Mailbag.

Dear editor:

I would like to respond to the letter written by Dr. Bill Watson published in your July 2001 edition, a diatribe against continued use of the Mooney viscometer as a standard test instrument by the rubber industry, and against its developer, Dr. Melvin Mooney. Dr. Watson's stated intent for writing this letter was to stir the pot and solicit comments. My intent is not only to continue stirring the pot, but to add more fuel to the fire.

Values for viscous and elastic behavior of rubber, as measured by your newly developed couette rheometer, are simply too elementary, my dear Watson, to fully predict processability of a given rubber compound in all of the manufacturing processes used within the industry. Do not be discouraged. My statement holds equally true for test results obtained by using other types of rheological instruments. Rheological values may be helpful for deciding what is, or what is not, contributing to a processing problem, but are not always the major source of the problem.

Some years ago I presented an integrated approach to efficient polymer processing which includes good processability (ref. 1). It is based on three parameter sets for which data must be obtained - polymer properties, equipment design and operating variables. Within these three parameter sets, a minimum of seven different types of information, or data, are necessary. These include: viscosity as a function of rate and temperature, plus scorch and curing kinetics as a function of temperature for the rubber compound; physical size, shape and power availability for a given piece of processing equipment, along with a precise description and size and shape of all tooling through which the rubber will flow; and operating variables for each type of equipment and tooling which includes all temperatures, pressures and flow rates the rubber compound will experience. Based on this approach, it is obvious that a model which describes only viscosity and elasticity cannot, by itself, solve a problem of processability.

Dr. Watson asks the question, "Why does the rubber industry continue to operate the Mooney viscometer at a temperature of 100 [degrees] C when many internal mixers operate at temperatures around 160 [degrees] C?" This is a valid question so long as one is testing a sample of raw rubber or a non-productive compound. On the other hand, why not test the sample at 25 [degrees] C? Most raw rubber and rubber compounds are about that temperature when first added to processing equipment. The most puzzling problems I have encountered as a troubleshooter, the ones for which viscosity, scorch and curing kinetics provide no answers, have usually involved difficulty in uniform feeding of ambient temperature compound to a cold-feed extruder, or to a mill or to an internal mixer. I believe this to be a matter of the surface frictional characteristics of a compound as compared to the internal frictional characteristics associated with resistance to flow. By comparison, I have found that problems occurring downstream in the operation were usually easier to solve.

We are a strange industry. We carefully roughen surfaces in the test chambers of many rheometers to prevent slippage of the test specimen so that we can get a more accurate measure of viscosity. We also add processing aids to the compound to reduce its viscosity and/or promote slippage along the hot, metal surfaces of processing equipment to minimize heat build-up during processing which can lead to scorching sooner than expected. We prepare a masterbatch one day, convert it to a productive batch and process it the next day. Because we operate on a five day a week schedule, the masterbatch prepared on Friday gets its final mixing the following Monday. There is a problem getting it to feed properly to the internal mixer. Why? Surface friction changed as the compound stiffened over the weekend. Can this be measured by a rheometer? No! The robber softens again during preheating of the sample to the temperature set for testing. Using a capillary rheometer, I once tested twenty different samples of productive stock which represented both good and poor feeding characteristics to a cold-feed extruder. Each sample represented an admixture created from four master-batches which was the producer's attempt at achieving more uniform feed. No significant differences in flow characteristics were determined by the rheometer. All twenty samples were judged equal based on precision of the test; yet another example of failure of the rheometer to distinguish between surface and internal friction. Attempts have been made to measure slippage of robber against metal by employing smooth-surfaced rotors in the rheometer. While some of these rheometers are capable of varying pressure within the test chamber, I have yet to read about test results, using these instruments, that could be correlated with slippage of feed stock in various types of manufacturing equipment. Until such time as a satisfactory test is developed, I guess we will have to continue putting the blame on one rheometer or another for their failure to address this problem.

I am retired. However, if I were to re-enter the robber business and had a say in what test equipment should be purchased, giving due consideration to budget constraints, my first choice would be a modern Mooney viscometer. Why? Because this rheometer provides the best reproducibility of test results of any type of rheometer built to date, both intra-laboratory and inter-laboratory, which is very important when dealing with both suppliers and customers. That Mooney viscometer would also be equipped to measure stress relaxation. The second instrument would be a rotorless, or moving-die curemeter. Because I am willing to perform tests at different and non-standard temperatures, I can obtain an excellent index of the viscosity and elasticity characteristics of both raw and compounded robbers rapidly, indices helpful for operating and, hopefully, troubleshooting the manufacturing processes. I would certainly not purchase your couette rheometer to obtain these data because I firmly believe that Dr. Mooney was correct when he made his decision back in the 1930s not to use that type of rheometer. It still is a research instrument, too complex in operation, and requires too long a testing time to be practical for control operations.

George P. Colbert, retired

(1.) J.D. Byam and G.P. Colbert, Plastics and Rubber Processing and Applications, 5, 95 (1980).