Introduction

Polymer nanocomposites have gained significant attention from both the scientific community and industry for many years [1, 2]. These materials consist of multiple phases, with one or more dispersed phases at the nanoscale within a polymer matrix [3, 4]. The incorporation of nanoadditives into polymer matrices can cause the enhancement of various properties such as thermal conductivity, mechanical strength, barrier properties, electrical conductivity, and biological characteristics. The enhancement of these properties is affected by several factors, such as the size of particles, surface development, and distribution of nanoparticles within the polymer matrix [4]. Various materials can be used as nanofillers, both in terms of chemical nature (inorganic, organic), physical structure (crystalline, amorphous, gas inclusions—nanofoams), and particle shape (3D-type—powder-like; 2D-type – lamellar-like; 1D-type, linear (rod, fibrous, tubular)). Nanoadditives categorized by shape include, among the others, i) carbon nanotubes (CNTs), carbon nanofibers (CNFs), halloysite nanotubes (HNTs) as 1D-type; ii) graphene derivatives forms (GDFs) like graphene, graphene nanoplatelets (GNPs), layered silicates as 2D-type; iii) spherical silica, carbon black, semiconductor nanoclusters, quantum dots as 3D-type [5,6,7,8,9,10,11,12]. One of the most commonly applied 1D-type nanoadditives is CNTs [13]. Polymer nanocomposites containing CNTs exhibit excellent functional properties (e.g., mechanical and thermal properties) [14]. Nonetheless, another type of 1D-type nanofillers still gaining scientific interest are CNFs that exhibit excellent mechanical, thermal, and electrical properties, with reported tensile strengths ranging from 1.5 to 7 GPa and Young’s modulus ranging from 228 to 724 GPa, depending on processing methods and fiber diameter [15]. Another example of 1D-type nanoadditives, but of a different chemical nature (mineral nanofiller) are HNTs, which are an attractive natural alternative to carbon nanotubes (CNTs) because their production is inexpensive and are available in several regions [16, 17]. The Young modulus value of HNTs is in the range of 230–340 GPa [18]. HNTs, while used as fillers, enhance the strength of polymer composites while preserving their ductility [19]. It is a notable advantage over CNTs and nanoclays, which often compromise flexibility for increased strength [19]. In addition to their excellent mechanical properties, HNTs exhibit good biocompatibility and low toxicity toward worms and microorganisms [20].

Among lamellar fillers (2D-type), GDFs and nanocomposites containing GDFs have been enjoying unwavering interest for over 20 years [21]. GNPs, characterized by their two-dimensional carbon-based structure, can enhance barrier properties, thermal conductivity, and mechanical strength when added to polymer matrices [8]. Reported values for Young’s modulus and tensile strength of GNPs are 1,100 GPa and 125 GPa, respectively [22]. One of the most common 2D-type mineral nanofiller used is montmorillonite (MMT). The structure unit of MMT is built of two tetrahedral silica layers separated by an octahedral alumina layer [23]. MMTs are widely used thanks to their easy modification with organic cations on the interlayer gallery [24]. MMTs’ attributes, including their widespread availability, cost-effectiveness, and natural abundance, resulted in their broad application for industrial use worldwide [25]. Currently, commercially produced MMTs are accessible at an affordable price. For instance, CLOISITE® 20A is an example of an industrially employed MMT that enhances thermoplastic compounds’ physical and barrier properties [26].

Incorporated nanofillers do not solely determine the performance of polymer nanocomposites (PNCs); the properties of the polymer matrix also influence them. Thermoplastic elastomers (TPEs) are a group of polymers characterized by the mechanical properties of elastomers and the processability of thermoplastics [26]. These properties can be achieved due to their structure, which can be categorized into rigid and flexible segments. In contrast to traditional elastomers, which rely on chemical cross-linking, TPEs rely on physical cross-linking in their rigid segments. Among the many TPEs subgroups poly(ester-ether)s (PEE)s stand out, featuring polyesters as the rigid component and polyethers as the flexible one. The most recognizable PEE is poly(butylene terephthalate)-poly(tetramethylene oxide) (PBT-PTMO), known under the trade name Hytrel® (DuPont) [27]. The increasing environmental risks and a decline in fossil fuels resulted in the demand for eco-friendly counterparts of traditional fossil-based polymers. In 2004, the US Department of Energy recognized furandicarboxylic acid (FDCA) as one of the top twelve critical renewable materials and the only aromatic monomer on the list [28]. The most popular route of the FDCA is through the oxidation of 5-hydroxymethylfurfural (HMF) and its derivatives [29, 30]. FDCA has a chemical structure similar to terephthalic acid (TPA), making it a viable monomer for synthesizing polyurethanes, polyamides, and polyesters. Furthermore, FDCA-based polyesters exhibit superior barrier properties and tensile modulus compared to their TPA-based counterparts. For instance, poly(ethylene 2,5-furanodicarboxylate) (PEF) exhibits 11 and 19 times superior barrier properties to O2 and CO2, respectively, in comparison with poly(ethylene terephthalate) PET [31,32,33].

The aim of this work is to compare how the incorporation of four types of nanoadditives that differ in shape (1D-type and 2D-type) and the chemical character (carbon and mineral) affect the synthesis process and the selected properties of the biobased polymer matrix. Therefore, one introduces two 1D-type nanofillers (CNFs and HNTs) and two 2D-type nanofillers (GNPs and C20A) into poly(propylene 2,5-furandicarboxylate)-block-poly(tetrahydrofuran) (PPF-b-F-PTMO) with the content of soft phase (F-PTMO) equals to 50 wt%. The dispersion of nanoadditives was studied using scanning electron microscopy (SEM). Limited viscosity number (LVN) and differential scanning calorimetry (DSC) investigated the morphology and phase structure. To verify processing temperature limitations and application potential resulting from the mechanical performance of nanofillers-induced changes in PPF-b-F-PTMO, the thermogravimetric analysis (TGA) and the tensile properties (static and cyclic) have been analyzed.

Materials and methods

Preparation of bionanocomposites

Four types of bionanocomposites containing carbon and mineral nanofillers were synthesized employing in situ polymerization, the scheme of which is presented in Fig. 1. Before the synthesis, the nanoadditives were dispersed in propylene glycol (bio-PDO) through a high-speed stirrer and ultrasonic homogenizer for 30 min.

Figure 1
figure 1

Schematic illustration of the preparation process of obtaining bionanocomposites.

Full size image

The following nanoadditives were used to prepare bionanocomposites:

  • 1D-type:

    1. o

      carbon nanofibers (CNFs, purity > 98%, Sigma-Aldrich, Burlington, MA, USA), diameter: 100 nm, length: 20–200 μm;

    2. o

      halloysite nanotubes (HNTs) (trademark “Dunino,” Intermark, Gliwice, Poland), a form of a powder containing rod-shaped HNTs of diameter 100–140 nm and length 1–5 μm, bulk density 450–600 g/dm3;

  • 2D-type:

    1. o

      graphene nanoplatelets (GNPs) (Angstron Materials, Dayton, OH, USA), according to providers’ data: thickness < 1 nm, C: ~ 97%, and O: < 2.1%;

    2. o

      organically modified Cloisite® 20A (C20A, Southern Clay Products, Austin, TX, USA) used as received, in the form of white powder and size of particles of D50: < 10 μm, with a bulk density of 350 kg/m3, the distance between the lamellas (XRD, d001): 2.7 nm.

One can categorize the polymerization procedure into two stages: transesterification and polycondensation. Transesterification takes place between dimethyl 2,5-furandicarboxylate (DMFDC, 99,9%, Henan Coreychem Co., Ltd., Zhengzhou, China) and the bio-PDO (99,5%, DuPont Tate & Lyle BioProducts, Loudon, USA), with which nanoparticles were mixed at the stage preceding the synthesis process. This occurs within the temperature range of 160–185 °C and atmospheric pressure. The initial phase is facilitated by a catalyst (Ti(OBu)4) (TBT, 97%, Merck, Darmstadt, Germany) for a duration of up to two hours. Methanol, the primary by-product, is distilled and collected during transesterification, with approximately 90% of the calculated methanol being distilled to indicate the completion of the first step. Prior to polycondensation, poly(tetramethylene oxide) (PTMO, 99%, BASF, Ludwigshafen, Germany) with a molecular mass of 1000 g/mol, a thermal stabilizer (Irganox 1010, Ciba Geigy, Basel, Switzerland), and the second catalyst dose (also TBT) are added. In the subsequent step, the temperature is raised to 240 °C, and a vacuum is applied to remove excess bio-PDO, reducing the final pressure to 25 Pa for up to 1.5 h. Progression in the second stage is monitored by observing changes in stirrer torque to estimate melt viscosity. The process concludes once the desired melt viscosity is attained, indicating the achievement of the desired molecular mass.

Subsequently, the material is extruded from the reactor into a water bath using compressed nitrogen. The yield of the polycondensation process is ca. 70%. All synthesized copolymers are denoted as PPF-b-F-PTMO since the diester unit is included with the PTMO sequence in the flexible segment. Since PPF-b-F-PTMO 50/50 wt% is referred to as PEE in this study, all nanocomposites in the obtained series are labeled as PEE/x% NP, where x represents the quantity of nanofiller, NP signifies the type of nanoparticle. For example, PPF-b-F-PTMO 50/50 containing 1 wt% CNF is labeled PEE/1% CNF. The remaining materials are designated similarly.

Preparation of the samples

The obtained samples were first granulated and then subjected to drying at 55 °C for 24 h under vacuum conditions prior to injection molding. Injection molding was carried out using a Boy 15 machine (Dr. Boy GmbH & Co., Neustadt, Germany) to form dumbbell-shaped samples of type A3 as per ISO 527 standards. The following conditions were applied during injection: injection temperature set at 180 °C, mold temperature maintained at 30 °C, injection pressure at 70 MPa, and holding pressure at 12 MPa.

Characterization methods

The nanocomposite structures were examined using field emission scanning electron microscopy (FE-SEM Hitachi SU-70, Tokyo, Japan). Prior to testing, the samples were cryofractured in liquid nitrogen and subsequently coated with a thin film of gold–palladium under vacuum conditions (JEOL JEE-4X, Japan). Secondary electron (SE) images were captured at an acceleration voltage of 5kV.

The samples’ phase structure was analyzed by differential scanning calorimeter (DSC) using the DSC 204 F1 Phoenix (Netzsch, Selb, Germany). The experiment employed a heating/cooling/heating cycle within the temperature range of − 85 °C to 250 °C. Three cycles of heating–cooling–heating were applied in the following order: heating 10 °C/min–cooling 10 °C/min–heating 10 °C/min–cooling 3 °C/min–heating 3 °C/min. 5-min isotherms were maintained between cycles. The synthesized materials’ glass transition temperature (Tg) was determined as the midpoint of the heat capacity change (∆Cp). The ΔCp for the materials was calculated using NETZSCH Proteus® software, according to ASTM E 1269 based on three measurements (baseline, sapphire, sample) by using DSC apparatus. The sensitivity is derived from the calibration measurement using a sapphire as a specific heat standard. The melting temperature (Tm) and crystallization temperature (Tc) were identified at the peaks of endothermic and exothermic reactions during the second heating scan, respectively. The degree of crystallinity of the hard phase constituting 50 wt% in the block copolymer was calculated using the following equation:

$$X_{cR} = \left( {\frac{{\Delta H_{m} }}{{\Delta H_{m}^{0} }}} \right) \cdot \begin{array}{*{20}c} {0,5} & {100\% } \\ \end{array}$$
(1)

where: ΔHm represents the enthalpy of melting obtained from the melting peak area observed on DSC thermograms, while ΔHm0 denotes the enthalpy of melting for fully crystalline material, which is 141.7 J/g for PPF [33].

The thermal stability of the bionanocomposites in an oxidizing atmosphere was assessed using thermogravimetry (TGA 92–16.18 Setaram, Caluire, France) with simultaneous TG-DSC measurements. The experiments were conducted under oxidizing conditions, specifically dry synthetic air (with a N2:O2 ratio of 80:20 vol.%) at a flow rate of 20 mL/min. The tests were carried out at a heating rate of 10 °C /min within the temperature range of 20–700 °C.

The limited viscosity number (LVN) was determined at 30 °C in a mixture of phenol and 1.1.2.2-tetrachloroethane (with a weight ratio of 60/40). The concentration of the polymer solution was 0.5 g/dL. LVN was measured using a capillary Ubbelohde viscometer (type Ic, with a calibration constant of K = 0.03294).

Static mechanical properties were assessed using an Autograph AG–X plus tensile testing machine (Shimadzu, Kyoto, Japan), which conforms to class 1.0 standards according to EN 10002–2, ISO 7500–1, BS 1610, ASTM E4, and JIS B7721. The machine was equipped with a 1 kN Shimadzu load cell, an optical extensometer (class 0.5 according to ISO 9513), and TRAPEZIUM X computer software (version 1.4.5, Shimadzu, Kyoto, Japan). Initially, specimens were elongated to 1% at a crosshead speed of 1 mm/min, followed by an acceleration of 100 mm/min. Static mechanical measurements were conducted by PN-EN ISO 527, with reported values representing the average of five samples. Additionally, cyclic tensile measurements were performed at a crosshead speed of 100 mm/min. Dumbbell-shaped samples were stretched until the specified strain value (5%, 15%, 25%, 50%, 100%, and 200%) was reached, after which the tensile force was released to zero until the sample fractured. Reported values are the average of five samples.

The hardness of the synthesized bionanocomposites was evaluated using a Zwick 3100 Shore D tester (Zwick GmbH, Ulm, Germany). The hardness test was conducted following ISO 7619–1 and ASTM D2240 standards. Reported values represent the average of ten independent measurements.

Results and discussion

Dispersion of nanofillers

The dispersion and distribution characteristics of the nanoadditives are the key factors that affect the supramolecular structure (mainly crystallization behavior) and tensile properties of the nanocomposites [34]. Therefore, to obtain a proper dispersion in the in situ obtained bionanocomposites, the process has proceeded with an additional step using mechanical mixing and sonication. The SEM images taken at two selected magnifications (1k and 5k) of nanocomposites containing 1%CNFs (a,b), 1%HNTs (c,d), 1%GNPs (e,f), and 1%C20A (g,h) are presented in Fig. 2. In general, it was found that all types of nanoadditives were evenly distributed in the polymer matrix (visible at 1k magnification, i.e., Fig. 2a, c, e, g). Slight differences result from nanofiller-matrix interactions; thus, morphological differences are only observed at higher magnification (5k). CNFs were well distributed throughout the polymer matrix (Fig. 2a); however, some voids were observed between the nanofibers and the matrix, signifying that the nanofibers extended beyond the matrix during the brittle fracture preparation, and there was a lack of strong interactions between the nanofibers and the polymer. Since it is known that the key role in improving the properties of nanocomposites is played by the size of the interphase between the phases, i.e., nanofiller-polymer, insufficient improvement in properties resulting from the addition of the filler can be expected, among others, affecting the mechanical properties (described later in the article). Thus, we can assume that this effect may result from the smooth surface of the CNFs, which leads to a limited mechanical interlocking with the polymer matrix [35]. In turn, for bionanocomposites with the addition of the second type of 1D-type nanoadditive, i.e., HNTs, small agglomerates of nanotubes were also observed here and there. Nonetheless, these agglomerations seem evenly distributed in the polymer matrix. The nanocomposites with oriented GNPs also show a good dispersion. However, compared to the nanoadditives mentioned above, GNPs seem entirely embedded in the matrix due to the numerous defects, free radicals, and other irregularities on the surface of GNPs [36]. On the other hand, nanoclays (C20A), which were also well dispersed (Fig. 2g), seem to be sticking out from the matrix. Therefore, the effectiveness of the in situ polymerization method, preceded by an additional step of mixing nanoparticles with a liquid monomer, was confirmed; however, the impact of nanofillers on the thermal and mechanical properties will depend only on the type of nanofiller and the nature of the interactions at the nanofiller-polymer interface.

Figure 2
figure 2

SEM images of: PEE/1% CNF (a, b), PEE/1% HNT (c, d), PEE/1% GNP (e, f), PEE/1% C20A (g, h).

Full size image

Morphology and structure

The impact of the incorporation of nanoadditives differing in shape and chemical nature on the phase transition temperatures in the block copolymer was analyzed based on DSC thermograms. The DSC thermograms recorded at heating/cooling rates of 10 °C/min and 3 °C/min of the 2nd heating and cooling are presented in Fig. 3. Additionally, the values of the characteristic phase transition temperatures and the accompanying values of change in heat capacity and enthalpies (ΔHcc, ΔHm, ΔHc) and the determined degree of crystallinity of the rigid segment are summarized in Table 1. PEE exhibits two distinct microphases: A hard microphase abundant in PPF and a soft microphase enriched in PTMO. Consequently, PEE exhibits two glass transitions: One associated with the glass transition of the rigid segment (TgR) and another related to the glass transition of the flexible segment (TgF). Only for two materials (with the addition of mineral nanoparticles), it was not possible to determine the TgR during heating at a heating/cooling rate of 3 °C/min since this transition was overlapped with the exothermic effect of rigid segments cold crystallization in these materials: Tcc onset for PEE/1% HNT equals to 37.5 °C and for PEE/1% C20A equals to 42.6 °C. The TgR values for all samples determined during the second heating, regardless of the heating rate, are comparable to the reference sample, of course, except for the previously mentioned materials with the addition of mineral nanoparticles at the heating rate of 3 °C/min. In turn, at a heating rate of 10 °C/min, only for the material with the addition of 1% of C20A, a slight decrease in the TgR value was observed compared to the reference sample. This is most likely due to the dimensionality of the “filler network structure,” meaning how the polymer/clay nanocomposite is constructed. Ajayan et al. [37] reported that polymer intercalated between clay lamellas usually has greater mobility. The enhanced mobility can be attributed to the ordering process between the layers, leading to the formation of regions with varying densities and providing more opportunities for increased mobility. As intercalation progresses, the mobility of chains between clay lamellae improves, resulting in heightened friction between phases. There is a greater propensity for frictional growth with the increasing concentration of clay [38]. In turn, at a lower heating rate (3 °C/min), of course, lower TgR values were recorded. In this case, no uniform trend was observed regarding the impact of the addition of carbon nanoadditives on the glass transition temperature values: The addition of CNFs caused a slight decrease. In contrast, the addition of GNP caused a slight increase, which results from the different impacts of these two nanofillers that differ in shape on molecular mobility. These temperature variations were not determined for nanocomposites by adding mineral nanoparticles (as mentioned above).

Figure 3
figure 3

Differential scanning calorimetry (DSC) thermograms were obtained for the synthesized materials during the second heating (represented by solid lines), and cooling stages. These measurements were conducted at two distinct heating/cooling rates: 10 °C/min (denoted as a and b) and 3 °C/min (indicated as c and d).

Full size image
Table 1 Thermal properties determined from DSC
Full size table

Moreover, the values of TgF of nanocomposites are slightly lower than those observed for neat PEE at a heating rate of 10 °C/min. The decrease in the values of TgF is related to the values of limited viscosity number (Table 3), which were lower for all nanocomposites if compared to the reference sample and which indicate, as it was, the values of molecular masses. A similar effect was previously observed for PPF-based nanocomposites [39]. Almost all synthesized bionanocomposites exhibited lower values of TgF than the reference material, which is caused by the increase in the nanofiller-matrix interactions that affect molecular mobility at low temperatures [40]. However, at a heating rate of 3°C/min, for PEE/1% GNP, one observed a slight increase in the TgF value, which is most probably due to the presence of a crystal structure that imposes constraints or hinders the diffusion of mobile polymer chains [39]. This phenomenon will be further discussed while analyzing crystallization behavior. Moreover, the values of ΔCp for all nanocomposites at the heating rates of 10 and 3 °C/min are lower than the reference sample, resulting from the nanofiller-polymer interactions that constrain the diffusion of amorphous chains. This phenomenon arises from increased crystallinity, which imposes limitations on the diffusion of mobile polymer chains, thereby reducing the ΔCp value.

Based on the thermograms recorded during cooling, only the sample containing 1% GNPs crystallizes from the melt at the cooling at rates of 10 and 3 °C/min. This indicates the nucleating behavior of nanoplatelets that could have enhanced the crystallization process. Generally, among the various nanoadditives that differ in structure and geometry, CNTs and graphene were the most efficient ones [41] to act as additional heterogeneous nucleants [40, 42, 43]. All synthesized materials demonstrated cold crystallization during second heating at both heating rates. For the series of nanocomposites containing mineral nanofillers (HNTs and C20A), one observed that the values of Tcc are comparable to the value for the reference sample at both heating rates. In turn, for the samples containing carbon nanofillers, first of all, the cold crystallization onset temperatures exhibited visibly lower values of about 12 and 15 °C. Besides, these two materials, containing CNFs and GNPs, exhibit a double exothermic peak during cold crystallization, probably caused by the saturation of nucleation efficiency, which agrees with the following studies [40, 44]. Based on the observations of melt and cold crystallization, it can be concluded that polymer chains fold prior to crystallization and nucleation, and the crystallization proceeds via rearrangements of polymer chains in the existing chain templates [45]. The change in the structural arrangement is possible only by adding energy during heating and the presence of PTMO, which is less stiff, as a result of which the copoly(ester-ethers) undergo cold crystallization.

The values of Tm for all materials do not differ significantly. The average value is approximately 156.5 °C (and the median is 157.2 °C), and the differences in values are within the range of 1.9 °C. As suspected, the highest Tm value was observed for the PEE/1% GNP sample when heated at a rate of 3 °C/min. Incorporating nanoadditives caused a slight increase in XcR (related to ΔHm). Nonetheless, only the incorporation of GNP affected the crystallization behavior (GNP acted as an additional crystallization nucleation agent) since only PEE/1% GNP crystallizes from the melt, while the addition of other nanofillers, tested with two heating/cooling rates, rather influenced the chain folding. The shift to a higher temperature value in the Tc for GNP-filled PEE is caused by a lower cooling rate for various thermoplastic polymers [46, 47], including thermoplastic biobased polyesters [32]. With a decreasing cooling rate, the measured crystallization temperature as a peak of exothermic effect increases for GNP-filled PEE. At lower cooling rates, the time the polymer chains reach the level of the nucleation energy barriers becomes longer; thus, Tc occurs at higher temperature values with a more substantial exothermic effect (ΔHc). At the same time, reducing the cooling rate allows for a longer time for macromolecule diffusion in crystallite lattice and affects less perfect crystallites [46]. It is well-known that nanoadditives can affect the crystallization behavior of polymer matrices [48]. In most studies, one can find that the nanoparticles can serve as heterogeneous nucleation sites for crystallization and act as effective nucleating agents. However, their impact on the crystallization rate remains contentious, as it has been observed that some nanoparticles decrease nucleation activity when they achieve good dispersion and compatibility/miscibility with the polymer matrix. Generally, the overall crystallization process is anticipated to be accelerated owing to the heterogeneous nucleation process. However, nanoadditives can also impede molecular chain mobility, thereby delaying the crystallization rate. Thus, the competition between the nucleation effect and polymer chain retardation dictates the entire crystallization process. Depending on the shape (aspect ratio), chemical nature, specific surface area, and functionalization of nanoparticles, as well as their dispersion in the matrix, both nucleation and crystallization rates can be influenced [48,49,50]. Therefore, this study will continue to gain a deeper understanding of the mechanisms underlying the influence of selected nanofillers on the crystallization behavior of the biobased PPF-b-F-PTMO copolymer.

Thermal stability in an oxidative atmosphere

Many prior studies indicated that nanofillers’ promoting and hindering effects on the thermal stability of polymers may be expected [51,52,53,54]. Therefore, a comparative analysis of the impact of carbon and inorganic nanofiller addition on PEE, grouped by different geometric features, is reasonable due to the different mechanisms of their interactions with the polymer during exposition to elevated temperatures. The influence of selected nanoadditives on thermal stability in an oxidative atmosphere was studied by thermogravimetric analysis (TGA). The TGA curves made in the air are presented in Fig. 4. The characteristic temperatures of the mass losses of 5%, 50%, and 90% (T5%, T50%, T90%) in the oxidative atmosphere, and the temperatures correspond to the maximum rate of mass losses (TDTG1 and TDTG2), the amount of residue as well as the degradation rates corresponding to the maximum rate of mass losses (RDTG1, RDTG2, and RDTG3) are summarized in Table 2. The onset of thermal degradation is typically defined as the temperature at which a 5% weight loss is observed. Herein, the incorporation of 1 wt% of carbon nanoparticles (CNFs, GNP) did not significantly improve thermal stability. The characteristic mass loss temperature values determined for composites with carbon-based fillers are comparable to the reference PEE sample. Generally, the thermal properties of carbon nanofillers/polymer composites have attracted much attention from polymer scientists because they exhibit very high thermal stability [55,56,57,58]. Additionally, it has been identified that the polymer matrix’s structure and the interaction between nanofillers and the matrix play crucial roles in determining the thermal degradation behavior of carbon nanofiller-filled polymer composites. For instance, in poly(trimethylene terephthalate)-block-poly(tetramethylene oxide) segmented copolymer (PTT-PTMO) [59] based nanocomposites containing CNTs and/or GNPs prepared via in situ polymerization, the noticeable improvement of thermal stability in the oxidative atmosphere was observed. Authors claimed that the thermal stability enhancement did not seem to depend upon a particular shape of the nanoadditives, and both nanofillers exhibit high electron affinities and, as such, seem to be capturing free radicals [60, 61].

Figure 4
figure 4

TG and DTG curves for the synthesized materials measured in an oxidizing atmosphere for 1D-type nanofillers (a) and for 2D-type nanofillers (b).

Full size image
Table 2 Properties determined from TGA analysis
Full size table

In turn, the addition of mineral nanoparticles resulted in a significant deterioration of the thermal stability of the nanocomposites, especially at lower temperatures (T5%, T10%). The C20A-modified series revealed a more significant negative impact on the stability of PEE. Generally, due to the large scope of conducted research for polyolefin–clay composites, adding mineral fillers is assumed to affect thermal stability improvement positively. For instance, the increase in thermo-oxidative stability was observed by Peila [62], who showed that the presence of nanoclays such as commercially available Cloisite had an apparent effect on the thermal stability and thermo-oxidative properties of the LDPE-based composites. Moreover, Polanský [63] indicated the phenomenon of improving heat dissipation from the polyethylene structure through the addition of nanofiller in the form of halloysite nanotubes, which results in improved thermal stability. Also, thermal stability is enhanced in polypropylene due to the addition of HNTs [64] even with only 1 wt% HNT addition. In that case, the T50% value shifted up to 31 °C toward higher temperatures.

On the contrary, in our previous study [65], where we analyzed the effect of the addition of C20A and HNTs on the properties of the cross-linked PE, also resulted in the deterioration of thermal stability at lower temperatures, despite the good dispersion of nanoparticles in the matrix. An improvement in thermal properties was noted only at higher temperatures (T50%). Taking into account the measurement conditions (oxidizing atmosphere), one could refer to the observations made by Upadhyaya et al. [66], who demonstrated a beneficial effect of Cloisite 20A as suppressing thermo-oxidative degradation of bismaleimide (BMI). Their proposed explanation was based on reducing oxygen diffusion after adding nanoclay in the composite. The difference between the unfavorable interaction between the specified polymers (polyolefins) and thermoset resin compared to the PEE-based composite described in this paper will result from the lack of susceptibility of PE/PP to hydrolytic degradation and the insufficient temperature accompanying the BMI processing. In poly(l-lactic acid)/organically modified vermiculite nanocomposites prepared by melt compounding, Jesús Fernández et al. [67] observed that in the nanocomposites containing 5 and 10 wt% of organoclays, T5% decreases around 7 °C for the nanocomposites containing 5 wt% of clay and around 25 °C for the nanocomposites with 10 wt% of clay, compared to unfilled PLLA. They attributed this phenomenon to reducing the polymer’s molecular weight after melt processing, which can encourage the thermal decomposition of the PLLA matrix. In the considered case, one also observed a slight decrease in LVN of all nanocomposites containing mineral nanofillers (Table 3); thus, such an explanation may be most reasonable herein. Despite the lower LVN values for carbon-based nanofillers, it can be assumed that their ability to scavenge free radicals caused a suppressive effect opposite to the changes in thermal stability caused by the reduced molecular weight.

Table 3 LVN values and mechanical properties of the synthesized materials
Full size table

A two-stage decomposition process is observed in the case of neat polymer matrix and composites containing carbon nanofillers (Fig. 4). However, it should be noted that the course of the TG and DTG curves for the material series containing mineral nanofillers (HNT and C20A) is characterized by the appearance of the additional (first) decomposition step at the lower temperature values. The thermal and thermo-oxidative degradation mechanism of copoly(ether-ester) was already widely discussed [44, 68]. The decomposition of copoly(ether-ester)s begins with the flexible segment (F-PTMO) degradation. Oxygen primarily affects the carbon atom located in the α position relative to the ether oxygen atom in ether [44]. The temperature corresponding to the maximum mass loss rate (the most profound peak) occurring in the temperature range from 350 to 450 °C is associated with the degradation of rigid segments. In the last stage, taking place in the range from 450 to 600 °C, decomposition of the non-volatile residue takes place [69]. A significantly greater amount of residue was observed for samples with the addition of carbon nanofillers, i.e., CNF and GNP (Table 2). In turn, in the case of nanocomposites with the addition of mineral nanoparticles, the value of the residue is actually identical to that of the reference sample. This is due to fact that carbon nanofillers themselves are more thermally stable [7, 13,14,15]; thus, most probably one would observe comparable residue values at higher temperatures. However, since we compare residue values at 700 °C, there are differences in the determined parameters. In the case of C20A-filled nanocomposites, the values of the first degradation step coincide with the previously reported degradation range of the organic fractions of C20A [70]. However, taking into account the low share of nanofiller (1 wt%) and the lack of significant changes in the molecular mass associated with LVN (Table 3), it should be concluded that substantial deterioration in the thermal stability of the nanocomposite containing C20A in the range of 270–310 °C will not be induced by the presence of the filler itself and changing the structure of the polymer, but by the PEE-degradation catalysis caused decomposition products of organic low-molecular-weight compounds used for filler organophilization. Another mechanism causing a decrease in thermal stability in an oxidizing atmosphere can be suspected for HNT-modified PEE (where the highest degradation rate corresponding to the maximum rate of mass loss RDTG3 was observed). This inorganic filler has not been modified with organic compounds. Despite the intensive pre-processing drying of the fillers, the bound water causes hydrolysis during mixing in the molten state. This negative effect of unmodified inorganic fillers on the change in molecular weight and deterioration of thermal stability is mainly observed for biobased polyesters. It should be emphasized, however, that all the materials considered showed high thermal stability, and its deterioration in the case of PEE/1% C20A does not adversely affect the application possibilities of this series.

Mechanical properties

The introduction of a small amount of nanofillers into the polymer matrix has the potential to influence the mechanical characteristics of the material directly. Additionally, it can indirectly influence the synthesis process and consequently affect the resulting structural properties. Therefore, the influence of 1D-type and 1D-type nanoadditives on limited viscosity number (LVN) and the mechanical properties of PEE were studied. The representative stress–strain curves are presented in Fig. 5, and the mechanical properties are summarized in Table 3. Moreover, the measured values of LVN can also be found in Table 3. PEE has the highest value of LVN (1.155dL/g). Incorporating various nanofillers caused a reduction of the LVN by 6% to 11%, indicating a relatively minor difference. Such difference is likely due to the nanoparticles’ effect on enhancing the melt’s bulk viscosity during the final synthesis stage. Furthermore, the LVN value is directly associated with the polymer’s molecular weight, according to the trend: the higher the value of LVN, the higher the molecular weight [71]. The increased molecular weight relates to superior properties such as yield strength, toughness, or melting temperature [72].

Figure 5
figure 5

Representative stress–strain curves for PEE and nanocomposites containing 1D-type nanofillers (a) and 2D-type nanofillers (b) registered at static tensile tests.

Full size image

Polymer nanocomposites demonstrate a superior tensile modulus (E) value compared to neat PEE, with the most noticeable improvements observed in those containing carbon-based nanofillers. These nanocomposites exhibit an E value around 90 MPa greater than neat PEE. The synthesized materials have a yield point; the addition of nanofillers resulted in a higher value of yield strength (σy) except for PEE/ 1% C20A. Nanocomposites with mineral-based nanofillers possess superior elongation at yield (εy) compared to neat PEE. On the other hand, nanocomposites with carbon-based nanofillers demonstrate a reduced εy. The elongation at break (ɛb) values for the nanocomposites also displayed a similar behavior. Nanocomposites that contain mineral-based nanofillers show a modest reduction or increase in ɛb value, 12% lower for the PEE/1%HNT nanocomposite, while for the PEE/1% C20A, it was about 49% higher compared to the PEE. However, incorporating carbon-based nanofillers significantly decreased εb, even up to 145.1% for PEE/1% GNP. Additionally, εb values for all tested materials surpass 200% when subjected to a 100 mm/min crosshead speed. Thus, these materials’ mechanical properties fall into the TPE category [73]. The tensile strength value at break (σb) was slightly higher for nanocomposites compared to unmodified PEE; the most significant change in σb is caused by adding HNT (2.4 MPa). None of the synthesized materials undergo strain hardening, which is probably caused by their insufficient Xc. Thus, the observed σy > σb correlation is justified and is supported by the literature [71,72,73]. The increase in the value of both E and σb results from a higher value of Xc and interactions between nanoparticles and polymer matrix. However, the difference between Xc among materials is relatively small, about a few percent (Table 1). Thus, good dispersion of nanoparticles and interactions between nanofillers and the polymer matrix play a crucial role in enhancing E, σy, and σb values [74]. The difference between the ɛb between nanocomposites filled with mineral-based nanofillers and carbon-based nanofillers results in good interfacial compatibility between the mineral-based nanofillers and polymer matrix [19, 75, 76]. Furthermore, the effective interfacial bonding in HNT is attributed to the hydroxyl and siloxane groups originating from the edges/defect walls and surface of nanotubes [77]. The discrepancy between the value of εb neat PLA and its nanocomposite with a maximum of 6 wt% HNT reaches up to 27%; it is worth mentioning that the difference between PLA and PLA/ 2 wt% HNT is only about 8% [75]. Moreover, the PA-6 with the 2% HNT exhibited almost twice as much of the value of εb when compared to the neat PA-6 [76]. The addition of 2 wt% MMT to Nylon-6 also resulted in a higher value of εb when compared to the neat polymer [78].

The influence of the incorporated nanofillers on the elastomeric properties of PEE was analyzed through cyclic tensile measurements. Figure 6a-b presents stress–strain curves observed during the cyclic tensile test, while Fig. 6c-d shows the permanent set values measured after each deformation cycle. A permanent set is defined as a non-recoverable elongation of the specimen’s original length [79]. The permanent set remains consistent for the synthesized materials under low strains (under ε = 50%). The permanent set values for nanocomposites with mineral-based nanofillers are lower than neat PEE and nanocomposites containing carbon-based nanofillers, supporting the previous analysis of the influence of mineral-based nanofillers on elastic properties. In addition, the results of cyclic tensile testing of PEE/1% CNF and PEE/1% GNP show that these materials reveal the highest stiffness, consistent with the static tensile test and the exhibited E values of the synthesized materials. The values of Shore D hardness are shown in Table 3. The incorporation of nanoparticles increased the hardness of the nanocomposites compared to PEE. The most noticeable change in hardness was observed for PEE/1% GNP, which is 7 ShD higher than PEE.

Figure 6
figure 6

Representative stress–strain curves for PEE and nanocomposites containing 1D-type nanofillers (a) and 2D-type nanofillers (b) registered at cyclic tensile tests.

Full size image

Conclusion

Two series of block copolymers based on poly(propylene 2,5-furandicarboxylate) and—poly(tetramethylene oxide) containing carbon and mineral nanofillers with different shapes (1D and 2D-type) were synthesized via in situ polymerization. The impact of incorporating 1D-type nanoparticles, namely carbon nanofibers (CNFs) and halloysite nanotubes (HNTs), as well as 2D-type nanoadditives, such as graphene nanoplatelets (GNPs) and organoclay (C20A), on the properties of a biobased block copolymer was examined. The dispersion of the nanoadditives in the nanocomposites was evaluated using scanning electron microscopy (SEM), and it was found that all nanofillers were uniformly distributed within the polymer matrix. The interfacial interactions were observed rather for carbon nanofillers, while in the case of mineral nanofillers, in a few places, one could have observed the agglomerates and the platelets sticking out of the matrix. Nonetheless, the efficiency of the in situ polymerization technique, preceded by an extra step of mixing nanoparticles with a liquid monomer, was validated.

Moreover, from the DSC analysis, it was possible to analyze differences in the polymer structure resulting from adding nanoparticles differing in aspect ratio and chemical nature. In general, the incorporation of nanoparticles led to increased crystallinity (Xc) but mainly affected the glass transition of rigid and flexible segments, chain folding, and, thus, cold and melt crystallization. In fact, only GNPs acted as an additional crystallization nucleation agent since only PEE/1% GNP crystallizes from the melt, while the addition of other nanofillers rather influenced the chain folding. The introduction of 1 wt% of carbon nanoparticles (CNFs, GNP) did not cause a significant improvement in thermo-oxidative stability. The values of characteristic temperatures of mass losses are comparable for these nanocomposites or slightly higher than for the reference sample. In turn, the addition of mineral nanoparticles resulted in a significant deterioration of the thermo-oxidative stability of the nanocomposite, especially at lower temperatures, resulting most probably from the decrease in LVN values. The incorporation of nanofillers resulted in a significant increase in the value of E. Nevertheless, all the synthesized nanocomposites maintained their elastomeric properties. The addition of the mineral-based nanofillers led to an enhancement of σy and σb without sacrificing the ductility of the neat PEE. The synthesized polymer bionanocomposites exhibited elastomeric mechanical properties during testing. Furthermore, the excellent processability of the produced materials via injection molding and the ability to tailor their mechanical and thermal properties by adjusting the type and content of nanofillers suggest potential applications in packaging, automotive, sports, construction, and various other industries.