Modification of the mechanical properties of polyamide 6 multifilaments in high-speed melt spinning with nano silicates

Abstract

In the last few years research activities have been focused on the modification of fiber properties with nano-scaled additives. One of the most important fields of research is the alteration of mechanical properties such as the tenacity and the specific breaking load. In this study, we determined the influence of nano-phyllosilicates on the drawability of polyamide 6 multifilament yarns. It was first demonstrated that the drawability of the fibers drastically increased in an industrially relevant high-speed melt spinning process. Structural properties of the material are identified by wide-angle X-ray diffraction (WAXD) and differential scanning calorimetry (DSC). Changes in the crystalline properties as well as in the alignment of the silicates are compared with the stress–strain curves of the fibers, and a molecular mechanism for the drawing process is derived from these experiments. In a first step, a significant phase transition in the crystalline structure unaffected by the silicates occurs for low draw ratios (DRs). Beyond this point, where unmodified fibers start to break, a gliding between the silicate layers takes place, which is responsible for an extended drawability of the fibers. This mechanism leads to new possibilities for fiber processing, which can be used to research ultra-fine filaments in future studies.

Keywords

melt spinning multifilament yarn nanocomposites polyamide 6 structure analysis

Polyamide 6 (PA6), also called Polycaprolactam or Perlon as a trademark, is a polymer of the group of polyamides. It was first synthesized by means of ring-opening polymerization reactions from ε-aminocaprolactam in 1938.¹ The ‘6’ in the name denotes the number of carbon atoms in the repeating base molecule. The resulting sequence in the polymer chain is $- [HN - ({CH}_{2})_{5} - CO]_{n} -$ .²

The polymer is a polymorphic, semi-crystalline material. Three different thermodynamic, metastable phases can coexist in the glassy or rubbery solid state.^3,4 The so-called γ phase is crystalline with a monoclinic unit cell. The lattice parameters are known as a = 9.33 Å and c = 4.78 Å perpendicular to the polymer chains as well as b = 16.88 Å in the polymer chain direction from wide-angle X-ray diffraction (WAXD) experiments. The value of the angle β between a and c axis is 121°.⁵ Since γ is around 120°, a pseudohexagonal version of this unit cell can be calculated and is often used in theory for the structural description.⁴ The second crystalline phase denoted as the α phase also has a monoclinic cell with lattice parameters a = 9.56 Å, b = 17.24 Å, and c = 8.01 Å with an angle β = 67.5°.⁶ As distinguished from the α and γ phase, the β phase as the third stable phase is characterized by a mesomorphic structure. The short-range order exceeds the typical values for an amorphous material. Since no long-range order exists in this phase, no detailed structure can be obtained from diffraction experiments. However, a molecular model, based on a nano-crystalline, hexagonal phase with a = 4.8 Å and c = 8.6 Å was proposed. The side groups in this structure model are randomly oriented.⁷

Characterization and phase transitions

Information on the properties of all phases is predominantly gathered from diffraction experiments. Since all reflections from the lattice planes perpendicular to the polymer chains appear in the same angular range, peak fitting is carried out for an identification of the phases.^3,4 Both γ and α form crystals appearing together with the β phase if a sample is cooled from the melt. The α formation is preferred at low cooling rates.^3,4,8 Pure β phase material can be obtained by quenching.^3,7

Thermal analysis as well as temperature-dependent diffraction experiments reveal information about phase transitions. The γ crystallites arise from the β phase at temperatures higher than 130 °C,³ whereas γ is transformed to α phase crystallites in the temperature regime of the melting point, which is located at 220 °C.⁹ In contrast, examples of unstable γ crystals, which transform far below the melting point,¹⁰^–¹² have been reported. Also β phase crystals are able to transform to α phase. Therefore, annealing at 190 °C is necessary.¹³ Since polymer chains are more stretched in the γ phase, these chains’ formation is also preferred in strain-induced crystallization under mechanical stress with the result that this phase is preferred after the drawing processes.⁸ The γ phase is gradually transformed to the α phase during annealing at temperatures higher than 160 °C.³

Melt spinning

Melt spinning is the classic method of fiber production of thermoplastic materials.² PA6 is one of the most common fiber polymers, where the processing of monofilaments and multifilament yarn has a long tradition. The material can be melt-spun as low oriented yarn (LOY) and partially oriented yarn (POY), whereas high-spinning speeds can be achieved. The material can be converted to fully drawn yarn (FDY) in a separate process as well as to inline drawn yarns (IDY) in one step by melt spinning.^2,14

The fiber formation process of PA6 has been widely examined with theoretical calculations and online measurements of fiber properties. In this case, the fiber diameter and velocity, mechanical forces, filament temperature, and development of molecular orientation have been investigated.¹⁴^–¹⁶ Other studies deal with the structure formation in the melt spinning process. The development of crystal phases was examined with in-situ measurements of diffraction patterns, which showed the qualitative development of the three phases.¹⁷^–²⁰

Nanocomposites

Nanocomposites of PA6 were characterized in many studies. Several types of particles at the nanoscale, including carbon nanotubes, carbon nanofibers, silicon dioxide, and many silicate-based particles, were doped in the polymer matrix.²¹^–⁴⁹ The composites are either produced by in-situ processing during polymerization or by compounding a nano-powder in the liquid state with a twin screw extruder.²²^–²⁹

Much research has focused on the modification of polyamide with layered silicates (or phyllosilicates).²²^–⁴⁹ These sheet-like structures are known for the improvement of fire retardancy, the modification of mechanical and electrical properties, and influencing the sorption.^27,30^–³⁴ The influence on the polymer structure and its physical properties were examined by WAXD, differential scanning calorimetry (DSC), transmission electron microscopy (TEM), nuclear magnetic resonance (NMR), Fourier-transform infrared spectroscopy (FTIR), and viscosimetry. In these experiments, the silicate concentration, the molecular weight of the polymer, and the silicate type were varied.^4,28,32,34^–⁴⁵ The presence of the silicates seems to affect the crystallization behavior, the occurring crystal phases, and the crystallite sizes.^28,35^–⁴² Further structural changes occur in the annealing process.⁴⁶ In addition, the silicates are supposed to support the formation of a rigid amorphous phase (RAF).⁴⁷

The use of silicates in fibers was mostly examined in electrospun nanofibers, where fiber diameter is in the same range as the silicate dimensions. The alignment of the particles was observed with TEM.⁴⁸^–⁵⁰ Only a limited amount of research projects on the effect of nano-silicates on the properties of filament yarns have been carried out.^8,28,51 One group detected changes in the crystal geometry.⁸ Another study deals with the fiber properties of yarns spun at different take-up velocities with different silicate concentrations. The presence of silicates leads to an increase in the density and the crystallinity, where mechanical properties mostly remain unchanged.⁵¹ Another group worked on the structural changes in the fibers, where nanocomposites with exfoliated silicate sheets were obtained by in-situ preparation. The monofilaments were spun at very low take-up velocities and showed a reduced drawability with higher clay content.²⁷

Aim of this study

Current research activities at the Institut für Textiltechnik (ITA) focus on the processing of new polymer materials to fibers as well as the modification of standard polymeric materials with dopants from organic and inorganic materials. These activities aim at the functionalization of fibers with new properties and the enhancement of fiber properties. In this study, we examined the influence of melt-compounded nano-phyllosilicates (or layered silicates) on the mechanical properties and drawability of PA6 multifilaments in a high-speed melt spinning process with industrially relevant winding speeds. Various experiments have been carried out to determine the influence of these particles and to understand the changes at a molecular scale. According to the crystallographic and thermal properties, we present a model for the extended drawability of the multifilaments.

Experimental details

Material

For the melt spinning of the multifilaments, a standard polymer B24N03 from BASF AG, Germany was used, which is suitable for high-speed spinning. According to the datasheet of the material, the melting temperature is 220 °C and the density at room temperature is about 1.14 g/cm³.⁵²

Silicate-modified compounds of the material are produced with the help of a twin screw extruder. Two different master batches with a total mass fraction of 2% and 1% of phyllosilicates are created. Commercially available phyllosilicates, Nanofil 5 from Rockwood Clay Products, Inc., USA, are used as nanoclays.⁵³ These silicates are based on a natural bentonite the most important component of which is montmorillonite. The surface of the silicate is treated by a dimethyl, di(hydrogenated tallow)alkyl ammonium salt.⁵³ As a consequence, Na⁺ ions are exchanged with an organic component, allowing for a better embedding in the polymer matrix. The typical distance between two layers is about 2.8 nm.⁵³

Melt spinning

Melt spinning is carried out on the spinning plant ITA Plus, which allows a broad variation of many spinning parameters at an industrial scale. The polymer in granule form is being transported into a single-screw extruder from Oerlikon Barmag GmbH & Co. KG, Germany, where the polymer is compressed and melted through friction and increasing pressure among the granules. The melt exits the extruder at pressure levels around 100 bar and is transported through heated pipes to the spin pump. Up to the spin pump the process and throughput is pressure controlled depending on the lead pressure at the extruder head. From the spin pump is a throughput-controlled process, where the mass flow rate Q is adjusted to 44.0 g/min. The polymer is pumped into the spin pack, which consists of a spinneret and multiple filtration layers. These can be woven metal structures, ceramic, or metallic nonwovens, and also stainless steel sand is commonly used in melt spinning. After passing through all filtration stages, the melt flows into the capillaries of the spinneret. For the melt spinning of PA6, a spinning plate with a number of n = 24 capillaries is chosen. The geometry of the capillaries is defined by their diameter (D = 250 µm) and their length (L = 500 µm). Together with the mass flow rate Q and the density of the polymer ρ, the geometry also defines the extrusion velocity v_e, where A is the total area of capillaries:
$v_{e} = \frac{Q}{ρ \cdot A} = \frac{Q}{ρ \cdot n \cdot π (\frac{D}{2})^{2}}$
(1)

At the end of the capillary, the melt exits and forms a fiber. The temperature of the extrusion equipment and thus of the melt is the major process parameter determining the viscosity and shearing of the polymer. The temperature in the extruder is chosen as 265°C and slightly increased in the heated pipes for lowering the viscosity. The extrusion temperature is chosen as 265°C. Over a length of 1500 mm the filaments are cooled by a laminar, conditioned air stream at 20°C with a velocity of 0.35 m/s perpendicular to the fiber axis and 75% relative humidity. In this step, the solidification and crystallization of the fibers take place. The as-spun fibers are drawn down vertically usually by at least one draw down godet rotating at a surface velocity v_G. Thus, the draw down ratio (DDR) is fixed as the ratio between godet velocity v_G and extrusion velocity v_e:
$DDR = \frac{v_{G}}{v_{e}}$
(2)

Depending on the use of the filaments, a wide variety of drawing states can be realized inline before winding the filaments on bobbins. The solid-state drawing of the fibers is conducted between two godets, where the consecutive godet revolves faster than the first. In this study the spinning is realized with inline drawing using a variation of the drawing ratio between 1.0185 and 1.8, whereas the drawing temperature is fixed. The filaments are winded on bobbins with a winder from Oerlikon Barmag GmbH & Co. KG, Germany, whereas the winding speed is fixed at 5000 m/min. The parameters used for melt spinning and inline drawing of the multifilaments are summarized in Table 1.
Table 1.
Process parameters and variations for melt-spinning of polyamide 6

Parameter Values

Capillary geometry (number of filaments × capillary diameter × capillary length) 24 × 0.25 mm × 0.5 mm

Extrusion temperature 265°C

Mass flow rate 44.0 g/min

Extrusion velocity 25.7 m/min

Draw down ratio 191.0/159.3/134.0/117.7/107.9

Inline draw ratio 1.0185/1.220/1.450/1.650/1.800

Godet temperature (godet 1/godet 2) 40°C/160°C

Winding speed 5000 m/min

Mass fraction phyllosilicates 0%/1%/2%

Quench air (velocity/ temperature/humidity) 0.35 m/s/20°C/75 %

WAXD

Diffraction experiments are carried out on a single-crystal diffractometer STOE & Cie IPDS II equipped with an image plate for digital readout. Two independent goniometer circles allow the exact positioning of the sample for detecting the reflections of the main lattice planes from the PA6 crystal structures. A special specimen holder guarantees the parallel alignment of the fibers and a reproducible measurement of the orientation distribution of crystallites in the fiber. Molybdenum K_α radiation with a wavelength of λ = 0.71073 Å was chosen for the experiments. The experimental setup and the specimen holder are displayed in Figure 1.
Figure 1.
Experimental setup (left) and specimen holder (right).

WAXD is used to detect changes in the lattice parameters, the crystallite dimensions and the orientation distribution due to the phyllosilicates. Further analysis of the diffraction patterns is done with the X-ray Analysis Tool developed at ITA. This tool allows the parameterization of the recorded two-dimensional pixel graphics in directions 2θ (diffraction angle) and φ (azimuthal angle). The diffraction angle is calculated from the radial position r’ with the help of the known detector distance d = 200 mm:
$2 θ = arctan (\frac{r^{'}}{d})$
(3)

With the help of this data, the independent integration of different regions in the diffraction pattern in both directions 2θ and φ is possible. Numerical calculations and peak fits on the resulting intensity diagrams are carried out with the OriginPro software from OriginLab, USA.

DSC

DSC measurements are carried out on a Mettler Toledo DSC 1 at the ITA polymer laboratory, which is equipped with a FRS5 sensor having 56 thermocouples for precise heat flow measurements. For all experiments, a heating rate of 10°C/min and a nitrogen atmosphere was chosen. The fibers were cut into pieces with a total sample weight of 10 mg. The samples were placed into 20 µl aluminum crucibles and pressed to the bottom of the crucible with a lid for good heat contact. An automatic sample changer was used to place the crucibles on the DSC sensor for reproducible measurements. Melting enthalpy is calculated by using the received thermograms. The melting temperature between fibers in the first and quiescent crystallization in the second heating process is also determined. Furthermore, the glass transition and corresponding relaxation processes are examined. The results are used for the identification of changes in the crystal phases α and γ as well as for the determination of the total crystallinity.

Mechanical properties

A Statimat 4U from Textechno GmbH, Germany was used to determine the stress–strain curve of all materials with a clamping length of 100 mm and an extension velocity of 100 mm/min under the standard condition. These curves are used for the calculation of the tensile strength and the elongation at break. The mechanical properties are compared with the structural data determined by the other experimental techniques. This combination reveals specific molecular mechanisms for the modification of the mechanical properties due to phyllosilicates.

Results

Crystalline properties

In Figure 2, typical diffraction patterns of the PA6 fibers are displayed and are representative for all measurements. The small-angle scattering reflections (2θ ≈ 2°) appear on the equator of the patterns only at high DRs for fibers with phyllosilicates.
Figure 2.
Typical diffraction patterns of PA6 fibers (left: 0% of phyllosilicates with a DR = 1.22, right: 2% of phyllosilicates with a DR of 1.8).

For further analysis, intensity profiles are calculated for selected regions. Background intensity due to air scattering and noise from the readout process is subtracted. The regions in the diffraction patterns and the corresponding intensity profiles are displayed in Figure 3.
Figure 3.
Regions in diffraction patterns and corresponding intensity profiles: (a) scans in radial direction, (b) scans in azimuthal direction (intensity distribution in region 3 corresponds to region 1 in (a) and vice versa).

The 2θ profiles can be used for the determination of the unit cell and crystallite dimensions meridional and parallel to the fiber axis. Equatorial profiles contain one reflection from the two lattice planes (200) and (001) of the γ phase (indices for pseudohexagonal description of the crystalline form) and (200) as well as (002) of the α phase. The meridional profiles (parallel to the fiber) axis contain one peak from the (020) lattice plane. For the description of the crystalline intensity distributions, a Lorentz peak model is chosen, where the center Bragg angle 2θ _c , the half width Δ2θ, and the total intensity I₀ are refined:
$I (2 θ) = I_{o} \cdot \frac{1}{2 π} \cdot \frac{Δ 2 θ}{(2 θ - 2 θ_{c}) + (\frac{Δ 2 θ}{2})^{2}}$
(4)

The distance (d) between the crystalline lattice planes is calculated from the center diffraction angle and the wavelength with the help of Bragg’s equation:
$d = \frac{λ}{2 \cdot sin (\frac{2 θ_{c}}{2})}$
(5)

For all fiber samples, the d values for (200), (020), and (001) correspond to well-investigated values of the unit cell of γ phase PA6 (4.79 Å in the pseudohexagonal description)^5,54 and do not change with the process parameters. Therefore, they cannot be used to evaluate the effect of the spinning process.

A qualitative value for the crystallite dimensions (D) is derived from the peak position and half width with the help of the Scherrer’s equation:
$D = \frac{λ}{Δ 2 θ \cdot cos (\frac{2 θ_{c}}{2})}$
(6)

Further evaluation is carried out for diffraction patterns with additional intensity distributions at small diffraction angles, where two different interpretations are possible. In the first case, the intensity is considered small-angle scattering, where the slope δ of the intensity distribution provides information about typical sizes of regions having other densities than the polymer in the fiber structure and follows Guinier’s law:⁵⁵
$I (Q) = Δ ρ^{2} N_{p}^{2} V_{p}^{2} \exp (- \frac{Q^{2} δ^{2}}{3})$
(7)

The variable q is the norm of the scattering vector, which can be determined from the diffraction angle:
$Q = \frac{4 π}{λ} \cdot sin (\frac{2 θ}{2})$
(8)

The other parameters are the density difference between void and polymer Δρ as well as the number N_p, and the volume of the density fluctuations V_p. Since the reflections appear on the equatorial direction of the patterns, these voids are perpendicular to the fiber axis. The resulting distances are interpreted as spaces between polymer chains due to the embedding of silicate layers. The corresponding parameter δ (Guinier slope) is determined by a linear regression, where the logarithmic intensity is plotted against the square of the scattering parameter Q².

The second interpretation is based on the shape transformation of the silicate crystal morphology, where the intensity distribution is a partially visible Bragg reflection of the silicate crystal lattice. A smaller amount of stapled silicate layers lead to an increase of the full width at half maximum (FWHM). The size D_s and number of stapled silicates is then calculated with the help of the Scherrer equation. Both interpretations lead to similar values for the number of embedded silicate layers between the polymer chains and the same adjacent molecular mechanism.

The azimuthal scans provide information on the orientation distribution of the crystallites (f_c) as well as the orientation of the phyllosilicates (f_s). The mean deviation from the peak center is used for the calculation of the orientation factor f ranging from –1/2 for samples oriented in the other direction (not possible for fibers) to 0 for unoriented samples up to 1 for perfect orientation:⁵⁶
$f = 1 - \frac{3}{2} < \sin^{2} (ϕ) >$
(9)

In Table 2, an overview of the crystalline properties of all samples is given.
Table 2.
Crystalline properties of the PA6 fibers with different mass fractions of silicates and draw ratios revealed from WAXD

Sample 1 2 3 4 5 6 7 8

Silicates 0 % 0 % 1 % 2 % 2 % 2 % 2 % 2 %

Draw ratio (DR) 1.0185 1.22 1.22 1.0185 1.22 1.45 1.65 1.8

Lattice spacing perpendicular d_II [Å] 8.47 8.45 8.52 8.52 8.46 8.49 8.48 8.45

Lattice spacing parallel D_II [nm] 20.2 17.9 18.5 19.8 18.8 17.6 17.7 17.5

Orientation factor of crystallites (perpendicular) f_c,┴ 0.85 0.91 0.89 0.86 0.92 0.93 0.91 0.92

Orientation factor of crystallites (parallel) f_c_,_II 0.86 0.90 0.88 0.87 0.91 0.90 0.92 0.93

Orientation factor of silicates f_s – – – – 0.88 0.89 0.92 0.93

Spacing due to silicates from Guinier slope δ [nm] – – – – 2.35 2.15 1.85 1.49

Spacing due to silicates from FWHM D_s [nm] – – – – 2.45 2.11 1.67 1.35

Thermal properties

Typical DSC thermograms with and without phyllosilicates as well as from quiescent crystallization (second heating process) are displayed in Figure 4. Calculated values for the glass transition temperature T_g, the corresponding change in heat capacity Δc_p, the melting temperature T_m and the melting enthalpy ΔH are indicated.
Figure 4.
DSC thermograms (with and without phyllosilicates, exothermal heat flow on positive axis).

With increasing DR from 1.0185 to 1.22, a significant change in the heat flow curves takes place for both types of fiber. First, a relaxation process can be observed at the glass transition for the less drawn fibers, which also appears in the second heating process. The second change is a modification of the peak form for the higher ratio together with the appearance of a shoulder at lower temperatures. For the fibers with phyllosilicates allowing higher DRs (1.8), no significant change in the thermograms takes place. For all fibers, as well as for quiescent crystallization, the changes in heat capacity Δc_p = (0.20 ± 0.02) J/gK at the glass transition T_g = (49.0 ± 0.4)°C as well as the melting enthalpy ΔH = (72.7 ± 0.4) J/g at the melting temperature T_m = (220.0 ± 0.6)°C remain constant. The degree of crystallinity is calculated from the fraction of the theoretical value for pure crystalline material of PA6:⁵⁷
$X_{c} = \frac{Δ H}{Δ H_{0}} = \frac{Δ H [J / g]}{190.6} \cdot 100 %$
(10)

The thermal properties of all samples are summarized in Table 3.
Table 3.
Thermal properties of the PA6 fibers with different mass fractions of silicates and draw ratios revealed from DSC (Q: quiescent crystallization, second heating)

Sample 1 2 4 5 6 7 8 9 Q

Silicates 0 % 0 % 1 % 2 % 2 % 2 % 2 % 2 % 0 %

Draw ratio (DR) 1.0185 1.22 1.22 1.0185 1.22 1.45 1.65 1.8 -

Glass transition temperature T_g [°C] 48.4 49.7 49.3 48.9 48.5 49.1 48.7 49.3 48.8

Change in heat capacity Δc_p [J/g·K] 0.19 0.19 0.21 0.22 0.18 0.2 0.18 0.23 0.21

Melting temperature T_m [°C] 219.3 220.5 221.1 219.5 220.3 219.9 219.2 220.8 219.5

Melting enthalpy ΔH [J/g] 73.3 73.0 72.1 72.3 73.0 72.8 73.1 72.5 72.3

Crystallinity X_c [%] 38.5 38.3 37.8 37.9 38.3 38.2 38.4 38.0 37.9

Mechanical properties

Stress strain curves of fibers with DR 1.22 with and without phyllosilicates are displayed in Figure 5. It is obvious that the silicates affect the mechanical properties of the fibers, whereas the elongation at break is increased and the tensile strength is reduced dramatically.
Figure 5.
Stress strain curves of fibers with DR 1.22 (with and without phyllosilicates).

The values for the tensile strength F_max and the elongation at break E_max are summarized in Table 4. In addition, the values for the fineness T are displayed. Since extrusion velocity and winding speed were constant for all materials, the fineness is nearly the same for all fibers.
Table 4.
Mechanical properties of the PA6 fibers with different mass fractions of silicates and draw ratios revealed from stress strain curves (fineness T from reeling)

Sample 1 2 3 4 5 6 7 8

Silicates 0 % 0 % 1 % 2 % 2 % 2 % 2 % 2 %

Draw ratio (DR) 1.0185 1.22 1.22 1.0185 1.22 1.45 1.65 1.8

Tensile strength F_max [cN/dtex] 5.68 5.75 4.06 3.08 3.19 3.72 4.05 4.31

Elongation at break E_max [%] 49.2 46.6 65.4 76.9 73.6 63.8 55.8 51.8

Fineness T [dtex] 91.4 91.4 91.2 91.2 91.4 91.3 91.5 91.5

Discussion

Crystallinity, phase contents and morphology

The DSC analysis of the multifilaments showed a nearly constant melting enthalpy in all samples, which also occurs in the second heating process. This indicates that the total crystallinity is not affected by the process conditions chosen for the production of the PA6 fibers. Furthermore, the phyllosilicates do not seem to affect the crystallization behavior of the material. Compared with other studies showing a nucleation of γ phase on the silicates,⁸ this emphasizes the importance of stress-induced crystallization in the high-speed spinning process. The comparison of the fiber samples with the second heating leads to the conclusion that the fraction of crystalline material is completely unaffected by the extreme cooling conditions in the melt spinning process. Although the total crystallinity X_c ≈ 38.1% is constant all over the processing conditions, the form of the DSC thermograms changes significantly. The appearance of the shoulder before the melting peak for higher DRs indicates a transition phase to a greater amount of the crystalline α phase during the heating process. This conversion to α phase during heat treatment is special for polymeric fibers since the reverse transition can be observed in samples with no orientation (e.g. in compression molding).⁵⁷ The temperatures for the shoulder observed in the experiments correspond to a phase transition to a high-temperature form similar to α phase known from another study.⁸ This transition occurs by increasing the DR from 1.0185 to 1.22, whereas the peak form remains constant for higher DRs in the silicate doped samples. This observation indicates that the γ to α transformation occurs in the early stages of drawing. In this step, only the polymer chains are affected by the mechanical stress of the inline drawing process. At higher DRs, the thermal response of the polymer is constant, so that further drawing is assumed to affect the silicates. However, the corresponding changes in the silicate system are not detectable due to the small mass fraction in the polymer matrix.

The structural phase transition is accompanied by two other effects. First, the relaxation process during the glass transition of the amorphous fraction is suppressed when the material has been transformed to the α phase. Since glass transition occurs in the noncrystalline regions, this different nature of the process cannot be generated directly by the crystal phases. However, it is assumed to be caused by a change in the morphology, where amorphous and crystalline regions are arranged differently. Like the transformation in the crystalline regions, this effect is unaffected by the presence of phyllosilicates in the polymer matrix. The second effect is a slightly smaller value of the crystallite size parallel to the fiber axis. This occurs in combination with the increasing γ to α transformation. Therefore, this observation also provides an indication of changes in the crystal morphology, whereas in this case the crystallite geometry (detectable in the crystallite size) is changed. Like the other observations affecting the crystal phases, this change is also independent of the presence and amount of silicate dopants.

Orientation distribution and dispersion of silicates

Information about the orientation distribution is obtained from the evaluation of the intensity profiles. Since small angle scattering in the patterns due to voids is assumed to be an effect of the silicates embedded in the polymer, the corresponding intensity distribution is used for the determination of the placement of silicates in the fiber. The orientation factor f of the crystallites as well as the silicates is displayed in Figure 6a. The crystalline regions are well oriented in all fibers, but this orientation increases even more together with the γ to α transformation. For higher DRs, the factor remains almost constant. Since values for f with different mass fractions of silicates are very similar, the dopant does not seem to affect the reorientation. Instead of that, the effect is also assumed to take place due to a significant change in the morphology of the fiber structure, where the fibrils of the α phase have a stronger orientation along the fiber axis.
Figure 6.
(a) Orientation factors of crystallites and silicates dependent to DR, (b) Guinier slope δ as a function of DR.

For the lowest DR, no small-angle scattering signal can be detected, so that no orientation information can be obtained. In this case, the silicates are assumed to be more or less disoriented respective to the fiber axis so that no orientation occurs by the drawing of the melt. For higher DRs in the solid state, the silicate orientation increases continuously and reaches values as high as the crystalline ones. Since small-angle scattering is observed in the equatorial plane of the diffraction patterns, the normal to the silicate layers defined by distance to the next silicate sheet is perpendicular to the fiber axis.

Further information on the embedding of the silicate layers in the polyamide matrix can be obtained from the slope δ determined in the Guinier plot described in the experimental section, or alternatively from the FWHM of the reflection and the Scherrer formula. Both values represent a medium distance of polymeric material, whereas this void is filled with silicates. Consequently, the parameter can be used for the calculation for an average thickness of silicates in the matrix. With the help of the typical distance between two silicate layers, the corresponding number of adjacent layers can be evaluated, which is a benchmark for the dispersion of silicates in polyamide. As seen in Figure 6b, the Guinier slope δ decreases with increasing DR. The final value for the highest DR 1.8 (1.49 nm) corresponds to the dimension of one silicate layer (with regard to the organic component, which is compatible with the polymer and does not cause voids). Therefore, a defoliation of silicates takes place in the drawing process and results in single silicate layers dispersed in the polymer matrix. The same effect of silicate defoliation, remarkable in smaller voids in the polymeric material, was detected by Ibanes et al.^8,58 Compared with this study with winding speeds less than 1000 m/min, the effect was demonstrated here for high-speed melt spinning (5000 m/min). Furthermore, the material was drawn at lower temperatures (40°C compared with 140°C). Both variations cause a slower phase transformation since kinetics is slower at lower temperatures and the transition has less time to take place at higher speeds. Therefore, the effect of α phase formation is less detectable here. It is remarkable that a complete defoliation of the silicates could be achieved in an industrially relevant spinning process.

Drawability

As seen from the mechanical characterization, the tensile strength decreases with an increasing amount of phyllosilicates, whereas the elongation and therefore the drawability increase drastically. These results are in contrast to prior work, where nanocomposites were created by in-situ polymerization and silicate layers were already dispersed. In this case, the maximum elongation was reduced.²⁷ Thus, this seems to be an effect of melt compounding, where silicates are first agglomerated in the granule. Since dispersion takes place in the drawing process, drawability can only be increased when the nanocomposites are prepared by melt compounding. With increasing DR, the tensile strength increases together with a decreasing value for the elongation but never reaches the higher values for pure PA6 filaments. The dependency of both mechanical properties from the mass fraction of phyllosilicates and the DR is almost linear. Compared with the study of Ibanes et al.,^8,58 tensile strength in particular is much higher here. This is the effect of the high-speed spinning process, showing the possibility of an industrial application of silicates for achieving higher drawability. The corresponding diagrams are shown in Figure 7.
Figure 7.
Tenacity in dependence of mass fraction of phyllosilicates and DR.

Molecular interpretation

With the help of information discussed in the previous paragraphs, a molecular interpretation of the drawing process is possible. For both pure and nano modified PA6 fibers, the mechanical stress in the process first affects the polymeric matrix. In this step, a significant change in the crystal morphology takes place, causing a structural phase transition from the crystal phase γ to α, a better molecular orientation, and smaller crystallites in the lamellar fiber structure. For the PA6 nanocomposites, a first orientation of the silicates takes place. In this step, the silicate layers are more or less agglomerated as revealed from the higher values for the Guinier slope. Beyond the point of break for pure PA6 multifilaments, extended drawability is caused by the presence of silicates in the case of agglomerated sheets. Mechanical stress acts on the silicate layers, where weak binding forces between two layers allow a gliding and a defoliation of stapled silicate sheets. These binding forces are stronger than the conformational and structural energy saved in the polymer matrix, since the process takes place at higher DRs with increased mechanical stress. However, they are weaker than the forces acting between a polymer chain and an adjacent silicate layer. The other way around, no gliding would be possible since no mechanical forces would be transmitted from the polymer to the silicate system. Weak binding forces can also explain the decrease of tensile strength. For the highest DR, the defoliation is fulfilled and single-layer silicates are well dispersed and oriented in the polyamide matrix. Since no further orientation and gliding processes can take place, the fiber breaks after this point. The complete transformations at the molecular scale occurring in the drawing process are exemplified in Figure 8.
Figure 8.
Overview of all transformations at molecular scale occurring in the drawing process.

Conclusion

The presence of phyllosilicates in PA6 multifilaments leads to an increase of the drawability in the inline drawing process. For achieving these properties, the preparation should be done by melt compounding since agglomerated silicate sheets are necessary. In this study, it was first demonstrated that this effect can be used in an industrially relevant high-speed spinning process at 5000 m/min. For lower DRs, only the PA6 polymer chains are affected by the mechanical stress. This effect is independent of the silicates and leads to a higher orientation of the polymer chains as well as a transformation in the crystalline regions. This transformation is assumed to be the conversion from the γ to the α phase. For higher DRs, beyond the elongation of break for pure PA6, a defoliation of the silicate layers takes place. Drawing of the fibers is possible until single silicate layers are dispersed in the polymer matrix. Compared with other studies with lower winding speeds, the structural phase transition from γ to α is less dominant, but the effect of the silicates is larger. In this case, the doping of PA6 with nano phyllosilicates could be used to reduce the fiber fineness and to produce ultrafine filaments. Even though mechanical properties are lower than for unfilled PA6, they are still good enough for an industrial application of these filaments. For this purpose, further experiments with a reduced mass flow rate have to be carried out. Furthermore, the reduction of tensile strength has to be compensated with an appropriate procedure of processing the fibers.

Parameter	Values
Capillary geometry (number of filaments × capillary diameter × capillary length)	24 × 0.25 mm × 0.5 mm
Extrusion temperature	265°C
Mass flow rate	44.0 g/min
Extrusion velocity	25.7 m/min
Draw down ratio	191.0/159.3/134.0/117.7/107.9
Inline draw ratio	1.0185/1.220/1.450/1.650/1.800
Godet temperature (godet 1/godet 2)	40°C/160°C
Winding speed	5000 m/min
Mass fraction phyllosilicates	0%/1%/2%
Quench air (velocity/ temperature/humidity)	0.35 m/s/20°C/75 %

Sample	1	2	3	4	5	6	7	8
Silicates	0 %	0 %	1 %	2 %	2 %	2 %	2 %	2 %
Draw ratio (DR)	1.0185	1.22	1.22	1.0185	1.22	1.45	1.65	1.8
Lattice spacing perpendicular d_II [Å]	8.47	8.45	8.52	8.52	8.46	8.49	8.48	8.45
Lattice spacing parallel D_II [nm]	20.2	17.9	18.5	19.8	18.8	17.6	17.7	17.5
Orientation factor of crystallites (perpendicular) f_c,┴	0.85	0.91	0.89	0.86	0.92	0.93	0.91	0.92
Orientation factor of crystallites (parallel) f_c_,_II	0.86	0.90	0.88	0.87	0.91	0.90	0.92	0.93
Orientation factor of silicates f_s	–	–	–	–	0.88	0.89	0.92	0.93
Spacing due to silicates from Guinier slope δ [nm]	–	–	–	–	2.35	2.15	1.85	1.49
Spacing due to silicates from FWHM D_s [nm]	–	–	–	–	2.45	2.11	1.67	1.35

Sample	1	2	4	5	6	7	8	9	Q
Silicates	0 %	0 %	1 %	2 %	2 %	2 %	2 %	2 %	0 %
Draw ratio (DR)	1.0185	1.22	1.22	1.0185	1.22	1.45	1.65	1.8	-
Glass transition temperature T_g [°C]	48.4	49.7	49.3	48.9	48.5	49.1	48.7	49.3	48.8
Change in heat capacity Δc_p [J/g·K]	0.19	0.19	0.21	0.22	0.18	0.2	0.18	0.23	0.21
Melting temperature T_m [°C]	219.3	220.5	221.1	219.5	220.3	219.9	219.2	220.8	219.5
Melting enthalpy ΔH [J/g]	73.3	73.0	72.1	72.3	73.0	72.8	73.1	72.5	72.3
Crystallinity X_c [%]	38.5	38.3	37.8	37.9	38.3	38.2	38.4	38.0	37.9

Sample	1	2	3	4	5	6	7	8
Silicates	0 %	0 %	1 %	2 %	2 %	2 %	2 %	2 %
Draw ratio (DR)	1.0185	1.22	1.22	1.0185	1.22	1.45	1.65	1.8
Tensile strength F_max [cN/dtex]	5.68	5.75	4.06	3.08	3.19	3.72	4.05	4.31
Elongation at break E_max [%]	49.2	46.6	65.4	76.9	73.6	63.8	55.8	51.8
Fineness T [dtex]	91.4	91.4	91.2	91.2	91.4	91.3	91.5	91.5

Footnotes

Funding

Contract grant sponsor: Deutsche Forschungsgemeinschaft (German Research Foundation, DFG); research project “FiberSage”, contract grant number: GR 1311/34-1.

References

Schlack P. Preparation of polyamides. Patent 2241321, USA, 2002.

Fourné

. Synthetische fasern, 1st edn. München: Hanser, 1995.

Wei

. Induced γ → α crystal transformation in blends of polyamide 6 and liquid crystalline copolyester. Macromolecules 2000; 33: 5181–5186.

Fornes

Paul

. Crystallization behavior of nylon 6 nanocomposites. Polymer 2003; 44: 3945–3961.

Arimoto

Ishibashi

Hirai

. Crystal structure of the γ-form of nylon 6. J Polym Sci Pol Phys 1965; 3: 317–326.

Holmes

Bunn

Smith

. The crystal structure of polycaproamide: nylon 6. J Polym Sci Pol Chem 1955; XVII: 159–177.

Ziabicki

. Über die mesomorphe β-form von polycapronamid und ihre umwandlung in die kristalline form α. Kolloid Z 1959; 167(2): 132–141.

Ibanes

de Boissieu

David

. High temperature behavior of the crystalline phases in unfilled and clay-filled nylon 6 fibers. Polymer 2006; 47: 5071–5079.

Rahbar

Mojtahedi

. Influence of hot multistage drawing on structure and mechanical properties of nylon 6 multifilament yarn. J Eng Fibers Fabrics 2011; 6(2): 11–11.

10.

Gianchandani J, Spruiell JE and Clark ES. J Appl Polym Sci 1982; 27: 3527–3551.

11.

Varlot K, Reynaud E, Kloppfer MH and Vigier G. Clayreinforced polyamide: preferential orientation of the montmorillonite sheets and the polyamide crystalline lamellae. J Polym Sci, Part B: Polym Phys 2001; 39: 1360–1370.

12.

Xiaohui Liu and Qiuju Wu. Phase transition in nylon 6/clay nanocomposites on annealing, Polymer Communication, Polymer 2002; 43: 1933–1936.

13.

Zhang

Liu

. Crystalline behaviors and phase transition during the manufacture of fine denier PA6 fibers. Sci China Press 2002; 52(11): 1835–1842.

14.

Craubner

. Untersuchungen über den schmelzspinnproze ß und die fadenbildung bei polyamid 6. Angew Makromol Chem 1974; 40/41: 173–217.

15.

Ishibashi

Aoki

Ishii

. Studies on melt spinning of nylon 6. I. Cooling and deformation behavior and orientation of nylon 6 threadline. J Appl Polym Sci 1970; 14: 1597–1613.

16.

Ishibashi

Ishii

. Studies on melt spinning of nylon 6. ii. effect of heating the threadline upon orientation and crystallization. J Appl Polym Sci 1976; 20: 335–344.

17.

Bankar

Spruiell

White

. Melt spinning of nylon 6: structure development and mechanical properties of as-spun-filaments. J Appl Polym Sci 1977; 21: 2341–2358.

18.

Hirami

Tanimura

. Structure development in the melt spinning of nylon 6. J Macromol Sci Phys 1981; 19(2): 205–210.

19.

Haberkorn

Hahn

Breuer

. On the neck-like deformation in high-speed spun polyamides. J Appl Polym Sci 1993; 47: 1551–1579.

20.

Samon

Schultz

Hsiao

. Structure development in the early stages of crystallization during melt spinning. Polymer 2002; 43: 1873–1875.

21.

Tjong

. Structural and mechanical properties of polymer nanocomposites. Mat Sci Eng R 2006; 53: 73–197.

22.

Ray

Okamoto

. Polymer/layered silicate nanocomposites: a review from preparation to processing. Prog Polym Sci 2003; 28: 1539–1641.

23.

Carrado

. Synthetic organo- and polymer-clays: preparation, characterization: materials applications. Appl Clay Sci 2000; 17: 1–23.

24.

LeBaron

Wang

Pinnavaia

. Polymer-layered silicate nanocomposites: an overview. Appl Clay Sci 1999; 15: 11–29.

25.

Mahfuz

Hasan

Rangari

. Investigation of thermal and mechanical response of nylon-6 filaments with the infusion of acicular and spherical nanoparticles. NSTI-Nanotech 2005; 2: 86–89.

26.

Cho

Paul

. Nylon 6 nanocomposites by melt compounding. Polymer 2003; 42: 1083–1094.

27.

Kim

Lee

Prabu

. Preparation and characterization of nylon 6/organoclay nanocomposite filament fibers. Polymer Composite 2009; 42: 1083–1094.

28.

Ming

Xiangqing

Wang

. Preparation and characterization of polyamide6/montmorillonite nanocomposites by reactive extrusion. Int J Polym Mat 2006; 55(2): 147–159.

29.

Shah

Paul

. Nylon 6 nanocomposites prepared by a melt mixing masterbatch process. Polymer 2004; 45: 2991–3000.

30.

Yang

Chen

. Mechanical properties and flame retardancy research of montmorillonite intercalate polyamide 66 composites. J Compos Mater 2009; 43(23): 2785–2792.

31.

Gilman

Morgen

Harris

. Polymer layered-silicate nanocomposites: polyamide 6, polypropylene and polystyrene: new advances in flame retardant technology. In: Proceedings. Fire Retardant Chemicals Association; 1999 Oct 24-27; Tucson, US, Lancaster: Fire Retardant Chemicals Assoc, 1999; 9–22.

32.

Reynaud

Jouen

Gauthier

. Nanofillers in polymeric matrix: a study on silica reinforced PA6. Polymer. 2001; 42: 8759–8768.

33.

Krištofič

Dulíková

Vassová

. PA 6/copolyamide/layered silicate fibers. Fibres Text East Eur 2007; 15(5–6): 64–65.

34.

Masenelli-Varlot

Reynaud

Vigier

. Mechanical properties of clay-reinforced polyamide. J Polym Sci Pol Phys 2002; 40: 272–283.

35.

. Structural analysis of polyamide/clay nanocomposites. J Macromol Sci Phys 2002; 41(1): 17–31.

36.

Lee

Park

Kim

. Structure and properties of polyamide-6 & 6/66 clay nanocomposites. J Macromol Sci Chem 2009; 46(1): 65–73.

37.

Tamura

Uno

Yamada

. Layered silicate-polyamide-6 nanocomposites: influence of silicate species on morphology and properties. J Polym Sci Pol Phys 2009; 47(6): 583–595.

38.

Hedicke

Wittich

Mehler

. Crystallization behavior of polyamide-6 and polyamide-66 nanocomposites. Compos Sci Technol 2006; 66(3–4): 571–575.

39.

Dinzhos

Privalko

. Enthalpy relaxation in the cooling/heating cycles of polyamide 6/organoclay nanocomposites I. nonisothermal crystallization. Macromol Sci Phys 2005; 44(4): 421–430.

40.

Mathias

Davis

Jarrett

. Observation of α and γ crystal forms and amorphous regions of nylon 6-clay nanocomposites using solid state ¹⁵N nuclear magnetic resonance. Macromolecules 1999; 32: 7958–7960.

41.

. Suppressed molecular orientation in nylon 6/clay nanocomposite at large strain: role of microvoiding. J Polym Sci Pol Phys 2010; 48(5): 514–519.

42.

Garofalo

Russo

Scarfato

. Nanostructural modifications of polyamide/MMT hybrids under isothermal and nonisothermal elongational flow. J Polym Sci Pol Phys 2009; 47(10): 981–993.

43.

Baldi

Fanceschini

Bignotti

. Rheological behavior of nano-composites based on polyamide 6 under shear and elongational flow at high strain rates. Rheol Acta 2009; 48: 73–88.

44.

Choi

Lim

Choi

. Viscoelastic properties of exfoliated polyamide-6/layered silicate nanocomposites. J Mater Sci 2006; 41: 1843–1846.

45.

Loo

Gleason

. Investigation of polymer and nanoclay orientation distribution in nylon 6/montmorillonite nanocomposites. Polymer 2004; 45: 5933–5939.

46.

Zhao

. Thermal history dependence of polymorphic transformation of polyamide 6/silicate nanocomposites. Polym Int 2009; 58(5): 469–474.

47.

Wurm

Ismail

Krezschmar

. Retarded crystallization in polyamide/layered silicates nanocomposites caused by an immobilized interphase. Macromolecules 2010; 43: 1480–1487.

48.

Fong

Liu

Wang

. Generation of electrospun fibers of nylon 6-montmorillonite nanocomposites. Polymer 2002; 43: 775–780.

49.

Bellan

Craighead

. Formation and properties of nylon-6 and nylon-6/montmorillonite composite nanofibers. Polymer 2006; 47: 6208–6217.

50.

Kim

Michler

Ania

. Temperature dependence of polymorphism in electrospun nanofibers of PA6 and PA6/clay nanocomposite. Polymer 2007; 48: 4814–4823.

51.

Giza

Ito

Kikutani

. Fiber structure formation in high-speed melt spinning of polyamide 6/clay hybrid nanocomposite. J Macromol Sci Phys 2000; 39(4): 545–559.

52.

BASF. Ultramid B24 N 03 product information. Ludwigshafen: BASF Plastics. 2007 Mar. Available from: http://prospector.ides.com/. Accessed: 14.10.2011.

53.

Rockwood Clay Additives. Product bulletin/nanofil. cheshire: rockwood additives limited. Available from: http://www.rockwoodadditives.com. Accessed: 14.10.2011.

54.

Miyasaka

Makishima

. Transition of nylon 6 γ-phase crystals by stretching in the chain direction. J Polym Sci A 1967; 5: 3017–3027.

55.

Richardson R. Scattering and reflection techniques. In: Cosgrove editor (ed) Colloid Science: Principles, Methods and Applications. 2nd ed. Chichester: Wiley, 2010, pp. 273–298.

56.

Booys

Hermans

. Zur ableitung eines mittleren orientierungswinkels aus dem röntgendiagramm. Colloid Polym Sci 1941; 97: 229–231.

57.

Krištofič

Marcinčin

Ujhelyiová

. The DSC study of PA6, polyamides and copolyamides. J Therm Anal Calorim 2000; 60: 357–369.

58.

Lincoln

Vaia

Wang

. Temperature dependence of polymer crystalline morphology in nylon 6/montmorillonite nanocomposites. Polymer 2001; 42: 9975–9985.