Recent trends in mixing technology.

There is a widespread belief that rubber manufacturing is a rather mature industry with only marginal projected growth and lack of opportunity for innovation. Although this contention has some limited validity, it does not truthfully reflect the actual situation, especially with reference to mixing technologies. Indeed, in the past two decades we have witnessed significant developments in equipment design leading to increased output and higher productivity. So, when it comes to continuous change, this sector is a perfect fit for today's business environment.

Early developments

In 1820, Thomas Hancock "pioneered" the beginnings of today's rubber industry by inventing the masticator, a rudimentary wooden roll mill with protruding spikes capable of disintegrating "that tenacious rubber" into a workable material. Edwin M. Chaffee of Roxboro, MA, received a patent in 1839 for a processing machine that was the forerunner of the modern two roll mill. Since 1854, the wooden rolls were replaced with smooth chilled iron rolls. In 1916, the mill rolls were enclosed in a cast iron shell and a ram added to keep the rubber between the rolls. This was the first internal mixer, developed by Fernley H. Banbury. Later on in the early 1930s, Francis Shaw developed an internal mixer with interlocking rotors.

Extruders have been in use for almost 150 years. The first extruders were hand driven devices patented and manufactured in England in 1845 by H. Bewly and R.A. Brooman. The invention of ram extruders to cover wires with guta percha dates back to 1854. In the 1880s, screw extruders were invented and utilized to process components for pneumatic tires. During the 1950s, the so called "normal stress pump" was first introduced. This facilitated the development of continuous compounding equipment such as the Kenics static mixer-extruder (ref. 1), in which a strip of compound is fed into a gap between a rotating disc and a stationary element consisting of a series of alternating right and left hand helices that continuously divide the compound, causing it to rotate around its own hydraulic axis.

Factors affecting the mixing of polymeric materials

With a little visualization, let's return to the mill room of a medium sized non-tire manufacturing facility in the late 1960s. Through the noisy and rather dusty environment, one can distinguish one or two internal mixers (#9 or #11) one huge 84" OD batch-off mill and nearby a circular cooling tank filled with a water/talcum suspension. To our left and right are several 42" two-roll mills in full action with skilled operators cutting, folding and rolling up the stock to ensure proper dispersion of minor ingredients such as curatives, etc. The finished sheets were subsequently taken on carts to other plant sections for further processing. There is a saying: "The more things change, the more they stay the same," and this is generally valid for the mill room--the place of action--to this very day. The reasons for this albeit generalized situation are the rather simple reluctance to change and the broad utility of the two-roll mill for a variety of elastomeric compounds in terms of a rather consistent, satisfactory and uniform ingredient dispersion. Today, in spite of the unquestionable advances in the design, performance and productivity of other mixing equipment, internal mixers, extruders, etc., to be reviewed later on, there is hardly one single location where a rubber mill is not either in daily use or at least maintained in good working condition. This should not surprise anyone, knowing that the roller mill always works with a once given cylinder speed and a once fixed friction, and this combination was found satisfactory for 20 different elastomers, and sometimes for several hundred different formulations. Over the years, rubber mills have been greatly improved via installation of automatic stockblenders, space saving direct hydraulic drives providing variable cylinder speed, and thus adjustable friction, safety features capable of reducing the cylinder movement to a maximum of one inch after hitting the emergency stop, etc. In spite of the above, mills still have several shortcomings, such as being difficult to fully automate, some inherent batch to batch variations, lingering safety concerns, low output, and it is generally still a labor intensive, dirty operation.

The nature and physical form of raw materials has had a major impact on the evolution of mixing techniques. Not too long ago, natural rubber was supplied only in huge 100 kg (220 lb.) bales requiring, after hot housing, cumbersome slicing via hydraulic guillotine cutters. These slices were subsequently taken to a weighing station. There, all the formula components were weighed out according to the amounts shown on the pertinent batch cards, neatly bagged and placed in plastic bins for subsequent transportation to the mill room via carts or conveyor. Needless to say, this was, and still is in smaller factories, a highly labor intensive and not too dependable operation. With the advent and increasing usage of SBR (alone or as carbon black masterbatches) and other synthetic rubbers supplied as bales not exceeding 20-50 kg, and the availability of other minor ingredients in customized pre-dispersed form, the above described situation has changed dramatically for the better, giving way to fully mechanized and automated weighing, metering and feeding installations which can be controlled via specialized software programs. This has lead to higher operational efficiencies, productivity and elimination of operator related errors. The high volume tire industry has greatly benefited from this modernization, especially after the large-scale introduction of SBR-carbon black masterbatches, which simplified the entire compounding operation. Challenged by this trend and the appearance of various polyisoprene grades, the natural rubber producers had little choice but to adjust to the changing and more quality-oriented marketplace. As a result, a series of technically specified rubbers (TSR) are now available under a variety of designations (SMR 5, etc.) in a convenient bale size: 33.3 kg (75 lbs.) as well as in crumb form. Of special interest for the compounder are TSR grades covering a range of Mooney viscosities. Because of its stereo-regular structure--100% cis-1,4 polyisoprene units--NR crystallizes upon storage at temperatures below 20[degrees]C. The rate of crystallization varies with the temperature, type and condition of the rubber. At -26[degrees]C, a bale of rubber becomes hard after a few hours, whereas at 8[degrees]C, it takes about one month before crystallization occurs. The melting temperature (Mt) of NR is always higher than that at which it crystallizes. Once the Mt is reached, crystallized rubber melts immediately. The Mt is determined by the conditions at the time of crystallization. Since in practice the heat history of the rubber is seldom known, to ensure melting, the temperature at the center of the bale must be at least 28[degrees]C. According to a pertinent study (ref. 2), at an environmental temperature of 40[degrees]C, it takes more than 300 hours for the center of a one ton TSR pellet to reach approximately 30[degrees]C from an initial temperature of 0[degrees]C (32[degrees]F). Generally, and depending upon practical circumstances, hot housing of NR may take between 48 hours and up to two weeks. Needless to say, any attempts to process "frozen" NR must be avoided, since it may cause poor ingredient dispersion, inconsistent physical properties and, at worst, damage to expensive mixing equipment.

In the mid 1970s, Polysar introduced their SS 250 series of high styrene content butadiene copolymers in granular form, primarily for injection molding, but also for other applications as reinforcing agents. This set the stage for the development of TPEs of the Kraton type (Shell), which were readily accepted by the adhesive and sealant manufacturers due to their free flowing and readily soluble pellet form. In a related development, five of the Ameripol (Goodrich) SBR grades were produced in a similar readily soluble crumb, also eliminating the need for milling.

More recently, as a follow-up to the above trend, elastomers are commercialized in powder form to facilitate injection molding, as well as in liquid form (NR, polyisoprene, butyl) offering easier processing at higher filler loadings. Of special interest for the automotive industry are the "in place" injected silicone rubber gaskets.

As an example of a unique opposite development, the plastics industry has taken advantage of the existing rubber processing equipment in the case of high molecular weight grades of PVC that exhibit elastomer-like viscoelastic properties within a certain temperature range.

Compound theology and formulation design are closely related and must be properly understood to achieve an efficient processing environment. Rheology can be defined briefly as the science of flow and deformation of matter. However, in the case of a polymer matrix present in any compound, the picture is more complex, since rheology is concerned with the relationship between several factors such as stress, strain, time, temperature, as well as the polymer's chemical and morphological structure. If we consider a desirable balance between service performance and ease of processing, a qualitative inverse relationship can be schematically represented, as shown in figure 1.

[FIGURE 1 OMITTED]

One can distinguish the areas, including along the slanted line; the region of common experience above the line; the target area of a successful product; below the line: the disaster zone. The task of formula design can be simplified if the compounder is focused on the "comfort zone" of common experience. This involves a thorough understanding of the chemical nature and polymer morphology for proper end use selection, expressed via the glass transition temperature Tg (ref. 3), as well as careful consideration of other criteria such as availability, nature of inert or reinforcing fillers, as well as of other common ingredients (process oils, antioxidants, curatives, etc.), their possible matrix interactions, and finally, the very important overall cost picture. According to a generalized estimate, the raw material cost represents approximately 33% of the finished product cost. It is also useful to look into the computer file of former similar end use formulations and check pertinent ratios, i.e., polymer:filler, filler:process oil, etc. In addition, a bibliographic search could reveal insights into over 9,000 recipes published between 1948-1957 alone (ref. 4). Compounders have also attempted to shorten the formulation work via contour maps based on a grid of experimental data points. Thus, extender oil plotted vs. SRF carbon black in phr can provide a range of useful data for a cured SBR compound in regard to tensile strength, elongation and hardness (ref. 4).

For the practitioner, a two-level factorial experimental design can reveal sometimes overlooked recipe component interactions. While the above mentioned techniques can be of some help, they cannot replace a compounder's experience, intuition and willingness to innovate by working some extra hours. Even so, for those interested, there is always something new to be learned, especially on the factory floor, and this is one of the reasons why some of the black magic of compounding is still an art rather than a purely scientific endeavor.

Compound rheological measurements are extremely important in the lab to assess processability. The purpose of these measurements, like the Williams plasticity test (in between plate compression) and especially rotary shear instruments like the Mooney viscometer, is to expose the sample to shear rates similar to those encountered in the actual factory process step. In spite of some unjustified criticism regarding certain limitations, the Mooney viscometer has gained worldwide acceptance, and its operation is described in detail in ASTM standard D-1646-81 : Results of a typical test are normally reported as follows:

50-ML 1 + 4 (100[degrees]C)

where 50 M is the Mooney viscosity number, L indicates the large rotor inserted in the cavity, 1 is the time in minutes to preheat the specimen, 4 is the time in minutes (after starting the motor) at which the reading is taken and 100[degrees]C is the temperature of the test. The Mooney viscometer is also universally used to determine the scorch (premature cure) characteristics of compounded rubber. It is thus possible to plot Mooney units vs. time. In this case, the test is run between 120 and 135[degrees]C.

The mixing proceeds

With a few exceptions, there are well-established sequences of component incorporation into a rubber matrix. Thus, in general, after some preliminary mastication, antioxidants, activators and accelerators are added gradually, followed by reinforcing and/or inert fillers, process oils and finally sulfur. The goal was, and is especially today, to optimize the dispersion of all ingredients in the most cost-efficient way while achieving a certain desirable Mooney viscosity for proper subsequent processing (calendering, extrusion, etc.). The end result is obviously a cured product capable of consistently meeting specification requirements in regard to physical properties (tensile strength, elongation, hardness, etc.), as well as certain performance test criteria, especially for automotive parts. Due to the highly competitive business climate and ever more demanding quality requirements only relative lately, in the past three decades, serious investigations were undertaken to gain a better understanding of the mixing process. Thus, according to Palmgren (ref. 5) and Funt (ref. 6), the ingredients in the mix undergo four physical changes:

* Incorporation--initial "wetting stage" in which the initially separate ingredients (rubber, filler, plasticizers, etc.) form a coherent mass.

* Distribution--the process of increasing homogenization--also called simple or extensive mixing.

* Dispersion--filler agglomerates are gradually distributed through the rubber by simple mixing and then broken down to particulate size, giving a fine mixed phase. This is where (in the case of carbon black) the bound rubber is formed because of the high shear stresses needed to break down the particle agglomerates and force the elastomer into the carbon black interstices.

* Plasticization--in contrast to natural rubber, most synthetic tubbers are not readily broken down in molecular weight by shear and heat. The viscosity changes result primarily from the intimate incorporation of plasticizers (process oils, etc.).

From these phase changes, subsequent work showed that incorporation and dispersion are of particular interest to the compounder.

Incorporation

Tokita and Pliskin (ref. 7) pointed out that during processing, rubbers do not deform smoothly or flow steadily, but rather yield and rupture. They were also the first researchers who graphically related deformation and elongation at break to the incorporation time. As a follow-up to the above finding, Nakajima (ref. 8) emphasized the significance of extensional or elongational flow in the mixing process. According to his findings, the incorporation phase has two mechanisms. In the first, the elastomer undergoes a large deformation, increasing the surface area for accepting filler agglomerates and next sealing them inside. In the second mechanism, the elastomer breaks down into small pieces, mixes with the filler agglomerates and once again seals them inside. The first mechanism can be easily observed in an open mill, whereas the latter cannot, because the breaking and sealing steps occur on a micro-scale.

Dispersion

As a result of another investigation, Gessler, Hess and Medalia (ref. 9) reported that at the end of the incorporation phase the majority of the filler is present as rubber filled pellet fragments. These act as large filler particles, whose actual volume is greater than that of the filler alone, because of the rubber inside the particles and the rubber trapped between them. As the dispersion proceeds and these agglomerates are broken down, the rubber between the particles is released into the matrix. Properties such as viscosity and die swell, which are related to the volume fraction of the rubber in the mix, reach their optimum at the end of the dispersion phase. Therefore it becomes self-explanatory why a well dispersed mixture always has a lower viscosity and a higher die swell than a comparable less homogenous compound. The validity of the described dispersion mechanism was confirmed by further studies showing that the presence of carbon black aggregates--present when the dispersion is incomplete--has a negative effect on the modulus and torque values (from curemeter curves) before and after curing, when compared to the data from an identical well dispersed compound. Based upon the above findings, the incorporation and dispersion steps must be carried out in the most efficient way.

Roll mills

An open two-roll mill consists of two parallel horizontal rolls rotating toward one another at different speeds. This opposite rotation of the tolls pulls the elastomer and other ingredients through the nip (gap). The back roll is the one rotating faster vs. the front roll by a ratio called the friction ratio. The actual work is done on the front roll, either manually via angular knife downward sheet cutting followed by off and on plug rolling into the nip, or mechanically via a stockblender. The very important temperature control is achieved via spray nozzles by circulating hot or more frequently cold water through the longitudinal core of the hollow equally gauged cylindrical rolls. Mixing is in general a function of:

* Roll speeds--the value (friction ratio) is fixed;

* number of foldings between passes;

* number of passes;

* the gap between the rolls.

The second and third items are functions of the elastomer's nature, the batch size, milling time and also depend on the viscoelastic properties of the compound being mixed. The value of the fourth item is easily changed on most machines and gradually increased as the batch is being mixed. Modern factories utilize semi-automatic stockblenders schematically shown in figure 2.

[FIGURE 2 OMITTED]

This more recent roller guided device is moving back and forth above the rear of the back roll, and via a take-up unit continuously recirculates a compound strip into the rolling stock plug. Thus, within a certain preset mixing time, it is possible to ensure batch uniformity while eliminating operator inconsistencies. From the mill, the finished sheets are placed on the racks of a conveyerized batch-off unit that performs spraying with a cooling water/talcum suspension, forced air flow drying and transportation to other factory departments.

More recently, efforts have been made to gain a better understanding of what is happening in the small gap region between the two rolls where the rolling stock is subjected to high shear stresses. It is in this high shear zone that filler particle agglomerates are subjected to hydrodynamic forces that tend to break them apart. One can visualize that the magnitude of the separating forces changes as agglomerates rotate freely along the streamlines in the flow field. It is now generally assumed (ref. 10) that the particles do not cross streamlines and accordingly:

* Dispersive mixing occurs exclusively in the high shear nip zone;

* once broken the agglomerates become randomly oriented in the flow field;

* agglomerates that were not broken enter the nip in the following pass, keeping the streamline pattern:

* at any time during the mixing process all fluid elements, particles and agglomerates, arc subjected to the same number of passes through the nip region.

For these reasons, the uniformity of the dispersive action on the two-roll mill is excellent and represents a constant challenge for the rotor designers of internal mixers.

Internal mixers

In a typical internal mixer, the material is subjected to less shearing, since the gap between the rotors and mixing chamber is not adjustable. At high filler loadings, rotor speeds and larger batch sizes combined with much shorter mixing cycles, heat transfer is more difficult, causing increases in batch temperature which requires close control. However, the incorporation step is more rapid due to the hydraulic ram pressure, and the mixing depends less on the viscoelastic properties in the batch. A comparative illustration of typical mill vs. internal mixing for a standard natural rubber recipe with 50 phr carbon black (ref. 11) is shown in table 1.

As seen, it takes almost twice as long to mix the standard recipe on an average size mill for a roughly ten times smaller batch size compared to an average capacity internal mixer. For highly loaded EPDM compounds, etc., the economics are even more favorable in the case of modem internal mixers. For this primary reason, major tire manufacturing plants have switched almost entirely to these modern machines, while still retaining some mills for downstream final homogenization to ensure optimum component dispersion.

For most rapid and efficient mixing, an overload of 10 to 15% is rather common. Even so, in practice the mixing chamber volume utilization is about 70%. Most compounds having a cured hardness of 40 to 80 durometer are thoroughly mixed after 3-4 minutes at 20 rpm. For highly filled compounds, especially those based on EPDM, the upside down addition of ingredients is most effective. This involves addition of all fillers and oils prior to that of the elastomer. In contrast, compounds with very high process oil levels, soft sponge stocks and low viscosity mineral filled stocks are generally dump-mixed via the conventional right side up way. There are also other innovative practices, such as seeding mixing, in which a small amount of a previously well mixed batch is added to the new batch, and sandwich mixing, in which part of the major component (elastomer) is added first, next minor components, followed by the balance of the elastomer. As a rule of thumb, hard additives, reinforcing fillers, are added as early as possible in the mixing cycle to generate maximum shear, whereas diluents like plasticizers are added as late in the mixing cycle as possible. It is understood that any particular technique is more or less suitable for a given polymer system and thus has to be modified accordingly. For instance, in the case of a PVC NBR blend, the PVC powder can be fused first on a hot mill followed by blending with pre-masticated NBR, and this pre-blend dumped next into an internal mixer for incorporation of fillers, etc. For carbon black reinforced stock, the dumping temperatures range from 125-150[degrees]C, whereas those containing blowing agents are normally dumped below 110[degrees]C.

Extruders

Mixing of any millroom stock in single screw extruders is of dispersive and distributive nature, which ensures uniformity through randomization. They have several functions, such as warm-up and stock feeding into calenders, covering wire for insulation, manufacturing of flat ribbons and various shaped profiles, production of mandrel supported and heavy wall self-supporting tubing (automotive radiator hose), palletizing, etc. The typical hot fed single screw extruder is a well-known machine where the screw rotates in a tightly fitted barrel, thereby transmitting mechanical energy into the compound. The length of the screw is relatively short; 3-6 x D compared to the more recent cold fed extruders in which the screw length can reach 12-24 x D. Some practitioners advocate the entire elimination of warm up mills and a total switch to cold ribbon stock fed extruders. While this might be valid for mechanical goods based on synthetic rubbers, and especially EPDM, it entails certain compounding limitations for other end uses in which natural rubber is part of the recipe. Since fractured appearance and excessive die swell are not desirable, any formulation error or improper mixing have a way of showing up in the semi-finished, non-cured extrudate. This may explain a certain conservative approach to the retention of warm-up mills to reduce scrap.

Novel mixing concepts and systems

In the past two decades, scientists and technologists have made major strides for a better understanding of the mixing phenomena taking place in extruders, and especially in internal mixers. This activity was driven, and still is, not only by the quest for higher efficiency in each zone of the machine or each step of the mixing procedure in terms of intensive and distributive effects, but also through the development of polymer blends and alloys resulting in new materials for special applications. According to work done by Nakajima (ref. 12), these are treated as viscoelastic solids rather than viscous fluids. This implies a material behavior as transient instead of steady state. The viscoelastic properties at large elongational deformations and ultimate cured properties are considered fundamental in terms of practical usefulness. Instead of the laminar flow model based on steady state shear flow mechanism, a comminution and lamination model has been proposed. This is considered as representative for a large elastomer deformation, rupture and recovery during mixing. The above concept leads to measurements suggesting the potential for energy savings via process improvements. So far, this was confirmed by the use of warm cooling water giving better heat transfer and a new design configuration for the four-wing rotors as shown schematically in figure 3, which avoids energy wasting impinging streams.

[FIGURE 3 OMITTED]

As a follow-up to the above elongational flow concept, a recent study by Clarke and Petera (ref.13) was focused on improving rotor geometry in the context of filler disagglomeration during the dispersive mixing process. By using a rather simple compound based on SBR and 30 phr carbon black, it was possible to correlate the disagglomeration process via a first order differential equation with the rate constant proportional to the power density experienced by the compound during mixing. The reported result is a kinetic model in original finite element software capable of providing useful guidance for improving cross-sectional rotor geometry in regard to more efficient dispersive mixing. Given the widespread use of internal mixers and their high replacement cost, the first phase of operational developments was focused on a number of variables such as:

* Ram pressure: Mixing studies showed that, in general, high ram pressures offer advantages for high Mooney viscosity compounds. However, a closer investigation revealed that these high pressures were more effective in the initial stage of forcing the charge of raw materials into the mixing chamber, but lower pressures are more desirable during the actual mixing process. Furthermore, in the case of new compounds, excessive ram pressures may exceed the limit of power capability of an existing installation. This led to the development of a ram position indicator (RPI), a device fitted with transducers which allows ram pressure adjustments related to the batch size and mixing action.

* Chamber loading: Rubber bale size and temperature can negatively influence the mixing quality. If these parameters change from day to day, mixing consistency is hardly achievable. For best results, a 1/2 or 1/3 bale size, weighing 30-35 kg, should be used. For bagged minor components, a new device was installed which checks the weight and can also signal that the batch has received the charge. The optimum batch weight for a particular compound depends on the type and level of rubber, filler and plasticizers. Arbitrary fill factors or rules of thumb are no longer recommended, especially when ram pressure can be controlled.

* Cooling water temperature: This factor was investigated by measuring the coefficient of friction of various elastomers at different temperatures on metal surfaces similar to the mixer's chamber. It was found that water temperatures up to 30[degrees]C were more suitable for NR compounds, while for SBR, more efficient mixing occurs with cooling water at 40-50[degrees]C. This is in agreement with previous reported findings. In addition, the warmer cooling water prevents condensation on the walls of the mixing chamber.

* Rotor speed: Variable speed rotors ranging from 18-100 rpm have now become customary on all new internal mixers. This feature provides additional flexibility so that, depending upon the nature of the compound, higher productivity can be achieved via reducing the mixing cycle. For heat sensitive compounds, lower rotor speeds can ensure better mixing randomization during longer mixing cycles. It was also possible to prove that a batch size that is either too small or too large can reduce productivity.

* Rotor design: As mentioned earlier, dispersive mixing has been recognized as a crucial element for achieving uniform quality and batch-to-batch consistency. As a further practical confirmation, it was realized that after 8-9 years of heavy around-the-clock usage in a tire operation, dispersive mixing can be compromised to an unacceptable degree due to the deterioration of the rotor tip geometry. As a result, designers have increased their efforts to investigate in detail the rotor's geometric elements and their influence on the mixing process. Thus, distinctive effects can be caused by changing the features such as tip width, clearance, leading and trailing angles, rotor waist diameter, etc. These efforts led to equal speed rotors and the synchronous tangential rotor design.

As a follow-up development, the same manufacturer introduced a mixer with intermeshing rotors that promote a positive circulation of the material in the mixing chamber and facilitate distributive and dispersive mixing of larger batch weights without affecting the loss of material or mixed compound quality. This new internal mixer also exhibits a better thermal efficiency via high velocity cooling water circulation through the rotors and mixing chambers, and offers additional mechanical design features, as well as a PLC-based operational control system.

Borzenski (ref. 14) compared the features of the improved mixing equipment and clearly defined the areas of application. Thus, the worldwide proportional utilization of the intermeshing rotor mixers by the manufacturers of technical rubber goods is as high as 79%, and only 5% for tires; whereas the synchronous tangential rotor design is used by 57% of all tire plants and 23% of technical goods manufacturers. There is also an overlap zone indicating that both machines can process medium viscosity formulations and general rubber goods equally well. Among the deciding factors to favor one machine vs. the other are:

--the degree of detailed attention to mixer operating conditions;

--the nature of materials to be mixed;

--production rate requirements vs. mixer batch size.

As indicated by the quoted proportional utilizations, the higher production rate demands of the tire compounders are a deciding factor. Because of the internal mixer's high utilization rate, Borzenski and Valsamis (ref. 15) reported further mixing improvements via the wing function technology (WFT) rotors.

Not to be outdone, Nortey (ref. 16) reported interesting details on the new Coflow internal mixer.

This is, in fact, a hybrid tangential-intermeshing rotor machine with larger L/D rotor ratio, clockwise/counterclockwise material circulation, higher fill factor (82%), etc. It is also claimed that, by using lab and plant mixers of identical design, there are no scale-up problems for compounds based on NR and a broad variety of synthetic elastomers. Finally, another manufacturer (ref. 17) offers a lower cost alternative for improved dispersive mixing with a tilt discharge multi-purpose (plastics and rubber) internal mixer. It comes with longer 2W rotors (length ratio 1:1.75) and improved cooling via a larger solid cast steel jacket. However, it may require larger floor space for installation and possibly longer mixing cycles, since the rotor speed is fixed at 30 rpm for all four available units with increased batch capacity.

* Extruders have found their way into the rubber industry via intrusion from the plastics processing industries. This is valid for the modern single-screw and especially the twin-screw extruders. Their design has evolved to a high level of sophistication with capabilities to process rubber in powder form, or even reprocess particulate vulcanized rubber. This has set the stage for continuous rubber compounding.

The prevailing actual usage of extruders is in the downstream material handling of batches discharged from an internal mixer, thus replacing the traditional 84" sheeter mills. Large single-screw extruders can be used either as a discharge unit connected to a pelletizer or for direct sheeting via a roller head or sheet-forming die. Compared to the former two-roll mills the single-screw roller head extruders offer a safe, operator-free alternative with a predictable productivity and relative lower power consumption. Through the introduction of the self-feeding twin extruders for this discharge operation, the mechanical pushers can also be eliminated, while retaining all of the above-mentioned benefits.

Further sophistication via design upgrades has facilitated the direct high volume conversion of a batch form product into a high quality sheet from a connected two-roll precision calender. Cold-fed extruders gained broader acceptance primarily through their ability to effectively mix and homogenize a compound without using high shear rates. This was achieved by projecting pins into the barrel along the path of the screw so that the flow of the compound is continuously divided. Thus, the relative longer pin barrel extruder can process higher Mooney compounds without major problems. Further screw design improvements have also made their impact in this area via mixing by flow division and subsequent mechanical shear in a section between the screw and barrel. It is claimed that this system can achieve faster extrusion speeds with only minimum viscous heating. One of the more recent developments involves direct extrusion through a gear pump. Thus, Skibba and Thoma (ref. 18) reported that a single screw transfer mix extruder in which the die head is replaced by a gear pump represents a stepped-up technology in rubber extrusion. By using a gear pump for pressure build-up, less demand is placed on the extruder, which leads to an increase in throughput for comparative extruder sizes. As major advantages, it is claimed that the system offers process flexibility, reduced scrap and improved productivity.

Continuous processing has reached the feasibility stage through the introduction of co-rotating intermeshing twin extruders. According to Dreiblatt and Eise (ref. 19), this development was prompted by an increased demand for high performance polymers. The involved screw design features kneading block elements shown in figure 4.

[FIGURE 4 OMITTED]

Continuous processing and rubber compounding can be achieved through the Werner-Pfleiderer modular system featuring improved dispersion, batch-to-batch consistency and reduced heat exposure. It is claimed that all of these requirements for higher quality special applications can be met via the ZSK system represented schematically in figure 5.

[FIGURE 5 OMITTED]

Instruments for quality control and process optimization

The quality Of a finished vulcanized product is only as good as the attention to details given during the execution of test procedures for the quality control of incoming raw materials and processability characteristics of compounds during each stage of the manufacturing process. Furthermore, the bottom line of each enterprise is greatly dependent on the organizational consistency of controlling what is happening in the mixing department. This is important because compounding or mixing errors are at the core of any subsequent factory problems resulting in scrap, frustrations due to wasted efforts and unsustainable financial losses.

For many years, the Mooney rotational viscometer was the commonly accepted way to measure the approximate average molecular weight of incoming raw rubber, the viscosity of friction of natural rubber batches, as well as the interim and final mixing quality of a given rubber compound. Although its usefulness for fast approximative measurements ranging up to 80 MU is still valid, this method is less accurate at higher viscosities. There are situations when this can be overcome by running the test at higher temperatures, however, this may lead to scorchiness and erratic results once the curatives are included in the mix. One of the fundamental problems is that the Mooney viscosities are measured at rather low shear rates and stresses compared to those prevailing in internal mixers, extruders, etc. Although Mooney viscometers are still being used for relative scorch measurements, the data are not always reproducible, since the rotors are not heated, creating non-isothermal conditions.

All the above difficulties have drastically changed for the better with the introduction of various torque rheometers capable of not only better analyzing incoming elastomers and other raw materials, but also providing a better way to assess processability in every respect. Van Burskirk, Turetzky and Gunberg (ref. 20) were able to quantify high shear mixing by using a work per unit-volume parameter. Via this concept it was possible to define a power law relationship between work input and Mooney viscosity independent of mixer size and speed. As a follow-up, a correlation was established between the Brabender Plastograph and internal mixers of various sizes for several tire tread compounds. This was an important step for successful lab-plant scale-ups, since the work input, which is the integral over time of the mixing power curve, can be measured on lab as well as factory machines using a power integrator.

The Brabender Plasticorder MIM (mixing in miniature) and similar instruments by other manufacturers have proven their practical utility, as long as certain critical parameters (loading sequence, fill factor, etc.) are closely matched with factory conditions.

Recently a Plasti-Corder Digi System, designed especially for mixing applications, has become commercially available with a broad range of capabilities including cure rate analysis. Two software programs for mixing and statistical data analysis support the system, ASTM D 2443-83 describes in detail three test methods to determine the degree of carbon black dispersion in rubber: Visual inspection; agglomerate count; and microprocessor roughness measurement. According to Medalia (ref. 21 ) physical properties such as tensile strength, die swell-extrusion shrinkage and electrical resistivity are also important. A rather simple way to assess carbon black dispersion is to insulate a wire via a lab extruder and check the insulation's resistivity. Any poor dispersion or presence of agglomerates will show up as fluctuations in the above data.

Extrudability testing of unvulcanized compounds has been, and is still, performed according to the Garvey Die Test described in ASTM D 2330-83. The test involves the use of a variable speed lab extruder--Brabender Plasticorder--to extrude a profile through a specially designed die having various shapes. After optimizing the test parameters on a standard compound, an extrudate is prepared with the test sample and compared visually. Although qualitative, this test has a proven practical value. More quantitative data can be obtained via the die swell tester described by Plaskin (ref. 22). This is a semiautomatic test involving the calculation of the die swell--ratio of extrudate cross-section to die cross-section. A new arrival is Winext, the custom-designed Brabender extrusion program for the Inteli-Torque Plasti-Corder. Once the extrusion trial is started, the software provides teal time process information on process torque, extruder speed, melt temperature and melt pressure. It is an excellent R&D tool.

The evolution of curemeters is fully described by Sullivan and Wise (ref. 23). It should be mentioned briefly that the oscillating disc rheometer (ODR) described in ASTM D 2084 has gained wide acceptance by compounders and technologists as a production control test for three major reasons:

* It can detect minor changes prompted by formula adjustments or different processing conditions;

* by operating at a temperature range of 175-200[degrees]C, a test can be completed within five minutes, which coincides with a typical internal mixer mixing cycle;

* via current software, the ASTM cure rate index can be calculated as follows:

Cure index = 100/(cure time-scorch time)

A more recent development, the MDR (moving die rheometer) is a pressurized cavity rotorless curemeter capable of achieving isothermic operating conditions. In addition to faster response, due to a thin sample cross-section, this instrument has facilitated automatic sample loading and unloading, as well as digital data acquisition. With comparative results vs. the ODR, this new instrument is especially useful for high volume QC testing.

Interim in-process testing

This should be part of any organized QC procedures. These are the so-called in-house test methods developed over the years in many factories to accommodate the pertinent finished product requirements. They are generally simple, fast and do not require highly skilled lab technicians or expensive equipment specified in standardized procedures. For example, compounding errors can be detected by measuring the densities of unvulcanized compounds. In the case of adhesives, a TI (thixotropic index) can be measured on any batch prior to and after final filler addition by running Brookfield viscosities at differential spindle speeds. In a similar way, a dynamic stress rheometer (DSR) (ref. 22) can be used for testing the acceptability of a stock at various stages of processing. This is done by determining the integral torque values at a fixed set of conditions for each stock being processed. The rejection or acceptance criteria are based on internal specification limits established with standard compounds. Whenever possible, it has been found useful and motivating to train the factory personnel (batch tenders, mill and mixer operators) to carry out these interim in-house test methods. By getting a feel of what they produce, they will try harder to become part of the desirable top quality team effort.

For operational control of internal mixers, power charts should be supplemented by torque-work recordings. This, as well as digital data recording and acquisition, is facilitated by control panels on all new and modernized equipment.

Trends and opportunities

For economic, safety and environmental reasons, many companies have either partially or fully outsourced their mixing operations to strategically located custom mixing enterprises. This cooperative trend can be mutually beneficial if the capabilities are properly coordinated so that stringent spec or performance test requirements, especially by automotive end users, can be met consistently. For those who prefer to do their own mixing, there are comprehensive software packages tailored to control the entire manufacturing process related to the mixing operation. The potential of continuous compounding, via special twin extruder design, is slowly being recognized in spite of some doubts and roadblocks, such as high cost of free flowing elastomeric raw materials and clean-up down time during compound changes. Among the benefits are end product uniformity, high level of automation, floor space savings, fine tuning for special end use requirements, flexibility via a modular concept and a window for cost/benefit opportunities. Another related area being closely watched is the successful injection molding of selected high volume products (hockey pucks, etc.).

Conclusions

The downstream mill room operation has witnessed a broad replacement of the 84" sheeter mill by a twin screw extruder which can be connected either to a pelletizer or used for direct sheeting via a high precision two-roll calender. Modern internal mixers featuring controlled ram pressure, equal speed synchronous tangential rotor design, wing function rotor technology, etc., have gained wide acceptance in the high volume tire industry. In contrast, manufacturers of technical goods have switched preferentially to the newer mixers featuring intermeshing rotors, better chamber material circulation and improved dispersive mixing. In spite of the above application-driven preferences, there is a broad area of overlapping potential in which either machine can achieve a satisfactory compound mixing quality.

The actual choice depends on economic and other factors such as already available mixers, output, product quality requirements, vendors experience and technical support capabilities, etc. The acceptance of design features from plastics processing machinery is a positive development. Partial automation via stockblenders has justified the retention of two roll mills, which can still provide unmatched flexibility and top quality dispersive mixing. Ultimate profitability is directly related to the organization and operational consistency of the mixing department.

Table 1 - operational comparison between mill
and internal mixing

                                        Mill      Internal mixer

Blade size                           16" x 24"       #9 or 11
Batch size (lb.)                      30 - 50        300-350
Motor (hp)                            70 - 75        300-600
Temp. [degrees]C                      70 - 75        110-125
Mixing speed (slow roll)               24 rpm      11 *-30 rpm
Friction (slow to fast roll/rotor)     1:1.4          1:1.25
Total mixing cycle (min.)            Approx. 17       10 **
Productivity (output/man hr.)            -              +
Safety (dust, etc.)                      -              +
Space and energy requirement             -              +
High volume mfg. cost                    -              +
Primary type of mixing
  Incorporation                          -              +
  Dispersive                             +              -

* Speed quoted for std. recipe

** Mixing cycle for SBR-carbon black masterbatched
compounds 4-6 min.

References

(1.) S.L. Rosen, "Fundamental principles of polymeric materials," John Wiley & Sons, Inc., NY 1982.

(2.) G.M. Brotow and A.G. Sears, NR Technology, 13 (4), 73 (1982).

(3.) A.W. Blum, Rubber World, 225 (4), 35 (2002).

(4.) F. Rodriguez, "Principles of polymer systems," p. 263, McGraw Hill, NY, 1982.

(5.) H. Palmgren, Eur. Rubber J., 15 (6), 30 (1974).

(6.) J.M. Funt, "Mixing of rubbers," Rapra, Shawbury, U.K. (1977).

(7.) N. Tokita and I. Plisken, Rubber Chem. Technnol., 46, 1,166 (1973).

(8.) N. Nakajima, Polymer Engng. Science 19, 215 (1979) and Rubber Chem. Technol., 54, 266 (1981).

(9.) A.M. Gessler, W.M. Hess and A.I. Medalia, Plast. & Rubber Process., 1 (37), 141 (1978).

(10.) I. Manas-Zloczower and Z. Tadmor, "Mixing and compounding of polymers," Hanser, NY (1994).

(11.) M. Morton, "Rubber technology," 3rd Ed., Chapman and Hall (1995).

(12.) N. Nakajima, "Energy measures of efficient mixing," ACS Rubber Division meeting, Cleveland, OH (1981).

(13.) J. Clarke and J. Petera, Rubber Chem. Technol., 72 (3), 807 (1999).

(14.) F.J. Borzenski, "Mixing technology and machinery," Ferris State Univ. (2001).

(15.) F.J. Borzenski and R. Valsamis, Rubber World, 226 (4), 40 (2002).

(16.) C. Nortey, Rubber World, 226 (4), 32 (2002).

(17.) Kobelco-Bolling brochure (2002).

(18.) O. Skibba and O. Thoma, Rubber World, 224 (6), 36 (2001).

(19.) A. Dreiblatt and K. Eise, p. 241 in C. Rauwendaal, "Mixing in polymers and processing," M. Decker (1991).

(20.) Van Burskirk, S. Turetzky and J. Gunberg, Rubber Chem. Technol., 48, 577 (1975) and 49 (1976).

(21.) A.I. Medalia, "Measurement of carbon black dispersion," ACS Rubber Division meeting, Cleveland, OH (1981),

(22.) I. Pleskin, Rubber Chem. Technol., 46, 1,218 (1973).

(23.) A.B. Sullivan and R.W. Wise, "Process and vulcanization test," p. 105 in M. Morton, "Rubber technology," 3rd Ed., Chapman and Hall (1995).

(24.) S.R. Moghe, Rubber Chem. Technol., 49, 247 (1976).