal report, 1992.