253
Cellular Polymers, Vol. 24, No. 5, 2005
Mechanical Performance of Auxetic Polyurethane Foam for Antivibration Glove Applications
Mechanical Performance of Auxetic Polyurethane Foam
for Antivibration Glove Applications
F. Scarpa*, J. Giacominº, Y. Zhang*, and P. Pastorino*
*Multidomain Cellular Solids Laboratory, University of Sheffi eld, UK
ºPerception Enhancement Systems, University of Sheffi eld, UK
Received: 25 July 2005 Accepted: 2 September 2005
ABSTRACT
In this study the static and dynamic characteristics of conventional open cell
polyurethane (PU), of auxetic (negative Poisson’s ratio) and of iso-density foams
were analysed. The specimens were produced from conventional gray open-cells
polyurethane foam with 30-35 pores/inch and 0.0027 g/cm
3
density, by means
of process which has been previously defi ned by the authors. Poisson’s ratio
measurements were performed under quasi-static conditions using an MTS 858
servo-hydraulic test machine and a video image acquisition system. For the auxetic
foams the results suggested similar behaviour to that previously reported in the
literature, with signifi cant increases in stiffness during compressive loading, and a
signifi cant dependence of the Poisson’s ratio on the applied strain. Transmissibility
tests, performed in accordance with the ISO 13753 procedure for antivibration
glove materials, suggested a strong dependence of the transmissibility on the
foam manufacturing parameters. Within the frequency range from 10 to 31.5 Hz
the transmissibility was found to be greater than 1, while it was less than 1 at all
frequencies greater than 31.5 Hz. The transmissibility results were similar to the
mean values for 80 resilient materials tested by Koton et. al., but were higher
than the fi ve best materials (not all polymeric) identifi ed by the same researchers.
In this study it has been suggested that the resilient behaviour of glove isolation
materials should also be evaluated in terms of the indentation characteristics. A
simple, linear elastic, Finite Element simulation was therefore performed, and
the indentation results suggested that auxetic foams offer a signifi cant decrease
in compressive stresses with respect to conventional PU foams.
To whom correspondence should be addressed:
Dr Fabrizio Scarpa, Department of Aerospace Engineering, University of Bristol, BS8 1TR
Bristol, UK Tel: +44 117 9289728 Fax: +44 117 9773371 Email: f.scarpa@bris.ac.uk
CP 5 05.indb 253CP 5 05.indb 253 27/10/05 12:11:07 pm27/10/05 12:11:07 pm
254
Cellular Polymers, Vol. 24, No. 5, 2005
F. Scarpa, J. Giacomin, Y. Zhang and P. Pastorino
INTRODUCTION
With the entry into force in 2002 of the physical agents directive
(1)
, minimum
requirements have been established in all European Union member states for
the protection of workers against the possibly damaging effects of mechanical
vibration. For hand-arm vibration, the physical agents directive establishes a
daily exposure limit value of 5 m/s
2
of Wh weighted acceleration
(2)
, standardised
to an eight-hour reference period. A daily exposure action value of 2.5 m/s
2
is also specifi ed for the same frequency weighting and reference period. For
many pneumatically or hydraulically assisted tools, vibration emission values
less than those fi xed by the physical agents directive can be challenging to
achieve in practice.
One of the most commonly adopted measures for lowering human exposure
to hand-arm vibration is the use of vibration isolating, or antivibration, gloves.
Antivibration gloves are attractive because they are applicable to existing tools,
can be moved with the operator from one tool to the next, and can be less
expensive than the purchase of tools characterised by lower vibration emissions.
The vibration isolation is largely determined by the material properties and
thickness of the cushioning materials interposed between the vibrating surface
and the human hand. The identifi cation of appropriate cushioning materials is
therefore a key aspect of the exposure-reduction problem.
A promising new class of material which may prove useful in the glove
cushioning application is that of auxetic materials. An auxetic material is defi ned
as one having a negative Poisson’s ratio, thus a material whose lateral extent
will expand, rather than contract, under uniaxial tension. The term “auxetic”
was fi rst defi ned by Evans and co-workers at the University of Exeter
(3)
, and it
is one of the defi nitions entered in the mainstream for these specifi c materials.
Since 1987, when isotropic auxetic foam was fi rst manufactured
(4)
, there has been
much interest in developing negative Poisson’s ratio materials for use in several
engineering fi elds. The negative Poisson’s ratio behaviour does not contradict
the classical theory of elasticity. A homogeneous, isotropic, thermodynamically
correct 3D solid has a potential Poisson’s ratio range between –1.0 and 0.5,
while the magnitude of the Poisson’s ratio can be even larger in the case
of anisotropic solids
(3)
. A negative Poisson’s ratio coeffi cient for a material
can lead to increased indentation resistance
(5)
, enhanced bending stiffness in
structural elements, increased shear resistance
(6)
in structural elements, optimal
passive tuning of structural vibration
(7)
and enhanced dielectric properties for
microwave absorbers
(8)
. The dynamic behaviour of auxetic materials has been
investigated by several authors
(9,10,11,12)
. The possible use of auxetic materials for
viscoelastic damping applications has been examined in
(13)
, where a biphasic
CP 5 05.indb 254CP 5 05.indb 254 27/10/05 12:11:08 pm27/10/05 12:11:08 pm
255
Cellular Polymers, Vol. 24, No. 5, 2005
Mechanical Performance of Auxetic Polyurethane Foam for Antivibration Glove Applications
auxetic composite showed a loss tangent exceeding the lower Voigt limit, and
close to the Hashin upper bound. In recent papers
(14,15)
increases in loss factor
and storage moduli have been shown for auxetic PU foams manufactured
using a modifi ed route from the classical production layout. The unusual
deformation mechanisms of auxetic foams have been also recently proposed
for displacement control via magnetic fi elds in foam composite with one phase
made of magnetorheological fl uid
(16)
. One possible application of auxetic foams
in ergonomics and human factors is that of special seat cushions
(17)
, where
negative Poisson’s ratio foams can be used to reduce interface pressures, thus
reducing discomfort and possibly preventing pressure sores.
The testing of antivibration glove cushioning materials has been the subject
of several research studies
(19,20)
which have lead to the establishment ISO
standard 13753:1999 Mechanical vibration and shock – Hand-arm vibration
– Method for measuring the vibration transmissibility of resilient materials
when loaded by the hand-arm system
(21)
, which defi nes a method for measuring
the mechanical transmissibility of glove cushioning materials. As stated in ISO
13753:1999, “The method determines the impedance of the material when
loaded by a mass providing a compression force equivalent to that found when
the material is gripped by the hand. This is done by measuring the transfer
function of the mass-loaded material at all the required frequencies. The vibration
transmission when loaded by the hand is computed using standard values of
hand-arm impedance and the measured values of the material impedance. The
impedances used in this International Standard are for the palm of the hand
when gripping a circular handle.” ISO 13753:1999 provides a useful, and
relatively simple, method for evaluating the isolation properties of resilient
materials for antivibration gloves or other products which attempt to reduce
the transmission of vibration to the human hand-arm system.
This paper describes a research study in which the possible antivibration
properties of auxetic polyurethane foam have been evaluated. The research has
been performed in light of the possibility of fabricating antivibration gloves
from auxetic polyurethane foam, therefore the ISO 13753 method of defi ning
material transmissibility has been adopted. In order to evaluate the performance
of the auxetic foams that were defi ned and produced, comparisons are made
with respect to a conventional polyurethane foam sample.
FOAM SAMPLE MATERIAL PROPERTIES
The specimens that were tested were produced from conventional grey open-
cells polyurethane foam with 30-35 pores/inch and 0.0027 g/cm
3
density,
CP 5 05.indb 255CP 5 05.indb 255 27/10/05 12:11:09 pm27/10/05 12:11:09 pm
256
Cellular Polymers, Vol. 24, No. 5, 2005
F. Scarpa, J. Giacomin, Y. Zhang and P. Pastorino
supplied by the McMaster-Carr Co. of Chicago Illinois. The manufacturing
process followed a procedure outlined in
(15)
. An aluminium mould consisting
of cylinders backed by pistons was used to impose concurrent axial and radial
compression on the parent foam samples. The compression ratio value ranged
from 3 to 6 along the two manufacturing directions, followed by heating
until the softening temperature was reached. The base foam was supplied
in squared blocks of 600 mm side by 50 mm thickness. Cylindrical samples
were obtained using a sharp-edge tube which was penetrated into the block
of conventional foam. These samples were then inserted inside lubricated
tubes, using a conical inlet to avoid creases. The tubes containing the parent
foam were then positioned in the mould frame and a piston pushed inside it.
The locking of the piston in the desired position followed compression. The
whole operation was then repeated for the other specimens. The mould was
then put into an industrial oven, and two heating profi les
(15)
were used. The
heating process was followed by room temperature cooling, which lasted two
hours on average. After this operation, the extracted specimens were gently
tensioned in order to relax the external surface. Following a procedure outlined
in
(15)
, the general designation of the samples was defi ned by an “XY” code,
where “X” was the position of the specimen in the mould and “Y” was the
progressive letter assigned to every specimen. Samples with higher uniformity
of mechanical properties belonged to the specimens C and L (parameters shown
in Table 1). For comparison, specimens with no imposed radial compression
were also tested. These latter manufacturing parameters were labelled as
“M” specimens. The specimens with no imposed radial load were produced
as a control sample, to check against the possibility that measured changes in
mechanical and transmissibility properties of the auxetic foams were due to
the increase in foam density.
In a manner similar to the procedure outlined in
(14,15)
, Poisson’s ratio measurements
were based on image data acquired with a SONY DCR-TRV50E digital camera
and processed with the MATLAB v6.1 software. The images were acquired
Table 1. Compression ratios used for the different auxetic and iso-density
foams
Sample series Compression ratio (%)
Radial Axial Volumetric
3C 36.7 46.9 78.7
5C 36.7 29.2 71.6
2L 62.0 50.0 92.8
5M 0 68.8 68.8
CP 5 05.indb 256CP 5 05.indb 256 27/10/05 12:11:10 pm27/10/05 12:11:10 pm
257
Cellular Polymers, Vol. 24, No. 5, 2005
Mechanical Performance of Auxetic Polyurethane Foam for Antivibration Glove Applications
during quasi-static tests performed using an MTS 858 servo-hydraulic machine.
The Poisson ratio measures were carried out using specimens which were cut
to 40 mm in length and glued to the end clamp of the machine with Loctite
®
adhesive, then stretched until breaking or ungluing at 0.1 mm/s (e = 0.0025 s
-1
).
A calibration photo was taken for each test in order to convert the units of
measurement from pixels to millimetres. A ruler was placed at a set distance
from the camera iris, and the conversion factor was determined geometrically
using the known distance of the ruler from the camera and the nominal focal
length of the camera lens. After calibration, a set of photos was acquired to
measure longitudinal and transverse strain during testing. This operation was
performed using a MATLAB routine which calculated ten values of diameter
and ten values of length along the specimen from every photograph, providing
as output both the mean value and the standard deviation. To avoid end effects
during the measurements, the extreme thirds of the dark-grey specimens were
painted white in order to consider only the central third in the calculation of
the lengths.
The MTS 858 servo-hydraulic system used for the materials properties testing
included a cross-head-mounted actuator, a hydraulic power unit and a test
controller with two channels of control (force and displacement). The load
unit is fatigue-rated at 25 kN, and can be operated at frequencies up to 30 Hz,
with an integral crosshead-mounted linear actuator with an attached manifold.
The MTS machine was used to perform tensile test (Poisson’s ratio measures)
and quasi-static compression tests on specimens 5C, 3C, 2L, 5M and parent
foam samples.
Quasi-static compression tests were carried out before the Poisson’s ratio testing.
A gap of 40 mm was left between the clamp ends of the machine, and samples cut
to 40 mm in length were placed between them, lubricating the contact surfaces to
minimize friction. The speed of the head was 0.01 mms
-1
, indicating a strain-rate
of 0.0025 s
-1
, with maximum compressive strain reached at 75%. Data were
acquired in displacement-crossing mode, acquiring displacement and force
every 0.1 mm. From the knowledge of the initial sizes of the specimens, it was
possible to derive a strain-stress curve for every case. All tests were carried
out at room temperature of 20 ºC. No control of humidity was performed
during the tests.
ISO 13753 TRANSMISSIBILITY TESTS
ISO 13753:1999 requires the measurement of the transfer function of the
resilient material across the frequency range from 10 to 500 Hz while loaded
CP 5 05.indb 257CP 5 05.indb 257 27/10/05 12:11:11 pm27/10/05 12:11:11 pm
