Design and manufacture of degradable polymers: Biocomposites of micro-lamellar talc and poly(lactic acid)

Accepted Manuscript Design and manufacture of degradable polymers: Biocomposites of micro-lamellar talc and polylactic acid (PLA)

M. Barletta, E. Pizzi, M. Puopolo, S. Vesco PII:

S0254-0584(17)30320-6

DOI:

10.1016/j.matchemphys.2017.04.036

Reference:

MAC 19638

To appear in:

Materials Chemistry and Physics

Received Date:

27 December 2016

Revised Date:

18 April 2017

Accepted Date:

20 April 2017

Please cite this article as: M. Barletta, E. Pizzi, M. Puopolo, S. Vesco, Design and manufacture of degradable polymers: Biocomposites of micro-lamellar talc and polylactic acid (PLA), Materials Chemistry and Physics (2017), doi: 10.1016/j.matchemphys.2017.04.036

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Highlights 

High performant bio-composite of Poly Lactic Acid (PLA) reinforced with micro-lamellar talc;



Talc modified by surface reaction with organic and hybrid organic-inorganic compatibilizers;



Custom-built formulations by the dispersion of the pre-treated micro-lamellar talc in PLA;



Differential Scanning Calorimetry (DSC) and Attenuated Total Reflection Fourier Transform - Infrared Spectrophotometry (ATR FT-IR);



Chemical and physical interaction among the functional groups of pre-treated talc and PLA chains.

ACCEPTED MANUSCRIPT Design and manufacture of degradable polymers: Biocomposites of micro-lamellar talc and polylactic acid (PLA) M. Barletta*1,2, E. Pizzi1, M. Puopolo1, S. Vesco1 1Dipartimento

di Ingegneria dell’Impresa, Università degli Studi di Roma Tor Vergata, Via del Politecnico 1,

00133 Rome (Italy) 2Dipartimento

di Ingegneria, Università degli Studi di Roma Tre, Via Vito Volterra 62, 00146 Rome (Italy)

*Corresponding

author e-mail: [email protected] Tel.: +39(0)672697168 Mob.: +393289816259

Abstract. Over the last decade, biopolymers have seen a sharp increase in their market share as an alternative to conventional oil-based plastic materials. Among them, degradable biopolymers are rarely employed, especially in the field of durable end-goods, because of their often limited chemical, mechanical and thermal behaviours. In contrast, stringent international regulations encourage the employment of lowimpact materials in several manufacturing domains, with degradable polymers considered par excellence. In the present work, an innovative route to control the final properties of an extrusion-compounded, highperformance biocomposite composed of polylactic acid (PLA) reinforced with micro-lamellar talc is proposed. For this purpose, talc was first modified via surface reaction with organic (i.e., belonging to the class of polyisocyanates) or hybrid organic-inorganic (i.e., belonging to the class of organosilanes) compatibilizers, respectively. Custom-built formulations were then achieved by the dispersion of the pretreated micro-lamellar talc in commercial-grade PLA. The resulting compounds were analysed by differential scanning calorimetry (DSC) and attenuated total reflection Fourier transform infrared (ATR FTIR) spectroscopy. The experimental findings show that PLAs with different properties can be achieved as a result of chemical and physical interactions among the functional groups on the surface of the pretreated talc and the terminal groups of the PLA chains. These results are extremely promising for the achievement of innovative grades of PLAs suitable for melt processing and to ensure the greatly improved chemical inertness, thermal stability and mechanical strength of the resulting end-goods. Key words: Polylactic Acid (PLA); Degradable Biopolymers; Biocomposite; Extrusion Compounding; Microlamellar Talc Compatibilization.

ACCEPTED MANUSCRIPT 1. INTRODUCTION Thermoplastic polyesters are widely used in a number of manufacturing segments, including the development of packaging for food and pharmaceuticals. Polyethylene terephthalate (PET) dominates the large market of bottles and containers. Alternative materials in the packaging domain are very limited in comparison due to the excellent properties of PET, such as its light weight, mechanical and impact resistance, good barrier properties to gases and liquids, safety, recyclability and inexpensiveness. In contrast, PET materials have severe environmental impact because of their chemical-physical inertness and persistence in the environment. However, stringent environmental regulations are shifting the focus to biodegradable and compostable bioplastics, which are derived from renewable resources and are already available on the market. Research efforts are currently concentrated on making bioplastics suitable for industrial applications. Among bioplastics, polylactic acid (PLA), which is a compostable, biodegradable, bio-derived polyester, can satisfy requirements for the formulation of eco-sustainable materials, especially for packaging applications [1], because PLA boasts superior mechanical properties and appreciable thermal stability. With respect to PET, PLA is still rather brittle and suffers from poor thermal resistance. Moreover, PLA is generally very expensive. For these reasons, the application of PLA is mostly limited to specialty fields, such as for the development of biomedical solutions, specifically for applications in resorbable devices, surgical implants and materials. Currently, PLA can be melt processed by the same routes used for conventional thermoplastic polyesters, but reprocessing can pose serious risk [2]. In fact, PLA can easily undergo rapid degradation during melt processing at high temperature. Reprocessing can, therefore, compromise the performance of PLA. The effect of degradation by reprocessing on PLA can be mitigated by co-formulating commercially available polymers with specialty additives and fillers [3]. High cost remains the biggest issue to face. Therefore, research efforts are focused on the development of innovative routes to formulate and process PLA in order to achieve a resulting material with good physical properties and mechanical response at a reasonable cost. In addition, co-formulation must not affect the biodegradability and/or compostability of PLA. The most common approach to improve the properties of PLA is by dispersing mineral fillers inside the polymer matrix. Various fillers, including clay [4], talc [5; 6], inorganic whiskers [7] and even carbon fibres [8], have been investigated. Fillers can form a tough and hard inorganic phase inside the soft organic PLA phase, thus enhancing the mechanical response and thermal stability of the resulting composite material. Achievement of the appropriate dispersion and distribution of the fillers in the polymeric matrix can be troublesome, especially when nanosized or small particles are involved. Clustering of particles can compromise the performance of the resulting composite. Similarly, the mechanical and physical performance of PLA is strictly dependent on the compatibility between the polymer matrix and fillers. The formation of an effective interface between the soft polymeric phase and the hard mineral filler in the PLA composite would, in turn, influence the mechanical properties of the material, enhancing the elastic modulus, strength and especially the ductility, toughness and impact resistance. Compatibilization between PLA and inorganic fillers can be achieved by many different routes [9], although reactive compatibilization is the most effective [3; 9]. Compatibilization can involve the seizing of fillers with organic agents that have high physical affinity to PLA. Otherwise, compatibilization can be achieved by the formation of chemical bonds among the fillers and polymeric chains [10; 11]. This result can be promoted by intermediate reactive molecules of appropriate molecular weight and steric hindrance, which can simultaneously combine with the functional groups on the mineral fillers (often, hydroxyls) and with the terminal reactive groups of the polymers. In the present work, PLA has been co-formulated with a lamellar inorganic filler. This filler is constituted by micro-sized natural talc that is cheap and can be used in all those sectors in which the presence of nano-scale

ACCEPTED MANUSCRIPT components represent a limiting factor or are forbidden by the current regulations. Talc is known to act as powerful nucleating agent for PLA [12; 13]. It is already effective at low concentration (2 wt. %). It increases the overall PLA crystallinity, thus reducing the degradation of the polymer and improving its mechanical resistance and durability [14]. In general, talc dispersion within a PLA matrix is hindered by the hydrophilic nature of talc itself, which does not allow a good interaction with the compostable polyester because PLA is essentially hydrophobic. Poor compatibility between the organic and inorganic fraction of the talc-PLA composite leads to a mediocre mixing of the two phases, thus limiting the achievable mechanical properties. The higher the interaction between the surface of the talc and the bulk of the polymeric matrix, the greater the capability of the filler to absorb stresses applied on the polymer. The stresses are both internal, due to shrinkage after moulding processes, and external, due to the end-use conditions of the material or its accidental impacts, its barrier properties and lastly its efficiency as a nucleating agent. In this framework, the present investigation concerns the direct functionalization of micro-lamellar talc by the exploitation of two different classes of chemical reagents employed as coupling agents: organosilanes and polyisocyanates. The functionalization consists of the formation of chemical bonds between the coupling agent and the surface of the mineral filler. The pre-treated talcs are thought to favor the so-called affinity interactions inside the PLA composite or, alternatively, to establish stronger chemical bridges with the PLA matrix by only relying on the chemical functionalities of the coupling agent. In particular, as the first type of chemical functionalization, two organosilane molecules, (3-aminopropyl)triethoxy silane (APTES) and (3glycidyloxypropyl)trimethoxysilane (GLYMO), were grafted to the talc surface via hydrolysis and condensation reactions between the silanol end-groups of the hydrolysed silanes and the hydroxyl groups on the talc surface. Two commercially available polyisocyanates with hydrophilic and hydrophobic character were also employed to functionalize talc through reaction of the isocyanate groups with the hydroxyl groups on the talc surface. The experimental analyses allowed the preparation of talc-PLA pellets by reactive extrusion compounding as well as the comparative evaluation of the effect of talc pretreatment on the physical (mostly, thermal) properties and chemical structures of the resulting PLA compounds.

ACCEPTED MANUSCRIPT 2. Experimental 2.1 Materials Poly(Lactic Acid) (PLA) Ingeo Biopolymer 3260 HP (Nature Works LLC, Minnetonka, Minnesota, USA), microlamellar talc Luzenac HAR W 92 (Imerys S.A., Paris, France), Acrylic Impact Modifier Paraloid BPM-515 (Dow, The Dow Chemical Company, City, Midland, Michigan, USA) and Acrylic Melt-Strength Enhancer Paraloid BPMS-260 (Dow, The Dow Chemical Company, City, Midland, Michigan, USA) are commercially available materials and they were used as received. For the functionalization of talc, (3-amino propyl)triethoxy silane (APTES, Evonik, Essen, Germany), (3-glycidyloxypropyl)trimethoxysilane (GLYMO, Evonik, Essen, Germany), hydrophilic poly-isocyanate Bayhydur 3100 (Bayer Material Science AG, Leverkusen, Germany) with an isocyanate groups content of 17.4 wt. % and hydrophobic poly-isocyanate Desmodur N3900 (Bayer Material Science AG, Leverkusen, Germany) with isocyanate groups content of 23.5 wt. % were used. Trizma base and proteinase K, both employed for selective hydrolysis of the composite surface, together with all the puregrade solvents used were purchased from Sigma-Aldrich (Sigma-Aldrich Italia, Milano, Italy). On one hand, (3-amino propyl)triethoxy silane (APTES) and (3-Glycidyloxypropyl)trimethoxysilane (GLYMO) can be in turn grafted to the talc surface via hydrolysis and condensation reactions taking place between silanol end groups of the hydrolyzed silane molecules and hydroxyls groups on talc surface. The APTES silane, bearing an amino group on its lateral chain, can also bond with the PLA chains, in principle, by aminolysis reaction between its amino group and ester group on PLA backbone, leading to the formation of an amidic derivative. Covalent bonds between the hydroxyl terminal groups on the polymeric chain of the PLA and the oxirane ring immobilized on talc through the reaction with GLYMO can also take place. On the other hand, the two commercially available poly-isocyanate molecules endowed with different hydrophilic/hydrophobic characteristics can be, alternatively, grafted to the talc surface via the urethane linkage formation between isocyanates and hydroxyl groups on the talc surface. This procedure should follow the reaction pathway of the nucleophilic addition of the hydroxyl groups to the isocyanate groups, thus forming urethane bonds. 2.2 Procedures of talc functionalization with silanes and polyisocyanates APTES was grafted on reduced graphene oxide, as described elsewhere, [15] to functionalize the microlamellar talc. In particular, 1 g of the micro-lamellar talc was treated in a round-bottom flask with 20 ml of a 1 M aqueous APTES solution for 18 h under magnetic stirring at room temperature. Then, the suspension of the treated talc was collected by vacuum filtration to remove the reaction solution and purified through 3 washing cycles in a centrifuge (5 min, 4000 rpm) with pure isopropyl alcohol. An analogous reaction procedure was followed to achieve the grafting of GLYMO (Figure 1). Grafting of the two polyisocyanates (Figure 2) was achieved by adapting a procedure reported in the literature [16]. In particular, by taking into account the stoichiometric ratio between the total hydroxyl content of talc and the isocyanate content of polyisocyanate, 1 g of micro-lamellar talc was treated in a round-bottom flask with 3 g of Bayhydur 3100 (Figure 1) that was previously dissolved in 30 ml of anhydrous toluene at 110 °C in a thermostatic bath for 30 min under reflux and magnetic stirring. Then, the treated talc was collected by vacuum filtration and purified through 3 washing cycles in a centrifuge (10 min, 4000 rpm) with pure butyl acetate. The same reaction procedure was followed for the treatment of talc with Desmodur N3900 polyisocyanate (Figure 2), by adjusting the quantity of the latter to account for the relative content of isocyanate groups with respect to the hydroxyl groups of talc. Each washing cycle was preceded by an ultrasonic bath treatment for 1 min. Once collected, the powders were oven-dried at 60 °C until a constant weight was reached (~72 h). To assess the effectiveness of the purification procedure and remove the unreacted species adsorbed on the talc surface, a control experiment was performed after washing, in which the previously purified samples were refluxed in hot solvent for 10 h. Following the characterization of the

ACCEPTED MANUSCRIPT talcs, the procedures were scaled-up to prepare a sufficient amount of material for the subsequent extrusion process of the PLA compounds.

Figure 1. Chemical condensation reaction between talc and a) hydrolysed APTES and b) hydrolysed GLYMO.

ACCEPTED MANUSCRIPT

Figure 2. Chemical reaction via urethane linkage formation between talc and a) commercial Bayhydur 3100 polyisocyanate and b) commercial Desmodur N3900 polyisocyanate. 2.3 Characterization of treated talcs Compositional analysis of unmodified talc and the treated talcs was carried out by Fourier transform infrared (FT-IR) analysis using an IR spectrophotometer (Jasco 6600, Jasco Inc., Easton, Maryland, USA). Before the FT-IR measurements, the talc powders were dried in a thermostatic oven (APT Line ED (E2), Binder Inc., Bohemia, NY, USA) at 60 °C overnight. After drying, the talc powders (3-3.5 mg) were ground in an agate mortar with 200 mg of spectroscopic-grade anhydrous KBr. After grinding, a pressure of 10 tons was applied to the sample for 10 min until a clear pellet, suitable for FT-IR analysis, was obtained. FT–IR spectra were recorded in transmittance mode by co-adding 30 scans in the 4000–600 cm-1 spectral range with a resolution

ACCEPTED MANUSCRIPT of 2 cm-1. The FT-IR spectra were submitted to baseline correction, normalized with respect to the Si-O bond stretching signal at 1052 cm-1 and displayed in absorbance mode. The presence of organic moieties on the talc surface and the resulting variation in the talc surface properties was confirmed by static contact angle experiments. Powder samples of the untreated and treated talcs were subjected to compression by applying a pressure of 10 tons for 10 min to obtain flat tablets. Contact angle tests were then carried out by depositing a 100 µl drop of double-distilled water on the surface of each tablet at room temperature. Water drop images were then taken at scheduled times. The contact angles were calculated by data image processing in the AutoCAD software (Autodesk, 2012). Scanning Electron Microscopy (SEM) images of the micro-lamellar talc before and after the pre-treatments were taken by a Field Emission Gun - Scanning Electron Microscope (FEG-SEM Leo, Supra 35, Carl Zeiss SMT, Inc. Thornwood, New York). 2.4 Manufacturing process of PLA pellets by extrusion compounding Six different formulations were prepared for extrusion compounding the PLA pellets (Table 1), where 3000 g of material was used in each formulation. The micro-lamellar talc was pre-mixed with the other materials for 5 min using a planetary stirrer before compounding extrusion. After mixing, the formulations were dried in a vacuum oven (M40-VT, MPM Instruments, Bernareggio, Italy) for 24 h at 70 °C and 150 mbar. Compounding extrusion of the formulations was performed by a twin-screw extruder (Haake Polylab OS PTW 24/40, Thermo Scientific, Karlsruhe, Germany) with a screw diameter of 24 mm equipped with a strand cutting variable length pelletizer. The formulations were supplied to the extruder, with a speed of 80 rpm for the volumetric single screw feeder (OS, Type DRS 28, Thermo Scientific, Karlsruhe, Germany), and extruded with a screw speed of 60 rpm under the following temperature profile of the heating elements within the body: 180/190/200/200/200/190/190 °C. The feeder temperature was set at 175 °C. The extrusion head temperature was set at 190 °C. The twin-screw was set with mixing elements that provided additional shear mixing of the materials. The extrudate was cooled in a water bath and immediately pelletized. The resulting pellets were vacuum dried, stored in sealed bags and stored under anhydrous conditions. The scheme of the extrusion compounding process is shown in Fig. 3. Material

Type

Formulation #1, wt. %

Formulation #2, wt. %

Formulation #3, wt. %

Formulation #4, wt. %

Formulation #5, wt. %

Formulation #6, wt. %

PLA

Polymer Mineral filler Mineral filler Mineral filler

91.9

91.9

91.9

91.9

91.9

96.2

4.4

-

-

-

-

-

-

4.4

-

-

-

-

-

-

4.4

-

-

-

Mineral filler

-

-

-

4.4

-

-

Mineral filler

-

-

-

-

4.4

-

1.8

1.8

1.8

1.8

1.8

1.9

1.8

1.8

1.8

1.8

1.8

1.9

Talc Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900 BPM-515 BPMS260

Impact modifier Meltstrength enhancer

Table 1. List of materials for extrusion compounding.

ACCEPTED MANUSCRIPT

Figure 3. Scheme of the talc treatment and extrusion compounding process of the composites. 2.5 Characterization of PLA/treated talc composites Attenuated Total Reflection Fourier Transform-Infrared analysis (ATR FT-IR) of the pellets was performed by an IR spectrophotometer (Jasco 6600, Jasco Inc., Easton, Maryland, USA) equipped with an ATR device (PRO ONE, Jasco Inc., Easton, Maryland, USA). FT-IR spectra were recorded in transmittance mode by co-adding 30 scans in the 4000–600 cm-1 spectral range with a resolution of 2 cm-1. The FT-IR spectra were submitted to baseline correction, normalized with respect to the ester carbonyl stretching band at 1715 cm-1 belonging to the PLA matrix and displayed in absorbance mode. For comparison, the ATR FT-IR spectra of the microlamellar talcs alone after the different treatments were also acquired under the same conditions. Before FTIR analysis, all the materials were dried in a thermostatic oven at 60 °C and then analysed directly. Thermal analysis of the PLA pellets after extrusion compounding was performed by differential scanning calorimetry (DSC) using a DSC 200 PC Phox calorimeter (Netzsch, Germany). Approximately 30 mg pellets were placed in a hermetic aluminium pan under 40 ml/min nitrogen flow. The samples were first submitted to a heating scan from 25 °C to 220 °C at a heating rate of 10 °C/min and then allowed to cool to room temperature in still air. Then, the samples were submitted to a second heating scan under the same conditions. The weight crystallinity degree (Xc, %) of the composites was calculated according to the following equation:

Xc (%) =

Hm H°m × PLA

× 100 %

where ΔHm and ΔH°m (J/g) are the experimental and theoretical melt enthalpy of PLA, respectively. The experimental melt enthalpy is calculated as the difference between the area under the PLA melting peak and the area of the cold-crystallization peak, which eventually appears in the thermogram. The theoretical melt enthalpy is 93.0 J/g [16]. PLA is the actual weight fraction of PLA in the formulation. Scanning Electron Microscopy (SEM) images were taken with the same equipment in correspondence of the external surface of the single pellet, in order to examine the morphology of the composites, particularly as an effect of the functionalization of talc. Similarly, the morphology of the correspondent composites was analyzed after selective etching treatment, by means of hydrolysis of ester linkage of the PLA at the surface. For the purpose, a reference method [18] involving an enzymatic procedure was adopted. In particular, the pellets were kept for 4 hours under stirring in a solution prepared with 1 mg of Trizma base and 4 mg of Proteinase K diluted in 5 ml of distilled water at 37 °C and pH 8.0. The applied conditions provided the enzyme

ACCEPTED MANUSCRIPT with the highest activity. After etching, the samples were washed with distilled water in an ultrasonic bath for 30 min, dried and prepared for SEM analysis. 3. Results and discussion 3.1. Compositional analysis by Transmission FT-IR of functionalized talcs Figure 4 shows the FT-IR spectra of the treated talcs compared with the spectrum of untreated talc. The untreated talc (Figure 4(a)) features a narrow band associated with the stretching vibration of hydroxyl functionalities belonging to Mg-OH groups (3676 cm-1), along with the wider and weaker double band centred at 3500 cm-1, generated by the vibrations of interlayer (3560 cm-1) and absorbed (3400 cm-1) water [18]. A weak signal can be identified at 1640 cm-1, which is related to trace water adsorbed on the mineral (water bending vibration). The signal at 1453 cm-1 is related to the presence of impurities, such as calcium carbonate. The strong absorption centred at approximately 1052 cm-1 is assigned to the stretching of the Si-O bond, and lastly, the signal centred at 663 cm-1 is related to the stretching vibration of the Mg-O-Si bond. The strong influence of the untreated talc signals is observed in the spectra of all the treated talcs. Indeed, the absorbance bands located below 1500 cm-1 overlap perfectly with the spectral profile of untreated talc (data not shown). The absorption bands of untreated talc also dominate the spectrum of the treated talcs in the range of 4000 to 2000 cm-1.

Figure 4. Complete Transmission FT-IR spectrum of the untreated talc (a). Comparison between the spectrum of the untreated talc and the spectrum of treated talc, within the 3800- 2600 cm-1 spectral range, for talc

ACCEPTED MANUSCRIPT treated with APTES (b), GLYMO (c), Bayhydur 3100 (d) and Desmodur N3900 (e), respectively. The spectra compared were normalized with respect to the talc signal at 3676 cm-1. The absorbance band of the treated talcs located at wavenumbers lower than 1500 cm-1 overlaps perfectly the spectral profile of the untreated talc (i.e., data not showed). The spectra of the treated talcs and untreated talc differ by the presence of peaks between 3000 and 2850 cm-1. These very weak bands originate from the overlap of the stretching absorption peaks of the alkyl groups belonging to the organic moieties on the surface of the treated talcs introduced by the functionalization process. The presence of these absorptions state the effectiveness of the functionalization procedure proposed in agreement with the results reported in [20]. Firstly, for the APTES molecule (Figure 4(b)), the typical absorption due to the stretching vibration mode of the terminal primary amino group, appearing in the 3500-3400 cm-1 range, is masked by the signal of the water absorbed in the talc. Secondly, for the GLYMO molecule (Figure 4(c)), there are two different types of methylene group, namely those belonging to the linear part of the chain and those located inside the oxirane ring. These two different C-H alkyl groups result in two different absorption frequencies appearing in the FT-IR spectrum as a double peak at 2930 and 28267 cm-1, respectively. In the same spectral area, however, some of the characteristic signals of the GLYMO molecule are masked by the presence of signals from the mineral. The absorption due to the stretching vibration of the epoxy ring at approximately 3500 cm-1 overlaps the water signal from talc. Absorptions that characterize the vibration of the C-O-C group (1160 cm-1) or the vibration of the epoxy ring (1250 cm-1) are completely masked due to the overlap with the wide and strong signal of talc at 1052 cm-1.

Figure 5. Comparison between the spectrum of talc treated with Bayhydur 3100 and the spectrum of Bayhydur 3100 itself over the whole spectral range.

ACCEPTED MANUSCRIPT Figure 6. Comparison between the spectrum of talc treated with Desmodur N3900 and the spectrum of Desmodur N3900 itself over the whole spectral range. The talc signals also hinder the signals of the talc functionalized by the reaction with the Bayhydur 3100 (Figure 4(d)). Despite numerous intense absorptions that characterize the spectrum of the Bayhydur 3100 alone, the spectrum of the talc treated with the Bayhydur 3100 only shows peaks that are ascribable to the absorptions of C-H stretching vibration of the alkyl moieties (Figure 5). In fact, as mentioned above, the characteristic signals of the polyether functionalities of the Bayhydur molecule are not visible due to overlap with the talc bands related to Si-O and Mg-O bonds. In addition, the C-N stretching vibrations of the tertiary ammines in the polyisocyanate ring, which should appear in the 1250-1020 cm-1 range, are completely masked by the strong talc signal in that region. The signal at 2272 cm-1, which appears in the spectrum of Bayhydur 3100 itself and indicates the presence of a large amount of isocyanate groups, almost completely disappears in the spectrum of the talc functionalized with Bayhydur 3100. The disappearance of this signal occurs along with a noticeable increase in the intensity of the alkyl bands in the range of 3000 to 2850 cm-1. The possible signal corresponding to the secondary amine in the urethane bond present in the molecule and formed by the functionalization reaction is also not visible. In fact, in that range (3600-3200 cm-1), the broad signals of water absorbed in the talc block the signal. However, despite the large similarity in the spectra of untreated talc and talc after treatment with Bayhydur 3100, variations in the relative intensity of the broad signals of water at 3600-3200 cm-1 can be observed. This observation can be explained as a result of the overlap with the signals of the amine groups belonging to the newly formed urethane bonds. Analogous considerations arise from the observation of the spectra of the talc treated with the Desmodur N3900 and of the Desmodur N3900 alone (Figure 6). At the same way, changes in water band intensity can be observed. This is possibly caused by the overlapping with the amine contribution of the urethane group. From the other hand, the presence of the signals at 2930 and 2860 cm-1 (Figure 4(e)) was considered the main evidence of the occurred talc functionalization reaction with the Desmodur poly-isocyanate.

Figure 7. SEM images acquired at 20K× magnification for a) talc, b) talc treated with APTES, c) talc treated with GLYMO, d) talc treated with Bayhydur 3100 and e) talc treated with Desmodur N3900.

ACCEPTED MANUSCRIPT The compositional analysis showed slight but clear changes in the spectral signal of the talc after the various functionalization reactions, both using the two silanes and the two poly-isocyanates. The reactions that occurs in heterogeneous phase lead to surface modification of the chemical properties of the talc particles, without changing substantially their morphological characteristics (i.e., see the SEM images in Figure 7). In particular, the reaction conditions and the purification processes herein modified neither the grain size nor the aggregation state of the filler. This represents an important element to control the structure and distribution of the filler in the preparation of PLA compounds and, hence, the mechanical and functional properties of the reinforced PLA, regardless of the type and nature of the additional constituents that are added to PLA during the extrusion compounding step. 3.2. Static contact angle on treated talcs The measurement of the static contact angle gives information about the hydrophobic or hydrophilic nature of the treated talc surface based on the specific molecule that is used for the functionalization reaction. Figure 8 shows the measured value of the static contact angle for each different scenario under investigation. Untreated talc showed the lowest contact angle of 54°. This result can be attributed to the hydrophilic nature of talc itself. Treatment with organic molecules led to an apparent change in the surface tension of the talc, which causes, as expected, an increase in the static contact angle. The talc functionalized with Desmodur N3900 polyisocyanate boasts the highest contact angle of 79°. This can be ascribed to the higher hydrophobicity of Desmodur N3900 compared with Bayhydur 3100. The other contact angle values fall within the previous limits corresponding to the specific polarity of the added molecules, as an effect of the functionalization reaction.

Figure 8. Photographs of the water static contact angle and its measurement for untreated talc and each treated talc sample in pellet form. Material

Static contact angle with Water at RT (°C)

Talc Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900

54 ± 0.5 66 ± 0.5 70 ± 0.5 62 ± 0.5 79 ± 0.5

ACCEPTED MANUSCRIPT Table 2. Calculated static contact angle on the various talcs reduced in form of pellets. 3.3. Thermal properties of extruded PLA/treated talc composites The as-received PLA is an extrusion grade biopolymer that was characterized by a single heating cycle. It showed a glass transition temperature of 71 °C, a melting temperature of 187 °C and a weight crystallinity content equal to 40 wt. %. Figures 9 and 10, which refers to the first and second DSC scan, respectively, display the thermographs and show the thermal transitions (i.e., the glass transition temperature Tg, the heat of crystallization ΔHc, and the melting peak temperature Tm) of the six formulations after compounding extrusion. The weight crystallinity content (Xc, %) of each material after the first and second DSC scans was also calculated. Table 3 summarizes the thermal transitions of the investigated formulations and the related heats absorbed or released after the first and second DSC scans. Analysis of the first DSC scan provides information on the thermal properties of the pellets immediately after the compounding extrusion of each formulation. The talc-containing formulations #1, #2, #3, #4 and #5 feature a glass transition temperature Tg in the range of 56-60 °C, which is slightly lower than the Tg of formulation #6, in which talc is not present. As is known, the Tg of a polymeric material is related to the physical interactions among the different chains in the amorphous phase. Talc could play a role by inserting in the amorphous phase among the polymeric chains. Talc can increase the average distance among the chains and reduce the extent of their physical interactions. Accordingly, talc can affect the Tg of the polymeric material in which it is dispersed. In this respect, formulation #1, with untreated talc, featured the lowest Tg (56.6 °C). The surface treatment of talc in formulations #2, #3, #4 and #5 mitigates the effect of the talc on the Tg of the polymeric material (i.e., Tg of 58.5, 57.9, 57.3 and 59.9 °C). These formulations show an increase in the Tg, which is closer to the Tg of formulation #6, which does not involve talc (Tg of 60.1 °C, the highest). This effect can be explained by an increase in the physical interactions among the polymeric chains and pretreated talc or, simply, by considering the establishment of chemical linkages and/or physical interactions between the functionalized talc surface and the polymeric chains, which restrict the chain mobility of the polymer (Table 4). Formulation ID Thermodynamic parameter Tg (°C) Tc1 (°C) ΔHc1 (J/g) Tc2 (°C) ΔHc2 (J/g) Tm (°C) ΔHm (J/g) ΔHcTOT (J/g) Xc (%)

Scan 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd

#1 (*)

#2 (*)

#3 (*)

#4 (*)

#5 (*)

#6 (**)

56.60 62.40 92.40 94.00 30.66 28.52 164.50 162.80 2.31 2.28 181.30 178.20 54.19 48.45 21.22 17.66 22.82 18.99

58.50 62.40 93.2 91.7 35.03 28.84 165.2 163.0 2.04 2.55 180.5 178.3 56.41 55.57 19.34 24.18 20.80 26.00

57.9 62.0 92.6 93.3 31.48 33.19 163.9 162.6 2.69 3.90 179.9 178.0 49.80 53.91 15.64 16.82 16.81 18.08

57.3 62.2 95.2 95.0 28.34 32.14 164.1 162.9 4.07 3.11 179.8 178.4 46.64 48.50 14.23 13.24 15.31 14.24

59.9 62.8 94.6 94.4 32.52 34.90 164.3 162.9 2.92 4.01 180.3 178.4 50.95 54.47 15.50 15.57 16.66 16.74

60.10 62.00 103.20 105.00 34.72 37.45 166.60 166.20 1.86 1.94 181.70 179.80 40.77 42.97 4.19 3.58 4.51 3.85

ACCEPTED MANUSCRIPT Table 3. Thermal transitions of the pellets after the first and second scanning pattern. Numerical values of the ΔH parameters were normalized respect to the 91.9 (*) and 96.2 (**) weight % of the PLA (PLA) in the formulation. Talc greatly influences the crystallization behavior of the PLA matrix, decreasing the temperature of cold crystallization, Tc1, and accelerating the crystallization kinetic. In fact, the formulations #1, #2, #3, #4 and #5 that contain talc exhibit a lower cold crystallization temperature Tc1 (~92-96 °C). The formulation #6 without talc presents a Tc1 of 103.2 °C. Micro-lamellar talc is well known to promote, at low concentration already, crystallites nucleation in polymers as other fillers do [21]. It also accelerates the crystallization kinetic of the Poly (Lactic Acid) (PLA) as reported in [22]. However, the talc treatments were found to play a significant role on Tc1 of PLA compounds. In fact, the formulations #2, #3, #4 and #5, which involve the pretreated talcs, show on average a Tc1 higher than that of formulation #1, which involves untreated talc. However, formulation #6, without talc, shows the highest Tc1 by far. The talc treatments mitigate the effect of talc on the crystallization behaviour of the PLA compounds, leading to intermediate values of Tc1. The talc treatment also reduces the PLA crystallization kinetics, especially for the formulations #4 and #5, which contain talc treated with highly reactive polyisocyanates (Tc1 = 95.2 and 94.6 °C). The talc treatments herein were designed to promote chemical/physical interactions, which in principle could favour compatibilization between the filler and the matrix and also reduce the polymer chain mobility during compounding extrusion. Amorphous PLA, which can interact closely (i.e., more effectively) with the pretreated (functionalized) mineral fillers, is less prone to organize into orderly structures (such as crystallites), especially in the presence of talc due to the aforementioned reduced mobility. As a result, the cold crystallization temperature of PLA with the pretreated talcs is expected to be higher than that of PLA with untreated talc. Accordingly, the crystallization kinetics of PLA with the pretreated talc compounds are slowed. Type of talc functionalization Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900

Possible chemical linkage with PLA matrix - Aminolysis of PLA chains - Amine reaction with PLA terminal carboxylic groups - Epoxy group reaction with PLA terminal carboxylic groups - Isocyanate reaction with PLA terminal hydroxyls - Isocyanate reaction with PLA terminal hydroxyls

Table 4. Hypothetical linkages established between the chemical functions introduced on talc by the various functionalization reactions and the PLA matrix. The peak melting temperatures (Tm) of the formulations remain roughly constant (~180-182 °C). The formulations containing talc (#1, #2, #3, #4 and #5) boast the highest melting enthalpy, especially formulation #1, which involves untreated talc (ΔHm = 54.19 J/g), and formulation #2, which involves APTES-treated talc (ΔHm = 56.41 J/g). The former boasts the highest weight crystallinity degree (Xc = 22.82 %) and the lowest Tc1 (92.4 °C). Formulation #1 crystallizes rapidly during compounding extrusion due to the talc. As mentioned earlier, talc acts as a nucleation promoter and accelerates the crystallization kinetics of the material. Formulation #6, without talc, has the lowest weight crystallinity degree (Xc = 4.51 %), a low melt enthalpy (ΔHm = 40.77 J/g) and the highest Tc1 (103.2 °C). Formulation #6 crystallizes very slowly during compounding extrusion. The weight crystallinity degree of this material after compounding extrusion is the lowest. In contrast, the weight crystallinity degree after compounding extrusion of formulation #1 is the highest. Therefore, the formulation that features the highest crystallization degree (i.e., formulation #1) is also characterized by the lowest cold crystallization temperature and vice versa. Accordingly, a decrease in the cold crystallization temperature is strongly related to the acceleration of the crystallization kinetics of PLA

ACCEPTED MANUSCRIPT compounds and vice versa. The formulations #2, #3, #4 and #5 display, more or less, an intermediate behaviour, highlighting a decrease in Xc and an increase in Tc1 with respect to formulation #1. In fact, as shown previously, talc promotes the crystallization kinetics of the polymer, while the surface treatments of talc mitigate its nucleating effect by promoting interactions with the PLA matrix. These PLA–treated talc interactions reduce the PLA chain mobility and, therefore, hinder the arrangement of the polymer into orderly spatial structures, such as crystallites. The Tg of the formulations after the second DSC scan do not greatly differ and average to 62 °C. Nevertheless, the Tg after the second DSC scan (#1, #2, #3, #4, #5, and #6) shows a significant shift (T = 2-6 °C) towards higher temperatures compared with the results of the first DSC scans. Specifically, formulation #6, without talc, shows a slightest increase in its Tg (T = ~2 °C) and the lowest value of Xc due to the absence of nucleating filler, as in the former case. Therefore, after cooling from 220 °C to room temperature after the first DSC scan, the amorphous phase shows an increase in the Tg, which should, more or less, result in the reduced flexibility of the corresponding PLA matrix. The stiffening (i.e., higher Tg) of the amorphous phase of PLA, inferred from the second DSC scan, can be attributed to 1) the establishment of stronger matrix/filler interactions or cross-linking reactions within the PLA matrix due to thermal degradation that occurred during the heating step of the DSC measurement, which in turn prevented PLA from crystallizing and thus lowered the Xc, and 2) an increase in Xc determined by thermodynamic and kinetic factors and can be ascribed to the specific thermal history of the material during DSC measurement (i.e., cooling). Formulations #2 and #3, both involving silane-treated talc, show an increase in Xc along with an increase in Tg after the second DSC scan with respect to the first scan. Formulations #4, #5 and #6 show nearly constant Xc values with an increase in Tg. Conversely, formulation #1 shows an increased Tg and a slight reduction in Xc. Indeed, other factors can influence the complex relationship between Tg and Xc after the second DSC scan, for example, the reduction in the molecular weight of PLA due to thermal (hydrolytic) degradation during DSC measurement. However, among the talc-containing formulations, formulations #4 and #5, which involve talc treated with the two polyisocyanates, show the lowest Xc, after both the first and second DSC scans, while formulation #2, which involves APTES-treated talc, shows the highest final Xc. This result can be explained in the light of the reduced reactivity of the APTES with the talc. APTES silane can graft to the talc surface via hydrolysis and condensation reactions taking place between silanol end groups of the hydrolyzed silane molecules and hydroxyls groups on talc surface. In particular, the APTES silane, bearing an amino group on its lateral chain, can also bond PLA chains, in principle, through aminolysis reaction between its amino group and ester group on PLA backbone, leading to the formation of an amidic derivative. In general, APTES silane shows a very high reactivity due to the alkaline nature of its lateral chain, which is able to catalyze the reaction of hydrolysis and condensation of silanol end group and, eventually, of other silanes with lower reactivity. Indeed, the reactivity among silicates and alkoxy silanes is high regardless the chemicals involved. In contrast, the rate of functionalization of a mineral filler like talc and the kinetic of these reactions not only depend on the nature of the lateral chain on the silane. They also depend significantly on the type of mineral filler involved. For example, in an earlier work, silica gel showed a higher reactivity with amino propyl triethoxy silane by hydrolysis and condensation mechanisms than pyrogenic silica, despite pyrogenic silica is characterized by a lower number of reactive hydroxyl groups [23]. The greater reactivity of the fumed silica was attributed to the smaller size of the particles. In fact, that gives a greater dispersion and mobility of the solid reactant that is able to rearrange after condensation with the silane, exposing new hydroxyls sites, which can react again [23]. In addition, a greater presence of water molecules replacing magnesium ions within the crystal lattice of phyllosilicates would increase the hydrolysis degree of talc [24]. This results in the following ascending order of reactivity: synthetic phyllosilicate > calcined synthetic phyllosilicate > natural talc. The final properties of the treated talc and its rate of functionalization also depend on spatial organization of the crystalline structure, that is, the distance between two different

ACCEPTED MANUSCRIPT layers. In fact, silanization reaction occurs mostly between phyllosilicate layers, making the penetration of silane within the multilayer structure a rate-determining step. More dramatic conditions during the functionalization process of natural talc such as higher reaction temperature, higher concentration of reagents, longer reaction times and addition of extra catalysts are necessary for optimization purposes.

Figure 9. Thermograms of the first heating scan of the non-reinforced PLA sample and the various PLA formulations reinforced with modified and unmodified talc.

Figure 10. Thermograms of the second heating scan of the non-reinforced PLA sample and the various PLA formulations reinforced with modified and unmodified talc. 3.4. Compositional analysis by ATR FT-IR of extruded PLA/treated talc composites Figure 11 displays the ATR FT-IR spectrum of formulation #6, that is, PLA after compounding extrusion with the impact modifier and melt-strength enhancer, as well as that of the as-received PLA. Commercial-grade PLA features overtone bands corresponding to water in the range of 4000 to 3481 cm-1, as well as a peak at 3294 cm-1 attributed to the stretching mode of the hydroxyl groups adjoining the –CH2– groups in the polymer. In this case, the –OH groups were attributed to the terminal hydroxyl groups of the PLA chains. In fact, after compounding extrusion of formulation #6, this peak disappears. The disappearance of the terminal –OH groups in formulation #6 can be attributed to their combination with the reactive counterparts in the impact modifier and/or melt-strength enhancer added to the formulation before compounding extrusion. In

ACCEPTED MANUSCRIPT addition, impact modifiers and melt-strength enhancers are known to act by promoting entanglement in the polymeric material by chain extension or by promoting branching reactions. In PLA, these reactions may involve the terminal hydroxyl groups, which therefore, can no longer be detected by FT-IR analysis after compounding extrusion. Commercial-grade PLA exhibits typical absorption bands at 2995 and 2946 cm-1 of – CH bond stretching belonging to the lateral –CH3 groups of the polymeric chains, in agreement with [24], even if these peaks are rather faint. Commercial-grade PLA also features two absorption peaks at 2916 and 2849 cm-1, attributed to additional stretching vibrations of –CH bonds. In addition, the spectrum features a small shoulder peak at 2885 cm-1 ascribable to the stretching vibration of the –CH bond on the tertiary carbon atoms of the PLA chain, in agreement with [25]. Formulation #6 also displays the typical absorption bands at 2995 and 2946 cm-1 of –CH bonds stretching. In contrast, absorptions related to –CH bond vibrations found in the spectrum of commercial-grade PLA at 2916 and 2849 cm-1 decrease in intensity (or almost disappear). The decrease or disappearance of the peaks attributed to –CH bond stretching vibrations at 2916 and 2849 cm-1 in formulation #6 can be attributed to the different assembly of the polymeric chains after compounding extrusion. As reported in Section 3.3, formulation #6 is characterized by an extremely low weight crystallization degree (Xc of ~4.5 wt. % for formulation #6 vs. Xc of ~40 wt. % for commercial-grade PLA). As reported in the literature [17], the mobility of the polymeric chains is greatly affected by the weight crystallization degree. In particular, an increase in the crystallinity of the polymer influences the mobility of the polymeric chains, which preferentially arrange in anti (trans) conformations. The different conformations of the polymeric chains significantly influence the vibrational modes and, accordingly, the detection and/or intensity of the corresponding absorption peaks. Figure 11 also reports the stretching vibration mode of the carbonyl groups of the commercial grade PLA and of the formulation #6 that are located at 1747 cm-1. Absorption peaks in the finger print range of commercial grade PLA and formulation #6 presents some remarkable differences. General absorption bands of commercial grade PLA include the peak at 1454 cm-1 and the double peak at 1378 and 1358 cm-1, respectively attributable to the bending deformation of -CH3 group and the deformation mode of the –CH bonds, the stretching vibrations of the ester groups at 1266, the coupling of the rocking vibration mode of the -CH3 with the stretching of the ester groups at 1212 and 1182 cm-1 in agreement with [26], the asymmetric rocking of -CH3 groups at 1126 cm-1, the symmetric stretching of the ester groups at 1081 cm-1 and the deformation of the C-CH3 bonds at 1035 cm-1. Other peaks are found at 955 and 921 cm-1. The peak at 955 cm-1 is elsewhere described as an amorphous band of the polymer, while the peak is characteristic of  crystals [27]. The signal of some vibrations related to C-C scheletal bonds is found at 754 and 686 cm-1 in agreement with [27]. Lastly, a rather big absorption peak, ascribable to terminal acrylate double bonds, can be found at 1560 cm-1. After compounding extrusion, this peak always disappears because of the combination of the terminal acrylate double bonds with their counterparts, found abundantly in both the acrylate-based impact modifier and melt strength enhancer additives. The adjoining peak at 1640 cm-1 is presumably attributable to the bending of entrapped water inside the commercial pellets. The formulation #6 features a reduction in the peak at 1212 cm-1, a decrease in the peak at 921 cm-1 and a corresponding increase in the absorption peak at 955 cm-1, a decrease in the signal of the scheletal C-C vibration bands at low wavenumber. These changes can be explained in the light of the reduced degree of weight crystallinity of the formulation #6. In particular, the concurrent decrease in the peak at 921 cm-1, associated to the  crystals in the polymer, and the increase in the absorption peak at 955 cm-1, associated to the amorphous phase, according to Meaurio et al. [27], corroborate the hypothesis that the changes in the absorption bands of the formulation #6 are essentiality related to the strong reduction in the crystallinity of the polymeric material.

ACCEPTED MANUSCRIPT

Figure 11. ATR FT-IR spectra in the range of a) 4000-2500 and b) 2000-600 cm-1 of formulation #6 and asreceived commercial-grade PLA, for comparison. Figure 12 displays the plots of the ATR FT-IR spectra in the fingerprint region of the formulations #1, #2, #3, #4, #5 after compounding extrusion process. They are compared, one by one, to the spectrum of the respective filler (i.e., untreated talc and treated talcs) and the spectrum of the formulation #6. Due to the rather low concentration of the talc employed in the formulation, all the spectra of the resulting PLA compounds containing talc almost resemble the features of the PLA spectrum (substantially, the formulation #6). The presence of talc comes out only from the observation of the weak signals at about 1018 (shoulder) and 664 cm-1 related to the typical absorptions of Si-O and Mg-O-Si bonds of talc, respectively. This result agree with what previously found, in FT-IR transmission mode, at 1052 and 663 cm-1, respectively. Decreasing in the crystallinity of the formulations compared to the as-received PLA spectrum is appreciable with good evidence by the analysis of the double peak at 1212 and 1182 cm-1 variation (Figure 13). In fact, as reported widely in the literature, a decrease of the intensity ratio of the peak at 1212 respect to the peak at 1182 cm-1 (I1212/I1182) is strictly related to structural changes from a more ordered structure towards an increase in the amorphous state of the PLA chains in samples obtained by melt crystallization [27]. The evolution of the spectra agrees with the decrease in crystallinity content of the respective formulations in accordance to thermal analysis. In particular, the as-received sample with 40 wt. % crystallinity content shows a value of the (I1212/I1182) ratio higher than that found for all the tested formulations, while it becomes lower for the formulation #1, which has a 23 wt. % crystallinity. For all the formulations with lower crystallinity, the ratio (I1212/I1182) results in an intermediate range between the value showed by formulation #1 and formulation #6, the latter being, indeed, characterized by the lowest crystallinity content (Figure 13(b)). Due to the low talc content with respect to the PLA phase, the ATR analysis cannot provide direct information on the presence of chemical functionalities generated by the reaction of functionalized talcs and the PLA matrix or terminal groups (i.e., the presence of amide signals in the sample containing talc treated with APTES). In this case, the instrumental response is outside of the limits of detection. Nevertheless, the effect of such functionalization reactions is apparent because of the resulting changes in crystallinity content, as evidenced by both spectroscopic and thermal analyses.

ACCEPTED MANUSCRIPT

Figure 12. ATR FT-IR spectra in the range of 2000-600 cm-1 for formulations a) #1 (untreated talc), b) #2, c) #3, d) #4 and e) #5 (treated talcs) compared to the spectrum of the respective filler (neat talc and treated talcs) and the spectrum of formulation #6.

ACCEPTED MANUSCRIPT

Figure 13. a) ATR FT-IR spectra in the range of 1250-1150 cm-1 for formulations #1 (untreated talc), #2, #3, #4 and #5 (treated talcs) compared to the spectra of formulation #6 and the as-received PLA sample. b) Superimposition of the spectra of formulations #1 and #6. The accordance of data from thermal and FT-IR analysis can be corroborated by the evaluation of SEM images obtained before and after the etching of the surface of the pellets by using an enzymatic solution of proteinase K. Proteinase K is known to degrade L-lactide units, preferentially in the amorphous phase. Figure 14 shows the modification of the aspect of the surface of the pellets after the exposure to the enzyme action during 4 h. Especially in formulations #3, #4 and #5, the etching treatment made the talc lamellae visible at the surface, which were instead not clearly recognizable in the images of pristine samples for all the formulations, as they were embedded in the PLA matrix. In particular, for such formulations the degradation resulted larger, showing more extended erosion areas than formulation #1 and #2. The etched regions should be mostly constituted by amorphous phase, because of the selectivity of the enzyme. Thus, this evidence can be reported because of the lowest crystallinity weight content of formulations #3, #4 and #5 in agreement with the results of the thermal analysis.

ACCEPTED MANUSCRIPT

Figure 14. SEM images acquired at 9-8 KX magnification for the various PLA formulations reinforced with modified and unmodified talc. Left: as-is external surface. Right: surface after 4 h hydrolysis with a proteinase K solution.

Conclusions In this work, commercial-grade PLA was compounded in a twin-screw co-rotating extruder with the addition of micro-lamellar talc as a mineral filler and nucleation promoter. Several formulations were investigated, one involving unmodified talc and the others involving chemically treated talc. The compatibilization between PLA and the mineral filler was, therefore, investigated with emphasis on the effect on the

ACCEPTED MANUSCRIPT crystallization of the PLA compounds. Based on the experimental findings, the following conclusions can be drawn: - The modification of talc through functionalization reactions was evidenced by FT-IR analysis, showing the contribution of the grafted coupling agents. These modification reactions resulted in a change in the surface properties of talc, which became more hydrophobic, as supported by static contact angle measurements; - Effects induced by the treatment of talc were examined by DSC analysis, which showed broad variations in the thermal properties of the resulting PLA compounds with respect to the extruded untreated PLA. In particular, the compatibilizer, intended to increase matrix/filler interactions, resulted in a significant reduction in the Xc of the polymer. The improved talc/PLA interactions also caused a change in the nucleating character of talc, where the effect was stronger for talc treated with the polyisocyanates, probably due to the higher reactivity of the isocyanate groups and stability of the urethane linkage compared with the organosilanes; - Pretreatments of talc with seizing agents allow the accurate control of the crystallization extent and PLA crystallization kinetics compared with the addition of untreated talc. This result has important technological applications, as it can be used during melt processing to avoid excessive crystallization in PLA, which can compromise the overall mechanical properties of the polymer, causing excess embrittlement of the resulting material. In conclusion, the experimental results demonstrate that the crystallization of PLA can be fine-tuned by the appropriate modification of the constituents in the formulation. In addition, accomplishment of the desired crystallization degree of PLA leads to polymeric materials that can boast the necessary thermal stability. Accordingly, the custom-built PLA formulations feature both promising thermal properties and mechanical behaviour, once crystallization is appropriately controlled. Hence, the design of innovative formulations has great potential to increase the possible applications of degradable biopolymers, even in the field of semidurable and durable end-goods. Thus, degradable biopolymers could represent a reliable alternative to both durable biopolymers and conventional plastics in larger market segments.

Acknowledgments This work is part of the Project co-funded by the European Commission within the LIFE + Programme (2014–2020). Grant agreement no.: LIFE-PLA4COFFEE ENV/IT/000744. The partners of LIFE-PLA4COFFEE ENV/IT/000744 are kindly acknowledged for their continuous support and stimulating insights. Mrs. Patrizia Moretti and Prof. Vincenzo Tagliaferri are also acknowledged for their contribution during the development of the Project.

ACCEPTED MANUSCRIPT References [1] E. T. H. Vink, K. R. Rábago, D. A. Glassner, P. R. Gruber, Application of life cycle assessment to NatureWorksTM polylactide (PLA) production, Polymer Degradation and Stability 80 (2003) 403 [2]L. T. Lim, R. Auras, M. Rubino, Processing technologies for poly(lactic acid), Progress in Polymer Science 33 (2008) 820 [3] X. Zhang, M.A. Geven, D.W., Grijpma, J.E. Gautrot, T. Peijs, Polymer-polymer composites for the design of strong and tough degradable biomaterials, Materials Today Communications, 8 (2016) 53 [4] N. Ogata, G. Jimenez, H. Kawai, T. Ogihara, Structure and thermal/mechanical properties of Poly (llactide)-clay blend, Journal of Polymer Science PartB: Polymer Physics, 35 (1997) 389 [5] K. A. M. Thakur, R. T. Kean, J. M. Zupfer, N. U. Buehler, M. A. Doscotch, E. J. Munson, Solid State 13C CPMAS NMR Studies of the crystallinity and morphology of Poly(L-lactide), Macromolecules, 29 (1996) 8844 [6] J. J. Kolstad, Crystallization kinetics of Poly(i-lactide-co-meso-lactide, Journal of Applied Polymer Science, 62 (1996) 1079 [7] H. Urayama, C. Ma, Y. Kimura, Mechanical and thermal properties of Poly(L-lactide) incorporating various inorganic fillers with particle and whisker shapes, Macromolecular Materials and Engineering, 288 (2003) 562 [8] Y. Z. Wan, Y. L. Wang, X. H. Xu, Q. Y. LI, In vitro degradation behavior of carbon fiber-reinforced PLA composites and influence of interfacial adhesion strength, Journal of Applied Polymer Science, 82 (2001) 150 [9] V. Nagarajan, A.K, Mohanty, M. Misra, Perspective on Polylactic Acid (PLA) based sustainable materials for durable applications: focus on toughness and heat resistance, Sustainable Chemistry & Engineering 4 (2016) 2899 [10] M. Barletta, P. Moretti, E. Pizzi, M. Puopolo, V. Tagliaferri, S. Vesco, Engineering of Poly Lactic Acids (PLAs) for injection and compression molding: material structure and thermal properties, Journal of Applied Polymer Science (2016). Article in printing [11] M. Barletta, P. Moretti, E. Pizzi, M. Puopolo, V. Tagliaferri, S. Vesco, Thermal behavior of injection and compression molded custom-built Poly Lactic Acids (PLAs) (2016). Article in progress [12] D. Battegazzore, S. Bocchini, A. Frache, Crystallization kinetics of poly(lactic acid) – talc composites, Express Polymer Letters, 5 (2011) 849-858 [13] A. Shakoor, N.L. Thomas, Talc as nucleating agent and reinforcing filler in poly(lactic acid) composites, Polymer Engineering and Science, 54 (2014) 64-70 [14] A. M. Harris, E. C. Lee, Improving mechanical performance of injection molded PLA by controlling cristallinity, Journal of Applied Polymer Science 107 (2008) 2246

ACCEPTED MANUSCRIPT [15] M. Barletta., S Vesco., M. Puopolo, V. Tagliaferri, Graphene reinforced UV-curable epoxy resins: Design, manufacture and material performance, Progress in Organic Coatings, 90 (2016) 178-186 [16] G. Siqueira, J. Bras, A. Dufresne, New process of chemical grafting of cellulose nanoparticles with a long chain isocyanate, Langmuir 26 (2010) 402 [17] D. Garlotta, A literature review of Poly(Lactic Acid), Journal of Polymer and Environment 9 (2001) 63 [18] J. Bojda, E. Piorkowska, Shear-induced nonisothermal crystallization of two grades of PLA, Polymer Testing 50 (2016) 172. [19] S. A. Parry, A. R. Pawley, R. L. Jones, S. M. Clark, An infrared spectroscopic study of the OH stretching frequencies of talc and 10-Å phase to 10 GPa, American Mineralogist 92 (2007) 525 [20] A. A. Baba, A. S. Ibrahim, R.B. Bale, F. A. Adekola, A. G.F. Alabi, Purification of a Nigerian talc ore by acid leaching, Applied Clay Science 114 (2015) 476 [21] Z. Su, Q. Li, Y. Liu, G. H. Hu, C. Wu, Multiple melting behavior of Poly(lactic acid) filled with modified carbon black, Journal of Polymer Science: Part B: Polymer Physics, 47 (2009) 1971 [22] Y.H. Cai, Crystallization and melting behavior of biodegradable Poly(L-lactic acid)/talc composites, EJournal of Chemistry, 9 (2012) 1569 [23] K. Chabrol, M. Gressier, N. Pebere, M. J. Menu, F. Martin, J. Bonino, C. Marichald, J. Brendle, Functionalization of synthetic talc-like phyllosilicates by alkoxyorganosilane grafting, Journal of Materials Chemistry, 20 (2010) 9695 [24] M. Barletta, S. Vesco, M. Puopolo, V. Tagliaferri, High performance composite coatings on plastics: UVcurable cycloaliphatic epoxy resins reinforced by graphene or graphene derivatives, Surface and Coatings Technology, 272 (2015) 322 [25] V. Krikorian, D. J. Pochan, Crystallization behavior of Poly(l-lactic acid) nanocomposites:  nucleation and growth probed by infrared spectroscopy, Macromolecules, 38 (2005) 6520 [26] G. Kister, G. Cassanas, M. Vert, Effects of morphology, conformation and configuration on the IR and Raman spectra of various poly(lactic acid)s, Polymer, 39 (1998) 267 [27] E. Meaurio, N. López-Rodríguez, J. R. Sarasua, Infrared spectrum of Poly(l-lactide):  application to crystallinity studies, Macromolecules, 39 (2006) 9291

ACCEPTED MANUSCRIPT List of Captions Table 1. List of materials for extrusion compounding. Table 2. Calculated static contact angle on the various talcs reduced in form of pellets. Table 3. Thermal transitions of the pellets after the first and second scanning pattern. Numerical values of the ΔH parameters were normalized respect to the 91.9 (*) and 96.2 (**) weight % of the PLA (PLA) in the formulation. Table 4. Hypothetical chemical linkages between the various functions on treated talcs and the PLA matrix. Figure 1. Chemical condensation reaction between talc and a) hydrolyzed APTES silane and b) hydrolyzed GLYMO silane. Figure 2. Figure 2. Chemical reaction via urethane link formation between talc and a) commercial Bayhydur 3100 poly-isocyanate and b) commercial Desmodur N3900 poly-isocyanate. Figure 3. Scheme of the talc treatment and extrusion compunding process of the composites. Figure 4. Top: complete Trasmission FT-IR spectrum of untreated talc. Bottom: comparison between the spectrum of untreated talc and the spectra of each treated talc, within the 3800- 2600 cm-1 spectral range. The spectra in comparison were normalized respect to the talc signal at 3676 cm-1. Figure 5. Comparison between the spectrum of talc treated with Bayhydur 3100 and the spectrum of the free reagent, within the whole spectral range. Figure 6. Comparison between the spectrum of talc treated with Desmodur N3900 and the spectrum of the free reagent, within the whole spectral range. Figure 7. SEM images acquired at 20 KX magnification for a) Talc, b) Talc AMEO, c) Talc Glymo, d) Talc Bayhydur 3100 e) Talc Desmodur N3900. Figure 8. Photographs of water static contact angle and its measurement for untreated talc and each treated talc sample in form of pellet. Figure 9. Thermograms of the first scan in heating of the non-reinforced PLA samples and the various PLA formulations reinforced with modified and unmodified talc. Figure 10. Thermograms of the second scan in heating of the non-reinforced PLA samples and the various PLA formulations reinforced with modified and unmodified talc. Figure 11. ATR FT-IR spectra in the 4000-2500 (a) and 2000-600 cm-1 range (b) of formulation #6 and asreceived commercial grade PLA samples, in comparison.

ACCEPTED MANUSCRIPT Figure 12. ATR FT-IR spectra in the 2000-600 cm-1 range of formulations #1 (untreated talc) (a), #2 (b), #3 (c), #4 (d) and #5 (e) (treated talcs) compared respectively to the spectrum of the relative filler (neat talc and treated talcs) and the spectrum of formulation #6. Figure 13. ATR FT-IR spectra in the 1250-1150 cm-1 range of formulations #1 (untreated talc), #2, #3, #4 and #5 (treated talcs) compared respectively to the spectra of formulation #6 and as-received PLA sample (a). Superimposition of the spectra of formulation #1 and #6 (b). Figure 14. SEM images acquired at 9-8 KX magnification for the various PLA formulations reinforced with modified and unmodified talc. Left: as-is external surface. Right: surface after 4 h hydrolysis with a proteinase K solution. List of Tables

Material

Type

Formulation # 1, wt. %

Formulation # 2, wt. %

Formulation # 3, wt. %

Formulation # 4, wt. %

Formulation # 5, wt. %

Formulation # 6, wt. %

PLA

Polymer Mineral filler Mineral filler Mineral filler

91.9

91.9

91.9

91.9

91.9

96.2

4.4

-

-

-

-

-

-

4.4

-

-

-

-

-

-

4.4

-

-

-

Mineral filler

-

-

-

4.4

-

-

Mineral filler

-

-

-

-

4.4

-

1.8

1.8

1.8

1.8

1.8

1.9

1.8

1.8

1.8

1.8

1.8

1.9

Talc Talc AMEO Talc Glymo Talc Bayhydur 3100 Talc Desmodur N3900 BPM-515 BPMS-260

Impact modifier Meltstrength enhancer

Table 1. List of materials for extrusion compounding. Material

Static contact angle with Water at RT (°C)

Talc Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900

54 ± 0.5 66 ± 0.5 70 ± 0.5 62 ± 0.5 79 ± 0.5

Material

Static contact angle with Water at RT (°C)

Talc Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900 Material

54 ± 0.5 66 ± 0.5 70 ± 0.5 62 ± 0.5 79 ± 0.5 Static contact angle with Water at RT (°C)

ACCEPTED MANUSCRIPT

Talc Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900

54 ± 0.5 66 ± 0.5 70 ± 0.5 62 ± 0.5 79 ± 0.5

Table 2. Calculated static contact angle on the various talcs reduced in form of pellets. Formulation ID Thermodynamic parameter Tg (°C) Tc1 (°C) ΔHc1 (J/g) Tc2 (°C) ΔHc2 (J/g) Tm (°C) ΔHm (J/g) ΔHcTOT (J/g) Xc (%)

Scan 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd 1st 2nd

#1 (*)

#2 (*)

#3 (*)

#4 (*)

#5 (*)

#6 (**)

56.60 62.40 92.40 94.00 30.66 28.52 164.50 162.80 2.31 2.28 181.30 178.20 54.19 48.45 21.22 17.66 22.82 18.99

58.50 62.40 93.2 91.7 35.03 28.84 165.2 163.0 2.04 2.55 180.5 178.3 56.41 55.57 19.34 24.18 20.80 26.00

57.9 62.0 92.6 93.3 31.48 33.19 163.9 162.6 2.69 3.90 179.9 178.0 49.80 53.91 15.64 16.82 16.81 18.08

57.3 62.2 95.2 95.0 28.34 32.14 164.1 162.9 4.07 3.11 179.8 178.4 46.64 48.50 14.23 13.24 15.31 14.24

59.9 62.8 94.6 94.4 32.52 34.90 164.3 162.9 2.92 4.01 180.3 178.4 50.95 54.47 15.50 15.57 16.66 16.74

60.10 62.00 103.20 105.00 34.72 37.45 166.60 166.20 1.86 1.94 181.70 179.80 40.77 42.97 4.19 3.58 4.51 3.85

Table 3. Thermal transitions of the pellets after the first and second scanning pattern. Numerical values of the ΔH parameters were normalized respect to the 91.9 (*) and 96.2 (**) weight % of the PLA (PLA) in the formulation. Type of talc functionalization Talc APTES Talc GLYMO Talc Bayhydur 3100 Talc Desmodur N3900

Possible chemical linkage with PLA matrix - Aminolysis of PLA chains - Amine reaction with PLA terminal carboxyls - Epoxy group reaction with PLA terminal carboxyls - Isocyanate reaction with PLA terminal hydroxyls - Isocyanate reaction with PLA terminal hydroxyls

Table 4. Hypothetical chemical linkages between the various functions on treated talcs and the PLA matrix.