Printer Friendly

Biaxial stretching of polymers using a novel and versatile stretching system.


Biaxial stretching of polymers is an important process with various applications in film packaging, in textile, in plastic geogrid used to reinforce dyke and roadbed and in some high tech applications, such as peculiar piezoelectric devises [1]. This is because biaxial stretching imparts the polymer with two-directions orientation of the polymeric chains that enhances mechanical properties such as tensile, elongation at fa: lure, and crazing strength in the stretch directions.

Various biaxial stretching machines are available on the market allowing one-step or two-steps biaxial stretching at controlled temperature, and controlled draw ratio and stretch rate. Unfortunately such devices are generally very expensive and require large polymer dimensions for production purposes. Only a limited number of laboratory devices are available with more or less complex engineered tools allowing testing several polymers for their stretchability, which simulate processes such thermoforming [2-4], film blowing and blow molding [5, 6]. Such material characterisations are crucial for understanding and quantifying the importance of factors such as ther-momechanical properties, the extent of elongation and orientation, the effect of draw ratio and temperature on crystallinity and thus on the end-use mechanical properties, the homogeneity of surface finish and controlled thickness.

The aim of this article is to present a new versatile laboratory stretching module that can be adapted to any uniaxial traction testing machine to generate one-step or two-steps biaxial stretching at various stretching ratios and draw rates. The system can be used both at high temperatures (if the traction machine is provided with an oven) and at room temperature to study at the same time biaxial stretching and mechanical properties of films both in uniaxial and in biaxial modes using consecutive or simultaneous two-directional stretching. The size of the device can be also adjusted to accommodate the dimensions of the traction machine and those of the oven.

Another application of this system is the study of the mechanical properties under biaxial stresses, for example, to determine the Young modulus in the two orthogonal directions, which is very important especially when the film shows different mechanical properties as they are tested along the axes of different directions. This is for instance the case of already uniaxially stretched films or composite materials involving fibers that are oriented in a preferential direction. Such mechanical characterization is achieved by the cell load or the strain gages of the traction machine. The shape of the sample can take various forms such as a cross form [7] or a squared one and there is also an engineering possibility of modifying the stretching elements for discs and circular samples.

In the first part of the paper, we will describe some details of the state-of-the art of the new technology which is the subject of a patent (provisional request N[degrees]61/129.127) and then present some results for the validation of the efficiency and the accuracy of the new system along with some results obtained on high density polyethylene (HDPE) and polypropylene (PP).

Detailed Description of the Device Design

The biaxial Stretching Device shown in Fig. 1 ensures processing of films by sequential or simultaneous biaxial stretching through the application of a uniaxial stress which is transformed to a biaxial one. The state of the stress and deformation is given by:


The system is designed such that uniaxial stretching is transformed to biaxial stretching with the possibility of adjusting the ratio (R) between displacements in the machine direction (11: MD) and in the transverse direction (22:TD) using classical and standard uni-axial traction machines. The home-made device is machined in the form of a separate module that can be adapted to any traction machine without any modification of the machine elements.

Figure 1 illustrates the general view of the developed Biaxial Stretching Module (BSM) along with some details of its various structural elements [8]


The hinged plate (11) of the BSM is provided with a series of holes (Tl to T4) (see Fig. 2), each of which provides a different stretching ratio (R) and this is obtained by positioning the pivots (T) of the rods (13) (see Fig. 3) in each of the positions (Tl, T2, T3, or T4). The hinged is rotated around the Z axis (see Fig. 1) to transfer the movement from the machine direction to transverse direction, enabling the biaxial displacement.

[D.sub.TD]=R [D.sub.MD] (2)

Where DTD and DMD are the transverse and the machine direction displacements, respectively.

The positions Ti provide different maximum draw ratios as indicated in Table 1. The technology is not limited to four positions, but additional holes or alternatively a sliding-blocking element might be adapted on the system to generate various draw ratios.

The module can be coupled to tensile machine by two interdependent lower and upper shafts (1 and 2) with the top and the bottom slides, respectively. The hole of the top shaft is oblong (15) to facilitate the installation of BSM on the tensile testing machine by giving a certain longitudinal freedom to the insert. The system is fixed by the knurled nut (5) to immobilize it.


Transfer of the main displacement movement from MD direction to TD direction is ensured by the rotation of the hinged plate (11), whose rotational movement is obtained using the rocs (14), which transmit the movement to the transverse rods (13), which in turn draw aside the two side arms as illustrated in Fig. 3.


The rates of displacement in both directions are imposed by the main rate of displacement of the tensile machine, which is transmitted directly with the same value to the longitudiral direction. The transverse speed is generally higher or equal to the longitudinal one and it is determined by:

[V.sub.TD] = [[V.sub.MD]/R] (3)

Where [V.sub.TD] and [V.sub.MD] are the transverse and the longitudinal displacement speeds, respectively. The pantograph system (7) [9] (Fig. 1) allows from one hand equidistant spacing between the grips (8) and in the other hand keep them aligned. The grips slide along the stems (12) minimize frictions through smooth slippage. The corners (6) relocate along the slides (3 and 4) and their displacement is facilitated by rollers assembled in each one of their ends.

The pinching of the film to be stretched is ensured by grips (see Fig. 4) designed such that self-clamping on the sample ends is ensured due to the thickness reduction during stretching. The surface of the grips is striated to effectively seize up on the surface of the sample pinching part and to avoid slippage.



Experimental Validation of the BSM Efficiency and Accuracy

Various tests without and with the presence of polymeric films were carried out to verify several issues such as (i) efficient transmission of the longitudinal displacement to the module, (ii) efficient and accurate transmission from MD direction to TD direction, (iii) accurate maximum draw ratio of the various Ti hole positions, (iv) uniformity of the thickness in the central and in the edge parts of the stretched films, and (v) the possibility of carrying out at the same time mechanical characterization such as stress-strain data and determination of the Young modulus in the two directions.

An example of such verifications is provided in Fig. 5 using the four Ti holes that impose given maximum draw ratios. First the figure shows that the machine (or longitudinal) direction speed in completely transmitted to the transverse direction. This can be verified by analysing the various curves corresponding to the different holes Ti (see Table 1). For instance, for T4, a perfect one-to-one speed is obtained as it is revealed by a straight line with a slope of 1. The second verification is that the different holes provide draw ratios in perfect proportions as indicated in Table 1 and as can be verified by the ratio of the slopes of the straight lines of the various Ti positions. For instance the line corresponding to T4 is characterised by a slope 1 and that corresponding to T2 is characterized by a slope 1/2, in perfect accordance with the specifications reported in Table 1.
TABLE 1. Maximum ratio at different hole position.

Hole  Maximum stretching ratio

Tl                        0.25
T2                        0.50
T3                        0.75
T4                        1.00

To verify the accuracy of measurements in the case where there is a difference in the longitudinal and the traverse directions, the experiments were conducted at various displacement ratios of and the gain in surface was determined as a function of the longitudinal displacement. The films of PP and HDPE with an initial surface of 100 cm2 were used. To calculate the surface gain rate, a portion of 10 mm was cut all around the sample (considered as unstretched zone), thus the surface gain rate (see Fig. 6) expressed by:




Where Si and Sf are the initial and the final film surfaces, DMD, R, and C are the displacement in the machine direction, the ratio between transverse and longitudinal displacements, and the characteristic sample dimension, respectively. In order to validate the relationship (4), BSM system was tested without sample and the results of such experiments are reported in Fig. 6. They are in perfect agreement with the relationship (4). The tests were made on a sample of initial dimensions 100 x 100 [mm.sup 2] For this purpose, the obtained surface gain rate is required only for the control of the MD displacement, which is completely transmitted from the tensile testing machine.

Biaxial Stretching of Polymer Films

The biaxial stretching experiments were carried out using a high density polyethylene (HDPE) and a polypropylene (PP). The two polymers are characterized by high enough mechanical strength and offer the possibility of biaxial stretching with a high draw ratio to make thin films. The two polymers were obtained from Basell Canada Company under the commercial names PP Pro-fax PDC 1274 and HDPE H6018. Their density, thermal conductivity, melting temperature, and their specific heat capacity are reported in Table 2 [10, 11].
TABLE 2. Polvmer characteristics.

        Density p      Thermal       Melting     Specific  Thermal
                                                   heal    expansion
      (g/cm.sup.3])  conductivity  temperature   capacity  Coefficient
                       k (W/mK)    ([degrees]c)     Cp     [([degrees]
                                                 (kJ/(kg    C.sup.-1])

pp            0.905          0.12           164     1.622        6.6 x
HDPE          0.960          0.52           130     1.555       7.14 x

The experimental setting of the system for the processing of thin films was carried out by use of a servo-hydrolic Instron traction machine equipped with an oven chamber that allows carrying out experiments a temperatures as high as 200[degrees]C. The homogeneous heating is ensured by radiating elements placed at equivalent distances within the chamber. Previous experiments using several thermocouples within the chamber revealed constant temperature mainly in the large central part of the chamber. Such homogeneity of the temperature in the oven was verified by both the manufacturer and in our lab.

The initial samples to be stretched were obtained by extrusion process using a twin-screw Leistritch corotating extrusion machine (Model ZSE27 with LID = 40) equipped with nine zones of heating and a flat die. Extrusion process was carried with a screw speed of 110 rpm and a barrel temperature profile of 180/190/190/180[degrees]C and 165/170/170/170[degrees]C from hopper to die for PP and HDPE, respectively.

The extruded films were continuously stretched at the exit of die using a calendaring system made of various parallel cylinders that allow controlling the film thickness by adjusting the rotation speed (draw ratio) and the gap between the cylinders. From the extruded films, square samples with the dimension of (100 X 100) m(m.sup.2 and 800 [micro]m in thickness were cut and were annealed for 48 h at 110[degrees]C in vacuum prior testing.

The biaxial stretching operation was carried out at high temperature. The heating temperature was selected to be high enough to enable film stretching with a minimum energy cost and sufficiently low to avoid melting of the samples that may generate sagging effects. Typically such a temperature was taken close to the material's melting point (150[degrees]C and 110[degrees]C for PP and HOPE, respectively).


The equilibrium heating time was chosen so that the temperature is kept uniform throughout the film, which was estimated using two methods. The first one is based on theoretical estimation using Fourier's law solution in the case of one dimensional conductive heat transfer.

T(x,t) = [T.sub.0] + ([T.sub.i] - [T.sub.0])erfc[x/[2[square root of [at]]]] (5)

where T0, Ti and [alpha] are the initial film temperature, the oven temperature and the thermal diffusivity given by:

a = [k/[p[C.sub.P]]] (6)

where k, Cp, and p are the thermal conductivity coeffi-cient, the specific heat and the material density, respectively. To calculate the equilibrium heating time, the real thickness should be considered, since the thickness is measured at room temperature, whereas the experiments are carried out at high temperature that causes thermal expansion of the polymer film. The thickness at high temperature is related to that at room temperature by:

[E.sub.Expanded] = [E.sub.initial](1 + [DELTA]T.[[alpha].sub.E]) (7)

where EExpanded, Einitial, AT, and [alpha]E are the expanded thickness, the initial thickness, the temperature gradient and the polymer thermal expansion coefficient. Taking into account thermal expansion, the estimated equilibrium high heating time for PP and HDPE with initial thicknesses, 500 [micro]m and 800 [micro]m are reported in Table 3.
TABLE 3. Heating time heorctically calculated for PP and HDPE

      Time (s) for film's          Time (s) for film's
      thickness 500 [micro]m       Ihickness 800 [micro]m

PP                    104                  265
HDPE                   63                  143

The second method is experimental and consists in measuring the involved mechanical work in transient regime as function of heating time. The work values are extracted from the stress-strain curves by measuring the area under the curve at a given heating time. The results of such experiments are reported in Fig. 7 for both PP and HDPE. The obtained results show that the required heating to generate homogeneous temperature throughout the sample is ~4 min and 3 min for the 800 [micro]m thick PP and HDPE films, respectively, time beyond which work is constant (see Fig. 7). These values are in rough agreement with the theoretical values reported in Table 3.


The sample dimensions were (100 X 100) m(m.sup.2) and 800 [micro]m in thickness. The stretching operation was carried out under the conditions (summarized in Table 4) in terms of temperature, heating time, longitudinal stretching speed (VMD). surface gain rate (%), and stretching ratio (R). Figure 8 shows the stretching system along with the clamped film before and after biaxial stretching.
TABLE 4. Operating conditions.

      Temperature     Healing     Stretching       Surface    Stretching
      ([degrees]C)  time (min)  speed (mm/s)   gain rate (%)    ratio
                                                              R =DTD/DMD

pp             150           4            40  45-100-180-300           1
HDPE           110           3            40  45-100-180-300           1


The thickness characterization consists in evaluating the thickness reduction ratio under the biaxial stretching effect, by comparison to a theoretical curve based on a constant volume as given by:

[T.sub.r] = [[[E.sub.i] - [E.sub.f]]/[E.sub.i]]

[T.sub.r] = 1 - [1/[[G.sub.s] + 1]]

where Ei and EF are sample thicknesses before and after stretching, respectively. The depth control was carried out using a dial gauge assembled on a magnetic base allowing estimation of the film thickness at various locations of the polymer film. To have more reliable data, the central region of the polymer films was subdivided into 16 squared areas as indicated in Fig. 9. First the thickness measurements in each square were averaged and the total film thickness was taken as the total average value of the 16 square areas. Only films with less than 10% difference between the thickness of the 16 square areas were considered.


The obtained results are reported in Figs. 10 and 11. For PP films, the experimental results in terms of thickness reduction according to the surface gain rate are in agreement with the theoretical estimation based on the hypothesis of a constant volume during stretching.




For HDPE films, the reduction ratio obtained experimentally was slightly higher than the theoretical estimation (Eq. 8% which could be explained by a reduction in volume during stretching. This is basically due the reorganization and the close up of the polymer chains under the effect of biaxial stretching, which is more likely to occur for HDPE than for PP due to methyl side group of PP than make the high closing up difficult to occur due to hindrance effects.

On the precision level, the maximum error recorded was 6% for a thickness reduction ratio of 49%, an error which can be due to the inaccuracy in thickness of the extruded film.


A new and versatile laboratory biaxial stretching system was developed to be coupled to any uniaxial traction testing machine to generate one-step or two-steps biaxial stretching at various stretching ratios and draw rates. The new system allows also simultaneous characterization of mechanical properties in transverse and longitudinal directions. Experiments conducted with and without polymer films showed the accuracy of the device in generating precise biaxial stretching with the same stretch in the two directions or with different draw ratios between the two directions.


Thanks to the Steacie fellowship (NESRC, Canada) and the Hassan II Academy of Science and Technology (Morocco) for their support.


(1.) J. Hillenbrand, N. Behrendt, V. Altstadt, H.W. Schmidt, and G.M. Sessler, J. Phys. D: Appl. Phys., 39, 535 (2006).

(2.) H. Angellier-Coussy, S. Torres-Giner, M.H. Morel, N. Gon-tard, and E. Gastaldi, J. Appl. Polym. Sci., 107, 487 (2008).

(3.) S.W. Hsiao and N. Kikuchi, Comput. Meth. Appl. Mech. Eng., 177, 1 (1999).

(4.) S. laboni and R.T. Woodhams, Polym. Compos., 8, 6 (1987).

(5.) Z.X. Tang and L. Ye, Plast. Rubber Compos., 32, 10, 459 (2003).

(6.) A.G. Fard, J. Plast. Film Sheeting, 15, 194 (1999).

(7.) N. Bhatnagar, R. Bhardwaj, P. Selvakumar, and M. Brieu, Polym. Test., 26, 154 (2007).

(8.) A. Qaiss and M. Bousmina, US Patent Appl 61/129, 127 (2008).

(9.) L. Capt, M.R. Kamal, S. Rettenberger, and H. Munstedt, ANTEC, Conf. Proc., 1, 997 (2003).

(10.) D.V. Howe, Polymer Data Handbook, Oxford University (1999).

(11.) L. Mandelkern and R.G. Alamo, Polymer Data Handbook, Oxford University (1999).

A. Qaiss,(1) M. Bousrnina(Bousmina (1), (2), (3)

(1) Department of Chemical Engineering, Canada Research Chair on Polymer Physics and Nanomaterials, Laval University Ste-Foy Quebec, G1K 7P4, Canada

(2) INANOTECH (Institute of Nanomaterials and Nanotechnology), MASclR (Moroccan Advanced Science, Innovation and Research Foundation), ENSET, Av. Des FARS, Madinat Al Irfane, 10100, Rabat, Morocco

(3) Hassan II Academy of Science and Technology, Rabat, Morocco

Correspondence to: M.Bousmina; e-mail:

DOI 10.1002/pen.21869

Published online in Wiley Online Library (

[c] 2011 Society of Plastics Engineers
COPYRIGHT 2011 Society of Plastics Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Qaiss, A.; Bousrnina, M.
Publication:Polymer Engineering and Science
Article Type:Report
Geographic Code:1CANA
Date:Jul 1, 2011
Previous Article:The nanostructure and dewetting of block copolymer thin films annealed in different neutral solvents.
Next Article:Stabilizations of molecular structures and mechanical properties of PVC and wood/PVC composites by tinuvin and Ti[O.sub.2] stabilizers.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters