Examples of Electrical Property Characterization and Application Experiences

1 | P a g e
Carbon Nanotube Alignment and Application as Field Effect Transistors
Jacob Feste
University of Arkansas, Biomedical Engineering, jtfeste@email.uark.edu

2 | P a g e
Abstract
The objective of this experiment was to determine how the frequency-dependent DEP
and electro-thermal forces influence the alignment process for CNTs. The impacts were
determined by measuring the I-V curves of aligned CNTs diluted 5 times at different frequencies
(0.5 MHz, 1 MHz, 1.5 MHz), and also by correlating them to a CNT stock solution at 1.5 MHz. A
20 V peak to peak voltage was applied four times to each sample with the corresponding
current values measured. The results supported an increase in current values with dilution. Of
the diluted samples, the 1 MHz measurements gave the smallest current values with the 0.5
MHz measurements giving slightly higher values due to increased DEP forces. The 1.5 MHz
measurements gave significantly higher current values as a result of reaching the crossover
frequency.
Introduction
Carbon nanotubes are cylindrical nanoparticles composed of a desired number of
carbon layers. These nanoparticles have a wide range of application due to their unique
properties. Carbon nanotubes exhibit exceptional strength, contain unique electrical properties,
and are efficient conductors of heat [2]. The alignment of such a nanoparticle, however, is
considered a complex process due to the small nature of the nanometer sized particles
involved. Nanoparticles are unable to be seen by the human eye, are associated with an
inherent degree of randomness and clumping, are very sensitive to the environment, and are
generally considered difficult to control in order to form a desired, uniform structure.
Therefore, the formation of nanostructures is typically done by utilizing the chemical and
electrical properties of the nanoparticles. Carbon nanotubes are long, cylindrical nanoparticles
that may be connected together via alignment. Such alignment is made possible using
dielectrophoresis. Ultimately, this process allows the nanoparticles to be carried, or aligned, as
a result of their dielectric properties [3]. Dielectrophoretic alignment of carbon nanotubes
begins with the placement of a dissolved carbon nanotube source between two triangular
electrodes on an electrode chip. A voltage difference is then applied between the two
triangular electrodes to generate a circular, non-uniform electric field beginning from one
triangular electrode and converging towards the outer point of the other triangular electrode. A
dielectrophoresis (DEP) force is then generated and is given by [1]:
Equation (1): 𝐹𝐷𝐸𝑃 = Γ𝜀 𝑚 𝑅𝑒{ 𝐹𝐶𝑀}∇|𝐸𝑟𝑚𝑠 |2
When this force is positive, the nanoparticles will migrate towards the high electric field
gradient, resulting in alignment [1]. Alignment accuracy is increased with an increasing DEP
force and decreasing triangular electrode angles for a more directional electric field. While the
nanoparticles in the dissolved solution begin to align, the remaining solution is intended to be
removed gradually in order to reduce mobility at the electrodes and allow for proper structure
formation. Dielectrophoresis also allows this process. The remaining solution, typically water

3 | P a g e
and surfactant for molecule separation, also has a force acting on it during the process. This
force is termed the electro-thermal force and is given by [1]:
Equation (2): 𝐹𝑒 = −𝑀(
𝜀 𝑚 𝜎 𝑚 𝑉𝑟𝑚𝑠
4
2𝑘 𝑇 𝜋3 𝑟𝑐
3
𝑇
)(1 −
2𝜃
𝜋
)
This force also depends on whether the frequency of the generated field is greater than or less
than the solution’s crossover frequency. Above this frequency, the electro-thermal force will be
positive. When the electro-thermal force is positive, the remaining fluid solution will be forced
away from the electrodes [1]. The processes of both alignment and solution removal are highly
dependent on the frequency of the generated field. More specifically, these processes are
dependent on the crossover frequencies of the nanoparticles. At or below the crossover
frequency, the electro-thermal forces given off by the electrodes are too weak to repel the
attracted fluids [6]. On the other hand, the DEP force will decrease as the frequency
approaches the crossover frequency [6]. It is therefore desired to maintain frequencies just
above the crossover frequency to allow for solution removal and to reduce the impact of the
decreasing DEP forces. The DEP alignment process is complete when the nanoparticles are
properly aligned and the remaining solution is removed.
The electrical properties of aligned carbon nanotubes give rise to a wide variety of
unique applications. The resulting alignment may primarily be a semi conductive material if
single-walled nanotubes are produced, or a conductive material if multi-walled nanotubes are
produced [1]. The semi conductive nature of single-walled carbon nanotubes (SWCNTs), for
instance, provides the potential application of SWCNTs as field effect transistors. A field effect
transistor (FET) is a transistor that controls the electrical conductivity of a channel of one type
of charge carrier by using an electrical field to control the shape of the channel for a semi
conductive material [4]. As a FET, aligned SWCNTs may be used for the detection of DNA. These
FETs include an electron source at one end (i.e. gold/chrome) and an electron drain at the other
end (i.e. gold/chrome), of which are connected by a semi-conductive channel (aligned
SWCNTs). This channel is attached to a silicon oxide layer that separates it from a P-type
substrate/backgate (silicon). A voltage difference is applied between the source and drain to
allow an electron current to flow through the channel. A voltage is also applied to the backgate
in order to further control conductivity. When a voltage is applied, the conductivity of the
channel changes due to conformational changes [5]. DNA detection may be performed with
such a device by attaching various single-stranded DNA (ssDNA) molecules to the aligned
SWCNT channel. When a complimentary strands of DNA reaches the device, DNA hybridization
will occur to form completed, double-stranded DNA structures. Soon after this process occurs,
the negative charges associated with some of the phosphate groups on the DNA strands are
transferred towards the positively charged electrodes (gold/chrome). The negative charges are
then donated to the carbon nanotubes at either the source or drain regions. The binding
process increases resistivity and decreases conductivity at those regions by decreasing the
cross-sectional area at which current is able to flow due to conformational changes. Binding of

4 | P a g e
one region of DNA to the nanotubes brings the remaining DNA structure near the SWCNT
channel. Close proximity of the entire DNA strand is maintained following this process and is
oriented in a “wrapping” manner due to intermolecular forces, with binding occurring at the
source and drain electrode regions. Ultimately, the process will reduce the current and increase
resistivity. Detection of these measured changes is then able to provide that DNA hybridization
has occurred with a complimentary DNA strand.
Procedure
Materials
 1 mg Carbon Nanotubes (NanoLab PD15L5-20)
 4 µl AQ Nanosperse surfactant
 10 ml distilled H2O
 15 ml centrifuge tubes
 Wax Paper
 Micropipettes
 Electrode Microchip
 Conductive Wire
Equipment
 Sonicator
 NIS software
 Microscope
 Solder Machine
 Voltage Source Meter
 Laptop
 BK Precision® 4005DDS Function Generator
 Agilent 54622A Oscilloscope
 Weight Scale
Procedure
5x Dilution Solution Preparation:
1. Measure 1 mg of carbon nanotubes using wax paper.
2. Transfer the nanotubes to a 15 ml centrifuge tube.
3. Using a micropipette, add 4 µl of surfactant to the centrifuge tube.

5 | P a g e
4. Mix the solution by pumping the solution with the micropipette and slowly retracting
three times.
5. Add 10 ml of distilled H2O to the centrifuge tube using a micropipette.
6. Sonicate the solution for 10 minutes.
Note: Steps 3-6 may be repeated as needed until the solution is completely dissolved.
Electrode Microchip Preparation:
1. Pull the microchip off the blue tape very carefully.
2. Place and tape the microchip on the microscope film.
3. Transfer the film to the microscope.
4. Under the microscope, solder conductive wire from one electrode on the microchip to
the other.
I-V Curve Measurements:
1. Prepare the stock solution of carbon nanotubes and the 5x diluted solution.
2. Clean the electrode chip using an appropriate acidic solution.
3. Deposit approximately 2 µl of solution on the microchip, repeated for each solution.
4. Using the function generator and oscilloscope, apply 1.5 MHz with 20 peak-peak voltage
to the electrodes. Run four tests for each sample. For the 5x solution, repeat with 500
kHz and 1 MHz.
5. Observe carbon nanotube alignment and turn off the voltage before the sample is dry.
6. Measure I-V curves using a source meter and a laptop.
Results
Figure (1): Illustration of the CNT alignment process from beginning to end.

6 | P a g e
Figure (2): I-V curve results for 5x diluted CNT solution at 500 kHz frequency.
Figure (3): I-V curve results for 5x diluted CNT solution at 1 MHz frequency.

7 | P a g e
Figure (4): I-V curve results for 5x diluted CNT solution at 1.5 MHz frequency.
Figure (5): I-V curve results for CNT stock solution at 1.5 MHz frequency.
Y= 0.0015x-6E-6

8 | P a g e
Figure (6): Average I-V results for each sample and at each frequency for comparison.
Discussion and Conclusion
Our results include I-V curves for both stock and 5x diluted CNTs at different
frequencies. By analyzing this data, assumptions can be made about the DEP and electro-
thermal forces, crossover frequencies, dilution impact, and alignment widths. While tests were
not performed involving CNTs as FET devices, some properties associated with CNT-FET devices
may be correlated with our results. The conformational changes of the SWCNT channel and
their effects on an I-V curve, for instance, may be analyzed by determining the impact that
alignment width has on the I-V curves of the results. For the diluted CNT solution, the impact
that signal frequency has on its I-V curve is illustrated by figures 2-4 and figure 6. Each of these
graphs are associated with a generally linear change in current with respect to voltage. The 500
kHz I-V curve resulted in peak currents around 0.00065 A while the 1 MHz I-V curve resulted in
slightly less peak currents around 0.00048 A. The 1.5 MHz I-V curve resulted in significantly
higher peak currents around 0.01 A, however the measured peak to peak voltages were around
13 V as opposed to the 20 V peak to peak voltages associated with the other frequencies. Based
on the trend line in figure 4, it can be suggested that peak currents would likely reach around
0.015 A for a peak to peak voltage of 20 V. However, the reduced peak to peak voltage may
have been due to reduced DEP forces. If the DEP forces were not strong enough, some of the
CNTs may have remained unaligned; leaving them disconnected from the channel and attached
to the outer regions of the electrodes. It is possible that the applied voltage was shared among
these nanoparticles and therefore reduced. Due to the rapid increase in current, it can be
suggested that the crossover frequency for the diluted sample is likely between 1-1.5 MHz. In
other words, the electro-thermal forces at 500 kHz and 1 MHz are likely negative and push the

9 | P a g e
remaining solution towards the electrodes, greatly increasing resistivity and decreasing current.
The large difference in current values between 1-1.5 MHz suggests that this force becomes
positive at some instance between these frequencies, allowing the remaining solution to be
expelled away from the electrodes. The decrease in current from 500 kHz to 1 MHz is most
likely due to decreasing DEP forces as the crossover frequency is approached. Therefore, the
500 kHz measurements likely gave the best alignments. The electrodes remained attracted to
the fluid, however, significantly increasing resistance and decreasing current. The 1 MHz
measurements also had their electrodes resisted by the fluid, with decreasing currents resulting
in decreased DEP forces as the crossover frequency is approached. The difference can also be
explained by the changes in CNT width. As frequency increases, CNT width also increases
relative to its height [1]. A width-to-height ratio of 1 is desired to provide minimal current
constriction. As the ratio increases further away from this value, resistance increases due to
conformational changes. This phenomena is the primary detection mechanism for CNT-FET
devices and is supported by the results. The 500 kHz measurements likely had width-to-height
ratios near the desired value while the 1 MHz measurements may have had ratios beyond the
desired value. For the 1.5 MHz measurements, the decrease in resistivity associated with fluid
removal was significantly greater than the resistivity increase associated with width-to-height
ratios and decreased DEP forces; suggesting that these increases are insignificant by
comparison. The I-V curve results for the CNT stock solution at 1.5 MHz are displayed in figure
5, with the comparison data for each sample type’s average I-V values illustrated in figure 6.
The I-V curve of the CNT stock solution resulted in peak currents around 0.00025 A with
voltages ranging from 0 V to 10 V. These values primarily represent the true conductivity of the
CNTs without alignment. By diluting the CNTs and aligning them, the conduction channel
becomes much larger, resulting in increased currents at each frequency. For instance, the
diluted and aligned solution at the same frequency resulted in currents around 40 times that of
the CNT stock. In conclusion, the results suggest much more efficient conduction channels for
dissolved and aligned CNTs. They also suggests that this large increase in efficiency is only made
possible at the correct frequencies, making them highly dependent on their crossover
frequencies. As long as the frequency conditions are ideal, aligned CNTs may serve as effective
conductive channels with various application possibilities.
References
[1] Tung, Steve. "Aligned Carbon Nanotube Based Field Effect Transistor."Department of
Mechanical Engineering (2015): n. pag. University of Arkansas. Web. 10 Nov. 2015.
[2] "Carbon Nanotubes." ScienceDaily. N.p., n.d. Web. 10 Nov. 2015.
[3] Pethig, Ronald. "Review Article—Dielectrophoresis: Status of the Theory, Technology, and
Applications." Biomicrofluidics. American Institute of Physics, 29 June 2010. Web. 10 Nov. 2015.

10 | P a g e
[4] Lesurf, Jim. "Field Effect Transistor." Scots Guide. University of St. Andrews, n.d. Web. 10 Nov.
2015.
[5] "Depletion-type IGFETs." : Insulated-gate Field-effectTransistors. All About Circuits, n.d. Web.
10 Nov. 2015.
[6] Kakaç, S. Microfluidics Based Microsystems: Fundamentals and Applications. Dordrecht:
Springer, 2010. Google Books. Google. Web. 10 Nov. 2015.

11 | P a g e
The Impact of Scaling Effects on Nano Composite MEMS Devices: Mechanical and Electrical
Properties
Jacob Feste
University of Arkansas, Biomedical Engineering, jtfeste@email.uark.edu

12 | P a g e
Abstract
The goal of this experiment was to determine the mechanical and electrical properties
of a PDMS nanoparticle composite with various percentages of carbon black. These properties
are also measured in order to relate them to characteristics associated with a composite
material, such as the Rule of Mixtures (ROM). Mechanical properties were measured via tensile
testing with carbon black percentages ranging from 0%-20% (in multiples of two), with brass,
polypropylene, and steel measured for reference. The results indicated a weak material and a
linear increase in tensile strength as carbon black percentage increased, satisfying the ROM for
this composite. Electrical properties were determined by measuring output voltages relating to
changes in resistance in order to determine changes in resistivity. A Wheatstone bridge circuit
staged by an instrumentation amplifier and measured through an oscilloscope was used with
carbon black percentages ranging from 14%-19%. The results included a large degree of error
for the 14%-16% samples, however the results from the remaining samples provided enough
accuracy to support an increase in resistivity as carbon black percentage decreases.
Nomenclature
R= Resistance (Ohms)
L= Length (m)
A= Cross-sectional Area (m2)
ρ= Resistivity (Ohm*m)
V= Voltage (V)
I= Current (A)
σ= Stress (Pa)
F= Force (N)
ԑ= Strain
E= Elastic Modulus (Pa)
Introduction
A composite material is defined as a single material composed of a mixture of two or
more materials. Composite materials have a wide range of application due to the dependence
of their properties on the individual properties of the various materials they are composed of.
This relationship allows for the engineering of a composite material with desired properties
based on the combined properties included in its composition. However, for composite
materials, properties such as mechanical properties may be negatively influenced by the
addition of one material to the surface of another. When two materials of different shapes and
sizes are combined, a small region of space is created where their dimensions are not perfectly
identical, resulting in a less stable and less uniform structure. The “porous” identity, for

13 | P a g e
example, interrupts conduction paths, decreases mechanical strength, and other properties
that favor a single, uniform structure. It is therefore desired to minimize this aspect in order for
a composite material to maximize its desired, combined properties and minimize the negative
impacts inherently involved with the combination of different materials. Minimizing the degree
of porosity in composites is performed by increasing the impact that an added material’s
surface (area) has compared to the overall volume. The impact that a material’s surface has on
its total volume increases as the size of a material decreases due to the effect of scaling laws.
These laws support an increased area-to-volume ratio with a decrease in length due to the
exponential relationship between length, area, and volume [1]. Therefore, the addition of
particles of smaller sizes are desired for the production of effective composite materials.
Nanoparticles fulfill this requirement, making nanoparticle composites effective at retaining the
combined material properties.
The nanoparticle composites involved in this experiment are aimed to highlight the
properties of a material with an elastomeric matrix phase (PDMS) and carbon black. The
composites are formed using DDPOST, a process that allows the formation of a thick layer of
polymer composite using micro- and nanoparticles with polymer matrix [1]. The resulting
polymer composite can be made electrically, mechanically, or chemically active by selecting
specific particle and matrix materials [1]. The nanoparticle composites of this experiment were
formed to allow chemical activation in order to serve as Micro-Electro-Mechanical Systems
(MEMS) based corrosion sensors. The carbon black nanoparticle inclusions are considered
electrically conductive nanoparticles and are applied to the PDMS matrix. When swelling and
etching agents are applied to this mixture, the PDMS matrix swells to a certain degree
dependent on the concentration of the swelling agents or chemical vapor exposure. Upon
swelling, the PDMS volume expands and extends the electrical pathways of the conductive
carbon black suspension [2]. The resistivity changes are then measured upon swelling
equilibrium in order to determine vapor concentration. Resistivity is given by the following
relationship:
Equation (1): 𝑅 = 𝜌(
𝐿
𝐴
) or 𝜌 = 𝑅(
𝐴
𝐿
)
PDMS has a high resistivity around 1*1013-1*1015 ohm*m [3]. When swelling occurs, the
resistivity value increases due to the extended electrical pathways [2]. These changes may be
evaluated to determine concentration due to the percolation theory. This theory claims that
clusters of particles attached to the surface of a material, such as the clusters of carbon black
nanoparticles attached to the PDMS matrix, accurately represent a uniform structure until the
randomness and separation of these clusters reaches a percolation limit at a certain percentage
of clustered particles [4]. At this limit, there is no longer a possible path connecting each
cluster, where this probability increases exponential as the percentage of clusters decreases.
The small sizes of the nanoparticles give an advantage in this aspect. Therefore, the
concentrations of the swelling and etching agents may be represented by the concentration of
removed nanoparticles until the percolation limit is reached and are given by the changes in
measured and known resistivity. Chemical vapor will provide effects similar to those of the
swelling and etching agents, allowing its concentration to be evaluated. Changes in resistance
must be measured in order to measure the changes in resistivity to make concentration

14 | P a g e
assumptions. This process is done with the use of a complex circuit and resistance changes
measured by an LCR meter. The relationship between voltage, current, and resistance is given
by:
Equation (2): 𝑅 =
𝑉
𝐼
or 𝑉 = 𝐼𝑅
The derivation of this relationship:
Equation (3): 𝑑𝑉 = 𝑅𝑑𝐼 + 𝐼𝑑𝑅
Is utilized in order to form a circuit with variable resistance only. By maintaining a constant
current, this equation becomes further reduced to:
Equation (4): 𝑑𝑉 = 𝐼𝑑𝑅
Therefore, a changing input voltage and constant current are necessary for the circuit to
measure changes in resistance. However, the voltage values must remain constant in order to
measure these changes. A “constant” voltage may be manipulated from a variable voltage
source by using voltage dividers and op amps. The circuit begins with the use of a voltage
divider in order to output a reference voltage close to 1 volt for the rest of the circuit. The
voltage divider equation is as:
Equation (5): 𝑉𝑜𝑢𝑡 = 𝑉𝑖𝑛
𝑅1
𝑅1+𝑅2
By using an R1 (10 kΩ) resistor with resistance significantly higher than that of R2 (10 Ω), the
output voltage will always remain about 1 volt (0.999V). This circuit is followed by an op amp
voltage follower in order to maintain the constant voltage, followed by a 10 kΩ resistor in order
to return the output voltage to its true value by countering the effects of the much higher R1
resistor of the voltage divider and maintain constant current. This component is then followed
by another op amp with one input side grounded and the other connected to the LCR device for
measurements. The grounded component serves to form as an adequate zero voltage
reference level while also having the ability to absorb as much current as possible without
disturbing the voltage potential, ultimately maintaining the desired constant values of the
circuit. Our device uses a 4-point probe that measures the voltage difference between the
middle two probes. The outer two probes connect to the non-grounded side of the op amp and
the op amp’s output, sending the constant current through the two ends. The middle two
probes are separated by a specimen acting as a resistor, where these voltage values are
measured following the addition of another voltage follower for the two probes to maintain
constant voltage for each path. The change in voltage across these two probes are measured to
determine the change in resistance, made possible by the constant current and voltage values
that are maintained throughout the circuit and differ only between these two points. The final
circuit is illustrated by figure (1) below.

15 | P a g e
In addition to the electrical properties of a nanoparticle composite, the mechanical
properties of these composites are also important. As stated previously, the combined
mechanical properties may be negatively influenced by the formation of such composites due
to inherent size differences between the components. Also, nanoparticles often reside in
clusters and are difficult to attach uniformly, giving rise to inherent randomness. The resulting
composites are often anisotropic in nature and therefore require estimates to determine
properties such as their overall mechanical strength. The Rule of Mixtures (ROM) may be
analyzed to provide these estimates. The ROM states that stress values are directly
proportional to the ratio of the volume of nanoparticles to that of the total composite [5].
Nanoparticles have an advantage in that they have more of a surface area effect in a composite
and therefore the remainder of the composite retains most of its volume and therefore
individual stress values. However, this property may eliminate the possible use of the ROM in
order to estimate the stress values for these composites types. The stress values may be
measured using a tensile test, where the samples are elongated until failure or until a certain
length, with the stress and strain values measured along the way. Stress and strain values are
given by the following relationships:
Equation (6): 𝜎 = 𝐸ԑ
Equation (7): 𝜎 =
𝐹
𝐴
Equation (8): ԑ =
∆𝑙
𝐿
By comparing this data to the relationship given by the ROM, it is made possible to determine
whether or not the ROMremains accurate for the composite. Proportionality should be seen
between the stress values and different volume percentages. If this pattern is not given, it can

16 | P a g e
be suggested that the ROM does not apply. This experiment will measure the mechanical
properties of nanoparticle composites composed of PDMS and different volumes of carbon
black. By comparing the data for the different percentages, it is possible to determine their
proportionality and whether or not the composite follows the ROM.
Procedures
Materials
1. Carbon nanoparticles (Alfa Aesar, 45527[42nm, 100%], 39724 [42nm, 50%], H30253)
2. Sylgard® 184 silicone elastomer kit, Dow Corning (Midland, MI)
3. PASCO Scientific Plastic (AP-8222) and Metal (AP-8223) tensile test specimens
4. Mixing cups and stirrer
5. Isopropanol alchohol and De-ionized water
6. Metal Spatula
7. Glass Beakers (10mL)
8. Microscope glass slide (1”x3”)
9. Gold/Silver nanoparticles from synthesis lab (D. Chen)
10. Microfabricated electrodes with SU-8 molds
11. Tensile test specimen mold
12. Instrumentation amplifier with Wheatstone bridge circuit
13. Chemical vapor mixing and injection system
Equipment
1. Dell Inspiron 1764 Laptop
2. Tenma 72-9365 200Mhz Oscilloscope
3. BK Precision Power supply-Model9310
4. Agilent Technologies U1733P LCR Meter
5. PASCO Scientific Stress-Strain Apparatus (Tensile Tester)
6. Extech EX540 Multimeter/Thermocouple Reader (with Type-K thermocouple probe)
7. Digital Hot-plate
8. Digital Scale
9. Microsoft LifeCamStudio Webcam
Software
1. LabView
2. Instrument drivers/software

17 | P a g e
Procedures
Electrical:
Preparation:
1. Weigh each particle of carbon black for desired volume ratios of 14%, 15%,
16%, 17%, 18%, and 19% when mixed with PDMS.
2. Convert the mass ratios from a systemic set of PDMS to the desired volume
ratios given in step 1.
3. Manually mix the particles and PDMS at 5 minutes each in a disposable
plastic mixing cup.
4. Squeegee the material into the pre-fabricated micro electrode SU8 mold by
using a standard microscope glass slide.
5. Release the mold after it has been cured for a couple of days.
LCR Measurements:
1. Connect either the multi-meter (Extech EX540), LCR meter (Agilent U1733P),
or Wheatstone bridge circuit staged by an instrumentation amplifier and
measured through an oscilloscope to the laptop and open their associated
link software.
2. Use the software to record all data digitally.
3. Begin with the multi-meter and measure the resistance and capacitance of
each sensor device.
4. Repeat step 3 with the LCR meter.
5. Connect the sensors to the bridge circuit, power up the circuit with the
power supply, and feed the output to the oscilloscope.
6. Repeat all the measurements with the oscilloscope while on a hotplate with
temperatures of (RT+10oC increments up to 100oC).
7. Monitor the temperature with the type-K thermocouple connected through
the multi-meter.
Note: Only the Wheatstone bridge method was performed for measurements.
Swelling and Resistance Measurements:
1. For each of the samples in preparation step 1, place a small disk of the
material into a 50mL beaker filled to the 40mL mark with one of the swelling
agents.
2. Use a 50:50 ratio of swelling and etching agents (12mL to 12mL) toluene and
acetic acid.
3. Report the swelling process using a webcam (Microsoft LifeCamCinema
720p) that has been calibrated for its pixel resolution and controlled by the
LabView™ Vision® software.
4. Record the resistance when the swelling reaches equilibrium.

18 | P a g e
Note: This procedure was demonstrated but not performed for this
experiment.
Mechanical:
Preparation:
1. Mix PDMS with various carbon black densities (42nm; 50%, 100%).
2. Pour each sample into the prefabricated aluminum molds for natural curing
over one week.
3. Weigh each particle of Carbon Black for desired volume ratios of 0%, 2%, 4%,
6%, 8%, 10%, 12%, 14%, 16%, 18%, and 20% when mixed with PDMS.
4. Convert the mass ratios from a systemic set of PDMS to the desired volume
ratios given in step 1.
5. Manually mix the particles and PDMS at 5 minutes each in a disposable
plastic mixing cup.
Tensile Test Measurements:
1. Connect the PASCO Scientific Stress-Strain Apparatus’s Passport Rotary and
Force Sensors to the individual USB Link, then connect it to the laptop.
2. Start the DataStudio software and select the stress-strain apparatus
experiment.
3. Load the PASCO tensile test specimens and crank the rotary handle at a
steady rate until the specimen breaks.
4. Save the data and repeat for each sample.
5. Remove the nanoparticle PDMS composite specimen from its mold by
unscrewing the mold covers and load it into the Stress-Strain Apparatus.
6. Pull the specimen until breakage.
Results
Figure (2): LCR Measurements using the Wheatstone bridge circuit measurement method for
PDMS composites with %17-19 Carbon Black.
LCR Measurements for PDMS/Carbon Black (%) Composites Vin=0.9999 V Vsource=12 V
PDMS Composites with % Carbon Black 17% 18% 19%
Vout (V) Vout (V) Vout (V)
Sample 1 11.6 6.9 5.1
Sample 2 8.2 6.5 4.1
Sample 3 7.3 7.4 5.9
Sample 4 7.3 5.8 4.8
Sample 5 10.5 4.9 5.7
Average 8.98 6.3 5.12

19 | P a g e
Figure (3): Tensile test results for PDMS composites with % Carbon Black.
Figure (4): Tensile test results for total PDMS composite averages and different materials to
serve as references.

20 | P a g e
Discussion
The results of this experiment include results for LCR and tensile test measurements.
The swelling and resistance procedure was demonstrated but not performed. The LCR
measurements were taken using a Wheatstone bridge circuit staged by an instrumentation
amplifier and measured through an oscilloscope. The results are given by figure 2. According to
equation 2 and equation 4, these results suggest an average increase in resistivity for a PDMS
and carbon black composite as the percentage of carbon black decreases. This relationship was
expected and likely due to the more resistant, or less conductive, nature of PDMS compared to
carbon black. These measurements were intended to be taken for samples of 14%-19% carbon
black. While some successful results remained, measurements taken for the composites with
carbon black percentages less than 17% included a significant degree of error and therefore
inconclusive results. This error was likely due to the large possibility of error associated with
such a complex measurement system. It could also be due to errors involved in the preparation
process in which inaccurate volume percentages were produced. The results of the tensile tests
were much more accurate and are illustrated in figure 3 and figure 4. According to figure 3,
there was a general increase in ultimate tensile strength, or maximum stress before rupture, as
carbon black percentage increased. The stress values upon rupture were between 40,000-
60,000 Pascals for the samples with no carbon black while they were between 110,000-140,000
Pascals for those with 20% carbon black. The 10% samples had stress values between 80,000-
100,000 Pascals before rupture, suggesting a linear relationship between carbon black
percentage and ultimate tensile stress. Most of the samples ruptured between stress values of
4-5. Many samples did not break, however, making it impossible to make correlations between
the carbon black percentages and rupture point strain values. Alone, PDMS has a relatively low
ultimate tensile strength around 15,000-90,000 Pascals. When a fraction of its volume is
replaced by carbon black, the combined properties should increase this value to satisfy the rule
of mixtures. The tensile test results exhibit a somewhat linear increase in stress values as
carbon black percentage increases, suggesting a valid ROMrelationship. Figure 4 illustrates the
average of the tensile test values combining each sample of each percentage, and compares
them to materials such as steel, brass, and polypropylene. Steel had the highest tensile strength
but was much more brittle, or broke with less strain, than the other materials. Brass also had a
high tensile strength but mimicked the ductility of the polymer materials, polypropylene and
the PDMS composite. Overall, the composite material was much weaker than the other
materials, likely due to its composition and identity as a nanoparticle composite. However, the
negative mechanical impacts associated with a composite material were minimized due to the
results suggesting ROMapplicability. Error was also a possibility for the tensile test
measurements as well due to testing being done manually. However, the measurements were
accurate enough to generate conclusive assumptions and disregard this error.

21 | P a g e
References
[1] Huang, Adam. "Experiencing Scaling Effects via Nano Composite MEMS Devices." University
of Arkansas, n.d. Web. 1 Nov. 2015.
[2] Huang, Adam, Victor Tak Sing Wong, and Chih-Ming Ho. "Silicone Polymer Chemical Vapor
Sensors Fabricated by Direct Polymer Patterning on Substrate Technique (DPPOST)." Sensors
and Actuators B: Chemical116.1-2 (2006): 2-10. Web.
[3] "Polydimethylsiloxane (PDMS)." CiDRA Precision Services. N.p., n.d. Web. 01 Nov. 2015.
[4] Gastner, Michael T. "Percolation Theory." Michael Gastner: Percolation Theory. Imperial
College London, n.d. Web. 01 Nov. 2015.
[5] Kopeliovich, Dmitri. "Estimations of Composite Materials Properties."Substech. N.p., 2 June
2012. Web. 1 Nov. 2015.

Examples of Electrical Property Characterization and Application Experiences

More Related Content

What's hot (19)

Similar to Examples of Electrical Property Characterization and Application Experiences (20)

Examples of Electrical Property Characterization and Application Experiences