Highly robust molecular information carriers for traceability of plastic materials Authors: Thibault Schutz, Nezha Badi, Filip E Du Prez Published: 2026 ABSTRACT AbstractSequence-defined oligomers offer programmable molecular architectures with potential in data storage, authentication, and anticounterfeiting. However, their deployment in real-world materials has been constrained by their low scale, limited thermal resilience and the need for specialized analytical methods. Here we introduce a molecular tagging strategy based on abiotic sequence-defined oligoamides incorporated at ultralow concentrations ( < 0.05 wt%) that endure the thermal and chemical conditions of industrial polymer processing. A four-monomer alphabet derived from unnatural amino acids enables the construction of digital molecular sequences that permit a first optical identification through ultraviolet absorption and/or fluorescence from chromophoric terminal groups. Sequence authentication, implemented via an automated Python-scripted readout strategy from tandem mass spectra, enables rapid and reproducible molecular code identification. Comparative studies of secondary and tertiary amide backbones reveal structure-stability correlations that govern the design of thermally robust taggants. The oligoamides retain structural integrity within both commodity thermoplastics and thermosets, surviving extrusion (up to 270 °C) and curing processes while remaining fully decodable. This approach establishes a versatile platform for secure molecular identification and traceability across several types of industrial plastic materials.Subject terms: Polymer synthesis, Information storage, Fluorescent labelling Introduction The development of anticounterfeiting and traceability technologies has become a pressing challenge in materials science, not only driven by economic and societal concerns but also by emerging regulatory requirements, which urge the tracking of materials throughout the supply chain. The growing prevalence of fraudulent plastics exacerbates these issues: according to the EU intellectual property office, counterfeit and pirated goods accounted for approximately USD 467 billion in 2021, representing about 2.3 % of global trade1. Beyond financial losses, counterfeit plastic products pose a serious risk to human health and the environment, as they often involve poor-quality and non-recyclable materials. These challenges have motivated the exploration of various authentication strategies, including nanoscale physical unclonable functions2, optical responses3, and supramolecular architectures4. Among these, molecular encryption represents a particularly promising pathway, as it enables the embedding of unique identifiers directly within a material. For example, the precise sequence of nucleobases in DNA can be used to encode digital information such as pictures, text or movies5. In this case, the four-monomer alphabet is translated into a binary code, enabling 2 bits of encryption per base6. Yet, the practical utility of DNA in industrial products remains constrained by its vulnerability to UV-light, moisture, and thermal degradation7. Some protocols have been developed to protect the DNA sequence, such as its incorporation in silica beads8 or in glassy polymer networks9, but they introduce additional complexity, requiring extra reagents, harsh extraction protocols and incorporating undesired inorganic species. Unnatural sequence-defined macromolecules have emerged in the past decade as robust alternatives for data encryption and anticounterfeiting10–16. These fully abiotic uniform structures, constructed from a wide range of synthetic monomers, provide stable and programmable molecular taggants that can be unambiguously decoded using tandem mass spectrometry (MS/MS)17,18. Several proof-of-concept studies on anticounterfeiting technologies using synthetic oligomers have emerged in recent years19,20. For example, Lutz and co-workers showed that sequence-defined oligourethanes (SDOUs) could serve as molecular taggants in 3D-printed polymer networks and implants21,22, while Zhang and co-workers reported multilevel barcoding using a polyester backbone23. Despite this progress, translation to industrial practice is hampered by two critical limitations. First, many proposed sequence-defined oligomers lack the chemical and thermal robustness needed to survive polymer manufacturing conditions, which routinely exceed 150 °C during thermoset design and range from around 200 °C for polypropylene (PP) or polystyrene (PS) to 275 °C for the polyamide 6-6 (PA66)24. Secondly, most existing labelling procedures rely on solvent casting or other laboratory-scale methods, whereas an ideal molecular taggant should behave like a conventional additive, directly dispersible in standard melt- or solution-processing workflows without requiring additional steps. Addressing these challenges requires molecular codes that are not only highly stable under harsh processing conditions but also scalable and compatible with existing polymer manufacturing. Our group recently demonstrated the feasibility of large-scale production of unnatural sequence-defined oligoamides (SDOAs), demonstrating the first promising thermal properties, with degradation temperatures exceeding 200 °C25. While natural peptides have been marginally explored for data encryption26–29, to the best of our knowledge, abiotic peptides have never been reported for anticounterfeiting applications. The use of such amino acid based architectures offer an attractive platform: they combine the synthetic versatility of peptide chemistry - including fine control over properties such as solubility, charge, and sequence readability - with the robustness of fully synthetic polymer backbones28. Importantly, peptide synthesis is a mature and automatable technology, routinely practiced at the kilogram scale in the pharmaceutical sector30, and relies on accessible monomers such as protected amino acids, rather than costly DNA phosphoramidites. In this study, we present the synthesis of tailor-made, information-rich SDOAs as highly robust molecular taggants for anticounterfeiting and traceability. These abiotic taggants are designed to withstand standard polymer processing conditions, enabling their incorporation into commodity plastics as actual additives. Their synthesis is based on a solution-phase synthesis protocol utilising non-cleavable, long-chain hydrophobic (HA) or fluorous-based (FA) anchors (Fig. 1). Solution-phase synthesis protocol for digital SDOA synthesis using a 2- or 4- amino acid (AA) alphabet. (i) Coupling using fluorenylmethyloxycarbonyl (Fmoc)-protected digital monomer: EDC.HCl, DMAP, CHCl3, 30 °C. (ii) Fmoc deprotection: 30 mol% diethylamine. (iii) SDOA end-capping using UV- or fluorescent-active carboxylic acid: EDC.HCl, DMAP, CHCl3, 30 °C. The thermal resilience of diverse molecular backbones, including urethane- and amide-based sequence-defined architectures, is investigated alongside the influence of hydrophobic anchoring groups and optically active end-groups. We introduce a four-monomer alphabet of abiotic amino acids for molecular barcode encryption and explore the incorporation of the resulting taggants at ultralow loadings (0.01–0.03 wt%) into thermoplastic and thermoset matrices. In addition, strategies for taggant recovery and identification are examined, combining solvent extraction with rapid optical prescreening through ultraviolet or fluorescence spectroscopy and automated sequence decoding from tandem mass spectrometry. Finally, the dispersion of the taggants within polymer matrices and their compatibility with processing conditions are evaluated to assess the potential for practical implementation in plastic materials. Together, these results will demonstrate the feasibility of robust, fully decodable digital molecular identifiers in industrial plastic products. Results Tailoring thermal performance through macromolecular design To compare the thermal stability of oligomers with different chemical nature for their backbone, a first set of uniform tetramers (T) with secondary urethanes, secondary amides or tertiary amide linkages was prepared starting from FA (i.e., FA-T1, FA-T2 and FA-T3, see supplementary information for full detail about their synthesis and Fig. 2). These oligomers were selected based on two key criteria: (1) the use of readily available building blocks, durable at high temperatures, and (2) the scalability potential of their synthesis to lab quantities (i.e., 10–100 g) via an efficient protocol. Our recently reported strategy to synthesise more than 50 g of telechelic SDOUs31 was adapted to produce the urethane-based FA-T1, while the oligoamides FA-T2 and FA-T3 were obtained following a classical solution-phase peptide synthesis (SPS), earlier developed by Carpino and co-workers to make short peptides on a 100 g-scale32. The two strategies were adapted by incorporating FA to enhance solubility and mass spectrometry (MS) sequencing. Purity of the three tetramers was confirmed by size exclusion chromatography (SEC), liquid chromatography – mass spectrometry (LC-MS), high resolution mass spectrometry (HRMS) and matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) analyses (Supplementary Figs. S1–S12 and Supplementary Table S1). Thermal stability comparison of different chemical backbones. (Left) Structure of the studied tetramers; (top right) table summarising the degradation temperatures at 5% weight loss for each oligomer; (bottom right) thermogravimetric analysis (TGA) traces obtained for the six tetramers. In comparison to FA-based anchors, we wanted to investigate the thermal resilience of hydrophobic chains with the use of a uniform soluble support containing long aliphatic chains (HA, Fig. 1 and Supplementary Figs. S13–S15) recently studied in our group to obtain SDOAs25,33. This support, employed industrially for kg-scale peptide synthesis34, facilitates the removal of side-products and excess reagents from the support-bound oligomer through simple precipitation in methanol or other polar solvents. While such support is typically cleaved to obtain the final sequence, it can also be used without a cleavable linker, taking advantage of the long hydrophobic chain to enhance MALDI-TOF detection and potentially improve thermal stability. In our study, this support was used to synthesise on the one hand an oligourethane tetramer (HA-T1) and on the other hand two oligoamides tetramers with secondary (HA-T2) and tertiary (HA-T3) amide groups (Fig. 2), whose structure and purity was again confirmed by MALDI-TOF and SEC (Supplementary Figs S16–S23 and Supplementary Table S1). Since most of the plastic production processes occur under non-controlled atmosphere (without inert gas), we investigated the thermal stability of the discrete oligomers under air using TGA (Fig. 2). As expected, oligoamides exhibited superior thermal stability compared to oligourethanes, with FA-T1 showing 5% mass loss (Td5%) at 186 °C and FA-T2 at 259 °C. This can be attributed to the stronger resonance stabilisation of amide functional groups, making them less likely to decompose at high temperatures and less prone to hydrolysis compared to urethanes. This observation is consistent with the only example found in literature about SDOU taggant thermal stability, where the structure starts to degrade around 150 °C21. N-methylation of the oligoamides further improved stability significantly, with FA-T3 exhibiting a Td5% at 296 °C, consistent with steric effects and the absence of hydrogen bonding, which hinder hydrolysis at elevated temperatures and prevent other types of degradation such as oxidation, depolymerisation, or random chain scission. Subsequent investigations showed that HA anchoring provides additional stabilisation: HA-T1 (urethane-based) and HA-T2 (amide-based) tetramers displayed a degradation temperature of about 40°C higher than their FA-based counterparts, FA-T1 and FA-T2 (Td5% = 225 and 301 °C, for HA-T1 and HA-T2, respectively). On the other hand, only a small improvement was observed for HA-T3 with a temperature increase of 15°C compared to its FA-based counterpart, FA-T3. The enhanced thermal stability observed for HA-based SDOAs compared to FA-based analogues can be attributed to differences in the electronic nature of the anchors. The FA anchor contains strongly electron-withdrawing trifluoromethyl substituents, which increase polarisation of the adjacent amide bond and could promote thermally induced cleavage, whereas the HA anchor lacks such activating groups. In any case, these results confirm that N-methylated amides are the optimal candidates to produce heat-tolerant molecular taggants. Notably, both HA and FA anchors showed remarkable thermal resistance over 295 °C, even exceeding processing temperatures of PA66 and polycarbonate, which are known to be among the highest (starting from 270–280 °C)35,36. Digital information encryption and readout Encoding strategies in molecular information storage rely fundamentally on the choice of the monomer alphabet. In a two-monomer system, each unit stores 1 bit of information, whereas a four-monomer alphabet encodes 2 bits per unit (Fig. 1). This strategy enables 64 (26) different encoding possibilities for molecular labelling. The number can be readily increased by extending the sequence length or expanding the monomer alphabet, as demonstrated by our group37 and others before28,38,39. Considering this exponential increase, we next explored digital information storage in abiotic amide-based hexamers. Guided by the demonstrated stability of N-methylated amides, we developed a four-monomer alphabet comprising N-methylated Fmoc-protected- glycine, β-alanine, γ-aminobutyric and aminohexanoic acids. The first two are commercially available, while the latter two were readily synthesised on a multigram scale in high yielding, two steps via ring-opening of N-methylpyrrolidone and caprolactam, respectively (Supplementary Figs. S5–S9). The molecular labelling concept was demonstrated using random binary codes (e.g., 010011100011 and 011100100001), stored in the monomer sequences of hexamers bearing either the fluorous (FA-H1 to FA-H7) or the hydrophobic (HA-H1 to HA-H3) anchor, and end-capped with optically active moieties (vide infra). SEC was used to follow the evolution of the SDOA synthesis, and MALDI-TOF confirmed the final structure (see Fig. 3 for FA-H1 and HA-H1, and Supplementary Figs. S24–S36 for the other SDOAs). Characterisation of SDOAs FA-H1 and HA-H1. MALDI-TOF and SEC (THF) traces showing the stepwise synthesis of FA-H1 and HA-H1, from n = 1-Fmoc (protected monomer) to n = 6-Fmoc (protected hexamer), illustrating each stage of hexamer design. FA-H1 and HA-H1 correspond to the final hexamers obtained after Fmoc deprotection. Although alternative readout techniques such as nanopore sequencing have recently been explored for natural peptide-based data storage systems29, MS/MS and MALDI-TOF/TOF MS remain the methods of choice27,29. In fact, MS/MS sequencing of unnatural oligoamides was already reported a few decades ago by the groups of Liskamp40, Zuckermann41 and Leigh42. However, to the best of our knowledge, this feature was only recently highlighted by Pentelute and co-workers in the context of digital information extraction from abiotic peptides28. Although MALDI-TOF/TOF MS was suitable for data readout of FA-H1, HRMS/MS was preferred for analysing fragments with a mass below 1000 Da. Figure 4 displays the fragmentation pattern of the precursor ion FA-H1+ (m/z = 964 [M + H]+). As defined by conventional fragment ion nomenclature in peptides sequencing43,44 (Fig. 4, grey box), the spectrum can be interpreted unambiguously from b- and y-type ions, yielding complete sequence coverage and thus digital information extraction. In practice, sequence reconstruction is feasible when at least one of the an, bn, cn, xn, yn or zn ions is detected for the corresponding n fragment. An additional mass of two Daltons must also be considered for c and y ions, as peptide structures retain the original protonating hydrogen while incorporating an additional hydrogen from the opposite terminus45. Using this approach, all FA-based SDOAs were successfully decoded (Supplementary Figs. S38–S46). Digital information extraction from FA-H1. a structure of FA-H1 and its fragmentation pattern with expected m/z values (N.O. means that the fragment was not observed). b LC-MS analysis of FA-H1 (254 nm); (c) resulting HRMS/MS spectrum in positive mode. HA-based SDOAs required a different strategy since the long hydrophobic chains display poor solubility in polar solvents and complicate ESI-MS/MS analysis. Therefore, MALDI-TOF/TOF MS was used instead in the presence of sodium trifluoroacetate to improve the ionisation during analyte preparation (See supplementary information, section 4.2 with Supplementary Table S2 and Supplementary Figs. S47 and S48 for more details). This technique enabled successful sequence reconstruction and digital information extraction from all HA-based SDOAs (Supplementary Figs. S49–S52). Ultimately, the digital information extraction from FA- and HA-based SDOAs has been automated. In fact, fragmentation modelling that is widely used in proteomics can be easily adapted to abiotic peptides28,46. We recently developed Oligoreader, a software tool that generates theoretical mass spectra through isotopic distribution simulations, designed for the readout of synthetic sequence-defined macromolecules47. To complement this, a more accessible Python script was implemented to simply extract digital information from experimental spectra (Supplementary Fig. S55). Despite the higher spectral complexity observed with HA-based SDOAs, digital information encrypted in all tetramers (i.e., FA-T2, FA-T3, HA-T2 and HA-T3) and hexamers (i.e., FA-H1 to FA-H7 and HA-H1 to HA-H3) was successfully and automatically decoded from both HRMS/MS and MALDI-TOF/TOF MS data (examples shown with hexamers Supplementary Figs. S56–S65). UV- and fluorescent-active molecular taggant detection In addition to sequence-defined encoding, incorporation of UV- and fluorescent-active end groups enables complementary optical identification. To explore this, a library of UV- and fluorescent-active building blocks was introduced by capping the terminal amine of FA- and HA-based SDOAs with available chromophoric carboxylic acids. Successful incorporation was confirmed by SEC, LC-MS, HRMS, and MALDI-TOF (Supplementary Figs. S24–S36). The structures and corresponding λmax values are summarised in Fig. 5. Normalised UV traces and TGA analysis (in air) of SDOAs taggants. a FA-based and (b) HA-based oligomers. Despite the limited availability of thermally robust chromophores with λmax above 350 nm, the selected library provides a diverse set of distinct UV signatures. FA-based SDOAs exhibit particularly characteristic absorption profiles because of the combined contribution of the bistrifluoromethyl aniline anchor (FA) and the capping agent (Fig. 5A), while HA-based SDOAs show slightly shifted signatures due to the aromatic nature of the anchor (Fig. 5B and Supplementary Figs. S66–68). Among fluorescent-active taggants, naphthalene and anthracene derivatives (FA-H1, HA-H1 and FA-H2, HA-H2) displayed a distinct emission, except for FA-H1 that showed complete quenching. As FRET can be excluded due to the absence of spectral overlap between absorption and emission (Fig. 5 and Supplementary Fig. S69)48, this quenching is attributed to intramolecular photoinduced electron transfer (PET) between the naphthyl group and the electron-deficient bistrifluoromethyl anchor. The flexible amide backbone enables transient proximity between these units, facilitating PET. This interpretation is supported by the FA/HA comparison because only FA-based oligomers exhibit quenching despite consisting of identical chromophores and backbone structures. Partial fluorescence remains detectable in FA-H2, likely ascribed to the higher quantum yield of anthracene in ethanol, which mitigates quenching effects49. This is further confirmed in Supplementary Fig. S70, showing a > 80% decrease in fluorescence intensity for FA-H2 compared to HA-H2 at equivalent concentrations. At the macroscopic scale, visual detection was demonstrated by incorporating a sufficient amount of SDOAs into PA66, exhibiting distinct responses under 365 nm UV light, including fluorescence and visible coloration depending on the chromophore (Supplementary Fig. S71). Thermal stability of the UV- and fluorescent-active molecular taggants We next investigated the influence of the capping agent on the thermal stability of the SDOAs using TGA. Oligoamide degradation is often determined by the cleavage of the terminal monomer25, thus thermal stability of the FA-based SDOAs (FA-H1 to FA-H7) featuring various capping agents was systematically compared (Fig. 5A, B). The first degradation pathway can be attributed to main chain hydrolysis, while the second one is potentially caused by the cleavage of the terminal amide bond. The enhanced temperature resistance of the latter pathway is attributed to the electronic stabilisation provided by conjugated systems, as confirmed by TGA-FTIR analysis (Supplementary Fig. S72). First analysis showed that naphthalene-capped FA-H1 exhibited only a slightly lower degradation temperature than the tetra-amide FA-T3 (282 °C vs 298 °C at 5% weight loss), indicating that chain length exerts only a minor effect on thermal stability. Although anthracene has a bulkier structure than naphthalene, it is known that this extended π-system has lower thermal stability50. This is indeed confirmed by the thermal degradation temperature of FA-H2 observed at 208 °C. On the other hand, the indole-capped structure FA-H3 demonstrated a much higher thermal resilience (Td5% = 276 °C). This reflects the inherent stability of the heterocyclic indole ring and its electronic stabilisation effects. We subsequently explored more challenging structures, including cinnamic (FA-H4) and coumaric (FA-H5) compounds. The presence of reactive double bonds or hydroxyl groups increases susceptibility to hydrolysis or side reactions at elevated temperatures, resulting in degradation at 254 °C and 241 °C, respectively. Nevertheless, these two conjugated systems remained thermally more stable than the phenyl-capped FA-H6 (231 °C), highlighting the stabilisation effect of vinylene extended conjugation. Also, FA-H6 bearing only a single aromatic substituent exhibited reduced thermal stability relative to its extended-conjugation counterpart, FA-H1. Finally, the benzylic-capped FA-H7 was analysed to further clarify the origin of the secondary degradation pathway. Notably, its TGA curve showed a remarkable diminution of the secondary degradation event, thereby again confirming that the additional thermal degradation observed in conjugated carbonyl-containing structures results from cleavage of the thermally robust aromatic amide bond present in the end-capping moiety. Ultimately, the effect of HA anchoring on the thermal resilience was evaluated using HA-H1, HA-H2 and HA-H3. As expected, the long hydrophobic alkyl chain enhanced the thermal stability of the SDOAs in comparison to corresponding FA-based analogues. Similar trends were observed for the naphthalene and cinnamic-capped HA derivatives, showing higher degradation temperatures (around 10°C increase, Fig. 5B). The anthracene-capped HA-H2 exhibited a more substantial increase ( + 37 °C), which further confirms that PET contributes significantly to the thermal degradation of FA-H2. In general, these results demonstrate that the nature of the UV- or fluorescent-active capping agent significantly influences the thermal stability of SDOAs. Such effects must be carefully considered when designing molecular taggants for high-temperature polymer processing. Notably, all SDOAs examined in this study maintained thermal stability above 230 °C (except FA-H2) and some of them above 270 °C. Material traceability in a thermoset and thermoplastics To demonstrate the feasibility of material traceability, the above described SDOA taggants were first applied in a case study using an unknown commercial formulation for highly crosslinked epoxy resins at a 20 g scale51. HA-H1 was incorporated in a very small amount (0.01 wt%), and a suspension of amine (A) and epoxy (E) compounds was added to a plastic cup, which was vigorously mixed using a speed mixer. Network formation induced heat release above 150 °C, indicating efficient conversion of A and E (Fig. 6A). After 24 h curing at room temperature, the taggant HA-H1 was extracted from the thermoset (See section 8.1 in the supplementary information). The solution was concentrated under air and analysed by MALDI-TOF MS, confirming the presence of HA-H1 and demonstrating successful extraction from the highly crosslinked material (Fig. 6A). Simultaneously, this result also shows the chemical resilience of SDOA taggants during thermoset design at elevated temperatures. Methodology for labelling, processing, extraction, and decoding of molecular taggants in polymer materials. Experiments conducted on both an epoxy thermoset material (a) and the thermoplastic PA66 (b). After molecular taggant recovery and MS analysis, the corresponding binary code can automatically be decrypted using a Python-based script. Subsequently, the robustness of this tagging technology was evaluated under extreme polymer processing conditions by assessing its resilience following high-temperature exposure. Prior to extrusion experiments, SDOAs detectability was tested by subjecting them to thermal annealing for 30 min at various temperatures in the presence of three different types of commodity plastics, i.e., PP, PS and PA66 (Supplementary Table S3). Besides their chemical nature, the processing temperatures differ substantially, ranging from 180–250 °C for PP and PS to 260 – 275 °C for PA66. Thermal annealing experiments were performed using 7 g of the corresponding commodity plastics, each containing 0.03 wt% (2 mg) of the selected molecular taggants. After thermal exposure for 30 min, the tagged polymer was allowed to cool to room temperature, cut into small pieces, and subjected to molecular taggant extraction using 5 mL of ethanol. The resulting solution was directly analysed by UV or fluorescence spectroscopy, and a concentrated aliquot was prepared for MS analysis. All SDOAs listed in Supplementary Table S3 were successfully detected and confirmed by MS, demonstrating thermal stability consistence with the degradation temperatures observed previously (some examples are illustrated in Supplementary Figs. S73–S77). In a final labelling/extraction experiment, a green polylactic acid (PLA) yogurt cup containing unknown additives and colourants was successfully tagged and analyzed, demonstrating the robustness of our multilevel identification method in real-world conditions (Supplementary Fig. S78). To further demonstrate versatility and resilience during high-temperature compounding (and extrusion), FA- and HA-based taggants with different UV-active groups and thermal degradation profiles were tested (Supplementary Table S4). The most thermally stable SDOAs, i.e., FA-H1, FA-H3, and HA-H1 with degradation temperatures above 270 °C, were tested in a PA66 matrix. On the other hand, the cinnamic and anthracene derivatives FA-H4, HA-H2, and HA-H3, which display slightly lower thermal stability in the range of 240–260 °C, were evaluated in PP. High-temperature compounding was first carried out in a mini-compounder (7 g scale) for 15 min under air. HA-based SDOAs (HA-H1, HA-H2 and HA-H3) listed in Supplementary Table S4 were extracted using the same procedure mentioned above (Supplementary Figs. S79–S82), demonstrating the remarkable robustness of this novel technology. On the other hand, a slightly modified extraction procedure was used to extract FA-based SDOAs (FA-H1, FA-H3 and FA-H4). After ethanol extraction, solutions were concentrated and washed with diethyl ether (Et₂O) to remove low molar mass species prior to analysis. In contrast, HA-based structures did not require preliminary washing steps to detect high molar mass taggants as single species, underscoring the benefit of on-demand synthetic design in tailoring detection properties. High-temperature compounding was also performed on a slightly larger scale (15 g) to achieve 0.01 wt% of taggants in the polymer matrix. This experiment was conducted with FA-H1, and after extraction, the molecular tag was again unambiguously detected through LC-MS analysis, indicating the absence of any degradation (Fig. 6B). Finally, the robustness of our SDOA technology was evaluated against comparable existing SDOU taggants21. For this, SDOUs FA-T1 and HA-T1 (Fig. 2) were compounded at 200 °C in PP to match their observed degradation temperatures. Following the same extraction protocol used for SDOAs, both SDOU taggants were successfully detected. However, as anticipated, neither SDOU taggant was observable using MS techniques after compounding in the PA66 matrix under the same processing conditions as those applied to SDOAs at 275 °C (Supplementary Table S4). In fact, isothermal TGA at 270 °C for 30 min in oxidative conditions revealed over 25% weight loss for the urethane backbone (HA-T1) versus approximately 5% for the amide backbone (HA-T2), demonstrating again the superior thermal stability of amide-based SDOAs (Supplementary Fig. S83). Dispersion, detection and stability of molecular taggants in polymers To assess practical implementation, taggant dispersion, detection, and stability were evaluated under melt-compounding conditions. Homogeneous dispersion was confirmed by reproducible UV–vis absorption profiles from independently extracted samples (Supplementary Figs. S84 and S85) and supported by X-ray photoelectron spectroscopy (XPS) at ultralow loading (0.02 wt%), which consistently detected taggant-specific elements across multiple locations, but no signals for virgin PP (Supplementary Fig. S86). Detection limits and extraction kinetics were evaluated by UV–Vis and MS. For this, 0.8 mg of taggant in 2 mL EtOH was required for reliable UV identification, or only around 0.4 mg using CHCl₃ (Supplementary Figs. S84 and S85). On the other hand, MS enables detection with at least 2 mg (Supplementary Figs. S79 and S80). Rapid extraction within minutes was observed, with higher efficiency in CHCl₃. In the presence of three common aromatic additives (antioxidant Irganox 1330, UV-absorber Tinuvin 329 and Pigment Blue 15), partial spectral overlap occurs, but this is fully mitigated by a simple washing step with diethyl ether (Et₂O), enabling clear recovery of the taggant signal (Supplementary Fig. S87). Leaching experiments (water, 10% ethanol, acidic and basic media at 40 °C) showed minimal migration across polymer matrices, with only trace amounts detected, typically below the quantification limit (Supplementary Tables S5, S6). In comparison to the SDOAs taggants developed in this study, only the UV-absorber showed measurable migration under the same conditions. Taggants remained detectable after exposure, confirming stable incorporation. In addition, thermal and rheological analyses of PS containing 0.1 wt% taggant showed no measurable changes in degradation temperature, Tg, or viscoelastic behaviour (Supplementary Fig. S88). From a practical perspective, the approach is compatible with industrial polymer formulations, as effective identification relies on ultralow taggant loadings. At these concentrations, the associated cost contribution is expected to be in the same range as that of speciality additives commonly used in technical plastics (e.g., nucleating agents). This suggests particular relevance for applications where traceability, anti-counterfeiting, or controlled material identification provide added value, such as for high-value, technical, or regulated products. In summary, this work establishes the first molecular tagging platform that achieves multilevel identification while enduring the extreme thermal conditions of commodity polymer processing. Sequence-defined oligoamides, synthesised as tetramers and hexamers with uniform masses between 800 and 1750 Da remain fully detectable by MS, MS/MS and/or UV spectroscopy, even at incorporation levels below 0.03 wt % and after compounding at temperatures above 270 °C in thermoplastic matrices such as PA66. Furthermore, homogeneous dispersion within polymer matrices, efficient extraction, and minimal leaching under relevant conditions were demonstrated, confirming the stability and practical applicability of the taggants. These highly robust molecular taggants set a benchmark for future developments in sequence-defined labelled materials, highlighting how precise macromolecular design enables scalable synthesis, chemical and thermal robustness, and straightforward automated readout. Incorporation of UV- or fluorescence-active terminal groups provides orthogonal optical detection modalities. Importantly, no measurable impact on thermomechanical properties was observed at relevant loadings, supporting seamless integration into existing materials. More broadly, this approach positions SDOAs as durable and versatile identifiers for anticounterfeiting and traceability, with potential to extend it across other material classes such as textile, wood and composites. Methods All experimental methods and analytical data are available in the supplementary information document. Supplementary information Supplementary information Transparent Peer Review file Acknowledgements The authors thank Bernhard De Meyer, Jan Goeman and Pieter Surmont (UGent) for technical support. Gabriel Mabille (UMons) and Sandro Lehner (EMPA) are also acknowledged for assistance with preliminary extrusion experiments and TGA-FTIR analysis, respectively. The authors also thank Matthias Minjauw and Prof. Christophe Detavernier (UGent) for XPS analysis. Author contributions The manuscript was written through contributions of T.S., N.B. and F.D.P., who contributed equally to the conception of the project and discussed the experimental results and their validation. T.S. designed and conducted the experiments. All the data were analysed and interpreted by T.S. All authors discussed the results and contributed to the final manuscript. Peer review Peer review information Nature Communications thanks Hongwei Ma, and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available Funding This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 101021081, ERC-AdG-2020, CiMaC project) and from the European Union’s Horizon Europe research and innovation programme (grant agreement No. 101248566, ERC POC, BigTop project). The NMR infrastructure used in this study has been funded by the Research Foundation Flanders (FWO, I006920N) and the Bijzonder Onderzoeksfonds (BOF.BAS.2022.0023.01). Data availability The data supporting the findings of this study are available within the article and its Supplementary Information. All data are available from the corresponding author upon request. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Thibault Schutz, Email: Thibault.Schutz@ugent.be. Nezha Badi, Email: Nezha.Badi@ugent.be. Filip E. Du Prez, Email: Filip.DuPrez@ugent.be Supplementary information The online version contains supplementary material available at 10.1038/s41467-026-73889-1. References Associated Data Supplementary Materials Supplementary information Transparent Peer Review file Data Availability Statement The data supporting the findings of this study are available within the article and its Supplementary Information. All data are available from the corresponding author upon request.