IR Spectroscopy as a Diagnostic Tool in the Recycling Process and Evaluation of Recycled Polymeric Materials

1. Introduction

The environmental sustainability of plastic object production can be significantly improved by developing recycling strategies that ensure reliable assessment of the quality and reproducibility of the recycled polymeric material [1,2,3].

Chemical recycling, which involves depolymerizing the polymeric component to obtain small molecules (monomers) that can be reused as raw materials in subsequent polymer synthesis processes, is potentially the most reliable route [4,5]. In principle, it can yield a final product (the polymer) that is identical to the one obtained from conventional sources. However, the complexity and cost of chemical processes have so far made mechanical/thermal recycling the most widely adopted strategy for polymer materials [3,4,5].

Ideally, a thermoplastic polymer can be melted and reformed an infinite number of times at relatively low temperatures, thus with limited energy consumption, without losing its physicochemical properties.

This unique property makes recycling feasible: plastic waste—once sorted by polymer type—is shredded into fragments, which are then melted and extruded to produce pellets ready for market use as secondary raw materials for manufacturing new plastic products. Although this process may appear simple when described schematically, each of its steps has critical aspects that strongly impact the quality of the final product. We can identify: (i) Issues related to the intrinsic characteristics of the plastic waste, and, (ii) issues related to the recycling process itself.

Waste material-related issues concern the accuracy of polymer sorting, and the presence of contaminants or foreign fragments that were not removed. These problems are especially common in post-consumer plastic waste, while post-industrial waste sources typically allow for predefined selection of the input polymer type.

It is also essential to recognize that polymers in finished products are always used in formulations. Part of the formulation depends on the raw material itself (virgin polymer pellets), while another part is adjusted during processing to meet the functional requirements of the final product. Common additives include: inorganic fillers (e.g., glass fibers, calcium carbonate, talc, carbon black), stabilizers, antioxidants, flame retardants, and color pigments. The same polymer may be compounded with different additives and filler concentrations, depending on the needs of the converter [6]

Another critical aspect is that plastic waste is aged material, and the polymer may be partially degraded due to wear and environmental exposure. Typically, this leads to oxidation phenomena, resulting in lower molecular weights and a decrease or loss of mechanical properties.

Process-related issues mostly stem from the inevitable partial thermal degradation during processing, which makes it necessary to add additional stabilizers (typically antioxidants) during the recycling process. The inherent limitations of mechanical/thermal recycling affect the properties of the recycled polymer in the following ways:

• The recycled material is different from virgin polymer: while it may be similar in chemical characteristics and performance, it is effectively a new material.
• Recycled polymers can vary from batch to batch, due to inconsistent and hard-to-control input materials. In some cases, inhomogeneities may be present even from pellet to pellet, or within a single pellet (especially if foreign inclusions are present, such as metal flakes).

To ensure proper use of secondary raw materials obtained from plastic waste recycling, it is essential to have a comprehensive understanding of their physicochemical properties [3]. This enables comparisons with the reference virgin polymer and with other recycled polymers available on the market.

In this article, we will discuss the potential of IR spectroscopy in this context, explicitly addressing the specific contributions in light of the aforementioned challenges. In particular, we will focus on methodological aspects and, through selected examples, propose guidelines for critical interpretation and the use of spectroscopic markers that are not typically considered in this field. The discussion includes examples from the spectroscopic analysis of commercially sourced recycled polymer pellets, as well as reference spectra from virgin polymers. A case study on the quantitative determination of polypropylene content —which is frequently present in recycled polyethylene samples—will also be presented.

2. Materials and Methods

Several commercial polymer samples were analyzed, including high-density polyethylene (HDPE), low-density polyethylene (LDPE), isotactic polypropylene (PP), polyamide (PA), polystyrene (PS), polyvinyl chloride (PVC), polyethylene terephthalate (PET), Carilon^®, and recycled HDPE pellets (R-HDPE) from various manufacturers. For the construction of the PP/PE calibration curve, isotactic polypropylene and HDPE purchased from Aldrich were used as reference materials.

Transmission-mode infrared (IR) spectra were acquired on thin-film samples using a Nicolet 6700 Fourier-transform infrared (FTIR) spectrometer over the range of 400–6500 cm⁻¹, with 128 scans and a spectral resolution of 4 cm⁻¹. Attenuated Total Reflectance (ATR) spectra were recorded using a Thermo-Electron Continuµm IR microscope coupled to the same Nicolet 6700 FTIR instrument, equipped with a slide-on single-bounce ATR accessory featuring a silicon crystal (128 scans, 4 cm⁻¹ resolution). ATR spectra were intensity-corrected using the ATR correction function implemented in the OMNIC 8.2.0.387 software (Thermo Fisher Scientific Inc.), and the corrected spectra are shown in the corresponding figures for comparison with transmission-mode spectra.

To investigate the PP concentration in R-HDPE samples, a series of reference mixtures with known HDPE/PP ratios were prepared. Pure HDPE and isotactic PP (Aldrich) pellets were mechanically ground using a blender with steel blades under liquid nitrogen to prevent thermal degradation. The resulting powder was sieved using a mesh with pore size ~100 µm to isolate the finest fraction. Six mixtures with varying HDPE/PP compositions were then prepared by weighing appropriate amounts of each component. Approximately 150 mg of each mixture was placed into a pellet press and compressed at 10 tons/cm² for 2 minutes. Thin films (~100 µm thick) were subsequently obtained by compression molding at 200 °C. Three replicate films were produced for each composition to ensure reproducibility. R-HDPE samples were prepared with the same procedure by compression molding the recycled polymer pellets at 200 °C.

The experimental conditions for natural sunlight exposure of Carilon^® and PP specimens, used to study degradation and oxidation processes, are described in detail in the cited literature.

3. Results and Discussion

3.1. General Characteristics of Polymeric Materials and Their IR Spectra

The discussion of polymer spectra must take into account the peculiar characteristics of these materials [7,8] and their spectroscopic features [9,10,11,12,13,14,15,16] that are independent of whether the material is a virgin polymer or a second-raw material obtained through recycling processes.

• A polymeric material is often a formulation: in addition to the polymer itself, it contains additives that meet processing requirements and ensure the durability of the final product. Additives may provide chemical stability, while fillers can modify mechanical properties (e.g., reinforcing glass fibers) or electrical properties (e.g., graphene particles). Many polymeric materials intended for specific technological applications consist of blends of two or more polymers or copolymers with variable composition. The resulting IR spectrum is therefore a superposition of the spectral responses of all components and cannot perfectly match that of a pure homopolymer or even samples of the same polymer/copolymer that differ in additive type and/or concentration.
• Automated identification (typically using spectral libraries) is an effective and valuable method for determining the polymer family, especially when combined with supervised analysis. This is commonly employed for plastic sorting in recycling processes. However, especially for semi-crystalline polymers, sample preparation can influence morphology [7,8], resulting in spectral differences compared to literature references or database entries. Furthermore, the measurement technique—such as transmission, attenuated total reflectance (ATR ), or reflection—can significantly alter the spectral pattern (band shape and intensity), even when spectra are mathematically converted to a common absorbance (or transmittance) scale. Note that absorbance conversion—except in transmission mode—is based on theoretical models and assumptions that are never perfectly met in practice.
• Effective use of spectral libraries often requires data post-processing: baseline correction and subtraction of solvent, substrate, or contamination bands may be necessary. Spectral comparison also requires proper normalization procedures, typically based on normalizing absorbance intensity to a reference band. These processes are not automatic and require expertise, as they may introduce artifacts.

Nevertheless, IR spectroscopy offers capabilities far beyond basic chemical identification via spectral databases or correlative analysis. Spectra analysis has greatly enhanced our understanding of polymer physics, particularly due to advances in first-principles modeling (e.g., DFT) and molecular dynamics simulations [17,18,19,20,21,22]. A substantial body of theoretical and experimental work is now available [10,11,12,13,14,15,20,21,22] enabling critical interpretation of polymer spectroscopic responses and offering valuable tools even for the analysis of complex materials such as recycled polymers.

Below is a brief list of chemical and structural information that can be obtained through IR spectroscopy, some of which will be discussed in the following sections:

• Chemical identification, including additives: this is especially useful when studying/identifying unknown formulations and can reach quantitative diagnosis through calibration curves based on suitably developed reference samples of known composition.
• Diagnosis of chemical modifications: particularly important for monitoring degradation phenomena during production (e.g., extrusion, molding, spinning) or aging under various environmental conditions.
• Structural diagnosis: identifying molecular conformation, degree of crystallinity, and structural defects in semi-crystalline polymers.
• Morphology analysis: identifying the presence of different phases (e.g., crystalline vs. amorphous), distinguishing between crystalline polymorphs, and even quantifying them. Also includes recognition of polymer chain orientation, as in fibrous materials.
• Thermal transformation analysis: the material’s thermal history can be traced by recording spectra at different temperatures during controlled heating/cooling cycles. This allows the monitoring of phase transitions (e.g., melting) or the effects of post-treatment processes (annealing, thermal cross-linking).

The above information is obtained by analyzing the frequencies and intensities of the IR absorption bands associated with vibrational transitions of the material. It is important to remember that Raman spectroscopy is complementary to IR spectroscopy for studying molecular and material vibrational transitions. In some cases, Raman can even replace IR, especially when sampling issues arise with IR measurements or when IR sensitivity/selectivity is low.

3.2. Techniques and Measurement Setups

Polymeric materials are generally solid samples, and when analyzing their structure and morphology, it is essential to use an appropriate experimental setup and/or properly prepare the sample. Several measurement techniques are well documented in the specialized literature [13,15,16]. In this section, we briefly list the most commonly used methods, highlighting which types of polymeric samples they are best suited for. In addition to transmission experiments, which directly yield the absorption spectrum, alternative setups exploit different physical phenomena—primarily reflection and scattering—associated with IR photons absorption. These methods can reconstruct an absorbance spectrum through appropriate mathematical processing. However, as such procedures often introduce some distortion in the spectral profile, it is considered best practice—particularly for quantitative analysis—to compare spectra obtained using the same technique and experimental setup. For solid polymer samples, the following methods are available:

Transmission measurements. Suitable samples include: fine polymer powders dispersed in KBr pellets; thin films deposited on a transparent substrate (e.g., KBr or ZnSe windows); thin films deposited on a reflective substrate (e.g., aluminum, gold, silicon wafer). In the last case, the setup is called double transmission, as the IR beam passes through the sample twice (incident and reflected beam) before reaching the detector. By compression in a diamond anvil cell (DAC) accessory, very small polymer particles can be prepared as a thin film suitable for transmission-mode analysis. It should be noted, however, that the application of high pressure may induce morphological changes in the material.

Reflection measurements comprise:

• ATR: One of the most convenient techniques for polymeric materials. It detects absorption of the evanescent wave generated at the surface of a high-refractive-index crystal in contact with the sample. Since surface layers often contain additives (e.g., slip agents, release agents), for bulk analysis, it may be necessary to remove the surface layer or section the sample. With special ATR accessories, direct measurements on finished products are also possible.
• Specular reflection: Suitable for polymeric materials with good reflectivity (also depending on surface finish). The sample must be thick enough to prevent the collection of transmitted photons (as in double transmission setups).
• Diffuse Reflectance (DRIFT): Captures diffusely scattered IR radiation over a large solid angle, from which the IR spectrum is obtained. Applicable to samples with high IR scattering (e.g., powders or granular solids), where transmitted IR is negligible and specular reflection is weak.

3.3. Spectral Analysis

There are well-established procedures used by companies and manufacturers for quality control of various polymeric materials via IR analysis. Due to their applied nature and confidentiality, these methods and algorithms are rarely detailed in the scientific literature, and general accessible guidelines are lacking. This section aims to provide practical methodological guidance for addressing the challenges in studying recycled or recyclable polymers, supported by selected examples. IR spectroscopy can assist in: (i) Separation of polymeric materials for recycling; (ii) Analysis/evaluation of recycled material (second-raw material)

3.3.1. Sorting

IR analysis, particularly in the near-infrared (NIR) region, is a well-established method for rapid sorting of polymers to be recycled. These applications often focus on identifying the specific polymer class (e.g., polyethylene, polypropylene, polyamide, polyethylene terephthalate, polyvynil chloride, etc.). Identification is enabled by the spectroscopic “fingerprints”—intense absorption bands corresponding to the fundamental vibrational modes of the polymer. In this context, in addition to supervised sorting by operators, automated techniques such as NIR/MIR hyperspectral imaging can be employed on industrial lines [6]. Effective sorting should not only selectively identify a specific polymer but also minimize contamination with other polymeric species that may or may not be miscible with the primary polymer, as such contamination can degrade material properties [23,26].

The sorting challenge is particularly relevant for post-consumer recycled material, which additionally requires preliminary removal of non-polymeric components (metal fragments, paper, etc.). In the case of post-industrial waste, the polymer and its formulation should, in principle, already be known—especially when the scraps are returned directly to the recycling company.

Figure 1 shows a comparison of NIR - IR spectra of common polymers, some of which are currently subject to recycling. The spectra clearly demonstrate that identification is possible by detecting intense and selective absorption peaks specific to each polymer. Spectra were recorded on manufactured samples (thick films) using both transmission mode (Figure 1.a) and ATR mode (Figure 1.b). The intense absorption bands in the mid-IR region — corresponding to fundamental transitions (3500–400 cm⁻¹) — enable highly effective polymer identification, which can be performed either automatically using spectral libraries or through empirical correlations based on characteristic group frequencies [9].

It is important to note that spectra acquired in transmission mode (Figure 1.a) show saturation effects in the intense fundamental bands, which hinder reliable identification. In contrast, spectra recorded in ATR mode overcome this limitation and can be used to identify the main polymeric component in the material by focusing on the most intense bands. Unlike ATR spectra, where NIR transitions are extremely weak and mostly indistinguishable from experimental noise, transmission spectra of films exhibit a good signal in the NIR region. In this case, overtone and combination bands can be used for both material sorting and quantitative diagnostics (see Section 3.3.3).

Polyethylene is one of the most abundant polymers found in post-consumer plastic waste. It is widely used in packaging — primarily as low-density polyethylene (LDPE) — and, in the form of high-density polyethylene (HDPE), in a variety of manufactured products (containers, furniture components, toys, pipes, sheets) where enhanced mechanical properties are required. During the recycling process, it is important to separate HDPE from LDPE, and IR spectroscopy is a powerful tool for this purpose, provided that one focuses on specific spectral features that support accurate classification. Figure 2 shows a comparison between the ATR spectra of two samples of virgin HDPE and LDPE polyethylene.

The differences between HDPE and LDPE relate to the relative abundance of amorphous and crystalline phases. These differences are reflected in a more pronounced crystal splitting in the rocking and bending bands of HDPE, as well as in the intensification of signals associated with polymer chains in distorted conformations in the CH₂ wagging region. An additional marker of LDPE is the appearance of a very characteristic band at 1372 cm⁻¹, attributed to the symmetric bending of the methyl group – often referred as “umbrella mode”. This band gains significant intensity in the presence of the branching of the polymer’s main chain, which ends in CH₃ groups. These branches are typical of LDPE but are less frequent in HDPE due to the tighter control of the molecular structure in catalyzed synthesis.

Any analysis based solely on the observation of the fundamental transitions of the main polymer cannot exclude the presence of contaminants, additives, or other polymers in small percentages. We will explore these aspects in greater detail in Section 3.3.2 and 3.3.3, by analyzing recycled polymer pellets, where mixtures of similar polymers—such as PE and PP, or HDPE and LDPE—represent one of the most common issues.

In the context of analyzing material intended for recycling, IR analysis based on the identification of oxidation markers—typically the C=O stretching band around 1700 cm⁻¹—can be useful [27,28,29]. More importantly, the analysis of band shapes can provide insights into the aging and degradation levels of the materials. This information could be used as a criterion to discard low-quality materials before undergoing mechanical recycling.

3.3.2. Comparative Analysis of Recycled Polymers: R-HDPE.

The IR analysis of recycled polymeric materials—usually supplied in pellet form to manufacturers—requires appropriate experimental strategies, and several example of application of IR spectroscopy to specifica cases are reported in the literature [28,29,30,31].

Depending on the type of information needed, it may be preferable to perform ATR measurements or transmission measurements on films obtained by melting and compressing the pellets. Quantitative determinations require the development of calibration methods using model formulations prepared ad hoc, as will be illustrated in a case study illustrated in Section 2.3.3. In this section, we present some comparisons between IR spectra of recycled HDPE pellets (R-HDPE), showing how a supervised analysis of ATR spectra can already highlight significant differences in chemical composition and structure.

Figure 3 shows the IR (ATR) spectra of various R-HDPE samples obtained from different manufacturers. Comparison with the spectrum of a virgin HDPE pellet reveals that each R-HDPE sample is distinct, and all differ noticeably from the virgin material. The inset (zoomed region) highlights the shape of the two strong rocking and bending bands, which exhibit characteristic crystal splitting—a marker of the crystalline phase [10,12,14]. This splitting is clearly visible in virgin HDPE, where crystallinity is high (usually around 60%), but is less pronounced in the recycled pellets, indicating a lower degree of crystallinity in R-HDPE. In some spectra, features attributable to additives—such as antioxidants—are also clearly observed.

In Figure 4, a zoomed-in view of the IR spectra highlights the marker band of the methyl group, which appears particularly intense in sample 4 (red line). This could be attributed to a significant presence of LDPE within the R-HDPE sample. However, since PP also exhibits a strong CH₃ bending band at nearly the same wavenumber, it is also plausible that sample 4 contains a certain amount of PP. Notably, the varying intensity of the band at 1650 cm⁻¹ (Figure 3) attributed to a common antioxidant, indicates different additive concentrations across the three samples.

The first conclusion drawn from the analysis of the R-HDPE samples in Figure 4 is that the label “R-HDPE” does not guarantee a crystallinity level comparable to that of virgin HDPE. This is likely because the polymeric fraction in R-HDPE consists of a blend of HDPE and some amount of LDPE and/or it contains a fraction of PP. It is well known that the degree of crystallinity strongly influences the mechanical performance of polymer materials [7,8].

Another important observation arises from comparing ATR spectra of different granules within the same R-HDPE batch (Figure 5). Some granules, when examined under an optical microscope, reveal fragments of extraneous materials. IR analysis confirms that different pellets exhibit varying chemical compositions. Spectral subtraction (not shown here) and comparison with reference spectra from a database reveal that sample A contains common anti-tack additives on the pellet surface.

3.3.3. Detection and quantification of Polypropylene in R-HDPE samples.

Figure 6 shows a comparison that highlights the presence of polypropylene within R-HDPE samples. The spectra of the recycled polymer, recorded in transmission mode on a film obtained by compression moulding from the pellet, exhibits the typical saturation of polyethylene’s fundamental absorption bands. However, it also clearly reveals weak absorption features attributable to the presence of a polypropylene fraction, likely introduced as contamination during the separation stage of the recycling process.

The presence of isotactic polypropylene (PP) in R-HDPE samples is quite common due to the intrinsic challenges during the sorting process. While small amounts of PP may have beneficial effects on the mechanical and rheological properties of the material, it is crucial to reliably control the PP content to ensure consistent material properties when using different commercial R-HDPE products or different batches from the same manufacturer. To address this, we developed methods based on intensity ratios of selected marker bands for HDPE and PP, enabling quantification of the PP weight percentage in the range of 0–15%. Interestingly, there is no single optimal strategy for selecting marker bands for this analysis. For example, absorption features from certain additives can mask useful PP and HDPE marker bands in the mid-IR region. In such cases, the near-infrared (NIR) region provides more informative spectral data. Reference calibration curves were obtained using lab-prepared samples consisting of known weight ratios of virgin PP and HDPE.

In order to use IR spectroscopy for the quantitative determination of polypropylene (PP) content in commercial recycled HDPE (R-HDPE), transmission IR measurements were performed on thin films of HDPE/PP blends prepared according to the method described in the Materials and Methods Section. Three films were prepared and analyzed for each sample. Table 1 reports the composition of the different reference samples, labeled with alphabetical letters.

Several commercial recycled PE samples (samples 2-8) and a commercial virgin HDPE (sample 1) were also prepared as thin films obtained directly from the pellets.

IR spectra of the reference samples and R-HDPE samples were recorded in the 350–7000 cm⁻¹ range to capture absorptions corresponding to first overtones and combination bands of fundamental transitions in the NIR region (Figure 7). As usual, the strong fundamental absorption bands of polyethylene exhibit saturation effects in the spectra recorded in transmission mode; however, PP features are clearly detectable, even in reference samples with low PP content (Figure 7c). In particular, PP features are also observed as weak but distinct peaks in the spectra of R-HDPE samples (Figure 7d).

For R-HDPE samples, the simultaneous presence of absorption bands from polymers, fillers, and additives makes the spectral region very crowded and masks several weak HDPE bands (see Figures 7c, 7d) that could otherwise be used to quantify the relative PP/HDPE content. Therefore, we focused our analysis on the NIR region (4000–7000 cm⁻¹), where overtone and combination bands are observed (Figure 8). The NIR analysis offers several additional advantages:

• Absorption intensities of overtone and combination bands are much weaker than fundamental transitions, reducing bands saturation effects even in relatively thick films. In fact, thicker samples can enhance the visibility of these weak bands.
• Additives and fillers exhibit very weak absorptions in the NIR, making their contribution negligible compared to the main polymer component. Hence, the obtained information primarily reflects the bulk polymer.
• Transmission mode ensures analysis of the bulk material, preventing overestimation of surface chemical species possibly present on pellets.

For the construction of the calibration curve, the HDPE marker band at 5664 cm⁻¹ (1797 nm) was selected, corresponding to the first overtone of symmetric C–H stretching vibration of the methylene (CH₂) group [32]. As for PP, the marker band at 5907 cm⁻¹ (1694 nm) was chosen, corresponding to the first overtone of asymmetric C–H stretching vibration of the methyl (CH₃) group. These two bands were selected due to their rather good separation and easy identification in both reference blend samples and commercial recycled HDPE samples (Figures 8a and 8b).

The peak height of the bands at 5664 cm⁻¹ and 5907 cm⁻¹ was measured after baseline correction by drawing a straight line between 6000 and 5000 cm⁻¹ (Figure 8c). For each concentration, three spectra from three different films were recorded, and the average peak intensity ratio of the marker bands was calculated and associated with the relative concentration. Numerical values are reported in Table 1, and the trend is shown in the plot in Figure 9. A linear correlation of the intensity ratio as a function of PP concentration was observed. From these data, linear regression allowed deriving a calibration function describing PP relative concentration C_(PP/PE) (w/w%)as a function of the band intensity ratio R=I₅₉₀₇/I₅₅₆₄.

$$C_{\mathrm{P}\mathrm{P}/\mathrm{P}\mathrm{E}}=\left({\displaystyle \frac{I_{5907}}{I_{5564}}}-0.067\right)\times {\displaystyle \frac{1}{0.004}}$$

(1)

This calibration curve was used to determine the PP content in commercial recycled HDPE samples. Spectra were recorded on three films per R-HDPE sample (spectra are reported in Figure 8). Values of R=I₅₉₀₇/I₅₅₆₄ and the estimated PP weight percentages from the calibration curve are summarized in Table 2.

In conclusion, the careful preparation of reference samples with known PP concentrations in HDPE enabled the establishment of a reliable calibration curve useful for detecting small percentages of PP. The method can discriminate samples with vanishing PP levels: sample 1, a commercial sample marketed as virgin HDPE, showed a negligible estimated PP content (~0.3%), confirmed by supervised spectral analysis in the fingerprint region (950–1000 cm⁻¹), where no PP-related absorptions were observed. All other samples (R-HDPE) showed clear PP signals in the fingerprint region, and result to contain estimated PP relative percentages higher than 1.5%.