Generating Tagged Micro‐ and Nanoparticles of Poly(ethylene furanoate) and Poly(ethylene terephthalate) as Reference Materials

TL;DR

Researchers created fluorescently tagged nanoparticles from PET and PEF polymers to help detect and study nanoplastics in the environment and biological systems.

Contribution

A novel method for producing fluorescently labeled PET and PEF nanoparticles with long-term stability and small sizes for environmental and biological monitoring.

Findings

Fluorescently tagged PET and PEF nanoparticles (200–700 nm) were successfully produced using mechanical and solvent-based methods.

The nanoparticles showed long-term stability in water (up to 57 days) and were characterized with minimal changes in polymer properties.

s-SNOM imaging confirmed the ability to detect individual PEF particles as small as 200 nm.

Abstract

Detecting nanoplastic particles in environmental samples and biological tissues remains a significant challenge, especially in view of newly emerging polymers, not yet commercially exploited. Fluorescent labeling provides a tagging strategy to overcome this limitation by reducing the detection limit of individual particles, especially for small‐sized particles. We present a method for producing labeled nanoparticles (NP/MP) of poly(ethylene terephthalate) (PET) and poly(ethylene furanoate) (PEF), tagged with Alexa Fluor 633 or Alexa Fluor 647. Our preparations used mechanical grinding or solvent‐based approaches (confined impinging jet mixing, ((CIJ, precipitation), generating particles with hydrodynamic diameters of 200–700 nm, displaying long‐term stability in water of up to 57 days. Stable suspensions with concentrations of the particles ranging from 10 µg/mL (surfactant‐free, by solvent mixing) to 5.88 mg/mL (precipitation, containing surfactant) were generated with zeta‐potentials from −5 to −50 mV. Characterization of the nanoparticles by SEC, DSC, and XRD showed no significant changes in molecular weight, thermal behavior, or crystallinity via the solvent‐based methods, compared to the pristine polymer, highlighting their suitability for producing standardized nanoparticle dispersions. Fluorescence spectroscopy of the Alexa‐dye‐labeled particles confirmed the successful incorporation of the Alexa dyes, so improving monitoring of their biological profiles of the PEF‐MP/NPs. s‐SNOM (near field imaging) could identify individual PEF‐particles sized ∼200 nm by direct imaging.

We successfully developed poly(ethylene terephthalate) (PET) and poly(ethylene furanoate) (PEF) nanoparticles and microparticles, covalently tagged with Alexa Fluor 633 or Alexa Fluor 647, using both mechanical and solvent‐based approaches. The particles displaying hydrodynamic diameters between 200 and 700 nm, exhibited long‐term colloidal stability in aqueous suspension.

Figures

Schematic representation of preparation routes to generate NP/MP (here indicated as NP1 – NP8) of PEF/PET as dispersion via mechanical grinding, confined impinging jet methods (CIJ), and precipitation from DMSO.

(A) Synthetic route for the used PEF by melt‐polycondensation using Sb2O3 as catalyst in a two‐step process, (B) 1H‐ and 13C‐NMR spectra, and (C) SEC chromatogram in HFIP.

DLS measurements of PEF/PET particles (NP1 and NP7 after preparation, NP2 – NP6 and NP8 after dialysis) and corresponding SEM images.

Hydrodynamic diameters and ζ‐potential of PEF NPs over time measured by DLS.

(A) AFM image of PEF‐NP3 with a nominal size of up to 200 nm on Silicon. (B) FTIR spectrum of a pressed powder PEF sample (reference, obtained by conventional FTIR spectroscopy) shown in black and nano‐FTIR spectra obtained for a continuous PEF film (obtained from heated PEF micro particles) shown in grey and for six individual PEF NPs A‐F as indicated in (A) shown in different colors, respectively. The different FTIR spectra have been vertically offset for clarity.

Tables

TABLE 1. PEF and PET ‐based particles prepared with different techniques, the corresponding hydrodynamic diameters, Dh, ζ‐potential, and the achieved concentrations.

Entry	Mode	Polymer	Solvent	Label	D h (nm)	ζ‐potential (mV)	Concentration
NP1	Dispersion	PEF	Acetone/H2O	—	680–430	−39	25 µg/mL
NP2	CIJ	PEF	HFIP/ H2O	—	92	−40	10 µg/mL
NP3	CIJ	PEF	HFIP/ H2O	—	215	−50	20 µg/mL
NP4	Precipitation	PEF	DMSO/ H2O	—	173	−15	3.30 mg/mL a
NP5	Precipitation	PEF	DMSO/ H2O	Alexa 647	123	−13	5.88 mg/mL a
NP6	Precipitation	PEF	DMSO/ H2O	Alexa 647	209	−12	4.70 mg/mL a
NP7	Dispersion	PET	Acetone/ H2O	—	145	−35	0.44 mg/mL
NP8	Precipitation	PET	DMSO/ H2O	Alexa 633	160	−5	5.0 mg/mL a

Full text

Introduction

1

Plastic micro‐ and nanoparticles, also termed “microplastic/nanoplastic”, are increasingly found in nature [1] due to rising global plastic production and improper waste disposal [2]. Limited reuse or recycling strategies result in most plastics ending up in landfills or the environment [3]. Degradation processes break macroplastics into microplastics and eventually smaller, nanosized plastic particles, often called nanoplastics (for the definition of MP/NP see AUTHOR FOOTNOTE), demanding enhanced techniques for their detection in the environment. Paul‐Pont et al. recommended laboratory tests on microplastics from the ‘Big Six’ polymers—PP, PE, PVC, PU, PET, and PS—so that priorities can be set for exposure [4], which make up about 80% of EU plastics. Current biological data on nanoplastics are limited [5], and risk assessment is still in an early stage with growing concern about their toxicity in living organisms. Size matters when having a closer look at the biological impact nano/microplastic particles can have due to their small size and reactive surfaces, inducing oxidative stress, inflammation, immune dysregulation, and other forms of cellular impact. Their ability to pass through biological barriers such as the gut lining, bloodstream, and even the placenta raises concerns about potential effects on the human immune system, but also on reproductive, and neurological health. Therefore, the detection of MP/NPs still requires significant improvement to assess the effects of long‐term exposure on living organisms [6, 7, 8, 9].

Poly(ethylene furanoate) (PEF) is gaining recognition as a sustainable alternative to PET due to its structural similarity and superior performance [10]. However, the potential environmental and health impacts of PEF MP and NP remain underexplored, with early studies suggesting potential oxidative stress effects in soil ecosystems, comparable to PET [11]. The emergence of micro/nanoplastic (MP/NP) particles highlights the critical need to understand the biological role of PEF‐MP/NP and its currently unknown effects on humans [12]. Since PEF is not yet in widespread commercial use, its associated (MP/NP) particles are not found in the environment, making laboratory synthesis the only avenue for toxicological assessment.

Therefore, this work details the preparation of PEF/PET‐nanoparticles (NPs) to enable studies on their cellular‐impact and their potential medical significance. A key aspect of this research is the generation of labeled particle dispersions, crucial for conducting reliable in vitro and in vivo toxicity assays to assess human health risks. Such assessments are vital for evaluating the overall safety and viability of PEF as a sustainable replacement for existing (mostly PET) based materials.

Whereas macroplastic (sized 5 mm—1 µm) is generated by classical sieving methods (after a conventional mechanical grinding process), micro‐ (MP) and nanosized plastic (NP) particles are significantly more challenging to address [13]. Cryomilling or ball‐milling [14] can generate small‐sized particles after sieving and further processing [15], applicable to polyolefins, mainly PE and PP, where the energy‐input of the milling process can still be counterbalanced by a proper cooling (usually liquid nitrogen). Most of the so‐generated particles still require subsequent stabilization (e.g., by TWEEN‐surfactants, such as Tween‐80), resulting in significant changes of their surface properties (zeta‐potential), but generating the required enhanced stability of the dispersions for further application in cell culture‐testing [15].

Thünemann et al. have generated model PP and PE‐nanoplastics (sized in the range of 180–220 nm) by mechanical breakdown of polymer‐granules in solution, generating surfactant‐free dispersions [16] characterized by physical properties (crystallinity, morphology) similar to those of the initial bulk‐material. Confined impinging jet mixing (CIJ) uses a polymer solution (e.g., in a solvent for PS, PP, PE, PET, …) that is precipitated into a non‐solvent, thus generating particles of useful size‐ranges depending on the experimental conditions used, such as the mixer setup, the temperature and the used solvent‐mixtures [17]. While this method requires the use of surfactants during the precipitation process, it allows for the incorporation of dyes (such as Nile red, fluorescein, and rhodamine) and results in sizes below 100 nm.

We here for the first time report PEF as a polymer to prepare NPs sized ∼100 nm up to 700 nm, based on three methods (see Figure 1): (a) grinding inside a solvent using the mechanical breakdown of PEF‐powders and (b) a solvent‐based confined impinging jet (CIJ) mixing and (c) a solvent‐based precipitation method that also allows the incorporation of Alexa‐dyes into the PEF‐NPs for further imaging‐purposes. Simultaneously, PET NPs with and without Alexa dyes were generated in comparison to PEF. As PEF presents in a limited availability on the market, we synthesized a polymer with a molecular weight of ∼12 kDa, and further checked the properties of the so‐generated PEF‐NPs in view of (a) crystallinity, (b) stability of the NPs, and (c) we imaged the NPs using AFM and performed nano‐scale IR spectroscopy on individual NPs.

Schematic representation of preparation routes to generate NP/MP (here indicated as NP1 – NP8) of PEF/PET as dispersion via mechanical grinding, confined impinging jet methods (CIJ), and precipitation from DMSO.

AUTHOR FOOTNOTE: For clarity, it should be noted that, in addition to the international report ISO/TR 21960:2020(en) [18], Hartmann et al. have proposed that nanoplastic particles refer to dimensions less than 1 µm [19]. We cite both publications for definitions and classifications of plastic debris. Conversely, Sieg et al. have adopted the term “nano” as is customary in the nanotoxicology domain, indicating particles ranging from 1 to 100 nm [20]. In academic literature, a prevalent gap exists in the definitional boundaries between microplastics (greater than 1 µm) and nanoparticles (less than 100 nm). For plastic particles with diameters between 100 and 1000 nm, the designation “submicron” is frequently employed [20]. To ensure precision, the present discussion refers to polymer nanoparticles as plastic particles within the size spectrum of 1 nm to 1 µm.

Results and Discussion

2

Synthesis of Poly(ethylene furanoate) (PEF)

2.1

Poly(ethylene furanoate) (PEF) is available only to a limited extent commercially; hence, it was synthesized via a two‐step polycondensation using a metal catalyst, antimony(III) oxide (Figure 2A). Polymerization started with transesterification of dimethyl furan‐2,5‐dicarboxylate and ethylene glycol, followed by melt‐polycondensation at elevated temperature (230°C–240°C). The resulting polymer was purified several times by precipitation from HFIP solution into acidified methanol, ensuring the removal of residual metal catalyst, which can be critical for subsequent in vitro testing. Purity of PEF was confirmed by proton and carbon NMR (Figure 2B). The spectra display the expected signals of the polymer backbone along with weak resonances resulting from diethylene glycol furandicarboxylate (DEGF) units, reaching up to 3.6%. The molecular weight of the obtained polymer was analyzed using SEC (HFIP as solvent and PMMA as standards), displaying a molecular weight of Mn = 12 kDa (Figure 2C). MALDI‐TOF mass spectrometry (Figure S1) confirmed the expected repeating unit structure (182 m/z) and the presence of DEGF units (226 m/z), both detected as potassium adducts. Notably, the molecular weights obtained by MALDI‐TOF were lower than those obtained by SEC, likely due to the limited ionization efficiency of relatively high molecular weight PEF chains. Furthermore, MALDI‐TOF analysis enabled the identification of polymer end groups, consistent with the simulated patterns (see Figure S1).

(A) Synthetic route for the used PEF by melt‐polycondensation using Sb2O3 as catalyst in a two‐step process, (B) 1H‐ and 13C‐NMR spectra, and (C) SEC chromatogram in HFIP.

Additionally, the so‐obtained PEF as a solid material was characterized by FT–IR spectroscopy (Figure S2A), revealing the characteristic absorption bands corresponding to the polymer backbone in the expected regions. The C─H stretching vibrations from both the furan ring and the ethylene glycol unit are observed in the range of 2900–3200 cm^−1^. A strong C═O stretching band is visible at 1724 cm^−1^, while the C═C stretching band appears at 1577 cm^−1^.

The morphology of PEF was investigated using DSC and XRD to gain deeper insight into its structural characteristics before particle formation. Figure S2B shows the thermal transitions, with a glass transition temperature (T_g_) at 83°C. Cold crystallization is observed at 172°C, followed by melting of the polymer at 211°C. Additionally, XRD measurements revealed the presence of three sharp signals, respectively, at 16.0°, 20.5°, and 27.6°, i.e., a pattern similar to the ones previously reported for PEF polymers (Figure S2C) [21, 22]. Crystallinity of the pristine PEF polymer was estimated as ∼35%. Commercial poly(ethylene terephthalate), PET‐Lighter C93 (standard bottle grade) was obtained from Equipolymers, Netherlands, displaying a glass transition at 78.5°C and a melt temperature of 247.8°C, with a crystallinity of ∼50% [23].

Preparation of Plastic Particles

2.2

Polymeric particles were prepared using three different approaches (see Figure 1). A top‐down method based on mechanical breakdown, and two bottom‐up methods, confined impinging jet (CIJ) mixing and precipitation, were used, discussed in detail below (Figure 1). The mechanical breakdown method (dispersing with the mixer) offers the advantage to reach higher concentrations of the particles without the need for added surfactants, inspired by Ekvall, who created nanoplastics by mechanically breaking down polystyrene products using a household immersion blender, generating nanoplastics directly in water [24]. This high‐density polyethylene nanoplastics (∼110 nm) caused size‐dependent toxicity in Daphnia magna [25]. However, these polyethylene particles were unstable over time due to aggregation and subsequent precipitation, with hardly any nanoparticles detected after 100 days. The zeta potential of about −10.9 ± 6.4 mV was likely insufficient for a long‐term stability, which usually required values below −30 mV [26]. To avoid surfactant stabilizers, we aimed for strongly negative zeta potentials to reach stable dispersions of the PET/PEF nanoparticles, to serve as reference materials to detect nanoplastics in liquids, in food and the environment, similar to methods reported by Thünemann et al. [16]. Although CIJ mixing can also be performed without surfactants, the resulting dispersions typically yield lower concentrations. We utilized two established methods (mechanical dispersion and CIJ mixer) for PET particles and applied them here to the PEF particles. In contrast, the precipitation method eliminates the need for a CIJ mixer and, with addition of surfactants, allows the preparation of high‐concentration dispersions. PEF (**NP_1_‐NP_6_ **) and PET (**NP_7_‐NP_8_ **) based nanoparticles prepared by these three different methods are summarized in Table 1.

For the preparation of dispersions, PEF/PET powder/granules were fragmented in acetone by an Ultra‐Turrax disperser (details in Experimental section). PET particles were obtained within 10 min at a rotation speed of 18 krpm, whereas PEF required longer processing (40 min in total) and higher energy input (rotation speed of 25 krpm). As most of the polymer (98%–99%) remained in the macroscopic range, residual debris was removed by filtration, followed by solvent exchange from acetone to water. The freshly obtained PEF‐based nanoparticles **NP_1_ ** exhibited relatively large hydrodynamic diameters (∼680 nm) (Figure 3). SEM analysis of **NP_1_ ** (Figure 3) revealed irregular particle morphologies presumably due to mechanical degradation. Moreover, particles exhibit a pronounced tendency to agglomerate into larger clusters. In contrast, PET particles **NP_7_ ** displayed significantly smaller sizes, ranging from 140 to 150 nm.

DLS measurements of PEF/PET particles (NP1 and NP7 after preparation, NP2 – NP6 and NP8 after dialysis) and corresponding SEM images.

In contrast, the bottom‐up approaches, CIJ mixing (**NP_2_ ** and **NP_3_ **) and solution precipitation (**NP_4_‐NP_6_ ** and **NP_8_ **), produced significantly smaller particles (215–90 nm). The flash nanoprecipitation process employed here relies on rapid mixing of a polymer solution with an antisolvent introduced through the inlet of the CIJ device. The confined impinging jet mixer used for this purpose was adapted from previously reported designs (detailed design in Figure S3) [17, 27] and further modified in this work to allow operation in multiple flow configurations, thereby improving flexibility for nanoparticle preparation. In a typical preparation, a PEF solution in HFIP was rapidly mixed with water in a CIJ setup at a rate of ∼2 mL s^−1^ (injection time < 3 s), resulting in precipitation of polymer particles. The obtained PEF particles were subsequently purified by dialysis. The initial PEF concentration in the solvent, hexa‐fluoro‐isopropanol, HFIP, strongly influenced the particle size: increasing the concentration from 2.5 to 5 mg mL^−1^ led to a decrease in the average size from ∼215 nm (**NP_3_ **, Figure 3) to ∼90 nm (**NP_2_ **, Figure 3). However, a higher initial solution concentration did not necessarily translate to a higher final dispersion concentration, as **NP_3_ ** contained nearly twice the particle amount compared to **NP_2_ **. SEM analysis of **NP_2_ ** and **NP_3_ ** (Figure 3) revealed predominantly spherical particle morphologies with a slight tendency toward agglomeration into larger clusters.

The precipitation method was employed to prepare **NP_4_‐NP_6_ **, simplifying the bottom‐up approach by eliminating the need for a CIJ mixer. In this procedure, PEF powder (10 g L^−1^) is dissolved in DMSO at elevated temperatures to form a metastable solution, into which fluorescent dyes (Alexa 647 for tagged particles) and surfactants were incorporated. Precipitation of the particles was then induced by adding a Pluronic‐containing aqueous phase to the PEF solution. The slightly adapted method was also applied for tagging of PET particles with Alexa 633 dye (**NP_8_ **).

The resulting particles display hydrodynamic diameters in the range of ∼120–200 nm (from thereon termed “NP”, Figure 3), including both fluorescently tagged and non‐tagged samples. Determining the concentration of these dispersions proved challenging. The reported concentration (for **NP_4_‐NP_6_ ** and for **NP_8_ **), includes the surfactant content, which overlaps with the thermal degradation temperature of PEF, not perfectly allowing a decoupling in the calculations (TGA data are presented in Figure S4). The successful incorporation of the fluorescent dyes into the NPs was confirmed by fluorescence spectroscopy of **NP_5_ ** and **NP_8_ ** (Figure S5). SEM analysis of non‐tagged particles **NP_4_ ** (Figure 3) revealed predominantly spherical particle morphologies, which were preserved after tagging with fluorescent dye (**NP_5_ ** and **NP_6_ **, Figure 3). However, tagged NPs tend to agglomerate into larger clusters (**NP_6_ **, Figure 3). In contrast, no agglomeration was observed in the case of PET nanoparticles (**NP_8_ **, Figure 3). A comparative analysis of the nanoparticle sizes obtained using DLS for particles in suspension and SEM for particles dried on a substrate shows that the hydrodynamic diameter (D_h_) determined by DLS was in general larger than the mean diameter measured from SEM images for most samples (e.g., NP_2_, NP_3_, NP_5_). This is an expected and well‐documented observation that originates from the fundamental principles of each characterization method [28].

To assess the effect of the preparation method on the properties of the resulting PEF nanoparticles, SEC analysis was performed and compared to the initial PEF polymer (Figure S6). Repeated mechanical dispersion with an Ultra‐Turrax disperser caused a progressive decrease in molecular weight. The molecular weight of the initial PEF polymer (12 kDa) reduced to 10 kDa after the first dispersion cycle (NP_1_‐1). When the residual debris from this dispersion was reused in subsequent preparations, the molecular weight decreased even further, reaching ∼7.5 kDa for NP_1_‐3. This pronounced reduction can be attributed to the substantial mechanical stress imposed on the polymer during processing. Simultaneously, an increase in polydispersity was observed. In the DSC thermograms the melting point slightly shifted to lower values (208°C for **NP_1_ ** compared to 211°C for PEF). However, XRD measurements revealed no significant change in crystallinity of the resulting NP compared to the pristine PEF (around 33%, Figure S7). Additionally, we investigated the residual microscopic debris, which exhibited no significant change compared to the pristine PEF (Figure S8).

In comparison, the bottom‐up approaches—CIJ mixing (**NP_3_ **) and solution precipitation (**NP_4_ **, **NP_6_ **) did not lead to notable changes in molecular weight or polydispersity of the resulting particles. Likewise, no significant variations in thermal behavior or crystallinity were detected, highlighting their suitability to generate standardized and stable NP dispersions (Figures S6–S8). This also underscores the advantage of those methods to produce PEF/PET particles with a comparable morphology as close as possible to the native materials, important for all future microbiological and toxicological assessments.

Stability of the Plastic Nanoparticles (NPs)

2.3

The stability of the prepared NPs was systematically evaluated over time, as maintaining a stable dispersion is critical for reliable cell‐based experiments. Figure 4 illustrates both the size stability and the corresponding ζ‐potential of the NPs, **NP_1–NP_6, ** during the observation period. The larger particles produced via mechanical dispersion (**NP_1_ **) exhibited the most pronounced changes in hydrodynamic diameter. Their average size decreased significantly from 680 to 430 nm, accompanied by a reduction in suspension concentration from 25 to 8.5 µg/mL. This trend is likely attributable to the sedimentation of larger aggregates, effectively reducing the concentration of dispersed PEF. In contrast, nanoparticles prepared using a confined impinging jet mixer demonstrated considerably enhanced stability over the monitoring period. Specifically, **NP_2_ ** and **NP_3_ ** retained relatively constant sizes of ∼90 and ∼215 nm, respectively. Similarly, NPs obtained through the simple precipitation method (**NP_4–NP_6 **) exhibited long‐term stability due to the presence of Pluronic F‐127 surfactant, which reduced the aggregation and precipitation of the particles. ζ‐potential of all formulations displayed slight fluctuations over time, initial values ranged from −30 to −15 mV, before gradually stabilizing at approximately −15 mV.

Hydrodynamic diameters and ζ‐potential of PEF NPs over time measured by DLS.

The long‐term stability of the **NP_7_ ** particles was probed for their use as a reference material. DLS measurements were performed for 30 months, revealing that the hydrodynamic diameter stays constant at 146 ± 15 nm. Additionally, the PDI and zeta potential did not change significantly over the 30 months of storage at ambient conditions. For details, see characterization and stability of the **NP_7_ ** nanoparticles in the Supporting Information. Tentatively, 10 vials with PET were heated for 2 h at 121°C to estimate any heat sensitivity of the particles. As a result, no significant changes in hydrodynamic diameter, polydispersity, or zeta potential were detected. The surprisingly high colloidal stability of **NP_7_ ** may result from the highly negative zeta potential.

s‐SNOM Analysis of PEF NP (sized below 200 nm)

2.4

In contrast to the detection of polymer microplastic/particles, which is easily accomplished by a wide range of analytical techniques (IR‐, Raman, FFF, SEM, etc.), the detection of smaller polymeric NPs with a sufficient sensitivity represents a significant challenge. Here, we demonstrate the ability to image PEF‐NPs using atomic force microscopy (AFM) and, at the same time, identify their chemical composition (nano‐FTIR). To this end, we use scattering scanning optical near field microscopy (s‐SNOM) and perform Fourier transform infrared (FTIR) spectroscopy on the nanoscale [29]. This method is described in detail in the Section 4 and Figure 5A) shows an atomic force microscope (AFM) image of PEF NPs (PEF‐NP_3_) with sizes ranging from 50 to 200 nm dispersed on a silicon substrate. In Figure 5B), nano‐FTIR spectra are shown obtained from six selected PEF NPs. The corresponding measurement locations are indicated in the AFM image (Figure 5A). Note that nano‐FTIR spectra are measured in terms of the phase of the response, while the standard FTIR spectrum of the PEF reference spectrum shows the transmission directly and was scaled to indicate the PEF absorption lines (Figure S8C). In the nano‐FTIR spectra, the characteristic C─O vibrations at 1260 cm^−1^ can be clearly identified for all PEF NPs. The signal amplitude in our spectra is limited by the light power of the broadband infrared light source, and therefore, only the strongest absorption can be resolved in the present data. However, we conclude that this method is sensitive enough to detect fingerprint absorption lines of individual PEF NPs.

(A) AFM image of PEF‐NP3 with a nominal size of up to 200 nm on Silicon. (B) FTIR spectrum of a pressed powder PEF sample (reference, obtained by conventional FTIR spectroscopy) shown in black and nano‐FTIR spectra obtained for a continuous PEF film (obtained from heated PEF micro particles) shown in grey and for six individual PEF NPs A‐F as indicated in (A) shown in different colors, respectively. The different FTIR spectra have been vertically offset for clarity.

Conclusion

3

This work demonstrates a straightforward strategy for the preparation of fluorescent‐labeled polymer nanoparticles, overcoming the limitation of the detection of nanoplastics in environmental and biological systems. We successfully prepared poly(ethylene terephthalate) (PET) and poly(ethylene furanoate) (PEF) nanoparticles and microparticles, covalently tagged with Alexa Fluor 633 or Alexa Fluor 647 for further biological testing, using both mechanical and solvent‐based approaches.

The resulting particles, with hydrodynamic diameters between 200 and 700 nm, exhibited long‐term colloidal stability in aqueous suspension for up to 57 days and covered a wide concentration range (10 µg/mL–5.88 mg/mL), with zeta potentials of −5 to −50 mV. Whereas mechanical stress during the preparation by the dispersion‐method resulted in a decrease in molecular weight and thus changes in the thermal properties of the particles in comparison to the initial bulk‐polymers, subsequent SEC, DSC, and XRD analysis confirmed that solvent‐based preparation preserved the molecular weight, thermal properties, and crystallinity of the pristine polymers. This also resulted in the generation of standardized and reproducible dispersions, now allowing to conduct biological testing without the use of added surfactant. Near‐field optical imaging, SNOM, allows the detection of nanosized particles and identifies their chemical identity—an achievement that, in the future, allows the detection of nanosized particles inside more complex materials, such as tissue. The successful incorporation of Alexa dyes was verified by fluorescence spectroscopy, enabling future biological tracking of PEF‐ and PET‐based particles. Together, these results provide a versatile platform for generating stable, fluorescently labeled nanoplastic reference materials, paving the way for systematic studies of their environmental fate, biological interactions, and potential impacts.

Experimental Section/Methods

4

Materials

4.1

All chemicals were used as received without further purification. Acetone (ACS reagent, ≥99.5%), Pluronic F‐127 (powder, BioReagent, suitable for cell culture), and dimethyl sulfoxide (DMSO; ACS reagent, ≥99.9%) and ethylene glycol (99.8%, anhydrous) were purchased from Sigma–Aldrich. Sodium chloride (NaCl; ≥99%) was obtained from Carl Roth GmbH. Sterile water for cell culture (high grade) was purchased from PAN Biotech. 1,1,1,3,3,3‐Hexafluoro‐2‐propanol (HFIP; 99%) was obtained from Carbolution Chemicals. Furan‐2,5‐dicarboxylic acid (98%) was obtained from abcr. Poly(ethylene terephthalate) PET Lighter C93 (standard bottle grade) was purchased from Equipolymers, Netherlands. Note that the reference material BAM‐P206 – a PET microplastic powder—was produced from the same product. The water for the preparation of PET samples was purified using a Sartorius Arium 611 DI purifier. Folded filters of grade 2105 (fast filtering, particle retention 12–15 µm, by LabSolute) were purchased from Th. Geyer.

For preparing PEF nanoparticles, all glassware was cleaned using a KOH–isopropanol base bath, thoroughly rinsed with distilled, endotoxin‐free water and acetone, and then dried in an oven at 150°C for at least 1 h prior to use.

Synthesis of PEF

4.2

Dimethyl furan‐2,5‐dicarboxylate (3 g, 16.3 mmol), ethylene glycol (2 g, 32.2 mmol), and Sb_2_O_3_ (8 mg, 0.027 mmol) were placed in a flask equipped with a magnetic stirring bar and distillation apparatus under a nitrogen atmosphere. The reaction was carried out at 180°C for 16 h, after which the collected methanol was removed from the receiver flask. Subsequently, the reaction temperature was increased to 230°C, and a vacuum of 100 mbar was applied for 3 h. The pressure was then further reduced to 10 mbar, and the reaction was run for an additional 2 h. The polymer was purified by repeated precipitation from HFIP into acidified (HCl) methanol (three cycles). Yield 60%.

Preparation of PEF NPs with Ultra Turrax Disperser (IKA)

4.3

500 mg of PEF powder was weighed into a high‐walled 200 mL glass tube and dispersed in 115 mL of acetone. The suspension was processed using a T18 digital Ultra‐Turrax disperser (IKA) at 25 000 rpm for a total of 40 min, with 5 min cooling intervals after every 10 min of operation, while maintained in an ice bath to suppress solvent evaporation. The resulting acetone dispersion was filtered through a 5–13 µm paper filter to remove aggregate debris, yielding a turbid filtrate. To exchange the solvent, the acetone was reduced to 40 mL with a rotary evaporator and poured into 115 mL endotoxin‐free water (sterile water‐suitable for cell culture; manufactured by PAN Biotech, Germany). The dispersion remained turbid, and residual acetone was subsequently removed by rotary evaporator. The dispersion was analyzed by DLS and ELS to confirm nanoparticle size and stability, respectively.

Preparation of PET NPs with Ultra Turrax Disperser (IKA)

4.4

A total of 6.0 g of PET granules was placed into a tall 400 mL glass beaker. Then, 115 mL of acetone was added to the beaker. The beaker was cooled in a water and ice bath. Then the polymer was processed using a T18 digital Ultra‐Turrax disperser (IKA GmbH Staufen, Germany) at 18 000 rpm for 10 min. After the processing, the disperser tool was cleaned with 5 mL of additional acetone to remove plastic debris. The acetone dispersion was filtered to eliminate larger plastic particles. To replace acetone with water as the solvent, acetone was evaporated to approximately 30 mL. Then, 115 mL of purified water was added to the dispersion, causing it to turn turbid immediately. The remaining acetone was evaporated, followed by the addition of 115 mL of acetone. The beaker was cooled in an ice bath to prevent acetone from evaporating during preparation. After dispersion, the solution was filtered through a folded filter to remove larger polymer aggregates. Acetone was evaporated until about 10% of the liquid remained. Then, 115 mL of water was added to the mixture, and the remaining acetone was evaporated to create an aqueous dispersion of the polymer nanoparticles. The dispersion was then filtered again with a folded filter to remove any particles that might have aggregated during transfer to the water.

Preparation of the reference materials—The reference material was prepared following the described method, repeated 30 times. Batches with a D h z‐score between −2 and +2 were combined in a 5‐L bottle, then gently swirled to mix. It was dispensed with a BRAND seripettor into small screw‐top bottles. After every 15th fill, the bottle was swirled again for thorough mixing. Ultimately, 200 bottles, each containing 10 mL, were filled. To improve durability, the samples were pasteurized at 90°C for 3 h. These samples were used for measurements.

Preparation of PEF with Confined Impinging Jet Mixer (CIJ)

4.5

Polymer solutions for CIJ injection were prepared in HFIP at 2.5–5 mg mL^−1^ by dissolving PEF powder (Mn ≈ 12 kg mol^−1^, Mw ≈ 22 kg mol^−1^) overnight at room temperature. Milli‐Q water was used as both antisolvent and capture medium; for nanoparticle preparation, both contained NaCl (2.5 mg mL^−1^). Polymer solutions (∼20°C) and antisolvent (∼20°C) were co‐injected through opposite CIJ ports at ∼2 mL s^−1^ (injection time < 3s). The effluent was collected in a vigorously stirred (1000 rpm) capture reservoir (100 mL Milli‐Q water, 10x injection volume). Typical injection volumes ranged from 6 to 10 mL, producing 66–110 mL nanoparticle dispersions. To remove HFIP and NaCl, dispersions were dialyzed with dialysis membrane (Spectra/Por7 Dialysis Membrane; Pre‐treated RC Tubing; MWCO: 1 kD; Nominal Flat Width: 38 mm, Diameter: 24 mm; Vol/Length: 4.6 mL/cm; manufactured by Repligen Corporation) against Milli‐Q water for 9–14 days. Final dispersions were filtered through 5–13 µm paper filters, and the dialysate was analyzed by DLS and ELS to confirm nanoparticle size and stability, respectively.

Preparation of PEF by Precipitation

4.6

In a 100 mL beaker, 20–50 mL of DMSO was heated to 180°C, and 200–500 mg of PEF powder was added with stirring at 250 rpm using a magnetic stir bar until fully dissolved (∼5 min). Subsequently, 200–500 mg of Pluronic F – 127 was added to the PEF solution, dissolving immediately, followed by the addition of 1–2.5 µL/mL of Alexa Fluor 647 in DMSO, which induced an immediate green coloration. The heat source was removed and the solution was cooled to 120°C. While stirring, 50 mL of 2 wt.% aqueous Pluronic – 127 solution was added in a single step, causing the solution to change to a light sky‐blue color as the temperature decreased to 70°C. The resulting dispersion was immediately filtered through a 5–13 µm paper filter. Then the dispersion was analyzed by DLS and ELS (Diluted by 100) to confirm nanoparticle size and stability, respectively. The obtained dispersions were dialyzed with dialysis membrane (Spectra/Por7 Dialysis Membrane; Pre‐treated RC Tubing; MWCO: 1 kD; Nominal Flat Width: 38 mm, Diameter: 24 mm; Vol/Length: 4.6 mL/cm; manufactured by Repligen Corporation) against Sterile water (Sterile water—suitable for cell culture; manufactured by PAN Biotech, Germany) until the dialysate conductivity stabilized in 10–14 days.

Preparation of PET by Precipitation

4.7

In a 100 mL beaker, 20 mL of DMSO was heated to 180°C, and 200 mg of PET powder was added with stirring at 250 rpm using a magnetic stir bar until fully dissolved. Subsequently, 200 mg of Pluronic F‐127 was added to the PET solution, which dissolved immediately. This was followed by the addition of 0.1 mL of Alexa Fluor 633 (concentration unknown) in DMSO. The heat source was removed, and the solution was cooled to 110°C. While stirring, 20 mL of 2 wt.% aqueous Pluronic F‐127 solution was added in a single step as the temperature decreased to 70°C. After 15 min, when the temperature reached 50°C, the product was filtered through a 5–13 µm paper filter. The obtained dispersions were dialyzed with dialysis membrane (Spectra/Por7 Dialysis Membrane; Pre‐treated RC Tubing; MWCO: 1 kD; Nominal Flat Width: 38 mm, Diameter: 24 mm; Vol/Length: 4.6 mL/cm; manufactured by Repligen Corporation) against Sterile water (Sterile water—suitable for cell culture; manufactured by PAN Biotech, Germany) until the dialysate conductivity stabilized in 10–14 days.

Size Exclusion Chromatography (SEC) measurements with 0.1 mol/L KTFAc were performed at 20°C on a Viscotek GPCmax VE 2001 from Malvern, applying a PSS PFG precolumn and a PSS PFG main column. As solvent HFIP with 0.1 mol/L KTFAc was used and the sample concentration was adjusted to approximately 3 mg mL^−1^ while applying a flow rate of 0.5 mL min^−1^. The refractive index detection was performed with a VE 3580 RI detector of ViscotekTM. For determination of the molecular weights, external calibration was done using polymethyl methacrylate (PMMA) standards (purchased from PSS) with a molecular weight range from 602 to 182 000 g mol^−1^. OmniSEC software (Version 5.12) was used for evaluating data.

Fourier‐transform infrared spectroscopy (FT–IR) measurements were done on a Bruker Tensor VERTEX 70 spectrometer, and Opus 8.2 was used for data analysis. For Attenuated total reflection‐infrared (ATR‐FTIR), a Golden Gate Heated Diamond ATR Top‐plate was used.

Nano‐Fourier‐transform infrared spectroscopy (nano‐FTIR) measurements were performed using a scattering scanning optical near field microscope (s‐SNOM) (Attocube/Neaspec Nanostation 1) using a home‐built mid infrared light source based on difference frequency generation in a GaSe crystal with similar specifications as reported previously [29]. Nano‐FTIR was used to obtain nanoscale‐resolved infrared spectra from individual NPs with a spatial resolution of about 30 nm. In this technique, a metal‐coated atomic force microscope (AFM) tip was illuminated with a focused, broadband IR beam. The tip acted as an antenna, localizing the infrared field to a region of tens of nanometers at its apex. The light scattered by the tip–sample interaction was collected and analyzed using a Fourier‐transform infrared spectrometer. Demodulation of the scattered signal at higher harmonics of the tip‐tapping frequency isolated the near‐field component, providing nano‐scale chemical contrast. The mid infrared light was focused onto a standard platinum‐coated AFM tip (purchased from NanoWorld), which was in close proximity to the sample. The AFM tip acted as an optical antenna and led to strongly enhanced near fields close to the tip apex. This enabled spatial resolution on the order of the tip radius, while the near fields also penetrated into the sample [30]. We detect at the second harmonic of the tapping frequency (2 Ω). Detection of the tip‐scattered light yielded the complex‐valued spectral scattering coefficient σ=seiφ, which can be decomposed into the near‐field (nano‐FTIR) amplitude s and phase φ spectra. For quantitative analysis, all spectra were normalized to the nano‐FTIR response of a clean silicon substrate according to σnorm=σ/σSi [30]. We study organic materials and for such systems, it is well‐established to evaluate the data as normalized nano‐FTIR phase spectra  φnorm(ω)=arg(σ/σSi), which qualitatively relate to the absorptive properties of molecular samples with high sensitivity [31, 32, 33, 34].

Matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry measurements were performed on a Bruker autoflex maX MALDI TOF/TOF System (Bruker Daltonics) using a BRUKER smartbeam‐II nitrogen laser, operating at 355 nm wavelength. The used Dithranol:PEF:KTFA ratio was 50:50:1, and 1 µL of the mixture was spotted on the MALDI target. The polymer samples were dissolved in HFIP with a concentration of 1.5 mg/mL, and dithranol was used as a matrix, adjusting the concentration to 20 mg/mL in HFIP. KTFA was used as salt with a concentration of 5 mg/mL in HFIP. Data evaluation was carried out via flexAnalysis software (3.4), and simulation of the isotopic pattern was performed by Data Analysis software (version 4.0).

Polymer Concentration Determined via Gravimetric Analysis

4.8

Gravimetric analysis was conducted using open aluminium crucibles with a volume of 85 µL. Each crucible was initially weighed on the TGA instrument, pre‐heated on the heating plate to 120°C, and subsequently loaded with a 2 mL aliquot of the sample, added dropwise, and evaporated to prevent spillage and ensure quantitative transfer. The crucibles were then maintained at 120°C for 15 min to achieve complete solvent evaporation. Afterward, they were reweighed on the TGA instrument to determine the net mass increase.

Fluorescence Spectroscopy

4.9

Fluorescence spectra were recorded using a Cary Eclipse spectrometer (Agilent) with Helma quartz cuvettes (10 mm path length). Temperature was controlled via a Cary Single Cell Peltier Accessory (Type SPVF 1×10, Agilent). Nanoparticle samples were diluted 20‐fold in sterile water prior to measurement. Excitation was set at 600 nm, with a detector voltage of 600 V. All measurements were performed in triplicate, and data were processed and visualized using Origin 2019 (OriginLab, USA).

Dynamic Light Scattering (DLS) and ζ‐Potential

4.10

Hydrodynamic diameter (D_h_; 0.3 nm–100 µm) and ζ‐potential were measured by Dynamic Light Scattering (DLS) and Electrophoretic Light Scattering (ELS), respectively, using a Litesizer 500 (Anton Paar, Germany) equipped with a wavelength of 658 nm, 40 mW single‐frequency laser diode. Scattering intensity was collected at 173° (backscattering) with the detector maintained at 35°C. Measurements were performed in a Univette cuvette (50–900 µL) with palladium electrodes, enabling simultaneous determination of particle size (3.8 nm—100 µm) and ζ‐potential (up to ± 1000 mV). Glass cuvettes were pre‐cleaned using acetone before being filled with the sample solution. Each measurements were repeated three times (20s × 10). All data was processed using Kalliope software.

Differential scanning calorimetry (DSC)

4.11

Data were collected on a calibrated heat‐flux DSC (Mettler‐Toledo, Greifensee, Switzerland) equipped with an FRS5 sensor, connected to a TC100 Intracooler (Huber, Offenbach, Germany). 5–10 mg samples were placed into aluminum crucibles and were heated under a nitrogen atmosphere. Initially, samples were preheated to 240°C with a rate of 10 K min^−1^. After cooling to 0°C with the rate of 10 K min^−1^ samples were heated a second time to 240°C with the same heating rate.

XRD

4.12

Measurements were performed with an Incoatec (Geesthacht, Germany) IµS equipped with a microfocus source and a monochromator for CuKα radiation (λ = 1.5406 Å). 2D scattering patterns were recorded with a Vantec 500 2D detector (Bruker AXS, Karlsruhe). All samples were measured in transmission mode at 2 different sample‐detector distances.

SEM Analysis

4.13

For morphological analysis, the nanoparticle suspensions were deposited onto silicon (Si) wafer substrates. The Si wafers were cleaved into 1 cm × 1 cm pieces and cleaned by ultrasonication in isopropanol, followed by drying in air. A 10 µL droplet of the nanoparticle suspension was then drop‐cast onto the cleaned surface of the Si wafer and allowed to dry fully under ambient conditions in a covered petri dish to prevent contamination. The samples were mounted on standard aluminum SEM stubs using metallic clips and analyzed without the application of a conductive sputter coating to observe the materials in their pristine state.

The morphology of the nanoparticles was characterized using a FEI Quanta 650 FEG Scanning Electron Microscope (SEM). Imaging was performed in high vacuum mode using an accelerating voltage of 3 kV and a working distance of approximately 10 mm to achieve high‐resolution images. Secondary electron (SE) imaging was utilized to visualize the surface topography of the nanoparticles.

Particle size distribution was determined from the obtained SEM micrographs using the ImageJ software [35]. After calibrating the image scale, a brightness threshold was applied to create a binary image that isolates the particles from the substrate background. The “Analyze Particles” function was then used to measure the projected area of individual nanoparticles, from which their equivalent diameters were calculated.

Author Contributions

Conceptualization, W.H.B., A.F.T., A.M., and Z.K.; data curation, R.G.S., A.F.T., and Z.K.; formal analysis, R.G.S., A.M., A.F.T., M.D., R.A., K.B., and Z.K; funding acquisition, W.H.B. and A.M.; investigation, R.G.S., A.M., A.F.T., R.H., G.W., M.D., K.B., J.M.S.S., and Z.K.; methodology, W.H.B., A.F.T., and Z.K.; project administration, W.H.B., A.F.T., and Z.K.; resources, W.H.B., A.F.T., and Z.K.; supervision, W.H.B., A.F.T., and Z.K.; validation, R.G.S., W.H.B., A.F.T., and Z.K.; visualization, A.M., M.D., K.B., J.M.S.S., A.F.T., and Z.K; writing – original draft preparation, R.G.S., W.H.B., A.F.T., A.M., and Z.K.; writing – review and editing, W.H.B., A.F.T., A.M., and Z.K.; All authors have read and agreed to the published version of the manuscript.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File: marc70179‐sup‐0001‐SuppMat.docx.