Advanced Materials Research Vols 332-334 (2011) pp 697-701 Online: 2011-09-02

© (2011) Trans Tech Publications, Switzerland

doi:10.4028/www.scientific.net/AMR.332-334.697

Transverse Compression Properties of Textile Materials

Piaras Alphonsus Kelly1,a, Simon Bickerton2,b and Jonathan Cheng1,c

1
Department of Engineering Science, School of Engineering, The University of Auckland, Private
Bag 92019, Auckland Mail Centre, Auckland 1142, New Zealand
2
Department of Mechanical Engineering, School of Engineering, The University of Auckland, Private
Bag 92019, Auckland Mail Centre, Auckland 1142, New Zealand
a b c
[email protected], [email protected], [email protected]

Keywords: Textile Materials, Compression, Mechanical Properties, Relaxation, Viscoplasticity.

Abstract. The response of textile materials to transverse compression is of great importance in many
applications. In this paper is discussed the mechanical properties of textile materials when subjected
to transverse compression. The response of a textile to load is very non-linear and inelastic. The
response is viscoelastic (there are rate effects) and plastic (there are permanent deformations). It is
shown that energy is stored in a textile when it is loaded, and that some of this energy is not released
when the material is unloaded, but instead is locked into its structure. This locked energy cannot be
released unless the structure is placed in tension. A thermomechanical framework is introduced
which incorporates the textile locked energy. Models are developed which predict well the response
of textiles to load.

Introduction
Textile materials are subjected to compression in a variety of technical and industrial applications [1],
and knowledge of their mechanical response in such circumstances is of great importance [2]. Many
models have been devised to capture and predict the response of textiles to load. Most of this work has
been carried out on the single compaction curve; this is the curve of stress versus volume fraction
during a single compaction. For example, Van Wyk [3] considered a random assembly of fibres and
assumed that the fibres displace under force according to the elastic beam theory, and hence derived
the power law relation

σ f = A(V f3 − V f30 ), (1)

where σ f is the compaction stress, V f is the fibre volume fraction, V f 0 is the free fibre volume
fraction (volume fraction at zero stress) and A is a material parameter related to the stiffness of the
fibres. Many other researchers have followed Wan Wyk’s methodology, making refinemenst to the
basic model [4-7]. Another micromechanical approach, particularly useful in the study of knitted and
woven fabrics, is to analyse the response of individual yarns and tows. These meso-scale models treat
yarns as deforming continuum solids and no allowance is made for the movement of individual fibres
withn yarns [8,9]. Due to the difficulty of incorporating the micro and meso effects in any model,
many simple empirical models have been devised to capture the single compaction curve (see, for
example, [10-12]).

The models just mentioned are all essentially elastic models, that is, the stress depends on the current
volume fraction, and is independent of the history of loading. However, textile materials are highly
inelastic. This inelastic response of textiles is discussed in the following section. After that, the
concept of locked energy is introduced and, finally, a mathematical model of a textile material
incorporating this locked energy is introduced. Results are given which show the effectiveness of the
model.

All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of Trans
Tech Publications, www.ttp.net. (ID: 152.14.136.77, NCSU North Carolina State University, Raleigh, USA-10/05/15,13:58:47)
698 Advanced Textile Materials

Inelastic effects in textile materials

During transverse compression, a number of micro-level deformation modes occur apart from fibre
bending, leading to inelastic effects. These include fibre re-arrangment and fibre sliding. The effect of
these inelastic interactions can be seen in Fig. 1, which shows the response of a sample of 10 layers of
E-glass Plain Weave Fabric, areal mass 232.8 g/m2, to a cyclic test, repeatedly loaded to a fixed
volume fraction of 60%, plotted on a graph of compaction stress versus volume fraction. This
response is typical of any textile, whatever its architecture or constituents. A notable feature here is
the cyclic softening, that is, the reduction in the stress required to reach the target volume fraction
with each subsequent loading cycle. The unloading curve falls below the compaction curve for any
given cycle, forming a hysteresis loop. With each subsequent cycle, the hysteresis loops decrease in
size and grow closer, until eventually a single asymptotic stabilised equilibrium hysteresis loop is
reached and repeats indefinitely; this usually occurs after about ten cycles. The free volume fraction,
that is, the volume fraction at zero stress, for this material was initially measured to be V f 0 = 0.251 .
This free volume fraction increases with cycling, as indicated in Fig 15.1, until it reaches an
equilibrium free volume fraction of V f 0 = 0.377 corresponding to the equilibrium hysteresis loop.

Figure 1: Cyclic loading of a random mat to 38% volume fraction; cycles 1 and 20 are shown for
compaction speeds of 1 mm/min and 25 mm/min.

This permanent deformation, or Permanent Set (PS), can be very significant in textile deformation.
For example, Susich & Backer [13] have shown that it arises for a wide range of fibres in tension,
including glass (3% of total deformation during loading is PS upon unloading ), polypropylene ( 7%)
and acrylic (23%). Somashekar et al. [14] carried out similar recovery tests on fabrics, showing
clearly the large inelastic effects which arise when textiles are subjected to compression (with PS up
to 36% reported).
The term plastic is used to denote a material which undergoes rate-independent permanent
deformation. The term is strongly associated with engineering materials such as metals; however,
fibrous materials respond quite differently to the standard engineering materials. For example, metals
“flow” with ever-increasing strain when stressed highly enough; fibrous materials, on the other hand,
display the opposite effect of the strain reaching a limit with the associated stress increasing
indefinitely.
 Advanced Materials Research Vols. 332-334 699

Locked Energy. When fibrous materials are compacted and then unloaded, the fibres will not
un-bend to their original configurations due to the frictional constraints at the contacts. This implies
that the elastic energy due to bending is not all released during unloading; some of this energy is
locked into the sample. This concept of locked strain energy was noted by Grosberg [15] in a
calculation of the fibre withdrawal force from a sample of fibres.
With negligible change to preform cross-sectional area, and assuming a quasi-static process,
the work done W1 (per unit un-compacted volume V0 ) in compacting a sample from an initial
thickness h0 to a final thickness hmin can be calculated from

hmin V fmax
1 σ
W1 = −
V0 ∫ F dh = V ∫ V
h0
f0
Vf 0 f
2
dV f . (2)

Here, V f 0 is the initial un-compacted free volume fraction, V fmax is the final volume fraction, F is the
varying compaction force and σ is the compaction stress. A similar calculation gives the work W2
done during unloading. The percentage work differences (W1 − W2 ) / W1 × 100 was calculated for a
number of materials, and it reaches up to 75%. The work difference was significant even after many
cycles. The work difference W1 − W2 is conventionally considered to be the energy loss (per unit
un-compacted volume) during the cycle (see, for example, [16]). However, this is not appropriate for
textiles, because some of this energy is not irreversibly lost. As mentioned, some of this energy
remains locked in the fibres, and cannot be accessed without applying tension to the material. Thus,

W1 − W2 = (W1s − W2 s ) + (W1d − W2 d ) (3)

where the subscript s denotes “stored” and the subscript d denotes “dissipated”. The term inside the
first parentheses on the right hand side of Eqn. 3 is the locked energy. The term inside the second
parentheses gives the energy loss. The energy loss during compaction is ( W1d ) and there is also the
possibility of energy loss during unloading ( W2 d ). A number of models have been used to describe
the permanent deformations occuring in textile materials (see, for example, [17-19]). However, none
of these models account for the locked energy. In the next section is introduced and discussed a
method by which this locked energy can be considered and used to predict the repsonse of textiles.

A Thermomechanical Model of Textile Materials

A convenient approach to take when modelling fibrous materials is to use a thermomechanical
framework, i.e. to build a model up from thermodynamic principles [20]. This is because (1) the laws
of thermodynamics will automatically be satisfied, and (2) the framework is simple, in the sense that
only a number of principles are necessary, and it is not necessary to develop an ad-hoc theory
involving separate yield criteria, flow rules, etc. On the contrary, one need only specify two energy
potentials. Following the thermomechanical procedure outlined in, for example, Ziegler [21], the rate
at which work is done by the external forces can be expressed as

σ :d = Ψ

+Φ, (4)

where σ is the stress, d is the rate of deformation, Ψ is the Helmholtz free energy, a measure of the
energy stored, and Φ is the dissipation, the rate at which energy is lost. One is required to specify the
forms of these two energy functions. The free energy and dissipation then act as potentials from
which the stresses and strains can be determined [22].
700 Advanced Textile Materials

Regarding the hypothesising of functional forms for Ψ and Φ , the following two key points are
to be made concerning fibrous materials in general:
(i) they have an ability to acquire locked energy, as discussed earlier; this energy cannot be accessed
without a reversal of any permanent deformation.
(ii) they are frictional materials, i.e. energy is dissipated through a frictional mechanism.

Point (i) implies that the free energy can be expressed in the general form

Ψ (e, α ) = Ψ1 (e,α ) + Hˆ (α ) , (5)

with the second term here being the locked energy, a function of permanent deformations, as
described by the internal variable α , but not of the current strain e. Point (ii) implies that the
dissipation is of the form

Φ(σ,α,α
) , (6)

that is, it depends explicitly on the stress [23].

This approach can be used to build up models of increasing complexity [24]. For example, in
[20] is described a model with power-law free energy functions and with the dissipation of the form
Φ = fσα
, where f is constant, playing the role of a friction coefficient. This leads to plastic
deformation hysteresis. Fig. 2 shows model results for a single-internal variable model for the
equilibrium hysteresis loop for a plain weave fabric. Here, dissipation is continuous during loading
and unloading, with the amount of locked energy and permanent deformation returning to their
original values after a complete loop; results compare well with cyclic experimental results for these
materials subjected to at least 20 cycles of loading.

Figure 2: Hysteresis loop: experimental and model data for a plain weave fabric
 Advanced Materials Research Vols. 332-334 701

References
[1] S. Jaganathan, H. Vahedi Tafreshi, E. Shim and Pourdeyhimi, Colloids and Surf A, Vol. 337
(2009), p. 173.
[2] V.K. Kothari and A. Das, Geotextiles and Geomemb, Vol. 11 (1992), p. 235.
[3] C.M. Van Wyk, J Text Inst, Vol. 37 (1946), p. T285.
[4] A. Stearn, J Text Inst, Vol. 62 (1971), p. 353.
[5] T. Komori and Makishima, Text Res J, Vol. 47 (1977), p. 13.
[6] D.H. Lee and G. Carnaby, Text Res J, Vol. 62 (1992), p. 185.
[7] S. Toll and J-A. E.Manson, J Appl Mech, Vol. 62 (1995), p. 223.
[8] K.F. Choi and S.K. Tandon, J Text Inst, Vol. 97 (2006), p. 49.
[9] S.V. Lomov, G. Huysmans, Y. Luo, R.S. Parnas, A. Prodromou, I. Verpoest and F.R. Phelan,
Comp Part A, Vo.. 32 (2001), p. 1379.
[10] E. Sebestyen and T.S. Hickie, J Text Inst, Vol. 62 (1971), p. 353.
[11] B. Neckář , Text Res J, Vol. 67 (1997), p. 123.
[12] P.A. Kelly, J Text Inst (2011).
[13] G. Susich and S. Backer, Text Res J, Vol. 21, (1951) p. 482.
[14] A.A. Somashekar, S. Bickerton and D. Bhattacharyya, Comp Part A, Vol. 37 (2006), p. 858.
[15] P. Grosberg, J Text Inst, Vol. 54 (1963), p. T223.
[16] D. Das D and Pourdeyhimi, Ind J Fibre Text Res, Vol. 35 (2010), p. 303.
[17] B. Olofsson B, J Text Inst, Vol. 58 (1967), p. 221.
[18] J.I. Dunlop, J Text Inst, Vol. 74 (1983), p.92.
[19] H. Lin, M. Sherburn, J. Crookston, A.C. Long, M.J. Clifford and I.A. Jones, Mod Sim Mater Sci
Eng, Vol. 16 (2008), p.1.
[20] P.A. Kelly, FPCM-9, Montreal, 8-10 July.
[21] H. Ziegler: An Introduction to Thermomechanics (North-Holland pub., Groningen, 1977).
[22] I.F. Collins and G.T. Houlsby, Proc. Royal Society London Series A, Vol. 453 (1997), p. 1975.
[23] I.F. Collins and P.A. Kelly, Geotechnique, Vol. 52 (2002), p. 507.
[24] J. Cheng, P.A. Kelly and S. Bickerton, J Comp Mat, (2011).
Advanced Textile Materials
10.4028/www.scientific.net/AMR.332-334

Transverse Compression Properties of Textile Materials

10.4028/www.scientific.net/AMR.332-334.697