Feasibility of Solvent-Cast PLLA/Iron Composites for Biomedical Applications

Abstract

Degradable polymers, such as poly(L-lactide) (PLLA), are widely investigated for biomedical applications, including drug delivery systems and temporary implants. Their functionality can be expanded by incorporating degradable metal microparticles that may influence degradation behaviour and enable additional surface modification strategies. In this study, the feasibility of composites consisting of PLLA and biodegradable iron microparticles was investigated. Composites were fabricated by solvent casting, providing a gentle alternative to thermal processing methods, which often compromise polymer integrity. Composites were evaluated by thermogravimetric analysis, differential scanning calorimetry, scanning electron microscopy (SEM), tensile testing, dynamic mechanical analysis, and X-ray photoelectron spectroscopy (XPS). Incorporation of iron altered thermal behaviour and crystallinity of PLLA, indicating interactions between polymer matrix and dispersed metal phase that may affect degradation kinetics and material stability. While iron addition reduced Young’s modulus, tensile strength, and elongation at break, composites maintained sufficient structural integrity for potential biomedical applications. XPS and SEM confirmed the embedding of particles within the polymer matrix, enabling potential post-processing approaches. In vitro direct contact and eluate tests demonstrated good cell viability, whereas exposure to free iron particles resulted in dose- and time-dependent cytotoxic effects. Overall, the results demonstrate the feasibility of solvent-cast PLLA–iron composites for resorbable biomedical applications.

1. Introduction

A combination of polymers with other materials results in composites with specific properties. That ranges from mechanical ones to surface characteristics and following biological behaviour, allowing the generation of materials with defined mechanics and biological activity [1]. Composites have several degrees of freedom that can be used to set the desired material structures and property profiles and to combine the advantages of several material components. These degrees of freedom include the number, type, proportions, geometry, size, and spatial distribution of the components. Fields of application using absorbable composites open up to medicine with regard to tissue engineering, regenerative medicine, drug release systems, as well as wound dressings and dental materials, orthopaedic or vascular grafts [1,2]. Overall, combinations of silver, zinc, magnesium, and iron as metallic components and polylactides or polycaprolactone as polymeric components are frequently studied.

Silver is of significant importance in polymer–metal composites due to its biological properties, such as biocompatibility, cytotoxicity, antimicrobial and antibacterial activity, as well as electrical properties, and is used in the form of submicron wire, nanowires, and nanoparticles [3,4,5,6,7,8]. Magnesium as a component in polymer–metal composites causes an improvement of mechanical properties compared to unloaded polymers. Moreover, a pH buffer effect is achieved by combining base-degrading magnesium with acid-degrading polymers [9].

Iron is another resorbable and biocompatible material, exhibiting higher mechanical strength and Young’s modulus than magnesium. It has a low degradation rate without gas release in contrast to magnesium [10,11,12]. Polymers like biodegradable poly(L-lactide) (PLLA) and polycaprolactone (PCL) are common ones to be used in composites. PLLA has been widely used in FDA-approved medical devices and drug-delivery systems and is considered to exhibit low toxicity compared with many other synthetic polymers [13,14]. It has wide application in medical implants since it maintains mechanical and structural integrity within in vitro and in vivo applications [13,15,16,17,18]. Nevertheless, its hydrophobic surface might inhibit the biocompatibility by affecting protein absorption and resulting in cell adhesion [19]. Mechanical properties of PLLA scaffolds include tensile strength in a range of 50–70 MPa, Young’s modulus of 2–16 GPa and a comparatively low elongation at break (2–6%) depending on molecular weight, material substrate or fabrication process [20,21,22]. Glass transition temperatures above 60 °C and melting temperatures between 170 °C and 180 °C depend on material purity [13]. PLLA offers high chemical stability, a manufacturing-dependent crystallinity, and a resistance to enzyme degradation [13]. Therefore, it shows slow degradation rates with altered mechanical properties [13] of about 30 to 40 weeks in vitro and in vivo, which might result in delayed inflammation [19]. Addition of particles is conducted to tailor this behaviour [23]. Estrada et al. showed an increase in degradation after incorporation of iron (31.5 wt%) and magnesium (5 wt%) particles into PLLA in PBS over 14 and 28 days, with resulting sample swelling, degradation products, and spot-wise colour changes on the surface as well as further improved cytocompatibility [23]. PLA micro-iron composites are mainly processed by additive manufacturing or electrospinning [24,25,26]. In some studies, the polymer is used as a binder material to enable additive extrusion and is removed by sintering after shaping [27,28,29]. In other investigations, the optical as well as thermal properties of the additively manufactured parts are improved by the addition of iron or steel particles [30,31]. Pandey et al. investigated a composite of nano iron (8% w/w) and the absorbable polymer PLLA, from which micro-particles for water purification were produced by electrospraying [32]. Wang et al. showed an improved thermal stability and tissue response by adding 20% Fe₃O₄ nanoparticles to PLLA for bone applications [33]. Three-dimensional-printed PLA and electrospun PLLA composites containing defined amounts of Fe₃O₄ to add magnetic properties for biomedical and X-ray applications [34,35] and PLLA-HAP-Fe₃O₄ scaffolds for heat-controlled biomedical applications [36] were recently described. Fe-ion nanoparticles embedded in PLLA sheets showed promising results as a potential wound dressing in skin burn treatment [37]. Rath et al. investigated the fabrication of PCL-carbonyl iron particle composites using the extrusion 3D printing technique for microwave shielding applications. Iron particles with a diameter of approx. 3 µm and a particle content of 2–18% were used to produce samples with dimensions in the 10 mm range [38,39]. Singh et al. characterised the biological and mechanical properties of composites of PCL and carbonyl iron (0–2%) produced by solvent casting, as well as solvent cast three-dimensional printing, and saw applications of the material for cardiovascular stents [40]. Composite scaffolds based on iron foam and PCL enable the regulation of mechanical and degradation properties [41]. With electrically conductive interconnects for transient implantable systems, Zhang et al. explored another application of PCL–iron composites [42,43]. The screen-printing process is used to produce samples with iron contents between 10 and 50% volume fraction and a length of several centimetres. Iron micro-particles may be used to alter the rheological behaviour of polymers, which was shown by Chuah et al. in polystyrene foam [44]. A combination of iron and PLLA is a common combination for piezoelectric biomedical applications, e.g., in bone fracture treatment [45]. Moreover, due to its magnetic properties, iron is able to add actuator properties to polymers, paving the way for the construction of soft robotics [46]. Overall, iron as a component in polymer–metal composites for biomedical or other applications has been described comparatively little so far.

Polymer processing methods, such as extrusion or injection moulding, that require the polymer to be melted, are known to lead to thermal degradation [47,48]. This may also be valid for a variety of drugs that need to be incorporated into a PLLA bulk to be released in a sustained manner. Moreover, the cooling step after the melt process leads to the formation of internal stresses that may consequently lead to deformation or even cracking of the biomaterial [13,24]. In contrast, solvent-based processes offer a gentler way of producing the desired biomaterials. For these reasons, a solvent-based casting process is used in this study that ensures control over the amount of embedded iron particles. This also takes into account that a simple and cost-effective processing is a particularly important requirement in an industrial context. Other solvent-based processes, such as electrospinning, spraying, or dipping processes, can additionally be established with less effort. However, these methods will be inferior in order to control the amount of the embedded particles.

The aforementioned studies show a substantial interest in polymer–metal composites with diverse property profiles and processing methods. The principal source of novelty lies in the design of biodegradable scaffolds that uniquely combine the proven biocompatibility and degradability of PLLA with the mechanical reinforcement, degradation modulation, and functional responsiveness provided by metallic iron inclusions. Therefore, in the current study, novel PLLA+Fe composites were manufactured by solvent casting and characterisation methods from the various fields were combined with the aim to provide a systematic physicochemical, mechanical, and biological characterisation. The broad range of used methods shall prove their feasibility in a wide range of applications, particularly in the field of biomedical engineering.

2. Materials and Methods

2.1. Materials and Production Process

Poly(L-lactide) (PLLA) (Homopolymer “RESOMER^® L210”, medical grade, Evonik Industries AG, Essen, Germany) with a Mw of ~300,000 g/mol, inherent viscosity of 3.3–4.3 dL/g) was used as received as a carrier matrix. Iron particles “EMSURE^® Iron” (purity ≥99.5%) with a mean diameter of ø = 10 µm were purchased from Sigma Aldrich (Merck KGaA, Darmstadt, Germany). Acetone, chloroform, and methanol were purchased from J.T.Baker (Fisher Scientific GmbH, Schwerte, Germany).

PLLA films with and without iron particles were prepared. PLLA was dissolved in chloroform to a final concentration of 4% (w/w) by means of a magnetic stirrer. After the addition of iron particles (5% w/w with respect to PLLA), the solution was homogenised by stirring (Heidolph Instruments GmbH Co. KG, Schwabach, Germany). Moreover, 25 mL of the resulting suspension was poured into a Petri dish with a diameter of 10 cm. The solvent was allowed to evaporate for at least 24 h, and the resulting iron–polymer films were stored flat under a compressive load of 1 kg for an additional three days to prevent warping. For PLLA films without iron, a washing procedure was carried out: initially, two consecutive washing steps with methanol, each for one day, were performed, followed by two additional washing steps in water, with each step maintained again for one day. The washing procedure was performed at room temperature without agitation. The films were stored in a vacuum chamber at 40 °C. In order to prevent alteration, e.g., corrosion of iron particles, iron–polymer films were not washed, but subsequently stored at 40 °C in a vacuum chamber to evaporate remaining solvents.

2.2. Morphology and Surface Analysis

To investigate the microstructure of the specimens, materialographic images of compounds were taken using an optical microscope (Leica DMI 5000M, Leica, Wetzlar, Germany) and a scanning electron microscope (SEM, Zeiss Merlin VP compact Co. Zeiss, Oberkochen, Germany) with an integrated detector for energy dispersive X-ray spectroscopy (EDX; XFlash 6/30) and analysis software (Bruker Quantax 400, Berlin, Germany).

The materialographic preparation of cross sections of the film was carried out with very fine abrasives (2500 and 4000 SiC) due to the deviating mechanical properties of the matrix and particles. This procedure avoids the detachment of particles from the matrix as much as possible. Subsequent polishing (3 µm diamond suspension) was only done on part of the samples. The solvent cast PLLA+Fe film was bonded to a flat epoxy block with adhesive conductive carbon tape (Co. Plano, Wetzlar, Germany) and then finely ground by hand (4000 SiC) for optical microscope images. SEM analysis was used to examine separated pieces of film without materialographic preparation using only carbon sputtering. Cut surfaces as well as untreated surfaces were examined by SEM. The Imagic IMS software Version 22 (Imagic Bildverarbeitung AG, Glattbrugg, Switzerland) with the particle analysis module was used to evaluate the particle content of the materialographic images. At least six images of each material were observed for the analysis.

X-ray Photoelectron Spectroscopy (XPS) qualitative analysis was performed as described elsewhere [48] with X-ray photons of 1486.6 eV from an Al Kα monochromatic source (Thermo K-Alpha, Thermo Scientific, Inc., Oxford, UK) with a spot size of 200 µm. The film was etched using an Ar+ ion beam with an energy of 4000 eV and step times of 60 s. In each etch cycle, Fe2p- and Cl2p- spectra were obtained by high-resolution scans. The flood gun was used to achieve charge compensation. Detailed scans were recorded with a pass energy of 30 eV. All spectra were referenced to the C1s peak of hydrocarbon at 285.0 eV binding energy.

2.3. Thermal Analysis

The thermogravimetric analysis was carried out in a Perkin Elmer Pyris 6 TGA. Samples with a diameter of 5 mm were punched from solvent-cast films and analysed in open pure aluminum crucibles under a nitrogen flush of 200 mL/min. Several samples have been stacked for a total mass of about 6–7 mg. The temperature programme consists of heating from 30 °C to 600 °C at a heating rate of 10 K/min, holding at this temperature for one minute, and cooling to 30 °C at 10 K/min. The TGA curves show the relative mass calculated according to Equation (1). A curvature in the measurement curves was mathematically removed by subtracting a baseline mass (m_BL) from the measurement mass (m_meas). The baseline measurement was performed with an empty pure aluminium crucible using the same temperature programme. For comparison, all curves were normalised by dividing the mass by the respective initial mass (m₀). The onset of decomposition was defined at 1% weight loss.

$$m_{rel}={\displaystyle \frac{m_{meas}-m_{BL}}{m_0}}$$

(1)

The differential scanning calorimetry (DSC) method is used to measure temperatures of glass transitions, crystallisation, and melting processes of polymers. In this work, the DSC experiments were carried out in a Mettler DSC 823e/700_cryo. Samples packed in closed pure aluminum crucibles and an empty crucible as a reference were used. Samples with a diameter of 5 mm were punched from solvent-cast films. Several samples have been stacked for a total mass of about 10–12 mg. The temperature programme used for all materials consists of the following steps: 1. cooling from room temperature to 100 °C, 2. holding for 15 min, 3. heating to 200 °C, 4. holding for 1 min, 5. cooling to 100 °C, 6. holding for 15 min, 7. repeating steps 3–6 one time, 8. heating to room temperature with a heating/cooling rate of 10 K/min. During the first heating, the influences of the manufacturing process are to be determined. As a result of heating to 200 °C, the thermal history of the material is erased [49,50], and during the second heating, a reference of the material is measured. Between two measurements, baseline measurements were carried out with two empty crucibles. Two experiments with material samples were conducted for each material, and a baseline measurement was taken in between. Disks of solvent casting films (diameter 6 mm) were punched out of the solvent casting materials. These discs were adapted to the geometry of the pure aluminum crucibles used by stacking four discs on top of each other.

2.4. Mechanical Analysis

Tensile tests were performed with a uniaxial tensile test instrument, zwickiLine Z 1.0 (Zwick/Roell, Ulm, Germany), equipped with a 100 N load cell and a crosshead speed of 25 mm/min. Standard dogbone-shaped samples according to DIN ISO 527 (type 1BB) were prepared by die cutting from cast film samples by means of a toggle press EP 500-40 TB (mäder pressen GmbH, Neuhausen ob Eck, Germany) [51]. The average sample thickness measured at three positions using a dial gauge Mitutoyo 543-394B with associated stand (Mitutoyo Corporation, Kawasaki, Japan) was 130 ± 20 µm for pure PLLA and 245 ± 50 µm for PLLA+Fe composites. Measurements were carried out at 37 °C, while the test temperature was ensured using an in-house constructed thermochamber. Tensile force as a function of elongation was measured. Uniaxial test data were analysed regarding Young’s modulus between 1% and 2% strain, tensile strength, and nominal elongation at break. Nominal elongation at break was reported as the nominal strain at break after necking, calculated from the change in grip-to-grip distance. Results were averaged over n = 5 samples.

Dynamic mechanical analyses (DMA) to determine the cyclic mechanical stability of pure PLLA and PLLA+Fe composites were performed using a DMA 850 test bench (TA Instruments Inc., USA) equipped with a tension test setup. Rectangular-shaped specimens (7 × 20 mm²) were cut out from cast films with a scalpel and were uniaxially deformed with a 5 mm clamping length. Therefore, a force-controlled test protocol with a force of 8 N was applied for 432,000 load cycles using a test frequency of 5 Hz, which results in a measuring time of 24 h. To ensure that the samples were not irreversibly deformed, a force below the yield strength determined in uniaxial tensile tests was selected. For comparability, the same force value was used for both PLLA and PLLA+Fe. The tests were conducted under a constant temperature of 37 °C. Results for the maximum strain per cycle were averaged over n = 3 samples.

2.5. Cell Biological Investigations

Cell cultivation was done in an incubator (Memmert ICO240med, Memmert GmbH, Schwabach, Germany) at 5% CO₂, 21% O₂, and 37 °C. Cell culture media and additives were purchased from PAN Biotech GmbH, Aidenbach, Germany, unless otherwise stated.

Tissue culture polystyrene (TCPS), named as the negative control (NC), further served as a common surface for growth control. A positive control group (PC) was treated with a cytotoxic concentration of 10⁻⁴ mol/L disulfiram/tetraethylthiuram disulfide (TETD). High-density polyethylene (HDPE) was included as a negative material control within eluate testing according to DIN EN ISO 10993-12.

Due to potential cardiovascular applications, the human endothelial hybrid cell line EA.hy926 (ATCC CRL^®-2922™) and fibroblast cell line HT-1080 (ATCC^® CCL-121™) (both: LGC Standards GmbH, Wesel, Germany) were used for initial in vitro tests. EA.hy926 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (with 2 mmol/L L-glutamine, 4.5 g/L glucose, and 3.7 g/L NaHCO₃) with 10% fetal calf serum (FCS) and 1% penicillin/streptomycin. HT-1080 cells were cultured in modified Eagle’s medium (MEM) (with 2 mmol/L L-glutamine, 1 g/L glucose, and 1.5 g/L NaHCO₃) with 10% FCS and 1% penicillin/ streptomycin. Cells were harvested from the culture flask using trypsin/EDTA.

For direct contact testing, material test samples (∅ 6 mm) were weighed down using PTFE rings in 96-well plates and directly seeded with cells at a density of 10,000 cells/well (∼29,500 cells per cm²) in 200 μL of culture medium in four parallel wells for every material configuration, respectively, according to DIN EN ISO 10993-5 [52] and cultured over 48 h to verify initial cell reaction and adhesion. Eluates of test samples were tested by generating material eluates according to DIN EN ISO 10993-12 [53] by incubating test material in cell culture medium (6 cm²/mL) for 24 ± 2 h at 37 ± 1 °C. Eluates were added to pre-cultivated cells (24 h) in logarithmically diluted concentrations of 100%, 31.6%, and 10% for further 48 h.

To analyse the influence of iron particles themselves, cells were cultivated with particle concentrations of 2.5, 1.5, and 0.5 mg/mL in cell culture medium (~1 mg/mL ≙ 5 wt% iron particles in PLLA+Fe) directly with seeding or after 24 h pre-cultivation of cells, respectively. Three independent experiments were conducted. Cultivation was carried out over 48 h without any medium changes.

The metabolic activity of cells was determined via CellQuanti-Blue assay (BioAssay Systems, Biotrend Chemicals GmbH, Cologne, Germany) after 48 h of cultivation. Non-fluorescent resazurin is transformed into high-fluorescent resorufin by cellular reductases from metabolically active cells. The fluorescence, which is directly proportional to the metabolic cell activity, was measured at an emission wavelength of 590 nm with an excitation wavelength of 544 nm using a microplate reader (FLUOstar OPTIMA, BMG Labtech, Offenburg, Germany). Values were normalised to the negative control (NC) set as 100%.

Cell morphology on polymer samples was analysed by fluorescence staining of cells using phalloidin (AlexaFluor™ 488 phalloidin) for actin filaments as well as DAPI (NucBlue™ Fixed Cell ReadyProbes™) for cell nuclei (both: Invitrogen/Thermo Fisher, Dreieich, Germany), respectively. Therefore, cells were fixed with formalin, washed with PBS, and afterwards stained using one drop of phalloidin and DAPI solution per mL of PBS, respectively, and 100 μL per well. Incubation was carried out for 30 min in the dark at room temperature. For imaging, the Olympus BX53M microscope (Olympus, Hamburg, Germany) was used.

Cell–particle contact was depicted using light microscopy with a NIKON TS 100 (Nikon, Tokyo, Japan) and scanning electron microscopy with a SEM QUANTA FEG 250 (FEI Company, Dreieich, Germany), equipped with an Everhart–Thornley secondary electron detector (ETD) and operating at 10 kV in high vacuum, after glutaraldehyde fixation (2.5%) of cells and subsequent ethanol dehydration series and Au sputter coating.

Potential synthesis of inflammation markers after 48 h of cell cultivation with iron particles was evaluated via flow cytometry (MACSQuant Analyzer 10, Miltenyi Biotec, Bergisch-Gladbach, Germany) using the MACSPlex human cytokine 12 kit according to the manufacturer’s instructions (Miltenyi Biotec, Bergisch-Gladbach, Germany). Supernatants of six wells were pooled per configuration in each experiment. Supernatants of three independent tests have been measured within the cytokine assay according to the protocol. Supernatants were analysed for granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon-α (IFN α), IFN-γ, interleukins 2 (IL-2), IL-4, IL-5, IL-6, IL-9, IL-10, IL-12p70, IL-17A, and tumour necrosis factor α (TNF-α). Concentrations were calculated via standards of known quantities by means of MACSQuantify software Version 2.13.2 (Miltenyi Biotec, Bergisch-Gladbach, Germany).

Data are shown as box plots. Boxes denote interquartile ranges, horizontal lines within the boxes denote medians and whiskers denote minimum and maximum values. Mean values are displayed as dotted lines. Statistical significance was calculated using SigmaPlot 13.0. Data were tested for outliers using the formulae according to Nalimov. Normal distribution of data was evaluated with the Shapiro–Wilk test, and Kruskal–Wallis tests for independent samples was used to determine statistical significance. Values of p < 0.05 were set to be significant.

3. Results

3.1. Microstructure

Microstructure images of the film of PLLA+Fe are shown in Figure 1: (a) macro-image (bottom view), (b) optical microscope image from the bottom side of the film, and (c,d) SEM images of cut cross sections of the film together with an EDS Fe-map of the area. In optical microscopy, particles are only visible when viewed from below (see Figure 1b). This is because heavier metal particles sink to the bottom during solvent casting. Most of the particles in this view are agglomerated. The SEM images show cross-sections through the approx. 100 µm thick film with clearly more iron particles visible in the lower area. Iron particles have been identified by EDS. The interfaces between iron particles and the PLLA matrix mostly show a dense connection of the components. Only in some cases, particles are partially detached from the matrix. Volume fractions of particles in materialographic images from the bottom side, amounted to 1.98%, which is clearly above the volume fraction of particles calculated from their mass fraction (0.76%). This confirms the sinking of the heavier metal particles to the bottom during solvent casting.

3.2. X-Ray Photoelectron Spectroscopy

XPS analyses were carried out with sequential etching steps in order to obtain information on the element distribution also into deeper regions of the film. For this purpose, a survey spectrum and detailed spectra of the C1s, O1s, Cl2p, and Fe2p orbitals were recorded after each etching step.

The C1s spectra reveal a change in the carbon signal during etching and measurement steps, which indicates polymer destruction due to X-ray load. In addition, the detailed spectra of the Fe2p orbital allow the identification of different iron compounds even when they are present side by side. In Figure 2c,d, the corresponding detail spectra of the surface and after five etching steps are shown. No signals for iron are detectable on the surface, while very clear signals can be recognised in the spectra after etching. Iron particles are therefore not visible on the surfaces of the top and bottom sides, respectively, most probably because they are surrounded by PLLA. Only etching exposes the particles and makes them accessible for analysis. Two signals are noticeable in the detailed spectra, for the Fe2p1 at a binding energy of 719 eV and for the Fe2p3 orbital at a binding energy of 706 eV. These signals are specific to metallic iron. Signals for iron oxide (709.6 eV) or ferric chloride (710.4 eV) are not detected. Therefore, oxidation or formation of ferric chloride during the preparation of the film seems unlikely to occur. Also, in deeper layers, no such signals were measured. This analysis is also supported by the Cl2p detail spectra. They were also used to clarify possible residual amounts of solvent. In Figure 2g, corresponding signals of chlorine can be recognised, which proves the presence of chlorinated species and has already been reported by our group [54]. Furthermore, only organic chlorine compounds corresponding to chloroform are visible in the spectra. No signals indicating ferric chloride have been recognised here either.

3.3. Thermogravimetric Analysis

Figure 3 shows the TGA curves of PLLA and PLLA+Fe composite. For solvent-cast PLLA films without Fe particles, a single-stage TGA curve was determined with a start at 321 °C. In comparison, the TGA curve for PLLA+Fe composite consists of two stages, which start at significantly lower temperatures of less than 120 °C and approx. 280–290 °C. EDX analyses show C and O in both samples and Fe and Cl in the composite sample only. This means that the chlorine from the solvent chloroform can still be detected in the films of the composite. According to the TGA curves, the residual chloroform is removed at 105–110 °C, which leads to the observed mass loss of 5%. PLLA+Fe samples stay at a higher residual mass above 350 °C because the iron content remains stable. Thermogravimetric analysis was not performed on the iron powder alone because metallic iron shows negligible mass loss and remains thermally stable up to at least ≈300 °C in TGA experiments, with significant changes only occurring at higher temperatures due to oxidation processes rather than decomposition [55].

3.4. Differential Scanning Calorimetry

Figure 4 shows the DSC curves from heating and cooling twice (a) and details (b) of heating the PLLA+Fe composite and the pure PLLA film, as well as (c) transformation temperatures. The peaks during the heating steps are the glass transition (peak “a”), the crystallisation (peak “b”), the transformation from α’ to α-crystal (peak “c”), and the directly following melting (peak “d”). All these transformation temperatures agree well with those of bulk PLLA material [49,56,57]. The difference between the first and second heating of PLLA+Fe composite, especially regarding peak “b”, indicates a strong influence of the solvent casting process on the subsequent crystallisation of PLLA. Comparing the heating of the PLLA+Fe composite with pure PLLA film reveals slightly lower glass transition and melting temperatures.

3.5. Mechanical Analysis

3.5.1. Uniaxial Tensile Testing

Representative stress–strain curves of PLLA and PLLA+Fe composites are displayed in Figure 5, mean values and standard deviations for Young’s modulus, tensile strength, and nominal elongation at break are summarised in

Table 1. Mechanical properties of pure PLLA and PLLA+Fe composite derived from uniaxial tensile tests.

	Young’s Modulus [MPa] Between 1% and 2% Strain	Tensile Strength [MPa]	Nominal Elongation at Break [%]
PLLA	2130 ± 270	58.9 ± 14.6	84.7 ± 23.1
PLLA+Fe	1450 ± 390	41.6 ± 7.4	47.0 ± 4.6

. The PLLA+Fe composite shows significantly lower values for the Young’s modulus (1450 ± 390 MPa) as well as for tensile strength (41.6 ± 7.4 MPa) and nominal elongation at break (47.0 ± 4.6%) than samples made from pure PLLA.

3.5.2. Dynamic Mechanical Analysis

The results of dynamic mechanical analyses (DMA) over 24 h (n = 3, solid line) and the average curve (dotted line) for pure PLLA and PLLA+Fe samples are illustrated in Figure 6. The curves show a typical course for the elongation behaviour under sinusoidal stress with a defined force. The strain curves of PLLA (black) and PLLA+Fe (red) decrease with increasing load cycles due to relaxation effects, especially during the first 20,000 cycles. After 20,000 load cycles, the measured strain decreases slightly until 100,000 load cycles, for both PLLA and PLLA+Fe. From 100,000 load cycles onwards, the strain values remain almost constant. During the initial 100,000 cycles, particularly the first 20,000 cycles, the microstructure of the material undergoes restructuring, making the material stiffer. The curve progressions of PLLA and PLLA+Fe samples are very similar, except for the strain-offset that can be observed in composite samples. The incorporation of iron microparticles, therefore, leads to a lower initial displacement with a load of 8 N. The curve below 20,000 cycles shows a slightly reduced decrease for the composite samples compared to the reference PLLA.

3.6. Cell Biological Analysis

PLLA reference samples as well as PLLA polymers with the addition of 5 wt% Fe were assessed regarding their biocompatibility within direct contact and eluat tests according to DIN EN ISO 10993-5. Furthermore, the influence of Fe particles alone on cells was examined.

Both human fibroblasts and endothelial cells show an overall metabolic cell activity similar to the negative control set to 100% in direct contact with the PLLA reference and the PLLA+Fe-composites (Figure 7). A limit of 70% is stated as biocompatible. Nevertheless, especially for the endothelial cells, high data variability is recognisable. Exemplary fluorescence images show the results of metabolic cell activity, with decreased cell density and partial loss of the physiological spread shape of endothelial cells on all PLLA samples (Figure 8).

Eluate testing reveals biocompatibility of all eluate concentrations (100, 31.6, 10%) of the PLLA reference and the composite (Figure 9). Again, endothelial cells show high data width, probably caused by the thawing and cultivation of three different cryotubes throughout the tests, necessary due to laboratory internal processes. Exemplarily, fluorescence images of the undiluted eluates show dense cell layers similar to the negative control (Figure 10).

Due to the usage of a degradable polymer, Fe particles were evaluated separately for biocompatibility (Figure 11). Metabolic cell activity of human fibroblasts is significantly reduced for 2.5 mg/mL and 1.5 mg/mL Fe (*** p ≤ 0.001) after direct and delayed addition of particles, respectively, and even for 0.5 mg/mL Fe after delayed particle addition (* p < 0.05) compared to the negative control. Cell activity is significantly reduced in a concentration-dependent manner in both particle evaluations. In comparison, delayed addition of particles after 24 h of cell pre-cultivation results in less decreased cell activity compared to the direct addition.

Metabolic cell activity of human endothelial cells (Figure 12) is significantly reduced for 2.5 mg/mL and 1.5 mg/mL Fe after delayed addition of particles (*** p ≤ 0.001) compared to the negative control and in a concentration-dependent manner when comparing the single concentrations to each other (## p < 0.01, ### p ≤ 0.001). Direct addition of particles shows similar tendencies, but without significant differences.

Light microscopic and SEM images of human fibroblasts and endothelial cells incubated with three concentrations of iron particles with direct and delayed addition show a homogeneous distribution of particles in culture and increasing particle amounts (Figure 13 and Figure 14). In particular, cells with direct particle addition exhibit decreased cell density with higher particle concentrations. SEM images display cell–particle contact, but no engulfment of particles, which might be due to short cultivation time and the size of particles.

Multiplex analysis of inflammation markers after 48 h incubation of cells with particles revealed negligible amounts of GM-CSF, IFN-α, IL-4, as well as IL-17A and TNF-α. A slight but dose-independent increase in IL-2 and IL-6 was detected in HT-1080 fibroblasts and EA.hy926 endothelial cells, respectively, for both direct and delayed particle addition (Figure 15 and Figure 16), except for EA.hy926 with delay, where IL-6 was below the detectable limit (Figure 16 right).

4. Discussion

4.1. Thermal Analyses

TGA measurements showed a lower start temperature (321 °C) of PLLA+Fe composite than of PLLA films without Fe particles (280–290 °C), which is in accordance with the literature and may be because of the better heat transfer through the Fe particles [58]. Therefore, the melting temperature of the composites was also reduced by about 50 °C compared to pure PLLA. For PLLA+Fe films, TGA measurements showed a mass loss at a temperature of 105–110 °C. Probably, this was due to the loss of residual chloroform in the film. This finding is supported by the XPS spectra for chlorine and iron, which showed no signals for oxidised iron species, such as Fe(II) or Fe(III). DSC measurements also showed decreased glass transition temperatures for the PLLA+Fe films, which may also be caused by the plasticizing effect of chloroform. Since our data showed residual solvents in the PLLA+Fe film, the need for sufficient cleaning procedures after manufacturing needs to be pointed out. For polymer processing, only chlorinated solvents, such as dichloromethane and chloroform, were able to dissolve PLLA. Investigation of the ability of the applied washing procedure for the PLLA film to extract all residual chloroform from PLLA+Fe without altering the iron particle surface was not focussed by this study and needs to be elucidated in the future. XPS measurements further showed no iron particles at the surface, neither the top side nor the bottom side, indicating a complete coverage by the polymer. This finding was also supported by the SEM measurements. However, the observed particle aggregation during the curing time may be undesirable for certain applications. Therefore, it may be effectively mitigated by the use of sonification. Especially for biomedical applications, this allows the usage of established surface modification methods, such as O₂ plasma and UV processes, to enhance cell growth or to immobilise active pharmaceutical substances onto the polymer surface. Moreover, building up further polymeric layers remains permitted, paving the way for more complex drug delivery systems, such as biodegradable drug-eluting stents, or actuating applications, such as artificial muscles or electrically conducting purposes.

DSC measurements showed that the peaks of the composites started at temperatures up to 8 K lower and were lower than the peaks of the pure polymer. The iron particles could have increased the thermal conductivity of the material, which explains the peak shift to lower temperatures. Additionally, residual chloroform acts as a plasticiser, also leading to a decrease in the glass transition temperature [59]. Further, iron particles are not involved in polymer reactions, which caused weaker reaction peaks in the mass-normalised curves. Peaks “a” and “d” occurred at up to 10 K lower temperatures during the second heating compared to the first heating, which can be attributed to the improved contact of the samples to the crucible after melting. Peaks “b” and “c” only occurred during the second heating. The DSC curves of the two cooling steps were approximately identical. There were no significant peaks in these curves, such as solidification peaks, so presumably amorphous solidification took place during cooling at 10 K/min. Overall, only a minor influence of the iron particles on the material could be observed compared to the pure polymer. For PLLA, a similar sequence of peaks is described in the literature: the glass transition occurred at 55 °C, the PLLA exothermal crystallisation at 100 °C, and the melting peak at 173 °C [49]. Furthermore, a transformation of the α’ crystal to α immediately before the melting peak was determined [56]. These four reactions could be observed in Figure 3 at similar temperatures for the curve of the second heating. Overall, similar DSC curves of composites could be determined in comparison with DSC curves of pure polymers from the literature. The addition of the metallic particles did not seem to have a substantial influence on the transformation behaviour of the investigated materials, except for minor changes due to overall deviating thermal properties.

Nevertheless, changes in thermal behaviour and crystallinity might occur in a dose-dependent manner when iron particles may act as heterogeneous nucleation sites within the PLLA matrix, which can influence crystallisation behaviour and potentially enable tuning of the degradation kinetics and structural stability of the composite for biomedical applications. Such effects have also been reported for particle-filled polymer systems, where dispersed particles influence crystallisation processes and thereby modify the degradation and mechanical performance of the material. Therefore, the incorporation of iron particles not only introduces a biodegradable metal phase but may also provide additional opportunities to tailor the performance of the composite for resorbable biomedical devices [60,61,62].

4.2. Mechanical Analyses

Tensile tests showed a decrease in the Young’s modulus and a reduction in tensile strength and elongation at break of the PLLA+Fe composite compared to pure PLLA. The same tendency regarding tensile strength was observed by Islam et al., with similar values compared to our results [58]. The incorporation of iron showed a significantly lower influence on the much higher values of the Young’s modulus, in contrast to our results. This is most likely due to different manufacturing processes (3D printing vs. solvent-based process) and possibly the use of PLLA with a different molecular weight or of different iron particle size and content (not specified). Oksiuta et al. analysed PLA-based composites with 5% iron, magnesium, as well as polyethylene [63]. Iron powder particles had an average size of 45 µm. Samples were prepared by melt extrusion with subsequent hydraulic compression at 10 MPa. Tensile tests were performed with a crosshead speed of 10 mm/min at room temperature. Iron filling led to minor changes in the tensile parameters with a slight increase in Young‘s modulus, tensile strength, and elongation at break. Tensile strengths were basically in the same range as our determined values, Young‘s moduli were significantly higher, and the elongations at break were significantly lower than in our experiments, which might be due to different sample preparation (extrusion and compression vs. solvent-based processing) and testing parameters (e.g., testing at room temperature vs. 37 °C).

Results of the DMA showed a basically equal curve progression and consistent mechanical integrity of PLLA and PLLA+Fe samples, with a slightly lower initial displacement for samples with incorporated iron microparticles compared to pure PLLA. The nearly parallel curve progression of PLLA and PLLA+Fe suggests that the fundamental viscoelastic relaxation mechanisms of PLLA are not substantially altered by the particles. The lower initial elongation of the PLLA+Fe samples indicates increased effective stiffness of the composite material, as rigid metal particles constrain deformation of the matrix and carry part of the load. The marked decrease in elongation in the first 20,000 cycles is typical for the short relaxation times of amorphous PLLA chains in the glass state: rapid segmental movements lead to a rearrangement of the chains and thus to a significant reduction in stress. The slightly reduced decrease for the composite samples compared to the reference PLLA can be attributed to improved load support and reduced free chain mobility near the particles. Between 20,000 and 100,000 cycles, slower relaxation processes (longer relaxation times, physical ageing) dominate, causing the elongation to decrease only slightly further. From about 100,000 cycles onward, a quasi-stationary equilibrium is reached in which the viscoelastic stress relaxation and any micro-reorganisation hardly change the macroscopic strain. The selected force of 8 N is within the range for determining the modulus of elasticity and is well below the yield strength of the tested materials (14.9 ± 2.1 N for PLLA and 19.6 ± 1.0 N for PLLA+Fe). Mechanical properties, determined by tensile tests and DMA, revealed that the addition of iron particles leads to a slight reduction in the stiffness and increase in the brittleness of the PLLA material and slightly reduced the viscoelastic properties, while, according to references [58,63], the magnitude of the mechanical properties and, thus, the application possibilities were hardly affected. It has to be considered that solvent residues influence the mechanical behaviour of the materials, which should be investigated in more depth in further studies. Ensuring uniform dispersion of iron particles within the polymer matrix and maintaining good interfacial adhesion between particles and polymer are important considerations for improving composite performance and reducing fluctuations of replicate experiments. In the present work, no specific surface functionalisation of the iron particles was applied. The particles are primarily mechanically embedded within the PLLA matrix. Nevertheless, the obtained mechanical results indicate sufficient interfacial interaction for effective load transfer within the investigated composition range. By forming cylinders via joining of solvent-cast films by application of pressure and temperature [19], the nonuniform distribution of particles in the PLLA+Fe film probably turns out to be averaged. For pressure-loaded applications such as biodegradable stents, this method may also be applied to the films generated in this study.

4.3. Cell Biological Analysis

Although an improved cytocompatibility was stated after the addition of iron particles to PLLA [23], this could not be confirmed within this study, since no significant differences were found between the PLLA reference and the iron-containing material on both sides. Both cell types reacted in a similar manner to the test materials within direct contact and eluate testing, even without differences from the negative control cultured on tissue culture polystyrene.

Since the degradation time of PLLA is mentioned as 30 weeks [13] and the cultivation time was only 48 h, the materials used should be “as manufactured”. Fibroblasts seeded on PLA films showed adequate cell adhesion and proliferation with elongated appearance and numerous cytoplasmic extensions and cell–cell interactions [64]. PLLA films resulted in high proliferation and colonisation of mouse and human fibroblasts after 72 h in eluate tests [65]. PLLA films without additional surface treatment caused a comparable low adhesion and proliferation of endothelial cells over five days [66]. In contrast, human endothelial cells on dip-coated PLLA films showed a relative cell proliferation of ~80% compared to the Thermanox control after six days [67]. Due to our chosen 48 h cultivation, material degradation or particle leakage is unlikely to influence cells. However, due to the biodegradability of PLLA, iron particles will be released at some point. Therefore, iron particles were tested, specifically showing significant dose-dependent influences on fibroblasts and endothelial cells when added directly or delayed, respectively. Most studies in the literature deal with iron oxide nanoparticles, which made comparison of the results difficult. Accumulation of iron might play a central role in senescence and fibrosis [68]. The uptake of iron particles by fibroblasts occurs rapidly, likely through endocytosis. Studies utilizing iron oxide nanoparticles have demonstrated that human dermal fibroblasts can internalise these particles efficiently, showcasing the dynamic interaction between cells and iron-based materials [69]. A significant increase in the permeability of human endothelial cells was caused by exposure to iron oxide nanoparticles [70]. Iron chloride exposure resulted in a dose-dependent reduction in endothelial barrier integrity due to cell morphological changes and increased production of reactive oxygen species [71]. Further, iron overload has been linked to apoptosis and ferroptosis in endothelial cells with elevated iron levels triggering pathways that lead to cell death [72]. Iron particles were without significant effects on permeability, cytotoxicity, and inflammation response in human endothelial cells up to 100 µg/mL [73]. Cells tend to maintain the typical fibroblast-like morphology after contact with iron nano- and micro-particles [74]. The cytotoxicity of iron nanoparticles was influenced by various factors, including size, shape, and surface coating [75]. Iron-containing materials intended for biomedical applications may, in principle, promote iron-dependent cell death pathways such as ferroptosis [76]. In the present study, pure iron particles exhibited a concentration-dependent cytotoxicity in vitro. In contrast, no significant differences in cell viability were observed between neat PLLA films and iron particle-loaded PLLA composites. This indicates that embedding the particles within the PLLA matrix effectively mitigates the cytotoxic effects observed for the free particles, likely by limiting direct cellular exposure and iron ion availability. Given the slow hydrolytic degradation of PLLA, iron release is expected to occur gradually rather than acutely, reducing the likelihood of ferroptosis induction under the tested conditions. Nevertheless, long-term in vivo studies and mechanistic investigations addressing iron release kinetics and ferroptosis-related pathways will be required to fully assess potential iron-mediated effects beyond the scope of the present work.

In our study, the synthesis of inflammation markers by fibroblasts and endothelial cells occurred after particle contact. Inflammatory responses within endothelial cells were proven to be activated after exposure to iron [71]. Fagali et al. found that cell reaction after exposure to degradation products of Fe-based biomaterials is more dependent on the presence of insoluble products than on soluble Fe species [77]. Overall, the influence of iron on cells strongly depends on the used cell line/type, dose, and exposure time, but further on the form of the element (e.g., nano- and micro-particle, salt, ions) [78].

5. Conclusions

Resorbable polymer–metal composites of PLLA matrices together with iron powder particles have been produced via solvent casting. Both PLLA and iron have been selected due to their general degradability in the human body. Metal particles amounted to 5% (w/w), and particle sizes ranged in the 10 µm scale. Solvent casting resulted in films ø100 mm × 0.1 mm. In the solvent-cast films, the heavier metal particles accumulated at the bottom side of the film. Thermal analysis by TGA and DSC revealed no significant differences in the glass transition and melting behaviour of PLLA in the composites compared to the pure polymer. The only exception occurred in solvent-cast PLLA+Fe films, where a potential loss of the solvent chloroform can still be detected by TGA. With XPS, it was also shown that residual chloroform was present. Iron chloride could not be found, which narrows the findings in TGA towards the loss of chloroform from the PLLA+Fe films. Mechanical testing showed a decrease in the stiffness and an increase in brittleness of the PLLA material due to the addition of iron particles, which has to be considered with regard to the particular application.

PLLA and PLLA+Fe composites showed an overall biocompatibility with human fibroblasts and endothelial cells when examined in direct contact and eluate testing according to DIN EN ISO 10993-5. Yet, evaluation of iron particle cytotoxicity revealed a time- and dose-dependent decrease in cell viability and slightly increased levels of IL-2 and IL-6, which might have an influence as soon as the PLLA is degraded.

The applied characterisation methods from the various fields, including structure, mechanics, surface chemistry, thermal properties, and biocompatibility, prove the feasibility of novel PLLA+Fe composites manufactured by solvent casting for biomedical applications. The solvent casting process can be improved regarding the homogeneity of the iron microparticle distribution. Further, cleaning processes need to be established to ensure the final absence of chloroform. This paves the way for future use of polymer metal composites in biomedical applications, such as biodegradable stent structures. Moreover, these data give insights into the alteration of thermal, optical, and electrical properties, which may be used in soft robotics and 3D-printing of implants.