recycling

Review
Mechanical Processing of GFRP Waste into
Large-Sized Pieces for Use in Concrete
Ardavan Yazdanbakhsh *, Lawrence C. Bank and Yuan Tian
City College of New York, New York, NY 10031, USA; lbank2@ccny.cuny.edu (L.C.B.);
ytian000@citymail.cuny.edu (Y.T.)
* Correspondence: ayazdanbakhsh@ccny.cuny.edu; Tel.: +1-212-650-6569

Received: 23 December 2017; Accepted: 12 February 2018; Published: 15 February 2018

Abstract: Recycling glass fiber reinforced polymer (GFRP) composite materials has been proven
to be challenging due to their high mechanical performance and high resistance to harsh chemical
and thermal conditions. This work discusses the efforts made in the past to mechanically process
GFRP waste materials by cutting them into large-sized (cm scale) pieces, as opposed to pulverization,
for use in concrete mixtures. These pieces can be classified into two main categories—coarse aggregate
and discrete reinforcement, here referred to as “needles.” The results from all the studies show that
using GFRP coarse aggregate leads to significant reductions in the compressive strength and tensile
strength of concrete. However, GFRP needles lead to sizable increases in the energy absorption
capacity of concrete. In addition, if the glass fibers are longitudinally aligned within the needles,
these elements can substantially increase the tensile strength of concrete. Processing GFRP waste into
needles requires less energy and time than that for producing GFRP coarse aggregate. Also, compared to
pulverized GFRP waste, which consists of broken and separate particles of glass and resin that at best
can be used as low-quality fillers, GFRP needles are high strength composite elements.

Keywords: aggregate; concrete; GFRP; mechanical recycling; needle; discrete reinforcement

1. Introduction
For the last three decades, fiber-reinforced polymer (FRP) composite materials have been used
extensively and increasingly in major industries. FRP typically consist of fibers encased in a matrix
of resins with fiber concentrations typically in the range of 12–60% by volume [1]. It is estimated that
in the United States, the annual demand for FRP materials, 95% of which made with glass fibers [2],
has climbed to nearly two million metric tons in 2017 (over 20% increase since 2007) [3]. When these
GFRP products reach the end of their service lives, the annual GFRP waste produced in the U.S. (including
the production scrap, which is approximately 5% of produced GFRP) will be at least 2 million metric tons,
with increasing values in the subsequent years.
GFRP materials have remarkable physical and mechanical properties; they can be over two times
stronger than steel in tension at only a quarter of the weight [4]. Although these materials are extremely
durable in corrosive environments at low temperatures (<80 ◦ C) their life spans can be relatively short
due to low design allowables and functional obsolescence. For example, wind turbine blades, with
GFRP as the main material used in the structure and outer shell of almost all of them, are designed for
only a 20-year service life [5,6]. Since 1999 the production of wind power in the United States has grown
30 times and the annual installation of new wind turbines has grown 10 times [7]. Despite the fact
that GFRP is non-biodegradable, the vast majority of GFRP waste in the United States is landfilled.
Recycling GFRP materials is a challenging task since approximately 75% of GFRP products are made
with thermoset polymers, which do not melt at high temperatures. A number of methods for recycling
FRP waste have been developed involving chemical and/or thermal processes for reclaiming the

Recycling 2018, 3, 8; doi:10.3390/recycling3010008 www.mdpi.com/journal/recycling

Recycling 2018, 3, 8 2 of 11

fibers [8]. These techniques are very costly and are justifiable only for carbon fiber reinforced polymer
composite materials because of the high price of carbon fibers.
Low-impact processing of GFRP into products which can be used in built infrastructure can have
a significant beneficial impact on the environment, as it reduces the demand for natural resources and
the need for landfilling. Mechanical recycling of GFRP is an attractive option in terms of lower energy
demand and the avoidance of chemical processes. The energy required for mechanical recycling is
between 0.5% and 5% of that required for chemical recycling and between 0.4% and 16% of the energy
used for thermal recycling (pyrolysis) [9]. One issue with traditional mechanical processing (pulverizing
or shredding) is that the processed GFRP is no longer a composite material. It consists of separate pieces
of broken damaged fibers and resin particles and therefore has a negative impact on the mechanical
properties of the new material in which it is incorporated [8,10–12]. Cutting GFRP waste into relatively
large pieces for use in new products, as opposed to grinding and shredding, is an attractive potential
recycling option for two reasons: (1) the energy demand for cutting GFRP to large pieces is less than
that required for grinding and shredding (since less surface is generated) and (2) cut pieces of GFRP are
composite materials, rather than separate damaged fibers and resin particles, with mechanical properties
the same as those of the GFRP before being processed.
This work presents the main investigations performed in the past on the incorporation of coarse
processed GFRP waste in concrete. In those studies, different types of GFRP products were processed
into elements with different geometries and surface textures and were added to concrete mixtures at
different volumetric ratios. Each study measured a number of primary mechanical characteristics of
waste incorporated concrete such as compressive strength, tensile strength and flexural load carrying
capacity. A brief summary of the studies, the results and the possible explanations for the observed
performance of concrete with GFRP waste is explained. A comprehensive review on the use of recycled
plastics in concrete is presented in an article by Gu and Ozbakkaloglu [13]. The present work focuses
only on the use of large-sized pieces of mechanically processed GFRP waste in concrete.

2. Use of Processed GFRP Waste in Concrete as Coarse Aggregate

In a study by Yazdanbakhsh et al. scrap from production of GFRP reinforcing bars (rebars) with
different diameters (6, 10, 13, 16, 19 and 25 mm) were cut into short cylindrical pieces with aspect ratio
of one. These pieces, referred to as GFRP Recycled Aggregate (GRA), were used as full and partial
(40% by volume) replacement of coarse natural aggregate (NA) in two types of concrete with water
to cement ratios of 0.57 and 0.44 [14]. Both the GRA and NA had the same particle size distribution
according to ASTM C33 standard [15]. The scrap consisted of short pieces of high-quality rebars that
could not be sold due to their length. The results of the study showed that the replacement of NA with
GRA resulted in significant reductions in the compressive strength and splitting tensile strength of
concrete, due to the weak bond between GRA particles (particularly the smooth saw-cut base surfaces
of the cylindrical pieces) and concrete matrix. Another possible reason for the lower strength of
GFRP-incorporated concretes is that there were only a limited number (six) of particle sizes in the GRA.
In addition, all the GRA particles had the same shape. The non-gradual size gradation and the round
shape of GRA particles can have a negative impact on the packing density of the aggregate and
the interlock between the particles. Replacing 40% and 100% of NA with GRA in both types of concrete
resulted in the reductions of 13% and 21% in compressive strength, respectively. For the concrete with
the water to cement ratio of 0.57, replacing 40% and 100% of NA with GRA resulted in the significant
reductions of 26% and 35% in splitting tensile strength, respectively. For the higher strength concrete
with the water to cement ratio of 0.44, replacing 40% and 100% of NA with GRA resulted in the less
significant reductions of 12% and 20% in splitting tensile strength, respectively. Increasing the cement
content of concrete improves the bond between aggregates and concrete matrix. Since the tensile
strength of concrete is highly affected by the aggregate-matrix bond, the replacement of NA with GRA
caused less reduction in the tensile strength of the concrete with higher cement content.
Recycling 2018, 3, 8 3 of 11

Alam et al. [16] used GFRP scrap that was excess from the composite manufacturing process for
casting waterslides. The GFRP was coated with a thin layer of gel to make it smooth. The long strips
of the scrap were processed by cutting them into small flat squares using an abrasive wet tile saw.
The flat GRA was used to replace 25% and 50% volumetric portions of NA in concrete with a water to
cement ratio of 0.4. The GRA and NA particles had similar maximum particle sizes. However, the GRA
particles had similar sizes while the NA particles were well-graded. Replacement of NA with GRA
resulted in up to more than 50% and 40% decreases in the compressive strength and flexural strength
of concrete, respectively. The very smooth surface of the GRA pieces resulted in a poor bond with
concrete matrix and was an important reason for the significant strength loss of concrete. The flat shape
and poor gradation of the GRA particles are possible reasons for the poor mechanical performance of
GRA incorporated concrete.
Fox used GFRP sheets acquired from a wind blade production plant and a wind energy manufacturing
laboratory to produce GRA [17]. He used a band saw and sliding wet saw to process the sheets into cubic
and rectangular blocks. One to three faces of each GRA piece were sheared and made rough with the aim
of enhancing the bond between GRA and concrete matrix. Two types of GRA with average side length
of 13 mm and 25 mm were produced and incorporated in concrete mixtures with the water to cement
ratio of 0.45. GRA was used as a partial replacement of NA with volumetric ratios of 0.25%, 32.5%, 37.5%
and 50% in concrete mixtures. The results showed that replacing 50% of natural coarse aggregate with
different combinations of 13 mm and 25 mm recycled wind blade aggregate led to approximately 44%
reduction in compressive strength (from 41 MPa down to 21 MPa). Replacing the natural coarse aggregate
with 25 mm recycled aggregates at volume ratios of 25% and 37.5% resulted in 22% and 38% reductions
in compressive strength, respectively. Since GRAs with fully smooth faces were not investigated in the
study, it is not known whether shearing the faces of GRA pieces had any positive effect on the compressive
strength of concrete.

3. Use of Processed GFRP Waste in Concrete as Discrete Reinforcing Elements

3.1. The Benefits and Challenges of Using the Existing Discrete Reinforcing Elements in Concrete
The use of discrete reinforcing elements in concrete is attractive since these elements can be added
during mixing to reduce time and labor cost for preparing (bending, cutting and connecting) and
placing traditional steel reinforcing bars (rebars). Fibers constitute the vast majority of existing discrete
reinforcing elements used commercially in concrete. Fibers, particularly those made from steel and
polymers or even composites, have been used to reduce or even replace rebars in various types of
structural contexts, particularly in flat members such as airport taxiways [18], slabs-on-ground [19]
and elevated slabs [20]. Since fibers are much smaller than traditional rebars and are, to a notable
extent, uniformly dispersed and oriented within concrete matrix, they can distribute stress effectively
in concrete and can resist the occurrence of local damage [21–23].
Fibers are not devoid of shortcomings: (1) Dispersing fibers in fresh concrete is difficult.
Fibers tend to agglomerate and form clumps. Therefore, in many cases the maximum amount
of fiber that can be used in concrete is limited to 1% and sometimes 0.5% of the total volume of
concrete; (2) Fibers reduce the workability of concrete significantly due to their high specific surface
area resulting from their small cross-section dimensions and high aspect ratios (typically above 50),
even when incorporated at the aforementioned low dosages. High range water-reducing admixtures
(superplasticizers) thus are used to avoid the need to increase the water to cementitious material
ratio in concrete. Only a limited dosage of superplasticizers can be used in concrete, beyond which
setting time and entrapped air content can increase significantly. Using the allowable dosage to
counteract the effect of fibers on workability is not the most efficient application for superplasticizers;
when superplasticizers are used in concrete without fibers, less cementitious binder (paste) is required
to achieve the desired target compressive strength and workability and the demand for Portland
cement can be reduced by 25%, which is a major contribution to the environmental sustainability of
only a limited dosage of fibers can be used, their effect on the tensile strength of concrete—an
important parameter in the design of structures such as rigid pavements—is typically low [25–27].

3.2. The Concept of “Needle” as Discrete Reinforcing Elements in Concrete

Recycling 2018, 3, 8 4 of 11

The concept of “needle” as concrete reinforcing element has been discussed in a number of past
studies concrete
[28,29].[24]; Needles resemble
(3) Although fibers can fibers
enhance used in concrete
the post-failure in two
toughness general
of concrete ways: (1)
significantly, A needle is
because
only a limited dosage of fibers can be used, their effect on the tensile strength
elongated, i.e. its aspect ratio (the ratio of length to nominal diameter) is larger than one; (2) needles of concrete—an important
parameter in the design of structures such as rigid pavements—is typically low [25–27].
can be mixed with fresh concrete to improve some of the properties of the hardened concrete.
However, 3.2.theThe mechanisms
Concept of “Needle” by aswhich needles
Discrete ReinforcingandElements
concrete fibers resist the growth of the cracks are
in Concrete
fundamentally different.
The concept of “needle” as concrete reinforcing element has been discussed in a number of past
In fiber
studiesor[28,29].
needle reinforced
Needles resembleconcrete
fibers usedthe angle between
in concrete in two general an embedded elongated
ways: (1) A needle element and
is elongated,
the planei.e.ofitsaaspect
crackratio
that(the
intersects that element
ratio of length to nominalrangesdiameter)from 0° tothan
is larger 90°.one;
The(2)probability
needles can be that the element
mixed
with fresh concrete
is perpendicular to improve
to the crack surfacesomeisofvery
the properties
low. In of the hardened
addition, sinceconcrete. However, the mechanisms
a crack—particularly a major crack
by which needles and concrete fibers resist the growth of the cracks are fundamentally different.
that contributes to the failure of a concrete member—intersects more than one elongated element, the
In fiber or needle reinforced concrete the angle between an embedded elongated element and
vast majority of the elements are “inclined,” i.e. lie at an angle to the crack surface. When fiber
the plane of a crack that intersects that element ranges from 0◦ to 90◦ . The probability that the element
reinforced concrete (FRC)
is perpendicular carries
to the crack increasing
surface is very low. external load
In addition, and
since a propagating acrack
a crack—particularly intersects an
major crack
inclinedthat
fiber, high tensile
contributes strain
to the failure of develops in the fiber in the
a concrete member—intersects more zone
thanbetween theelement,
one elongated crack surfaces,
the vast which
majority
translates into ofa the
highelements
shearare “inclined,”
stress in thei.e. lie at an angle
interface to the crack
between surface.
the fiberWhen
andfiber reinforced concrete
cementitious matrix in the
(FRC) carries increasing external load and a propagating crack intersects an inclined fiber, high tensile
zone close to the crack. This shear stress may result in debonding which progresses toward the end
strain develops in the fiber in the zone between the crack surfaces, which translates into a high shear
of the fiber, after which the fiber, if not anchoraged at the end, will start to move out of its grove at
stress in the interface between the fiber and cementitious matrix in the zone close to the crack. This shear
one sidestress
of the crack.
may resultAs shown inwhich
in debonding Figure 1, in order
progresses towardfortheanendinclined fiberafter
of the fiber, to slip
which out,
the it must
fiber, bend over
if not
crack surfaces,
anchoraged which maywill
at the end, lead
startto crumbling
to move out of itsofgrove
concrete
at one sidein aof relatively
the crack. Assmall
shown region.
in Figure Even
1, stiff
in order for an inclined fiber to slip out, it must bend over crack surfaces,
commercial fibers, such as those made with steel, bend in plastic mode during crack propagation. If which may lead to crumbling of
concrete in a relatively small region. Even stiff commercial fibers, such as those made with steel, bend in
during debonding or pull-out the tensile stress at a point in the fiber exceeds its ultimate tensile
plastic mode during crack propagation. If during debonding or pull-out the tensile stress at a point in
strength,thethe fiber will break.
fiber exceeds its ultimate tensile strength, the fiber will break.

Figure Figure 1. Fiber

1. Fiber pull-out
pull-out andand matrixcrumbling
matrix crumbling during
duringthethe
growth of a crack
growth in concrete.
of a crack in concrete.

An “ultimate needle” is defined as an elongated element that is fully rigid and infinitely strong in
An tension
“ultimate needle” is defined as an elongated element that is fully rigid and infinitely strong
and shear. Therefore, ultimate needles do not deform or break. The only way for a concrete
in tension and
crack shear.byTherefore,
bridged an ultimateultimate needles
needle to grow do not
in width is ifdeform or break.
the concrete encasingThetheonly
needleway for a concrete
crumbles
crack bridged
and/or by an in
spalls ultimate needle
at least one side to grow
of the in width
crack, is if the concrete
either progressively encasing
or at once. thepossible
It is not needletocrumbles
produceinultimate
and/or spalls needles
at least one since
side materials that areeither
of the crack, infinitely strong and rigidor
progressively doat
notonce.
exist. However,
It is not itpossible
is to
possible to create very stiff and strong needles by selecting materials with high Young’s
produce ultimate needles since materials that are infinitely strong and rigid do not exist. However, it modulus and
tensile and flexural strengths and by choosing a large needle cross-section area. Figure 2 illustrates
is possible to create very stiff and strong needles by selecting materials with high Young’s modulus
schematically the mechanism of the growth of a crack bridged by a needle. When intersected by
and tensile and flexural
a propagating crack, strengths
the needle: (1)anddobynot choosing a large
deform in plastic needle
mode cross-section
or fracture; (2) undergoarea.
veryFigure 2
illustrates schematically
small elastic bending.the mechanism
Therefore, the crackofopening
the growth
mechanism of will
a crack bridged
be similar to thatby a needle.
described for When
cracks
intersected bybridged by ultimatecrack,
a propagating needles.
theIn needle:
addition, (1)
suchdoneedles, due to their
not deform large cross-section
in plastic mode or area,
fracture; (2)
undergo very small elastic bending. Therefore, the crack opening mechanism will be similar to that
described for cracks bridged by ultimate needles. In addition, such needles, due to their large cross-
section area, are expected to have specific surface areas much greater than those of concrete fibers
and, therefore, do not reduce the workability of fresh concrete significantly.
Recycling 2018, 3, 8 5 of 11

are expected to have specific surface areas much greater than those of concrete fibers and, therefore,
do not reduce the workability of fresh concrete significantly.
Recycling 2018, 3, x 5 of 11

Recycling 2018, 3, x 5 of 11

Figure 2. (Left) a needle bridging a concrete crack at the outset; (Right) Local failure and crumbling
Figure 2. (Left)
during cracka opening.
needle bridging a concrete crack at the outset; (Right) Local failure and crumbling
during crack opening.
Figure 2. (Left) a needle bridging a concrete crack at the outset; (Right) Local failure and crumbling
3.3. Past Studies on Recycling GFRP Waste into Needles for Use in Concrete
during crack opening.
3.3. Past Studies on Recycling
In a recent GFRP
study, GFRP Waste
rebar into Needles
production forwas
scrap Usecut
in Concrete
into needles to be used as discrete
3.3. Past Studies on Recycling GFRP Waste into Needles for Use in Concrete
reinforcing elements in concrete [28]. The Rebar
In a recent study, GFRP rebar production scrap(RB)was needles,
cut into hadneedlesa diameter
to be usedof 6 as
mm and a length
discrete reinforcing
of 100
elements In mm (Figure
in aconcrete 3).
recent study, Incorporating
GFRP
[28]. The the
rebar(RB)
Rebar RB needles
production
needles, in
scrap concrete
hadwas mixtures
cut into of
a diameter to replace
needles
6 mmtoand volumetrically
be used
a length 5% mm
as discrete
of 100
and 10%
reinforcing of coarse aggregate
elements in (1.76%
concrete and
[28]. The 3.52% of
Rebar (RB)concrete) led
needles,to to
had 22% and 33%
a diameter increase
of 6 mm and in the tensile
(Figure 3). Incorporating the RB needles in concrete mixtures replace volumetrically 5%a and
length10% of
strength
of 100 mm of (Figure
concrete, respectively.
3). Incorporating Intheaddition,
RB the results
needles in of the mixtures
concrete splitting tensile
to strength
replace tests showed
volumetrically 5% of
coarse aggregate
that 10%
the use
(1.76% and
of theaggregate
3.52%
GFRP needles
of concrete)
in and
concrete
led to 22% and 33% increase in the tensile strength
and of coarse (1.76% 3.52%led to significant
of concrete) led toincreases
22% andin33% theincrease
post-peak toughness
in the tensile
concrete, respectively.
(energy absorption Incapacity)
addition,ofthe results Figure
concrete. of the splitting
4 shows tensile
that strength
after peak testsconcrete
load showedspecimens
that the use of
strength of concrete, respectively. In addition, the results of the splitting tensile strength tests showed
the GFRP needles in concrete
incorporating led to significant
to resist increases in theup post-peak toughness (energy absorption
that the use ofRB theneedles continued
GFRP needles in concrete external
led load
to significant to displacements
increases several
in the post-peak times larger
toughness
capacity) of
(energy concrete.
than those absorption Figure
associatedcapacity) 4 shows
with peak that
ofload. after
Concrete
concrete. peak
Figure load
specimens concrete
4 showswith that5% specimens
andpeak
after 10% needleincorporating
replacement
load concrete RB needles
were,
specimens
respectively
continued to resist5%
incorporating RB and
external9%load
needles weaker
up toin displacements
continued compression than
to resist external the
loadcontrol
several times specimens.
up to larger than The
displacements those observation
associated
several of
times largerthe peak
with
fresh
load. than and specimens
Concrete
those harden concretes
associated withpeak
with 5%showed
and
load.10% thatneedle
Concretethespecimens
RB needles
replacementwith were
were,
5% randomly
and dispersed
respectively
10% needle 5% and in 9%
replacement concrete.
weaker in
were,
Although
compression the5%
than
respectively diameter
the control
and ofspecimens.
the cylinders
9% weaker Thewas
in compression only
than50%
observation theoflarger than
the fresh
control the harden
and
specimens.length
The of the needles
concretes
observation showedthe that
of the
distribution
fresh and and orientation
harden concretes of the needles
showed that thein RB
theneedles
cylindricalwere specimens
randomly were relatively
dispersed in random
concrete.
the RB needles were randomly dispersed in concrete. Although the diameter of the cylinders was only
(Figure 5). the diameter of the cylinders was only 50% larger than the length of the needles the
Although
50% larger than the length of the needles the distribution and orientation of the needles in the cylindrical
distribution and orientation of the needles in the cylindrical specimens were relatively random
specimens were relatively random (Figure 5).
(Figure 5).

Figure 3. Needles obtained by cutting the scrap from producing GFRP rebars [28].

Figure 3. Needles obtained by cutting the scrap from producing GFRP rebars [28].
Figure 3. Needles obtained by cutting the scrap from producing GFRP rebars [28].
Recycling 2018, 3, 8 6 of 11
Recycling
Recycling 2018,
2018, 3,
3, xx 66 of
of 11
11

400
400
NA
NA (control)
(control)
350
350 FRP-NDL-5
FRP-NDL-5
300
300 FRP-NDL-10
FRP-NDL-10
250
250

kN
Load, kN
200
200

Load, 150
150
100
100
50
50
00
00 0.4
0.4 0.8
0.8 1.2
1.2 1.6
1.6
Displacement,
Displacement, mm
mm

Figure
Figure 4.
4. Load-displacement
Load-displacement curves
curves from
from the
the splitting
splitting tensile
tensile tests.
tests. FRP-NDL-5
FRP-NDL-5 and
and FRP-NDL-10
FRP-NDL-10
Figure 4. Load-displacement curves from the splitting tensile tests. FRP-NDL-5 and FRP-NDL-10 represent
represent
represent concrete
concrete specimens
specimens in
in which
which 5%
5% and
and 10%
10% ofof NA
NA isis replaced
replaced volumetrically
volumetrically with
with the
the rebar
rebar
concrete specimens in which 5% and 10% of NA is replaced volumetrically with the rebar (RB) needles.
(RB)
(RB) needles.
needles.

Figure
Figure 5.
5. Distribution
Distribution of
of the
the rebar
rebar (RB)
(RB) needles
needles in
in aa cylindrical
cylindrical specimen
specimen tested
tested for
for splitting
splitting tensile
tensile strength.
strength.
Figure 5. Distribution of the rebar (RB) needles in a cylindrical specimen tested for splitting tensile strength.
In
In another
another recent
recent study,
study, GFRPGFRP needles
needles with with aa geometry
geometry similar
similar toto that
that ofof rebar
rebar needles
needles werewere
produced
In by
another cutting
recent a wind
study, turbine
GFRP blade
needles using
with a a table
geometry saw with
similar to
produced by cutting a wind turbine blade using a table saw with a diamond blade [29]. These wind- a diamond
that of rebar blade
needles[29]. These
were wind-
produced
blade
by
blade (WB)
(WB)aneedles
cutting wind turbine
needles had
had aa blade
length of
lengthusingof 100
100a mm
mm and
table sawthe
and with
the majority
a diamond
majority of
of themblade
them had
had square
[29]. These
square cross-sections
wind-blade (WB)
cross-sections with
with
the
the side
needles
sidehadlength
length of
of 66 mm.
a length of 100
mm. Twommtypes
Two and the
types of
of WB
WB needles
majority were
of them
needles werehadproduced; those
those with
with smooth
square cross-sections
produced; with the
smooth and
and grooved
side length
grooved
surfaces
of 6 mm.(Figure
surfaces Two types
(Figure 6).
6). The
of
TheWB second
needles
second type
type of
of needles
were produced;
needles had
had grooves
those withwith
grooves smooth
with the width
theand
width of
of 11 mm
grooved mm and
and depth
surfaces (Figure
depth of 22
of6).
mm
The spaced
mmsecond
spacedtype at
at 17
17ofmm
mm intervals.
needles Reference
had grooves
intervals. Reference with and
andtheneedle-incorporated
width of 1 mm and depth
needle-incorporated concretes
concretesof 2 were produced
mm spaced
were produced at 17using
mm
using
the
the same
same mix
intervals. proportions
Reference
mix as
as in
in the
the study
and needle-incorporated
proportions study of of rebar needles.
concretes
rebar were
needles. Various
produced
Various concrete
usingspecimens
concrete the same mix
specimens were
were produced
proportions
produced
and
as
andintested
the study
tested to
to measure
of rebarthe
measure the compressive
needles.
compressiveVarious strength,
concretesplitting
strength, specimens
splitting tensile
were
tensile strength
produced
strength and
andandflexural
testedtoughness
flexural to measure
toughness of
of
concrete.
the
concrete. Equivalent
compressive
Equivalent flexural
strength, strength
splitting
flexural tensilewas
strength was calculated
strength as
as aa measure
and flexural
calculated toughness
measure of
of toughness
of concrete.according
toughness Equivalent
according to ASTM
toflexural
ASTM
C1609
C1609 [30].
strength was calculated as a measure of toughness according to ASTM C1609 [30].
[30].
The results
The results
resultsshowshow
showthatthat
thatthe the incorporation
incorporationofof
theincorporation both
ofboth
both plain
plain
plain and
andand grooved
grooved
grooved WB WB
WB needles
needles
needles in in
in concrete
concrete
concretehashasan
has
an
an insignificant
insignificant
insignificant impactimpact
on the
impact on the
the compressive
oncompressive
compressivestrength strength of
of concrete.
of concrete.
strength Replacing
concrete. Replacing
5% and 10%
Replacing 5%
5% andof NA
and 10%
10% of
of NA
with NA with
plain WB
with
plain
plain WB
needles WB needles
resulted
needles resulted
in 1.5% in
in 1.5%
reduction
resulted reduction
and 0.1%
1.5% increase
reduction and
and in 0.1%
the
0.1% increase
increase in
compressive the
the compressive
strength
in of concrete, strength
compressive strength of
respectively. of
concrete,
concrete, respectively.
Replacing 5% and 10% of
respectively. Replacing
NA with5%
Replacing and
and 10%
grooved
5% of
of NA
WB needles
10% NA with grooved
resulted
with in 7.2%
grooved WB
WB needles
increase
needles and resulted in
in 7.2%
4.6% reduction
resulted 7.2%
increase
in and
and 4.6%
the compressive
increase 4.6% reduction
strength ofin
reduction the
the compressive
inconcrete, respectively.
compressive strength
These
strength of concrete,
ofchanges
concrete, arerespectively. These
These changes
within the expected
respectively. marginare
changes of
are
within
error
within the
andthedo expected
not indicate
expected margin
margin any of error
error and
beneficial
of and do
do not
not indicate
or detrimental any
impact
indicate anyofbeneficial
the needles
beneficial or
oron detrimental
compressive
detrimental impact of
of the
strength
impact of
the
needles
needles on
concrete. Thecompressive
on better performance
compressive strength
strength of of
of concrete.
concrete
concrete.withThe wind
The better
bladeperformance
better of
of concrete
needles in compression
performance with
with wind
concretecompared wind blade
to that of
blade
needles
concrete in
withcompression
rebar needles compared
is expectedto that
to of
be concrete
due to the with
more rebar
angularneedles
needles in compression compared to that of concrete with rebar needles is expected to be due to the surfaceis expected
of wind to
bladebe due
needles.to the
more
more angular
angular surface
surface of of wind
wind blade
blade needles.
needles.
Recycling 2018, 3, x 7 of 11
Recycling 2018, 3, 8 7 of 11
Recycling 2018, 3, x 7 of 11

Figure 6. (a) Plain and (b) grooved wind blade needles obtained from cutting a wind blade shell.
Figure 6. (a) Plain and (b) grooved wind blade needles obtained from cutting a wind blade shell.
Figure 6. (a) Plain and (b) grooved wind blade needles obtained from cutting a wind blade shell.
Test results also show
Test results that that
also show the the
splitting
splittingtensile
tensilestrength
strength decreases
decreases as asaaresult
resultofofincorporating
incorporating
WB needles
Test in concrete.
results also The
show splitting
that the tensile
splitting strength
tensile of the
strength concrete
decreaseswith as5%
a volumetric
result
WB needles in concrete. The splitting tensile strength of the concrete with 5% volumetric replacement of replacement
incorporating WB
ofneedles
NA of with plain
in concrete.
NA and
with plainThegrooved
andsplittingneedles
groovedtensile were
needlesstrength14%
were 14% and
of and 13% lower
the concrete
13% lower than
with
than5% that of the
volumetric
that reference concrete.
replacement
of the reference concrete.of NA
The splitting
with plain andtensile
The splitting strength
tensile
grooved needlesof the
strength wereconcrete
of the14% andwith
concrete with
13% 10%10%volumetric
lower volumetric
than that of replacement
replacement
the reference ofofNANA with
with
concrete. grooved
grooved
The splitting
tensileneedles
needles was 8%
strength was 8% concrete
oflower
the lower
thanthanthatthat
with of
of the
10% the reference
reference
volumetric concrete.ofThe
concrete.
replacement The concrete
NA withconcrete with
groovedwith 10%
10%volumetric
needles volumetric
was 8% lower
replacement
replacement
than that of NA
of the ofwith
NA with
reference plainplain
concrete. needles
needles
The andand
concrete the the reference
reference
with concretereplacement
concrete
10% volumetric (without needles)
(without of NA had
needles) had
with the same
the
plain same
needles
average average splitting tensile strength. These results are in contrast with those from investigating GFRP
and thesplitting
referencetensile
concrete strength.
(withoutThese
needles)results
had are in contrast
the same average with thosetensile
splitting from investigating
strength. TheseGFRP results
rebar needles, which as stated earlier, significantly increased the splitting tensile strength of concrete.
rebar needles,
are in contrast which
with as stated
those fromearlier, significantly
investigating GFRP increased the splitting
rebar needles, whichtensile strength
as stated earlier,ofsignificantly
concrete.
A resin burn-off test revealed that most of the glass fibers in wind blade needles were not aligned
Aincreased
resin burn-off test revealed
thetosplitting tensile that mostofofconcrete.
strength the glassAfibers
resin in wind test
burn-off blade needles were notofaligned
parallel the axis of the needles. However, in a rebar needle all the fibersrevealed that
are precisely most
aligned the glass
along
parallel
fibersthetowind
in the axis of
blade the needles.
needles were However,
not aligned in a rebar
parallel needle
to the all
axis ofthe
thefibers
needles.
axis of the needle. Therefore, the average tensile strength of the wind blade needles was are precisely
However, aligned
in a rebar along
needle
the axis of the
all thesignificantly needle.
fibers are precisely
lower thanTherefore,
aligned the
that of along average tensile strength of the wind blade
the axis of the needle. Therefore, the average tensile strength
rebar needles. needles wasof
significantly
the wind blade lower than that
needles was of rebar needles.
significantly lower than that of rebar needles.

Figure 7. Load-displacement results from third point bending tests of concrete specimens with (a)
plain and (b) grooved needles. The numbers 5 and 10 in the specimen notations represent the
volumetric ratios of replacing NA with needles.
Figure
Figure7.7.Load-displacement
Load-displacementresults
results from third point
from third pointbending
bendingtests
testsofof concrete
concrete specimens
specimens with
with (a)
(a) plain
plain and
and (b) (b)
Thegroovedgrooved needles.
results needles. The
The numbers
of the toughness numbers 5 and
5 and 10that
test showed 10
in the in the
thespecimenspecimen
concretenotations notations represent
represent theWB
beams incorporating the
volumetric
needles
volumetric
continued
ratios ratios
of replacingof replacing
to carryNAload NA
after
with thewith needles.
applied
needles. load reached its peak value (Figure 7). Replacing 5% and
10% of coarse aggregate by volume with the plain needles led to the increase of average equivalent
The results
flexural of theRtoughness
strength 150
from 1.1test
to showed
14.0 thatrespectively.
and 32.3, the concrete Whenbeams incorporating
grooved needles wereWB usedneedles
The results of theT ,150
toughness test showed that the concrete beams incorporating WB the
needles
continued to carry load after the applied load reached its peak value (Figure 7). Replacing 5% and
continued
average toequivalent
carry loadflexural
after the applied
strength load
for 5% andreached its peak value
10% replacements were (Figure
12.6 and7). Replacing
24.7, 5% and
respectively.
10% ofThe
coarse aggregate
comparison of by test
the volume with
results with the
a plain needles
number of past led toon
studies thefiber
increase of average
reinforce concrete equivalent
revealed
10% of coarse aggregate by volume with the plain needles led to the increase of average equivalent
flexural strength
that
flexural strength RT150
the equivalent
R,150 from
150 flexural
from1.1strength
to 14.0
1.1 andand
values
to 14.0 32.3, respectively.
for32.3,
the concrete When
in which
respectively. 10%grooved
When needles
ofgrooved
NA is replaced were used
plainthe
withwere
needles used
T,150
average wind blade needles
equivalent are strength
flexural similar to for
concretes
5% incorporating (1) steel fibers with volumetric dosages in
the average equivalent flexural strength forand
5% 10%10%
and replacements
replacements were 12.612.6
were andand24.7, respectively.
24.7, respectively.
the range of 0.2–1.0%; (2) glass fibers with volumetric dosages in the range of 1.0–2.0%; and (3)
The
Thecomparison
comparisonofofthethetest
testresults
results with
with a number
number of ofpast
paststudies
studieson onfiber
fiberreinforce
reinforce concrete
concrete revealed
revealed that
that
thethe equivalent
equivalent flexural
flexural strength
strength values
values for for
the the concrete
concrete in which
in which 10%10%of NAof NA is replaced
is replaced withwith plain
plain wind
wind
bladeblade needles
needles are similar
are similar to concretes
to concretes incorporating
incorporating (1)fibers
(1) steel steelwith
fibers with volumetric
volumetric dosagesdosages in
in the range
the
of range of 0.2–1.0%;
0.2–1.0%; (2) glass
(2) glass fibers withfibers with dosages
volumetric volumetric dosages
in the range in the rangeand
of 1.0–2.0%; of 1.0–2.0%; andfibers
(3) polymeric (3)
 needles glass fibers were transversely-aligned. It is expected that if the wind blade shell was cut along
the fibers, so that the fibers in all the needles were longitudinally-aligned, the flexural resistance of
the beam specimens would be notably higher.

4. Summary of the Past Studies

Recycling 2018, 3, 8 8 of 11
Compressive strength, tensile strength and flexural strength are among the main material-level
properties of concrete that significantly affect the performance of load bearing structural concrete
with volumetric
members dosages
and used in theby
frequently range of 0.4–0.5%.
structural designersTherefore,
a. The few the
pastreported
studies onincreases in toughness
recycled GFRP waste
caused by using needles are significant considering that in most of the wind blade
incorporated concretes have measured all or some of those properties. The findings from those needles glass fibers
were transversely-aligned.
studies are presented in It is expected
Table that if the
1. The results showwind
thatblade
usingshell was cut
recycled along
GFRP the fibers,
aggregates so that
have a
thedetrimental
fibers in alleffect on the mechanical
the needles performance of concrete.
were longitudinally-aligned, However,
the flexural recycled
resistance of GFRP needles
the beam have
specimens
significant
would contributions
be notably higher. on enhancing the overall mechanical performance of concrete. In particular,
GFRP needles with longitudinally aligned fibers enhance the tensile and flexural performance of
4. Summary of the Past
concrete. However, in Studies
the past investigations needles did not enhance the compressive strength of
concrete. Needles with round cross-sections tend to cause a small reduction in the compressive
Compressive strength, tensile strength and flexural strength are among the main material-level
strength of concrete.
properties of concrete that significantly affect the performance of load bearing structural concrete members
Figure 8 presents the change (increase or decrease) in the compressive strength and tensile
and used frequently by structural designers a. The few past studies on recycled GFRP waste incorporated
strength of concretes as result of incorporating recycled GFRP aggregates and needles in concrete
concretes have measured all or some of those properties. The findings from those studies are presented in
with different volumetric ratios (the ratios of GFRP volume to concrete volume). In the figure, the
Table 1. The results
diamond show that
shape markers using recycled
represent GFRP
the results of aggregates have with
testing concrete a detrimental effect on and
GFRP aggregates the mechanical
the dash
performance of concrete. However, recycled GFRP needles have significant contributions
markers represent those of concrete with GFRP needles. Although there is not a linear relationship on enhancing
thebetween
overall mechanical performance
the GFRP dosage of concrete.
and mechanical In particular,
performance, theGFRP
figureneedles
clearly with
shows longitudinally
that increasingaligned
the
fibers enhance
content the tensile
of recycled GFRPand flexural in
aggregates performance
concrete leadsof concrete. However,
to the reduction in the
in both the past investigations
compressive and
needles did
tensile not enhance
strengths the compressive
of concrete. Figure 8 strength
also showsof concrete. Needles with
that incorporating round
small cross-sections
amounts tend
of needles to to
cause a small
concrete reduction
can in the compressive
have significant strengthimpacts
(typically positive) of concrete.
on the tensile strength of concrete.

Figure 8. Change (increase or decrease) in the (a) compressive strength and (b) splitting tensile
Figure 8. Change (increase or decrease) in the (a) compressive strength and (b) splitting tensile strength
strength of concrete as a result of incorporating recycled GFRP aggregates and needles in concrete at
of concrete as a result of incorporating recycled GFRP aggregates and needles in concrete at different
different volumetric portions of concrete. The diamond shape markers represent the results of testing
volumetric portions of concrete. The diamond shape markers represent the results of testing concrete
concrete with GFRP aggregates and the dash markers represent those of concrete with GFRP needles.
with GFRP aggregates and the dash markers represent those of concrete with GFRP needles.

Figure 8 presents the change (increase or decrease) in the compressive strength and tensile strength
of concretes as result of incorporating recycled GFRP aggregates and needles in concrete with different
volumetric ratios (the ratios of GFRP volume to concrete volume). In the figure, the diamond shape
markers represent the results of testing concrete with GFRP aggregates and the dash markers represent
those of concrete with GFRP needles. Although there is not a linear relationship between the GFRP
dosage and mechanical performance, the figure clearly shows that increasing the content of recycled
GFRP aggregates in concrete leads to the reduction in both the compressive and tensile strengths
of concrete. Figure 8 also shows that incorporating small amounts of needles to concrete can have
significant (typically positive) impacts on the tensile strength of concrete.
Recycling 2018, 3, 8 9 of 11

Table 1. Results from the past studies on the mechanical properties of concrete with GFRP waste.

Coarse Aggregate GFRP to Concrete

Ref. Element Type Size mm Slump mm fc0 Mpa ft Mpa Flexural Strength MPa
Replacement Ratio (Vol.) % Ratio (Vol.) %
- - 0.0 0.0 160 50.03 2.32 4.13
Alam et al. 2013 [16] Flat GRA sl < 25 25.0 11.1 210 31.2 2.1 2.91
Flat GRA sl < 25 50.0 22.0 220 24.47 2.53 2.38
- - 0.0 0.0 - 40.5 - -
GRA sl = 13 50.0 7.8 - 22.3 - -
GRA sl = 13, 25 50.0 7.8 - 22.1 - -
GRA sl = 25 50.0 7.8 - 22.6 - -
Fox, 2016 [17] GRA sl = 25 25.0 3.9 - 31.2 - -
GRA sl = 25 37.5 5.9 - 25.0 - -
GRA sl = 25 50.0 7.8 - 22.1 - -
GRA sl = 25 32.5 5.1 - 39.2 - -
GRA sl = 25 32.5 5.1 - - 3.8 -
- - 0.0 0.0 - 37.5 4.0 -
GRA D = 19,25 40.0 15.4 - 32.8 3.0 -
Yazdanb-akhsh et al. GRA D = 6,10,13,19,25 100.0 38.5 - 29.5 2.6 -
2016 [14] - - 0.0 0.0 - 46.3 4.5 -
GRA D = 19, 25 40.0 16.0 - 40.4 4.0 -
GRA D = 6,10,13,19,25 100.0 40.0 - 36.6 3.6 -
- - 0.0 0.0 70 40.2 3.42 -
GRA D = 6,10,13,19 5.0 1.8 70 37.9 3.06 -
Yazdanb-akhsh et al.
GRA D = 6,10,13,19 10.0 3.5 65 38.9 3.42 -
2017 [28]
Needle D = 6 L = 100 5.0 1.8 75 38.0 4.18 -
Needle D = 6 L = 100 10.0 3.5 70 36.7 4.55 -
125
- - 0.0 0.0 36.0 3.35 5.0
160
L = 100 115
Plain needle 5.0 1.8 35.5 2.89 4.5
sl = 6 110
Yazdanb-akhsh, L = 100 125
2018 [29] Plain needle 10.0 3.5 36.1 3.35 4.6
sl = 6 110
Grooved L = 100 115
5.0 1.8 38.6 2.91 5.0
needle sl = 6 120
Grooved L = 100 110
10.0 3.5 34.4 3.09 4.7
needle sl = 6 95
Notes: L represents length, D represents diameter, sl is the approximate average side length of cubic and rectangular sections, fc0 and ft are the 28-day compressive and tensile strengths of concrete.
Recycling 2018, 3, 8 10 of 11

5. Guidance for Future Research

The findings of the past studies on using coarse processed GFRP waste in concrete show that
recycled needles can be efficient discrete reinforcing elements. Extensive research needs to be conducted
to optimize the production and performance of recycled needles. Needles can be produced from GFRP
waste only by means of cutting. Cutting technologies need to be developed that minimize the energy
and the time required for producing recycled needles. In addition, experimental and theoretical studies
need to be performed to optimize the geometry of needles to achieve the desired properties of concrete
in load-carrying members. Durability of recycled GFRP needles embedded in concrete, which is highly
alkaline and the long-term mechanical performance of concrete incorporating recycled GFRP needles
need to be studied. Finally, assessing the environmental and economic impacts of different alternatives
for managing and recycling GFRP waste, in addition to investigating the performance of recycled
waste, leads to finding the most efficient and environmentally-conscious practices for the production
and reuse of GFRP composite materials.

Acknowledgments: This work was supported by the New York State Energy Research and Development Authority
(NYSERDA) under the grant C2CUNY1 by the PowerBridgeNY Program, and by the U.S. National Science Foundation
(NSF) under the grant 1701694.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Reynolds, N.; Pharaoh, M. An introduction to composites recycling. In Management, Recycling and Reuse of
Waste Composites; Goodship, V., Ed.; Woodhead Publishing: Cambridge, UK, 2010; pp. 3–19.
2. Witten, E. Composites Market Report 2015 Market Developments, Trends, Outlook and Challenges; Federation of
Reinforced Plastics (AVK): Frankfurt, Germany, 2015.
3. Holmes, M. US demand for fibre reinforced plastic composites to rise. In Reinforced Plastics; Elsevier: Oxford, UK, 2013;
Available online: http://www.materialstoday.com/composite-industry/news/us-demand-for-fibre-
reinforced-plastic-composites (accessed on 13 February 2018).
4. Bank, L.C. Composites for Construction; John Wiley & Sons: Hoboken, NJ, USA, 2006.
5. Beauson, J.; Lilholt, H.; Brøndsted, P. Recycling solid residues recovered from glass fibre-reinforced composites
—A review applied to wind turbine blade materials. J. Reinf. Plast. Compos. 2014, 33, 1542–1556. [CrossRef]
6. Hardee, C. Improving wind blade manufacturability. Wind Syst., 2 February 2018; 26–30.
7. Hunt, H.; Wanner, C.; Hensley, J. U.S. Wind Industry Annual Market report—Year Ending 2015; American Wind
Energy Association (AWEA): Washington, DC, USA, 2016.
8. Yazdanbakhsh, A.; Bank, L. A critical review of research on reuse of mechanically recycled FRP production
and end-of-life waste for construction. Polymers 2014, 6, 1810–1826. [CrossRef]
9. Job, S.; Leeke, G.; Mativenga, P.T.; Oliveux, G.; Pickering, S.; Shuaib, N.A. Composite Recycling—Where Are We Now?
Composites UK: Hertfordshire, UK, 2016.
10. Tittarelli, F.; Kawashima, S.; Tregger, N.; Moriconi, G.; Shah, S.P. Effect of GRP by-product addition on
plastic and hardened properties of cement mortars. In Proceedings of the 2nd International Conference on
Sustainable Construction Materials and Technologies, Ancona, Italy, 28–30 June 2010.
11. Tittarelli, F.; Shah, S.P. Effect of low dosages of waste GRP dust on fresh and hardened properties of mortars: Part 1.
Constr. Build. Mater. 2013, 47, 1532–1538. [CrossRef]
12. Asokan, P.; Osmani, M.; Price, A.D.F. Assessing the recycling potential of glass fibre reinforced plastic waste
in concrete and cement composites. J. Clean. Prod. 2009, 17, 821–829. [CrossRef]
13. Gu, L.; Ozbakkaloglu, T. Use of recycled plastics in concrete: A critical review. Waste Manag. 2016, 51, 19–42.
[CrossRef] [PubMed]
14. Yazdanbakhsh, A.; Bank, L.C.; Chen, C. Use of recycled FRP reinforcing bar in concrete as coarse aggregate
and its impact on the mechanical properties of concrete. Constr. Build. Mater. 2016, 121, 278–284. [CrossRef]
15. ASTM C33. Standard Specification for Concrete Aggregates; STM International: West Conshohocken, PA, USA, 2013.
16. Shahria Alam, M.; Slater, E.; Billah, A.M. Green concrete made with RCA and FRP scrap aggregate: Fresh and
hardened properties. J. Mater. Civ. Eng. 2013, 25, 1783–1794. [CrossRef]
Recycling 2018, 3, 8 11 of 11

17. Fox, T.R. Recycling wind turbine blade composite material as aggregate in concrete. In Department of Industrial and
Manufacturing Systems Engineering; Iowa State University: Ames, IA, USA, 2016.
18. Murrell, S. Concrete Airport Pavements Make a Comeback; New York Construction News; McGraw Hill: New York,
NY, USA, 1993.
19. Roesler, J.R.; Altoubat, S.A.; Lange, D.A.; Rieder, K.-A.; Ulreich, G.R. Effect of synthetic fibers on structural
behavior of concrete slabs-on-ground. ACI Mater. J. 2006, 103, 3–10.
20. Soranakom, C.; Mobasher, B.; Destrée, X. Numerical Simulation of FRC Round Panel Tests and Full-Scale Elevated Slabs;
ACI Special Publication: Farmington Hills, MI, USA, 2007; Volume 248.
21. Bentur, A.; Mindess, S. Fiber Reinforced Cementitious Composites, 2nd ed.; Modern Concrete Technology; Taylor & Francis:
New York, NY, USA, 2007.
22. Behnood, A.; Verian, K.P.; Gharehveran, M.M. Evaluation of the splitting tensile strength in plain and steel
fiber-reinforced concrete based on the compressive strength. Constr. Build. Mater. 2015, 98, 519–529. [CrossRef]
23. Yazdanbakhsh, A.; Altoubat, S.; Rieder, K.A. Analytical study on shear strength of macro synthetic fiber reinforced
concrete beams. Eng. Struct. 2015, 100, 622–632. [CrossRef]
24. Jeknavorian, A.A. Innovations in chemical admixture technology as related to sustainability. In ACI Spring
2012 Convention; American Concrete Institute: Dallas, TX, USA, 2012.
25. Choi, Y.; Yuan, R.L. Experimental relationship between splitting tensile strength and compressive strength of
GFRC and PFRC. Cem. Concr. Res. 2005, 35, 1587–1591. [CrossRef]
26. Nanni, A. Splitting-tension test for fiber reinforced concrete. ACI Mater. J. 1988, 85, 229–233.
27. Wafa, F.; Ashour, S. Mechanical properties of high-strength fiber reinforced concrete. ACI Mater. J. 1992, 89, 449–455.
28. Yazdanbakhsh, A.; Bank, L.C.; Chen, C.; Tian, Y. FRP-Needles as discrete reinforcement in concrete. ASCE J. Mater.
Civ. Eng. 2017, 29, 1–9. [CrossRef]
29. Yazdanbakhsh, A.; Bank, L.C.; Rieder, K.-A.; Tian, Y.; Chen, C. Concrete with discrete slender elements from
mechanically recycled wind turbine blades. Resourc. Conserv. Recycl. 2017, 128, 11–21. [CrossRef]
30. ASTM C1609. Standard Test Method for Flexural Performance of Fiber-Reinforced Concrete (Using Beam with
Third-Point Loading); ASTM International: West Conshohocken, PA, USA, 2012.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).