Cite as: Shah DU, Schubel PJ. Journal of Reinforced Plastics and Composites. 2016, In press.
On recycled carbon fibre composites manufactured
through a liquid composite moulding process
Darshil U. Shah1*, Peter J. Schubel2
1
Centre for Natural Material Innovation Dept. of Architecture, University of Cambridge, Cambridge
CB2 1PX, UK
2Polymer
Composites Group, Division of Materials, Mechanics and Structures, Faculty of
Engineering, The University of Nottingham, Nottingham, NG7 2RD, UK
*Corresponding author. E-mail: dus20@cam.ac.uk; darshil.shah@hotmail.co.uk; Tel: +44
(0)1223760124.
Abstract
The recovery of carbon fibres from waste and end-of-life carbon fibre reinforced
plastic materials is both economically lucrative and environmentally necessary. Here, we
characterise the physical and mechanical properties of recycled carbon fibre reinforced
plastics (rCFRPs) composed of random and oriented non-woven recycled carbon fibre mats
that were impregnated with liquid epoxy matrices using a vacuum-infusion set-up. The low
areal density and poor compactability of the non-woven mats implied that press-moulding
upon impregnation was essential to control laminate thickness and improve fibre content; this
may limit the applications of the resulting rCFRPs. Moreover, the press consolidation process
is thought to degrade fibre length, and is a likely cause for the lower-than-expected tensile
properties of the rCFRPs. Expectedly, the oriented rCFRPs exhibited better tensile and
compressive properties than the random rCFRPs. Notably, while the tensile strength of the
rCFRPs was only up to 2.5 times better than the matrix, the tensile modulus was 4-10 times
enhanced. Through a comparative literature survey, we found that the liquid composite
moulded rCFRPs were outperformed by rCFRPs fabricated through other manufacturing
processes (e.g. prepregging), particularly those employing high compaction pressures, and
utilising long fibres recovered through pyrolysis and chemical processes, rather than the
fluidised-bed process.
Keywords: composite recycling; fluidised bed process; liquid composite moulding; carbon
fibre reinforced plastics; recycled carbon fibre
Page 1 of 16
1 Introduction
Composite materials can offer significant weight savings and performance advantages
over monolithic materials. This, alongside the former’s potential for complex shape forming,
which in turn facilitates optimised product design, has propelled their increasing usage in
high technology sectors. The global fibre reinforced plastics (FRPs) market was at 8.7
Mtonnes in 2011, projected to grow at about 5% per annum up to 2015 [1]. However, the
limited end-of-life re-use, recycling and disposal options of FRPs, particularly thermosetbased (which accounted for 63% of the global FRP market by volume, in 2011 [1]), is seen as
a key barrier in their continued use in some markets [2, 3].
Traditionally, the principal waste management options that have been adopted for
composites are landfilling and incineration (in which the resulting ash is landfilled) [2-5].
However, ever-stringent government legislations (e.g. EU Directive on Landfill of Waste
(Directive 99/31/EC), End-of-life Vehicle Directive (Directive 2000/53/EC), EU Directive on
Waste Incineration (Directive 2000/76/EC)) discourage, restrict, penalise, and in some
countries even ban such practices [6]. Consequently, to alleviate some of the environmental
issues associated with FRPs, over the past decade or so there have been two areas of
particular interest: i) developing recycling methods and strategies for conventional FRPs [2-4,
7], and ii) replacing synthetic constituent materials in conventional FRPs with bio-based (if
not fully green) fibres and matrices [8-11].
While recycling composites is challenging, various approaches (categorised into
mechanical, thermal and chemical routes) have been pursued, many of which have reached
commercial readiness [2, 3, 12]. Notably, while glass fibre composites (GFRPs) account for a
sizeable fraction of the global FRP market (87% by volume in 2011 [1]), there is little
economic incentive to recover glass fibres from GFRP waste due to the low cost of the virgin
fibre and limited applications of the recyclate. Rather, it is more viable to mechanically
recycle GFRPs and dispose them in cement kilns [12]. In contrast, recovering carbon fibres
from CFRP waste through thermal and chemical recycling methods is very attractive due to
the high value of the virgin and recycled fibre, almost maintained mechanical properties
(post-reclamation) of the carbon fibres due to their high thermal and chemical stability, and
wide range of applications, both non-structural and structural, of the recyclate [13].
The reclaimed carbon fibres are re-impregnated to manufacture recycled CFRPs
(rCFRPs). While there have been several in-depth studies on the three main fibre reclamation
Page 2 of 16
processes (pyrolysis, fluidised bed process, chemical process; reviewed in [2, 13]), there have
been fewer studies on the mechanical properties of the resulting rCFRPs. Part of the
bottleneck has been re-processing the recovered fibres, which are often filamentised,
entangled, and de-sized, into intermediate products (e.g. mats, tows), and then converting
these into aligned, high fibre fraction composites using traditional manufacturing processes
[12-14]. Nonetheless, a variety of manufacturing methods have been trialled to fabricate
rCFRPs; including, injection moulding [15], bulk moulding compounds (BMCs) and sheet
moulded compounds (SMCs) [16, 17], resin film infusion [14, 18], compression moulding
[19], and (out-of-autoclave and autoclave) prepregging [14, 16, 17]. However, all of these
studies incorporate the matrix material in solid form (e.g. pellets, sheets, prepregs). There are
limited studies (e.g. [20]) investigating the flow of liquid resin through the recycled carbon
fibre intermediate product, or characterising the resulting rCFRPs.
Here, we characterise the physical and mechanical properties of rCFRPs composed of
random and oriented non-woven recycled carbon fibre mats that were impregnated with
liquid epoxy matrices using a vacuum-infusion set-up, and thereafter press-moulded to
control laminate thickness and improve fibre content.
2 Experiments
2.1
Materials
For this study, two forms of recycled carbon fibre non-woven mats were employed as
reinforcement materials: mats with nominally random orientation (90.2 ± 1.6 gsm) and partial
alignment (24.3 ± 0.3 gsm; with 80% unidirectional alignment). These mats were produced
through a proprietary technique based on a wet-laid, paper-making process [16, 21]. The mats
composed of discontinuous carbon fibres (up to 10 mm in length) that were reclaimed from
mechanically-resized CFRP scrap through a fluidised bed process [2, 12, 22, 23]. For
reference, the recyclate had a tensile modulus Ef, tensile strength σf and interfacial shear
strength τ (with epoxy resin) of 218 GPa, 3.2 GPa and 62 MPa, respectively. The reclamation
method maintained the stiffness and interfacial properties of the virgin carbon fibre, but
reduced the tensile strength by over 25%. In addition, there was no change in fibre diameter d
(~7 μm) or density (1.74 g/cm3). The critical fibre length lc of the recycled carbon fibres was
determined to be ~180 μm using Eq. 1.
lc 
fd
2
Page 3 of 16
Eq. 1
Two low-viscosity (~100 mPa·s at 40 °C) thermosetting epoxy matrices, namely
RIMR135/RIMH137 and RIMR035c/RIMH038, were obtained from Hexion Specialty
Chemical GmbH (Germany). The RIMR135/RIMH137 system is a standard matrix for the
wind energy industry, with high mechanical properties (tensile stiffness Em and strength σm of
2.7-3.2 GPa and 70 MPa, respectively). On the other hand, the RIMR035c/RIMH038 system
is an attractive new alternative due to its comparable mechanical properties (tensile stiffness
Em and strength σm of 2.9 GPa and 70 MPa, respectively), significantly lower exotherm and
better cost-to-performance ratio.
2.2
Composite manufacture
Flat composite plaques (300 mm square) were manufactured by first impregnating a
constant mass of reinforcement material with the preheated (40 °C for 30 min) resin system
through a vacuum-bagging process at 100 ± 25 mbar absolute pressure (Fig. 1). A line gate
injection configuration was employed (Fig. 1c), with flow parallel to the fibres in the oriented
mat. Interestingly, all preforms (nominally-random and partially-oriented) exhibited a wetout time between 13 and 15 minutes, although on average the random mat preforms required
slightly lower infusion times. The use of a distribution medium ensured that the flow front
evolution was axial and gradual. The impregnated reinforcement was then hot (85 °C for 150
min) press-moulded at 2.5 MPa to achieve a nominal laminate thickness of ca. 3.0-3.3 mm,
corresponding to a nominal fibre volume fraction of ~30%.
Six different types of composites were prepared (Table) to test primarily the effects of
reinforcement mat form/orientation and matrix system type. Samples 1 and 3 also enabled
studying the anisotropy ratio in partially-aligned rCFRPs. In addition, Sample 6 was
fabricated without applying any compaction pressure to study properties achieved through a
conventional vacuum infusion manufacturing process; note that mechanical tests on this
sample could not be conducted due to its large thickness (of ~14 mm).
2.3
Property characterisation
2.4
Physical properties
The fibre weight fraction wf of a laminate was calculated using the ratio of the mass of
the reinforcement material and the resulting composite laminate. The composite density ρc
was measured using a Micromeritics AccuPyc 1330 helium pycnometer (n = 5). The fibre
and matrix densities were taken as 1.74 g/cm3 and 1.15 g/cm3, respectively. The composite
fibre volume fraction vf and porosity content vp were then determined using Eq. 2.
Page 4 of 16
vf 
2.5
c

w f ; v m  c (1  w f ); v p  1  (v f  v m )
f
m
Eq. 2
Mechanical testing
Tensile tests were conducted as per ISO 527-4:1997 using an Instron 5985 testing
machine equipped with a 50 kN load cell and a 50 mm clip-on extensometer. At least five
specimens were tested for each type of composite at a cross-head speed of 2 mm/min. The
tensile modulus Ec, ultimate tensile strength σc, and tensile failure strain εc were measured
from the stress-strain curve.
Compression tests were conducted according to ASTM D3410, on an Instron 5581
testing machine equipped with a 50 kN load cell and an IITRI compression test fixture. At
least five specimens were tested at a cross-head speed of 1 mm/min. Specimens were speckle
coated for video strain measurement. The compressive modulus, ultimate compressive
strength, and Poisson’s ratio were measured.
Short-beam shear tests were carried out according to ASTM D2344, where unnotched specimens were loaded in a three-point bending configuration at a cross-head speed
of 1 mm/min. An Instron 5969 testing machine equipped with a 5 kN load cell was used for
these tests. The ‘apparent’ interlaminar shear strength τ was calculated using Eq. 3, where P
is the maximum applied load. At least five specimens were tested for each type of composite.

3 P
4 bt
Eq. 3
3 Results and discussion
3.1
Volumetric composition and density
The physical properties of the fabricated rCFRPs are presented in Table and Fig. 2.
The density of the all the rCFRPs ranged between 1.19-1.31 g/cm3, owing to the high
proportion of the lower density resin. The fibre volume fraction ranged from 25% to 34% for
the press-moulded samples, and was only 8.5% for Sample 6 which was not press-moulded.
This indicated, firstly that high pressure compaction is necessary to produce rCFRPs with
reasonable fibre content. This is particularly because of the low packing density of the
recycled carbon fibre mats employed in this study, resulting from the discontinuous and
fluffy nature of the recyclate (reclaimed through the fluidised bed process), and the fibre
misalignments in the mats (produced through a modified paper-making process). We also
Page 5 of 16
observed that as the reinforcement mats have a low areal density (25-100 gsm), a number of
layers are required to produce the composites. Subsequently, compaction is necessary also to
reduce and control laminate thickness (Fig. 2). Notably, carbon fibre reinforcements
reclaimed from other methods, such as pyrolysis, can be consolidated into composites with
thickness and fibre volume fractions of 2-3 mm and 40-60%, respectively, even at no
compaction pressure [18].
The low compactibility of recycled carbon fibre mats fabricated through the wet-laid
process has been raised previously in literature [12, 16, 17, 24]. Consolidation through
vacuum alone produces fibre volume fractions of ca. 10-12% [17, 24], similar to our results.
Autoclave and press consolidation at pressures between 1 and 10 MPa can yield higher fibre
volume fractions ranging between 10% and 65% [12, 17, 24]. However, increasing
compaction pressures also introduces, in some cases substantial, fibre breakage [14, 19, 24].
There are contrasting reports on the ratio of cost (from fibre length reduction) to benefit
(from fibre volume fraction increase) on mechanical properties of the resulting rCFRPs, with
some suggesting a net advantage of high-pressure compaction [24], while others reporting
considerable degradation of composite strength [14]. Therefore, it is of interest to develop
processes to improve the compaction characteristics of the mats, ensuring reduced fibre
breakage. Possible routes include improving fibre alignment in the mats, as aligned mats
show significantly higher compactability [12, 16, 17, 24] and are likely to exhibit reduced
fibre breakage due to fewer fibre-fibre cross-over points and reduced fibre bending
deformation during consolidation. Indeed, our results showed that aligned rCFRPs (Samples
1, 3 and 4) had relatively higher fibre fractions than nominally random rCFRPs (Samples 2
and 5) (Fig. 2). Some researchers have also successfully trialled the application of (multiple)
precompaction cycles of the reinforcement mat (typically, at pressures higher than the
moulding pressure) prior to composite lay-up [12, 24]. This approach may be more suitable
for the manufacture of larger composite products (such as wind turbine components), where
the application of high consolidation pressures during or post resin impregnation may not be
practical. We also suggest the development of mats of higher areal density (closer to 300-800
gsm rather than the current 25-100 gsm) for the manufacture of high fibre content composites
without the use of a press.
Regarding porosity content, our measurements (Fig. 2) revealed that Sample 6, which
was not pressed, had the lowest void content of 1%. In contrast, Samples 1,4 and 5 had void
contents of 3-4%, while Samples 2 and 3 had much higher void fractions of around 9%.
Page 6 of 16
Therefore, we believe that while the vacuum infusion technique employed allows the
manufacture of low-void content composites, perhaps using the press to compact the infused
preform has a role in the higher void content. There was no clear effect of matrix type or
reinforcement form on void content. Notably, void contents of 3-8% (even upto 21% [18])
are commonly reported for press-moulded rCFRPs [12, 14, 25], which may be too high for
components where high quality control is required.
3.2
Mechanical properties
The measured mechanical properties of the samples are presented in Table 1. Physical
properties of the fabricated rCFRPs.
Sampl
e ID
1
2
3
4
5
6*
Nonwove
n mat
type
Partially
aligned
Nominally
random
Partially
aligned
Partially
aligned
Nominally
random
Partially
aligned
Testing
directio
n
0°
90°
0°
0°
Matrix system
Laminat
e
thickness
[mm]
Composit
e density
[gcm-3]
RIM135/RIMH13
7
RIM135/RIMH13
7
RIM135/RIMH13
7
RIM035/RIMH03
8
RIM035/RIMH03
8
RIM135/RIMH13
7
3.04 ±
0.14
3.22 ±
0.25
3.02 ±
0.23
3.17 ±
0.27
3.29 ±
0.40
1.31 ±
0.01
1.19 ±
0.02
1.22 ±
0.02
1.27 ±
0.01
1.25 ±
0.01
1.19 ±
0.01
~14
Fibre
volume
fractio
n
[%]
33.9 ±
0.4
25.6 ±
0.8
28.1 ±
0.6
26.6 ±
0.3
24.8 ±
0.3
8.5 ±
0.2
Porosit
y
content
[%]
3.2 ±
0.5
9.5 ±
1.9
9.0 ±
2.2
3.0 ±
0.6
3.8 ±
0.6
1.0 ±
0.4
* No press-moulding stage.
Table 1.
3.2.1
Effect of reinforcement alignment
Comparing Samples 1 and 4 with Samples 2 and 5, respectively, it was obvious that
the oriented rCFRPs performed much better than the random rCFRPs in terms of tensile
properties. Noting the differences in volumetric composition, the oriented samples had 1.32.1 times higher tensile strength and 1.5-2.4 times higher tensile modulus. However, the
oriented rCFRPs also exhibit lower tensile fracture strains. Nonetheless, as the tensile
modulus of the matrix was about 3 GPa, the reinforcing effect of the stiff carbon fibres was
clearly observed in the rCFRPs with tensile stiffness ranging between 12-32 GPa. Notably,
the tensile strength of the random rCFRPs (85-120 MPa) was not substantially higher than
Page 7 of 16
that of the matrix (~70 MPa), while aligned rCFRPs exhibited over two times the matrix
tensile strength.
The back-calculated (using the ‘rule of mixtures’ in Eq. 4 [8, 26]) fibre tensile
modulus Ef ranged from 60-130 GPa and 110-120 GPa, for the partially-oriented (fibre
orientation distribution factor η = 0.875 (= 1∙0.8 + 0.2∙0.375) for 80% unidirectional fibres
and 20% randomly-oriented fibres) and nominally randomly-oriented (η = 0.375) rCFRPs,
respectively. This was 40-70% smaller than the true elastic modulus of the recycled carbon
fibres (218 GPa). The discrepancy may be accounted by a reduction in length distribution
factor (implicitly assumed to be unity in Eq. 3). Although the reinforcement mats employed
fibres up to 10 mm long, the high moulding pressure may have led to severe fibre damage,
consequently degrading fibre length. For instance, Pimenta et al. [14] and Wong et al. [19]
found that compression moulding of similar recycled carbon fibre reinforcement mats at 7
MPa resulted in less than 40% of fibres longer than the critical length (calculated to be 180
μm in Section 2.1); this would result in length distribution factors of less than 0.5, which
would possibly account for most, if not all, of the discrepancy between back-calculated and
true fibre moduli. Porosity, particularly for Samples 2 and 3, may also explain some of the
discrepancy, although this is unclear as despite having a substantially higher void content
Sample 3 has higher tensile properties than Sample 5.
Ef 
E composite  v m E m
v f
Eq. 4
Compression tests revealed that the oriented rCFRPs had a ~15% smaller compressive
strength (due to easier kink band formation), but a 10-25% higher compressive modulus than
the random rCFRPs. In addition, the Possion’s ratio of the oriented rCFRPs was ~75% higher
than the random rCFRPs, suggesting that the latter expectedly exhibit less lateral expansion
when compressed.
The apparent inter-laminar shear strength of the composites ranged between 17 and 43
MPa, which is comparable to that of chopped strand E-glass/epoxy composites (~30 MPa)
[27]. However, there was no clear trend on the effect of fibre alignment on shear strength.
Anisotropy in aligned rCFRPs
We also found that although the aligned rCFRPs tested in the longitudinal direction (Sample
(Sample 1) had tensile and compressive properties higher than those tested in the transverse
Page 8 of 16
direction (Sample 3), the difference in properties was not substantial (Table 1. Physical
properties of the fabricated rCFRPs.
Sampl
e ID
1
2
3
4
5
6*
Nonwove
n mat
type
Partially
aligned
Nominally
random
Partially
aligned
Partially
aligned
Nominally
random
Partially
aligned
Testing
directio
n
0°
90°
0°
0°
Matrix system
Laminat
e
thickness
[mm]
Composit
e density
[gcm-3]
RIM135/RIMH13
7
RIM135/RIMH13
7
RIM135/RIMH13
7
RIM035/RIMH03
8
RIM035/RIMH03
8
RIM135/RIMH13
7
3.04 ±
0.14
3.22 ±
0.25
3.02 ±
0.23
3.17 ±
0.27
3.29 ±
0.40
1.31 ±
0.01
1.19 ±
0.02
1.22 ±
0.02
1.27 ±
0.01
1.25 ±
0.01
1.19 ±
0.01
~14
Fibre
volume
fractio
n
[%]
33.9 ±
0.4
25.6 ±
0.8
28.1 ±
0.6
26.6 ±
0.3
24.8 ±
0.3
8.5 ±
0.2
Porosit
y
content
[%]
3.2 ±
0.5
9.5 ±
1.9
9.0 ±
2.2
3.0 ±
0.6
3.8 ±
0.6
1.0 ±
0.4
* No press-moulding stage.
Table 1). For instance, even in the transverse direction, the tensile modulus of the
aligned rCFRPs was five times that of the cured resin, and higher than that of the randomlyoriented rCFRPs. Typically, for unidirectional composites, the tensile modulus in the
transverse direction is similar to the matrix tensile modulus [26, 28]. This confirmed that the
oriented mats were only partially-oriented. Interestingly, the Possion’s ratios of the aligned
rCFRPs were very different in the longitudinal (0.48 ± 0.10) and transverse direction (0.08 ±
0.05). Indeed, this is typical of composites in which fibres are perpendicular to the loading
direction. While the former will show significant lateral expansion when compressed, the
latter will show negligible lateral expansion when compressed.
3.2.2
Effect of matrix type
Comparing Samples 1 and 2 with Samples 4 and 5, respectively, enabled evaluating
the effect of the matrix type on the composite properties. Despite having lower fibre volume
fractions and comparable void content (except for Sample 2), it was observed that the tensile
modulus of composites made from RIM035/RIMH038 was 1.1-1.7 times higher than that of
RIM135/RIMH137 composites. This was surprising as both resins have fairly similar tensile
properties (modulus included). On the other hand, tensile strength didn’t show any clear trend
with respect to matrix type. rCFRPs employing RIM135/RIMH137 had a 1.2-1.4 times
higher failure strain than composites employing RIM035/RIMH038 resin. This was expected
Page 9 of 16
as the failure strain of RIM135/RIMH137 resin (8-12%) was higher than that of
RIM035/RIMH038 resin (8%).
Interestingly, rCFRPs made from RIM135/RIMH137 had a ~15% higher compressive
strength, 25-40% higher compressive modulus and ~50% higher Poisson’s ratio than rCFRPs
employing RIM035/RIMH038. However, in terms of inter-laminar shear strength, rCFRPs
based on RIM035/RIMH038 performed better, suggesting better interfacial adhesion.
3.2.3
Comparison with other manufacturing processes in literature
Table 2 compares the properties of the liquid composite moulded rCFRPs in this
study with properties reported in literatures on rCFRPs manufactured through other
processes. Intriguingly, most of the processes employ external pressures (of up to 10 MPa)
for consolidation to achieve higher fibre volume fractions.
Our random mat rCFRPs had tensile modulus and strength comparable to that of the
injection moulded rCFRPs (with nominally 3D-random fibre orientation) fabricated by
Connor [15], and bulk moulded compound panels (with nominal 2D-random fibre
orientation) produced by Turner et al. [17]. Random mat composites produced by
compression moulding at high consolidation pressures of 5-10 MPa had only slightly higher
fibre fractions that our random mat rCFRPs (fabricated at 2.5 MPa), but had double the
tensile stiffness and strength. This was an interesting result as one would envisage greater
fibre damage and length degradation in the compression moulded rCFRPs, and therefore
reduced performance [19].
Our aligned mat rCFRPs exhibited tensile properties comparable to those fabricated
by resin film infusion, although much lower than those manufactured using prepregging and
through sheet moulding compounds. Notably, the highest mechanical properties ever reported
for rCFRPs are based on prepregging with press-moulding at 5-10 MPa [17], where aligned
non-woven mats, similar to the ones used in this study, are the reinforcements. Carbon fibres
recovered through pyrolysis [18] (which preserves the original fibre architecture, e.g. woven
structure) and chemical methods [20] (which preserved the length of the fibres, but not the
architecture) have been produced into composites with very high mechanical properties as
well.
Page 10 of 16
4 Conclusions
This article characterised properties of rCFRPs manufactured through a liquid
composite moulding process. Nominally-random and partially-oriented non-woven mats of
carbon fibres, recovered through a fluidised bed process, were impregnated with two different
commercially available
epoxy resin
systems.
While
preform
impregnation
was
straightforward (and similar to conventional preforms), the low areal density and poor
compactability of the non-woven mats implied that a consolidation press had to be used to
achieve useful fibre volume fractions (of about 30%). The poor compaction characteristics of
the non-woven mats may be a barrier to their use in some industries. Potential routes of
improving the compaction behaviour of the mats are to improve fibre alignement, increase
areal density, and employ (multiple) precompaction cycles prior to lay-up.
The tensile, compressive and interlaminar shear properties of the different rCFRPs
were evaluated. The results clear demonstrated that while the oriented rCFRPs performed
better than the random rCFRPs, which in turn substantially reinforced the matrix in terms of
stiffness, the observed stiffness of the rCFRPs was below what would be expected from the
rule of mixtures for the fibre properties of the recycled carbon fibre. The high void content of
up to 10%, and fibre damage associated with the high compaction pressures may be
responsible for this.
A comparison of the mechanical properties of rCFRPs achieved through various
manufacturing routes has revealed that the liquid composite moulding process needed to be
further optimised to achieve comparable mechanical properties, although similar fibre
volume fractions were already being achieved. For instance, light resin transfer moulding,
with a semi-rigid composite top tool, may be more appropriate than vacuum-bagging to
reduce void content, control laminate thickness and complement press-consolidation.
References
1.
2.
3.
4.
5.
6.
Reux, F. Worldwide composites market: Main trends of the composites industry. in 5th
Innovative Composites Summit - JEC ASIA 2012. 26-28 June 2012. Singapore.
Pickering, S., Recycling technologies for thermoset composite materials - current status.
Composites Part A: Applied Science and Manufacturing, 2006. 37: p. 1206-1215.
Yang, Y., Boom R, Irion B, van Heerden D, Kuiper P, de Wita H, Recycling of composite
materials. Chemical Engineering and Processing: Process Intensification, 2012. 51: p. 53-68.
Larsen, K., Recycling wind turbine blades. Renewable Energy Focus, 2009. 9(7): p. 70-73.
Rybicka, J., Tiwari A, Campo PAD, Howarth J, Capturing composites manufacturing waste
flows through process mapping. Journal of Cleaner Production, 2015. 91: p. 251-261.
WR1202 - Landfill bans and restrictions in the EU and US. 2009, Green Alliance, Defra:
London, UK.
Page 11 of 16
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
Cherrington, R., Goodship V, Meredith J, Wood BM, Coles SR, Vuillaume A, Feito-Boirac
A, Spee F, Kirwan K, Producer responsibility: Defining the incentive for recycling composite
wind turbine blades in Europe. Energy Policy, 2012. 47: p. 13-21.
Shah, D., Developing plant fibre composites for structural applications by optimising
composite parameters: a critical review. Journal of Materials Science, 2013. 48(18): p. 60836107.
Koronis, G., Silva A, Fontul M, Green composites: A review of adequate materials for
automotive applications. Composites Part B: Engineering, 2013. 44(1): p. 120-127.
La Mantia, F., Morreale M, Green composites: A brief review. Composites Part A: Applied
Science and Manufacturing, 2011. 42: p. 579-588.
Faruk, O., Bledzki AK, Fink HP, Sain M, Biocomposites reinforced with natural fibres: 20002010. Progress in Polymer Science, 2012. 37(11): p. 1552-1596.
Pickering, S. Recycling and disposal of thermoset composites. in Workshop on Life Cycle
Assessment (LCA) for Composites. 2013. Dartington, UK.
Pimenta, S., Pinho ST, Recycling carbon fibre reinforced polymers for structural
applications: Technology review and market outlook. Waste Management, 2011. 31: p. 378392.
Pimenta, S., Pinho ST, Robinson P, Wong KH, Pickering SJ, Mechanical analysis and
toughening mechanisms of a multiphase recycled CFRP. Composites Science and
Technology, 2010. 70: p. 1713-1725.
Connor, M., Characterization of recycled carbon fibers and their formation of composites
using injection molding. 2008, North Carolina State University: Raleigh, NC, USA.
Turner, T., Pickering SJ, Warrior NA. Development of high value composite materials using
recycled carbon fibre. in SAMPE’09. 2009. Baltimore, MD, USA.
Turner, T., Warrior NA, Pickering SJ, Development of high value moulding compounds from
recycled carbon fibres. Plastics, Rubber and Composites, 2010. 39: p. 151-156.
Pimenta, S., Pinho ST, The effect of recycling on the mechanical response of carbon fibres
and their composites. Composite Structures, 2012. 94: p. 3669-3684.
Wong, K., Pickering SJ, Turner TA, Warrior NA. Compression moulding of a recycled
carbon fibre reinforced epoxy composite. in SAMPE'09. 2009. Baltimore, MD, USA.
Feraboli, P., Kawakami H, Wade B, Gasco F, DeOto L, Masini A, Recyclability and
reutilization of carbon fiber fabric/epoxy composites. Journal of Composite Materials, 2012.
46(12): p. 1459-1473.
Howarth, J., Jeschke M. Advanced non-woven materials from recycled carbon fibre. in
Carbon Fibre Recycling and Reuse 2009 Conference. 2009. Hamburg, Germany:
IntertechPira.
Pickering, S., Kelly RM, Kennerley JR, Rudd CD, Fenwick NJ, A fluidised bed process for
the recovery of glass fibres from scrap thermoset composites. Composites Science and
Technology, 2000. 60: p. 509-523.
Yip, H., Pickering SJ, Rudd CD, Characterisation of carbon fibres recycled from scrap
composites using fluidised bed process. Plastics, Rubber and Composites, 2002. 31(6): p. 278282.
Turner, T., Warrior NA, Pickering SJ, Re-use of carbon fibres in high value moulding
compounds & pre-pregs, in 17th International Conference on Composite Materials (ICCM17). 2009: Edinburgh, UK.
Pimenta, S., Pinho ST, Robinson P, Experimental study on the mechanical performance of
recycled CFRP, in 17th International Conference on Composite Materials (ICCM-17). 2009:
Edinburgh, UK.
Harris, B., Engineering composite materials. 1999, London: The Institute of Materials.
Shah, D., Schubel PJ, Clifford MJ, Licence P. Mechanical characterization of vacuum infused
thermoset matrix composites reinforced with aligned hydroxyethylcellulose sized plant bast
fibre yarns. in 4th International Conference on Sustainable Materials, Polymers and
Composites. 6-7 July 2011. Birmingham, UK.
Page 12 of 16
28.
Shah, D., Schubel PJ, Clifford MJ, Licence P, The tensile behavior of off-axis loaded plant
fiber composites: an insight on the non-linear stress-strain response. Polymer Composites,
2012. 33(9): p. 1494-1504.
Page 13 of 16
Tables
Table 1. Physical properties of the fabricated rCFRPs.
Sample ID
1
2
3
4
5
6*
Nonwoven mat type
Testing
direction
Matrix system
Laminate
thickness
[mm]
Composite
density
[gcm-3]
Partially aligned
Nominally random
Partially aligned
Partially aligned
Nominally random
Partially aligned
0°
90°
0°
0°
RIM135/RIMH137
RIM135/RIMH137
RIM135/RIMH137
RIM035/RIMH038
RIM035/RIMH038
RIM135/RIMH137
3.04 ± 0.14
3.22 ± 0.25
3.02 ± 0.23
3.17 ± 0.27
3.29 ± 0.40
~14
1.31 ± 0.01
1.19 ± 0.02
1.22 ± 0.02
1.27 ± 0.01
1.25 ± 0.01
1.19 ± 0.01
Fibre
volume
fraction
[%]
33.9 ± 0.4
25.6 ± 0.8
28.1 ± 0.6
26.6 ± 0.3
24.8 ± 0.3
8.5 ± 0.2
Porosity
content
[%]
3.2 ± 0.5
9.5 ± 1.9
9.0 ± 2.2
3.0 ± 0.6
3.8 ± 0.6
1.0 ± 0.4
* No press-moulding stage.
Table 1. Mechanical properties of the fabricated rCFRPs.
Sample ID
Tensile
modulus
[GPa]
Tensile
strength
[MPa]
Tensile failure
strain
[%]
Compressive
modulus
[GPa]
Compressive
strength
[MPa]
Poisson’s ratio
(compressive)
Inter-laminar
shear strength
[MPa]
1
2
3
4
5
19.0 ± 0.8
12.3 ± 0.5
14.4 ± 0.5
32.1 ± 2.4
13.2 ± 1.3
133 ± 7
100 ± 4
119 ± 5
174 ± 20
84 ± 3
0.75 ± 0.04
0.93 ± 0.06
0.87 ± 0.03
0.61 ± 0.07
0.68 ± 0.06
8.5 ± 0.7
7.6 ± 0.5
7.9 ± 0.9
6.9 ± 1.6
5.5 ± 0.2
146 ± 13
169 ± 10
128 ± 14
126 ± 16
147 ± 5
0.48 ± 0.10
0.28 ± 0.04
0.08 ± 0.05
0.32 ± 0.01
0.18 ± 0.06
16.7 ± 2.1
25.1 ± 3.4
24.8 ± 4.0
43.0 ± 2.6
27.8 ± 6.0
Page 14 of 16
Table 2. Comparison of the mechanical properties of rCFRPs manufactured through various routes in literature. Note that non-woven mats
employ short fibres (typically less than 10 mm in length).
Manufacturing technique
Injection moulding
Bulk moulding compound
Sheet moulding compound
Prepregging
Compression moulding
Resin film infusion
Reinforcement form
Pellets
Fibres mixed with resin
and filler
Aligned non-woven
mat
Aligned non-woven
mat
Aligned non-woven
mat
Random non-woven
mat
Aligned non-woven
mat
Woven fabric
Woven fabric
Long fibre random mat
Liquid composite moulding
Random non-woven
mat
Aligned non-woven
mat
Matrix and
moulding
pressure
Fibre
volume
fraction
[%]
Tensile
modulus
[GPa]
Tensile
strength
[MPa]
Source
Polycarbonate,
14 MPa
16
12-14
87-124
[15]
Epoxy, 2 MPa
10
20
71
[17]
23
44
282
[17]
17
26
203
[17]
44
80
422
[17]
30
25
207
[19]
Epoxy, 7MPa
30
28
194
[14]
Epoxy, 0 MPa
Epoxy, 0.7
MPa
Epoxy, vacuum
only
Epoxy, 2.5
MPa
Epoxy, 2.5
MPa
55
66
271
[18]
50
56
634
[18]
33
29
197
[20]
25-26
12-13
84-100
This study
27-34
19-32
133-174
This study
Epoxy, 5-10
MPa
Epoxy, 0.1
MPa
Epoxy, 5-10
MPa
Epoxy, 5-10
MPa
Page 15 of 16
Figures
b)
a)
c)
Fig. 1. Lay-up (a), vacuum-bagging (b) and liquid resin impregnation of the recycled carbon
fibre mats (c).
Not pressed
Porosity fraction
Matrix volume fraction
Fibre volume fraction
14.5
14.0
1.0%
Laminate thickness [mm]
13.5
2.5 MPa compaction pressure
13.0
3.5
9.5%
3.2%
9.0%
3.0%
3.8%
3.0
2.5
2.0
62.9%
64.9%
62.9%
70.4%
71.4%
90.5%
33.9%
25.6%
28.1%
28.1%
26.6%
26.6%
24.8%
24.8%
8.5%
8.5%
1
2
5
6
1.5
1.0
0.5
0.0
3
4
Sample ID
Fig. 2. Thickness and constituent volume fractions of the fabricated rCFRP laminates. The
error bars indicate ± 1 std. dev. in laminate thickness.
Page 16 of 16