**1. Introduction**

Polymer electrolytes are regarded as one of the most promising candidates in advanced electrochemical applications, such as "smart" windows, displays, sensors, and more importantly, rechargeable lithium batteries [1–4]. For this last one, in particular, the research has focused for decades on gel-type membrane [5], generally achieved by immobilizing a liquid solution (for instance, a polar aprotic organic solvent or mixtures with a lithium salt) into a hosting polymeric matrix, such as poly(ethylene oxide) (PEO) and its derivatives (e.g., polyacrylonitrile (PAN), poly(vinylidene fluoride) (PVDF), poly(methyl methacrylate) (PMMA)) [6,7]. Respect to liquid electrolytes, in fact, gel polymer electrolytes (GPEs) are able to conjugate high ion conductivities with good mechanical strength, flexible geometry, reducing of liquid leaking and, thus, higher safety [8].

Owing to its ability to dissolve a large variety of salts, through interaction of its ether oxygen with cations, PEO has been one of the most extensively studied polymer used to prepare solid-state electrolytes, lighter, thinner, and safer for lithium-ion polymer batteries [9,10].

Thought, the low ionic conductivities at room temperature (10−6–10−<sup>8</sup> S cm<sup>−</sup>1), the Li+ transference number lower than 0.5 and the poor mechanical strength, still hinder the large scale diffusion of PEO-based device. Conversely, PAN ensures an ionic conductivity of circa 10−<sup>3</sup> S cm<sup>−</sup>1, satisfactory flame and mechanical resistances, but the dimensional stability of gels is poor [11,12]. After GPE preparation, in fact, a phase separation between the encapsulated electrolyte solution and the PAN matrix typically occurs, leading to a leakage problem and, thus, the passivation phenomena

of the lithium electrode when in contact with the gel, as well as failure of the electrode/electrolyte contact both resulting in a dramatic reduction of the ionic conductivity.

One of the strategy undertaken to bypass the drawbacks is the blending method, according to which two or more polymers are mixed to obtain a blend electrolyte. As already probed [13–16] the method allows to easily control a large number of factors, directly affecting the thermal, mechanical and electrical properties of the final polymer electrolytes. By mixing PMMA and PVdF polymers, Nicotera and coworkers obtained a blend with remarkable improvement of mechanical stability respect to unblended polymers [17]. Helan et al. have been reported outstanding thermal stability up to 230 ◦C for PAN/PMMA blends, but with quite low ionic conductivity, of the order of 2 × 10−<sup>7</sup> S cm<sup>−</sup><sup>1</sup> [18]. Very interesting electrical behavior and dimensional stability have been obtained by Choi et al. on PEO-PAN blend gel electrolytes, despite no evidence regarding mechanical resistance being provided [19].

An alternative approach for creating gel electrolyte system with improved mechanical properties and electrochemical performances foresees the incorporation of nanoscale organic/inorganic fillers within the polymer matrix [20]. The addition of SiO2 [21], Al2O3 [22], TiO2 [23], and other metal oxides [24,25] generally act as solid plasticizers, softening the polymer backbone and, thus, enhancing the segmental motion of the hosting polymer which, in turn, results in improved ion conductivity.

Among inorganic fillers, layered nanoparticles based on clays have been actively investigated lately since they offer a large number of interesting properties such as high cation exchange capacity, large chemically active surface area, outstanding swelling ability, intercalation, catalytic activity, and high chemical and thermal stability. Finally, the properties of the smectite nanoclays can be tailored using simple chemical methods such as intercalation with organic or inorganic gues<sup>t</sup> molecules. From the above, the dispersion of proper clay minerals within the polymer matrix could enhance the ionic conductivity improving at the same time the strength and heat resistance of the GPE.

Smectite clay with different particle sizes has been effectively tested as filler for the preparation of PEO nanocomposite electrolytes, demonstrating a discrete improvement of ionic conduction [26]. Kurian et al. [27] have shown that the surface modification of clay by ion exchange reactions with cationic organic surfactants such as alkyl amines, enhance the chemical affinity with the polymer matrix, leading to exfoliation of the clay particles and improving the gel's strength. Organic montmorillonite (MMT) prepared by ion exchange with HTAB was dispersed in PAN polymer, obtaining a composite GPEs with improved thermal stability and ionic conductivity [28].

Despite the efforts, however, there is still the need to design a gel electrolyte able to guarantee adequate electrical performance without sacrificing mechanical strength and thermal resistance. In the present study, PAN/PEO blend (80:20 weight ratio) polymers were used in order to prepare nanocomposite GPEs with an organo-modified clay. Specifically, hydrated sodium calcium aluminum magnesium silicate hydroxide (SWy-2, Nanocor) was the natural montmorillonite/smectite clay selected since it is relatively inexpensive, widely available and has small particle size as well as it shows good intercalation capability. The organo-modification of the SWy-2 (org-SWy) was achieved by ion exchange reaction with hexadecyltrimethyl ammonium bromide (CTAB). The filler loading of org-SWy in the GPE was 10 wt % with respect to the polymers PAN/PEO. For the gel preparation, a mixture of ethylene carbonate (EC) and propylene carbonate (PC), with molar ratio EC:PC 1:0.4, was used as plasticizer, while lithium trifluoromethanesulfonate (LiTr) was the salt chosen.

In order to compare the effect of the clay on the gel properties, also not blended and filler-free GPE membranes were also prepared.

All the GPEs were investigated by thermal (DSC), morphological (scanning electronic microscopy-SEM) and mechanical (DMA) analysis, while the ion transport studies were conducted by electrochemical impedance spectroscopy (EIS) and by multinuclear NMR spectroscopy. In particular, the 1H, 7Li, and 19F pulse-field-gradient (PFG) method was employed to obtain a direct measurement of the self-diffusion coefficients both of ions and solvents plasticizers (EC/PC), while the spin-lattice relaxation time (T1) was obtained by the inversion recovery sequence.

The combination of the electrochemical and NMR data has provided a wide description of the ions dynamics inside the so complex systems, as well as information on ion associations and interactions between polymers, filler and ions.

#### **2. Materials and Methods**
