«A Dissertation Presented to The Academic Faculty By Parisa Pour Shahid Saeed Abadi In Partial Fulfillment of the Requirements for the Degree Doctor ...»
MECHANICAL BEHAVIOR OF CARBON NANOTUBE FORESTS
UNDER COMPRESSIVE LOADING
The Academic Faculty
Parisa Pour Shahid Saeed Abadi
In Partial Fulfillment
of the Requirements for the Degree
Doctor of Philosophy in the
George W. Woodruff School of Mechanical Engineering
Georgia Institute of Technology
Copyright © 2013 Parisa Pour Shahid Saeed Abadi
MECHANICAL BEHAVIOR OF CARBON NANOTUBE FORESTS
UNDER COMPRESSIVE LOADING
Dr. Samuel Graham, Advisor Dr. Hamid Garmestani School of Mechanical Engineering School of Materials Science and Engineering Georgia Institute of Technology Georgia Institute of Technology Dr. Baratunde A. Cola, Advisor Dr. Ting Zhu School of Mechanical Engineering School of Mechanical Engineering Georgia Institute of Technology Georgia Institute of Technology Dr. Satish Kumar School of Materials Science and Engineering Georgia Institute of Technology Date Approved: April 3, 2013 To my husband (Hassan Masoud), and my parents (Pari Shekouh and Seyyed Ata PourShahid)
Then, I would like to express my deep appreciation to my advisors, Dr. Samuel Graham and Dr. Baratunde Cola, for their incessant guidance and assistance throughout my entire Ph.D research process. I really appreciate their constructive comments and criticism and the freedom they gave me to explore my ideas. I would also like to thank Dr. Satish Kumar, Dr. Hamid Garmestani, and Dr. Ting Zhu for serving on my committee. Their comments added a lot to my research.
Many other people contributed to this work. I would like to thank Dr. Julia Greer from Caltech who offered me her lab equipment for conducting in situ mechanical experiments. I’m grateful to Dr. Matthew Maschmann from AFRL for all his help throughout this work including conductance of mechanical testing experiments. I wish to express my gratitude to others who assisted me throughout this research specially Dr.
Shelby Hutchens form Caltech, John Taphouse, and Dr. Anuradha Bulusu from Georgia Tech, and Denzell Bolling, undergraduate student visitor from Howard University.
I would like to acknowledge my lab mates, colleagues and friends for their support and help, and useful conversations throughout my stay at Georgia Tech. Special thanks go to NEST and EMRL group members, to faculty and staff of the school of mechanical engineering in Georgia Tech who helped me in different ways, specially Dr.
Jason Nadler and Dr. Yogendra Joshi for offering their lab equipment to be used for this dissertation, and to the staff of Microelectronic Research Center in Georgia Tech
assistance in performing experiments for this dissertation.
I am ever thankful to my husband for his love and support. I could not have accomplished this work without his support and encouragement during the tough times. I can never thank my parents enough. They sacrificed a lot so I can be where I am now.
Thank you and I love you. I wish to thank my siblings and all family members and friends who loved me and supported me.
LIST OF TABLES
LIST OF FIGURES
CHAPTER 1 INTRODUCTION
1.1 Applications and motivation
CHAPTER 2 BACKGROUND AND RELEVANT LITERATURE
2.1 CNT forest growth
2.2 CNT forest structure
2.3 CNT forest adhesion to substrate
2.4 CNT forest coating
2.6 CNT forest mechanical behavior
CHAPTER 3 EFFECTS OF MORPHOLOGY ON THE MICROCOMPRESSION RESPONSE OF PRISTINE CARBON NANOTUBE FORESTS 28
3.2.1 CNT forest growth
3.2.2 Characterization of CNT Forests
3.2.3 Mechanical testing
3.3 Results and Discussion
CHAPTER 4 BUCKLING-DRIVEN DELAMINATION OF CARBONNANOTUBE FORESTS
4.3 Results and discussion
CHAPTER 5 DEFORMATION AND FAILURE MECHANISM OFCONFORMALLY-COATED CARBON NANOTUBE FORESTS
5.3 Results and discussion
CHAPTER 6 EFFECT OF SOLVENTS ON MECHANICAL BEHAVIOR OFCARBON NANOTUBE FORESTS
6.3 Results and discussion
CHAPTER 7 SUMMARY AND RECOMMENDATIONS
Various modulus values measured for different CNT forests with different indenters and measurement methodologies. (adapted from ref. )
Overview of the dissertation
Parameters for LPCVD, APCVD, and P-LPCVD recipes
Unloading stiffness, effective elastic modulus, elastic recovery ratio, largest buckle width ratio, and maximum value of negative stress during unloading for different CNT forests.
Table 5.1. Theoretical estimation of EA and EI for bare and coated CNTs.
Hamaker constant for solvents and graphite-graphite interactions in solvents.
The values for graphite-graphite interactions in solvents are calculated using eq.
Fig. 1.1. Schematic showing a bulk material (left) and three types of nanomaterials (right). The dimension limitations [1, 2] are demonstrated.
Fig. 1.2. SEM images of a side edge of a CNT forest. Left: larger scale in which individual CNTs are not discernible, and Right: smaller scale in which change of morphology from top to bottom of the forest is clear.
Fig. 2.1. CVD multiwall CNT growth mechanism [61, 62]
Fig. 2.2. CNT forest growth stages . a) schematic demonstration of the CNT forest growth stages, and b-f) SEM images of locations marked on the right figure in part (a).
Fig. 2.3. Shear stress versus displacement in a tape test on CNT forests grown on an Inconel wire. The inset shows a schematic of the test setup. 
Fig. 2.4. A schematic of electron beam physical vapor deposition 
Fig. 2.5. Au coating on top of a CNT forest deposited by e-beam PVD method ..... 19 Fig. 2.6. Atomic layer deposition of alumina 
Fig. 2.7. TEM image of dispersed CNTs coated with ALD alumina .
Fig. 2.8. TEM image of a MWCNT coated with 25 nm 
Fig. 2.9. A schematic representation of load-displacement curve for nanoindentation .
Pmax: maximum load, S: stiffness, hmax: maximum displacement, and hf: final depth of the indenter impression after unloading.
Fig. 2.11. Schematic comparison of compression of a CNT forest pillar versus CNT forest indentation. The indentation experiment shows the effect of interaction between CNTs under the indenter and those on the sides.
Fig. 3.1. Black Magic Pro 4” CVD system. Left: growth chamber when plasma in on, and right: the outside look of the system.
Fig. 3.2. Low magnification SEM image of a CNT forest grown on a Si substrate......... 31 Fig. 3.3. TEM images taken by JEOL 4000EX from CNTs grown with a) LPCVD, b) APCVD, and c) P-LPCVD method.
Fig. 3.4. Density data for the three types of CNT forests. a) average mass density vs.
normalized height, and b) density vs. normalized distance from top surface; the solution to eq. 3.1
Fig. 3.5. counted CNTs per unit length vs. normalized distance from top surface for the three types of CNT forests
Fig. 3.6. a) A sample high magnification (400 kX) SEM image used for CNT counting and orientation measurements showing highly entangled CNTs at the bottom of a LPCVD forest. b) Left image with straight lines drawn on CNT pieces for the calculation of average angle between CNTs and vertical direction.
Fig. 3.7. Average angle between CNTs and vertical line vs. normalized distance from top surface for the three types of CNT forests.
Fig. 3.8. A schematic of SEMentor, the in situ nanoindentation instrument (in California Institute of Technology) .
Fig. 3.10. Schematic diagram (not to scale) of on-edge and in-bulk locations................. 42 Fig. 3.11. CNT forest edge after indentation of LPCVD forest. a) shortest, b) medium, c) tallest forest, and d) magnified view of local buckles close to substrate in medium case which is also representative of shortest case.
Fig. 3.12. a) Axial indentation stress-strain curves for indentation on the edges of LPCVD forests of three heights. b) Axial indentation stress-strain curves for shortest LPCVD forest for edge and bulk points.
Fig. 3.13. a) APCVD CNT forest edge after indentation, b,c) high magnification view of the details of buckles inside rectangles in (a), d) Axial indentation stress-strain curves for edge and bulk points.
Fig. 3.14. Indentation of a P-LPCVD forest on edge. a) side view of the deformed spot, b) side magnified view of the section under indenter before (top image) and after (bottom image) appearance of CNT tips, c) top view of the deformed spot with magnified view of the out of plane deformation in the drawn rectangle, d) Axial indentation stress-strain curves for edge and bulk points.
Fig. 3.15. SEM images at 80 kX magnification showing orientation and entanglement of
CNTs along the height of LPCVD CNT forests with three different heights - left:
290 m, middle: 180 m, and right: 70 m. The scale bar is 1 m for each image. The distance from the substrate (Z) is labeled in each image. The locations of the first buckle are also marked.
Fig. 3.16. SEM images at 80 kX magnification showing orientation and entanglement of CNTs along the height of LPCVD, APCVD, and P-LPCVD CNT forests. The distance from the substrate (Z) is labeled in each image and the images from the locations of the first buckle are also marked.
xi Fig. 4.1. Buckling driven delamination in a mica thin film on an aluminum substrate  (left), and a schematic showing the side view and the stress that cause buckling (right).
Fig. 4.2. Tensile testing set-up. Left: schematic showing the grips and details of the sample bonded to the grips and the soft string to eliminate the bending moments, and right: real tensile testing set-up.
Fig. 4.3. SEM images of the bottom edges of CNT forests illustrating permanent deformation in the CNT forests after macro-compression. a) SEM image of an edge region with multiple buckles but no delamination, b) SEM image of a location illustrating multiple buckles and delamination of CNTs c) a magnified view of a region where CNTs buckled and interface delamination occurred...... 65 Fig. 4.4. SEM images illustrating the deformation of a CNT forest coated with 1 µm Al that was indented on the edge. Corresponding points on the indentation stressstrain curve are labeled a-e.
Fig. 4.5. Stress-strain curve of indentation on the edge of a CNT forest coated with 1 µm Al. Points corresponding to those in Fig. 4.4 are labeled a-e.
Fig. 4.6. a) The same image as in Fig. 4.4c with a vertical line at the center to guide the eye. The inward arrow shows the direction of the buckling, b) side view of the out-of-plane deformation of the CNT forest along the vertical line in (a). The left arrow shows the direction of the buckling, c) schematic of the bending moment and stress distribution in an element highlighted in (b), d) Curvature of the section of CNT forest under the buckle location.
Fig. 4.7. Permanent deformation morphologies that resulted from flat punch indentation on the edge of a) an uncoated CNT forest, and CNT forests coated with b) 100 nm and c) 1 µm Al coatings. The scale bars are 50 µm.
xii Fig. 4.8. Deformed CNT forests showing the effects of uncoated CNT forest height nonuniformities on the buckling-induced delamination. a) Two locations of local delamination –showed by arrows - directly under two hills. b) a taller – 300 µm CNT forest buckled and delaminated similar to the case of 100 µm tall CNT forest. The right side showed by arrow is a region with lower height in which no buckling and delamination occurred.
Fig. 5.1. Micro-indentation apparatus used to visualize compression loading of CNT forests (in Air Force research lab). a) a schematic showing the functional components of the sample stage and loading unit – the inset shows the apparatus configured inside the SEM, b) photograph of the apparatus outside of the SEM.
Fig. 5.2. Deformed uncoated and alumina-coated 10 um tall CNT forests after microindentation. a) uncoated CNT forest buckled from bottom, b) CNT forest with 2 nm coating buckled from bottom similar to uncoated CNT forest, c) magnified view of the area inside the rectangle in (b) showing the multiple buckles close to substrate, d) magnified view of the area inside the rectangle in (c) showing curved CNTs in the top most buckle with no fracture, e) CNT forest with 10 nm coating showing nanotube fracture, f), and g) magnified view of the rectangles in (e) showing fractured nanotubes – arrows indicate some of the bare CNTs pulled out of the alumina coating.
Fig. 5.3. Deformed uncoated and alumina-coated 40 μm tall CNT forests after microindentation. a) uncoated CNT forest buckled from bottom, b) CNT forest with a 2 nm coating buckled from bottom similar to uncoated CNT forest, c) CNT forest with a 10 nm coating showing nanotube fracture