Mechanical Reinforcement of Polyamide 6 by Cold Hydrostatic Extrusion

This paper presents the effect of severe plastic deformation obtained using the cold hydrostatic extrusion (HE) method on the mechanical and structural properties of polyamide 6 (PA6). As a result of the plastic strain, a significant increase in ultimate tensile strength and tensile modulus were observed. Tensile strength rose by almost 500%, up to the level of 508 MPa, whereas the tensile modulus rose by about 65%. Flexural modulus increase was also observed to 3230 MPa, i.e., by approx. 160%. As a result of high plastic deformation, the structure of the polyamide 6 changed significantly, as evidenced by its fibrous nature as presented in the results of the scanning electron microscopy inspection (SEM). The surface quality of products investigated was tested using profilometry.


Introduction
Hydrostatic extrusion (HE) is one of the techniques used to obtain severe plastic deformation (SPD) in order to change a material's structure, as well as to alter mechanical and physical properties of metals and metal alloys. The effect of severe plastic deformation in metals is the transformation of a coarse-grained structure into an ultra-fine-grained or nanometric structure [1][2][3][4][5][6]. Many studies have shown that the refinement of the structure in HE processes leads to a significant improvement in the strength of metals and their alloys, such as aluminium, copper and their alloys [1,2], iron [3], titanium [4], steels [5] or cast iron [6]. Due to the deformation of the material in the HE processes using high hydrostatic pressure, the generation and propagation of cracks is inhibited, so it is possible to maintain the materials' cohesion, even at a very high level of strain, and thus drastically increase their strength and plasticity.
The HE processing of polymers was first described in 1964 by Pugh and Low [7]. Inoue et al. [8] described the HE processes for, inter alia, high-density polyethylene (HDPE), poly(vinyl chloride) (PVC), acrylonitrile-butadiene-styrene copolymer (ABS), polypropylene (PP), polyamide 6 (PA6), polyoxymethylene (POM) and poly(methyl methacrylate) (PMMA). HE processes were carried out with reduction in (R) within the range of 1-10 (R is the ratio of the cross-section area before and after the HE processes) at a temperature of 50-90 • C. The dependence of the extrusion pressure (p HE ) on the R ratio was analysed, confirming their commonly known linear dependence [9]. The linear dependence of p on R was also confirmed, inter alia, for polyethylene (PE) and PP in the works of Buckley and Long [10]; Davis [11] for PE; and Yoon et al. [12] and Ariyama et al. [13] for PP. They found that for amorphous polymers: poly(methyl methacrylate) (PMMA) and high impact polystyrene (HIPS), HE has a significant impact on their thermal properties. The strength tests of HDPE, PP, PVC and ABS subjected to the HE process were described by

Hydrostatic Extrusion
The principle of the HE method is presented in the work by Pachla et al. [37]. Rods with a diameter of 15.6 mm made of PA6 polymer were subjected to the process of cold hydrostatic extrusion using a press designed and manufactured at the Institute of High Pressure Physics of the Polish Academy of Sciences (Warsaw, Poland). The press, with a working chamber of 22 mm, enables extrusion at pressures of up to 2 GPa. Prior to the HE processes, the rods were coated with a silicone oil-based grease (Dow Corning Corporation, Midland, MI, USA) and squeezed out at a linear speed (v) in the range of 3.5-111 mm s −1 through forming a matrix with die the vertex angle 2α = 45 • and diameters of 6, 7 and 8 mm (Figure 1).
authors' knowledge, no information has been ever published on PA6 cold-deformed in the solid state using the HE method in a form that would allow its commercial use, e.g., in the form of a rod. The aim of this study was to investigate the influence of variable parameters of the HE processes on the structure and properties of PA6.

Material
The initial state material used was unmodified extruded polyamide PA6 in the form of a rod with a diameter of 15.6 mm, purchased from MEGA-TECH s.c., Grodzisk Mazowiecki, Poland [36]. The properties of the PA6 polyamide before HE processing are given in Table 1.

Hydrostatic Extrusion
The principle of the HE method is presented in the work by Pachla et al. [37]. Rods with a diameter of 15.6 mm made of PA6 polymer were subjected to the process of cold hydrostatic extrusion using a press designed and manufactured at the Institute of High Pressure Physics of the Polish Academy of Sciences (Warsaw, Poland). The press, with a working chamber of 22 mm, enables extrusion at pressures of up to 2 GPa. Prior to the HE processes, the rods were coated with a silicone oil-based grease (Dow Corning Corporation, Midland, MI, USA) and squeezed out at a linear speed (v) in the range of 3.5-111 mm s −1 through forming a matrix with die the vertex angle 2α = 45° and diameters of 6, 7 and 8 mm (Figure 1).  In order to avoid the effects of strong heating of the material during plastic deformation, the plastic strain zone and the material extracted from the die were intensively cooled with cold water. It was observed that the diameter of each rod after the hydrostatic extrusion process was larger than the diameter of the die. This is a phenomenon similar to the Barus effect, so-called post-extrusion swelling [38]. Depending on the die used, a variable degree of reduction ratio (R) and true strain (ε) are obtained. The reductions (R), calculated as the ratio of the diameter before and after the HE processes, correspond to the true strains (ε) in the range of 0.68-1.90, where the true strain is given by the natural logarithm of the extrusion ratio R. During the HE processes, a strong adiabatic heating effect occurs (AH), which is the result of changing the mechanical work of pressing into heat. The adiabatic temperature (T ad ) used to describe the parameters of the HE process was estimated using the following Formula (1) [39]: where: W-mechanical work of plastic deformation performed during extrusion (J); p HE -extrusion pressure (MPa); c-specific heat (J/g K) and ρ-density (g cm −3 ) and βshare of mechanical work that has been converted into thermal energy (0.95 was adopted for the HE processes).
In the case of polyamide PA6 deformed at room temperature within an extrusion pressure range between 80 and 430 MPa, ρ = 1.14 g cm −3 and c = 1.7 J/g K, the adiabatic temperature increase ∆T fell within the range of 50 to 211 • C.
The PA6 polymer is a hygroscopic material and has a tendency to absorb water. Due to the necessity of using water cooling in the HE processes to limit the effect of released heat on the properties of the final product, the materials were cured for 14 days after completion of the HE processes.

Experimental
The roughness tests were carried out using the HOMMEL TESTER T8000 profilometer by Hommelwerke (Villingen-Schwenningen, Germany) and a HOMMEL-ETAMIC TKU300/600 contact head with an apex radius of 2 µm. The head travel speed was 0.5 mm/s. The roughness parameter R a was obtained, i.e., the arithmetic mean value of the ordinates Z(x) of the roughness inside the elementary Section l. This value was calculated by the TurboWave software (V7.4, 2015, Hommelwerke GmbH, Villingen-Schwenningen, Germany) using Equation (2): As part of the research, a series of five measurements was performed using the contact method. The measuring sections were 4.8 mm long. Roughness profiles were obtained for each of the measurements.
The static tensile test (Zwick/Roell Z250, Ulm, Germany) was performed for samples with a diameter of 5 mm and a length of 10 mm at a tensile speed of 1 mm·min −1 to determine the tensile strength σ M in MPa and elongation at break ε M in %. A 1-axis compression test (Zwick/Roell Z250, Ulm, Germany) using samples with the diameter of the extruded rod and the height equal to 1.5 times the diameter was performed with a compression speed of 1 mm min −1 . The compression test was continued until the sample was compressed by 70% of its original height in order to determine the compressive modulus E c in GPa, compressive stress at deformation 0.1% σ 0.1 in MPa, and compressive strength σ M in MPa. A 3-point bending test (Zwick/Roell Z250, Ulm, Germany) was performed using samples with a diameter of 5 mm and a length of 160 mm, with a spacing between supports of 125 mm, a maximum deflection of 69 mm and a bending speed of 2 mm min −1 in order to determine the flexural modulus E f in MPa, flexural strength σ fM in MPa, and the flexural strain at maximum flexural strain ε fM in mm. An impact tensile toughness test was carried out using an Instron Dynatup 9250 HV testing machine (Norwood, MA, USA) on samples with a diameter of 5 mm and a length of 10 mm with an impact energy of 50 J and an impact speed of 1.8 ms −1 in order to determine the Charpy impact toughness K in kJ m 2 . At least 3 samples were used for each mechanical test. The test results of the PA6 rods subjected to the cold hydrostatic extrusion process were compared with the properties of the initial state rods (non-deformed) with a diameter of 15.6 mm.
Structural analysis was conducted using a Scanning Electron Microscopy-ZEISS SEM microscope, model ULTRA PLUS, with a Gemini column (Jena, Germany). The fracture surfaces were analysed on samples frozen for 5 min in liquid nitrogen and broken through bending. Before the observations, the samples were dusted with a layer of gold.
The DSC analysis of PA6 was performed using a differential scanning calorimeter DSC Q1000 (TA Instruments, New Castle, DE, USA). Samples (ok. 5 mg) were closed in hermetic aluminium cups and heated at 10 • C/min in the temperature range 0 to 250 • C.
Moisture content w was determined by Equation (3): where: m and m 1 are the mass of the weighing bottle with the material sample before and after drying, respectively, and m 2 is the mass of the investigated material sample. The density (ρ) test was conducted using hydrostatic weighing, on samples of the same dimensions, in accordance with ISO 2781.

Hydrostatic Extrusion of Polyamide PA6 with Quality of Surface Analysing
During the HE processes, the change in process pressure as a function of time was registered. Example pressure characteristics of the hydrostatic extrusion process for three degrees of plastic deformation ε are summarised in Figure 2.
flexural strain at maximum flexural strain εfM in mm. An impact tensile toughness test was carried out using an Instron Dynatup 9250 HV testing machine (Norwood, MA, USA) on samples with a diameter of 5 mm and a length of 10 mm with an impact energy of 50 J and an impact speed of 1.8 ms −1 in order to determine the Charpy impact toughness K in kJ m 2 . At least 3 samples were used for each mechanical test. The test results of the PA6 rods subjected to the cold hydrostatic extrusion process were compared with the properties of the initial state rods (non-deformed) with a diameter of 15.6 mm.
Structural analysis was conducted using a Scanning Electron Microscopy-ZEISS SEM microscope, model ULTRA PLUS, with a Gemini column (Jena, Germany). The fracture surfaces were analysed on samples frozen for 5 min in liquid nitrogen and broken through bending. Before the observations, the samples were dusted with a layer of gold.
The DSC analysis of PA6 was performed using a differential scanning calorimeter DSC Q1000 (TA Instruments, New Castle, DE, USA). Samples (ok. 5 mg) were closed in hermetic aluminium cups and heated at 10 °C/min in the temperature range 0 to 250 °C.
Moisture content was determined by Equation (3) where: and are the mass of the weighing bottle with the material sample before and after drying, respectively, and is the mass of the investigated material sample. The density (ρ) test was conducted using hydrostatic weighing, on samples of the same dimensions, in accordance with ISO 2781.

Hydrostatic Extrusion of Polyamide PA6 with Quality of Surface Analysing
During the HE processes, the change in process pressure as a function of time was registered. Example pressure characteristics of the hydrostatic extrusion process for three degrees of plastic deformation ε are summarised in Figure 2. At the beginning of the extrusion process, pressure jumps ('breakthrough pressure') are observed (Figure 2), which are the result of changing the static (higher) friction to kinetic (lower) friction. The increase in pHE caused by the increase in ε is a well-known phenomenon characteristic of metals and metal alloys, polymers, composites, etc. The increase in pHE during deformation with ε > 1.25 ( Figure 2) is caused by an increase in the At the beginning of the extrusion process, pressure jumps ('breakthrough pressure') are observed (Figure 2), which are the result of changing the static (higher) friction to kinetic (lower) friction. The increase in p HE caused by the increase in ε is a well-known phenomenon characteristic of metals and metal alloys, polymers, composites, etc. The increase in p HE during deformation with ε > 1.25 ( Figure 2) is caused by an increase in the packing density of the polymer macromolecules. The pressure then stabilises. The flat runs of the pressure change curve during the process indicate a constant, linear rate of extrusion of the product, which is a necessary condition for obtaining a homogeneous structure along the entire length of the rod and the resulting repetitive physical and mechanical properties [40]. The constant rate of extrusion determined on the basis of the flat courses of the pressure change curves was recorded for samples tested at a true strain of 1.25 and 1.57.
However, at a true strain ε = 1.9, the pressure was not stable and the product after the HE process was inconsistent ( Figure 3). packing density of the polymer macromolecules. The pressure then stabilises. The flat runs of the pressure change curve during the process indicate a constant, linear rate of extrusion of the product, which is a necessary condition for obtaining a homogeneous structure along the entire length of the rod and the resulting repetitive physical and mechanical properties [40]. The constant rate of extrusion determined on the basis of the flat courses of the pressure change curves was recorded for samples tested at a true strain of 1.25 and 1.57.
However, at a true strain ε = 1.9, the pressure was not stable and the product after the HE process was inconsistent ( Figure 3). Lack of pressure stability during the HE process for sample E_1.90_7 is caused by high pHE, causing the Tad temperature to rise to 211 °C. In order to explain the reasons for the occurrence of defects in the profile of sample E_1.90_7, a DSC analysis of sample A-PA6 was performed before the HE process ( Figure 4). Based on the course of the (DSC) curve, the glass transition temperature (Tg) was determined, which was 35 °C. The analysed HE processes of the PA6 polymer were performed at adiabatic temperature in the range of Tad = 40-211 °C, so at a temperature above the Tg level. The second transformation in the DSC thermal curve is the endothermic transformation of the polymer crystalline phase melting. This transformation is described by the initial melting temperature (Tonset = 208 °C), the melting point temperature Tm = 221 °C and the melting enthalpy ΔHm = 67 J/g. The melting process begins at around 166 °C (Ts). Since the Tad of the sample E_1.90_7 subjected to the HE process is higher than the melting process beginning, some of the crystallites melt during the HE process. The consequence Lack of pressure stability during the HE process for sample E_1.90_7 is caused by high p HE , causing the T ad temperature to rise to 211 • C. In order to explain the reasons for the occurrence of defects in the profile of sample E_1.90_7, a DSC analysis of sample A-PA6 was performed before the HE process ( Figure 4).
packing density of the polymer macromolecules. The pressure then stabilises. The flat runs of the pressure change curve during the process indicate a constant, linear rate of extrusion of the product, which is a necessary condition for obtaining a homogeneous structure along the entire length of the rod and the resulting repetitive physical and mechanical properties [40]. The constant rate of extrusion determined on the basis of the flat courses of the pressure change curves was recorded for samples tested at a true strain of 1.25 and 1.57.
However, at a true strain ε = 1.9, the pressure was not stable and the product after the HE process was inconsistent (Figure 3). Lack of pressure stability during the HE process for sample E_1.90_7 is caused by high pHE, causing the Tad temperature to rise to 211 °C. In order to explain the reasons for the occurrence of defects in the profile of sample E_1.90_7, a DSC analysis of sample A-PA6 was performed before the HE process ( Figure 4). Based on the course of the (DSC) curve, the glass transition temperature (Tg) was determined, which was 35 °C. The analysed HE processes of the PA6 polymer were performed at adiabatic temperature in the range of Tad = 40-211 °C, so at a temperature above the Tg level. The second transformation in the DSC thermal curve is the endothermic transformation of the polymer crystalline phase melting. This transformation is described by the initial melting temperature (Tonset = 208 °C), the melting point temperature Tm = 221 °C and the melting enthalpy ΔHm = 67 J/g. The melting process begins at around 166 °C (Ts). Since the Tad of the sample E_1.90_7 subjected to the HE process is higher than the melting process beginning, some of the crystallites melt during the HE process. The consequence Based on the course of the (DSC) curve, the glass transition temperature (T g ) was determined, which was 35 • C. The analysed HE processes of the PA6 polymer were performed at adiabatic temperature in the range of T ad = 40-211 • C, so at a temperature above the T g level. The second transformation in the DSC thermal curve is the endothermic transformation of the polymer crystalline phase melting. This transformation is described by the initial melting temperature (T onset = 208 • C), the melting point temperature T m = 221 • C and the melting enthalpy ∆H m = 67 J/g. The melting process begins at around 166 • C (T s ). Since the T ad of the sample E_1.90_7 subjected to the HE process is higher than the melting process beginning, some of the crystallites melt during the HE process. The consequence of this is the formation of a porous and rough surface on the rod after the HE process. When the HE process pressure is stable, the rods are characterised by a smooth surface with low roughness along the entire length (R and = 0.82 µm) as is shown in Figure 5a for the sample D_1.57_5. For the sample E_1.90_7 (Figure 5b) with ε = 1.9, when the pressure during the HE process was unstable, the surface of this sample had a much greater roughness (R and = 2.68 µm), and spiral cracks were noticed on the extruded rod. of this is the formation of a porous and rough surface on the rod after the HE process. When the HE process pressure is stable, the rods are characterised by a smooth surface with low roughness along the entire length (Rand = 0.82 μm) as is shown in Figure 5a for the sample D_1.57_5. For the sample E_1.90_7 (Figure 5b) with ε = 1.9, when the pressure during the HE process was unstable, the surface of this sample had a much greater roughness (Rand = 2.68 μm), and spiral cracks were noticed on the extruded rod. For the same values of plastic deformation with increasing extrusion rate (v) an increased pressure during the HE process was observed ( Figure 6). The effect of the pressure increases when the HE processes speed is increased is similar, regardless of the strain value. The pressure change as a function of the strain was also analysed for the same cold HE process speed v = 5 mm s −1 (Figure 7). It was found that increasing the degree of plastic strain results in the occurrence of a significantly higher pressure during the plastic deformation process.  For the same values of plastic deformation with increasing extrusion rate (v) an increased pressure during the HE process was observed ( Figure 6). The effect of the pressure increases when the HE processes speed is increased is similar, regardless of the strain value. The pressure change as a function of the strain was also analysed for the same cold HE process speed v = 5 mm s −1 (Figure 7). It was found that increasing the degree of plastic strain results in the occurrence of a significantly higher pressure during the plastic deformation process. of this is the formation of a porous and rough surface on the rod after the HE process. When the HE process pressure is stable, the rods are characterised by a smooth surface with low roughness along the entire length (Rand = 0.82 μm) as is shown in Figure 5a for the sample D_1.57_5. For the sample E_1.90_7 (Figure 5b) with ε = 1.9, when the pressure during the HE process was unstable, the surface of this sample had a much greater roughness (Rand = 2.68 μm), and spiral cracks were noticed on the extruded rod. For the same values of plastic deformation with increasing extrusion rate (v) an increased pressure during the HE process was observed ( Figure 6). The effect of the pressure increases when the HE processes speed is increased is similar, regardless of the strain value. The pressure change as a function of the strain was also analysed for the same cold HE process speed v = 5 mm s −1 (Figure 7). It was found that increasing the degree of plastic strain results in the occurrence of a significantly higher pressure during the plastic deformation process.  Plastic deformation of the PA6 polymer with true strain ε~0.7 increases the process pressure to approx. 100 MPa, which is accompanied by an increase in the adiabatic temperature T ad to about 70 • C., so at the true strain value ε~1.6 the pressure reaches the value of approx. 320 MPa, and the T ad of the material is 153 • C. To achieve such a high T ad temperature, it is necessary to supply much more heat than in the case of ε~0.7. The level of heat supplied depends on the specific heat of the material. The specific heat is the heat needed to change body temperature by one unit per unit mass of the material: where: c is the specific heat (J/kg K), ∆Q-heat delivered (J), m-material mass (kg), and ∆T temperature difference (K). Plastic deformation of the PA6 polymer with true strain ε ~ 0.7 increases the process pressure to approx. 100 MPa, which is accompanied by an increase in the adiabatic temperature Tad to about 70 °C., so at the true strain value ε ~ 1.6 the pressure reaches the value of approx. 320 MPa, and the Tad of the material is 153 °C. To achieve such a high Tad temperature, it is necessary to supply much more heat than in the case of ε ~ 0.7. The level of heat supplied depends on the specific heat of the material. The specific heat is the heat needed to change body temperature by one unit per unit mass of the material: where: is the specific heat (J/kg K), ∆ -heat delivered (J), -material mass (kg), and ∆ temperature difference (K).
The specific heat of the PA6 polymer amounts to 1700 J/kg K and is over 4 times higher than the specific heat of Cu (385 J/kg K). This means that in order to increase its adiabatic temperature during the HE processes, it is necessary to provide much more thermal energy for the same volume of PA6 polymer than in the case of Cu. To generate such a large amount of thermal energy in the material for the extrusion process, a much greater extrusion pressure must be used. Moreover, the PA6 polymer has a low thermal conductivity λ = 0.24-0.28 W/m K, i.e., the ability to conduct heat, so under the same conditions much less heat is removed from it than, for example, from Cu (400 W/m K), which is a material of over 1500 times greater thermal conductivity. Therefore, the PA6 polymer cools down slowly, holding its temperature much longer, which causes the heat to act on the material for a longer time, promoting changes in its structure and, consequently, its properties.

Mechanical Properties
Changes in the mechanical properties of the PA6 polymer after the hydrostatic extrusion process, and measured statistic tensile tests, are shown in Figures 8 and 9. The specific heat of the PA6 polymer amounts to 1700 J/kg K and is over 4 times higher than the specific heat of Cu (385 J/kg K). This means that in order to increase its adiabatic temperature during the HE processes, it is necessary to provide much more thermal energy for the same volume of PA6 polymer than in the case of Cu. To generate such a large amount of thermal energy in the material for the extrusion process, a much greater extrusion pressure must be used. Moreover, the PA6 polymer has a low thermal conductivity λ = 0.24-0.28 W/m K, i.e., the ability to conduct heat, so under the same conditions much less heat is removed from it than, for example, from Cu (400 W/m K), which is a material of over 1500 times greater thermal conductivity. Therefore, the PA6 polymer cools down slowly, holding its temperature much longer, which causes the heat to act on the material for a longer time, promoting changes in its structure and, consequently, its properties.

Mechanical Properties
Changes in the mechanical properties of the PA6 polymer after the hydrostatic extrusion process, and measured statistic tensile tests, are shown in Figures 8 and 9.    The percentage changes in the mechanical parameters are also summarised in Table  2. In the analysed cold HE processes, it was observed that the tensile modulus (Et) and tensile strength (σM) of materials increases with the degree of true strain. At maximum true strain value ε = 1.57 an over 10-fold increase in the tensile modulus and an almost 5- The percentage changes in the mechanical parameters are also summarised in Table 2. In the analysed cold HE processes, it was observed that the tensile modulus (E t ) and tensile strength (σ M ) of materials increases with the degree of true strain. At maximum true strain value ε = 1.57 an over 10-fold increase in the tensile modulus and an almost 5-fold increase in tensile strength was observed ( Table 2). As the degree of strain degree increases, the value of strain at break decreases to the level of ε M~3 80% for the true strain ε = 1.57. Figure 9 shows a significant increase in the strength of PA6 with the degree of true strain during the HE process, while decreasing elongation. The higher increase in strength of PA6 polymer indicates that the higher true strain during HE cold deformation is a key factor to promote the increase in crystallinity degree and homogeneity of the oriented structure, which will be discussed in the next article.
It was also found that the tensile modulus, tensile strength and strain at break at the same strain value do not depend on the HE process linear speed extrusion, see Figure 10a,b.
Analysis of the mechanical properties determined in the compression test, bending test and impact strength test of the PA6 polymer as a function of the degrees of actual strain applied are given in Table 3. ε = 1.57. Figure 9 shows a significant increase in the strength of PA6 with the degree of true strain during the HE process, while decreasing elongation. The higher increase in strength of PA6 polymer indicates that the higher true strain during HE cold deformation is a key factor to promote the increase in crystallinity degree and homogeneity of the oriented structure, which will be discussed in the next article. It was also found that the tensile modulus, tensile strength and strain at break at the same strain value do not depend on the HE process linear speed extrusion, see Figure  10a,b. Analysis of the mechanical properties determined in the compression test, bending test and impact strength test of the PA6 polymer as a function of the degrees of actual strain applied are given in Table 3. With an increase in the degree of true strain during the HE processes, an increase in all measured mechanical properties is observed. However, compared to the initial state material (A), the values obtained in the compression test, i.e., the compressive modulus (Ec) and the compressive strength (σM), are lower over the entire range of strain values. However, the improvement in the mechanical properties is visible in the bending and impact tests. The increase in these features is not as significant as in the case of the features determined in the tensile test.

Comparison of Selected Mechanical Properties of PA6 Processed Using Various Methods
Polyamide 6 can be processed with various techniques, including: selective laser sintering (SLS) [41], twin-screw extrusion (TSE) and cold-drawing (DR)-TSE DR [42], Equal-Channel Multiangular Extrusion (ECAP) [43], solid-phase extrusion (CE) [44], injection With an increase in the degree of true strain during the HE processes, an increase in all measured mechanical properties is observed. However, compared to the initial state material (A), the values obtained in the compression test, i.e., the compressive modulus (E c ) and the compressive strength (σ M ), are lower over the entire range of strain values. However, the improvement in the mechanical properties is visible in the bending and impact tests. The increase in these features is not as significant as in the case of the features determined in the tensile test.

Comparison of Selected Mechanical Properties of PA6 Processed Using Various Methods
Polyamide 6 can be processed with various techniques, including: selective laser sintering (SLS) [41], twin-screw extrusion (TSE) and cold-drawing (DR)-TSE DR [42], Equal-Channel Multiangular Extrusion (ECAP) [43], solid-phase extrusion (CE) [44], injection moulding (IM) [45,46] and twin-screw extruder + injection (TSE + IM) [47,48]. Figures 10 and 11 show a comparison of selected properties of PA6 samples formed using different processing methods. moulding (IM) [45,46] and twin-screw extruder + injection (TSE + IM) [47,48]. Figures 10  and 11 show a comparison of selected properties of PA6 samples formed using different processing methods. The tensile modulus of the PA6 samples after the HE process is similar to the modulus of the polymer processed using the SLS technique and is almost seven times lower than the modulus of samples processed with the IM technique (Figure 11), whereas the tensile strength is the highest in samples subjected to the HE process-six times higher than in samples processed with the IM technique. Samples subjected to the HE process The tensile modulus of the PA6 samples after the HE process is similar to the modulus of the polymer processed using the SLS technique and is almost seven times lower than the modulus of samples processed with the IM technique (Figure 11), whereas the tensile strength is the highest in samples subjected to the HE process-six times higher than in samples processed with the IM technique. Samples subjected to the HE process have the highest elongation at break compared to the samples subjected to other processes to which they were compared.
For most of the compared processing methods it was necessary to use a temperature above the PA6 melting point, which is >220 • C (IM, TSE, TSE + DR, TSE + IM, or slightly lower, i.e.,~195 • C (SLS, CE), or 150 • C (ECAP). During the HE processes, the PA6 material remained solid and was not additionally heated. The changes that took place in the material were caused by the heat generated by high pressure, i.e., the mechanical work of extrusion converted into heat. The compressive modulus of the PA6 samples subjected to the HE process is more than two times greater, whereas the compressive strength is more than three times greater than in the polymer samples subjected to the CE process [49]. As with the properties in the static tensile test, these changes are the result of changes in the structure of the materials ( Figure 12).   (Figure 13b) have characteristic brittle fractures, typical of samples with a high degree of crystallinity and low impact strength [50]. The fracture of the samples made of initial state material (sample A) during the TSE process structure of the PA6 is oriented. The consequence of such formation of the structure in samples A is the formation of the stepped structure of the fracture. In the samples of the material subjected to HE process with a true strain ε = 0.68, the structure undergoes a distinctive change compared to the material subjected to the TSE process. The structural elements of this sample are finer and more regularly distributed, whereas samples of the material subjected to HE process with a true strain ε = 1.57 show a fibrous material structure (Figure 13c).  Fractures of significantly different structure were observed. The initial state material samples ( Figure 13a) and samples of material subjected to the HE process with ε = 0.68 (Figure 13b) have characteristic brittle fractures, typical of samples with a high degree of crystallinity and low impact strength [50]. The fracture of the samples made of initial state material (sample A) during the TSE process structure of the PA6 is oriented. The consequence of such formation of the structure in samples A is the formation of the stepped structure of the fracture. In the samples of the material subjected to HE process with a true strain ε = 0.68, the structure undergoes a distinctive change compared to the material subjected to the TSE process. The structural elements of this sample are finer and more regularly distributed, whereas samples of the material subjected to HE process with a true strain ε = 1.57 show a fibrous material structure (Figure 13c). such formation of the structure in samples A is the formation of the stepped structure of the fracture. In the samples of the material subjected to HE process with a true strain ε = 0.68, the structure undergoes a distinctive change compared to the material subjected to the TSE process. The structural elements of this sample are finer and more regularly distributed, whereas samples of the material subjected to HE process with a true strain ε = 1.57 show a fibrous material structure (Figure 13c). Fibres arranged in layers in the direction of extrusion are shown in Figure 13c,d. During the hydrostatic extrusion process, the PA6 macromolecules are drawn in the extruding direction. The distribution of fibres within the entire PA6 polyamide rod diameter is even, but a solid layer formed by fibres was found around the rod circumference (Figure 13d). This may be due to the shear forces occurring during the HE processes, which increase as the die angle increases. The shear force was greater at the circumference zone of the rod, which, together with the strongest adiabatic heating effect (AH) in the die/polymer subsurface layer, led to a strong coalescence of molecular chains and the structure presented (Figure 13d). The layer on the circumference of the PA6 polyamide rod becomes stiffer than the rod core and hence the rod breaks in a characteristic manner (Figure 13d).

Structural Investigations
The initial analyses of PA6's structural changes before and after the HE process are described in Supplementary Materials. Based on these studies, it was found that the reason for the changes in strength properties is the increase in the degree of crystallinity and the increase in physical cross-linking with hydrogen bonds connecting PA6 chains after hydrostatic extrusion. ( Figure S1, Table S1, Figure S2, Table S2 from Supplementary Material).

Conclusions
As part of the study, the influence of variable parameters of the hydrostatic extrusion process on the mechanical properties of PA6 was analysed. The rods subjected to the HE process were strained with variable true strain values in the range of 0.68 to 1.9. The extrusion process was accompanied by adiabatic heating effects increasing the temperature during the process to values exceeding the melting point at the true strain ε = 1.9. The Fibres arranged in layers in the direction of extrusion are shown in Figure 13c,d. During the hydrostatic extrusion process, the PA6 macromolecules are drawn in the extruding direction. The distribution of fibres within the entire PA6 polyamide rod diameter is even, but a solid layer formed by fibres was found around the rod circumference (Figure 13d). This may be due to the shear forces occurring during the HE processes, which increase as the die angle increases. The shear force was greater at the circumference zone of the rod, which, together with the strongest adiabatic heating effect (AH) in the die/polymer subsurface layer, led to a strong coalescence of molecular chains and the structure presented (Figure 13d). The layer on the circumference of the PA6 polyamide rod becomes stiffer than the rod core and hence the rod breaks in a characteristic manner (Figure 13d).
The initial analyses of PA6's structural changes before and after the HE process are described in Supplementary Materials. Based on these studies, it was found that the reason for the changes in strength properties is the increase in the degree of crystallinity and the increase in physical cross-linking with hydrogen bonds connecting PA6 chains after hydrostatic extrusion. (Figures S1 and S2, Tables S1 and S2 from Supplementary Material).

Conclusions
As part of the study, the influence of variable parameters of the hydrostatic extrusion process on the mechanical properties of PA6 was analysed. The rods subjected to the HE process were strained with variable true strain values in the range of 0.68 to 1.9. The extrusion process was accompanied by adiabatic heating effects increasing the temperature during the process to values exceeding the melting point at the true strain ε = 1.9. The process performed under these conditions does not allow a regular-shaped bar with low roughness to be obtained, as part of the PA6 crystalline phase melts during its course. Based on the analyses conducted, it can be concluded that the solid-phase cold hydrostatic extrusion process should be carried out under conditions generating T ad higher than the T g of the PA6 material and preferably below the temperature at which the crystalline phase begins to melt (T s ). Such conditions occurred in the strain range of 0.68 to 1.57.
As the true strain value increases, material with a greater tensile modulus, tensile strength, flexural modulus and flexural strength is obtained. It also has a lower deformation at break. PA6 subjected to the HE process with a true strain ε = 1.57, compared to the material in its initial state, i.e., after the twin-screw extrusion process Sample A, it has a tensible modulus that is ten times higher, tensile strength that is almost five times higher, and elongation at break that is 35 times lower. In the case of the properties determined during the compression and bending tests, the changes are not as significant. The modulus marked during the bending test increases by 28%. A decrease in compressive strength is, however, observed.
As a result of the analyses conducted, it was found that for a given strain, the process pressure increases with the increase in the linear extrusion rate, whereas the mechanical properties in the entire range of the strain tested as a function of the strain rate do not change significantly.
The changes in the properties of the PA6 polymer subjected to the HE process are the result of structural changes. At a true strain ε = 0.68, a more regular structure is formed in the material than in the samples made of the initial state material (A). In contrast, a higher true strain ε = 1.57 in the HE process causes the structural elements to be formed into fibres that are connected and arranged along the extrusion direction. Macroscopic images of broken rods show that their cross-section structure is heterogeneous.
In further work on PA6 subjected to solid state HE processes, analyses of the structure of materials using various techniques will be performed to explain the mechanism of its changes during this process. In addition, the heterogeneity of the structure on the cross-section of the extruded rods will be evaluated.

Supplementary Materials:
The following are available online at https://www.mdpi.com/article/10 .3390/ma14206045/s1, Figure S1: DSC thermograms of PA6 samples before the HE process (PA6_A) and after the HE process (PA6_D), Figure S2: Results of the DMA analysis of samples PA6_A and PA6_D, Figure S3. Results of dynamic viscosity of PA6 samples using a rheometer; Table S1: Results of DSC thermograms analysis of tested samples, Table S2: DMA analysis results.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.