Drying Before Extrusion: Hours Are Not a Specification

Why some polymers have to be dried

Polyolefins do not absorb water. A polyethylene or polypropylene pellet has moisture on its surface and essentially none inside, and a vented extruder handles it without a dryer. This is why a plant that runs mostly polyolefins can operate for years without a drying problem.

Polyamides, polycarbonate, PET, PBT, ABS and TPU are hygroscopic. Water diffuses into the pellet and sits between the polymer chains, where it acts as a plasticiser and, at processing temperature, as a reactant. A polyamide at 280 degrees Celsius with 0.2 percent moisture will hydrolyse, and the molecular weight loss is permanent.

The visible symptom is silver streaking, which is steam expanding out of the melt as it exits the die. The invisible symptom is a drop in mechanical properties that shows up in a test months later, and a lot of warranty claims have been traced to a dryer that was switched off overnight.

The important point is that drying is a diffusion process, so it has a rate, a capacity and a floor. All three are polymer specific and none of them are captured by a number of hours.

The curve and its floor

The chart on the left shows moisture content against drying time for PA6, PC and PET in a desiccant dryer at 120 degrees Celsius with a minus 40 degree dew point. PA6 starts at 0.32 percent, falls to 0.09 by two hours and levels at 0.02 after six. PC starts at 0.09 and levels at 0.010. PET starts at 0.05 and levels at 0.007.

The floor is not zero and it is not a limitation of the dryer. It is the equilibrium moisture content at that temperature and that dew point. To go lower you have to raise the temperature or lower the dew point, and both have limits: too hot and the pellets stick or degrade, too dry and you are paying for regeneration energy that buys nothing.

This is why four hours and eight hours are often the same result. Once the curve has flattened, extra time in the hopper only risks re-contamination from a poorly sealed system.

It is also why a dryer that is working perfectly can still deliver wet resin. If the dew point is minus 10 instead of minus 40, the floor is higher and the resin never reaches the specification, no matter how long it stays.

What residual moisture does

The right hand panel shows silver streak rejection against residual moisture. At 150 ppm the rejection is 0.4 percent, at 300 ppm 0.9, at 600 ppm 2.6, at 1,200 ppm 7.1 and at 2,400 ppm 15.8. The curve is gentle below 600 and steep above it, which is why a specification at 600 ppm is a sensible line.

Below the visible threshold the damage is still happening. Hydrolysis reduces molecular weight, which reduces impact strength and elongation. A part made from resin at 900 ppm will look perfect and fail a drop test that the same part passed last year.

The effect compounds if the resin is reground, because regrind has more surface area and picks up moisture faster, and because it has already been through one heat history.

So the specification should be on moisture content, measured, not on drying time. A moisture analyser costs less than a single rejected shipment.

Table 1 — Where moisture starts to cost money.

Setting a drying regime

Start from the resin supplier data sheet, which gives a drying temperature, a time and a target moisture. Treat the time as a minimum and the moisture as the specification. Then verify with a measurement on your own dryer, because the data sheet assumes a dryer that is working to specification.

Check the dew point, not just the temperature. A desiccant dryer with saturated desiccant will hold temperature perfectly and deliver damp air, and the resin will never dry. Dew point should be below minus 20 degrees Celsius for most engineering resins and below minus 40 for polyamide and PET.

Check the airflow. A dryer that is full to the top has less airflow per kilogram of resin than one that is half full, and the drying rate falls accordingly. The hopper capacity is a maximum, not a target.

Check the residence time in the hopper against the consumption rate. If the machine consumes 20 kilograms an hour and the hopper holds 100 kilograms, the residence time is five hours and the drying is fine. If the machine consumes 60, the residence time is under two hours and the resin is not dry no matter what the controller says.

Table 2 — Starting points, to be verified on your own dryer.

The five ways a dryer fails quietly

Saturated desiccant is the first. The dryer holds temperature, the controller shows no alarm, and the dew point climbs from minus 40 to minus 5 over a week. Regeneration cycles that are too short or a blocked exhaust cause it.

A leaking hopper seal is the second. A dryer running at minus 40 will pull ambient air through any gap, and the moisture in that air is far more than the resin needs to fail.

A blocked process air filter is the third. Flow falls, drying slows, and the temperature reading is unaffected because the heater is still cycling.

Overfilling the hopper is the fourth. Residence time falls below the drying time and the resin at the bottom is dry while the resin at the top is not, so the problem is intermittent and looks like a machine fault.

And running the dryer only when the machine runs is the fifth. A hopper that cools overnight draws moist air in as it contracts, and the first hour of the next shift runs wet resin.

Verifying that the resin is dry

Use a moisture analyser, either a loss on drying balance or a Karl Fischer titration. The balance is fast enough for production and accurate enough for the 0.02 to 0.1 percent range that matters. Take the sample from the bottom of the hopper, not the top.

Log the dew point daily. It is the single most informative number about dryer health and it is almost never recorded.

Trend the silver streak rate as a process indicator. A rise that is not explained by a resin change is a dryer problem, and it will show there before it shows in a mechanical test.

Finally, verify after any change: a new resin lot, a new dryer, a maintenance intervention. Drying regimes are not transferable between dryers, because airflow and dew point differ.