DNA Detection Using Organic Thin Film Transistors:Physical Origins of Electrical Transduction Behavior
and Optimization of Sensitivity
Lakshmi JagannathanVivek Subramanian
Electrical Engineering and Computer SciencesUniversity of California at Berkeley
Technical Report No. UCB/EECS-2009-11
http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-11.html
January 22, 2009
Copyright 2009, by the author(s).All rights reserved.
Permission to make digital or hard copies of all or part of this work forpersonal or classroom use is granted without fee provided that copies arenot made or distributed for profit or commercial advantage and that copiesbear this notice and the full citation on the first page. To copy otherwise, torepublish, to post on servers or to redistribute to lists, requires prior specificpermission.
1
DNA Detection Using Organic Thin Film Transistors:
Physical Origins of Electrical Transduction Behavior
And Optimization of Sensitivity
by Lakshmi Jagannathan
Submitted to the Department of Electrical Engineering and Computer Sciences, University of
California at Berkeley, in partial satisfaction of the requirements for the degree of Master of
Science, Plan II.
Approval for the Report and Comprehensive Examination:
Committee:
Professor Vivek Subramanian
Research Advisor
Date:
* * * * * * *
Professor Ali Javey
Second Reader
Date:
2
Acknowledgements I am indebted to many people for directly and indirectly helping me with this thesis. I want to
thank my advisor, Vivek Subramanian, for his advice and support throughout. He has been a great
role model, and I have learned a lot working with him for the past few years. I would also like to
thank our entire EECS Organic Electronics group at Berkeley, for their advice, encouragement,
knowledge transfer, immediate willingness to help, and in general, just making me love to come to
the office everyday . I am grateful to Dr. Qintao Zhang, a graduate from our group, for the
knowledge transfer of the biosensor system, and his willingness to help to this day. I would also like
to thank Dr. Kanan Puntambekar, a postdoc graduate from our group, for helping me better
understand and explain different aspects of my project. I want to thank Daniel Forchheimer for his
help with the DNA microfluidics experiment, and his contribution of valuable ideas to my project. I
would like to express thanks to Professor Ali Javey for being a second reader for my thesis, and for
his encouragement during my first years in graduate school at Berkeley. I want to also acknowledge
Dr. Steve Ruzin and the staff at the College of Natural Resources, for their training and help with the
fluorescent microscopy system. I want to thank Ruth Gjerde and Pat Hernan at Berkeley for their
constant encouragement; they always believed in me more than I did! I would like to thank
Semiconductor Research Corporation (SRC) and Intel for sponsoring and funding me for graduate
school. I am grateful to Karen, Soundarya, Bharath, Kristen, Sridevi, Dhevi, and Shalini, for their
constant support and encouragement, and in particular for proofreading this document. I want to
also thank my entire youth group for their encouragement and help with everything. I would like to
thank all of my friends for being there for me when I needed them. Finally, I want to express my
heartfelt gratitude to my spiritual guru and guide, Sri Sathya Sai Baba, my entire family in Dallas,
and my family in San Jose for their love and incredible support, without which this truly would have
not been possible!
3
Table of Contents
Acknowledgements ......................................................................................................................... 2
Table of Contents ............................................................................................................................ 3
List of Figures ................................................................................................................................. 4
List of Tables .................................................................................................................................. 5
I. Introduction ............................................................................................................................... 6
A. Abstract .................................................................................................................................. 6
B. Motivation .............................................................................................................................. 6
C. Pentacene Organic Thin Film Transistors for Biosensing ................................................... 10
II. DNA Immobilization and Doping........................................................................................... 13
A. DNA Immobilization on Pentacene Film ............................................................................ 13
B. DNA and Pentacene Interaction .......................................................................................... 14
III. Investigation of Physical Origins of Electrical Transduction Behavior .................................. 19
A. Fluorescent Microscopy ....................................................................................................... 19
B. Time-of-Flight Secondary Ion Mass Spectroscopy (TOF-SIMS) ........................................ 23
i. Definition and Technical Capabilities .............................................................................. 23
ii. Experiment, Procedure, and Results ............................................................................... 25
IV. Optimization of DNA Immobilization and Sensor Sensitivity ............................................... 30
A. Pentacene Characterization Experiment .............................................................................. 30
i. Experimental Setup: Response Surface Design of Experiments (DOE) ........................... 31
ii. Experiment Protocol ......................................................................................................... 33
ii. Results ............................................................................................................................... 34
B. Optimization of DNA Immobilization and Sensor Sensitivity ............................................ 51
i. Experiment Protocol ......................................................................................................... 51
ii. Control Experiments ........................................................................................................ 52
iii. Results: Optimization of DNA Immobilization and Sensor Sensitivity........................... 53
V. Conclusion and Future Work .................................................................................................. 57
VI. Sources ................................................................................................................................... 58
APPENDIX ................................................................................................................................... 60
4
List of Figures
Figure 1. DNA is a building block for everything within a human being. ..................................... 7
Figure 2. Ideal DNA Detection System. ........................................................................................ 7
Figure 3. Process flow for formation of printed FETs. .................................................................. 9
Figure 4. Pentacene TFT Structure. ............................................................................................. 10
Figure 5a. Double stranded DNA chemical structure ......................................................................
Figure 5b. Pentacene (5-benzene ring) .......................................................................................... 14
Figure 6. Current-Voltage characteristics from an experiment show a positive Vt shift and a
higher hole current after DNA immobilization. ............................................................................ 15
Figure 7. Surface potential measurement of DNA molecules on a pentacene film. ..................... 16
Figure 8. Representative SCLC curves of original pentacene-based transistors and DNA ......... 18
Figure 9. Fluorescent Microscope System.. ................................................................................. 20
Figure 10. Axioimager M1 Fluorescent Microscope Set-up at the College of Natural Resources
(CNR) in Berkeley. ....................................................................................................................... 20
Figure 11. Fluorescent density images captured using the digital CCD camera in the fluorescent
microscope system. ....................................................................................................................... 21
Figure 12. Central Composite Design Methodology. .................................................................. 31
Figure 13. A set of 9 points was measured for each of the 17 experiments in the DOE to study
variation within the wafer and wafer-to-wafer variation. ............................................................. 33
Figure 14. Pentacene TFT structure. ............................................................................................ 34
Figure 15. Idsat (electrical characteristics) for all 17 experiments. ............................................... 35
Figure 16. Explanation of the t-test mean diamonds used for statistical purposes. ..................... 35
Figure 17. Leverage Plots relating input and output parameters using the measurements
performed.. .................................................................................................................................... 37
Figure 18. Relationship between the input parameters and Idsat. .................................................. 38
Figure 19. Schematic diagram depicting the phases and grain orientation in various pentacene
film. ............................................................................................................................................... 38
Figure 20. Interaction Profiler Plot for Idsat. ................................................................................. 40
Figure 22. Id- Vd characteristics from 2 different experiments with the lowest and highest
performing samples. ...................................................................................................................... 42
Figure 23. Relationship between the input parameters and mean roughness. ............................. 42
Figure 24. Interaction Profiler Plot for Mean Roughness. ........................................................... 44
Figure 25. AFM Images of samples from 2 different experiments with the lowest and highest
surface roughness parameters. ..................................................................................................... 45
Figure 26. Relationship between the input parameters and grain size. ........................................ 46
Figure 27. Interaction Profiler Plot for Grain Size. ...................................................................... 48
Figure 28. AFM images of samples from 2 different experiments consisting of the smallest and
biggest grain sizes ......................................................................................................................... 48
Figure 29. Relationship between the input parameters and evaporation rate and coverage area. 49
Figure 31. Control Experiments performed to neutralize buffer solution and DI water effects on
pentacene TFTs. ............................................................................................................................ 52
Figure 32. Relationship between input parameters and Idsat ratios. .............................................. 53
Figure 33. AFM Image of terrace-like formation of pentacene thin film on SiO2 substrate. ....... 54
Figure 35. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-
200................................................................................................................................................. 60
5
Figure 36. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio:
100-300 ......................................................................................................................................... 61
Figure 37. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio:
300-700 ......................................................................................................................................... 62
Figure 38. Mass Spectra of negative ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-
100................................................................................................................................................. 63
Figure 39. Mass Spectra of negative ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-
140................................................................................................................................................. 64
Figure 40. Mass Spectra of negative ions of interest for samples 1 to 3; Mass to Charge Ratio:
100-400 ......................................................................................................................................... 65
List of Tables
Table 1. Normalized positive ions of interest .............................................................................. 27
Table 2. Normalized negative ions of interest ............................................................................. 27
Table 3. Input and Outputs for Pentacene Characterization Experiment. .................................... 30
Table 4. JMP Surface Response DOE for Pentacene Characterization Experiment. .................. 32
6
I. Introduction
A. Abstract
Advancements in research and technology are happening now more than ever in the
biotechnology industry. Better detection and treatment methods are constantly being sought by
researchers. Collaboration among different fields of science and technology has brought us one
step closer towards achieving this goal. The vision of achieving a personalized system of disease
detection and treatment has triggered the ultimate purpose of my project: to be able to detect
genetic diseases using electrical means of sensing and detection. Previous work has shown the
potential of organic (pentacene) thin film transistors (OTFT) for DNA detection by showing
different electrical performance shifts in response to single and double stranded DNA[5]. The
goal of this thesis is to present two aspects of using OTFTs for genetic disease detection, namely
the physical origins of the observed electrical shifts, and the characterization of the pentacene
surface to allow optimization of the same for DNA immobilization and sensor sensitivity.
B. Motivation
Using OTFTs for DNA (Deoxyribonucleic acid) microarray technology paves the path for an
ideal DNA detection system. DNA detection is important to detect mutations that cause genetic
diseases. A genetic disorder is a disease that is caused by an abnormality in an individual's DNA.
Abnormalities can range from a small mutation in a single gene to the addition or subtraction of
an entire chromosome or set of chromosomes.
7
Figure 1. DNA is a building block for everything within a human being [1].
An ideal detection system would consist of an ultra low-cost, disposable DNA detection chip,
which can be analyzed with a handheld device (e.g. PDA), using a quick and easy testing
protocol. Moreover, the results/feedback from the chip would be available within a few hours.
Figure 2. An Ideal DNA Detection System. (Image: Courtesy of Dr.Qintao Zhang, Graduate from EECS Organics Electronics Group, UC Berkeley)
Gene: Portion of an organism‟s DNA with coding and non-coding
sequences.
Chromosome: formed from a single DNA molecule that contains many genes.
8
Use of organic transistors facilitates the possibility of an ultra low-cost, disposable DNA chip
through printed electronics. Ultra-low cost printed electronics have gained a great deal of
interest over the past few years because of their promise to greatly reduce the cost of many
electronic applications. They seek to reduce the manufacturing cost of electronics with less
expensive, all-additive printing methods that conventional silicon manufacturing cannot
replicate. There are several applications for low-cost printed electronics including radio
frequency identification (RFID) tags, electronic sensors, displays, smart cards, packaging, and
printed circuit boards (PCB). Based on performance requirements for the above applications,
suitable electronic materials for printing are typically examined, characterized, and implemented.
For example, soluble gold nanoparticle ink for metallization, printable organic dielectrics
including polyimide and PVP, and a high performance organic semiconductor such as a
pentacene precursor semiconductor ink have all been used to make a fully printed organic thin
film transistor [2]. All of these materials have plastic substrate compatible activation
temperatures (<200°C).
Printed electronics using organic semiconductors have many advantages over typical silicon
processes. Although silicon processing uses many of the ideas in printing to achieve low costs
and rapid manufacturing, certain limitations make it difficult to reduce costs further. Some of the
high costs associated with silicon processing include energy because of the high thermal budget
and requirement of ultra-clean processes, and the capital and process cost expenditures associated
with pattern transfer or photolithography due to the extra time and material (>90% material
wasted overall) needed by the lithography processing steps [2]. The processing steps for printed
electronics have a low thermal budget, allowing for a variety of low-cost and flexible substrates,
and additive processing, through which material is only deposited where needed, reducing the
overall consumption of materials. Figure 3 depicts the process flow for printed OTFTs.
9
Figure 3. Process flow for formation of printed FETs. (Image: Courtesy of Dr.Vivek Subramanian, UC Berkeley)
Printed OTFTs are particularly suitable for sensor applications because of the ability to print
multiple active layers on the same substrate and the bottom gated structure. The ability to print
multiple layers on the same substrate is very valuable for biosensor applications because it gives
the DNA microarray technology (explained in the following section) another dimension. The
bottom gated structure allows for the channel layer to be exposed. This allows for easy and non-
destructive inclusion of analytes, bio-molecules/proteins, etc., as the final „layer.‟ Moreover, the
open channel layer allows for easy electrical measurements before and after inclusion of the
objects that are to be sensed. DNA sensitivity is measured using the saturation current before
and after the immobilization of DNA. Having this layer open is critical for this step in the
experiment to take place.
In this work, evaporated pentacene transistors were used instead. As we will discuss later in
the thesis, the morphology of the pentacene surface is critical to the immobilization of the DNA
and the sensitivity of the sensor. By evaporating pentacene instead of printing it, we are able to
control this surface structure very precisely. For the purposes of the work covered in this thesis,
the control of this surface was very important for optimization. Therefore, evaporated pentacene
10
transistors were used. Translating from evaporated to printed transistors, in terms of the
fabrication processes and output characteristics, has previously been mastered by our group and
will be implemented in the future work of this project.
C. Pentacene Organic Thin Film Transistors for Biosensing
The pentacene TFT bottom gated structure used throughout this project is shown in Figure 4
below.
Figure 4. Pentacene TFT Structure.
The thickness of the pentacene can be varied as necessary, but for our purposes, it ranges from
10-30nm. The steps for fabricating the transistor and including DNA on the pentacene surface
will be discussed further in chapter IV of the thesis.
Pentacene TFTs are potentially useful for biosensors because of the advantages they provide
over current state of the art fluorescent detection technology. The current DNA microarray/chip
technology involves a meticulous process. For example, to determine whether an individual
possesses a mutation for BRCA1 or BRCA2, genes whose mutations are known to cause as many
as 60% of all cases of hereditary breast and ovarian cancers [3], a scientist first obtains a sample
of DNA from the patient's blood as well as a control sample, one that does not contain a mutation
in either gene. The researcher then denatures the DNA in the samples, a process that separates
100nm SiO2
Source
(100nm)
Drain
(100nm)
Gate (n++ Si)
Pentacene
11
the two complementary strands of DNA into single-stranded molecules. The long strands of
DNA are then cut into smaller, more manageable fragments. Each fragment is then labeled by
attaching a fluorescent dye. The individual's DNA is labeled with green dye and the control
(normal) DNA is labeled with red dye. Both sets of labeled DNA are then inserted into the chip
and allowed to hybridize (bind) to the synthetic BRCA1 or BRCA2 DNA on the chip.
If the individual does not have a mutation for the gene, both the red and green samples will
bind to the sequences on the chip. If the individual does possess a mutation, the individual's
DNA will not bind properly in the region where the mutation is located. The scientist can then
examine this area more closely to confirm that a mutation is present [3].
With our pentacene TFT sensor, we propose to do the same mutation detection electrically.
Instead of using fluorescent laser technology to correlate the extracted fluorescent signals with
mutations, we will detect the mutations by analyzing conductance and threshold voltage shifts.
Our technology has several advantages compared to the current optical detection technology:
1) Higher Sensitivity: As many previous researchers have already shown, electrical detection
is known to be more sensitive than optical detection [4]. If the detector is more sensitive,
needless to say, the diseases can be detected earlier (important for early cancer detection,
for example) and much more accurately.
2) Label free method: The DNA used with our proposed technology does not need to be
tagged with fluorescent molecules. This means that the expensive fluorescent laser
detection system can also be eliminated. The whole process becomes simpler and cheaper.
3) Disposable/Portable/Ultra-low cost: Adding to the previous point, the elimination of the
fluorescent laser detection system makes it cheaper and easily accessible. Also, since
12
OTFTs can potentially be printed on plastic substrates, the disposability and low-cost
factors are further made easier. More importantly, it is impossible to make a disposable
DNA detection chip unless it is done with an electrical or a label-free method.
4) Faster Process: The method of hybridization, which takes 12-24 hours, poses a bottleneck
for the current DNA microarray technology process. Amongst other steps that could
potentially save time in our proposed technology, the possibility of pulse-enhanced
hybridization [5], which could potentially reduce the hybridization time to milliseconds,
makes our proposed electrical detection process much faster.
Having thus motivated and introduced the topic, the next few sections will discuss the sensor
system further. Section II describes the immobilization and doping mechanisms of DNA on
pentacene. Section III discusses the methods that have been used and will be used to correlate
the amount of DNA on the surface with the electrical shift observed. Finally, Section IV will
present the effect of pentacene film formation on DNA detection behavior, and optimization of
the pentacene surface in terms of morphology and topology for highest DNA immobilization, and
more importantly, highest sensitivity for DNA detection.
13
II. DNA Immobilization and Doping
A. DNA Immobilization on Pentacene Film
Previously, it has been shown that pentacene can be used to detect DNA [5]. Through
physical absorptive immobilization, DNA immobilizes on pentacene film. The pentacene film is
hydrophobic, and DNA is diluted in saline-sodium citrate (SSC) buffer solution. The 20x
solution of buffer consists of 3M sodium chloride and 300 mM trisodium citrate. The buffer
solution is further diluted in water, as it is usually done in practice, to give a 2x concentration
solution. The buffer can be considered salt water for analytical purposes.
The hydrophobic interaction immobilization (interactions between ssDNA molecules and
crevices on hydrophobic substrate surface) was a very common method in the early stage of DNA
microarray development and has been studied very well [6]. Using this method, we can explain
the pentacene film interaction with DNA. When the DNA in buffer solution is pipetted onto the
pentacene surface, the DNA segregates to hydrophobic „holes‟ on the surface. During this
process, the salt water, in which the DNA is diluted, gets pushed out, „physically‟ immobilizing
the DNA in these crevices.
Initial hypotheses led us to believe that roughness and grain boundaries in particular
immobilize more DNA, and would give the highest sensitivity. Recent experiments, which will
be presented in later sections of the thesis, have shown that in addition to the morphological
features, the overall sensitivity is more strongly dependent on how much DNA is able to diffuse
to the part of the pentacene film which contains the channel. The channel is contained within the
first few monolayers of pentacene film.
14
B. DNA and Pentacene Interaction
The sensing of DNA is possible because of DNA interaction with pentacene. The DNA and
pentacene chemical structures are shown in Figure 5a and 5b below:
Figure 5a. Double stranded DNA chemical structure [7]. Figure 5b. Pentacene (5-benzene ring)
chemical structure [8].
Pentacene TFTs are p-type accumulation devices (negative gate voltage and negative
current). After DNA immobilization on surface, the hole current of the TFT increases.
Consequently, the threshold voltage (Vt) becomes more positive (i.e. shifts to the right). The
result from one of the experiments, shown in Figure 6 below, delineates the sensor characteristics
observed after DNA interaction with pentacene.
Ion Ratio is defined as:
Ion Ratio = onmobilizatiBeforeDNAI
onmobilizatiAfterDNAntDrainCurreI
d
d
Im
Im)(
The bigger the Ion Ratio, the higher the sensitivity of the sensor. The DNA used in the
experiment below and all of the experiments/results that will be presented in the thesis, is a 125-
base pair designed single strand DNA sequence, synthesized by Biosynthesis, Inc.
Pentacene (5 benzene-ring) chemical structure
Benzene Structure
Single Strand DNA chemical structure
15
Figure 6. Current-Voltage characteristics from an experiment show a positive Vt shift and a higher hole current after
DNA immobilization.
Two possible explanations of the interaction of DNA and pentacene will be discussed below.
One is the direct doping of DNA on pentacene. The net-effect of the DNA on the pentacene
surface is electron withdrawing, therefore increasing the hole current of the transistor. Although
DNA is known to be negatively charged because of the phosphate groups on its backbone, the
possible interaction of the two chemical structures shown in Figure 5 might be causing the shift
in hole current. More specifically, areas of interaction include the aromatic structures in base
pairs (such as adenine) of the DNA, and the inherent aromaticity of the 5 benzene-ring pentacene
structure. The result of this electron withdrawing interaction is supported and explained below
using the Kelvin Probe Microscopy technique.
Another explanation uses the trap filling caused by the electrical interactions with DNA to
improve the overall performance of the transistor, resulting in the shift seen in Figure 6. Using
SCLC (Space Charge Limited Current) measurement technique, Zhang et al are able to support
this hypothesis. A short summary of Dr. Zhang‟s work on these explanations is discussed below.
Id (drain current) before DNA immobilization Id (drain current) after DNA immobilization
Positive Vt shift
Ion Ratio
16
Further details on the doping mechanisms of DNA can be found in his thesis, OTFT Based DNA
Detection System.
Dr. Zhang performed Atomic Force Microscopy (AFM) measurements to verify the DNA
doping of pentacene. Figure 7 below shows an AFM image of a DNA molecule along with
potential measurements done through Kelvin probe microscopy [5].
a)
b) Figure 7. Surface potential measurement of DNA molecules on a pentacene film. a) An Atomic Force Microscopy
(AFM) image, where the single bright thread is an immobilized DNA molecule. b) A surface potential image matching the same area of the AFM image on the top. It is clear that the DNA molecule has negative potential, which
is possibly caused by electron-withdrawing from the bottom pentacene film [5].
Kelvin Probe Force Microscopy (KPFM) is a scanned probe method where the potential
offset between a probe tip and a surface can be measured using the same principle as a
macroscopic Kelvin probe. The cantilever in the AFM is a reference electrode that forms a
17
capacitor with the surface, over which it is scanned laterally at a constant separation. With KPFM
the workfunction of surfaces can be observed at atomic or molecular scales. The work function
relates to many surface phenomena, including catalytic activity, reconstruction of surfaces,
doping and band-bending of semiconductors, charge trapping in dielectrics and corrosion [12].
Unlike workfunction mismatch induced surface potential seen in solid thin film study though, the
surface potential of macro bio-molecules is caused by dipoles. DNA hybridization or antigen-
antibody reactions have been proved to increase the amplitude of potential differences [11]. The
bright thread seen in Figure 7a (shown by the green and red arrows on the image) represents
DNA. The KPFM results in Figure 7b show that at the location of the DNA, the potential
becomes negative indicating that DNA, as a result of the immobilization and interaction with the
pentacene surface, has become more negatively charged. Given these results of the KPFM
measurements, Dr. Zhang is able to conclude that the DNA molecule has negative potential,
possibly caused by electron-withdrawing from the bottom pentacene film [5].
A space charge limited current (SCLC) measurement explains the doping mechanism as well.
SCLC is a result of the space charge region that forms within a conductive semiconductor
material where the charge carriers have diffused away, or have been forced away by an electric
field. Space charge regions have been shown to exist in pentacene films with gold electrodes in
previous studies [29]. The space charge region is characterized by generation-recombination
centers, steep impurity gradients, and rapidly changing populations of holes and electrons. This
space charge region limits the current carried by the majority carriers because of the presence
traps within this region. The electrical impact of DNA, with its negative charge, is to fill these
traps, facilitating the flow of hole current.
An SCLC setup measures two-terminal devices. A voltage from 0 to 100V is swept between
the source/drain with a floating gate electrode, and the current flowing into the drain is measured.
18
A typical SCLC curve is plotted on a log-log scale and has more than two different slopes. The
slope=1 part of the curve is called ohmic regime and the slope≥ 2 part is the trap-filling regime.
As the electrical field is increased to a certain level, the concentration of the injected carriers
from the electrode overwhelms the intrinsic carrier concentration, thus starting to fill traps in the
films. At a slope very close to 2, all traps been filled. This trap-free regime, slope equal 2, is also
called SCLC regime [5]. The SCLC curve, measured by Dr. Zhang, is shown in Figure 8 below.
Figure 8. Representative SCLC curves of original pentacene-based transistors and DNA immobilized transistors [5].
SCLC measurements before and after DNA immobilization shown above illustrate the doping
mechanism of DNA molecules on the pentacene film. In the ohmic regime, the magnitude of current
increased, relating to the free carrier concentration increase or mobility increase, because of
immobilized DNA segments. In the trap-filling regime, the decrease in slope from 3.4 as measured
in the original pentacene transistor to 2.7 in the DNA immobilized pentacene transistors indicates
that deep traps have been filled by DNA molecules. Figure 8 shows that the trap-filling action in
hole-rich pentacene film has actually increased the hole concentration. This is because trapped
electrons in the pentacene film have been released by DNA segments, thereby facilitating the flow of
hole current [5, 6].
19
III. Investigation of Physical Origins of Electrical Transduction
Behavior
An important step in making a viable DNA sensor is to show that what is being detected is
just the DNA, and to eliminate any other noise that might affect the sensitivity. One way to do
this is to correlate the amount of DNA on the surface with the electrical shift that is seen after
DNA immobilization. A few methods were used to obtain this data, namely atomic force
microscopy (AFM), fluorescent microscopy, and time-of-flight secondary ion mass spectroscopy
(TOF-SIMS). Useful results and insight were derived from the latter two techniques and are
discussed in this section.
A. Fluorescent Microscopy
Fluorescent microscopy is a light microscope used to study properties of organic or inorganic
substances using the phenomena of fluorescence and phosphorescence instead of, or in addition
to, reflection and absorption. A component of interest in the specimen is specifically labeled
with a fluorescent molecule called a fluorophore (such as green fluorescent protein (GFP),
fluorescein).
The ssDNA used in this experiment was tagged with carboxyfluorescein molecule on the
5‟end of the DNA strand. Carboxyfluorescein is a fluorescent dye with an excitation and
emission of 492/517 nm, respectively. In other words, the DNA with the fluorescent molecule is
illuminated with light of wavelength of 492nm which is absorbed by the fluorophores. It then
emits a light of wavelength of 517nm, which is in turn passed by an emission filter specific to the
emitted wavelength. The illumination light is separated from the much weaker emitted
fluorescence through the use of the emission filter. Typical components of a fluorescence
microscope are the light source (xenon arc lamp or mercury-vapor lamp), the excitation filter, the
20
dichroic mirror (or dichromatic beamsplitter), and the emission filter [14]. Figure 9 below
delineates the fluorescent microscopy system.
Figure 9. Fluorescent Microscope System [13].
A Zeiss AxioImager M1 fluorescence microscope system from the College of Natural
Resources, Berkeley, was used in this experiment. The microscope contains the Photometrics
Quantix digital CCD camera that allows visualization of fluorescence with high spatial resolution
(1200X1300 pixels) and high bit depth (12-bit gray scale). A second camera, QImaging 5MPix
MicroPublisher, captures color images. The set-up is shown in Figure 10 below.
Figure 10. Axioimager M1 Fluorescent Microscope Set-up at the College of Natural Resources (CNR) in Berkeley.
The CNR website (microsopy.berkeley.edu) provides further details on the mechanism and
capabilities of the fluorescent microscope system shown above. Experimental results/images
21
obtained from fluorescent density measurements are shown below. The area shown below is just
that of the channel area, i.e. 110um*1mm (horizontal orientation), where the DNA is pipetted.
a)
b)
Figure 11. Fluorescent density images captured using the digital CCD camera in the fluorescent microscope system. The green dots in images a) and b), which are to represent green-fluorescent tagged DNA, are from 2 substrates with
different pentacene thicknesses and concentrations of DNA.
The green dots seen above represent the green-fluorescent tagged DNA. The substrate in
11a, given the pentacene evaporation conditions that were used, is expected to have a rougher
surface (referring to results shown in Section IV) than that of 11b. Correspondingly, more DNA
can be seen immobilized in 11a. The quantitative measurement that was used to make this
conclusion was fluorescent density measurements. Fluorescent density or intensity
measurements were made using the camera, and its in-built segmentation software tool that adds
up the intensity of different segments of green fluorescence within the imaged channel area.
Before trying to correlate the these results with electrical results, control experiments were done
to see if the fluorescent tag had any effects on the electrical characteristics. No-tag DNA and
fluorescent tagged DNA were pipetted and treated with the same experimental protocol. Results
showed that the fluorescent tag, by itself, causes no additional shift to the electrical
characteristics.
channel area where DNA is pipetted
Pentacene TFT structure on wafer
22
While trying to correlate the electrical and visual results, some potential problems were
discovered. After doing some control experiments on the pentacene surface, it was observed
that the pentacene substrate also fluoresces, contributing its own fluorescent signal to the
measurements above. This is to be expected of an organic semiconductor such as pentacene.
Fluorescence occurs when a molecule, atom or nanostructure relaxes to its ground state after
being electrically excited. State S0 is called the ground state of the fluorophore (fluorescent
molecule) and S1 is its first (electronically) excited state [14].
A molecule in its excited state, S1, can relax by various competing pathways. It can undergo
non-radiative relaxation in which the excitation energy is dissipated as heat (vibrations) to the
solvent. Excited organic molecules can also relax via conversion to a triplet state which may
subsequently relax via phosphorescence or by a secondary non-radiative relaxation step.
Relaxation of an S1 state can also occur through interaction with a second molecule through
fluorescence quenching. Finally, molecules that are excited through light absorption can transfer
energy to a second 'sensitized' molecule, which is converted to its excited state and can then
fluoresce [14]. The latter case is likely the cause of the fluorescent „noise‟ that we are seeing
with the measurements above. It is not easy to separate this from the actual signal unless the
DNA is tagged with a fluorescent molecule which fluoresces at a wavelength that does not excite
the pentacene molecule. Alternatives that are being considered will be discussed in the last
paragraph of this section.
The second problem with fluorescent microscopy using the digital CCD camera is the
resolution of the camera. The Max XY resolution of this fluorescent microscope is about 200nm.
With the resolution of the available CCD camera, 100s of molecules needs to be clumped
together in the 200nm space for the system to detect the fluorescent signal. As discovered
23
through AFM, the distribution of DNA on the surface is random. In other words, the DNA
density in different scan areas is not expected to be consistent. Given the resolution of the digital
CCD camera, the random distribution makes it difficult to accurately capture the amount of DNA
on the surface.
To keep the pentacene substrate from fluorescing, different control experiments with a range
of emission filters were tried on the surface. We were able to conclude that pentacene does not
fluoresce at blue light wavelengths (450-495nm [15]) or below. In addition, to get as close as
possible to the resolution of the Single Molecule Spectroscopy (SMS) [16], which would allow
measurement of a single fluorescent molecule or a single strand of DNA, the College of Natural
Resources at Berkeley has recently obtained a new EMCCD (electron-multiplying charge
coupled device) camera, which is highly sensitive and super bright. It is much more sensitive
than the camera that has been used so far.
Since results with the current system are promising, future experiments for correlation will be
performed with blue fluorescent tagged DNA and the new EMCCD camera.
B. Time-of-Flight Secondary Ion Mass Spectroscopy (TOF-SIMS)
This section covers the use of TOF-SIMS technique for correlation and the results obtains from
the experiment.
i. Definition and Technical Capabilities
Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS) is a surface analytical
technique that focuses a pulsed beam of primary ions onto a sample surface, producing secondary
ions in a sputtering process. Analyzing these secondary ions provides information about the
24
molecular and elemental species present on the surface. For example, if there were organic
contaminants, such as oils adsorbed on the surface, TOF-SIMS would reveal this information,
whereas other techniques may not. Since TOF-SIMS is a survey technique, all the elements in
the periodic table, including H, are detected. Moreover, TOF-SIMS can provide mass spectral
information, image information in the XY dimension across a sample, and also depth profile
information on the Z dimension into a sample [17].
The surface sensitivity of TOF-SIMS makes it a good first pass at problem solving. Other
techniques can be used to obtain in depth information. The imaging capabilities of TOF-SIMS
can provide elemental and molecular information from defects and particles on the micron scale.
TOF-SIMS can also be used for depth profiling and compliments dynamic SIMS. The
advantages for TOF-SIMS for profiling are its small areas capabilities and also its ability to do
survey depth profiles [17].
TOF-SIMS is known to be one of the most sensitive surface analytical techniques, with a
detection limit of 107- 10
10 atoms/cm
2 sub-monolayer and a depth resolution of 2-3nm. It
provides specific molecular information on thin (sub-monolayer) organic films/contaminants
with an excellent detection limit (ppm) for most elements. The TOF-SIMS technique was used
instead of SIMS because TOF-SIMS is able to perform the SIMS analysis on organic thin films
and more specifically soft substrates. It is a non-destructive process tailored to work with soft
substrates such as pentacene. TOF-SIMS has a few limitations. The samples that need to be
analyzed must be vacuum compatible. In addition, this technique can be too surface sensitive
with sample packaging; prior handling may also impact quality of results. Despite the
limitations, the TOF-SIMS technique is extremely sensitive, reliable, and very useful in many
applications.
25
ii. Experiment, Procedure, and Results
The TOF-SIMS analysis was performed in the Evans Analytical Group labs in Sunnyvale,
CA. The TOF-SIMS analysis was performed to get a visual proof of DNA on surface and to see
if it is possible to correlate these results with the electrical measurements.
More specifically, the purpose of this analysis was to determine the relative levels of DNA on
three samples: Sample #1- Control: Pentacene substrate with no buffer or DNA but which has
gone through the same experimental protocol (fabrication and storage) as the other two; Sample
#2-Buffer: Pentacene substrate with just pure buffer and NO DNA. The buffer, as discussed
earlier is saline-sodium citrate solution; Sample #3-DNA in buffer: DNA was pipetted on the
channel of the transistor. Note that all 3 samples were taken from the same substrate; each of the
samples went through the same cleaning, evaporation, and storage procedures. The relative
levels of DNA were assessed by monitoring the levels of sodium (Na) and phosphate (POx) ions.
Experiment:
Data was obtained using a gallium liquid metal ion gun (LMIG) primary ion source (Ion
Potential: 12 kV for +ions and 18kV for –ions; DC Ion Current: 2nA). The instrument was
operated in an ion microprobe mode in which the bunched, pulsed primary ion beam was rastered
across the sample's surface. Three positive and three negative ion spectra were acquired from
each sample in order to investigate the reproducibility of the data. Acquisition of multiple spectra
serves to highlight possible chemical heterogeneity across the sample‟s surface. In each case the
analytical area was 80m × 80m. Spectra were acquired by specifying a particular area of the
sample as a “region of interest” (ROI). This allowed data to be acquired only from the channel,
eliminating signal from the surrounding gold lines.
The elements of interest were POx (for DNA) and Na (for buffer solution).
26
Summary of Results:
The highest levels of POx and Na were observed on sample #3. These species were observed
at significantly lower levels from sample #2. No POx was observed from sample #1 which had
the lowest levels of Na.
Other species observed from all samples included the pentacene substrate (C22H14),
Polydimethylsiloxane (PDMS), which is a signal observed because of the hexamethyldisilazane
(HMDS) layer in the dielectric/pentacene interface, and F, Fluorine.
Discussion of Results:
The results are presented as mass spectra, which are displayed as the number of secondary
ions detected (Y-axis) versus the mass-to-charge (m/z) ratio of the ions (X-axis). The ion counts
are shown on linear intensity scales, and probable empirical formulae for a number of peaks are
identified in the figures. The multiple spectra obtained from each sample were generally very
similar so only one representative spectrum from each sample in each polarity is included in this
report. Mass Spectra results can be found in the appendix of this report.
The results are also presented as tables of normalized intensities. These tables can be used to
compare the level of a given species between samples; however, they cannot be used to compare
the levels of different species either on the same sample or between samples. This is because
different species have different secondary ion yields so TOF-SIMS has different sensitivities for
different species. The tables show the mean of three measurements and also the standard
deviations.
27
1, control 2, buffer 3, (DNA + buffer)
m/z formula mean mean mean
Elements
23 Na 16.1 6.0 569 146 4790 580
28 Si 3670 180 3610 160 587 38
Pentacene
278 C22H14 105 11 40.4 7.7 55.0 6.2
PDMS
73 C3H9Si 408 31 317 42 95.9 12.0
147 C5H15Si2O 16.5 5.6 73.1 13.0 15.6 1.7
Table 1. Normalized positive ions of interest (normalized relative to total ion counts ×10000)
Table 1 above presents the normalized intensities of the positive ions of interest. Elemental
species observed from all three samples included Na, Si, and F. The levels of Na varied
significantly among the samples with the highest levels observed from sample #3 and the lowest
levels on Sample #1. The highest levels are expected from sample #3 because the DNA
molecules themselves will have the buffer solution (Na) attached to them. Sample #1 most likely
has Na because of contaminants from the environment. Comparable levels of Si were observed
on samples #1 and #2 while lower levels were found on #3. The lower levels in sample #3 are
possibly due to the strong signal exhibited by the Na in this sample. This can also be observed in
the mass spectra included in the appendix of this report.
1, control 2, buffer 3, DNA + buffer)
m/z formula mean mean mean
47 PO nd - 0.337 0.018 0.551 0.150
63 PO2 nd - 0.295 0.022 2.91 0.81
79 PO3 nd - nd - 2.45 0.73
“nd” indicates species not detected on that sample Table 2. Normalized negative ions of interest
(normalized relative to total ion counts ×10000)
28
Table 2 above presents the normalized intensities of the negative ions of interest. The focus was
on the POx ions to determine the presence of DNA on the surface and the relative amounts of
DNA on the different samples. Phosphate ions (POx) were observed in negative polarity only on
samples #2, and #3. The levels of POx on sample #3 were significantly higher than those
observed from sample #2. The PO3¯ ion in particular was observed only on sample #3.
Phosphate ions were not clearly observed in the mass spectra (shown in appendix) due to the
many strong hydrocarbons in the spectra.
Many ions due to HMDS (shown as PDMS in the mass spectra and tables) were observed
from all samples. Several hydrocarbon ions were also observed from all samples, and these are
shown in the mass spectra included in the appendix. Representative low mass positive ions
include C2H3, C3H5, and C4H7. Representative negative ions include C2H, C4H, and C6H.
Species such as these have multiple possible sources. They may be adsorbed low mass molecular
species or they may be fragment ions from other higher mass molecular species. The parent
molecule of the pentacene substrate (C22H14, m/z 278) was clearly observed on all samples. This
indicates that the coverage of adsorbed species (e.g. DNA and HMDS) is discontinuous and/or
less than a monolayer thick.
TOF-SIMS analysis has shown the presence of phosphate groups, particular PO2 and PO3
(from DNA) and sodium (from buffer solution) in highest concentrations in sample #3, the
sample with DNA immobilized in the channel. The control sample (sample #1) shows small
traces of sodium and the buffer sample (sample #2) shows small traces of PO possibly due to
environment contribution and/or contamination. Since Sample #3 has shown significantly higher
amounts of the species of interest, the presence of DNA on the pentacene surface has been
confirmed.
29
Fluorescent Microscopy and TOF-SIMS have both shown useful results to ascertain the
physical origins of electrical transduction behavior. TOF-SIMS results have confirmed the
presence of DNA on pentacene surface while modified fluorescent microscopy measurements
have shown potential for use in directly correlating the amount of DNA immobilized on the
surface with the shift of the electrical performance characteristics.
30
IV. Optimization of DNA Immobilization and Sensor Sensitivity
DNA immobilization has been previously discussed in Section II of this report. Physical
absorptive immobilization, the mechanism through which DNA immobilizes on pentacene,
highlights the importance of the topology of the pentacene film surface for immobilization of
DNA and the sensitivity of the sensor. DNA is known to segregate to topological features on
pentacene surface. We are able to exploit the control of pentacene evaporation conditions to tune
pentacene film morphology to maximize sensitivity. In this section, we demonstrate DNA
detection using optimized films.
The optimization experiment is split up into two parts. The first is a set of experiments that
characterize the pentacene (vary the pentacene morphology) by varying the input parameters.
The second integrates the DNA into the experiments, thereby arriving at the best surface and
evaporation conditions for highest sensor sensitivity.
A. Pentacene Characterization Experiment
By carefully controlling input parameters to set different evaporation conditions, we analyze
pentacene morphological and electrical characteristics. The input parameters that were varied
and the outputs that were extracted are shown in Table 3 below.
Input Variables Output Characteristics
Thickness of Pentacene film
Input Current (Temperature of the crucible)
Substrate Temperature
Idsat (Saturation Current)
Mean Roughness
Grain Size
Evaporation Rate
Coverage Area
Table 3. Input and Outputs for Pentacene Characterization Experiment.
31
i. Experimental Setup: Response Surface Design of Experiments (DOE)
Because the goal of this part of the experiment was to relate the input and output parameters,
the experiment needed to be tailored to ultimately result in providing these relationships. To
perform this comprehensive analysis, design of experiments (DOE) set-up was used. In
particular, a surface response DOE was used because of the importance of gaining a thorough
understanding of the output characteristics for optimizing DNA immobilization.
The DOE set-up was configured using JMP, a statistical software. The DOE platform in JMP
is an environment for describing the factors, responses and other specifications, creating a
designed experiment. JMP allows the design of different types of DOE including screening
design and response surface design. The former is to get an overall understanding while the
latter is gain an in-depth understanding of the relationships between input and output parameters.
Response surface designs are useful for modeling a curved surface (quadratic) to continuous
factors. If a minimum or maximum response exists inside the factor region, a response surface
model can pinpoint it. Three distinct values for each factor are necessary to fit a quadratic
function, so the standard two-level designs (such as screening design) cannot fit curved surfaces.
The most popular response surface design is the central composite design, illustrated by Figure
12 below [19].
Figure 12. Central Composite Design Methodology [18].
32
CCD combines a two-level fractional factorial (used in screening designs) and two other
kinds of points, namely center points, for which all the factor values are at the zero (or midrange)
value and axial (or star) points, for which all but one factor are set at zero (midrange) and that
one factor is set at outer (axial) values. There are two main types of central composite designs,
namely uniform precision and orthogonal designs. The properties of these central composite
designs relate to the number of center points in the design and to the axial values. For orthogonal
designs, the number of center points is chosen so that the second order parameter estimates are
minimally correlated with the other parameter estimates. Uniform precision, which was used for
the following DOE, means that the number of center points is chosen so that the prediction
variance at the center is approximately the same as at the design vertices [19].
Using a surface response design and more specifically, a uniform precision CCD set up, the
DOE shown in Table 4 below was followed. A set of 17 evaporation conditions was designed
for the experiment.
Table 4. JMP Surface Response DOE for Pentacene Characterization Experiment.
33
The input variables are shown in columns 3, 4, and 5. The output characteristics are in columns
6-10. For the DNA immobilization part of this experiment, 2 or 3 more columns will be added,
namely Idsat ratio, Threshold Voltage (Vt) shift, and mobility shift.
A set of 9 points was measured within each of these experiments in order to study the
variation within wafer and wafer to wafer variation. The measurement setup used for each of the
17 experiments is shown in Figure 13 below.
Figure 13. A set of 9 points was measured for each of the 17 experiments in the DOE to study variation within the
wafer and wafer-to-wafer variation.
ii. Experiment Protocol
The experimental procedure and specifications for the pentacene characterization experiment
is discussed below:
1) N++ silicon as substrate
2) 1000Å wet oxidation for SiO2 (using Tystar oxidation furnace)
3) HMDS on SiO2 surface (to optimize the interface between hydrophobic (photoresist) and
hydrophilic (SiO2) surfaces)
4) Photoresist spinning followed by lithography to define source and drain contacts
5) Evaporation of 1000Å of gold for source and drain contacts
6) Lift-off for 45 minutes in acetone
11 33
22
44
55 66 77
88
99
34
7) Sonication/Cleaning of Substrate- another 45 minutes in acetone followed by 45 minutes
in IPA (isopropyl alcohol)
8) HMDS on surface (to optimize interface between SiO2 (hydrophilic) and
pentacene(hydrophobic))
9) Evaporation of Pentacene using the specified input parameters on the DOE (the pentacene
evaporation was performed under vacuum with a chamber pressure of ~2.2 x10-7
Torr)
Figure 14. Pentacene TFT structure.
The sonication/cleaning step (step 7) is especially important for sensor sensitivity since this
establishes the interface between the dielectric and the semiconductor. DNA immobilizes and
interacts very close to this interface to cause the shift in the transistor characteristics.
ii. Results
Results: Within Wafer and Wafer to Wafer Variation
The results shown in Figure 15 below highlight the variation within each experiment and
across the different experiments. In addition, the wide range of electrical characteristics across
the DOE further emphasizes the importance of controlling these evaporation conditions to
determine the optimum immobilization for highest sensor sensitivity. The x axis of the graph
corresponds with the experiments on the DOE in Table 4.
100nm SiO2
Source
(100nm)
Drain
(100nm)
Gate (n++ Si)
Pentacene
35
The output shown here is saturation current (Idsat). The variation within each wafer and from
wafer to wafer will have a direct correlation with the other output characteristics. If there‟s more
variation (bigger standard deviation) within one experiment, then this is a result of lack of
uniformity of the pentacene surface in that experiment. This relates back to the morphological
characteristics. For example, higher input current or thicker pentacene, depending on which
factor dominates, likely causes non-uniform morphology within a wafer, increasing the variation
in electrical performance. This will be further discussed in the next part of this section.
Figure 15. Idsat (electrical characteristics) for all 17 experiments. These results display the different performance
characteristics across the DOE as well as the variation across the experiments.
The green diamonds seen in Figure 15 do not represent standard deviation. A means diamond
illustrates a sample mean and 95% confidence interval, as shown in Figure 16 below.
Figure 16. Explanation of the t-test mean diamonds used for statistical purposes [19].
36
The line across each diamond represents the group mean. The vertical span of each diamond
represents the 95% confidence interval for each group. Overlap marks are drawn above and
below the group mean. For groups with equal sample sizes, overlapping marks indicate that the
two group means are not significantly different at the 95% confidence level. The confidence
interval computation assumes that variances are equal across observations. Therefore, the height
of the confidence interval (diamond) is proportional to the reciprocal of the square root of the
number of observations in the group [19].
The horizontal extent of each group along the x-axis (the horizontal size of the diamond) is
proportional to the sample size of each level of the x variable. It follows that the narrower
diamonds are usually the taller ones because fewer data points yield a less precise estimate of the
group mean [19]. The t-test/anova analysis shown in Figure 15 also shows the sameness data.
Between two experiments, if the overlap/sameness value (measured by t-test) is <5%, then we
can be „confident‟ that they are statistically different.
The variation can be characterized by the standard deviation value for each of the
experiments. Experiment 9 had the highest standard deviation value of 0.000022, and the
corresponding variation can be seen clearly in Figure 15. Experiment 14 and Experiment 6, had
the lowest variation with a standard deviation of 0.000001.
Results: Relating Input and Output Parameters
1. Saturation Current (Idsat)
Figure 18 below shows the effect that the three input parameters (Input Current, Thickness of
Pentacene, and Substrate Temperature) have on Saturation current (Idsat). Saturation current and
other electrical characteristics were measured using the Agilent HP4156 probe station purged in
37
nitrogen. The measurements were done under nitrogen, as is usually the practice when measuring
pentacene transistors, because pentacene is oxygen-sensitive.
Figures 17 and 19 are extracted by JMP using the 153 experimental/electrical measurements
(17 experiments * 9 measurements for each) performed. Using leverage plots, some of which are
shown in Figure 17 below, and relating the input parameters to outputs, JMP is able to provide a
relationship between the input and output characteristics, which are shown in Figures 17 and 19.
Figure 17. Leverage Plots relating input and output parameters using the measurements performed. Some parameters show a cause-effect relationship while others do not affect the saturation current.
Figure 18 below shows how the three input parameters independently affect the saturation
current.
38
Figure 18. Relationship between the input parameters and Idsat.
As mentioned earlier, the pentacene TFT device is a p-type accumulation device; the Y-axis in
the figure above is negative current representing hole current. The figure above shows that as the
thickness of the pentacene increases, the hole current (magnitude) also increases. This result can
be justified using a schematic of the known growth modes of pentacene shown in Figure 19. The
stacking of the pentacene and surface coverage causes this increase in electrical characteristics.
Figure 19 below shows the schematic diagram of the pentacene throughout a range of
thicknesses, and its effect on the electrical characteristics is further explained below.
Figure 19. Schematic diagram depicting the phases and grain orientation in various pentacene film [20].
The schematic diagram shown above has been produced as a result of GIXD (grazing incidence
X-ray diffraction) characterization done by Dr. Sandra Fritz [20]. The schematic diagram shows
-8e-5
-6e-5
-4e-5
-2e-5
0e+0
2e-5
Idsat
Act
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted P<.0001 RSq=0.41
RMSE=1.4e-5
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.409977
0.369748
0.000014
-0.00002
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
1.8854e-8
2.71339e-8
4.59879e-8
Sum of Squares
2.0949e-9
2.056e-10
Mean Square
10.1911
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
1.1973e-8
1.51609e-8
2.71339e-8
Sum of Squares
2.3946e-9
1.194e-10
Mean Square
20.0591
F Ratio
<.0001
Prob > F
0.6703
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
-0.000014
-0.000008
-0.000003
-0.000007
-0.000007
2.3039e-7
0.0000011
-0.000001
-0.000001
-0.000004
Estimate
0.000003
0.000001
0.000001
0.000001
0.000002
0.000002
0.000002
0.000002
0.000002
0.000002
Std Error
-4.05
-5.89
-2.56
-5.24
-4.03
0.14
0.66
-0.74
-0.89
-2.25
t Rati o
<.0001
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
7.12763e-9
1.34661e-9
5.65133e-9
3.33073e-9
3.8216e-12
9.0222e-11
1.1331e-10
1.6346e-10
1.03827e-9
Sum of Squares
34.6742
6.5509
27.4924
16.2032
0.0186
0.4389
0.5512
0.7952
5.0510
F Ratio
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob > F
Effect Tests
-0.00005
-0.00004
-0.00003
-0.00002
-0.00001
0
0.00001
0.00002
0.00003
0.00004
Idsat
Resid
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted
Residual by Predicted Plot
Whole Model
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P=0.0116
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P<.0001
Leverage Plot
Input Current(4.951,5.249)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000026 -0.000021 -0.000017 -0.000013
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P<.00
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019-0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.8
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019 -0.000018
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.5
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.4
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000018-0.000017
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.3
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000025-0.000021-0.000018-0.000015
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.0263
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
-0.000001
.
.
-0.000007
-0.000001
.
2.3039e-7
0.0000011
-0.000004
-0.000008
-0.000003
-0.000007
Coef
Thi ckness of Pentacene(140.54,259.46)Substrate Temperature(34.122,60.878)Input Current(4.951,5.249) Idsat
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
Variabl e
186.39104
33.056779
4.9301363
Crit ical Val ue
Soluti on i s a SaddlePoint
Predi cted Value at Soluti on-0.000007
Solution
Response Surface
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
Thi ckness of Pentacene
34.122
60.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.12260.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Idsat
0.00002
-7.5e-5
-0.00001
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
Prediction Profiler
Response Idsat
39
that after about 400 Å, the pentacene starts exhibiting „bulk-like‟ characteristics. 200-400 Å
represents the transition. The increase in current in our experiment is mainly due to the stacking
of the monolayers and uniform coverage till about 200 Å or so. In our experiment, the
thicknesses were varied from 100-300 Å . Throughout this range, the stacking helps increase the
mobility and conductivity. The coverage starts getting worse after about 400 Å .
The results in Figure 18 also show that as the substrate temperature increases, the saturation
current increases. As the substrate temperature increases, the grain size increases due to
thermodynamic factors such as sticking coefficient and diffusion of the pentacene grains during
evaporation. As many groups have shown, as grain size increases, the mobility increases. Since
electrical conductivity is limited by charge transfer across grain boundaries the formation of
larger crystallites leads to enhanced conductivities [23]. Mobility increase leads to the current
increase that we see in our experiment.
Increase in input current, which directly relates to the temperature of the crucible holding the
pentacene, interestingly increases the overall saturation current as well. An interplay of multiple
factors is most likely causing this shift.
The interplay among the different factors is shown in the interaction plot shown in Figure 20.
When the lines within an interaction box are not parallel, it indicates that the two inputs
corresponding to the „interaction‟ box are dependent on each other in causing an effect on the
output, which in this case in Idsat. Note that Figure 18 delineates how the inputs are
independently affecting Idsat while Figure 20 presents the dependent relationships. In Figure 20
below, the two factors that are interacting in affecting Idsat are thickness of pentacene and
substrate temperature. These are circled in the figure below.
40
Figure 20. Interaction Profiler Plot for Idsat.
In Figure 18, the values of the inputs were set at midpoint to get the general trend. The
interaction plot reveals that since thickness of pentacene and substrate temperature are
interacting, if the thickness of pentacene for example changed from the midpoint value of 200 Å
to 100 Å , then the way the substrate temperature affects the Idsat will be different (i.e. different
slope). An illustration of this is shown in Figure 21 below. When tuning the output
characteristics, these interactions need to be taken into consideration as well.
-8e-5
-6e-5
-4e-5
-2e-5
0e+0
2e-5
Idsat
Act
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted P<.0001 RSq=0.41
RMSE=1.4e-5
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.409977
0.369748
0.000014
-0.00002
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
1.8854e-8
2.71339e-8
4.59879e-8
Sum of Squares
2.0949e-9
2.056e-10
Mean Square
10.1911
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
1.1973e-8
1.51609e-8
2.71339e-8
Sum of Squares
2.3946e-9
1.194e-10
Mean Square
20.0591
F Ratio
<.0001
Prob > F
0.6703
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
-0.000014
-0.000008
-0.000003
-0.000007
-0.000007
2.3039e-7
0.0000011
-0.000001
-0.000001
-0.000004
Estimate
0.000003
0.000001
0.000001
0.000001
0.000002
0.000002
0.000002
0.000002
0.000002
0.000002
Std Error
-4.05
-5.89
-2.56
-5.24
-4.03
0.14
0.66
-0.74
-0.89
-2.25
t Rati o
<.0001
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
7.12763e-9
1.34661e-9
5.65133e-9
3.33073e-9
3.8216e-12
9.0222e-11
1.1331e-10
1.6346e-10
1.03827e-9
Sum of Squares
34.6742
6.5509
27.4924
16.2032
0.0186
0.4389
0.5512
0.7952
5.0510
F Ratio
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob > F
Effect Tests
-0.00005
-0.00004
-0.00003
-0.00002
-0.00001
0
0.00001
0.00002
0.00003
0.00004
Idsat
Resid
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted
Residual by Predicted Plot
Whole Model
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P=0.0116
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P<.0001
Leverage Plot
Input Current(4.951,5.249)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000026 -0.000021 -0.000017 -0.000013
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P<.00
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019-0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.8
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019 -0.000018
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.5
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.4
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000018-0.000017
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.3
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000025-0.000021-0.000018-0.000015
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.0263
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
-0.000001
.
.
-0.000007
-0.000001
.
2.3039e-7
0.0000011
-0.000004
-0.000008
-0.000003
-0.000007
Coef
Thi ckness of Pentacene(140.54,259.46)Substrate Temperature(34.122,60.878)Input Current(4.951,5.249) Idsat
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
Variabl e
186.39104
33.056779
4.9301363
Crit ical Val ue
Soluti on i s a SaddlePoint
Predi cted Value at Soluti on-0.000007
Solution
Response Surface
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5Id
sat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
Thi ckness of Pentacene
34.122
60.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.12260.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Idsat
0.00002
-7.5e-5
-0.00001
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
Prediction Profiler
Response Idsat
41
Idsa
t
0.00002
-7.5e-5
-7.43e-6
Thi ckness of Pentacene
140
.54
259
.46
141
Substrate Temperature
34.
122
60.
878
47.5
Input Current4
.951
5.2
495.1
-8e-5
-6e-5
-4e-5
-2e-5
0e+0
2e-5
Idsat
Act
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted P<.0001 RSq=0.41
RMSE=1.4e-5
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.409977
0.369748
0.000014
-0.00002
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
1.8854e-8
2.71339e-8
4.59879e-8
Sum of Squares
2.0949e-9
2.056e-10
Mean Square
10.1911
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
1.1973e-8
1.51609e-8
2.71339e-8
Sum of Squares
2.3946e-9
1.194e-10
Mean Square
20.0591
F Ratio
<.0001
Prob > F
0.6703
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
-0.000014
-0.000008
-0.000003
-0.000007
-0.000007
2.3039e-7
0.0000011
-0.000001
-0.000001
-0.000004
Estimate
0.000003
0.000001
0.000001
0.000001
0.000002
0.000002
0.000002
0.000002
0.000002
0.000002
Std Error
-4.05
-5.89
-2.56
-5.24
-4.03
0.14
0.66
-0.74
-0.89
-2.25
t Rati o
<.0001
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
7.12763e-9
1.34661e-9
5.65133e-9
3.33073e-9
3.8216e-12
9.0222e-11
1.1331e-10
1.6346e-10
1.03827e-9
Sum of Squares
34.6742
6.5509
27.4924
16.2032
0.0186
0.4389
0.5512
0.7952
5.0510
F Ratio
<.0001
0.0116
<.0001
<.0001
0.8918
0.5088
0.4591
0.3742
0.0263
Prob > F
Effect Tests
-0.00005
-0.00004
-0.00003
-0.00002
-0.00001
0
0.00001
0.00002
0.00003
0.00004
Idsat
Resid
ua
l
-0.00008-0.00005 -0.00002 0 .00002
Idsat Predicted
Residual by Predicted Plot
Whole Model
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P=0.0116
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P<.0001
Leverage Plot
Input Current(4.951,5.249)&RS
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000026 -0.000021 -0.000017 -0.000013
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P<.00
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019-0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.8
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.00002 -0.000019 -0.000018
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.5
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000019-0.000018
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.4
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000021 -0.00002 -0.000018-0.000017
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.3
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
-0.00008
-0.00006
-0.00004
-0.00002
0
0.00002
Idsat
Le
vera
ge
Resid
ua
ls
-0.000025-0.000021-0.000018-0.000015
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.0263
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
-0.000001
.
.
-0.000007
-0.000001
.
2.3039e-7
0.0000011
-0.000004
-0.000008
-0.000003
-0.000007
Coef
Thi ckness of Pentacene(140.54,259.46)Substrate Temperature(34.122,60.878)Input Current(4.951,5.249) Idsat
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
Variabl e
186.39104
33.056779
4.9301363
Crit ical Val ue
Soluti on i s a SaddlePoint
Predi cted Value at Soluti on-0.000007
Solution
Response Surface
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
-9e-5
-7e-5
-5e-5
-3e-5
-1e-5
1e-5
3e-5
Idsat
Thi ckness of Pentacene
34.122
60.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.12260.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Idsat
0.00002
-7.5e-5
-0.00001
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
Prediction Profiler
Response Idsat
a)
b)
Note that in Figure 21b, the thickness of pentacene has been set at 141 Å (i.e. the middle red line
has been moved). In this case the substrate temperature has the opposite effect on Idsat as
compared to Figure 21a. The reason for the interaction here can be supported by the
explanations that have been already provided on how the thickness of pentacene affects the
stacking and consequently the saturation current.
Figure 22 shows the wide range of electrical characteristics obtained by varying the
evaporation conditions. The saturation current varied from about ~2 A in the lowest
performing sample to ~60 A on the highest performing sample. Figure 22 below shows the
characteristics from two different experiments, representing the lowest and highest performance.
Figure 21. Illustration of the significance of the interaction plot.
42
Figure 22. Id- Vd characteristics from 2 different experiments with the lowest(left) and highest(right) performing samples.
To summarize, the three input factors affect the Idsat in the following way. Increase in
thickness of pentacene causes an increase in the hole current because of the increase in film
stacking and coverage explained using the schematic diagram. The increase in substrate
temperature leads to bigger grain sizes, which consequently causes the enhanced conductivity.
Finally, the input current increase also enhances the performance because of a combination of an
increase in grain size and interplay with the other input factors.
2. Mean Roughness of Pentacene Film Surface
Using a similar system of leverage plots as shown in Figure 17, the relationship between the
input factors and mean roughness has been extracted and shown in Figure 23. Mean roughness
was measured using Atomic Force Microscopy (AFM).
Figure 23. Relationship between the input parameters and mean roughness.
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
Act
ua
l
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
Mean Roughness Predi cted P<.0001
RSq=0.75 RMSE=0.4609
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.746553
0.729272
0.46086
4.243979
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
82.58211
28.03579
110.61790
Sum of Squares
9.17579
0.21239
Mean Square
43.2021
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
9.611127
18.424668
28.035795
Sum of Squares
1.92223
0.14508
Mean Square
13.2498
F Ratio
<.0001
Prob > F
0.8334
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
4.2663565
0.7709168
-0.231355
-0.041969
-0.080681
0.1065694
-0.139403
-0.024359
-0.006368
0.0063131
Estimate
0.108316
0.041569
0.041569
0.042661
0.054313
0.054313
0.054313
0.050573
0.050573
0.051564
Std Error
39.39
18.55
-5.57
-0.98
-1.49
1.96
-2.57
-0.48
-0.13
0.12
t Rati o
<.0001
<.0001
<.0001
0.3270
0.1398
0.0519
0.0114
0.6308
0.9000
0.9027
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
73.047982
6.578877
0.205558
0.468673
0.817707
1.399186
0.049276
0.003367
0.003184
Sum of Squares
343.9294
30.9751
0.9678
2.2066
3.8500
6.5877
0.2320
0.0159
0.0150
F Ratio
<.0001
<.0001
0.3270
0.1398
0.0519
0.0114
0.6308
0.9000
0.9027
Prob > F
Effect Tests
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
Me
an
Ro
ug
hn
ess
Re
sid
ual
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
Mean Roughness Predi cted
Residual by Predicted Plot
Whole Model
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P<.0001
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P=0.3270
Leverage Plot
Input Current(4.951,5.249)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.164.184.204.224.244.264.284.304.32
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.1
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.15 4.20 4.25 4.30 4.35
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.0
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.10 4.15 4.20 4.25 4.30 4.35 4.40
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.21 4.22 4.23 4.24 4.25 4.26 4.27 4.284.29
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.6
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.22 4.23 4.24 4.25 4.26 4.27
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.9
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.21 4.22 4.23 4.24 4.25 4.26 4.27
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.9027
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Me
an
Ro
ug
hn
ess
6.127
2.208
4.266357
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
Thi ckness of Pentacene
34.122
60.878
4.9515.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Prediction Profiler
Response Mean Roughness
Id-Vd
-2.50E-06
-2.00E-06
-1.50E-06
-1.00E-06
-5.00E-07
0.00E+00
5.00E-07
-50.00 -40.00 -30.00 -20.00 -10.00 0.00 ID
ID2
ID3
ID4
ID5
Id-Vd
-7.00E-05
-6.00E-05
-5.00E-05
-4.00E-05
-3.00E-05
-2.00E-05
-1.00E-05
0.00E+00
1.00E-05
-50.00 -40.00 -30.00 -20.00 -10.00 0.00ID
ID2
ID3
ID4
ID5
43
As the thickness of pentacene increases, the film roughness increases because the higher the
number of monolayers, the rougher the film usually tends to be. This is further confirmed when
looking at the schematic diagram in Figure 19. As the thickness of the pentacene film increases,
the stacking becomes misaligned, especially after about 160 Å or so. Despite the increase in
roughness, the current has still gone up with higher thickness possibly because other factors that
affect the current such as better transport within a monolayer dominated.
As the substrate temperature increases, the roughness decreases. When evaporation of
pentacene is done at higher substrate temperatures, the formation of the pentacene crystals on the
surface is more ordered, thereby reducing the overall roughness. The effect of substrate
temperature on the ordering of thin films of pentacene has been analyzed previously [24].
The input current does not independently affect the mean roughness according to Figure 23
above, but it does interact with both substrate temperature and to a lesser extent, thickness of
pentacene as shown in Figure 24 below. Although at a pentacene thickness of 200 Å and a
substrate temperature of 47.5 C , the input current does not affect the mean roughness, at other
values of thicknesses and substrate temperatures, it does. Depending on what thickness and
substrate temperature turns out to be ideal for DNA immobilization and overall sensitivity, the
input current will have to be tuned accordingly.
44
Figure 24. Interaction Profiler Plot for Mean Roughness.
As previously discussed, the topological features on the pentacene film are critical for
DNA immobilization and sensor sensitivity. By doing AFM on these samples, surface
roughness data on the samples were extracted. Figure 25 shows AFM images of samples from 2
different experiments, representing surfaces with the lowest (2.208nm) and highest (6.127nm)
mean roughness. The wide range of the surface roughness data further emphasizes the need and
importance of controlling the input parameters to get optimum morphological conditions for
highest sensitivity.
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
Act
ua
l
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
Mean Roughness Predi cted P<.0001
RSq=0.75 RMSE=0.4609
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.746553
0.729272
0.46086
4.243979
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
82.58211
28.03579
110.61790
Sum of Squares
9.17579
0.21239
Mean Square
43.2021
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
9.611127
18.424668
28.035795
Sum of Squares
1.92223
0.14508
Mean Square
13.2498
F Ratio
<.0001
Prob > F
0.8334
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
4.2663565
0.7709168
-0.231355
-0.041969
-0.080681
0.1065694
-0.139403
-0.024359
-0.006368
0.0063131
Estimate
0.108316
0.041569
0.041569
0.042661
0.054313
0.054313
0.054313
0.050573
0.050573
0.051564
Std Error
39.39
18.55
-5.57
-0.98
-1.49
1.96
-2.57
-0.48
-0.13
0.12
t Rati o
<.0001
<.0001
<.0001
0.3270
0.1398
0.0519
0.0114
0.6308
0.9000
0.9027
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
73.047982
6.578877
0.205558
0.468673
0.817707
1.399186
0.049276
0.003367
0.003184
Sum of Squares
343.9294
30.9751
0.9678
2.2066
3.8500
6.5877
0.2320
0.0159
0.0150
F Ratio
<.0001
<.0001
0.3270
0.1398
0.0519
0.0114
0.6308
0.9000
0.9027
Prob > F
Effect Tests
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
Me
an
Ro
ug
hn
ess
Re
sid
ual
2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
Mean Roughness Predi cted
Residual by Predicted Plot
Whole Model
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P<.0001
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P=0.3270
Leverage Plot
Input Current(4.951,5.249)&RS
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.164.184.204.224.244.264.284.304.32
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.1
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.15 4.20 4.25 4.30 4.35
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.0
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.10 4.15 4.20 4.25 4.30 4.35 4.40
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.21 4.22 4.23 4.24 4.25 4.26 4.27 4.284.29
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.6
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.22 4.23 4.24 4.25 4.26 4.27
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.9
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
Me
an
Ro
ug
hn
ess
Leve
rage
Re
sid
uals
4.21 4.22 4.23 4.24 4.25 4.26 4.27
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.9027
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
-0.024359
.
.
-0.080681
-0.006368
.
0.1065694
-0.139403
0.0063131
0.7709168
-0.231355
-0.041969
Coef
Thi ckness of Pentacene(140.54,259.46)Substrate Temperature(34.122,60.878)Input Current(4.951,5.249)Mean Roughness
Thi ckness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)
Variabl e
416.52362
75.914608
4.509781
Crit ical Val ue
Soluti on i s a SaddlePoint
Crit ical values outsi de data range
Predi cted Value at Soluti on5.5074304
Solution
Response Surface
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
Me
an
Ro
ug
hn
ess
Thi ckness of Pentacene
34.122
60.878
4.9515.249
150 200 250
140.54
259.46
Substrate Temperature
4.9515.249
40 50 60
140.54
259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Response Mean Roughness
45
a)
b)
Figure 25. AFM images of samples from 2 different experiments with the lowest (a) and highest (b) surface roughness parameters. The lowest extracted roughness value was 2.208nm while the highest was 6.127nm.
3. Pentacene Grain Size
Grain boundaries are important topological features where DNA can immobilize.
Therefore, grain size is an important parameter to characterize. Atomic Force Microscopy
(AFM) was used to measure grain size. After extracting AFM images from tapping mode scans,
the grain size was measured by boxing the area around an average-size grain in a particular scan.
46
Keeping the systematic variation in consideration, this measurement technique is valid to make a
general comparison across the different experiments. Figure 26 below, a derivation from the
leverage plots discussed earlier, shows how the input parameters affect the grain size.
Figure 26. Relationship between the input parameters and grain size.
To begin with, the thickness of pentacene, as expected, does not affect the grain size. This is true
in Figure 26 above as well as in the interactive profile shown below in Figure 27. As substrate
temperature increases, the grain size also increases (which in turn leads to better electrical
performance). The increase in grain size can be attributed to Oswaldt Ripening effects [25].
Oswaldt Ripening is an observed phenomenon in solid (or liquid) solutions which describes the
evolution of an inhomogeneous structure over time. When a phase precipitates out of a solid,
energetic factors will cause large precipitates to grow, drawing material from the smaller
precipitates, which shrink. With higher substrate temperature, this thermodynamically-driven
spontaneous process occurs because larger particles are more energetically favored than smaller
particles. This stems from the fact that molecules on the surface of a particle are energetically
less stable than the ones already well ordered and packed in the interior. Large particles, with
their lower surface to volume ratio, results in a lower energy state (and have a lower surface
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize A
ctu
al
0 1 2 3 4 5 6 7 8 9
Grain Size Predicted P<.0001 RSq=0.61
RMSE=1.0864
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.608112
0.581392
1.086427
2.150019
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
241.76662
155.80280
397.56942
Sum of Squares
26.8630
1.1803
Mean Square
22.7590
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
116.38324
39.41955
155.80280
Sum of Squares
23.2766
0.3104
Mean Square
74.9916
F Ratio
<.0001
Prob > F
0.9008
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
1.0581313
-0.170463
0.8965621
-0.355127
0.0342315
0.157945
0.4270159
-0.030844
0.2693819
1.0824579
Estimate
0.255342
0.097995
0.097995
0.100568
0.128037
0.128037
0.128037
0.119221
0.119221
0.121556
Std Error
4.14
-1.74
9.15
-3.53
0.27
1.23
3.34
-0.26
2.26
8.90
t Rati o
<.0001
0.0843
<.0001
0.0006
0.7896
0.2195
0.0011
0.7963
0.0255
<.0001
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
3.571505
98.799308
14.717925
0.084369
1.796158
13.128664
0.079002
6.026082
93.598258
Sum of Squares
3.0259
83.7052
12.4694
0.0715
1.5217
11.1229
0.0669
5.1054
79.2988
F Ratio
0.0843
<.0001
0.0006
0.7896
0.2195
0.0011
0.7963
0.0255
<.0001
Prob > F
Effect Tests
-2
-1
0
1
2
3
4
Gra
in S
ize R
esi
du
al
0 1 2 3 4 5 6 7 8 9
Grain Size Predicted
Residual by Predicted Plot
Whole Model
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P=0.0843
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
-1
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P<.0001
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P=0.0006
Leverage Plot
Input Current(4.951,5.249)&RS
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.12 2.13 2.14 2.15 2.16 2.17 2.18
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.7
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.00 2.05 2.10 2.15 2.20 2.25 2.30
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.2
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.112.122.132.142.152.162.172.182.19
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.7
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
.5 1.0 1.5 2.0 2.5 3.0 3.5
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P<.0001
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Gra
in S
ize
8.514
0.007
1.051445
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.4
Input Current
4.9
51
5.2
49
5.1
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
Thi ckness of Pentacene
34.122
60.878
4.9515.249
150 200 250
140.54259.46
Substrate Temperature
4.9515.249
40 50 60
140.54259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Prediction Profiler
Response Grain Size
47
energy). As the system tries to lower its overall energy, molecules on the surface of a small
(energetically unfavorable) particle will tend to diffuse through solution and add to the surface of
larger particle. Therefore, the smaller particles continue to shrink, while larger particles continue
to grow [27], leading to the larger grain size with higher substrate temperatures.
The input current shows an interesting non-linear relationship with grain size. The expected
relationship here would be that as input current decreases, the grain size would increase linearly
because of the way the grains are formed. When the pentacene crystals hit the substrate as they
are being evaporated, they grow till they hit another crystal around them. With lower
evaporation rate, the crystals have „time‟ to grow bigger on the surface. This linear relationship
is not seen because of two possible reasons. One is that there are more factors/variables
interacting that we are not aware of and/or we will see in the interaction plot below. The other
reason could be the way the grain size was measured. Since the grain size was measured as an
average measurement from sample to sample in AFM, a more precise way of measuring the grain
size may need to be implemented to get a clearer relationship between input current and grain
size.
Figure 27 addresses the interaction among the different parameters. The main interaction
happens between substrate temperature and input current, which might help better explain the
non-linear relationship being seen between input current and grain size.
48
Figure 27. Interaction Profiler Plot for Grain Size.
Again, since grain size and grain boundaries are an important part of sensitivity of our sensor,
the range of our grain size from 8.514 m2
down to 0.314 m2, illustrated in Figure 28,
emphasizes the importance of tuning the morphology to suit the necessary conditions for DNA
interaction. This will be discussed in depth in the following sections.
Figure 28. AFM images of samples from 2 different experiments consisting of the smallest grain size (left) of 0.314
m 2 and the biggest grain size (right) of 8.514 m
2.
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize A
ctu
al
0 1 2 3 4 5 6 7 8 9
Grain Size Predicted P<.0001 RSq=0.61
RMSE=1.0864
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.608112
0.581392
1.086427
2.150019
142
Summary of Fit
Model
Error
C. Total
Source
9
132
141
DF
241.76662
155.80280
397.56942
Sum of Squares
26.8630
1.1803
Mean Square
22.7590
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
127
132
DF
116.38324
39.41955
155.80280
Sum of Squares
23.2766
0.3104
Mean Square
74.9916
F Ratio
<.0001
Prob > F
0.9008
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
1.0581313
-0.170463
0.8965621
-0.355127
0.0342315
0.157945
0.4270159
-0.030844
0.2693819
1.0824579
Estimate
0.255342
0.097995
0.097995
0.100568
0.128037
0.128037
0.128037
0.119221
0.119221
0.121556
Std Error
4.14
-1.74
9.15
-3.53
0.27
1.23
3.34
-0.26
2.26
8.90
t Rati o
<.0001
0.0843
<.0001
0.0006
0.7896
0.2195
0.0011
0.7963
0.0255
<.0001
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
3.571505
98.799308
14.717925
0.084369
1.796158
13.128664
0.079002
6.026082
93.598258
Sum of Squares
3.0259
83.7052
12.4694
0.0715
1.5217
11.1229
0.0669
5.1054
79.2988
F Ratio
0.0843
<.0001
0.0006
0.7896
0.2195
0.0011
0.7963
0.0255
<.0001
Prob > F
Effect Tests
-2
-1
0
1
2
3
4
Gra
in S
ize R
esi
du
al
0 1 2 3 4 5 6 7 8 9
Grain Size Predicted
Residual by Predicted Plot
Whole Model
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P=0.0843
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
-1
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P<.0001
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
-2.0 -1.5 -1.0 -0.5 .0 .5 1.0 1.5 2.0
Input Current(4.951,5.249)&RS Leverage,
P=0.0006
Leverage Plot
Input Current(4.951,5.249)&RS
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.12 2.13 2.14 2.15 2.16 2.17 2.18
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.7
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.00 2.05 2.10 2.15 2.20 2.25 2.30
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.2
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5 2.6
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
2.112.122.132.142.152.162.172.182.19
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.7
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize L
eve
rag
e R
esi
du
als
.5 1.0 1.5 2.0 2.5 3.0 3.5
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P<.0001
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Gra
in S
ize
8.514
0.007
1.051445
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.4
Input Current
4.9
51
5.2
49
5.1
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
0
1
2
3
4
5
6
7
8
9
Gra
in S
ize
Thi ckness of Pentacene
34.122
60.878
4.9515.249
150 200 250
140.54259.46
Substrate Temperature
4.9515.249
40 50 60
140.54259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Prediction Profiler
Response Grain Size
5 m 5 m
5 m 5 m
49
4 & 5. Evaporation Rate and Coverage Area
The last two output parameters that were analyzed were evaporation rate and coverage area.
Evaporation rate was calculated using the data from the crystal monitor within the pentacene
evaporator. The coverage area was calculated from the images scanned using AFM. According
to the relationships seen in Figure 29 below, evaporation rate is affected only by the input
current, and the coverage area does not show any strong relationships with the input factors.
Figure 29. Relationship between the input parameters and evaporation rate and coverage area.
The higher the input current, the higher the temperature of the crucible holding the pentacene
for evaporation. This in turn will cause a higher evaporation rate. For coverage area, none of the
input factors show a statistically significant effect since the slopes are all smaller than the error
bars. This is most likely due to some noise from the data extracted. One way to improve this
0
5
10
15
20
Eva
p R
ate
Actu
al
0 5 10 15 20
Evap Rate Predicted P=0.0836 RSq=0.83
RMSE=2.9697
Actual by Predicted Plot
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
5.0650119
1.4729952
0.5291968
3.4397451
1.219993
2.067977
0.8805256
-0.239633
-0.472243
0.5363697
Estimate
2.093768
0.803606
0.803606
0.803606
1.049961
1.049961
1.049961
0.975699
0.975699
0.975699
Std Error
2.42
1.83
0.66
4.28
1.16
1.97
0.84
-0.25
-0.48
0.55
t Rati o
0.0519
0.1165
0.5346
0.0052
0.2894
0.0964
0.4338
0.8142
0.6455
0.6024
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
29.63148
3.82459
161.58580
11.90706
34.21223
6.20260
0.53198
2.06602
2.66522
Sum of Squares
3.3598
0.4337
18.3217
1.3501
3.8792
0.7033
0.0603
0.2343
0.3022
F Ratio
0.1165
0.5346
0.0052
0.2894
0.0964
0.4338
0.8142
0.6455
0.6024
Prob > F
Effect Tests
-4
-3
-2
-1
0
1
2
3
4
Eva
p R
ate
Re
sid
ua
l
0 5 10 15 20
Evap Rate Predicted
Residual by Predicted Plot
Whole Model
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P=0.1165
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P=0.5346
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Input Current(4.951,5.249)&RS Leverage,
P=0.0052
Leverage Plot
Input Current(4.951,5.249)&RS
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
3.5 4.0 4.5 5.0 5.5 6.0
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.2
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
3 4 5 6 7
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.0
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
4.0 4.5 5.0 5.5 6.0
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.4
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
4.6 4.7 4.8 4.9 5.0 5.1 5.2 5.3
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.8
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
4.3 4.5 4.7 4.9 5.1 5.3 5.5 5.7
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.6
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
0
5
10
15
20
Eva
p R
ate
Le
vera
ge R
esid
ua
ls
4.0 4.2 4.4 4.6 4.8 5.0 5.2 5.4 5.6 5.8
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.6024
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Response Evap Rate
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1
Cove
rage
Are
a A
ctu
al
.91 .92 .93 .94 .95 .96 .97 .98 .991.00
Coverage Area Predi cted P=0.7170
RSq=0.50 RMSE=0.0251
Actual by Predicted Plot
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
0.9881251
-0.00136
-0.007235
-0.001655
0.0056331
-0.015849
0.007974
-0.003956
-0.001823
-0.004648
Estimate
0.017694
0.006791
0.006791
0.006791
0.008873
0.008873
0.008873
0.008245
0.008245
0.008245
Std Error
55.85
-0.20
-1.07
-0.24
0.63
-1.79
0.90
-0.48
-0.22
-0.56
t Rati o
<.0001
0.8479
0.3277
0.8156
0.5489
0.1243
0.4034
0.6483
0.8323
0.5934
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
0.00002524
0.00071484
0.00003740
0.00025386
0.00200961
0.00050867
0.00014501
0.00003080
0.00020016
Sum of Squares
0.0401
1.1350
0.0594
0.4031
3.1908
0.8077
0.2302
0.0489
0.3178
F Ratio
0.8479
0.3277
0.8156
0.5489
0.1243
0.4034
0.6483
0.8323
0.5934
Prob > F
Effect Tests
-0.03
-0.02
-0.01
0.00
0.01
0.02
Cove
rage
Are
a R
esid
ua
l
.91 .92 .93 .94 .95 .96 .97 .98 .991.00
Coverage Area Predi cted
Residual by Predicted Plot
Whole Model
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P=0.8479
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P=0.3277
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Input Current(4.951,5.249)&RS Leverage,
P=0.8156
Leverage Plot
Input Current(4.951,5.249)&RS
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.973 .975 .977 .979 .981 .983 .985
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P=0.5
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.965 .970 .975 .980 .985 .990 .995
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.1
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.970.972.974.976.978.980.982.984.986
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.4
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.974 .976 .978 .980 .982 .984 .986
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.6
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.977 .978 .979 .980 .981 .982
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.8
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
0.91
0.92
0.93
0.94
0.95
0.96
0.97
0.98
0.99
1.00
1.01
Cove
rage
Are
a L
evera
ge
Resid
ua
ls
.973 .975 .977 .979 .981 .983 .985 .987
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.5934
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Response Coverage Area
Eva
p R
ate
18.3905
-2.0682
5.065012
Cove
rage
Are
a
1.03241
0.91491
0.988125
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
0
5
10
15
20
Eva
p R
ate
0
5
10
15
20
Eva
p R
ate
0
5
10
15
20
Eva
p R
ate
Thi ckness of Pentacene
34.12260.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.12260.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
0.91
0.94
0.96
0.99
Cove
rage
Are
a
0.91
0.94
0.96
0.99
Cove
rage
Are
a
0.91
0.94
0.96
0.99
Cove
rage
Are
a
Thi ckness of Pentacene
34.12260.878
4.951
5.249
150 200 250
140.54259.46
Substrate Temperature
4.9515.249
40 50 60
140.54
259.46
34.12260.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Prediction Profiler
Least Squares Fit
50
noise would be to use a more precise and statistically accurate tool to measure coverage area.
The purpose of using a contrast tool to obtain coverage area data was to get a general idea of the
trend across all of the experiments in the DOE.
Similar to the figures shown previously, Figure 30 shows the lowest and highest coverage
area from samples of 2 different experiments. The evaporation rate, depending on the input
current, ranged from 0.838 Å /min to 18.391 Å /min.
The left image with the lowest coverage area exposes a lot of the dielectric surface. The stacking
of the image on the right is better, and coverage area is much higher since the dielectric surface is
completely covered by the pentacene crystals.
Sections 1-5 presented thus far have highlighted the importance of the relationship of the
different input parameters and their effect on the morphological and electrical characteristics of
pentacene TFTs. The following section takes this a step further and integrates the DNA
immobilization experiment into the data obtained so far, to arrive at the optimum conditions for
highest sensitivity.
5 m
5 m
5 m
5 m
Figure 30. AFM Images of samples from 2 different experiments that resulted in the lowest (left) and highest (right) coverage area. The yellow area in the images represents the pentacene crystals, and the dark red area is the SiO2 (dielectric) surface.
51
B. Optimization of DNA Immobilization and Sensor Sensitivity
The purpose of this part of the experiment is to determine pentacene evaporation conditions
that facilitate optimum DNA immobilization and highest sensitivity.
i. Experiment Protocol
The experiment protocol that was followed is enlisted below. For each of the experiments in the
DOE, the following steps are followed:
1) Pre-measurements of electrical characteristics of transistors on substrate (153 transistors from
all of the 17 experiments) using the probe station (same set up as described on page 37)
2) Pipetting of 1.5 L of DNA (125 Base-Pairs single strand DNA sequence synthesized
by BioSynthesis, Inc) on channels of pre-measured transistors; the concentration of DNA
used in this experiment was 0.5 g/ L
3) Immobilization of DNA on pentacene surface (in air) until buffer solution is dry
4) Washing of sample for 1-2 minutes with DI Water
5) Drying of substrate (gently)
6) Storage of transistors for a few hours in nitrogen (to neutralized any buffer solution and DI
water effects)
7) Post-Measurements of electrical characteristics
8) Calculation of sensitivity of particular substrate using Ion ratio and threshold voltage shifts
The storage of the sample in nitrogen (step 6) before the post-measurement of electrical
characteristics is important in order to neutralize the effects that buffer and DI water have on the
pentacene surface. This was concluded based on control experiments performed using different
52
protocols on the similar substrates. The control experiment performed and the results obtained
are discussed in the following section.
ii. Control Experiments
Since pentacene is moisture sensitive, control experiments had to be performed to make sure
that the effects of buffer solution and water on the pentacene transistors do not confound the
DNA immobilization results. Dr. Qintao Zhang had previous shown results from control
experiments he had performed using buffer solution and DI water on pentacene. The set of
control experiments done this time around was to figure out what step in the process actually
neutralized the effects. A few protocols were followed, and the one presented in the previous
section produced the desired results. The results from the control experiments performed using
this protocol is shown in Figure 31 below. Idsat ratio of 1 means that the electrical characteristics
have not been affected.
Idsat
ratio
s
0.5
1
1.5
2
2.5
3
3.5
4
buffer di water ssDNA
Type of Experiment Figure 31. Control Experiments performed to neutralize buffer solution and DI water effects on pentacene TFTs.
53
The storage in nitrogen neutralized the buffer solution and DI water effects. Further experiments
need to be performed to understand why the storage in nitrogen helps in this process so that the
storage time and atmosphere can be optimized.
iii. Results: Optimization of DNA Immobilization and Sensor Sensitivity
This section presents the results from integrating the pentacene characterization experiment
with DNA. To review, the Idsat ratio is defined as follows:
Idsat Ratio = onmobilizatiBeforeDNAI
onmobilizatiAfterDNAntDrainCurreI
d
d
Im
Im)(
As with previous experiments, JMP leverages the 153 points that were measured before and
after DNA immobilization (for Idsat Ratio), to produce the graphs and relationships in Figure 32
below.
Figure 32. Relationship between input parameters and Idsat ratios.
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios A
ctua
l
1 2 3 4 5 6
DNA-Idsat_Rati os Predi cted P<.0001
RSq=0.56 RMSE=0.6127
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.563176
0.533617
0.612651
2.547461
143
Summary of Fit
Model
Error
C. Total
Source
9
133
142
DF
64.35996
49.92033
114.28029
Sum of Squares
7.15111
0.37534
Mean Square
19.0523
F Ratio
<.0001
Prob > F
Analysis of Variance
Lack Of Fit
Pure Error
Total Error
Source
5
128
133
DF
13.668493
36.251841
49.920335
Sum of Squares
2.73370
0.28322
Mean Square
9.6523
F Ratio
<.0001
Prob > F
0.6828
Max RSq
Lack Of Fit
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Term
2.2922838
-0.382604
0.4274611
-0.252851
-0.360626
-0.066864
-0.1492
0.0255407
0.2273335
0.0428153
Estimate
0.143982
0.055503
0.055503
0.055503
0.072741
0.072741
0.072741
0.067126
0.067126
0.067126
Std Error
15.92
-6.89
7.70
-4.56
-4.96
-0.92
-2.05
0.38
3.39
0.64
t Rati o
<.0001
<.0001
<.0001
<.0001
<.0001
0.3597
0.0422
0.7042
0.0009
0.5247
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Source
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
DF
17.835760
22.263095
7.789717
9.225263
0.317140
1.579068
0.054338
4.304920
0.152699
Sum of Squares
47.5188
59.3143
20.7537
24.5784
0.8449
4.2070
0.1448
11.4694
0.4068
F Ratio
<.0001
<.0001
<.0001
<.0001
0.3597
0.0422
0.7042
0.0009
0.5247
Prob > F
Effect Tests
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
DN
A-I
dsa
t_R
atios R
esi
du
al
1 2 3 4 5 6
DNA-Idsat_Rati os Predi cted
Residual by Predicted Plot
Whole Model
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Thi ckness of
Pentacene(140.54,259.46)&RS Leverage,
P<.0001
Leverage Plot
Thickness of Pentacene(140.54,259.46)&RS
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Substrate
Temperature(34.122,60.878)&RS
Leverage, P<.0001
Leverage Plot
Substrate Temperature(34.122,60.878)&RS
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
-1.5 -1.0 -0.5 .0 .5 1.0 1.5
Input Current(4.951,5.249)&RS Leverage,
P<.0001
Leverage Plot
Input Current(4.951,5.249)&RS
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0
Thi ckness of
Pentacene(140.54,259.46)*Substrate
Temperature(34. Leverage, P<.00
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.462.482.502.522.542.562.582.602.62
Thi ckness of
Pentacene(140.54,259.46)*Input
Current(4.951,5.249 Leverage, P=0.3
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.40 2.45 2.50 2.55 2.60 2.65 2.70
Substrate
Temperature(34.122,60.878)*Input
Current(4.951,5.249) Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.502.512.522.532.542.552.562.572.58
Thi ckness of
Pentacene(140.54,259.46)*Thi ckness of
Pentacene(14 Leverage, P=0.7
Leverage Plot
Thickness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.2 2.3 2.4 2.5 2.6 2.7 2.8
Substrate
Temperature(34.122,60.878)*Substrate
Temperature(34.1 Leverage, P=0.0
Leverage Plot
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios L
eve
rage
Re
sid
uals
2.48 2.50 2.52 2.54 2.56 2.58 2.60
Input Current(4.951,5.249)*Input
Current(4.951,5.249) Leverage, P=0.5247
Leverage Plot
Input Current(4.951,5.249)*Input Current(4.951,5.249)
DN
A-I
dsa
t_R
atios
5.90977
1.04594
2.292284
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
Thi ckness of Pentacene
34.12260.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thick
ne
ss o
f Pe
nta
cene
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Prediction Profiler
Response DNA-Idsat_Ratios
54
The input parameters have already been related to the morphological and electrical
characteristics. In this section, we relate the input parameters to sensor sensitivity. To begin
with, as the thickness of pentacene increases, the Idsat ratio or sensitivity goes down. Thinner
films allow the DNA to be immobilized in the channel part of the film and affect the
performance more. The channel is formed in the first few monolayers (monolayer = 15 Å ) of the
pentacene film. Although thicker pentacene might immobilize more DNA because of higher
surface roughness, the sensitivity will not be high because not all of the DNA will be
immobilized in the channel part of the pentacene film to actually cause an effect.
As substrate temperature increases and input current decreases, the sensitivity increases.
Higher substrate temperature and lower input current (lower deposition rate) lead to bigger grains
on the surface. At thin film phases, the pentacene film growth produces a terrace like structure
shown in an AFM image in Figure 33.
Figure 33. AFM Image of terrace-like formation of pentacene thin film on SiO2 substrate [27].
Because of the terrace formation, the bigger the grain size, the more exposure there is to the first
few monolayers of the film. In other words, bigger size allows for a wider area of exposure of
55
the first few monolayers of the film, where the channel forms. This is in contrast to smaller grain
size, where even with the terrace formation, the channel part of the film is not widely exposed.
Therefore, the bigger the grain size, the more area (in the channel part of the film) the DNA has
to immobilize. This agrees very well with the data seen in Figure 32. With lower input current
and higher substrate temperature, bigger grains are grown, allowing the DNA to more effectively
immobilize and affect the channel part of the film, thereby causing the highest electrical shift and
giving the highest sensitivity. The sensitivity values range from 1.0456 up to 5.909. Figure 34
takes this analysis a step further and shows how the different input parameters interact together to
affect the sensitivity.
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios A
ctu
al
1 2 3 4 5 6
DNA-Idsat_Rati os Predi cted P<.0001
RSq=0.59 RMSE=0.6013
Actual by Predicted Plot
RSquare
RSquare Adj
Root Mean Square Error
Mean of Response
Observati ons (or Sum Wgts)
0.587339
0.548652
0.601328
2.553529
141
Summary of Fit
Model
Error
C. Total
Source
12
128
140
DF
65.87610
46.28418
112.16029
Sum of Squares
5.48968
0.36160
Mean Square
15.1818
F Ratio
<.0001
Prob > F
Analysis of Variance
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Grain Size
Mean Roughness
Idsat
Term
2.5584276
-0.306932
0.4108072
-0.24085
-0.335803
-0.066425
-0.170855
0.0224374
0.2202644
0.0623213
0.0196847
-0.053072
4211.2101
Esti mate
0.577503
0.108987
0.071038
0.066476
0.076951
0.073853
0.074986
0.066335
0.067605
0.089048
0.053302
0.126044
3702.706
Std Error
4.43
-2.82
5.78
-3.62
-4.36
-0.90
-2.28
0.34
3.26
0.70
0.37
-0.42
1.14
t Rati o
<.0001
0.0056
<.0001
0.0004
<.0001
0.3701
0.0244
0.7357
0.0014
0.4853
0.7125
0.6744
0.2575
Prob>|t|
Parameter Estimates
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Grain Size
Mean Roughness
Idsat
Source
1
1
1
1
1
1
1
1
1
1
1
1
Nparm
1
1
1
1
1
1
1
1
1
1
1
1
DF
2.867860
12.092395
4.746658
6.885906
0.292512
1.877241
0.041370
3.838481
0.177112
0.049317
0.064106
0.467733
Sum of Squares
7.9311
33.4418
13.1270
19.0431
0.8089
5.1916
0.1144
10.6154
0.4898
0.1364
0.1773
1.2935
F Ratio
0.0056
<.0001
0.0004
<.0001
0.3701
0.0244
0.7357
0.0014
0.4853
0.7125
0.6744
0.2575
Prob > F
Effect Tests
Conti nuous factors centered by mean, scaled by range/2
Intercept
Thi ckness of Pentacene(140.54,259.46)&RS
Substrate Temperature(34.122,60.878)&RS
Input Current(4.951,5.249)&RS
Thi ckness of Pentacene(140.54,259.46)*Substrate Temperature(34.122,60.878)
Thi ckness of Pentacene(140.54,259.46)*Input Current(4.951,5.249)
Substrate Temperature(34.122,60.878)*Input Current(4.951,5.249)
Thi ckness of Pentacene(140.54,259.46)*Thickness of Pentacene(140.54,259.46)
Substrate Temperature(34.122,60.878)*Substrate Temperature(34.122,60.878)
Input Current(4.951,5.249)*Input Current(4.951,5.249)
Grain Size
Mean Roughness
Idsat
Term
2.2949018
-0.306932
0.4108072
-0.24085
-0.335803
-0.066425
-0.170855
0.0224374
0.2202644
0.0623213
0.0807066
-0.103994
0.1998661
Scaled Esti mate
0.153313
0.108987
0.071038
0.066476
0.076951
0.073853
0.074986
0.066335
0.067605
0.089048
0.218536
0.246984
0.175732
Std Error
14.97
-2.82
5.78
-3.62
-4.36
-0.90
-2.28
0.34
3.26
0.70
0.37
-0.42
1.14
t Rati o
<.0001
0.0056
<.0001
0.0004
<.0001
0.3701
0.0244
0.7357
0.0014
0.4853
0.7125
0.6744
0.2575
Prob>|t|
Scaled Estimates
DN
A-I
dsa
t_R
atios
5.90977
1.04594
2.294902
Thi ckness of Pentacene
140
.54
259
.46
200
Substrate Temperature
34.1
22
60.8
78
47.5
Input Current
4.9
51
5.2
49
5.1
Grain Size
0.3
14
044
8.5
14
2.15762
Mean Roughness
2.2
08
6.1
27
4.24957
Idsat-0.0
000
7
0.0
00
02
-1.9e-5
Prediction Profiler
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
1
2
3
4
5
6
DN
A-I
dsa
t_R
atios
Thi ckness of Pentacene
34.12260.878
4.951
5.249
150 200 250
140.54
259.46
Substrate Temperature
4.951
5.249
40 50 60
140.54
259.46
34.122
60.878
Input Current
5 5.1 5.2 5.3
Thic
kne
ss o
f Pe
nta
ce
ne
Sub
stra
te T
em
pe
ratu
reIn
put C
urre
nt
Interaction Profiles
Response DNA-Idsat_Ratios
Figure 34. Interaction Profiler Plot for Idsat Ratio/Sensitivity.
56
We have studied impact of evaporation parameters on pentacene morphology, and tuned/
analyzed morphology to maximize the sensitivity of our OTFT platform. With thinner pentacene
films, higher substrate temperature, and lower input current, we are able to achieve highest
sensitivity. Results suggest that sensitivity of DNA sensor can be optimized through morphology
of the film to expose as much of channel part of the film as possible for effective DNA
immobilization.
57
V. Conclusion and Future Work
By finding reliable methods to relate the electrical transduction behavior to physical origins,
and by optimizing the pentacene film surface for highest DNA sensitivity, we have gotten closer
to our goal of making a viable DNA sensor using pentacene TFTs.
Section IV.B. discussed the experimental protocol used for the DNA optimization
experiment. The storage in nitrogen to neutralize effects of buffer solution and DI water was
followed as a result of the performed control experiments. In order to optimize the time and
atmosphere of storage in the protocol, further work needs to be done to understand the nitrogen
interaction with buffer solution and DI water and its effects on the pentacene surface. Also, to
make the DNA immobilization more stable and predictable, future work should focus on
chemically modifying the pentacene surface to covalently bind to the DNA. This way, the
orientation of the DNA is clear, and future experiments with hybridization will be more
controllable. Since the goal is to be able to use this DNA detection chip for genetic disease
detection, more work on hybridization detection needs to be performed. Particularly, more
research needs to be performed into detecting single point mutation, known as „single nucleotide
polymorphism‟ with a 20-30 base pair sequence as done in medical practice today.
58
VI. Sources
[1] Knowledge Systems Institute, Image of Biological Structure, Knowledge Systems Institute.
[Online]. Available: http://distancelearning.ksi.edu/demo/bio378/DNA_files/image023.jpg .
[Accessed: December 2, 2008].
[2] S. Molesa, “Ultra-Low-Cost printed electronics,” Ph.D. dissertation, University of California-
Berkeley, Berkeley, CA, USA, 2006.
[3] National Human Genome Research Institute, “DNA Microchip Technology,” National
Human Genome Research Institute. [Online]. Available: http://www.genome.gov/10000205.
[Accessed: November 23, 2008].
[4] M. Gabig-Ciminska, “Developing nucleic acid-based electrical detection systems,” Microbial
Cell Factories., Mar. 2006.
[5] Q. Zhang, “OTFT-Based DNA Detection System,” Ph.D. dissertation, University of
California-Berkeley, Berkeley, CA, USA, 2007.
[6] D. Gillespie, “A quantitative assay for DNA-RNA hybrids with DNA immobilized on a
membrane,” Journal of molecular biology., vol. 12, pp.829, 1965.
[7] Molecular Station, Molecular Biology Images, Molecular Station. [Online].
Available:http://www.molecularstation.com/molecular-biology-images/data/502/557px-
DNA_chemical_structure.png. [Accessed: November 23, 2008].
[8] Penn State Department of Chemistry, Image of Pentacene Structure, Penn State Elberly
College of Science. [Online]. Available: http://www.chem.psu.edu/images/
Pentacene_JPEG.jpg. [Accessed: December 7, 2008].
[9] “Atomic force microscope,” [Online]. Available: http://en.wikipedia.org/wiki/Atomic_force_
microscope. [Accessed: December 7, 2008].
[10] C. Bai, Scanning Tunneling Microscopy and Its Application. Springer, 1995.
[11] M. J. Heller, “DNA Microarray Technology: Devices, Systems, and Applications,” Annu. Rev.
Biomed. Eng., vol. 4,, pp.129, 2002.
[12] “Kelvin probe force microscope,” [Online]. Available: http://en.wikipedia.org/wiki/Kelvin_
probe_force_microscope. [Accessed: December 5, 2008].
[13] Science Education Resource Center at Carleton College, Image of Fluorescent Microscopy System, Science Education Resource Center at Carleton College. [Online}. Available: http://serc.carleton.edu/images/microbelife/research_methods/microscopy/fluorescent_filters.jpg. [Accessed: December 7, 2008].
[14] “Fluorescence”, [Online]. Available http://en.wikipedia.org/wiki/Fluorescence. [Accessed: November 23, 2008]
[15] Wikipedia, ”What is the wavelength of blue light?” Wiki Answers. [Online]. Available: http://wiki.answers.com/Q/What_is_the_wavelength_of_blue_light. [Accessed: December 8, 2008].
59
[16] Arizona State University, “Single Molecule Spectroscopy,” Arizona State University. [Online].
Available: http://www.public.asu.edu/~laserweb/woodbury/smf.html. [Accessed: December 11, 2008].
[17] Evans Analytical Group, “Time-of-Flight Secondary Ion Mass Spectrometry (TOF-SIMS),” Analytical Techniques. [Online]. Available: http://www.cea.com/techniques/analytical_ techniques/tof_sims.php. [Accessed: December 2, 2008].
[18] R. Plasun, “Optimization of VSLI Semiconductor Devices,” Ph.D. dissertation, Vienna University of Technology, Vienna, Austria, 1999.
[19] JMP Statistical Discovery Software, JMP 5.0.1a Software Manual, 1989.
[20] S. E. Fritz Vos, “Structure and Transport in Organic Semiconductor Thin Films,” Ph.D.
dissertation, University of Minnesota, USA, May 2006.
[21] G. Horowitz and M .E. Hajlaoui, “Mobility in polycrystalline oligothiophene field-effect
transistors dependent on grain size,” Advanced Materials, vol.12, no.14, pp. 1046-1050,
2000.
[22] D. Knipp, D.K. Murti, B. Krusor, R. Apte, L. Jiang, J.P. Lu, B.S. Ong, and R. Street,
“Photoconductivity of Pentacene Thin Film Transistors,” Materials Research Society
Symposium Proceedings, Vol. 665, pp. 207-212, 2002.
[23] A. J. Salih, S. P. Lau, J. M. Marshall, J. M. Maud, W. R. Bowen, N. Hilal, R. W. Lovitt, and
P. M. Williams, “Improved thin films of pentacene via pulsed laser deposition at elevated
substrate temperatures,” Appl. Phys. Letters, vol. 69, pp. 2231, 1996.
[24] T. Minakata, H. Imai, and M. Ozaki, “Electrical properties of highly ordered and
amorphous thin films of pentacene doped with iodine,” Journal of Applied Physics, vol. 72,
pp. 4178, 1992.
[25] J.W. Chang, H. Kim, J.K. Kim, B. K. Ju, J. Jang, and Y.H. Lee, “Structure and Morphology
of Vacuum-Evaporated Pentacene as a Function of the Substrate Temperature,” Journal of
the Korean Physical Society, Vol. 42, pp. S647-S651, February 2003.
[26] Wikipedia, “Ostwald ripening,” Wikipedia, [Online]. Available: http://en.wikipedia.org/wiki /Ostwald_ripening. [Accessed: December 8, 2008].
[27] I. P. M. Bouchoms, W. A. Schoonveld, J. Vrijmoeth, and T. M. Klapwijk, “Morphology
identification of the thin film phases of vacuum evaporated pentacene on SiO2 substrates,”
Synthetic Metals, Vol. 104, Issue 3, pp.175-178, July 1999.
[28] K. Puntambekar, “Characterization of Structural and Electrostatic Complexity in Pentacene
Thin Films By Scanning Probe Microscopy,” Ph.D. dissertation, University of Minnesota,
USA, October 2005.
[29] K.O Lee and T.T. Gan, “Space-Charge-Limited Currents in Evaporated Films of
Pentacene,” Physics Stat. Solutions, vol. 43, p. 565, 1977.
60
APPENDIX MASS SPECTRA RESULTS FROM TOF-SIMS Analysis
Figure 35. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-200
61
Figure 36. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio: 100-300
62
Figure 37. Mass Spectra of positive ions of interest for samples 1 to 3; Mass to Charge Ratio: 300-700
63
Figure 38. Mass Spectra of negative ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-100
64
Figure 39. Mass Spectra of negative ions of interest for samples 1 to 3; Mass to Charge Ratio: 0-140