Development of a Predictive Shielding Effectiveness … Abstract “Development of a Predictive Shielding Effectiveness Model for Carbon Fiber/Nylon Based Composites” The need for
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Development of a Predictive Shielding Effectiveness Model for Carbon Fiber/Nylon Based Composites
By
Nicholas B. Janda
Bachelor of Science, Case Western Reserve University, 2003
This thesis, “Development of a Predictive Shielding Effectiveness Model for Carbon Fiber/Nylon Based Composites,” is hereby approved in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE in the field of Chemical Engineering.
DEPARTMENT – Chemical Engineering
Signatures:
Thesis Advisor: ____________________________
Dr. Julia A. King
Thesis Co-advisor: ___________________________
Dr. Jason M. Keith
Department Chair: ______________________________
Dr. Michael Mullins
Date: ________________________________________
i
Abstract
“Development of a Predictive Shielding Effectiveness Model for Carbon Fiber/Nylon Based Composites”
The need for electromagnetic interference (EMI) shielding materials has increased recently due to the more prevalent use of personal communications devices (cell phones, pda’s). Metals have typically been used the material of choice for shielding applications. Design weight limitations for highly portable devices, however, has limited the applicability of metals for these applications. A need for light weight materials capable of providing EMI shielding exists. Through the addition of conductive fillers to normally electrically insulating polymer resins, electrically conductive composites can be used for shielding applications, providing light weight shielding materials. Shielding theory for composite materials, however, is largely undeveloped, unlike for metals. These models developed for metals cannot be used to accurately predict the shielding effectiveness provided by a composite containing a wide range of conductive fillers . The shielding effectiveness (SE) of two different carbon fiber/nylon based composites was studied over the radio frequency range (300 to 1000 MHz). The effects of incident electromagnetic wave (EM) frequency, filler volume percent, filler size (radius), and filler orientation on the measured SE were examined. The objective of this analysis is to characterize the factors involved in determining the SE of a composite from first principles. From this analysis, a model predicting shielding effectiveness for carbon fiber/nylon based composites is developed. The model is expected to perform well at low filler loadings, but also can be used to accurately predict shielding effectiveness at filler loadings above the percolation threshold, as seen from comparisons of the model to experiments with ThermalGraph™ and Fortafil carbon fibers in nylon 6,6.
ii
Acknowledgements
I must first acknowledge the Michigan Technological University Graduate School for
providing me the opportunity to further pursue my degree. I have traveled a long road to get
to this point and feel fortunate to complete my degree at Michigan Tech.
I also need to thank Dr. Warren Perger for providing the guidance, knowledge and focus
needed to complete the project. Thank you for teaching me how to “think/see like a wave”
and providing such truisms as “in the land of the blind the one eyed man is king.”
I must thank my co-advisors, Dr. Julia King and Dr. Jason Keith. Your input and aid was
always valued. I truly appreciated your consistent enthusiasm and support when the project
left the realm of typical Chemical Engineering. Thank you for your patience in dealing with
an atypical situation, project and student.
I would be remiss if I did not acknowledge the MATLAB assistance of Troy Oxby and
Dr. Jason Keith. Thank you for helping me rediscover my programming skills and showing
me that the program has more to offer than just Simulink.
The generosity of the National Science Foundation must be acknowledged. The funding
provided through Award Number DMI-9973278 allowed for prior fabrication of the samples
investigated in this study.
I must thank Brian Ott and Chris Copeland for providing a nearly endless amount of
distractions. Carrie Majkrzak, you deserve a medal of honor for sharing an office with me for
the past year.
Finally, I need to thank my Mom and Dad. Thank you for putting up with my educational
pursuits and never losing faith when things did not go smoothly. Your help and support along
the way has never gone unappreciated. A special thank you to my Mom: thanks for never
letting your level of frustration reach a point to where you felt it was necessary to strangle
your son.
iii
Table of Contents Abstract.......................................................................................................................... i Acknowledgements ...................................................................................................... ii Table of Contents ........................................................................................................ iii List of Figures................................................................................................................v List of Tables .............................................................................................................. vii CHAPTER 1: Introduction..........................................................................................1
1.1 Electromagnetic Radiation and Interference...................................................1 1.2 Polymer Based Composite Materials..............................................................3 1.3 Predicting Shielding Effectiveness in Composite Materials...........................3 1.4 Project Outline ................................................................................................5
CHAPTER 2: Project Materials and Sample Formulation ......................................6
4.3 Balance of Power Results .............................................................................26 4.4 Orientation Results........................................................................................29
CHAPTER 5: Electromagnetic Theory ....................................................................32
5.3 Scattered Field Theory..................................................................................33 5.4 Scattered Field Equation Derivation.............................................................35
5.5 Scattering Width ...........................................................................................49 CHAPTER 6: Shielding Effectiveness Model Design..............................................52
6.1 Introduction...................................................................................................52 6.2 Review of Problem Description and Focus ..................................................52 6.3 Analysis of Scattering Equations ..................................................................53
6.3.1 Dependence on Frequency, Optical Radius and Distance From Scatterer to Observer............................................................................................................53 6.3.2 Deterministic Nature of Scattering Equations ......................................55
6.4 Accounting for Collision Probability............................................................57 6.5 Scaling Factor Analysis ................................................................................59 6.6 Shielding Effectiveness Model Results ........................................................62 6.7 Scaling Factor - Linear Fit ............................................................................65 6.8 White Model Comparison.............................................................................68
CHAPTER 7: Conclusions and Future Work..........................................................71
7.1 Thesis Goal ...................................................................................................71 7.1.1 Conclusions from Electrical Resistivity/Conductivity Experiments ....71 7.1.2 Conclusions from Shielding Effectiveness Experiments......................71 7.1.3 Conclusions from Power Balance Analysis ..........................................72 7.1.4 Conclusions from Fiber Orientation Studies.........................................72 7.1.5 Conclusions from Model Development and Analysis ..........................72
7.2 Future Work ..................................................................................................74 CHAPTER 8: References ...........................................................................................76 Appendix A: Formulation Summary .......................................................................78 Appendix B: Shielding Effectiveness Experiment Results .....................................81 Appendix C: Balance of Power Results (mW) ........................................................88 Appendix D: Reflected, Absorbed and Transmitted Signal Results in dB...........95 Appendix E: Scaling Factor Analysis.....................................................................102 Appendix F: Shielding Effectiveness Model Results ............................................109 Appendix G: Shielding Effectiveness Model Results............................................116 Appendix H – White Model Derivation ..................................................................123 H.1 Introduction ....................................................................................................123 H.2 Absorption Term Derivation ..........................................................................124 H.2 Reflection Loss Term Derivation ...................................................................125 Appendix I: Proposed Model Comparison to White Model ................................128
v
List of Figures Figure 1.1-1: IEEE Standard for Safety Limits on Human Exposure to RF Fields (3) .2 Figure 2.3-1: Leistritz Extruder Used for Compounding of Composites ......................8 Figure 2.3-2: Extruder Screw Design, Note Flow is From Right to Left ......................9 Figure 2.3-3: Niigata Injection Molder........................................................................10 Figure 2.3-4: Four Cavity Mold...................................................................................11 Figure 2.3-5: Shielding Effectiveness Disk .................................................................11 Figure 3.2-1: Bar From Which Longitudinal Electrical Resistivity Samples Were Cut
...............................................................................................................14 Figure 3.2-2: (A) Experimental Set-up for Four Probe Test Method, .........................15 Figure 3.3-1: Shielding Test Fixtures With Support................................................16 Figure 3.3-2: Transmission Holder Without Sample...................................................16 Figure 3.3-3: Cross Sectional View of Transmission Holder (24) ..............................17 Figure 3.3-4: Reference and Load Shielding Effectiveness Disks (24).......................17 Figure 3.3-5: Reference Disk Alignment on Trasmission Fixture (25) .......................18 Figure 3.4-1: Shielding Test Apparatus Schematic (25)..............................................19 Figure 3.6-1: Dipole Antenna and Sample Holder ......................................................21 Figure 4.2-1: Shielding Effectiveness for Pure Nylon 6,6...........................................23 Figure 4.2-2: Shielding Effectiveness As a Function of Filler Volume Percent At
Select Frequencies ................................................................................24 Figure 4.2-3: Shielding Effectiveness Results for ThermalGraph™ DKD X .............25 Figure 4.2-4: Shielding Effectiveness Results for Fortafil 243 ...................................26 Figure 4.3-1: Balance of Power Results (mW) for NCN20 (ThermalGraph™)..........28 Figure 4.3-2: Balance of Power Results (mW) for NDN20 (Fortafil 243) ..................28 Figure 4.4-1: NDN40 Fiber to Incident Wave Orientation Dependence for
Transmitted Signal Strength ...............................................................29 Figure 4.4-2: Depictions of Perpendicular and Parallel Fiber to Wave Orientations ..30 Figure 4.4-3: Carbon Fiber/Epoxy Sheet Fiber to Incident Wave Orientation
Dependence for Transmitted Signal Strength ......................................31 Figure 5.2-1: Representation of Shielding Phenomena for Plane Waves Passing
Through a Homogeneous Barrier (10)..................................................33 Figure 5.3-1: A cylinder Impinged by a Uniform Plane Wave....................................34 Figure 5.4-1: Electromagnetic Frequency Spectrum (25)............................................37 Figure 5.4-2: Cross Sectional View of Transmission Holder (24) ..............................39 Figure 5.4-3: Cylindrical Coordinate System ..............................................................41 Figure 5.4-4: Block diagram Depicting the Two Step Process for Solving for the
Radiated Fields Given a Current and Charge Source (32)..............43 Figure 5.4-5: Diagram of the Position Vectors. The vector potential A at ρ is
obtained by integrating the current J at 'ρ . (3) .................................46 Figure 5.4-6: Uniform Plane Wave of TMz Orientation Impinging a Single
Cylindrical Scatterer With Radius a (32)............................................48 Figure 5.5-1: Cross Sectional View of Transmission Holder (24) ..............................50 Figure 6.3-1: Near Zone (ρ = 1.0 x 10-4 m) Scattering Width for Both Fibers............54 Figure 6.3-2: Far Zone (ρ = 50 m) Scattering Width for Both Fibers .........................55
vi
Figure 6.3-3: Theoretical Shielding Effectiveness of a Single Carbon Fiber Scattering an Incident Wave ..................................................................................56
Figure 6.4-1: Sample Wavelength Sized Window For Shielding Disk .......................58 Figure 6.5-1: Scaling Factor Analysis for NCN05 ......................................................59 Figure 6.5-2: Scaling Factor Analysis for NDN05 ......................................................60 Figure 6.6-1: Model Predicted and Experimentally Determined Shielding
Effectiveness for NCN05......................................................................63 Figure 6.6-2: Model Predicted and Experimentally Determined Shielding
Effectiveness for NDN05......................................................................63 Figure 6.6-3: Model Fit Quality Analysis for ThermalGraph™ Based Composites ...64 Figure 6.6-4: Model Fit Quality Analysis for Fortafil Based Composites ..................65 Figure 6.7-1: Linear Fit Applied to ThermalGraph™ Scaling Factor Data.................66 Figure 6.7-2: Linear Fit Applied to Fortafil Scaling Factor Data ................................66 Figure 6.7-3: Model Predicted and Experimentally Determined Shielding
Effectiveness for ...................................................................................67 Figure 6.7-4: Model Predicted and Experimentally Determined Shielding
Effectiveness for ...................................................................................68 Figure 6.8-1: White Model and Proposed Model Comparison and Experimentally
Determined Shielding Effectiveness for NCN10..................................69 Figure 6.8-2: White Model and Proposed Model Comparison for NDN40 ................70
vii
List of Tables Table 2.2-1: Properties of DuPont Zytel 101 NC010 (15).............................................6 Table 2.2-2: Properties of BP/Amoco ThermalGraph DKD X (16)..............................7 Table 2.2-3: Properties of Akzo Nobel Fortafil 243 PAN based 3.2mm Chopped and
Fiber Mean Length 3.2 mm (entire range is 2.3 mm to 4.1 mm) Carbon Assay 95%
Binder Content 2.6 wt% proprietary polymer that adheres pellet together and promotes adhesion with nylon matrix
8
2.3 Sample Preparation
For this project, the fillers were used as received. The Zytel 101 NC010 was dried in an
indirect heated dehumidifying drying oven (dewpoint of the recirculating air = -40oC). After
drying, the polymer was stored in moisture barrier bags.
2.3.1 Extrusion
An American Leistritz Extruder Corporation Model ZSE 27 was used for all polymer
extrusion throughout the course of the project. The extruder, shown in Figure 2-1, has a 27
mm co-rotating intermeshing twin screw with 10 zones and a length/diameter ratio of 40. The
screw design used produced minimal filler degradation while still providing adequate dispersal
of the filler within the polymer. This screw design is shown in Figure 2-2
Figure 2-1: Leistritz Extruder Used for Compounding of Composites
9
The polymer pellets (Zytel) were introduced in Zone 1. The second side stuffer, located
at Zone 7, was used to introduce the carbon fibers into the polymer melt. Two Schenck
AccuRate gravimetric feeders were used to accurately control the amount of each material
added to the extruder. A complete list of all formulations extruded is provided in Appendix A
and the extrusion conditions for each are discussed in detail by Weber (18), Clingerman (19)
and Heiser (20). Typical extrusion conditions are listed in Table 2.3-1.
Atmospheric Vent
Atmospheric Back Vent Side Stuffer Side Stuffer Main Feed
40D 36D 32D 28D 24D 20D 16D 12D 8D 4D
GFA
2-3
0-30
GFA
2-3
0-90
GFA
2-4
0-90
GFA
2-3
0-60
GFA
2-4
0-90
KB
5-2
-30-
60
KB
5-2
-30-
30
KB
5-2
-30-
90
KB
5-2
-30-
60
KB
5-2
-30-
30
KS
1-2-
10 E
GFA
2-3
0-60
GFA
2-4
0-90
GFA
2-4
0-90
KB
5-2
-30-
90
KB
5-2
-30-
60
KB
5-2
-30-
30
KB
5-2
-30-
60
GFA
2-2
0-30
GFA
2-3
0-90
GFA
2-4
0-90
KS
1-2-
10 A
0D
For Screw Type Elements
GFA-d-ee-ff G = co-rotating F = conveying A = Free-Meshing d = number of threads
ee = pitch (length in millimeters for one complete rotation)
ff = length of screw elements in millimeters Kneading disks
KBj-d-kk-ll KB = kneading block J = number of kneading segments d = number of threads
k = length of kneading block in millimeters l = twisting angle of the individual kneading segments
Kneading disks
KS1-d-hh-i KS1 = Kneading disc d = number of threads h = length of kneading disc in millimeters i = A for initial disc and E for end disc Zones
0D to 4D is Zone 1 (water cooled, not heated) 4D to 8D is Zone 2/Heating Zone 1 8D to 12D is Zone 3/Heating Zone 2 12D to 16D is Zone 4/Heating Zone 3 16D to 20D is Zone 5/Heating Zone 4 20D to 24D is Zone 6/Heating Zone 5 24D to 28D is Zone 7/Heating Zone 6 28D to 32D is Zone 8/Heating Zone 7 32D to 36D is Zone 9/Heating Zone 8 36D to 40D is Zone 10/Heating Zone 9 Nozzle is Heating Zone 10
Figure 2-2: Extruder Screw Design, Note Flow is From Right to Left
Table 2.3-1: Extrusion Conditions for Nylon 6,6 (19)
Zone 1 Temperature (by feed hopper) 210oC Zone 2 Temperature 250oC
Zone 3 to Zone 5 Temperature 270oC Zone 6 to Zone 7 Temperature 275oC
Zone 8 to Zone 10 Temperature 280oC Total Throughput 19.0 kg/hr
Screw rpm 300 rpm
10
2.3.2 Injection Molding
The test specimens were molded using a Niigata injection molding machine, model
NE85UA4, (Figure 2-3). Implementing a 40 mm diameter single screw with a length/diameter
ratio of 18, the lengths of the feed, compression and metering sections of the single screw
were 396 mm, 180 mm and 144 mm, respectively. Two different molds were used for this
project. The four cavity mold shown in Figure 2-4 was used to produce 3.2 mm thick ASTM
Type I tensile bars (end gated) and 6.4 cm diameter disks of 3.2 mm thickness. The tensile
bars were used for longitudinal electrical conductivity measurements while the disks were
used for transverse electrical conductivity tests. Figure 2-5 shows the mold from which the
shielding disks of 130 mm diameter and 3.2 mm thickness were created. The molding
conditions for each formulation using the four-cavity mold are discussed in detail in
Clingerman, Weber and Heiser (18-20). The typical operating conditions for the injection
molding machine can be found in Table 2.3-2.
Figure 2-3: Niigata Injection Molder
11
Table 2.3-2: Injection Molding Conditions for Conductive Nylon (20)
Zone 1 Temperature (by feed hopper) 285oC Zone 2 Temperature 290oC Zone 3 Temperature 299o C
Zone 4 Temperature (die nozzle heater) 310 oC Mold Temperature 88oC
Screw rpm 54 rpm Injection Pressure 154 MPa
Hold Pressure 109 MPa Back Pressure 3 MPa Injection Time 15 seconds Cooling Time 15 seconds Interval Time 2 seconds
Figure 2-4: Four Cavity Mold
Figure 2-5: Shielding Effectiveness Disk
Mold
12
2.4 Formulations
Test specimens were labeled according to the material, weight percent filler, and the
order that the specimen came out of the injection molder using the following nomenclature:
N – W – X – Y - ##
N = National Science Foundation Project W = Filler used X = Polymer Y = Weight percent of conductive fiber ## = Sample Number, indicating the order that the sample came out of the injection molder
All formulations were designated with an N as the first letter to denote that they were
from a previous NSF project (Award Number DMI-9973278). Following was a multi-letter
combination to denote the filler (W). “C” denoted the ThermalGraph carbon fiber, while “D”
referred to Fortafil 243. X was used to designate the polymer matrix used, with “N” referring
to nylon 6,6. The Y in the above formula was the weight percent of the conductive filler.
Following the above naming convention, a sample labeled NCN15-3, refers to the third
composite sample from the mold containing 15 wt% ThermalGraph DKD X carbon fiber in a
nylon 6,6 matrix.
Table 2.4-1 shows the concentrations of the resins produced for use in this project.
Table 2.4-1: Loading Levels for Composite Samples Studied
CHAPTER 3: Experimental and Characterization Methods
3.1 Introduction
In this section, the techniques used to determine the properties of the composite samples,
which were used to test the shielding effectiveness model developed in this thesis, are
discussed. These properties include: transverse (through-plane) and longitudinal (in-plane)
electrical resistivity (inverse of the electrical conductivity), shielding effectiveness, filler
volume fraction, filler orientation, filler length and aspect ratio.
3.2 Electrical Resistivity
3.2.1 Transverse Electrical Resistivity Test Method
For samples with an electrical resistivity greater than 104 ohm-cm, a through-plane (also
called transverse), volumetric electrical conductivity test was conducted on the as molded test
specimen. In this method, a constant voltage (typically 10 V or 100 V) was applied to the test
specimen and the resistivity was measured according to ASTM D257 using a Keithley 6517A
Electrometer/High Resistance Meter and an 8009 Resistivity Test Fixture (21). The Keithley
6524 High Resistance Measurement Software was used to automate the conductivity
measurement. For each formulation, a minimum of six specimens were tested. Each test
specimen was an injection molded disk that was 6.4 cm in diameter and 3.2 mm thick. Since
the presence of water can affect a sample’s conductivity, all samples were tested dry as
molded (DAM).
3.2.2 Longitudinal Electrical Resistivity Test Method
The volumetric longitudinal electrical resistivity (in-plane) was measured on all samples
with an electrical resistivity less than 104 ohm-cm. Test specimens cut from the center gauge
portion of a tensile bar, Figure 3-1, were surface ground on all sides and cut into sticks 2 mm
14
wide, 2 mm thick and 25.4 mm long. As with the transverse electrical resistivity test method,
the presence of water can have a marked effect on the measured conductivity of the sample,
so all were tested dry as molded.
Typically for each formulation, a total of six specimens were cut from a single tensile
bar, with four tensile bars generally used to obtain a total of twenty four test specimens.
After machining, test specimens were dried in a vacuum oven at 660 mmHg and 60°C for
two hours and then sealed in moisture barrier bags. These samples were then tested using a
four probe technique, as shown in Figure 3-2(a). This technique measures resistivity by
applying a constant current (typically 5 to 10 mA) with a Keithley 224 Programmable
Current Source and measuring the voltage drop over the center 6 mm of the sample with a
Keithley 182 Digital Sensitive Voltmeter. The electrical resistivity is then calculated from
Equation 3.2.1 (22):
Li
twVER
∆
= [3.2-1]
Where: ER = electric resistivity (ohm-cm) V∆ = voltage drop over center 0.6 cm of sample (volts) w = sample width (cm) t = sample thickness (cm) i = current (amps) L = length over which V∆ is measured (0.6 cm)
Figure 3-1: Bar From Which Longitudinal Electrical Resistivity Samples Were Cut
Flow & Measurement
15
2mm
2mm
25mm
∆V fromcenter 6mm
Constant current in through sample
Constant current out of
sample
SampleVolt
Meter
Current Source
(A) (B)
2mm
2mm
25mm
∆V fromcenter 6mm
Constant current in through sample
Constant current out of
sample
SampleVolt
Meter
Current Source
2mm
2mm
25mm
∆V fromcenter 6mm
Constant current in through sample
Constant current out of
sample
2mm
2mm
25mm
∆V fromcenter 6mm
Constant current in through sample
Constant current out of
sample
SampleVolt
Meter
Current Source
SampleVolt
Meter
Current Source
(A) (B) Figure 3-2: (A) Experimental Set-up for Four Probe Test Method,
(B) Sample Dimensions and Longitudinal Current Flow (19)
3.3 Shielding Effectiveness
The electromagnetic shielding effectiveness of each formulation was measured
according to ASTM D 4935-89 (Reapproved 1994), for planar materials using a plane-wave,
far-field EM wave. Although it provides a method of measuring far-field SE, the nature of
the shielding test apparatus used in this study allowed for measurement of near-field shielding
effectiveness values (23). To be able to measure near-field power values, one must be able to
fully characterize the impinging wave directly before it collides with the shielding media.
The method is valid over a frequency range of 30 MHz to 1.5 GHz.
An Electro-Metrics, Inc. shielding effectiveness test fixture (model EM-2107A) was
used to hold the sample with a HP 8752C network analyzer generating and receiving the EM
signals. Figure 3-3 and Figure 3-4 show the shielding test apparatus and sample holder.
Figure 3-5 shows a cross-sectional view of the test fixture.
16
Figure 3-3: Shielding Test Fixtures With Support
Figure 3-4: Transmission Holder Without Sample
For each formulation, one reference sample and at least 5 load samples were tested over a
frequency range of 30 MHz to 1.0 GHz. A reference sample consists of a large ring and a smaller
inner disk as shown in Figure 3-6. The shielding effectiveness (SE) of a material is the ratio of
the power received with and without a material present for the same incident power. For these
experiments, therefore, it is the difference ratio of the load sample to the reference sample. It is
expressed in units of decibels (dB), as shown in Equation 3.3-1 (4).
2
110PP
log SE 10dB = [3.3-1]
Where: 1P = received power with the material present (watts)
2P = received power without the material present (watts)
The input power used was 0 dBm, corresponding to 1 mW. The dynamic range (difference
between the maximum and minimum signals measurable by the system) of the system was 80 dB.
17
Figure 3-5: Cross Sectional View of Transmission Holder (24)
Figure 3-6: Reference and Load Shielding Effectiveness Disks (24)
18
Figure 3-7: Reference Disk Alignment on Trasmission Fixture (25)
Figure 3-7 shows the placement of the reference sample on the transmission fixture. The
small disk and larger outer ring must be precisely aligned on the fixture to obtain accurate
readings. The nylon 6,6-based samples were tested DAM. The results from the analysis are found
in Appendix B and discussed in Chapter 4.
3.4 Balance of Power Analysis
The shielding effect test apparatus was also used to determine the contribution of reflection
(scattered) and absorption to the overall shielding effectiveness of a sample. The HP 8752C
Network analyzer is capable of measuring the transmitted power from test fixture and reflected
power from the top of the sample holder. Accounting for cable loss for both the input and output
cables from the fixture, as seen in Figure 3-8, the amount of signal reflected and transmitted
through the sample can be directly measured. The absorbed signal power can then be calculated
Figure 3-8: Shielding Test Apparatus Schematic (25)
The transmitted and reflected power was measured for at least 6 load samples for each
formulation. Results from the balance of power analysis are discussed in Chapter 4 and listed in
Appendices C and D.
3.5 Fiber Volume Fraction, Fiber Length and Aspect Ratio
A solvent digestion method was used to determine the weight percent of the filler in the
composite sample. As described in ASTM Standard D5226, this method completely dissolves the
polymer, leaving only clean filler particles (26). A 0.2 g sample cut from the center of a
transverse ER disk was used. Formic acid was used to dissolve the nylon 6,6 based composites at
23 oC. The filler was separated from the solvent/polymer mixture through vacuum filtration. The
mass of the dried filler particles was then compared to the weight of the original mass of the
composite/filler sample to determine the weight percent of the filler within the sample. 2 to 4
samples were tested per formulation. These filler volume fraction results are shown in detail
elsewhere (20,25). In all cases, the actual filler content of each formulation matched the target
amount within acceptable tolerances.
HP Analyzer
Shielding Apparatus
Regulator
Input Cable
Output Cable
Air Cylinder
20
After the fibers were extracted frm the nylon 6,6 matrix, they were dispersed onto a glass
slide and viewed using an Olympus SZH10 optical microscope with an Optronics Engineering
LX-750 video camera. The images (at 60x magnification) were collected using Scion Image
version 1.62 software and then processed using Adobe Photoshop 5.0 and the Image Processing
Tool Kit v. 3.0. The length of each fiber was measured and the aspect ratio (AR), Diameter
LengthAR = ,
was calculated. For each formulation, between 200 and 3000 individual fibers were measured
(18-19,27). These results are shown in Appendix A.
3.6 Orientation
3.6.1 Fiber Orientation
The orientation of the carbon fibers within the composite was determined by viewing a
polished sample with an optical microscope. For each formulation, a 12.7 mm x 12.7 mm section
was cut from a SE test disk. The sample was mounted in epoxy and positioned such that the
depth of the sample could be viewed (3.2 mm). The samples, in the epoxy plug, were polished
and then viewed via an Olympus BX60 reflected light microscope at a magnification of 200x.
Scion Image version 1.62 software was used to collect the images, which were later processed in
Adobe Photoshop 5.0 using Image Processing Kit v. 3.0. The average orientation of 1000 to 4000
fibers per formulation was determined (28). Appendix A shows the results of this analysis.
3.6.2 Transmission Orientation Dependence
The effect of fiber orientation on the transmitted signal strength was investigated using the
fixture shown below in Figure 3-9. A large circular metal plate was affixed to a sheet of
plexiglass that was held in place by slots cut into a PVC pipe. The plate reduced the possibility of
wave diffraction around the composite sample interfering with the measured transmitted signal
strength. The shielding disk samples were placed in the remaining slot on the sample holder, in
front of the antenna and large metal plate. A dipole antenna was positioned directly behind the
metal plate with a transmission cable connecting it to the HP 8752C network analyzer. An
21
electric field of known orientation (parallel to the plane of the floor) and strength was sent
through the sample to be received by the antenna. The transmitted signal strength was measured
over a frequency range of 500 to 2000 MHz. The shielding disk sample was then rotated 90
degrees and the process repeated. All measurements were conducted in an anechoic chamber to
reduce the error inducing effects of outside interference and incident field reflection.
Figure 3-9: Dipole Antenna and Sample Holder
22
CHAPTER 4: Experimental Results
4.1 Introduction
The results from the balance of power analysis and shielding effect experiments, described
in Chapter 3, are discussed in this chapter. Also presented are the results from the fiber
orientation studies.
4.2 Shielding Effectiveness Results
From the data measured using the techniques discussed in Chapter 3, the shielding
effectiveness for each individual sample for each formulation was calculated using Equation
4.2.1.
2
110PP
log SE 10dB = [4.2-1]
Where: 1P = received power with the material present (watts)
2P = received power without the material present (watts) The SE results compiled in this investigation compared favorably to the work of Krueger (25) and
Heiser (20).
4.2.1 Pure Nylon 6,6
As expected, the pure matrix of only nylon 6,6 showed essentially no ability to shield
electromagnetic fields due to its dielectric nature. Ideally, an impinging electromagnetic wave
should encounter no resistance when passing through a dielectric material. Assuming the
material exhibits no conversion of the incident energy into heat while the wave travels through
the dielectric (condition known as a non-lossy dielectric), the shielding effectiveness should be
zero. Figure 4.2-1 shows the pure nylon 6,6 matrix following this behavior. Little shielding
effectiveness was measured, approximately 0.1 dB at the higher frequencies. This corresponds to
shielding only 2% of the incident field strength.
23
Figure 4-1: Shielding Effectiveness for Pure Nylon 6,6
The solid line in Figure 4.2-1 represents the mean shielding effectiveness for the
formulation. For each formulation, the mean was calculated from at least 4. Typically, 6 samples
were measured. The upper dashed line corresponds to the highest SE value recorded for any of
these samples. Similarly, the lower dashed line refers to the lowest SE value recorded. It is
possible for a single trial to produce a maximum at one frequency and a minimum at another.
This, however, was frequently not the case. A single specific specimen of a formulation typically
would produce SE values that were either high, average or low.
4.2.2 ThermalGraph™ DKD X
The introduction of ThermalGraph™ carbon fiber into the nylon 6,6 matrix resulted in
enhanced EM shielding characteristics. Increasing the amount of filler within the sample resulted
in decreased electrical resistivity (ER) and increased shielding effectiveness. Also observed was
the effect of increased frequency on the measured SE values. This trend is expected and has been
300 400 500 600 700 800 900 1000 -0.25
-0.2
-0.15
-0.1
-0.05
0
0.05
0.1
0.15
0.2
Frequency (MHz)
24
reported elsewhere (9-14). As frequency is increased, the wavelength of the EM wave decreases
and becomes for comparable to the size of the fiber. Thus, higher frequency waves are more
likely to encounter fiber embedded in the polymer matrix. Similarly, as the weight percent of
fiber is increased, there is an improved probability that the wave will collide with a fiber. The
fibers, as opposed to the polymer rich areas, are more likely to scatter or absorb the wave, as the
nylon is virtually invisible to the wave. Hence, SE increases as frequency increases.
For all formulations studied, listed in Table 2.4-1, SE increased at higher frequencies. The
SE results for the ThermalGraph™ DKD X composites at 300, 500 and 800 MHz are shown in
Figure 4-2. Figure 4-3 shows both how shielding effectiveness directly increased as a function of
filler weight percent and frequency.
Figure 4-2: Shielding Effectiveness As a Function of Filler Volume Percent At Select Frequencies
0 5 10 15 20 25 30 0
2
4
6
8
10
12
14
Volume Percent Fiber (%)
300 MHz 500 MHz 800 MHz
25
Figure 4-3: Shielding Effectiveness Results for ThermalGraph™ DKD X
4.2.3 Fortafil 243
The addition of Fortafil 243 fibers into the matrix produced similar SE trends. Like the
ThermalGraph™ DKD X, the SE for the Fortafil samples increased with both frequency and filler
weight percent. The Fortafil samples, however, showed markedly better shielding behavior. For
example, NCN40 was found to have the best shielding effect performance among the
ThermalGraph™ samples, approximately 14 dB at 1.0 GHz. In comparison, NDN40 was found
to have a maximum shielding effectiveness of 72 dB at 1.0 GHz. This disparity between the
behaviors of the two fillers tracked with the ER results (Appendix A). As shown in Tables 2.2-2
and 2.2-3, both the Fortafil and ThermalGraph™ fibers have similar electrical resistivities. When
both fibers, however, were introduced into the nylon 6,6 matrix in equal weight percents, the
Fortafil sample was found to be two orders of magnitude more conductive than the
ThermalGraph™. Thus, improved shielding for the Fortafil samples was observed.
300 400 500 600 700 800 900 1000 0
2
4
6
8
10
12
14
Frequency (MHz)
5 wt% 10 wt% 15 wt% 20 wt% 30 wt% 40 wt%
26
A corresponding trend was noticed in the percolation thresholds for Fortafil and
ThermalGraph™ electrical resistivity. A prior investigation determined the thresholds to be 9.5
and 3.4 volume percent, respectively (19). Previous work has suggested that the increased
shielding effectiveness afforded by the Fortafil 243 filler may be due in part to the increased
heteroatoms present on the surface of the individual fibers. Fortafil 243 results in improved
adhesion with the nylon matrix material which might explain increased composite SE (20,28-29).
The Fortafil based formulations are listed in Table 2.4-1. The SE results for the Fortafil based
composites are shown in Figure 4-4. Again, as frequency and filler weight percent were
increased, shielding effectiveness increased.
300 400 500 600 700 800 900 10000
10
20
30
40
50
60
70
80
Frequency (MHz)
Shi
eldi
ng E
ffect
iven
ess
(dB
)
5 wt%7 wt%10 wt%15 wt%20 wt%30 wt%40 wt%
Figure 4-4: Shielding Effectiveness Results for Fortafil 243
4.3 Balance of Power Results
From the frequency dependent transmitted and reflected power data accumulated from the
balance of power experiments, the relative effects of electric field reflection (scattering) and
27
absorption on the SE performance of a the composite samples was determined. Although the
experimental apparatus does not allow for direct measurement of the absorption power loss, a
simple power balance accounting for all methods of signal degradation allows for indirect
calculation of the absorption term, Equation 4.3-1.
7. Bigg, D. M., Adv. Polym. Tech., 4, 3/4, p.255 (1977).
8. Schulz, R. B., Plantz, V. C., and Brush, D. R., IEEE Trans. Elect. Compat., 30, 5, p.187 (1988).
9. Bushko, W. C., Stokes, V. K., Wilson, J., ANTEC ’99, p.1499 (1999).
10. White, D.R. J., EMI/EMC Handbook Series, Germantown, MD: Don White Consultants, Inc., 4 (1971).
11. Chu, H. C., Chen, C. H., IEEE Trans. Elect. Compat., 38, 1, p.1 (1996).
12. Lin, M. S., Lin, C. M., Wu, R. B., Chen, C. H., IEEE Trans. Elect. Compat., 35, 3, p.357 (1993).
13. Lin, M. S., Chen, C. H., IEEE Trans. Elect. Compat., 35, 1, p.21 (1993).
14. Krohn, T. L., Medgyesi-Mitschang, L. N., IEEE Trans. Antenna Propagat., 37, 2, p.219 (1989).
15. DuPont Zytel Nylon Resin Product and Properties, Version 95.9, Printed in USA. (2001).
16. Amoco Performance Products: High Thermal Conductivity Pitch Based Graphite Fibers, Amoco Polymers; Alpharetta, GA 30005. (2001).
17. Akzo Nobel Electrically Conductive Fortafil 243 Product Literature, Akzo Nobel Chemicals Inc., 300 S. Riverside Plaza, Chicago, IL, 60606.
18. Weber, E., PhD Dissertation, “Development and Modeling of Thermally Conductive Polymer/Carbon Composites”, (2001).
19. Clingerman, M. L., PhD Dissertation, “Development and Modeling of Electrically Conductive Composite Materials”, (2001).
77
20. Heiser, J. A., MS Thesis, “Conductive, Shielding, Tensile and Impact Properties of Carbon Filled Nylon 6,6 Based Resins”, (2003).
21. “Standard Test Methods for D-C Resistance or Conductance of Insulating Materials,” ASTM Standard D257-91, American Society for Testing and Materials, Philadelphia. (1998).
23. Annual Book of ASTM Standards, D4935-89. Standard Test Method for Measuring the Electromagnetic Shielding Effectiveness of Planar Materials. D4935-89. Copyright ASTM, (1989).
24. Shielding Effectiveness Test Fixture Electro-Metrics Model EM-2107-A Manual, Jan, (1999).
25. Krueger, Q. J., M.S. Thesis, "Electromagnetic Interference and Radio Frequency Interference Shielding of Carbon-Filled Conductive Resins", (2002).
26. “Standard Practice for Dissolving Polymer Materials”, ASTM Standard D5226-98, American Society for Testing and Materials, Philadelphia, Pennsylvania, (1998).
27. Konell, J. P., Ph.D. Dissertation "Characterization and Tensile Modulus Modeling of Conductive Resins", (2002).
28. Krueger, Q. J., and King, J. A., Adv.Polym. Tech, 22, 2, p.96 (2003).
29. J. A. Heiser, J. A. King, J. P. Konell, and L. L. Sutter, Polym. Comp., 25, 4, p.407 (2004).