Trends in rubber mixing.

Economical manufacturing of elastomer products requires the right equipment operating at the lowest possible "off-spec" rates and scrap-rates.

Many rubber product manufacturers are not yet aware of the fact that the mixing room can be the main reason for severe problems in the following processing steps. Their revisions and improvements should include targets like:

* Reduction of mixing stages;

* reliable temperature control within the mixer;

* decrease of mixing times and/or higher batch weights;

* identical quality from batch to batch.

This article will summarize the most important trends in mixing technology and demonstrate the benefits by representative examples.

Internal mixer trends

Tangential or intermeshing?

The discussion about the two typical kinds of mixers, mixers with tangential and mixers with intermeshing rotor systems, is quite old. A definite general answer to which system is the better one is not possible. To make up your mind about the best choice, the specific application must be considered.

The areas of material flow within tangential and intermeshing rotors are quite different, as shown in figure 1.

[Figure 1 ILLUSTRATION OMITTED]

The tangential machine has a relatively big gap between the rotors so material can be dragged in quite fast. Feeding times are short and mixing occurs mainly between the tip of the rotor blade and the chamber side surface. The larger portion of material is moving in front of the rotor wing, transported along the wings, which means it is moved in the direction of the end-plates.

In the case of intermeshing rotors, the gap between the rotors is relatively small. The ram forces the material between the rotors and into the mixing chamber, thus giving high strain deformations to the raw material, resulting in high normal stresses which lead to excellent contact between rubber and metal surfaces and consequently to high heat transfer coefficients.

In tangential as well as in intermeshing systems, similar effects of mixing occur between rotor tips and chamber-side surface. The tips of intermeshing rotors have a much larger surface than those of tangential rotors, so the heat exchange between rotor tip and chamber-side is much better.

This very efficient heat exchange permits also a relatively exact temperature control, and the big strain deformations lead to a dispersion mixing which is greatly superior when compared to tangential machines.

In general, four types of tangential rotors are in use today:

* Rotors with a "softer" energy input. These rotors have modifications in flight length and angle so that the distributive mixing is improved (i.e., ZZ-2 of Krupp Elastomertechnik and the new rotor of Pomini). Moreover, the thermal behavior of the mixer can be effectively improved if the rotor is designed in two pieces.

* Rotors with "high" energy input. These rotors have been developed to reduce masterbatch mixing times. To achieve this, flight lengths of the main rotor have been stretched to force the material to pass over the wing tip (i.e., so called full 4 wing-flighted rotors).

* Rotors with priority on improved distributive mixing. In addition to the change of rotor flight lengths, these rotors improve the material transport from side to side and rotor to rotor. One way to do so is to elongate the long flight extremely so that material is intensively transported to the side. In combination with the wing contour an intensive distributive mixing can be achieved (i.e., H-swirl rotors of Kobelco).

* Rotors needing certain wing positioning. Traditionally in tangential machines the two rotors operate with slight differences in speed so that the position of the wings of the two rotors to each other changes permanently. Rotor geometries that need a certain and constant configuration of the wings to each other to be effective must be operated at even speed (i.e., Farrel ST rotors).

This concept which showed at the beginning some advantages in favor of the ST-machines can and has been easily transferred to other types of rotors if improvements could be achieved.

Independent from the described rotor geometries' development on tangential machines, there is no doubt about the general trend in favor of intermeshing machines. They conquer more and more market shares in all fields of sophisticated mixing.

Figure 2 summarizes the properties of both types of machines. Tangential mixers have the advantages of quick drag in/drop out times. Also, higher fill factors might be possible. Intermeshing machines have clear advantages under quality aspects like temperature control, mastication, oil integration and dispersion.

Figure 2 - performance characteristics of intermeshing and tangential mixers

                                           GK-N         GK-E
                                        tangential   intermeshing

Specific energy input                      (+)           (++)
Thermal performance                        (+)           (+++)
Mixing efficiency                          (+)           (+++)
Comastication at partial fill              (-)            (+)
Plasticizer blending capability            (-)            (+)
Machine utilization                        (++)           (+)
Feed intake and discharge performance      (++)           (+)

(+++) = excellent;

(++) = good;

(+) = satisfactory;

(-) = unsatisfactory

Traditional suppliers of intermeshing mixers are Krupp and Francis Shawl Farrel and Farrel's former licencees Pomini and Kobelco are trying with more or less success to "jump on the train."

A relatively new trend also in the U.S. is the substitution of 270-tangential mixers by GK 250Es putting the GK 250E with intermeshing rotors directly into the position of the 270. In most cases, the main drive stays the same, and only the last stage of the gear box is modified to even speed at relatively low costs. When you ask industrial rubber manufacturers why they do this, they might (or might not) show you their ROI-calculation.

Based on results achieved in the mixer manufacturers' testing facilities, by using their own recipes and materials, they made theoretical calculations which in the meantime have been confirmed in practice and came out even better than foreseen.

The batch weight on both machines is about the same. A cycle time of five to six minutes on a 270 has been reduced to three minutes on a GK 250 E. Instead of eleven batches per hour, they are getting 17 to 20 batches, just a production advantage of about 50% better. If 70% of their costs per kilogram mix is material and 30% is depreciation, energy, people, etc., and their production is up 50%, they reduce their total costs from 100 to 85%. A GK 250 E should give them an output of about 17 million kilograms/year. If one kilogram batch costs them, in average, $1.00, their annual savings could be as high as $2.5 million.

A relatively new idea, especially for intermeshing machines, is the change of the rotor gap during mixing. Pomini offers this system under the name variable intermeshing clearance. Experimental studies, however, show that the impact of this clearance variation is relatively small. The mixer itself gets a more complex mechanical design. Taking into account that customers like to have the highest availability of the machine to reach good economic results, a more demanding or complicated design of mixer components seems to be quite questionable.

Hard-steel coatings

A large inconsistency in mixing is the wear of all components of the mixer coming in touch with the raw materials. This wear leads over a period of time to larger net volumes of the mixing chamber, rounded edges of the rotor wings and other setbacks which force the rubber manufacturer to correct his recipes.

In the meantime there are hard steel coating materials on the market which reduce the wear of the components, bringing lifetime up to twice of the past. We only have figures from a crack-free hard-coating material available, which Krupp has been using for about five years. Compared to the normally used Stellite #1, this material, called WP 53, results in the following features:

* Hardness above 52 HRC (Stellite 1 is about 56);

* tensile strength above 700 N/[M.sup.2] (Stellite 1 is about 635);

* homogeneous structure;

* crack-free surface, which means

- holes for cooling nearer to the surface;

- increased output/efficiency;

- reduced potential for batch contamination;

- no corrosion through the cracks into the basic material; corrosion through cracks results in outbreaks of the hard coating.

* surface resistant to corrosion (Cr-content above 13%).

Krupp is following up the wear in the field. A 250 liter masterbatch mixer (tangential) in a tire plant operating in three shifts, seven days a week, coming up to 8,000 hours of operation per year had after 3.5 years in operation a wear of 0.6 mm on their chamber halves. Another mixer, intermeshing, operating in 1-1/2 shifts per day at five days per week, mainly one step mixing, shows after four years in operation a wear of 0.3 mm in the mixing chamber side.

Hydraulic rams

Another promising trend, also very important and with a broader application during the last years, is the exchange of pneumatic against hydraulic rams.

The main reason to do so has to be seen in the influence of the ram pressure on batch quality. Figure 3 shows the influence of different ram pressures on the dispersion quality of carbon black at different fill factors.

[Figure 3 ILLUSTRATION OMITTED]

The evaluation done on a seven liter tangential lab mixer shows that ram pressure variations result in drop temperature changes and dispersion quality deviations. Thus, especially for highly sophisticated compounds, variations in ram pressure (i.e., due to pneumatic net pressure fluctuations) within a batch and from batch to batch are totally unacceptable. As a hydraulic ram has its own individual hydraulic loop and pump and hydraulic oil cannot be compressed like air, quality problems related to the ram moving system can be eliminated completely.

Energy cost savings of the hydraulics don't have the importance in the U.S. as in other parts of the world, but operating globally and exchanging experience and recipes, global standards become more and more important. So thinking global is another argument in favor of hydraulic rams because of the economics. Compressed air is one of the most expensive energy transfer media in the plant and energy costs, for example in Europe, are much higher than in the United States.

The trend is not only to buy new machines with hydraulic rams, but also to exchange existing pneumatic operated rams for hydraulic ones. This will be enforced more in the future if the real and additional processing benefits of the hydraulic ram will be in use (development trend). Until now, a hydraulic unit for the ram movement is, from the processing view, more or less a copy of the pneumatic system.

The fact in both cases is that at the beginning of the mixing cycle the ram moves completely uncontrolled. Its position is fixed by the balance of power under the ram. At this point, the dragging force of the rotors, the polymer piece size and the amount of material under the ram show big variations. As a consequence, the fill factor of the mixer is not clearly defined. This has a huge influence on the batch quality. A large improvement (development trend) could be seen in the real control of the ram position during the entire mixing cycle.

A new control concept is shown in figures 4 and 5. It consists of the following characteristics: At the start-up of a new recipe, the ram operates with constant pressure. When the desired batch quality is established, the ram/time curve is used as a mastercurve for further batches where it is reproduced by position control of the ram. Figure 4 shows a typical curve ram position vs. time and figure 5 demonstrates that with ram position control, the deviations in batch quality can be minimized. This reproduction of a master curve is only possible with hydraulics and not with pneumatics.

[Figures 4-5 ILLUSTRATION OMITTED]

Optimizing the mixing process further

Although this article discusses the steps of optimizing the mixer itself a correct optimization has to include all aspects of the mixing room.

A very useful tool to optimize the mixing cycle is a variable speed drive. Figure 6 shows the dispersion quality of a SBR/carbon black compound as a function of rotor speed and fill factor. As the results show, the impact of the speed is huge and stronger than other parameters (i.e., fill factor, time, etc.); consequently, the mixer should be equipped with a variable speed drive. A possible process optimization is shown in figure 7.

[Figure 6-7 ILLUSTRATION OMITTED]

The upper half of the chart shows the conventional mixing procedure. Since at high rotor speed the temperature increase cannot be stopped, the mixing process has to be finished when a critical level has been reached. To obtain a high quality compound a second or even a third mixing stage has to follow.

Using a variable speed drive, as shown in the lower part of figure 7, can save additional steps of mixing. When the rotor speed is reduced the mixer can be used as a cooling unit to bring the compound temperature down to a limit where final mixing can be done.

Here again, especially intensively cooled intermeshing machines can reach excellent cooling rates. In a practice test we achieved on a GK 190 E, which is a 190 I machine, 1 [degrees] C down-cooling per second.

Another possibility to maximize the productivity is to use extremely high speeds at the beginning of the mixing cycle. Thus high shear stress can be established, which is quite effective in terms of dispersion of pigments or mastication of the polymer.

The trend today is to merge the possibilities variable speed drives (in very specific applications, even hydraulic drives) are offering with modern process controls which are able to split the mixing process into certain steps. Each step can be driven to another criterion such as temperature, time, torque, specific energy input or the like.

Thus interesting advantages can be gained, especially for soft compounds where high amounts of oil have to be incorporated. Only with a computer controlled mixer is it possible to inject the oil economically with sensitivity to torque, temperature or time. By practical tests we experienced a time savings of about 40% with the replacement of a manually controlled oil injection with a computer controlled system.

In the European industrial rubber goods industry, all these advantages are used to a large extent because industrial rubber manufacturers have a high interest in one stage mixing.

Silica mixing

Arising new silica compounds present mainly the tire industry with tasks which their standard equipment is nearly unable to handle under economic aspects.

Silica itself is a standard filler that has been in use in the rubber industry for many years. Recently a French tire producer introduced new tires that showed some remarkable advantages including better wet grip, lower abrasion and lower energy consumption

The secret behind these "green tires" is the use of silica whose surface has been activated by organosilanes and which, during vulcanization of the tire, establishes chemical bonds between filler and polymer.

From the processing point of view, two chemical reactions have to be performed:

* The reaction to bind organosilanes to the silica (= first reaction) [right arrow] this should be done within the internal mixer;

* the reaction to bind organosilanes to the polymer (= second reaction) [right arrow] this should not happen before vulcanization of the complete tire.

A problem during performance of the first reaction within internal mixers is that it is relatively slow at moderate temperatures of, for example, 120 [degrees] C.

To achieve a short reaction time, higher batch temperatures within this first step of mixing have to be established and must be kept for a time of up to 15 minutes. At elevated temperatures, however, the second reaction might start so that first scorch problems appear.

With typical mixers in the tire industry (size about 250 liters and above, tangential concept) the material temperature in the mixer has a steady slope upwards and it is not possible to run it at a speed where the mix temperature stays constant. As a consequence, tire producers drive the first reaction-mixing cycle to a certain temperature and are forced to terminate the mixing process. Thus, reaction times at elevated temperatures are in the range of two to three minutes per stage of mixing. To achieve 10-15 minutes in total reaction, the batch has consequently to pass the machine four to five times.

It seems to be obvious that this is not an economic way of compounding. The silica technology is therefore heavily forcing the tire industry to start using intermeshing mixers. Figure 8 shows that it is possible to hold the batch temperature within intermeshing machines. The compound is first mixed at high speeds to disperse the silica and to incorporate all fillers, oils and chemicals into the compound. Then the rotor speed is slowed down and fixed on a level where the dissipation and the heat flow to the chamber-side wall and the rotor-surface are in an equilibrium.

[Figure 8 ILLUSTRATION OMITTED]

Practical tests of large tire companies have shown that the type of process shown in figure 8 can easily be realized in intermeshing machines. With this way of operation, several stages of mixing could be saved and the process of organosilane bonding (or "hydrophobing" since it leads to hydrophobic compounds) can be done in one mixing stage of up to 10 minutes.

The economic benefits are easy to estimate. But there are also some further processing advantages: it is well known that silica itself has very strong filler-filler interactions. When no hydrophobing has been done, the dropped compounds exert therefore a dramatic high viscosity. Batch strips behave more like wood than like a rubber compound. This means the handling of not hydrophobed or only partly hydrophobed silica compounds is very difficult or even impossible. Mooney viscosities MS at 100 [degrees] C can reach values up to 100.

When the hydrophobing occurs, the surfaces of the silica particles are covered with the organosilanes and the filler/filler interactions are destroyed completely. Fully hydrophobed compounds show Mooney values of only 25-40 ML at 100 [degrees] C.

These examples show that silica (tread) compounds need a good temperature controlled mixer because different levels of hydrophobing will lead to very uneven properties in further processing.

Outlook

Looking at today's internal mixers might lead to the (wrong) conclusion that nothing principal has changed since Werner & Pfleiderer invented the first internal mixer in the year 1879 and Fernley Banbury received a patent on his ram and drop-door developments in 1916-17.

This article highlighted only some major innovations of the last years, which lead to very much improved internal mixers.

The development of silica compounds shows dramatically that new materials demand new mixing concepts.

The internal mixer with ram and drop-door design, in operation for 80 years now, is an indisputable versatile tool for all rubber product manufacturers. The developments on the material side still continue. Other mixing principles than the internal mixers show up until today large restrictions in flexibility and overall results.

We are sure that this internal mixer will continue or even gain importance in the mixing of rubber in the next millennium.

References

[1.] J. W. Pohl and A. Limper, "New angles in rubber mixing technology," Tire Technology International 1993.

[2.] W. M. Wiedmann and H. M. Schmid, "Optimization of tangential and intermeshing rotor geometries in internal mixers," KGK 34/1981.

[3.] H.M. Schmid, "Requirements and optimization in the production of compounds on an internal mixer," Thesis 1991 at the University of Aachen/Germany.

[4.] P.R. Wood, "Rubber mixing," Rapra Review Reports, volume 8, number 6, 1996.

[5.] A. Limper and J. W. Pohl, "With advanced mixing room technology into the next century," Tyre Technology Conference, Basel, Switzerland, 1993.

[6.] J. W. Pohl, "Rubber processing from the view of a systems supplier - reality and wishes," Tire Tech Asia 96, Sing-apore.

[7.] Limper, A. and Pohl, J., Gummi-Mischtechnik, Altbewahrte Prinzipien und neue Technologien, Gummi, Fasern, Kuns-tstoffe, GAK 9/1992 - Jahrg. 45.

[8.] Schmid, H.-M, Qualitats-und Produktivitatsverbesserungen im Innenmischer. Vortrag VDI-Seminar "Der mischbetrieb in der gummi industrie," VDI-Verlag, Dusseldorf 1984.

[9.] Cotton, G., "Mixing of carbon black with rubber 111 analysis of the mixing torque curve," Kautschuk + Gummi, Kunststoffe, 38, (8, 1985), S. 118.

[10.] Cotton, G., "Mixing of carbon black with rubber 11 mechanism of carbon black incorporation," Rubber Chemistry and Technology (Vol. 57) 1984.

[11.] Buskirk, B., "Practical parameters for mixing," Rubber Chemistry and Technology (Vol. 48) 1975.

[12.] Funt, J., "Mixing of rubbers," publication of Rapra, Shawbury, England 1977.

[13.] Sunder, J., "Analyse des mischprozesses im innenmischer," dissertation an der RWTH Aachen 1992.

[14.] Sunder, J., Neue Regelungsmoglichkeiten am innenmischer, " Vortrag auf dem 16. IKV-Kolloquium Aachen 1992.

[15.] Pohl J. and Limper, A., "New angles in rubber mixing technology," Tire Technology International 1993, U.K. & International Press.

[16.] Pomini, L., "VIC internal mixers offer greatest flexibility in mixing," Kautschuk, Gummi Kunststoffe 1994, Heft 3.

[17.] Michaeli, W. and Meiertoberens, U., "The influence of clearance variation between the rotors of an internal mixer in the mixing process," Kautschuk Gummi Kunststoffe, 1994, Heft 5.

[18.] Bomal, Y., Cochet, P., Dejean, B., Maruchat, J. and Gelling, 1. "Influence of mixing procedure on the properties of a silica reinforced tire tread," paper presented at the ACS Meeting, May 1992, Louisville, KY.

[19.] Freakley, P.K. and Patel, S.R., "Internal mixing: The interpretation of experimentally determined in chamber pressure and temperature fields," paper of the Institute of Polymer Technology, Loughborough, Leicestershire, 1990.

[20.] Valsamis, L.N., Canedo, E.L. and Donoian, G.S., "New methods in evaluating the performance of internal mixers," paper presented at the ACS meeting, Oct. 1989, Detroit, MI.