Need help understanding impact of temperature and content of water/glycol on strength (crush, attrition) of methocel (methyl cellulose) extrudates.

In a given process, methocel is mixed at 150oF at 2wt% with a sand-water mixture (50:50) and 2% glycol. The mixture is then dried to extrudable moisture (20-30%) and extruded into spaghetti-like pills.

Recently the pill strength has been poor, with high level of dusting.

Is the temperature of dispersion too low (150oF)?

Does methocel need to ‘hydrate’ to provide strength?

Does the mixture need residence time and/or low temperature (<100oF) to hydrate? What is the ideal temperature?

Does the methocel loose strength if it dehydrates? Is the process reversible?

At what temperature does methocel dehydration take place?

Does high level of glycol prevent methocel hydration?

Can methocel be extruded with glycol only and still provide strength?

Does presence of certain contaminants (electrolytes, organic acids) prevent hydration?

What is the ideal temperature of extrusion?

Thank you!


For the customs of ChemSE, the question is rather broad; perhaps even too broad to provide you specific compositions, times and temperatures to improve your process of producing extrudates.

On the other hand, you already collected a number of plausible parameters influencing the target property, which you named as «pill strength». As general picture, you are heading to a systematic analysis -- preferably by design of experiments.

  • Aim to reformulate «pill strength» by a quality you may quantify easily like a scale, e.g. temperature, mass. This could be the Young modulus, which may be defined for either compression, or traction. For one batch of your extrudates, determine this «monitor property», to determine how well this property may be determined (arithmetical mean value $\mu$, standard deviation $\sigma$). Determine in advance where you like to see your process -- sometimes, it is not a single property describing your optimum.

    Assuming the Young modulus were the sole criterion to describe the optimum, you may determine in advance how many experiments $N$ per factor level are to be performed to recognize a relevant effect $\Delta\mu$ with a known standard deviation $\sigma$ by $$ N \approx 60 \left(\frac{\sigma}{\Delta\mu}\right)^2$$ and with a screening with $m$ combinations in a screening, each of them would be $n$ times realized $$ n = \frac{N}{m} $$

  • Because of the number of parameters already compiled, it is useful to screen their influence on «pill strength» in a two-level factor experiment. Think of them like a light switch, either «low level / off», or «high level / on». This applies to (i) temperature of dispersion, (ii) residence time and temperature in the extruder, (iii) high / low level of glycol perhaps preventing the methocel hydration. Run batches of extrusion with these parameter levels, and determine «pill strength» accordingly.

    This allows you to determine which of the parameters influence most the target property, and subsequently focus the optimization on a lesser number of them.

There are many publications about this large subject, both general textbooks (e.g., Box, Hunter, Hunter), in chemistry (e.g., ref, ref, ref, or ref), or entries on SECrossValidated (e.g., here).


Not the answer you're looking for? Browse other questions tagged or ask your own question.