Maximizing Polymer Recycling Efficiency: Real-Time Viscosity Modeling

Competence Center CHASE
Jun 8
3 min read

To make advanced polymer and especially polyethylen recycling scalable and profitable, processing plants must cut energy costs and prevent machine downtime. Predicting exactly how recycled plastic flows is the key to achieving this optimization.

Mixing solvents or gases into recycled plastic melts lowers viscosity and reduces machine pressure. While controlling this process in real-time was previously impossible due to highly variable waste streams, our study has validated a practical mathematical framework. This solution enables real-time factory automation—allowing recycling plants to run smarter lines that minimize equipment wear, prevent material degradation, and maximize production throughput at significantly lower energy costs.

Achieving Precision in Plastic Recycling

To automate a recycling plant using modern Industry 4.0 software—like a "Digital Twin"—the control computer must predict how the plastic will flow at any given second. If the prediction is wrong, filters can clog, pressures can spike to dangerous levels, and production must be stopped for manual cleaning.

Up until now, creating these vital flow predictions introduced severe operational hurdles:

The Mixed Waste Dilemma: Post-consumer recyclates (PCR) change from batch to batch, making standard settings useless.
The Missing Data Wall: High-precision academic models require complex molecular weight distribution data. Testing every batch in a laboratory is far too slow and expensive for a continuous recycling process.
The Risk of Blind Spots: Solvents act as lubricants, creating a high risk of "wall slip" inside the machine, which can distort sensor data and lead to false process adjustments.

To eliminate this guesswork, the research team analyzed real plastic melts (high-density PE-HD and recycled low-density PE-LD PCR) under harsh, industry-matching conditions (200°C to 240°C) using an in-line slit-die rheometer mounted directly on a single-screw extruder.

Validating Practical Models for Digital Twins

To find the fastest, most cost-effective way for recycling plants to calculate melt behavior, the study systematically compared three modeling approaches:

The Pure Physics Approach (Kelley–Bueche EOS): This method tries to predict behavior from fundamental thermodynamic principles. While scientifically valuable, the study proved that it consistently underestimates how much the solvents actually thin out the plastic. Relying on it blindly would cause engineers to miscalculate machine pressures.
The Molecular Approach (Schausberger Model): This method achieved the highest physical precision (with an error margin under 3%). However, because it completely depends on hard-to-get molecular mass data, it is only practical for premium virgin materials—not for variable plastic waste.
The Semiempirical Approach (The Winner for Industrial Automation): This compact, four-parameter model ignores complex molecular data. Instead, it uses a simplified mathematical term calibrated by easily accessible process data.

Actionable Results for Your Bottom Line

The practical evaluation delivered clear, actionable insights that directly benefit machine manufacturers and recycling plant operators:

Proven Data Accuracy Without the Cost: Specialized parallel-plate testing confirmed the complete absence of wall slip. This proves that the measured viscosity drops are pure material responses, meaning the data used to program the machines is 100% reliable.
Robust Control Over Variable Waste: The simple four-parameter semiempirical model achieved an exceptional accuracy level, keeping relative errors below 5% across all tested systems. Because it does not require complex molecular lab data, it can handle highly fluctuating post-consumer waste streams without breaking down.
Immediate Digital Integration: Because the winning model uses a simplified mathematical formulation, it can be programmed directly into real-time control software and Digital Twins. The computer can now adjust machine settings on the fly as the recycling feedstock changes.

Real-Time Process Optimization and Machinery Sizing

By providing a ready-to-use computational framework, this research allows the polymer industry to transition from reactive troubleshooting to proactive process control:

Engineers can precisely size pumps, filters, and extrusion dies because they finally know exactly how the diluted melt will behave under pressure.
Plant Operators can maximize the use of low-cost post-consumer recyclates while maintaining stable extrusion, efficient cell expansion in thermoplastic foaming, and zero unplanned downtime.