Polyamide 6/silica hybrid materials by a coupled polymerization reaction

Please note that technical editing may introduce minor changes to the text and/or graphics, which may alter content. The journal’s standard Terms & Conditions and the Ethical guidelines still apply. In no event shall the Royal Society of Chemistry be held responsible for any errors or omissions in this Accepted Manuscript or any consequences arising from the use of any information it contains. Accepted Manuscript


Introduction
Polyamide 6 (PA6) is one of the most important engineering plastics, due to the combination of high mechanical and thermal stability, chemical resistance and processability. 1 Physical properties can be further improved by combination with other components, e.g. layered silicates [2][3][4] , glass or carbon fibres [5][6][7][8] and metal oxides, [9][10][11][12] to fabricate hybrid materials or composites. These materials are suitable for several applications, especially in automobile industry, for example as inlet for fuel systems, wheel trims and engine covers. 13 PA6/silica hybrid materials are of special interest because incorporation of SiO 2 into PA6 improves mechanical properties such as hardness and elastic modulus. 14,15 Furthermore, silica is non-toxic, colorless and nanoparticles as well as hybrid materials could be obtained by the sol-gel process at mild reaction conditions. Strategies for producing PA6/SiO 2 hybrid materials can be classified in different categories. In the simplest way, both components, the preformed SiO 2 and PA6, are mixed together by extrusion, melting or another appropriate procedure. [16][17][18][19][20][21] More elegant ways use the in situ formation of the SiO 2 component, i.e. by sol-gel processing 22 or the in situ polymerization of ε-caprolactam in presence of preformed silica particles. 15,[23][24][25][26] The grafting of PA6 on surface functionalized SiO 2 particles is also a suitable route to fabricate polymer/SiO 2 hybrid materials. 27,28 As an additional way, the simultaneous formation of both polymer components within one procedure is known. 14,29 So far, there is no report on the simultaneous synthesis of PA6 and SiO 2 from a combined monomer within one coupled polymerization reaction.
In the literature the terms composite and hybrid material have been used in different ways for those types of materials. 16 Composite materials are mixtures of both components on a length scale of 100 nm to 10 m, whereas inorganic/organic hybrid materials are combined at the molecular level up to several nm. The challenge is to create nanostructured hybrid materials, which show a stronger improvement in physical properties than macroscopic mixtures even at low filler contents of a few weight percent. To achieve a nanostructured hybrid material, the simultaneous formation of both components in vicinity is required. 30 Therefore, monomers have been constructed in such a way that two polymers are formed from one single source monomer. This strategy has been established for monomers which contain two different moieties suitable for polymerization, one for chain polymerization and another one, i.e. for sol gel processes. [31][32][33][34] However, in this case both groups do polymerize independently of each other. For step-growth polymerization processes, the twin polymerization has been established as an elegant route to fabricate nanostructured inorganic/organic hybrid materials. The formation of both polymers occurs mechanistically coupled, which is the reason for the smooth nanostructure formation. 30,35,36 Monomers which are used in twin polymerization are called twin monomers. The objective of this publication is the development of a coupled polymerization procedure for synthesis of PA6/SiO 2 hybrid materials within one process. Therefore, three different reactants, namely 1,1',1'',1'''-silanetetrayltetrakis-(azepan-2one) (Si(ε-CL) 4 ), 6-aminocaproic acid (ε-ACA) and ε-caprolactam (ε-CL) are used in different ratios to adjust the amount of formed silica. The combined monomer Si(ε-CL) 4 1 which contains the ε-caprolactam moiety covalently bound via the N-atom to the silicon is used as precursor for the formation of silica as well as polyamide 6. Basically for the purpose of PA6 production cyclohexanone oxime (CHO) as industrial precursor molecule for ε-CL seems also eligible as component in monomer 2 Si(CHO) 4 (Scheme 1). However, in this case the BECKMAN rearrangement of this monomer would be an essential step before polymerization takes place. Both types of silicon monomers are investigated for PA6/SiO 2 hybrid material synthesis. It must be mentioned, that the single polymerization of 1 or 2 is unsuitable to produce PA6/SiO 2 as the overall stoichiometry is not complied. Water is essential as co-component (Scheme 2). Therefore it must be emphasized that monomer 1 and 2 are not ideal twin monomers but they are related to deficient twin monomers due to the possible mechanistically coupled formation of the inorganic and organic polymer. The polymerization process as shown in Scheme 2 is related to the apparent twin polymerization. 37 The challenge of PA6/SiO 2 hybrid material synthesis according to Scheme 2 is the improvement of the operative coupling of several reactions. One crucial aspect is how the water equivalent can be realized among the occurring processes. Water can be used directly or from a suitable source like the polycondensation of 6-aminocaproic acid. ε-ACA seems to be eligible, because it polymerizes to PA6 and initiates the hydrolytic lactam polymerization. The coupling of the water delivering and the water consuming reaction (equation 1 and 2 in Scheme 2) to produce PA6/SiO 2 has been explored as function of the ratio of 1 and ε-ACA. As a third reaction component ε-caprolactam has been found suitable for a good homogenization of the reaction melt (according to equation 3 in Scheme 2). By variation of reactant concentrations different SiO 2 amounts are adjustable. The reactions have been carried out in a typical procedure, which is suitable to fabricate PA6 from ε-CL.

Materials and methods
ε-Caprolactam (> 99%) and silicon tetrachloride (99%) were purchased from Sigma Aldrich. 6-Aminocaproic acid (99%) was purchased from Alfa Aesar. Cyclohexanone oxime (97 %) was ordered from Acros. Toluene was dried by standard methods and distilled before use in argon atmosphere. CDCl 3 was dried with molecular sieve 4 Å and stored under argon.  Solid state NMR spectra were recorded using a Bruker Digital Avance 400 spectrometer, equipped with double tuned probes capable of MAS (magic angle spinning). 13

Synthesis of monomer 1 and 2
In a typical procedure, either ε-caprolactam or cyclohexanone oxime (20.0 g, 0.177 mol) was dissolved under stirring in anhydrous toluene (300 mL). Triethylamine (20.2 g, 0.200 mol) was added in slight excess in a single portion. This solution was cooled by water bath and SiCl 4 (7.5 g, 0.044 mol), dissolved in 100 mL anhydrous toluene, was added slowly through a dropping funnel under vigorous stirring. Immediately a white precipitate from triethylammonium chloride was observed. Subsequently, the solution was stirred at room temperature for 16 h. Triethylammonium chloride was separated by filtration. After removing of the solvent under vacuum, a white to beige solid was obtained.

Synthesis of hybrid materials
Composites were synthesized with a high pressure lab autoclave of Berghof company. A Teflon beaker was used as insert. Before heating to reaction temperature of 230 °C, the reactants ε-ACA, ε-CL and 1 were filled in the autoclave in the specified molar ratios (Table 1) at room temperature and then it was purged with Argon up to 8 bar and afterwards relaxed to atmospheric pressure thrice. The reaction took place under 8 bar initial pressure (argon) for at least 210 min, including ca. 60 min heating phase. During reaction, pressure increased to approximately 14 bar. 15 min before termination of reaction time, pressure was released slowly to start post condensation phase. After cooling down to ambient temperature, the white to beige-colored monolithic samples were crushed into smaller pieces. To remove residual monomers or oligomers the hybrid materials were purified by soxhlet extraction for 48 h with methanol and then dried in a vacuum oven at 40 °C to constant mass. The reference was synthesized according to the same polymerization procedure except the addition of 1.
To increase average molecular weight a post condensation reaction can be carried out at 200 °C and 5 mbar for 12 h afterwards. The hybrid material synthesis procedure is reproducible for several times. For further information of the reproducibility by the example of P2 see ESI † ( Fig. S1 for ATR-FTIR spectra, Fig. S2 for DSC traces and Table S1 for amount of extractables and quantitative elemental analysis). Both monomers 1 and 2 were not described accurately in the literature. 1 is only briefly mentioned in a patent. The synthesis for 1 and 2 has been done by reaction of SiCl 4 with ε-CL or cyclohexanone oxime using an appropriate amine base to bind the HCl (see Experimental Part). Both new monomers have been characterized by spectroscopic methods and single X-ray structure analysis. Figure 1 shows the molecular structures of the synthesized monomers. For bond lengths and angles, as well as data acquisition details see ESI † (Table S2,  octacoordination with a tetrahedral SiN 4 core and four oxygen atoms in the "outer sphere", capping the tetrahedral planes. [43][44][45] The chemical shift of the 29 Si NMR signals of monomer 1 in solid state and solution state are similar which indicates the same bonding motif (Fig. S3 in ESI †). The solid state 29 Si NMR signal appears at δ = -43.0 ppm, which could be explained by the electron withdrawing effect of the carbonyl groups next to the nitrogen atom. Therefore, no strong effect of an octacoordination could be observed by 29 Si NMR spectroscopy.
Consequently, it is unclear whether the short Si−O distances are due to an additional stabilization or just steric reasons. All attempts to polymerize 2 to any PA6 hybrid materials failed. It remains intractable for BECKMANN rearrangement towards the ε-CL component. Neither acid treatment nor heating in different melt compositions have been successful (Table S5 in ESI †). Therefore, solely 1 was further investigated for hybrid material synthesis. In spite of the fact that water is suitable to induce the thermal polymerization of ε-CL, the use of free water as source for the synthetic procedure was not favorable because of a lower conversion or discoloration of the products (Table S6 in ESI †). The amino acid ε-ACA as water source has been found to be the most convenient way. In preliminary studies, the reaction of 1 with ε-ACA has been studied by DSC measurements to optimize the polymerization temperature for the overall process (Fig. 2). Monomer 1 shows a complex temperature dependence. However melting and polymerization of ε-ACA at temperatures ≥209 °C is in sum The overall process for synthesizing PA6/SiO 2 hybrid materials has been carried out in a high pressure autoclave suitable for PA6 synthesis (Experimental Part). It must be mentioned, that the order of reactant addition can be varied. Prepolymerization of the reactants ε-ACA and ε-CL and subsequent addition of 1 is possible, but products show discoloration due to contact with air while heating. Reaction time and polymerization temperature are important for homogeneity and extractable amounts, so that for a better comparability all experiments are done with constant reaction conditions in an autoclave. All reactants are inserted at the beginning of the heating period and the reactions are done under inert atmosphere at a temperature of 230 °C for 3.5 h including the heating and post condensation phase. For further informaƟon see Table S7 in ESI †. The observed extractables of the hybrid materials amount 10−32 % and depend on the molar ratios of reactants (Table  1). Especially for the samples P1 and P5 a decreasing amount of extractables can be detected with a higher ε-ACA ratio.
The resulting PA6/SiO 2 hybrid materials are homogeneous solid materials (Fig. 3). Primary monolithic products were received but also granules can be fabricated. Furthermore, the resulting thermoplastic materials can be extruded to films.   Table 1) as monolith, granules and film.    TEM images of P2 show SiO 2 agglomerates with primary particles of 35−60 nm in size (Fig. 4).
Homogeneity of samples as well as agglomeration tendency are affected by the containing SiO 2 amount as can be seen from electron microscopic images (Table 2 and Fig. S4−Fig. S9 in ESI †). The higher the SiO 2 content which is formed during polymerization, the more agglomeration of inorganic particles can be observed. The resulting SiO 2 agglomerates build particles with different shapes for example up to 30 µm long needle-like (P1_1) or shell-like particles around hybrid material (P5_1) but often spherical as can be seen from the other examples. At highest SiO 2 content (P5_2) particles larger than 100 µm are obtained. Furthermore, molar ratios of reactants seem to have an influence on agglomeration tendency. Hence, for example at experiment with constant SiO 2 amount (P1 and P5) an increasing ε-ACA ratio causes a decrease of SiO 2 particle size. The obtained wide particle size distribution could have influence on mechanical behavior but this is part of further work. The solid state 13 C and 29 Si NMR spectra of the hybrid materials P2 evidence the molecular structure which relates to the PA6/SiO 2 (Fig. 5). The solid state 13 C NMR spectrum is in agreement with literature data of pure PA6 high in α-crystallinity. 46 The solid state 29 Si NMR spectrum of the hybrid material shows Q 3 and Q 4 signals, which indicate Si atoms bound to 3 or 4 other Si atoms over siloxane-bridges. It is not possible to distinguish if the Q 3 signal is caused by Si−OH or Si−OC groups. Therefore, bonding of carboxylic acid groups to silicon cannot be excluded.  DSC traces of the samples R, P1_1, P2 and P5_1 show the typical thermal behavior of thermoplastic PA6 with a melting point around 220 °C (Fig. 6). As already described for the solid state 13 C NMR spectra, the crystallinity of PA6 is predominated by α-modification.
Variation of the SiO 2 amount has no influence on the melting or crystallization temperature and the crystal modification.
The degree of crystallization with values between 30−40 % is independent of the SiO 2 content and indicates the presence of crystalline and amorphous regions in the organic polymer. For further information see  Fig. 7). Additionally the ATR-FTIR spectra give information about the crystallization behavior of PA6. The position of the Amide V band at 690 cm −1 and Amide VI band at 580 cm −1 indicate crystallization mainly in α-modification as evidenced from DSC and solid state 13 C NMR measurements, too. Amorphous polymer shows broad signals, which can also be observed. Crystallization in the γ-modification would induce bands at 712 cm −1 and 625 cm −1 for Amid V and VI. [47][48][49] For additional characterization of the obtained hybrid materials the molar mass of the obtained PA6 and the influence of post condensation on the molecular weights and the polydispersity index (PDI) were investigated by size exclusion chromatography (SEC, Table 3, Fig. 8 and Fig. S14 in ESI). All polymers show a monomodal distribution of the molar mass. The SiO 2 content seems to have a slight influence on molecular weight and its distribution. Therefore M w as well as M n and PDI increase with increasing the SiO 2 amount that attends with an arising ratio of ε-ACA in comparison to the other reactants, namely 1 and ε-caprolactam. Generally, high amino acid concentrations lead to an increasing reaction rate of polycondensation reaction and therefore higher chain lengths and higher conversion. For example, a decrease in ε-ACA:1 ratio for P1_1 to P1_3 causes a decrease in obtained molar mass due to a reduced conversion of the reactants. A diametrical effect is determined for P5_1 and P5_2. It must be mentioned, that for these cases a higher ε-ACA content leads to lower chain lengths due to a decrease of ε-caprolactam unit amount and a higher concentration of amino or carboxylic acid end groups. As expected, the post condensation reaction of P2 at 200 °C leads to much higher M w . Additionally, a broader distribution of the molar masses with a higher PDI after post condensation is observed. This is due to the not reached equilibrium of the post condensation reaction. Additionally, the water content in the polymer grains decreases from the inner to the outer sphere. Therefore a different molecular weight distribution over the polymer grain is ascertained. 50 In some cases a microscopic nonuniformity of the solid polymers, especially a distribution of crystallite sizes, is discussed as a further reason. 51 Kinetic and thermodynamic investigations of the presented polymerization type are still under study.

Conclusions
In this study polyamide 6/SiO 2 hybrid materials were produced by a coupled polymerization reaction of three monomeric components namely 1,1',1'',1'''-silanetetrayltetrakis-(azepan-2one), 6-aminocaproic acid and ε-caprolactam within one process. The amount of SiO 2 is tunable by variation of reactant stoichiometry, whereas the ratio of 1,1',1'',1'''silanetetrayltetrakis-(azepan-2-one) to 6-aminocaproic acid is of great importance for a high conversion and homogeneity. The investigations have shown that the filler has no significant effect on thermal properties. The average molecular weight slightly rises with increasing the SiO 2 content. Furthermore, the latter increased during post condensation reaction and, additionally, a broader PDI was determined. Examination of the crystallization behavior of the obtained PA6 by NMR as well as FTIR spectroscopy and DSC show a predominated α-crystallinity and amorphous regions. Unfavorable is the increasing agglomeration tendency of the primary nanoparticles with higher SiO 2 amounts. In contrast to conventional in situ procedures or the compounding for the synthesis of PA6/filler hybrid materials, the preceding modification of the filler and negative effects of modifier, like the thermal degradation, 52,53 can be omitted with the described method. Additionally, problems in terms of processing potential toxic nanoparticle powders are avoided. 54