Optimization of the production of EPDM sponge rubber seals for automotive use.

The production of rubber profiles on continuous vulcanization units belongs to the most important processes of the rubber industry. The value of the annual sold profiles is around 0.59 to 0.72 million U.S. dollars in Germany. Automotive body seals represent a world market close to 100 kt/a. Unique performance properties in the fields of weather resistance, physical properties and processability are attributed to the automotive seals. To fulfill these demands they mostly are a combination of solid and cellular materials (refs. 1-3). Since the beginning of the 1990s, ethylene-propylene-diene elastomer (EPDM) has been increasingly used for automotive seals and profiles, as it complies with the required properties. Although the foam/sponge represents only a minor part of the profile, its production constitutes an important part of the EPDM application technology, which requires a high level of know-how. All aspects of EPDM-compounding in general are included in this manufacturing process. The increasing product performance attributes required for the sponge body seals, like the production of complex configurations to precise tolerance, a temperature independent load deflection and a smooth skin, lead to a complicated process due to the large number of process variables during fabrication and additionally due to the formulation effects.

In this article, results concerning the influences of this large number of parameters on product properties are summarized. Therefore, the effects of different acceleration systems on the product properties will be investigated. Different accelerator systems are chosen without taking notice of possible synergy effects between the different types of the accelerators. Then, the influence of different process parameters on the properties of the product (sponge rubber tube) is investigated, like belt velocity and power of the UHF-generator. Further analysis is performed, retaining the same acceleration types, to avoid synergy effects as much as possible. With each test, the amount of the accelerator is reduced in order to determine the relation between the used quantities and the properties of the products.

Additionally, extensive investigations are performed on a rheometer to measure rheological parameters describing the uncured compound. A mathematical description of the variation of the temperature profile in the seal during processing is used to find a temperature profile for the rheological measurement, which is similar to the production process. This is to make sure that the rheological parameters are measured under conditions similar to processing conditions. Furthermore, the mechanical properties of seals manufactured in an industrial production plant for automotive seals are measured. Finally, correlations are found between the rheological parameters and the mechanical properties, which allow the prediction of the product and sponge characteristics.

Sponge rubber technology

Although automotive body seals can be manufactured by a number of the elastomers, like EPDM, NR, SBR, CR, NBR or LSR, EPDM is the most important polymer for the sponge rubber production worldwide (ref. 3).

The manufacturing process of the EPDM-seals can be divided into three typical steps:

* Mixing operation: The quality of the dispersion and distribution of all ingredients is important for the final product quality, especially for the smooth surface of the profile.

* Extrusion: The temperature setting must be controlled to avoid an early start of the degradation of the blowing agent.

* Curing: The vulcanization and simultaneous blowing reactions have the most important influence on the quality of the final seal.

There is still little knowledge about the optimum choice of curing technology. Liquid curing medium (LCM), hot-air or microwave, ultra-high-frequency (UHF) curing belong to the common systems. One of the most important systems worldwide is the UHF/hot-air combination. It leads to a fast and homogeneous heating of even very complex shaped profiles within a short distance in the UHF unit. The following hot-air system keeps the profile on curing temperature to complete the curing and blowing reaction.

The investigations described below are concentrated on the sponge compound and the UHF curing process.

Materials and experimental techniques

For the following investigations, three different types of industrial processed sponge rubber compounds are investigated, as shown in table 1. For the chemical variations of the composition of the compound, the materials A and B are used. The aim of the variations in between compound A1 to A4 is to find an optimum temperature difference between the temperature at the start of the vulcanization and before the blowing reaction, without taking notice of possible synergy effects between the different types of the accelerators. Therefore, four different accelerator systems have been chosen to investigate their influence on the vulcanization reaction. The formulations reach from a very fast acceleration (A1) by using a maximum amount of accelerator, that depends on the solubility in the EPDM, up to a very slowly accelerated compound (A4) containing a low amount of accelerators. Due to the variations of the amount of accelerators, a vulcanization reaction should be realized, which takes place in a large temperature interval. The variations of the second compound B1 to B5 are produced by avoiding synergy effects as much as possible. Using the same types of accelerators, only their amounts are varied up to the maximum values of their solubility in the EPDM.

Table 1 - formulations of the compounds A, B and C

Compound in [phr]           A1       A2     A3       A4

EPDM                                    100
Carbon black N990                        25
Carbon black N550                        50
Calcium carbonate                        20
Mineral oil (naphthenic)                 70
Zinc oxide                                5
Stearic oxide                             1
Sulfur                                  1.5
Accelerator DPG              0      0.25    0.5     0.75
Accelerator CBS             0.5      1      1.5      2
Accelerator ZBEC            1.5      1      0.5      0
Accelerator MBT              3       3       3       3
Accelerator ZDBP             -       -       -       -
Accelerator thiuram         0.6     0.4     0.2      0
Blowing agent ACR            5       5       5       5
[Sigma]                    283.1   283.1   283.2   283.25

Compound in [phr]           B1      B2      B3      B4      B5

EPDM                                       100
Carbon black N990                           25
Carbon black N550                           50
Calcium carbonate                           20
Mineral oil (naphthenic)                    70
Zinc oxide                                   5
Stearic oxide                                1
Sulfur                                      1.5
Accelerator DPG             1.5     1.2     0.9     0.6     0.3
Accelerator CBS              2      1.6     1.2     0.8     0.4
Accelerator ZBEC             2      1.6     1.2     0.8     0.4
Accelerator MBT              2      1.6     1.2     0.8     0.4
Accelerator ZDBP             -       -       -       -       -
Accelerator thiuram          2      1.6     1.2     0.8     0.4
Blowing agent ACR            5       5       5       5       5
[Sigma]                     287    285.1   283.2   281.3   279.4

Compound in [phr]           C     C1      C2     C3    C4

EPDM                                     100
Carbon black N990                         30
Carbon black N550                         70
Calcium carbonate                         20
Mineral oil (naphthenic)                  85
Zinc oxide                 10     10      15     10    10
Stearic oxide                              1
Sulfur                                    1.5
Accelerator DPG             1     0.5     0.5    0.5   0.5
Accelerator CBS            1.5    1.5     1.5     2     2
Accelerator ZBEC            2      2       2      2     2
Accelerator MBT             2      2       2      2     2
Accelerator ZDBP            4      4       4      4     2
Accelerator thiuram         -      -       -      -     -
Blowing agent ACR           5      5       5      5     5
[Sigma]                    333   332.5   337.5   333   331

In order to find correlations between the rheological parameters and the final product properties, the production compound C is used. The vulcanization reaction and the degradation of the blowing agent should be optimized. The results measured under the following conditions are shown in figure 1. Both the start of the degradation of the blowing agent can be delayed (C2) and the effect of the acceleration system can be increased by delaying the vulcanization process. Hence, changes of the scorch time and the vulcanization rate take place simultaneously: The decrease of the amount of the guanidine accelerator (C1) and an increase of the amount of the sulfenamide accelerator (C3) increase the vulcanization velocity. The decrease of the amount of the thiophosphate accelerator decreases the vulcanization rate (C4).

[Figure 1 ILLUSTRATION OMITTED]

For the investigations of the uncured compounds, a moving die rheometer was used (ref. 4) which is able to measure the pressure. Isothermal and non-isothermal measurements are performed. Although the way of the heat transfer into the material by the rheometer (heat conduction) is still different to the production process (UHF), heating rates similar to the manufacturing process can be realized. In the following investigations, the course of the torque, which is a measure of the compound viscosity, and the gradient of the torque, standing for the vulcanization rate, are mainly taken into account.

For correlation purposes, samples of the different compounds are produced in an industrial production plant, described in figure 2. The manufactured profiles are analyzed relating to their physical and mechanical properties.

[Figure 2 ILLUSTRATION OMITTED]

Before the evaluation of the results of these trials are discussed, the calculation of the temperature distribution in the profile during the UHF-heating is described.

Calculation of the temperature profile in the UHF unit

First, the temperature profile for the rheological measurements is calculated. Relating to the microwave theory, the power density [P.sub.w] absorbed by the compound can be calculated according to equation 1 (refs 5 and 6). (The nomenclature used is summarized in table 2) (refs. 5 and 6).

(1) [p.sub.w] = 2[Pi] f [multiplied by] [E.sup.2] [multiplied by] [[Epsilon].sub.0][[Epsilon].sub.r] [multiplied by] tan [Delta]

Table 2 - nomenclature

DPG                 = Diphenylguanidine
CBS                 = Cyclohexylbenzothiazole-2-sulfenamide
ZBEC                = Zinc-dibenzyldithiocarbamate
MBT                 = 2-Mercaptobenzothiazole
ZDBP                = Zinc-dibutyldithiophosphate
[p.sub.w]           = Power density absorbed by the compound in
                      the UHF unit [W/[m.sup.3]]
f                   = Frequency of the UHF unit [[s.sup.-1]]
E                   = Electric field of the UHF unit [V/m]
[[Epsilon].sub.0]   = Electric field constant: 8.85 [multiplied by]
                      [10.sup.-12] [As/Vm]
[[Epsilon].sub.r]   = Relative dielectric constant [1]
tan [Delta]         = Dissipation factor [1]
[c.sub.p]           = Specific heat of rubber compound [J/kgK]
[Delta]v            = Temperature difference within the profile in
                      the UHF unit [K]
[Rho]               = Density of the rubber compound [kg/[m.sup.3]]
v                   = Take-off velocity of the profile in the UHF
                      unit [m/s]
A                   = Cross-section of the profile [[m.sup.2]]
L                   = Length of the UHF unit [m]
t                   = Residence time in the UHF unit [s]
[[Delta].sub.m]     = Penetration depth [m]
[S.sub.min]         = Minimum torque in the rheometer (vulcameter)
                      [Nm]
[P.sub.a,max]       = Maximum pressure in the rheometer
                      (vulcameter) [Pa]
[S.sub.degrad]      = Torque at the moment of increase of the
                      pressure in the rheometer (vulcameter) [Nm]
[Delta]w            = Specific lost energy [J/[m.sup.3]]
RSD                 = Residual stress under deflection [%]
[t.sub.s]           = Scorch time [s]
Q                   = Power absorbed by the rubber compound [W]
m                   = Mass throughput of the rubber compound
                      [kg/s]

In this equation, [[Epsilon].sub.0] is the electric field constant. The frequency f and the electric field E are parameters of the equipment, whereas the relative dielectric constant [[Epsilon].sub.r] and the dissipation factor tan 8 are material parameters of the sponge rubber compound, which are strongly influenced by the temperature. Values are taken from Fedtke and Ippen (refs. 7 and 8). Due to the energy balance, the temperature difference caused by the absorbed UHF-power is calculated by equation 2 (ref. 9).

(2) Q = m [c.sub.p] [Delta][Upsilon]

with in = [Rho] v A

After relating equation 2 to the volume of the material in combination with equation 1, the temperature difference [Delta][Upsilon] can be calculated as follows:

(3) [MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

If [[Epsilon].sub.r] tan [Delta] is constant, the rise of temperature in the UHF-unit is directly proportional to the residence time (figure 3, curve 1), whereas the temperature dependence of [[Epsilon].sub.r] tan [Delta] leads to curves 2 and 3 (ref. 8). As a first approximation, we use the linear relation (curve 1).

[Figure 3 ILLUSTRATION OMITTED]

McCrum et al. (ref. 10) explain the dependence of the parameters of the uncured material on the temperature and frequency with the fact that the resonance frequency of the dielectric is shifted to higher frequencies by an increasing temperature. The frequency of the UHF-generator is fixed at 2.450 MHz for the production of sponge rubber profiles by law. If the resonance frequency is higher than the UHF frequency, [[Epsilon].sub.r] tan [Delta] increases with increasing temperature. If the resonance frequency is lower than the UHF frequency, [[Epsilon].sub.r] tan [Delta] decreases with increasing temperature. Therefore, the heating up of the profile becomes unequal. If [[Epsilon].sub.r] tan [Delta] increases, parts in the profile of higher temperatures heat up faster. The decrease of [[Epsilon].sub.r] tan [Delta] leads to a more equal and homogeneous heating up.

The penetration depth [[Delta].sub.M] of the microwaves within the material also influences the temperature distribution inside the profile. The penetration depth [[Delta].sub.M] is caused by the absorption of the microwaves within the material (ref. 11). If the absorption is high and the penetration depth low, the core of the material is cooler than the surface. Therefore, a temperature gradient arises in the material.

A higher [[Epsilon].sub.r] tan [Delta] in the UHF-unit leads to a higher temperature increase (equation 3), whereas the penetration depth [[Delta].sub.M] of the microwaves is reduced, as the absorption increases (ref. 11). This also results in an inhomogeneous temperature profile in the compound. The temperature increase [Delta][Upsilon] of the material is in inverse proportion to the penetration depth [[Delta].sub.M] of the microwaves, with [[Delta].sub.M] = f([Delta][Upsilon], [c.sub.p]). If the temperature is changed due to other additives in the compound, [[Epsilon].sub.r] tan [Delta] is different, and therefore the specific heat capacity of the rubber compound is changed as well. Hence, a compromise between these two effects, a maximum penetration depth and a maximum heating rate, has to be found.

Measurements

The compounds A1 to A4 and B1 to B5 are taken into account, while the influence of the accelerator system on the theological and mechanical properties is analyzed. Rheological investigations of all uncured compounds are performed on the MDR. Then, seals are produced on the production plant (figure 2). The influence of different process parameters on the properties of the product, like belt velocity and power of the UHF-generator, has been investigated (table 3). The profiles are analyzed due to their cell size, geometry, Shore A hardness and residual stress under deflection.

Table 3 - experimental data

Compound                               A1-A4          B1-B5

Belt velocity [m/s]                 0.09, 0.18         0.13
Temperature
(UHF entrance) [[degrees] C]            80              80
Temperature
(UHF exit) [[degrees] C]                               140
Power UHF [kW]                    3, 3.6, 4.8, 6        6
Heating rate [K/s]                                     2.33

Compound                                C-C4

Belt velocity [m/s]                     0.08
Temperature
(UHF entrance) [[degrees] C]             80
Temperature
(UHF exit) [[degrees] C]           130, 145, 155,
                                      167, 185
Power UHF [kW]                      1.8, 2.1,2.7,
                                      3.6, 4.44
Heating rate [K/s]                0.89, 1.15, 1.33,
                                     1.54, 1.86

In order to find correlations between the rheological test results and the final product quality, all compounds of group C are analyzed on the rheometer and then seals out of these compounds are produced on the production plant. The course of temperature for the rheometer is calculated by using equation 3, the material parameters of the different compounds and the residence time in the UHF. Heating rates from 0.17 K/s to 0.5 K/s can be realized, however, these heating rates are still lower than in the UHF.

The rubber compound in the cavity of the rheometer is sheared by a rotor (sinusoidal oscillation). Due to the measurement of its maximum torque as a function of time and elevated temperature, the information regarding the vulcanization and blowing behavior and the viscosity can be obtained. Following, it is concentrated on the elastic properties of the materials. At the beginning of the analysis, the torque decreases due to the decrease of the viscosity of the sample. Then the curing starts and torque and viscosity respectively increase. The minimum of the torque [S.sub.min] is a characteristic value of the compound. Simultaneously, the blowing agent degrades and the pressure in the cavity increases, reaching a maximum [P.sub.a,max]. The very moment of a significant increase of the pressure is measured as a fold of the torque. The value [S.sub.degrad] of this point is, like the maximum pressure [P.sub.a,max], characteristic for the blowing reaction. These characteristic rheological values are used for the correlation with the properties of the product.

With all profiles (compound C), compression set and tensile tests are carried out. The residual stress under deflection, the compression deflection force, the strain at break, the tensile strength at break and the specific lost energy are measured. Only the density [Rho], the specific lost energy [Delta]w, and the residual stress under deflection (RSD) are taken into account in the following discussion.

Test results and interpretation

Rheological and mechanical influence of the acceleration system on the material and profile properties Compounds A1-A4

In the following, first, the results of the investigations concerning the influence of the acceleration system on the vulcanization behavior, the final profile geometry and the cell size are described. Afterwards, the influence of different belt velocities and power of the UHF on the cell size of the profiles are discussed only for compound A2.

The increasing size of the cells of the profiles made by compounds A1 to A3 is shown in figure 4. However, compound A4 has got the smallest cells. The behavior of the first three compounds can be explained by the acceleration system, which becomes slower each time. A1 has got the fastest acceleration system, therefore the vulcanization of the compound starts later than for compounds A2 and A3. However, it is still much faster than the vulcanization velocity of the other compounds. Compound A2 has got a shorter scorch time and a faster vulcanization velocity than compound A3. Therefore, the cells of A2 are smaller than the ones of A3. As the tensile stress at break and therefore the degree of vulcanization of compound A2 are higher than the ones of compound A1, the vulcanization velocity of A2 obviously becomes slower than the one of A1. Summarizing, the cell size is mainly dependent on the vulcanization velocity and only little on the degree of vulcanization or scorch time. The plots of the rheological measurements show an early blowing for all materials, which has not been expected but leads to even larger cells. During the investigations, the combinations of different types of accelerators lead to synergy effects which strongly influence the scorch time of the curing reaction and therefore, the size of the cells in the final profile.

[Figure 4 ILLUSTRATION OMITTED]

The investigations show that the geometry of the profiles is hardly influenced by the variations of the acceleration system. A simple tendency cannot be found.

Exemplarily, the influence of the belt velocity and the power of the UHF-unit on the cell size is discussed below. Using a high belt velocity of 0.18 m/s in the UHF, the cell size is only a little dependent on the power of the UHF-unit. As the belt velocity is very high, the profile is not sufficiently heated up. A maximum cell size is achieved at a generator power of 4.8 kW using a slow belt velocity in the UHF unit of 0.09 m/s. The cells might grow together due to the slow belt velocity and the vulcanization process.

The investigations show that the cell size of the profiles increases at lower belt velocities, however, no proportionality can be found. That means that the generation of gas depends on the residence time of the compound in the UHF-unit or the hot-air-channel. However, the gas generation and the cell size, respectively, are less influenced by the power of the UHF-unit.

Compound B1-B5

Figure 5 shows the results of the analysis of the bubble sizes of the samples B1 to B5. A lower amount of accelerators leads to smaller cells. The amount of accelerator influences the rate of vulcanization. A lower amount leads to an increased degree of vulcanization. So, the cells become smaller, as the degree of vulcanization increases and hence, limits the cell growth. Therefore, the geometry of the profiles becomes smaller with the lower amount of the accelerator. A connection between the cell size and the final product geometry exists.

[Figure 5 ILLUSTRATION OMITTED]

Figure 6 shows the results of the residual stress under deflection (RSD) test of the compounds B1 to B5. Normally, a low value is required for the RSD, as a low RSD stands for a high degree of vulcanization. The highest values of the RSD are seen for sample B3, whereas a much better, lower value exists for sample B4. B4 has got the highest degree of vulcanization. This shows that the acceleration system of this compound is well suited for this application. Higher amounts of accelerators do not always lead to higher degrees of vulcanization.

[Figure 6 ILLUSTRATION OMITTED]

Correlations between the theological parameters and the final product quality

Rheological investigations (compounds C-C4)

The pressure course, measured in the rheological experiment described above, is a main clue for the characterization of the expansion process. Figure 7 shows the evaluation of the maximum pressure dependent on the temporal heating rate for the different compounds. The pressure increases from compound C to C3 and decreases at compound C4, as the vulcanization starts earlier. That means that the blowing agent degrades more easily at a lower vulcanization level. It is also remarkable, that the highest amount of zinc oxide (compound C2) leads to a much higher pressure in the cavity related to compound C.

[Figure 7 ILLUSTRATION OMITTED]

The scorch time drops down to half of its initial value if the heat input is three times faster than regular. The heating rate has an important influence on the material properties, as seen in figure 8 for the maximum vulcanization rate. The dependence of the pressure on the heating rate causes density p decreases with an increasing heating rate. Previous rheological investigations have shown that the process temperature is very important. It has to be kept constant to avoid quality fluctuations in the final product quality.

[Figure 8 ILLUSTRATION OMITTED]

Further rheological results, which are not described here, allow the conclusion: If the material has got a lower viscosity, the blowing agent can degrade more easily. These results show the influence of changes of the acceleration system, the zinc oxide and of the different heating rates on the measurable rheological parameters.

Mechanical investigations (compounds C-C4)

The results of the investigations of the profiles also show an influence of the heating rate realized in the UHF unit on the density of the profiles. An increasing heating rate leads to a lower density. The influence of the heating rate is stronger than that of the variations in the compound formulation. Similar results are found for the mechanical properties.

Figure 9 shows the results of the RSD test concerning the different samples of compound C. The residual stress under deflection decreases from sample C to sample C3, and increases again for sample C4. Due to the variation of the mixture, the scorch time increases. Therefore, the blowing agent can degrade more easily and larger cells can grow. Simultaneously, the sponge will tend to show cells that are mainly open so that the residual stress under deflection decreases. The residual stress under deflection increases only for sample C4, which has a shorter scorch time. Summarizing, the variations of the formulation do not lead to an improved product, which mainly should have a closed cell structure. Even higher heating rates cause mainly open cell foams, as the residual stress under deflection decreases.

[Figure 9 ILLUSTRATION OMITTED]

A lower density of the sponge rubber profiles can be realized by an increasing heating rate. However, the cells in the rubber are mainly open then. To avoid the open cell structure, variations of the acceleration system or the EPDM itself become necessary. If the sponge rubber has got an open cell structure, the gas within the cells can easily escape if the profile is stressed. That results in a low value for the residual stress under deflection for a low density. If a mainly closed cell structure can be produced at a low density, the dependence of the residual stress under deflection on the density is contrary: A high residual stress at a low density and a low one at a low density.

The cells in the sponge rubber are on average smaller at a higher density and therefore, the volume in the sample that is filled by gas is smaller as well. Accordingly, at a lower density the cells are larger, like the amount of gas in the cells. If both rubber samples are compressed up to 50% of the original height (ref. 12), the pressure in the cells increases. Therefore, the gas molecules, which are smaller than the rubber molecules, diffuse out of the compound to equal the pressure between the gas cells and the surroundings. If the load is taken away, a partial vacuum arises within the cells due to the elastic behavior of the rubber to adopt the original form. The residual stress under deflection is measured 30 minutes after unloading the specimen. During that period of time, smaller cells can be filled faster by air than bigger cells due to their smaller volume. Therefore, higher values are measured at lower densities and vice versa.

The RSD is measured for block pressure (compact material). However, the results can be reduced by special designs of the profile. Due to the mainly very complicated designs, it is difficult to find correlations between the rheological parameters and the residual stress under deflection behavior, as the influencing factor "geometry" cannot be described by the rheological tests.

Finally, the lost energy [Delta]w is discussed. Being representative for the cross-linking density, a low lost energy means that the irreversible deflection is low and the elasticity is higher due to the higher cross-linking density. The lost energy decreases with higher heating rates so that the cross-linking density increases.

Correlations

A good correlation is found between the physical or mechanical properties of the profiles and the rheological values. Density, tensile strength at break, residual stress under deflection and compression deflection force are chosen as the main quality criteria. In this article, only the results for the density p and the residual stress under deflection are described. The regression equation 4 can be found, using the rheological values to predict a final product density:

(4) [Rho] = 0.50468 [multiplied by] [kg/[m.sup.3]] -0.015628 [multiplied by] [kg] [multiplied by] [P.sub.a,max]/[S.sub.degrad] + 0.2334 [multiplied by] [kg/[Nm.sup.4]] [multiplied by] [S.sub.min]

The density of the sponge rubber is adjustable by the combination of the amount of blowing agent and the acceleration system. The amount of the blowing agent shows a good correlation with the pressure [P.sub.a,max] and the setting of the acceleration system correlates with the value of the torque at the moment of degradation of the blowing agent [S.sub.degrad]. Furthermore, the experiments have shown that the viscosity of the compound is important to get the desired density of the sponge rubber profile. For that, the minimum of the torque [S.sub.min] can be used. Figure 10 shows the measured and calculated values of the density and the regression line. The correlation coefficient [R.sup.2] stands for the quality of the correlation (equation 4). In this case, the correlation coefficient for the calculated density is [R.sup.2] = 0.81. A high amount of blowing agent in the compound leads to a higher pressure, and therefore to a lower density. If the acceleration system causes an early curing, the value [S.sub.degrad] will increase, and consequently the density will increase as well. A low viscosity of the compound results in a lower [S.sub.min] and a lower density.

[Figure 10 ILLUSTRATION OMITTED]

The regression equation 5 can be found, using the rheological values to predict a final product residual stress under deflection for an open cell sponge rubber:

(5) RSD = 52.671 [%] - 0.17014 [multiplied by] [P.sub.a,max]/[S.sub.degrad] [multiplied by] [kg] + 12.0746 [multiplied by] [kg/[Nm.sup.4]] [multiplied by] [S.sub.min]

The higher the degree of blowing, the lower the density of the sponge rubber and the higher the tendency of the rubber to perform open cells. Therefore, the RSD increases. Due to this connection between the density and the RSD, a correlation between the rheological parameters an J the RSD, as shown in equation 5, can be found. The discussion of the correlation is similar to the one of the density. Figure 11 shows the measured values. The correlation coefficient for the calculated RSD is [R.sup.2] = 0.86.

[Figure 11 ILLUSTRATION OMITTED]

The correlation does not describe the RSD for a closed cell structure of the profile, as in this case the mechanism is different. A closed cell structure arises, if there exists an equilibrium between the toughness of the compound or the polymer, respectively and the pressure in the cells or the gas production of the blowing agent respectively. The gas volume of the blowing agent can easily be measured. It can also be described by the pressure measured during the rheological tests. However, it is more difficult to describe the toughness of the compound by a rheological value. Even the viscosity ([S.sub.min]) is not appropriate.

Summary

The investigations described have shown that both the temperature, which allows start of vulcanization and blowing reactions, and the acceleration system are of an important influence on the final product quality. It could be shown that a low amount of accelerator leads to the best properties of the profiles. The process temperatures have to be well known and must be kept constant to avoid quality fluctuations in the final product. Rheological results measured under temperature conditions similar to the manufacturing process lead to a more significant prediction of the product properties. Therefore, a calculation of the temperature development within the profile in the curing system e.g., UHF system, is useful. Furthermore, the investigations have shown that correlations between the rheological material properties and the product properties can be described, so that a prediction of final product quality criteria becomes possible.

References

(1.) J.W.M. Noordermeer, Rapra the 4th Cellular Polymer Conference, UK, 1 (1997).

(2.) G. Stella and N.P. Cheremisinoff, International Rubber Conference, Paris, 43 (1990).

(3.) A. Hill, Technischer Handel 70 (7), 292 (1992).

(4.) J.A. Sezna and H. Burhin, paper No. 93 presented at 146th meeting of the Rubber Division, American Chemical Society, Pittsburgh, USA, 1994.

(5.) H. Focht, Kautschuk Gummi Kunststoffe 29 (4), 187 (1976).

(6.) W. Hoffmann, "Rubber Technology Handbook," Hanser Publishers, Munich, Vienna, New York, 1989.

(7.) M. Fedtke and F. Schramm, Kautschuk Gummi Kunststoffe 51 (3), 201 (1998).

(8.) J. Ippen, "Mischungen fur die kontinuierliche vulkanisation im UHF-feld," Bayer AG, unpublished report, Leverkusen, 1969.

(9.) W. Wagner, "Warmeubertragung," Vogelverlag, Wurzburg, 1984.

(10.) McCrum, Read and Williams, "Anelastic and Dielectric Effects in Polymeric Solids," 1997.

(11.) H. Puschner, "Warme durch Mikrowellen," Philips Technische Bibliothek, 1964.

(12.) ISO-No. 815, 1972.

A. Krusche and E. Haberstroh, Institut fur Kunststoffverarbeitung IKV, Aachen, Germany