fibers

Article
Effect of Fabric Integration on the Physical and
Optical Performance of Electroluminescent Fibers
for Lighted Textile Applications
Alyssa Martin * ID
and Adam Fontecchio
Department of Electrical, Computer Engineering, Drexel University, Philadelphia, PA 19104, USA;
[email protected] or [email protected] or [email protected]
* Correspondence: [email protected]; Tel.: +1-(215)-895-3232




Received: 20 April 2018; Accepted: 13 July 2018; Published: 17 July 2018


Abstract: The advent of electroluminescent (EL) fibers, which emit light in response to an applied
electric field, has opened the door for fabric-integrated light emission and displays in textiles.
However, there have been few technical publications over the past few years about the performance
of these light emitting fibers inside functional fabrics. Thus, there is limited information on the effect
of integration on the physical and optical performance of such devices. In this work, alternating
current powder-based EL (ACPEL) fibers were evaluated under a range of operating conditions both
inside and outside of a knit matrix to understand how the EL fiber device performance changed
inside a functional fabric. The device efficiency, adjustable brightness, and mechanical properties of
these fibers are presented. The effects of fabric integration on the light-emitting fibers as well as the
supporting knit fabric are discussed as they relate to the practical applications of this technology.

Keywords: smart fabric; light-emitting textile; electroluminescent fiber; light-emitting fiber; physical
characterization; optical characterization

1. Introduction
The merging of electronics and textiles has given rise to garments and upholstery with new abilities
such as sensing, biomedical monitoring, power storage, movement, and communication. Within this
smart fabric field, light emitting fibers are garnering great attention for applications in fashion,
entertainment, optical physiological monitoring, safety lighting in garments, and automotive and
aircraft interior lighting [1]. The integration of electronics and textiles has typically been accomplished by
mounting prefabricated devices into garments [2,3] and incorporating discrete components (e.g., sensors,
batteries, controller chips) with laminated or knit conducting interconnects [4,5] that compromise the
most desirable characteristics of the textile including conformability, softness, strength, and washability.
Therefore, the smart fabric arena is rapidly pushing towards fabric-integrated electronic systems and
fibers and fabric structures exhibiting electro-optic properties [6,7].
A number of light emitting fibers have emerged in the literature including side-emitting optical
fibers [8–10], mechanoluminescent fibers [11], electroluminescent (EL) fibers [12–19], and fibers
exhibiting photoluminescence [20–23]. Electroluminescent and photoluminescent fibers emit light
upon the application of electrical or optical power, respectively. Both optical and photoluminescent (PL)
fibers require an external light source for illumination; optical fibers must be illuminated at one end by a
luminescent source and PL fibers must be charged by the sun or external luminescent source. Fibers that
emit light via electroluminescence are especially desirable for lighted fabrics and fabric-integrated
displays due to the fast switching times, inherent luminescence, and adjustable brightness of these
electrically controllable devices [24]. EL fibers based on inorganic powder phosphor [17], organic [15],

Fibers 2018, 6, 50; doi:10.3390/fib6030050 www.mdpi.com/journal/fibers

Fibers 2018, 6, 50 2 of 13

and polymer [12,13,16,18] material systems have been reported in literature. Table 1 compares the
optical performance and bending radius of fibers fabricated from different material systems.

Table 1. Comparison of electroluminescent fiber properties reported in literature.

Maximum Emission Turn On Bending

Type of Fiber Wavelength Reference
Intensity Voltage Radius
ACPEL 485 nm 49.39 cd/m2 50 V 3.1 cm [17,25]
LEC (iTMC) 855–984 nm 23 cd/m2 4.2 V 6 mm [13]
OLED N/A 104 mW/cm2 ~0.5 V N/A [15]
pLEC N/A 125 cd/m2 4.2 V N/A [12]
PLED (MEH-PPV) ~575 nm N/A 15–16 V N/A [18]
PLED (SY solution) ~550 nm 1458.8 cd/m2 5V 2.5 mm [16]

Despite the higher efficiency and flexibility of some of these other fibers, complicated fiber
fabrication processes and the sensitivity of the resulting fibers to handling and environmental factors
like heat, moisture, and humidity have prevented many of the reported EL fibers from achieving full
fabric-integration [12,13,16,24,26]. The only fibers to be integrated into a knit fabric are those fabricated
by Dias et al. [17], which were inlaid into the knit fabric, and Coyle et al. [18], which required the fibers
to be placed under tension while testing to maintain contact between the two fibers acting as the top
and bottom electrodes in the electroluminescent structure. The fibers produced by Zhang et al. [12]
and Kwon et al. [27] produced light while bent, but were not placed inside a fabric structure for
testing. A more detailed comparison of fibers produced by these material systems exists in previous
literature [25]. The ACPEL material set has demonstrated reliable light emission over large surface
areas in planar films and coated fibers due to the robustness of the thick layers [28]. This materials
system was chosen for testing due to its simplicity of fabrication, robustness of the thick layers,
long lifetimes, and brightness visible to the naked eye.
However, a lot is still unknown about the effect fabric integration has on the physical and
optical performance characteristics of these fibers. In this work, the optical, electrical, and mechanical
performance of ACPEL fibers inside and outside of a knit fabric were compared to understand the
effect of integration on both the light-emitting fibers and knit fabric. The flexibility, robustness,
brightness, and power requirements of the light-emitting fabric resulting from the integration of
ACPEL fibers into a knit fabric are important in identifying the applications and limitations of the
system. These properties and the methods of obtaining them for EL fibers inside and outside of a
fabric are reported in this work. These methods provide a foundation for experimentally quantifying
and comparing the performance of fabric-integrated EL fibers.

2. Materials and Methods

2.1. Fiber Fabrication

Electroluminescent (EL) fibers were fabricated by coating an alternating current phosphor-based
EL (ACPEL) structure onto a supporting conductive fiber. The cross-sectional ACPEL device structure,
depicted in Figure 1, consisted of a conductive bottom electrode, an isolation layer with a high dielectric
constant to focus the electric field on the emitting layer and protect it from heating at this electrode,
an emitting layer, and a top electrode layer that was translucent to allow light through. The Dupont
Luxprint® material system was used due to its robustness, simplicity of deposition, and ease of
handling after curing. The solution-processed layers were deposited in the following order onto an
Aracon® XS0400E-018 silver-coated Kevlar® yarn (manufactured by Micro-coax, Inc., Pottstown, PA,
USA): Dupont Luxprint® 8153 dielectric paste, Dupont Luxprint® 8154L phosphor paste, and Dupont
Luxprint® 7164 transparent conductive paste. Upon application of the first isolation layer, the dielectric
paste filled the gaps between the individual silver coated fibers within the underlying conductive yarn,
Fibers 2018, 6, x FOR PEER REVIEW 3 of 13
Fibers 2018, 6, 50 3 of 13

the underlying conductive yarn, creating a monofilament onto which subsequent layers were
deposited.
creating The individualonto
a monofilament strands of subsequent
which the supporting conductive
layers yarn were
were deposited. randomly
The individualdispersed
strands ofin the
dielectric
supporting coating layer.
conductive yarn were
Fibers 2018, 6, x FOR PEER REVIEW randomly dispersed in the dielectric coating layer. 3 of 13

the underlying conductive yarn, creating a monofilament onto which subsequent layers were
deposited. The individual strands of the supporting conductive yarn were randomly dispersed in the
dielectric coating layer.

Figure
Figure 1.
1. Schematic
Schematic diagrams
diagrams of
of the
the ACPEL
ACPEL fiber structure.
fiber structure.

Slot die devices, like that Figureshown in Figure

1. Schematic 2, with
diagrams of thevarying
ACPEL fibercylindrical
structure.die openings were used to
Slotthe
control diedeposition
devices, like that shown
of each layer onto in Figure 2, with varying
the previous layer of cylindrical die openings
the device structure. were
These used
devices
to control the deposition
Slot diein
were described devices, of
detail like each layer
that shown
in previous workonto the
in Figure previous
2, with
[29]. The layer
varying
fibers of the
werecylindrical device
cured inside structure.
die openings
an oven were These devices
used to the
to evaporate
were described
control the in detail
deposition in
of previous
each layer work
onto [29].
the The
previous fibers
layer were
of thecured
solvent from the fluid coating material after deposition. The dielectric and phosphor pastesdevice inside an
structure. oven
These to evaporate
devices were
the solvent from
were described theinfluid
detailcoating material
in previous work after deposition.
[29]. The fibers were The dielectric
cured inside an
cured at 130 °C for 15 min, while the translucent conductive paste took only 5 min to cure at the sameand phosphor
oven pastes
to evaporate the were
curedsolvent ◦ C for
at 130 from the
15fluid coating
min,many
whileothermaterial
the after deposition.
translucent conductive The dielectric
paste took and phosphor
only 5 [24],
min to pastesatwere
cure the same
temperature [30]. Unlike organic and polymer emissive systems this material set is
cured
temperature at 130 °C for
[30]. Unlike15 min, while the
manyroughness, translucent
other organic conductive
and and
polymer paste took
emissive only 5
systems min to
[24],cure
thisat the same
material set is
not highly sensitive to surface oxygen, particulates from the
temperature [30]. Unlike many other organic and polymer emissive systems [24], this material set isair due to its thick layers.
not highly
Therefore, sensitive
entiretofabrication
the sensitive surface roughness,
process can oxygen, and
take and
place particulates
in air and from
outsidethe airclean
due room,
to its thick layers.
not highly to surface roughness, oxygen, particulates from the airadue which
to its thick layers. is a
Therefore, the
greatTherefore,
advantage entire
for fabrication
scaling process
production can take
of the place in air and outside a clean room, which is a great
the entire fabrication process can fibers andin
take place using them
air and in commercial
outside a clean room, applications.
which is a
advantage for scaling production of the fibers and using them in commercial applications.
great advantage for scaling production of the fibers and using them in commercial applications.

Figure 2. Model of slot die devices used to deposit ACPEL layers onto supporting yarn substrate [29].
Figure 2. Model of slot die devices used to deposit ACPEL layers onto supporting yarn substrate [29].
Figure 2. Model of slot die devices used to deposit ACPEL layers onto supporting yarn substrate [29].
Fibers 2018, 6, 50 4 of 13
Fibers 2018, 6, x FOR PEER REVIEW 4 of 13

2.2. Power
2.2. Power Application
Application and
and Fabric
Fabric Integration
Integration
An Agilent
An Agilent 33220A
33220A arbitrary
arbitrary waveform
waveform generator
generator (AWG)
(AWG) amplified
amplified by by aa TREK
TREK model
model PZD700PZD700
high voltage amplifier supplied power to the fibers. Power was applied
high voltage amplifier supplied power to the fibers. Power was applied between the top electrode between the top electrode and
across
and all individual
across fibers of
all individual the supporting
fibers conductive
of the supporting yarn electrode
conductive simultaneously
yarn electrode to illuminate
simultaneously to
the fiber. This was accomplished by twisting the individual fibers
illuminate the fiber. This was accomplished by twisting the individual fibers of the supportingof the supporting conductive yarn
together andyarn
conductive contacting
together themandusing flat alligator
contacting clips as
them using shown
flat in Figure
alligator clips as3a.shown
The more tightly3a.
in Figure packed
The
the individual fibers of the supporting yarn electrode, the more efficient
more tightly packed the individual fibers of the supporting yarn electrode, the more efficient the the devices because there
is less power
devices because lostthere
to the is dielectric
less power isolation layer.
lost to the However,
dielectric as long
isolation as the
layer. average as
However, isolation
long aslayer the
thickness between the emissive layer and conductive fibers of the supporting
average isolation layer thickness between the emissive layer and conductive fibers of the supporting yarn closest to that
emissive
yarn closestlayer
to has 35 ± 10 layer
that aemissive µm tolerance,
has a 35 ±light10 μmemission across
tolerance, theemission
light yarn willacross
appear theuniform
yarn will to
an observer. This thickness is controlled by the diameter of the cylindrical
appear uniform to an observer. This thickness is controlled by the diameter of the cylindrical opening opening of the slot die
coating
of the slotdevice.
die coating device.
The illuminated
The illuminated fiber fiber in in Figure
Figure 3b 3b had
had non-uniform
non-uniform emission
emission due due to to an
an uneven
uneven dispersion
dispersion of of
supporting conductive fibers. The thicknesses of the layers has been
supporting conductive fibers. The thicknesses of the layers has been reported in previous work [29].reported in previous work [29].
Although the
Although the total
total thickness
thickness of of the
the isolation
isolation layer
layer was
was consistent
consistent across,
across, the
the parts
parts ofof thethe fiber
fiber that
that
appear dark had an isolation layer thickness between the emissive layer
appear dark had an isolation layer thickness between the emissive layer and conductive fibers of the and conductive fibers of the
supporting yarn greater than 50 µm. These dark areas were eliminated
supporting yarn greater than 50 μm. These dark areas were eliminated by keeping the fibers of the by keeping the fibers of the
supporting yarn
supporting yarn tightly
tightly twisted
twisted usingusing clamps
clamps during
during deposition
deposition and and curing
curing to to control
control the the dispersion
dispersion
of the fibers in the isolation
of the fibers in the isolation layer. layer.
A cm ××33cm
A 33 cm cmweft
weftknit knitfabric
fabriccomposed
composedof ofaa3-ply
3-plycotton
cottonyarnyarn(Size
(Size1010 crochet
crochet cotton
cotton thread,
thread,
Coats & Clark, Inc., Charlotte, NC, USA) with a gauge of 4 rows and 3 stitches = 1 cm was hand hand
Coats & Clark, Inc., Charlotte, NC, USA) with a gauge of 4 rows and 3 stitches = 1 cm was knit.
knit.
A A single
single fiberinlaid
fiber was was inlaid
into the into the fabric
fabric as shown
as shown in Figure
in Figure 3c. The 3c.conductive
The conductiveends of ends
the of the fibers
fibers were
were accessible
accessible on eitheron either
end of end of thesample,
the fabric fabric sample, which allowed
which allowed power topower to be using
be applied applied flatusing flat
alligator
alligator
clips. clips.

(a)

(b) (c)
Figure 3. Macro images of the ACPEL fibers (a) outside of knit indicating where electrodes are
Figure 3. Macro images of the ACPEL fibers (a) outside of knit indicating where electrodes are
connected
connected to
to fiber
fiber during
during testing,
testing, (b)
(b) illuminated
illuminated at
at 100
100 V
Vrms, 400 Hz, and (c) inlaid into 3 cm 2 knit
rms , 400 Hz, and (c) inlaid into 3 cm 2 knit
fabric sample.
fabric sample.

2.3. Optical Characterization Methods

2.3. Optical Characterization Methods
Electrical, optical, mechanical tests were performed on EL fibers inside and outside of the cotton
Electrical, optical, mechanical tests were performed on EL fibers inside and outside of the cotton
knit matrix. For each test, a 3-cm segment of fiber was tested due to space constraints in the fabrication
knit matrix. For each test, a 3-cm segment of fiber was tested due to space constraints in the fabrication
and testing set ups. However, using a longer continuous segment of fiber or multiple fibers would
increase the light output and effect the mechanical strength and flexibility of the light-emitting fabric.
Optical testing was performed inside a 12-inch diameter Gamma Scientific integrating sphere to
eliminate any geometric and ambient light effects on the intensity of light. The intensity of the emitted
Fibers 2018, 6, 50 5 of 13

and testing set ups. However, using a longer continuous segment of fiber or multiple fibers would
increase the light output and effect the mechanical strength and flexibility of the light-emitting fabric.
Optical testing was performed inside a 12-inch diameter Gamma Scientific integrating sphere to
eliminate any geometric and ambient light effects on the intensity of light. The intensity of the emitted
light was measured by an Ocean Optics USB-4000 spectrometer attached via fiber optic cable to the
integrating sphere and Spectrasuite software was used to process the data. Prior to beginning each
optical measurement, a voltage waveform with amplitude below the threshold voltage of the device
was applied to the fiber for half an hour to allow the device to reach the steady-state.
The voltage, frequency, and shape of the applied waveform can affect the optical output of the
device. ACPEL devices are most efficient during the rising edge portion of the applied waveform,
and least efficient when the voltage is held constant above the threshold voltage. A sine wave was used
to illuminate the ACPEL fiber devices except for the device efficiency measurements. The threshold
voltage of the EL fiber device was 49.833 V, which is independent of frequency. This voltage is the
magnitude of power necessary to accelerate electrons inside the EL layer to speeds high enough to
produce light via impact ionization with phosphor particles in the emitting layer.
The frequency of the applied waveform plays a large role in the brightness of light emission in
inorganic phosphor devices. The field must switch fast enough to have continuous excitation and
decay of electrons, but cannot exceed the lifetime of electrons. The brightness-frequency (B-F) curve
was determined by applying waveforms with increasing frequency to the device, while keeping the
amplitude constant, and measuring the luminance at each increased frequency. The frequency was
increased in increments of 100 Hz from 0 Hz to 50 kHz, while the voltage remained constant at 100
V. A constant voltage of 100 V was selected for the frequency sweep as it is in the middle of the fiber
operating range. The brightness-voltage (B-V) curve was determined by applying waveforms with
increasing amplitude (higher voltage) at a constant frequency of 400 Hz to the device and measuring
the luminance at each successive amplitude. The voltage sweep was performed on fibers at a constant
frequency of 400 Hz, while the voltage was increased in increments of 10 V from 0 V to 230 V. To
determine if the knitted matrix absorbed or blocked any light from the fibers, voltage and frequency
sweeps were performed on fibers inlaid into and outside a knitted fabric matrix.

2.4. Device Efficiency Calculation

A bipolar trapezoidal waveform, shown in Figure 4, was used to drive the fiber in the circuit as
there are distinct points on the waveform that can be identified and plotted to obtain a Q-V curve.
These points have been explained in great detail in prior work [25,31]. Most of the points designated
where the applied voltage began increasing and decreasing, except for B and G, which represent
the points in the waveform where the device began conducting charge, commonly called the
turn-on voltage.
The device efficiency can be derived from the charge-voltage (Q-V) curve of the ACPEL fibers.
The density of electrical power delivered across the ACPEL device per pulse is equal to the area
inside the Q-V curve. This measurement divided by the input power equals the device efficiency.
The Sawyer-Tower method is commonly used to determine this curve in EL devices [32,33]. In this
method, the voltage drop across a sense element, typically a capacitor or resistor, is monitored to gain
information about device operation. The Sawyer-Tower circuit is depicted in Figure 5. The signal
applied to the circuit consisted of a 1 kHz sequence of bipolar pulses with 100 V amplitude, rise,
and fall times of 5 µs and a pulse width of 30 µs.
 A bipolar trapezoidal waveform, shown in Figure 4, was used to drive the fiber in the circuit as
there are distinct points on the waveform that can be identified and plotted to obtain a Q-V curve.
These points have been explained in great detail in prior work [25,31]. Most of the points designated
where the applied voltage began increasing and decreasing, except for B and G, which represent the
points
Fibers in50the waveform where the device began conducting charge, commonly called the turn-on
2018, 6, 6 of 13
voltage.

Fibers 2018, 6, x FOR PEER REVIEW 6 of 13

The device efficiency can be derived from the charge-voltage (Q-V) curve of the ACPEL fibers.
The density of electrical power delivered across the ACPEL device per pulse is equal to the area inside
the Q-V curve. This measurement divided by the input power equals the device efficiency. The
Sawyer-Tower method is commonly used to determine this curve in EL devices [32,33]. In this
method, the voltage drop across a sense element, typically a capacitor or resistor, is monitored to gain
information about device operation. The Sawyer-Tower circuit is depicted in Figure 5. The signal
applied to the circuit
Figure 4. consisted of a 1 kHz
Bipolar trapezoid sequence
waveform of bipolar
indicating pulses with
important 100 V amplitude,
measurement points. rise, and
Figure
fall times of 4. Bipolar
5 µs and a pulsetrapezoid
width of waveform
30 µs. indicating important measurement points.

Figure 5. Schematic diagram of Sawyer-Tower measurement circuit set up.

Figure 5. Schematic diagram of Sawyer-Tower measurement circuit set up.
The output of the high-voltage amplifier drives the characterization circuit, which is a series
The outputofofa the
combination high-voltage
resistor, the ACPELamplifier drives
fiber device, and a the
sensecharacterization
element. The sense circuit,
element which
in thisis a series
case
was a capacitor,
combination which the
of a resistor, wasACPEL
chosen to be much
fiber larger
device, andthan the capacitance
a sense element. Theof the ACTFEL
sense elementdevice to case
in this
minimize its effect on the measurement. The resistor acts as a current-limiter to
was a capacitor, which was chosen to be much larger than the capacitance of the ACTFEL device toprotect the ACPEL
fiber device
minimize fromon
its effect catastrophic failure. TheThe
the measurement. capacitance
resistor of a pixel
acts as ain the 3 cm segment
current-limiter to of ACPEL
protect fiber
the ACPEL
was measured with a GW Instek LCR-819 LCR meter as 0.27 nF, so an 82 nF capacitor was used as
fiber device from catastrophic failure. The capacitance of a pixel in the 3 cm segment of ACPEL fiber
the sense element (Csense) in the characterization circuit. The resistor acts as a current-limiter to protect
was measured with a GW Instek LCR-819 LCR meter as 0.27 nF, so an 82 nF capacitor was used as the
the ACPEL fiber device from catastrophic failure, so a 1.5 kΩ resistor was used. The voltages V2 and
senseVelement (C6 and) in
3 in Figures sense
the characterization
7 were circuit. MSO-X
monitored by an Agilent The resistor
2014Aacts as a current-limiter
oscilloscope to protect the
during testing.
ACPEL fiber device from catastrophic failure, so a 1.5 kΩ resistor was used. The voltages V2 and V3 in
Figures 6 and 7 were monitored by an Agilent MSO-X 2014A oscilloscope during testing.
 minimize its effect on the measurement. The resistor acts as a current-limiter to protect the ACPEL
fiber device from catastrophic failure. The capacitance of a pixel in the 3 cm segment of ACPEL fiber
was measured with a GW Instek LCR-819 LCR meter as 0.27 nF, so an 82 nF capacitor was used as
the sense element (Csense) in the characterization circuit. The resistor acts as a current-limiter to protect
the ACPEL fiber device from catastrophic failure, so a 1.5 kΩ resistor was used. The voltages V2 7and
Fibers 2018, 6, 50 of 13
V3 in Figures 6 and 7 were monitored by an Agilent MSO-X 2014A oscilloscope during testing.

Figure 6. Effect of frequency on the luminance of the inlaid fiber (Voltage = 100 VAC).
Fibers 2018, 6, xFigure 6. Effect
FOR PEER REVIEWof frequency on the luminance of the inlaid fiber (Voltage = 100 VAC). 7 of 13

Figure 7. Comparison of the theoretically and experimentally derived relationship between

Figure 7. Comparison of the theoretically and experimentally derived relationship between brightness
brightness and voltage of an ACPEL fiber inside a knit fabric.
and voltage of an ACPEL fiber inside a knit fabric.

The voltage drop across the sense capacitor, V3(t), is related to the amount of charge stored on
The voltage
the external drop across
terminals of the the sensefiber
ACPEL device,Vq3 (t),
capacitor, is related to the amount of charge stored on the
ext(t), and described by
external terminals of the ACPEL fiber device, qext (t), and described by
𝑞 𝑡 =𝐶 𝑉 𝑡 (1)
qext (t) = Csense V3 (t) (1)
The value of qext(t) is also called the instantaneous external charge because it is a transient charge
measured externally with respect to the phosphor of the ACPEL device. The Q-V characteristics are
The value of qext (t) is also called the instantaneous external charge because it is a transient charge
obtained by plotting this instantaneous electric charge against the instantaneous voltage drop across
measured externally with respect to the phosphor of the ACPEL device. The Q-V characteristics are
the ACPEL fiber device, Vi(t), which is determined by
obtained by plotting this instantaneous electric charge against the instantaneous voltage drop across
the ACPEL fiber device, Vi (t), which is𝑉determined
𝑡 =𝑉 𝑡 − by𝑉 𝑡 (2)

Vi (t) = V2 (t) − V3 (t) (2)

2.5. Mechanical Characterization Methods
The force vs. displacement curves of the EL fibers inside and outside of the knitted matrix and
the bending radius of the fibers were previously reported [34] and demonstrated that the fibers were
not flexible enough to be used in garments. To further explain the effect in both the knitted fabric and
EL fiber after integration and determine what applications the fibers can be used for, flexural stress-
Fibers 2018, 6, 50 8 of 13

2.5. Mechanical Characterization Methods

The force vs. displacement curves of the EL fibers inside and outside of the knitted matrix and the
bending radius of the fibers were previously reported [34] and demonstrated that the fibers were not
flexible enough to be used in garments. To further explain the effect in both the knitted fabric and EL
fiber after integration and determine what applications the fibers can be used for, flexural stress-strain
curves were derived from the three-point flexural test detailed by ASTM D6856/D6856M [35,36].
Experimental tests were performed with a Mark-10 Force Gauge model MS-20 on fibers inside and
outside of a fabric matrix. Details of this testing set up are given in a previous work [34].
Equations (3) and (4) are based on homogeneous beam theory and define the flexural stress for
a circular cross section (σf ) and flexural strain (εf ) on the outer surface of the fiber, respectively [36].
In these equations, F is applied load, L is the support span, R is the radius of the fiber, D is the deflection
of the fiber, and d is the diameter of the beam [37].

FL
σf = (3)
πR3
6Dd
εf = (4)
L2
These equations were used to calculate the flexural stress-strain response of fibers inside and
outside of the knit fabric as an increasing load was applied. The three-point bend test method
is sensitive to the geometry of the fiber, and the accuracy of the calculated values depends upon
the ratio of the beam length (L) to the cross-sectional height of the fiber (h), called the span to
thickness ratio (L/h). A L/h ratio of at least 20:1 is recommended for determining the flexural
modulus of fiber-reinforced composites [36]. During experimentation, the L/h ratio was 32:1, which is
in the recommended range of values for determining the flexural modulus according to ASTM
D6856/D6856M.28 [35].

3. Results

3.1. Optical Characterization

Figure 6 shows the results of the frequency sweep. The intensity of light increased with increasing
frequency until approximately 45 kHz, when darkening at the center began to appear and the overall
intensity decreased. This darkening was a result of the increase in degradation of the electroluminescent
devices under very high frequencies. There is minimal gain in luminance with a frequency above
5 kHz and continuous operation of the fiber devices at higher frequencies will decrease the lifetime of
the fiber, so it is better to run the device at frequencies below this point [30].
The relationship between brightness and voltage is well defined in the literature for ACPEL
material systems and is described by Equation (5), where U is the applied voltage and A and c are
empirical constants [38,39]. These empirical constants are dependent on the material properties of the
ACPEL system, and are unique to the material under investigation. Once the empirical constant has
been derived, the brightness of the fiber at different voltages can be theoretically predicted for this
fiber. Alternatively, one can predict what voltage would be needed to achieve a desired brightness
from the material and determine if these fibers will fit into an application based on the limitations of
the device.  
c
B = Aexp − 1 (5)
U2
The empirical constants of the ACPEL material system used in this work were derived using a
nonlinear regression analysis to fit the equation to the average of five experimental voltage sweeps
performed on a single pixel in the display. Figure 7 demonstrates the accuracy of the brightness-voltage
(B-V) curve predicted by Equation (5), where the empirical constants A and c for this ACPEL system
were calculated as 22.2678 and 8.7396, respectively. The error bars at each voltage increment of the
Fibers 2018, 6, 50 9 of 13

experimentally derived curve show the standard deviation to the average of the five voltage sweeps
performed on the pixel. The difference in luminance of a fiber inside and outside of a knitted fabric
matrix was measured as 0.00015 ± 0.00005 cd/m2 regardless of the voltage or frequency applied.

3.2. Device Efficiency

Figure 8 depicts the Q-V curve of an EL fiber at 100 V. The density of power delivered to the
device per pulse was 323.64 W/cm3 . The power efficiency of the device could then be derived by
dividing this delivered power by the input power, which gave an efficiency of 0.016% for this device at
40 V above threshold. This efficiency depends upon a number of factors including the concentration of
phosphor particles in the emitting layer, applied waveform, length of the fiber, environmental factors,
and the dielectric constant of both the isolation layer and suspension medium of the emitting layer [31].
Thus,2018,
Fibers the 6,efficiency
x FOR PEERofREVIEW
the device can vary over time and under different operating conditions. 9 of 13

Figure 8. Charge-Voltage (Q-V) curve of the 3 cm ACPEL fiber at 100 V.

Figure 8. Charge-Voltage (Q-V) curve of the 3 cm ACPEL fiber at 100 V.

3.3. Mechanical Characterization

3.3. Mechanical Characterization
The curves resulting from the mechanical tests, shown in Figure 9, indicated that the knitted
The curves resulting from the mechanical tests, shown in Figure 9, indicated that the knitted
matrix absorbed some of the applied stress and allowed the inlaid fiber to withstand a much higher
matrix absorbed some of the applied stress and allowed the inlaid fiber to withstand a much higher
load than the fiber on its own. The flexural strength is a measure of how much force or pressure is
load than the fiber on its own. The flexural strength is a measure of how much force or pressure is
needed to break the fibers, which is quantified by the maximum stress before fiber failure. Assuming
needed to break the fibers, which is quantified by the maximum stress before fiber failure. Assuming
that the maximum stress occurs at the outermost layer in the EL fiber, the flexural strength can be
that the maximum stress occurs at the outermost layer in the EL fiber, the flexural strength can be
calculated by Equation (3), where F is the maximum force applied before failure [40]. The flexural
calculated by Equation (3), where F is the maximum force applied before failure [40]. The flexural
strength of the fibers inside and outside the knitted matrix was calculated as 5.23 MPa and 2.02 MPa,
strength of the fibers inside and outside the knitted matrix was calculated as 5.23 MPa and 2.02 MPa,
respectively. Based on these values, the knitted matrix allows the fiber to withstand more than twice
respectively. Based on these values, the knitted matrix allows the fiber to withstand more than twice
the load that it would be able to on its own.
the load that it would be able to on its own.
needed to break the fibers, which is quantified by the maximum stress before fiber failure. Assuming
that the maximum stress occurs at the outermost layer in the EL fiber, the flexural strength can be
calculated by Equation (3), where F is the maximum force applied before failure [40]. The flexural
strength of the fibers inside and outside the knitted matrix was calculated as 5.23 MPa and 2.02 MPa,
respectively.
Fibers 2018, 6, 50 Based on these values, the knitted matrix allows the fiber to withstand more than10twice
of 13
the load that it would be able to on its own.

Figure 9.
Figure 9. Bipolar
Bipolar trapezoid
trapezoid waveform
waveform indicating
indicating important
important measurement
measurement points.
points.

The flexural modulus of the fibers is the ratio of stress to strain in flexural deformation, which
The flexural modulus of the fibers is the ratio of stress to strain in flexural deformation,
describes the material’s tendency to bend. The higher the flexural modulus, the more resistant it is to
which describes the material’s tendency to bend. The higher the flexural modulus, the more
resistant it is to bending. Ideally, this flexural modulus is equivalent to the elastic modulus of the
material, but in reality, these values may differ due to the presence of normal and shear stresses
throughout the beam span. The contribution of the shear stress can be reduced by using a high
L/h ratio. During experimentation, a sufficiently high L/h ratio was maintained to minimize this
effect. The flexural modulus of an EL fiber outside the knit textile was determined using Equation (6),
where m is the initial slope of the load-deflection curve given in Figure 9 [41].

FL3
Ef = (6)
48πR4 D
The flexural modulus of a single fiber was determined to be 4.9 GPa, which means that these
fibers are stiff and have a high resistance to bending.

4. Discussion
The frequency has a more pronounced effect on the brightness of the ACPEL system than the
voltage because of the time dependence of light emission in ACPEL devices. Upon application of an
electric field with a high enough magnitude, electrons are excited by impact ionization and emit light
as they relax back to ground state. This relaxation process occurs on the order of nanoseconds, and the
emission produced has an equivalent decay time. To have a continuous emission, the applied field
must be constantly switched as the emission and subsequent decay of the devices occurs twice per AC
cycle. Generally, a higher frequency will result in a higher brightness due to the time dependence of
the emission [28]. However, once the AC cycle becomes comparable to or surpasses the lifetime of
excited electrons, the brightness will decrease.
A number of textile properties such as gauge, yarn thickness, and yarn material could affect the
visible light output of the fibers. Specifically, a tighter gauge, thicker yarn, or denser yarn material
could decrease the amount of visible light. The apparent brightness of the fibers will also change
based on the amount of ambient light. In applications where the surroundings are dark, like a wall
mounted screen in a movie theater or night time concert, the fibers could be run at lower voltages and
frequencies, thus increasing the lifetime of the device.
Fibers 2018, 6, 50 11 of 13

The efficiency of the fiber is an internal property of the fiber that will not change with fabric
integration. However, the wear and tear that the fiber experiences during normal use in a functional
fabric will decrease this efficiency over time. The efficiency is an important property of the fiber as it
determines the power requirements of the fiber. The low device efficiency exhibited by the ACPEL
fibers is a consequence of the indirect method of light emission in these devices, which gives rise to
high power requirements. However, this AC power can be supplied directly in airplanes and through
wall electrical outlets, enabling automotive interior lighting and wall-mounted lighting applications.
Moisture, and therefore humidity, can accelerate the natural decrease in luminous efficiency that
occurs over time in inorganic phosphor devices [42]. The transparent conductive top layer reduces
the exposure of the phosphor layer to such environmental effects. However, fabrics incorporating
these fibers should not be subjected to excessive moisture like soaking or washing without additional
precautions to protect the fibers. Additionally, the high power requirements significantly reduce the
portability of the fibers and fabrics supporting these fibers and require that the fibers be isolated from
human skin. In the future, an encapsulation layer could be added to protect the fibers and electrically
isolate them to prevent human contact. However, the addition of more layers could further reduce the
flexibility of the fibers, so a highly flexible polymer encapsulation layer is recommended.
The ACPEL fibers exhibit a high resistance to bending, which has been discussed in previous
work [29,34]. Thus, when introduced into the knitted matrix, they reduced the overall flexibility, but
significantly increased the strength of the fabric. This increased strength results from the distribution
of force across the fiber and matrix. The fiber absorbs some of the force and as it has a high flexural
strength, it reinforces the fabric strength. This increased strength is useful in applications where the
fibers is subjected to a great deal of perpendicular stresses such as automotive carpet lighting or a
display screen directly integrated into the fabric of a car seat.

5. Conclusions
In this paper, ACPEL fiber devices were fabricated and experimentally analyzed. To demonstrate
the performance and potential of these fibers in functional fabrics, electrical, optical, and mechanical
tests were performed on the fibers inside and outside of a knit fabric. The simple fabrication procedures
and ease of handling streamlines the integration of ACPEL fibers into knitted fabrics. The integration
of ACPEL fibers into a knit structure improves the strength and robustness of the resulting light
emitting fabric, making it useful for applications in automotive/airplane interior lighting and emissive
wall displays.

Author Contributions: A.M. fabricated the EL devices, performed the characterization experiments, analyzed the
data and wrote the paper. A.F. contributed all other materials/analysis tools and revised the manuscript.
Funding: This material is based upon work supported by both the National Science Foundation Graduate
Research Fellowship under Grant No. DGE-1104459, and the National Science Foundation STEM K-12 Fellowship
under grant No. DGE-0947936.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Dalsgaard, C.; Sterrett, R. White paper on smart textile garments and devices: A market overview of smart
textile wearable technologies. In Market Opportunities for Smart Textiles; Ohmatex: Viby, Denmark, 2014.
2. Janietz, S.; Gruber, B.; Schattauer, S.; Schulze, K. Integration of OLEDs in textiles. Adv. Sci. Technol. 2013,
80, 14–21. [CrossRef]
3. Wang, G.F.; Tao, X.M.; Huang, H.M. Light-emitting devices for wearable flexible displays. Color. Technol.
2005, 121, 132–138. [CrossRef]
4. Orth, M.; Post, R.; Cooper, E. Fabric computing interfaces. In CHI 98 Conference Summary on Human Factors in
Computing Systems—CHI ’98; ACM Press: New York, NY, USA, 1998; pp. 331–332.
5. CuteCircuit. Available online: https://cutecircuit.com/ (accessed on 8 July 2016).
Fibers 2018, 6, 50 12 of 13

6. Mauriello, M.; Gubbels, M.; Froehlich, J. Social fabric fitness: The design and evaluation of wearable
E-textile displays to support group running. In Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems, Toronto, ON, Canada, 26 April–1 May 2014.
7. Jost, K.; Stenger, D.; Perez, C.R.; McDonough, J.K.; Lian, K.; Gogotsi, Y.; Dion, G. Knitted and screen printed
carbon-fiber supercapacitors for applications in wearable electronics. Energy Environ. Sci. 2013, 6, 2698–2705.
[CrossRef]
8. Koncar, V. Optical fiber fabric displays. Opt. Photonics News 2005, 16, 40–44. [CrossRef]
9. Spigulis, J. Side-emitting fibers brighten our world. Opt. Photonics News 2005, 16, 34–39. [CrossRef]
10. Deflin, E.; Koncar, V.; Weill, A.; Vinchon, H. Bright optical fiber fabric: A new flexible display. Text. Asia 2001,
33, 25–28.
11. Jeong, S.M.; Song, S.; Seo, H.-J.; Choi, W.M.; Hwang, S.-H.; Lee, S.G.; Lim, S.K. Battery-Free,
Human-Motion-Powered Light-Emitting Fabric: Mechanoluminescent Textile. Adv. Sustain. Syst. 2017, 1,
1700126. [CrossRef]
12. Zhang, Z.; Guo, K.; Li, Y.; Li, X.; Guan, G.; Li, H.; Luo, Y.; Zhao, F.; Zhang, Q.; Wei, B.; et al. A colour-tunable,
weavable fibre-shaped polymer light-emitting electrochemical cell. Nature 2015, 9, 233–238. [CrossRef]
13. Yang, H.; Lightner, C.R.; Dong, L. Light-emitting coaxial nanofibers. ACS Nano 2012, 6, 622–628. [CrossRef]
[PubMed]
14. Samei, E.; Badano, A.; Chakraborty, D.; Compton, K.; Cornelius, C.; Corrigan, K.; Flynn, M.J.; Hemminger, B.;
Hangiandreou, N.; Johnson, J.; et al. Assessment of display performance for medical imaging systems.
Med. Phys. 2005, 32, 1205–1225. [CrossRef] [PubMed]
15. O’Connor, B.; An, K.K.H.H.; Zhao, Y.; Pipe, K.P.P.K.; Shtein, M. Fiber Shaped Light Emitting Device.
Adv. Mater. 2007, 19, 3897–3900. [CrossRef]
16. Kwon, S.; Kim, W.; Kim, H.C.; Choi, S.; Park, B.-C.; Kang, S.-H.; Choi, K.C. P-148: Polymer Light-Emitting
Diodes Using the Dip Coating Method on Flexible Fiber Substrates for Wearable Displays. In Proceedings of
the 53rd Annual SID Symposium, San Jose, CA, USA, 21 May–5 June 2015; Volume 46, pp. 1753–1755.
17. Dias, T.; Monaragala, R. Development and analysis of novel electroluminescent yarns and fabrics for localized
automotive interior illumination. Text. Res. J. 2012, 82, 1164–1176. [CrossRef]
18. Coyle, J.P.; Li, B.; Dion, G.; Fontecchio, A.K. Direct integration of a 4-pixel emissive display into a knit fabric
matrix. Adv. Disp. Technol. III 2013, 8643, 864308. [CrossRef]
19. Zhang, Z.; Cui, L.; Shi, X.; Tian, X.; Wang, D.; Gu, C.; Chen, E.; Cheng, X.; Xu, Y.; Hu, Y.; et al. Textile Display
for Electronic and Brain-Interfaced Communications. Adv. Mater. 2018, 30, 1800323. [CrossRef] [PubMed]
20. Services, S.A.T. Glow Yarns for Special Effects in Textiles from Swicofil. Available online: http://www.
swicofil.com/glow_yarn.html (accessed on 23 June 2017).
21. Bortz, T.; Agrawal, S.; Shelnut, J. Photoluminescent Fibers, Compositions and Fabrics Made Therefrom.
U.S. Patent 8,207,511, 26 June 2012.
22. Gauvreau, B.; Guo, N.; Schicker, K.; Stoeffler, K.; Boismenu, F.; Ajji, A.; Wingfield, R.; Dubois, C.;
Skorobogatiy, M. Color-changing and color-tunable photonic bandgap fiber textiles. Opt. Express 2008,
16, 15677–15693. [CrossRef] [PubMed]
23. Sayed, I.; Berzowska, J.; Skorobogatiy, M. Jacquard-woven photonic bandgap fiber displays. Res. J. Text. Appar.
2010, 14, 97–105. [CrossRef]
24. O’Connor, B.; Pipe, K.; Shtein, M. Fiber based organic photovoltaic devices. Appl. Phys. Lett. 2008, 92, 172.
[CrossRef]
25. Bellingham, A. Direct Integration of Dynamic Emissive Displays into Knitted Fabric Structures; Drexel University:
Philadelphia, PA, USA, 2017.
26. Kim, W.; Kwon, S.; Lee, S.-M.; Kim, J.Y.; Han, Y.; Kim, E.; Choi, K.C.; Park, S.; Park, B.-C. Soft fabric-based
flexible organic light-emitting diodes. Org. Electron. 2013, 14, 3007–3013. [CrossRef]
27. Kwon, S.; Kim, W.; Kim, H.; Choi, S.; Park, B.-C.; Kang, S.-H.; Choi, K.C. High Luminance Fiber-Based
Polymer Light-Emitting Devices by a Dip-Coating Method. Adv. Electron. Mater. 2015, 1, 1500103. [CrossRef]
28. Bredol, M.; Dieckhoff, H.S. Materials for Powder-Based AC-Electroluminescence. Materials 2010, 3, 1353–1374.
[CrossRef]
29. Bellingham, A.; Bromhead, N.; Fontecchio, A. Rapid Prototyping of Slot Die Devices for Roll to Roll
Production of EL Fibers. Materials 2017, 10, 594. [CrossRef] [PubMed]
Fibers 2018, 6, 50 13 of 13

30. Dupont. Processing Guide for DuPont Luxprint Electroluminescent Inks. 2012. Available online: http:
//www.dupont.com/content/dam/dupont/products-and-services/electronic-and-electricalmaterials/
documents/prodlib/MCM-EL-Processing-Guide.pdf (accessed on 16 July 2018).
31. Keir, P. Fabrication and Characterization of ACTFEL Devices, Oregon State. 1999. Available online: https://
ir.library.oregonstate.edu/concern/graduate_thesis_or_dissertations/n009w490c (accessed on 16 July 2018).
32. Aleksandrova, M.P.; Dobrikov, G.H.; Andreev, S.K.; Dobrikov, G.M.; Rassovska, M.M. Electrical Properties
Characterization of Thick Film Organic Electroluminescent Structures. Annu. J. Electron. 2011, 5, 183–186.
33. Wager, J.F.; Keir, P.D. Electrical Characterization of thin-film electroluminescent devices. Annu. Rev. Mater. Sci.
1997, 27, 223–248. [CrossRef]
34. Bellingham, A.; Fontecchio, A. Direct integration of individually controlled emissive pixels into knit fabric
for fabric-based dynamic display. IEEE Photonics J. 2017, 9, 1–10. [CrossRef]
35. Standard Guide for Testing Fabric-Reinforced Textile Composite Materials. ASTM International:
West Conshohocken, PA, USA. Available online: http://www.astm.org/cgi-bin/resolver.cgi?D6856D6856M
(accessed on 16 July 2018).
36. Zweben, C.; Smith, W.; Wardle, M. Test methods for fiber tensile strength, composite flexural modulus,
and properties of fabric-reinforced laminates. In Composite Materials: Testing and Design (Fifth Conference);
ASTM International: West Conshohocken, PA, USA, 1979.
37. Oberg, E.; Jones, F.; Horton, H.; Ryffel, H. Machinery’s Handbook; Industrial Press: New York, NY, USA, 2004.
38. Zalm, P.; Diemer, G.; Klasens, H. Some aspects of the voltage and frequency dependence of electroluminescent
zinc sulphide. Philips Res. Rep. 1955, 10, 205–215.
39. Alfrey, G.F.; Taylor, J.B. Electroluminescence in Single Crystals of Zinc Sulphide. Proc. Phys. Soc. Sect. B 1955,
68, 775–784. [CrossRef]
40. Hodgkinson, J.M. Mechanical Testing of Advanced Fibre Composites; Elsevier: New York, NY, USA, 2000;
ISBN 9781855738911.
41. Mallick, P.K. Fiber-Reinforced Composites: Materials, Manufacturing, and Design; CRC/Taylor & Francis:
Boca Raton, FL, USA, 2007; ISBN 1420005987.
42. Smet, P.F.; Moreels, I.; Hens, Z.; Poelman, D. Luminescence in Sulfides: A Rich History and a Bright Future.
Materials 2010, 3, 2834–2883. [CrossRef]

© 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/).