Hide and peak - polymer recyclate analysis

How differential scanning calorimetry and intelligent software can improve polymer recyclate analysis.

In an era of growing environmental responsibility and regulatory pressure, the drive to replace virgin plastics with recycled polymers has never been stronger. Yet, recycling polymers isn’t as straightforward as melting down old plastic and reshaping it. Accurately identifying the materials in a recyclate stream to ensure consistent quality, safety and performance is a significant challenge.

Recyclates often consist of heterogeneous mixtures – different polymers, additives, contaminants, or even degradation products. Unlike virgin materials with standardised compositions and well-documented properties, recyclates are often unpredictable. A batch of recycled polyethylene (PE) might contain traces of polypropylene (PP), ethylene-vinyl acetate, PET, or even barrier layers and fillers – depending on its source. 

Furthermore, additives like stabilisers, flame retardants, or plasticisers – while essential in the original application – may interfere with the recyclate’s intended reuse. Degradation products from multiple lifecycles only add to the challenge.

These compositions vary not only from batch to batch, but also based on their history, origin and previous use. These unknowns significantly complicate downstream processing and 
product performance.

For compounders, converters and original equipment manufacturers (OEMs) aiming to replace virgin polymers with recycled alternatives, knowing exactly what’s inside a material is essential. 

Mismatched or unknown compositions can lead to defects during processing, unstable mechanical properties, or even safety hazards in the final product. The ability to ‘see’ and distinguish what’s in the mix makes advanced analytical tools so critical. Identifying these polymers quickly and reliably is the first step towards turning waste into high-quality raw material.

Taking fingerprints

Differential scanning calorimetry (DSC) is one of the most trusted tools for thermal analysis. It measures how a material’s heat flow changes as it is heated or cooled, revealing critical data such as melting points, glass transitions and crystallisation behaviour. For polymers, these thermal transitions act like a fingerprint for identifying polymer types and assessing purity.

However, when it comes to complex mixtures – like those found in recycled plastics – interpretation becomes significantly more difficult. Overlapping peaks and subtle thermal events can obscure meaningful insights. 

Many recyclers and compounders still rely on visual inspection, bulk melt flow index (MFI) measurements or density checks. While these methods are quick and low-cost, they rarely provide detailed insight into the actual composition. 

For example, MFI can be affected by a range of factors beyond polymer type, including molecular weight distribution and degradation – making it more of a symptom than a diagnosis.

Spectroscopic techniques like fourier transform infrared or near-infrared spectroscopy are helpful for surface identification, but struggle with blends, additives, or bulk characterisation. DSC, especially when combined with smart software, provides a clearer path.

Software tools such as NETZSCH’s Identify and Peak Separation assist in comparing thermal fingerprints against extensive databases and distinguishing overlapping signals, helping to better understand polymer blends. Together, these tools enable a deeper, faster and more practical understanding of what recycled materials are truly made of.

Further developments in advanced data analysis will continue to refine and simplify the process. 

Building an identity

The Identify software acts like a thermal search engine. It compares experimental DSC curves with a growing database of reference materials. Analysts can quickly determine the most likely candidates for a sample based on peak positions, enthalpies and overall curve shape.

This is especially useful in recycling, where mixtures of PP, PE, polystyrene (PS) and engineering plastics may coexist. Instead of guessing based on one or two broad peaks, Identify provides a data-driven basis for comparison.

It is worth noting polymers don’t just melt at single, sharp temperatures. Depending on crystallinity, additives, molecular weight, or processing history, even a ‘pure’ polymer can show broad, overlapping thermal transitions. 

This is compounded in recyclates, where multiple polymer types co-exist and interact. For example, high-density polyethylene (HDPE) and PP can show overlapping melting peaks in the 120-160°C range, but these differ in shape and enthalpy. 

Additives can mask or shift peaks, and similar compounds like PP and PE often overlap, making it difficult to quantify content by eye. 

The ability to mathematically resolve these transitions, rather than relying on expert interpretation, is key to accurate analysis.

The Peak Separation tool uses mathematical deconvolution to distinguish individual peaks, even if they blend together in the raw data.

The result is a clearer picture of the polymer blend composition, including estimated weight fractions. This kind of resolution helps manufacturers determine the suitability of a recyclate for specific applications and informs whether further purification is required.

Take the example of a packaging manufacturer evaluating a batch of post-consumer recyclate. Standard tests like MFI or density checks might not reveal hidden incompatibilities. 

Using DSC, supported by NETZSCH’s Identity and Peak Separation, two overlapping melting peaks in the second heating cycle are deconvoluted into individual contributions from PP and HDPE, revealing a blend of 80/20, respectively. This insight helps explain deviations in processing or mechanical performance and guides decisions about further purification or blending.

Another common case involves engineering plastics such as polyamides. When switching from virgin PA6 to recyclate, unexpected melting or crystallisation may occur. 

A subtle second peak at around 260°C in the DSC curve often indicates the presence of PA66 – a structurally similar polymer that significantly impacts crystallisation kinetics 
and process behaviour. 

Depending on the degree of mixing, prior thermal history and mixing ratio, multiple heating cycles may show the evolution from two distinct melting points towards a broader, merged signal due to co-crystallisation. 

One curve at a time

When a sample contains multiple polymers with overlapping melting ranges after the second heating curve is recorded, the user can activate the Identify function, which compares the measured curve with a library of reference materials. In our example (below), the software matches the first peak with HDPE and the second with PP – confirming the identity of the two constituents.

Next, the analyst applies the Peak Separation tool. This feature mathematically deconvolutes the measured heat flow curve into individual peak contributions – even when those peaks overlap significantly. 

The software uses fitting algorithms to assign each portion of the signal to a specific polymer. In this case, it separates the HDPE and PP melting transitions and quantifies the respective enthalpy for each.

In the example shown in the bottom image, the HDPE portion of the curve has a melting enthalpy (area under the peak) of 44J/g. To estimate the mass fraction of HDPE in the blend, we need reference data. The melting enthalpy of 100% crystalline HDPE is approximately 293J/g and HDPE has on average a 75% crystallinity.

Using this information, we calculate the HDPE mass fraction as:

This simple calculation provides a quantitative estimate of the HDPE content in the recyclate without needing elaborate separation or spectroscopy. Similar estimates can be made for other polymers, provided that reliable reference enthalpies and typical crystallinity values are known. 

By combining thermal fingerprinting with mathematical curve fitting, DSC becomes more than a qualitative screening tool. It enables confident identification and quantitative insight, supporting key decisions in recyclate selection, formulation adjustment and quality control.

This kind of software-supported analysis supports both quality assurance and regulatory compliance.

This offers value across the entire plastics lifecycle as material buyers gain confidence in incoming recyclate quality, compounders can fine-tune formulations and blends, OEMs can validate recycled content in finished goods, and R&D teams can develop new sustainable recipes.

Knowing your material

As legislative frameworks like the European Green Deal or Extended Producer Responsibility continue to evolve, accurate and scaleable analysis tools become essential.

They allow quality control and acceptance testing, recyclate supplier benchmarking, R&D of recycled formulations and obviously certification of recycled content.

The common thread is knowing your material. From quick assessments to full material breakdowns, software-supported analysis bridges the gap between complex thermal data and practical decision-making. In doing so, it helps turn plastic waste into valuable raw materials – one curve at a time. 

And although tools like NETZSCH’s Identify and Peak Separation still require some interpretation and a certain level of expertise, that’s about to change.

Soon software solutions will apply machine learning models trained on thousands of DSC measurements to instantly interpret thermal data, separate overlapping peaks automatically and output confident, quantifiable results in seconds.

Users will simply upload their data and receive a clear, expert-level interpretation – with transparency about the reliability of each result.

This shift will help democratise polymer analysis, enabling technicians, buyers, and recyclers alike to work with recyclates more effectively regardless of their background or training.