五苯環有機薄膜電晶體的氨氣感測器研究

國立交通大學

光電工程研究所

碩士論文

五苯環有機薄膜電晶體的氨氣感測器研究

Development of Highly Sensitive Ammonia Gas

Sensor using Pentacene-Based Organic Thin

Film Transistors

研 究 生: 吳玉玫

指導教授: 冉曉雯 博士

五苯環有機薄膜電晶體的氨氣感測器研究

Development of Highly Sensitive Ammonia Gas Sensor using

Pentacene-Based Organic Thin Film Transistors

 

研 究 生: 吳玉玫 Student: Yu-Mei WU

指導教授: 冉曉雯 Advisor: Hsiao-Wen ZAN

 

國立交通大學

光電工程研究所

碩士論文

 

A Thesis

Submitted to Department of Photonics and Institute of Electro-Optical Engineering College of Electrical Engineering and Computer Science

National Chiao Tung University in partial Fulfillment of the Requirements

for the Degree of Master in Electro-Optical Engineering June 2010 Hsinchu, Taiwan

中華民國九十九年六月 

五苯環有機薄膜電晶體的氨氣感測器研究

研究生：吳玉玫 指導教授：冉曉雯 博士

國立交通大學

光電工程研究所碩士班

摘要

近年來有機薄膜電晶體感測器引起了大量的研究，在醫療的肝病檢測上，非 侵入式的氨氣感測器可藉由檢測病人的呼氣氨濃度，取代傳統的血氨濃度檢測。 為了能夠將有機薄膜電晶體真正應用到醫療的氨氣檢測上，我們必須將其對氨氣 的感測能力提升至0.5~5 ppm。 透過五苯環有機薄膜電晶體在不同氨氣濃度下的電性量測分析，如電流、臨 界電壓、載子遷移率、次臨界電壓等參數，形成多參數的氣體感測器，為了能提 升其對氨氣的感測能力，我們透過UV 光改變其 PMMA 介電層的官能基，使元 件對氨氣的反應提升至0.5 ppm。此外，我們也探討了環境中的水氣分子對元件 的影響，以及元件對其他氣體的反應，如氮氣、酒精、二氧化碳、丙酮、甲烷等， 建立其對氣體選擇性。綜合以上的研究，我們可以確立五苯環有機薄膜電晶體在 氨氣感測的可能性，使其更進一步的應用到醫療檢測上。

Development of Highly Sensitive Ammonia Gas Sensor using

Pentacene-Based Organic Thin Film Transistors

Student: Yu-Mei WU Advisor: Hsiao-Wen ZAN

Institute of Electro-Optical Engineering

Nation Chiao Tung University, Taiwan

Abstract

Non-invasive ammonia sensors are attractive for the diagnoses of a variety of

chronic diseases such as liver cirrhosis. A low cost pentacene-based organic thin film

transistor (OTFT) fabricated by a novel and simple process was demonstrated to be

highly sensitive and specific for ammonia gas. Measurement parameters of OTFT

device characteristics for ammonia detection were investigated. The significant

variations of the turn-on current, intrinsic mobility, subthreshold swing and threshold

voltage (Vth) were observed. The OTFT device detected low concentration (0.5~5

ppm) ammonia gas at room temperature that can distinguish between healthy person

further enhanced with a simple UV irradiation treatment to modify the functional-end

groups of poly(methyl methacrylate) (PMMA) dielectric layer. Possible interferences

for ammonia detection such as humidity effect, recovery phenomenon, and sensing

selectivity among nitrogen, alcohol, carbon dioxide, acetone, methane and ammonia

were also discussed. We concluded that the proposed pentacene-based OTFT is a

致謝

時光飛逝，轉眼間我就要碩士畢業了，回首碩士班的日子，最感謝的是我的 指導教授-冉曉雯老師帶領我進入半導體物理的知識殿堂，實驗的過程中進一步 地教導我研究的態度，不但培養了我解決問題的能力，也激發了我的潛能，此外， 也謝謝老師願意給我機會、鼓勵我去法國攻讀雙聯學位，使我更能獨當一面，能 更勇敢的迎向未知的挑戰。 我也非常的謝謝實驗室的大家，謝謝國錫、士欽、周董、武衛、蔚宗學長們 在我有問題時都能像你們請教；謝謝文馨、芸嘉、顏志宇、方哥、吳權陵學長姐 們教導我機台操作；更感謝淑玲、小寶、繁琦、歐陽你們陪我度過苦悶的研究生 活，有你們一起熬夜作 run 再晚也不會累，即使低潮也能一起互相加油打氣；還 有 michael、小能、達欣、威豪、王建敏、小哈、羅世益、長紘、小辛，和你們 聊天總是很開心；也很謝謝伍佰幫我喚回記憶、蠶北鼻讓我沒有成為孤單老人、 古明哲口試前的玩具直升機；還有 AOSO 實驗室的大家：真的很強的強哥、大帥 哥恩禎、燈一直被打開的撒撒、做事認真的楊哥、可愛的葉翰政和葉寶貝、同鄉 的 shut down 哥、老實的林洪正、總是很嗨的戴銘志學長、榮總的小白鼠們，謝 謝你們讓我在交大的最後三個月有著快樂的回憶。 此外，也非常感謝一起合作的楊裕雄教授實驗室，謝謝楊老師提供設備以及 對於研究方向的指導，也謝謝羅淵仁學長還有榮總醫院的協助。 最後，謝謝我寶貝父母親對我的栽培和疼愛，也謝謝我男朋友對我的鼓勵與 包容。

CONTENTS

Chinese Abstract

i

English Abstract

ii

Acknowledgment

iv

Contents

v

Figure Captions vii

Table Captions ix

Chapter 1: Introduction 1

1.1 Introduction of Organic Thin Film Transistors (OTFTs) 1

1.1.1 Overview of Pentacene-Based Thin Film Transistors 1

1.1.2 Device Structure 2

1.1.3 Organic Semiconductor Material 2

1.1.4 Operating mechanisms of OTFTs 3

1.2 Ammonia Sensor and Their Applications 4

1.2.1 Application of Ammonia Sensors 4

1.2.2 Different Types of Ammonia Sensors 5

1.3 OTFTs Ammonia Gas Sensors 7

1.4 Motivation 8

Figure 10

Chapter 2: Experiment Setup 13

2.1 Fabrication of Organic Thin-Film Transistors 13

2.3 Parameter Extraction 16

2.3.1 Field Effect Mobility 16

2.3.2 Threshold voltage 17

2.3.3 Subthreshold swing 18

Figure 19

Chapter 3: Results and Discussion 20

3.1 Ammonia-sensing phenomenon of standard OTFTs 20

3.1.1 Electrical Properties of Standard OTFTs 20

3.1.2 Gas Diffusion Model 20

3.1.3 Ammonia Concentration Effect 22

3.2 Ammonia-sensing Phenomenon of UV-treated PMMA OTFTs 23

3.2.1 Electrical Properties of UV-treated PMMA OTFTs 23

3.2.2 Sensing phenomenon of UV-treated PMMA OTFTs 25

3.3 Selectivity of Gas Sensing 26

3.4 Phenomenon of Recovery 27

3.5 Influences of Environment Humidity 28

Figure 31

Chapter 4: Conclusion

4 4

References

45

Figure Captions

Figure 1-1. (a) Top contact and (b) Bottom contact structure

Figure 1-2. Pentacene molecular structure

Figure 1-3. Energy scheme of the gold-pentacene interface

Figure 2-1. Structure of OTFTs

Figure 2-2. Photo images of sensing system

Figure 3-1. Transfer characteristics of standard OTFTs

Figure 3-2. I-V curve of standard OTFTs under 1ppm NH3 with different time

Figure 3-3. The parameters variation of OTFTs under 1 ppm NH3 with different time

Figure 3-4. Illustration of scattering effect and traps for gas sensing

Figure 3-5. Illustration of screen effect for gas sensing

Figure 3-6. Energy band of (a) standard OTFTs and (b) Ammonia sensing PMMA OTFTs

Figure 3-7. (a) Threshold voltage shift and (b) mobility variation versus different ammonia concentration measured in 1000 seconds and 2000 seconds

Figure 3-8. Simulation of PMMA and UV-treated PMMA estimated by Gaussian 03 with ab initio calculation.

Figure 3-10. Energy band of (a) standard OTFTs and (b) UV-treated PMMA OTFTs

Figure 3-11. AFM images of pentacene film deposited on (a) PMMA and (b) UV-treated PMMA

Figure 3-12. (a) Threshold voltage shift (Vth) (b) mobility variation (/0) and (c)

drain current variation (I/I0) versus ammonia concentration in different waiting time

of standard OTFTs and UV-treated PMMA OTFTs

Figure 3-13. Threshold voltage shift (Vth) and mobility variation ((0-)/0)

percentage for standard and UV-treated OTFTs when devices were exposed to

different kinds of gas molecules.

Figure 3-14. Mobility variation versus different time of (a) standard OTFTs and (b) UV-treated PMMA OTFTs under different ammonia concentration from 0 second to

3000 seconds

Figure 3-15. Threshold voltage shift versus different time of (a) standard OTFTs and

(b) UV-treated PMMA OTFTs under different ammonia concentration

Figure 3-16. Threshold voltage shift versus different NH3 concentrations in the dry

ambient (RH=0%) and in the wet ambient (RH=50%) of (a) standard OTFTs and (b)

UV-treated PMMA OTFTs

Figure 3-17. (a) Mobility variation and (b) drain current variation versus different NH3 concentrations in the dry ambient (RH=0%) and in the wet ambient (RH=50%)

of standard OTFTs or UV-treated PMMA OTFTs

Table Captions

Table 1. Diseases associated with unusual breath odors

Table 2. Gas concentration with different healthy condition

Chapter 1

Introduction

1.1 Introduction of Organic Thin Film Transistors (OTFTs)

1.1.1 An Overview of Pentacene-Based Thin Film Transistors

Organic thin film transistors (OTFTs) based on conjugated polymers, oligomers,

or other molecules have received much attention for more than a decade. Efforts have

also been made to understand the material properties and device physics [1,2,3,4].

Compared to amorphous silicon device, OTFTs have advantages for the low cost,

flexibility, low-temperature, and large area processing, etc. Despite the extensive

research efforts to discover better materials for the active layer of OTFTs, pentacene

is one of the most widely used small molecules for device applications due to its high

field effect mobility and easy film formation properties [1,5,6,7]. Recently,

pentacene-based thin-film transistors (TFTs) are developed for sensors, display

backplane and radio frequency identification devices [8,9,10]. Nevertheless, there are

still several issues to be considered before realizing the applications of pentacene

TFTs: threshold voltage hysteresis, gate voltage stress resistance and the environment-

or ultraviolet (UV) - induced degradation of the pentacene channel [11,12,13,14]. It

1.1.2 Device Structures

In general, OTFTs are made of three parts: an insulator, a thin film of organic

semiconducting material, and three electrodes. Two of the electrodes, the source and

the drain, are in direct contact with the semiconductor to apply a source-drain voltage

and measure the source-drain current that flows through the organic thin film. The

third electrode, the gate, is isolated from the semiconductor by the insulator to

modulate the magnitude of the source-drain current. Because most organic

semiconductors are fragile materials, the deposition of organic semiconductors on the

insulator is much easier than the converse. Thus, the majority of OTFTs are built with

the bottom-gate architecture, which is divided into two structures: top contact and

bottom contact, in Fig1-1. In the bottom contact structure, contacts are deposited on

the insulator. Thus, the contact resistance in bottom contact is higher than in top

contact.

1.1.3 Organic Semiconductor Material

Organic semiconductors are traditionally classified as polymers or small

molecules. In our study, we chose pentacene as the organic semiconductor material.

show in Fig1-2. It is the p-type material, because holes are easily to transport than

electrons. The principle can be illustrated with Fig1-3, the energy scheme of the

gold-pentacene interface [15]. HOMO presents highest occupied molecular orbital,

and LUMO presents lowest unoccupied molecular orbital. When a positive voltage is

applied to the gate, negative charges are induced in the source. However, the LUMO

level of pentacene is quite far from the Fermi level of gold. It is very hard for the

electrons to inject into the pentacene film. In contrast, when a negative voltage is

applied to the gate, positive charges are induced. Because the HOMO level of

pentacene is close to the Fermi level of gold, holes can easily inject into the

pentacene.

1.1.4 Operation mechanisms of OTFTs

The operation mechanisms of OTFTs are originated from MOSFET. Both OTFTs

and MOSFET have semiconductor layer, insulator layer, and three electrodes. But

traditional MOSFET are operated in the inversion mode, while the OTFTs are

operated in the accumulation mode. Since pentacene is a p-type semiconductor

material, holes are the major transport carriers than electrons. When a negative gate

voltage is applied, the energy band of the pentacene film bends upwards and causes

like a capacitor. Because the conductance of the channel is proportional to the charge,

it is also proportional to the gate voltage. If we apply a source-drain voltage, the

accumulated holes transport from source to drain and result in the source-drain

current.

The operation current of OTFTs can be divided into two regions: linear region

and saturation region. At low source-drain voltage, the current is proportional to both

the gate and drain voltages. When the drain voltage increases to approach the gate

voltage, there is a pinch off of the channel, and the channel current becomes

independent of the drain bias. So, we call it the saturation region.

1.2 Ammonia Sensor and Their Applications

1.2.1 Application of Ammonia Sensors

Ammonia is an importance compound for living system. It can be widely utilized

in many fields, such as chemical industries, fertilizer factories, refrigeration systems,

food processing, fire power plants, and medical diagnosis, etc. For applications above,

the concentration control of ammonia is very important. For example, in chemical

industries, ammonia concentration is related to the quality of fertilizers and frozen

foods. Additionally, ammonia concentration plays a role in medical diagnosis. Tab1 is

diseases. It shows that ammonia is a disease marker for the uremia, liver cirrhosis, and

renal failure. Tab2 is the gas concentration with different healthy condition. The

ammonia concentration is 1~5 ppm for renal failure and 0.5~1 ppm for liver cirrhosis.

The most common method we use for diagnosis is by examining the ammonia

concentration of the blood. Recently, developments of a non-invasive diagnostic

method have received considerable attention. If we can develop a non-invasive,

inexpensive, portable and disposable diagnostic device, patient’s breath can be easily

detect and traced by the gas sensor.

1.2.2 Different Types of Ammonia Sensors

(a) Metal-oxide gas sensors

Some well-known materials for ammonia gas sensing are ZnO [16], iridium

oxide [17], molybdenum oxide [18], polyaniline [19,20,21], polypyrrole [22], Au and

MoO3-modified WO3[23,24], Pt- and SiO2-doped SnO2[25], etc. When the device is

exposed to analytes, the gas removes some of the adsorbed oxygen and modulates the

height of the potential barriers, thus changing the conductivity and creating the sensor

signal. However, various ammonia sensors reported work at high temperature such as

350℃, but it is not convenient to sense at such high temperature. Despite of high

sensors remains limited.

(b) Catalytic ammonia sensors

Catalytic ammonia sensor is based on the catalytic reaction of a metal layer with

ammonia gas. The reaction will cause a change in electrode potential and the charge

carrier concentration which can be quantified by using a field effect device, like a

capacitor or a transistor [26,27]. The detection limit can be 1 ppm.

(c) Conducting polymer gas detectors

Conducting polymer ammonia gas detectors use polymer, like polypyrrole and

polyaniline, to react with ammonia. During the process, ammonia can reversibly

reduce the oxidized form of polymer. Because the reduction of the polymer film

causes a change in the conductivity of material, we can use it to make resistometric or

amperometric ammonia detection [28,29]. However, the irreversible reaction between

ammonia and the polymer causes the sensitivity of the sensor decrease when exposed

to ammonia. The detection limit is about 1 ppm.

(d) Optical gas analyzers

which uses coloration reaction of ammonia with phenol and hypochlorite in aqueous

solutions [30,31]. One drawback is the slow kinetics of the reaction. The detection

limit is about 5 uM of ammonia in water or 90 ppb. The second method is optical

absorption ammonia detection [32]. By using a laser and a spectrograph, we can get a

spectrum of the light influenced by the gas composition. Although the method is very

sensitive and selective for ammonia sensing, the equipment is very expensive and it is

not suitable for miniaturized ammonia sensors.

1.3 OTFTs Ammonia Gas Sensors

The research of OTFT sensors started in the late 1980s, just after the first OTFTs

was proposed. Such sensors offer the advantages of simple process, low fabrication

cost, remarkable response repeatability [33] and selectivity [34,35]. Because many

types of organic molecules exhibit sensing behavior based on their chemical

compositions, sensibility and selectivity can be pursued by choosing ad hoc

chemically or biologically functionalized semiconducting polymer active layers for

the use of OTFTs in compact sensing systems or in bio-chips [33,36,37,38]. In

addition, [36,39] provided evidences that the OFET sensor has better performance

than a similar resistor-type sensor in drift, sensitivity, signal-to-noise ratio, response

multi-parameter sensors [40]. By measuring the parameters of field-induced

conductivity, threshold voltage, and field-effect mobility, we can get a fingerprint of

each gas. It makes the transistors to be more selective than conventional

chemiresistors. Besides, the morphology of active layer plays an important role on the

sensibility of OTFT sensors. The device response increases when the grain size is

reduced [41]. The effect of a channel length comparable to the grains size has also

been observed. In order to lower the limit of detection, nanoscale organic transistors

have been explored [42].

So far, the active layer of OTFT sensors include substituted thiophene polymers,

oligomers, naphthalenes, copper-phthalocyanine, pentacene and others. These devices

were exposed to different analytes, such as alcohols, ketones, thiols, nitriles, ester and

ring compounds. However, it does not have much research of ammonia sensors with

OTFTs. Our studies of pentacene-based OTFTs applied on ammonia sensor were the

first.

1.4 Motivation

In order to develop a non-invasive, inexpensive, portable and disposable

diagnostic device, we use pentacene-based OTFT to act as ammonia sensor. It is very

parameters in different ammonia concentrations. Furthermore, to enhance the

response to ammonia gas, we try to modify the sensing surface to obtain different

functional groups which has strong interaction with ammonia gas. Besides, we need to

discuss the recovery phenomenon to know whether it can be reverse or not. The

(a)

(b)

Figure 1-2. Pentacene molecular structure

Table 1. Diseases associated with unusual breath odors

Table 2. Gas concentration with different healthy condition

William, 2007 H. Pylori infection Carbon dioxide Manolis, 1983 Periodontal disease Hydrogen sulfide Manolis, 1983 Liver Cirrhosis Ethanethiol Manolis, 1983 Liver Cirrhosis Butyric acid Manolis, 1983 Davies et al., 1997 Uremia; Liver Cirrhosis

Renal failure Ammonia

Ebeler et al., 1997 Grote et al., 1997 Diabetes; lung cancer

Acetone References Diseases Breath component as a disease marker William, 2007 H. Pylori infection Carbon dioxide Manolis, 1983 Periodontal disease Hydrogen sulfide Manolis, 1983 Liver Cirrhosis Ethanethiol Manolis, 1983 Liver Cirrhosis Butyric acid Manolis, 1983 Davies et al., 1997 Uremia; Liver Cirrhosis

Renal failure Ammonia

Ebeler et al., 1997 Grote et al., 1997 Diabetes; lung cancer

Acetone

References Diseases

Breath component as a disease marker

Chapter 2

Experiment Setup

2.1 Fabrication of Organic Thin-Film Transistors

We chose highly-doped p-type silicon wafers with 100 nm thick silicon oxide as

the substrate. The p-type silicon was used as gate electrode, and the silicon oxide

layer was used as gate insulator. After cleaned with 5min de-ionized water-5min

acetone-5min de-ionized water, the particles and the impurities on the substrate were

removed in order to avoid gate leakage which may cause instability [43].

Then, soluble poly(methyl methacrylate) (PMMA) was spun on the silicon

oxide layer by the spin coater to improve the electric performance and increase the

grain boundaries [44]. The PMMA was obtained from MicroChem. Corp. with a

molecular weight of 95000 and was dissolved in anisole at 10 wt %. The spin speed

was accelerated from 0 to 1000 rpm during the first 10 seconds and further increased

to 7500 rpm during the following 10 seconds. After kept as 7500 rpm for 40 seconds,

the spin speed was decreased from 7500 to 1000 in the following 10 seconds and

further decreased to 0 rpm during 10 seconds. After the process of spin coating, the

PMMA layer was annealed by hot plate with 90℃ for 30 minutes. The capacity of

were then exposed to UV-light, whose wavelength is 175~285 nm and the output

power is 40 mW, for 60 seconds to change the functional group of PMMA.

Following, the pentacene material obtained from Aldrich with 99.9% purity was

evaporated through a shadow mask on the UV treated and non-UV treated

PMMA/SiO2 dielectric layer as the active layer. The deposition was started at the

pressure around 3 10 torr and the temperature was kept at 20℃. The deposition

rate of 1000-Å-thick pentacene was 0.1 Å/sec at the first 100Å and was smoothly

increased to 0.5 Å/sec at the 200Å to the 1000Å. The deposition temperature, the

deposition pressure and the deposition rate are the important parameters to decide the

ordering quality of the organic film [45].

Finally, we used gold as the source and drain electrodes. Due to its similar work

function with pentacene, we can get better injection from gold to pentacene film.

Before gold deposition, we deposited 50-Å-thick nickel through a shadow mask on

the pentacene film as the adhesion layer. Then, we deposited 1000- Å-thick gold on

the samples. The width (W) and length (L) of the device’s channel were 800 μm and

2.2 Gas Sensing System

In this study, we used OTFTs devices to sense five different kinds of gas as

following: ammonia (NH3), methane (CH4), acetone (CH3COCH3), alcohol (C2H5OH)

and carbon dioxide (CO2). To control the environment in the precise condition, we

measured the devices with a semiconductor analyzer (Keithley 4200-SCS) in a sealed

chamber. The inside volume of the chamber is about 42 L and its configuration image

is shown in Fig2-2. After putting the device into the chamber, we vacuumed it to the

pressure of 10 torr and then purged it with high purity nitrogen (99.99% N2) to the

pressure of 1 atm. This process can avoid the influence of moisture and other gases

we did not want to sense. Thus, the data measured in the N2 ambiance was the

standard data. Then, we injected the gas we wanted into the chamber and measured

the device’s characteristic.

The concentration of NH3 was controlled as 0.5, 1, 3 and 5 ppm (mg/L) by a

mass flow controller (MFC). CH4 and CO2 gases were also injected through a MFC

and their concentrations were controlled as 2 ppm. To get CH3COCH3 and C2H5OH

gases, we used a Tedlar bag and put liquid CH3COCH3 and C2H5OH inside. After

liquid CH3COCH3 and C2H5OH evaporated to saturate pressure, we injected the gases

into the chamber with a syringe which can control the concentration as more than 2

To discuss the influence of moisture during the gas sensing process, we also

provided different relative humidity (RH) environment. By using a N2 flow to pump

water, water vapor can be introduced into the chamber. We controlled the RH value as

0% and 50% with different input time and measured the RH with a hygrometer.

2.3 Parameter Extraction

It is also important to analyze the data with a good method. In this section, I introduce

the method we used to extract the three important parameters of the devices from the

electrical characteristics: field effect mobility, threshold voltage, and subthreshold

swing.

2.3.1 Field Effect Mobility

Field effect mobility is an important parameter because it is directly related to the

orientation of the active layer. When other molecular transports into the bulk of the

active layer, the orientation of the active layer will be influenced and the field effect

mobility will also be changed. Because the characteristics of pentacene-based thin

film transistors are similar to conventional MOSFETs, we used the same method to

D OX t cons V G D m V L WC V I g D             tan ( ) m o x D g w C V L  

, where W is the width, L is the length of the channel, and Cox is the capacitance of

the dielectric layer.

2.3.2 Threshold Voltage

Threshold voltage is also an important parameter because it strongly dependents on

dielectric surface states. When the device was exposed to a gas, which might create

charge sites on the dielectric surface, the threshold voltage shifted. With the drain

current in the linear region:

2

which can be simplify for V_D V_G V_T to

(1)

, we extracted the threshold voltage by finding out the intersection point of the drain

2.3.3 Subthreshold Swing

Subthreshold swing represents how rapidly the device turn on from the off state. It is

related to the interface quality and the defect density of the device. We extract

subthreshold swing with



_log



constant  



D V D G

I

V

S

Figure 2-1. Structure of OTFTs

Chapter 3

Result and Discussion

In this chapter, we first discussed the gas-sensing phenomenon of standard OTFTs.

Then, we introduced the UV-treated PMMA method to improve the sensitivity of

OTFTs. Besides, we also discussed the selectivity and the phenomenon of recovery.

Finally, we increased the relative humidity to find out the environment influence of

the water.

3.1 Ammonia-sensing Phenomenon of Standard OTFTs

3.1.1 Electrical Properties of Standard OTFTs

Before exposing the device on the analytic gas, we measured the electrical

characteristics of the device at constant drain bias, V 3 V, and sweeping gate

bias from 5 V to -40 V. With the Fig3-1, we can extract the threshold voltage as -16.4

V, the mobility as 0.3 cm2_{/Vs, and subthreshold swing as 0.66 V/decade. Those data}

was taken as standard one to be compared with other data under gas-sensing.

3.1.2 Gas Diffusion Model

shown in Fig3-3, the variations of the turn-on current, field-effect mobility, and

threshold voltage shift were decreaed while the subthreshold swing was increased.

The turn-on current variation (Id/Id0), according to Eq(1), is affected by both threshold

voltage shift (Vth) and mobility variation (/0), where Id0 and 0 are the initial drain

current and the initial field-effect mobility.

The reason for the decreasing of mobility and threshold voltage is still not very

clear. A possible reason is that positive ammonia ions (NH4+) or polar ammonia

molecular (NH3) penetrated through the grain boundaries into the bulk of pentacene

layer and created scattering centers or traps [47], as Fig3-4. In addition, the polar

molecules may decrease the rate of charge transport in organic materials by increasing

the energetic disorder through charge–dipole interactions [48]. Because the

concentration of ammonia was fixed, the mobility decreased to saturate very soon.

On the other hand, the threshold voltage was decreased gradually. The reason

may be the hole-traps, which were attributed to the NH3 or NH4+ near dielectric

interface and caused lower concentration of gate-induced mobile carriers [49], as

Fig3-5. Therefore, it needs more negative gate voltage to induce holes in the

p-channel, as Fig3-6 Because it needed more time for the formation of the hole-traps

near the interface between pentacene and dielectric layer, the threshold voltage shift

conformed that the density of interface traps between pentacene and dielectric layer

were increased [46].

3.1.3 Ammonia Concentration Effect

To confirm the gas sensing model, we exposed the devices to different ammonia

concentration from 0.5ppm to 5ppm. Fig3-7 (a)(b) shows the threshold voltage shift

and mobility variation versus different ammonia concentration measured in different

time (1000 seconds and 2000 seconds).It clearly shows that the threshold voltage shift

and mobility variation were increased with the increasing of ammonia concentration,

which caused more NH4+ or NH3 penetrate through the grain boundaries of pentacene

film. We can also find that after waiting for 2000 seconds, the threshold voltage shift

more than waiting for 1000 seconds. However, the mobility variation did not change

much between 2000 seconds and 1000 seconds waiting time. The phenomenon

conformed to the gas sensing model we mentioned above.

For application to non-invasive diagnostic sensor for cirrhotic patients, it is

necessary to monitor ammonia concentration at 0.5 ppm or lower so that the breath

samples between healthy person (breath ammonia level: 0.278 ppm) and a patient

(breath ammonia level: 0.745 ppm) can be distinguished [50]. For the patients with

(dangerous) [51]. However, it was hard to distinguish the variation difference

between 0.5 ppm and 1 ppm ammonia with the standard OTFTs. Thus, we needed to

find another way to improve the sensing ability.

3.2 Ammonia-sensing Phenomenon of UV-treated PMMA

OTFTs

To enhance the sensitivity, we used a UV-light irradiation on PMMA to modify

the dipole moment of the dielectric surface [52]. The PMMA functional end-groups

changed from −COOCH3 to −COOOH, which will result in the negative charge sites

near the PMMA surface. Fig3-8 is the simulation of PMMA and UV-treated PMMA

which were estimated by Gaussian 03 with ab initio calculation. The dipole moment

of standard PMMA and UV-treated PMMA were 1.81~1.91 Debye and 2.42~2.5

Debye, respectively.

3.2.1 Electrical Properties of UV-treated PMMA OTFTs

Fig3-9 is the I-V curve of UV-treated device which was measured at constant

drain bias, V 3 V, and sweeping gate bias from 5 V to -40 V. The threshold

voltage we extracted is -11.1 V, mobility is 0.31 cm2_{/Vs, and subthreshold swing is}

turn on and its subthreshold swing was larger. It may due to the change of functional

end-groups (from –CH3 to –COOOH) produced by UV-treatment on PMMA surface,

which will result in the negative charged-states near PMMA surface. With the

comparing of the energy band of standard OTFTs Fig3-10(a) and UV-treated PMMA

OTFTs Fig3-10(b), the negative charge sites, which produced by UV treatment,

caused a surface potential change that induced a bending of the HOMO level at the

interface and increased the carrier density in the channel [53,54,55,56,57]. Thus, it did

not need so much negative gate voltage to induce holes for turn-on. However, the

mobility of both devices was the same. It indicated that the structure of the pentacene

film was not affected by UV treatment. Fig3-11 (a)(b) are the AFM images of

pentacene film deposited on PMMA and UV-treated PMMA, respectively. We can

find that the grains and roughness were almost the same.

To verify the hypothesis, we calculated the number of interface states Nss with

the equation [59],

. . log( )

[

1]

/

total SS

C

S S

e

N

kT q

q





 

where S.S. is the subthreshold swing, e is the Napierian logarithm, k is the

is the absolute temperature.

For standard OTFTs (subthreshold swing = 0.64 V/decade), the number of interface

states were 1.431012 _cm-2_eV-1_{; for UV-treated OTFTs (subthreshold swing = 0.68}

V/decade), the number of interface states were 1.531012 _cm-2_eV-1_{. It was reasonable}

to observe that UV-treated device exhibited higher interface state density than that of

standard device.

3.2.2 Sensing Phenomenon of UV-treated PMMA OTFT

We measured the UV-treated PMMA devices at the same condition as standard

devices. After extracting the parameters from I-V curve, we found that the threshold

voltage shifted much, the mobility and drain current also decreased much compared

with standard OTFTs. Fig3-12(a)(b)(c) are the threshold voltage shift, mobility

variation and drain current versus ammonia concentration in different waiting time of

standard OTFTs and UV-treated PMMA OTFTs, respectively. We proposed that UV

radiation increased the dipole moment of dielectric surface and attracted ammonia gas

molecules to accumulate near the dielectric surface. Also, the negative charge sites

which caused by UV treatment enhanced the attraction of positive ammonia ions

standard OTFTs. The variation difference between 0.5 ppm and 1 ppm ammonia can

be large enough to be distinguished. Therefore, UV treatment can enhance the

ammonia sensitivity.

3.3 Selectivity of Gas Sensing

To research and design an ammonia gas sensor, it is very important to confirm

the sensing selectivity. Thus, we exposed the devices to five different kinds of gas that

may exist in human’s breath, including ammonia (1 ppm), methane (2 ppm), acetone

(more than 1 ppm), alcohol (more than 1 ppm) and carbon dioxide (1000 ppm). With

Fig3-13, we can find that at a fixed sensing time (2000 sec), the threshold voltage

shift was not distinct with increased time under the condition of methane (CH4),

acetone (CH3COCH3), alcohol (C2H5OH) and carbon dioxide (CO2). Although the

concentration of ammonia was the lowest, the threshold voltage shift was evident

after 1000 seconds. The variation of mobility under ammonia environment was also

the biggest. With the sensing index, threshold voltage shift and mobility variation, we

can confirm that both standard and UV-treated OTFTs exhibited good selectivity

3.4 Phenomenon of Recovery

In order to verify the sensing mechanism, we discuss the phenomenon of

recovery. At 0 second, both the standard and UV-treated devices were exposed to

nitrogen. Then, during 0 to 3000 seconds, the devices were exposed to 0 ppm

(nitrogen), 1 ppm, 3 ppm, 5ppm ammonia gas individually. After that, during 3000 to

4500 seconds, the ambience was purged with nitrogen.

With Fig3-14 (a)(b), we can find that the mobility of both the standard device

and UV-treated PMMA device had recovery. Although the value could not recover to

the beginning, the phenomenon were immediately when the ammonia was purged out.

The recovery of mobility may due to the decreasing of ammonia which may create

scattering centers or traps[47].

However, with Fig3-15 (a)(b), the threshold voltage shift of both devices did not

have evident recovery. Although the slope of threshold voltage shift was gradual when

the ammonia was purged out, the value could not be increased. The reason may be

that the ammonia molecular NH3 or ammonia ions NH4+ had been trapped on the

interface of the dielectric and pentacene layer. Thus, the reaction could not be

3.5 Influences of Environment Humidity

In order to know the influence of humidity to the sensing model, we changed the

measure environment by increasing the relative humidity (RH) in the chamber. In the

wet nitrogen ambient (RH=50%), comparing with dry nitrogen ambient (RH=0%), we

can find that the threshold voltage of standard device had a negative shift, as

Fig3-16(a). However, in the same condition, the UV-treated PMMA device had a

positive threshold voltage shift, Fig3-16(b). It may due to different functional-end

group of dielectric. The polar -COOOH functional-end group of UV-treated PMMA

can interact with water molecules and create acceptor-like traps [60]. Because extra

holes were induced by trapped electrons, the threshold voltage turned positive [58].

Nevertheless, it is hard for less-polar -COCH3 functional-end group of standard

devices to interact with water molecules. When the polar water molecules diffused

into the interface between dielectric and pentacene layer, it may attribute hole-traps

near dielectric interface and caused lower concentration of gate-induced mobile

carriers. Therefore, it needs more negative gate voltage to induce holes in the

p-channel.

With the increasing of ammonia concentration, both the standard device and

UV-treated PMMA device had negative threshold voltage shift, especially in the wet

opposite for the UV-treated PMMA device, we can confirm that something must

happen in the wet ammonia ambient. The chemical reaction between water and

ammonia,

H O

NH

NH

OH

, may be enhanced due to the increased H2O, which resulted in more NH4+ and less

H2O. Thus, the threshold voltage shifted much to the negative side in wet ammonia

ambient.

From Fig3-17(a), we can find that the mobility variations of both standard

devices and UV-treated PMMA device were decreased more in wet ambient than in

dry ambient. The reduction of carrier mobility was because polar water molecules

residing at grain boundaries interact with carriers [61]. Scattering effect or the field

screening effect may be the mechanism to describe interactions between polar water

molecules and carriers [62]. Because UV-treated PMMA devices have polar surface

which can attract more dipole water, the decreased difference between wet and dry

ambient of UV-treated PMMA devices were more than standard devices.

For the UV-treated PMMA devices, since threshold voltage moved to positive

value in wet nitrogen ambient, there should be an increase of the drain current

variation. However, from Fig3-17(b), the drain current variation of UV- treated device

attributed to a reduction of mobility [58]. But, we can see the influence of threshold

voltage in the difference of decreased current between wet and dry ambient, the

Figure 3-2. I-V curve of standard OTFTs under 1ppm NH3 with different time

Figure 3-3. The parameters variation of OTFTs under 1 ppm NH3 with different time

0 -10 -20 -30 -40 10-12 10-11 10-10 10-9 10-8 10-7 10-6 1 ppm NH 3

I (A

)

V (V)

0 s 500 s 1000 s 1500 s 2000 s 0 500 1000 1500 2000 0.90 0.95 1.00 1.05 1.10 0.90 0.95 1.00 -1.0 -0.5 0.0 0.7 0.8 0.9 1.0 S.S. (m V/decade)

time (second)



/



0  Vth(V) Id/Id 0

Figure 3-4. Illustration of scattering effect and traps for gas sensing

(a) (b)

Figure 3-6. Energy band of (a) standard OTFTs and (b) Ammonia sensing PMMA OTFTs

(a) (b)

Figure 3-7. (a) Threshold voltage shift and (b) mobility variation versus different ammonia concentration measured in 1000 seconds and 2000 seconds

0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1 Standard  Vt h (V) NH 3 concentration (ppm) 0 second 1000 seconds 2000 seconds 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25 Standard  /  0 NH 3 concentration (ppm) 0 second 1000 seconds 2000 seconds

Figure 3-8. Simulation of PMMA and UV-treated PMMA estimated by Gaussian 03 with ab initio calculation.

(a) (b)

(a)

(b)

Figure 3-11. AFM images of pentacene film deposited on (a) PMMA and (b) UV-treated PMMA

(a)

(b)

0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1

Standard

 Vth ( V ) NH₃ concentration (ppm) 0 (second) 1000 (second) 2000 (second) 0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1

UV-treated

 V th (V ) NH₃ concentration (ppm) 0 (second) 1000 (second) 2000 (second) 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

Standard

  NH 3 concentration (ppm) 0 (second) 1000 (second) 2000 (second) 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

UV-treated

  NH₃ concentration (ppm) 0 (second) 1000 (second) 2000 (second)

(c)

Figure 3-12. (a) Threshold voltage shift (Vth) (b) mobility variation (/0) and

(c) drain current variation (I/I0) versus ammonia concentration in different waiting

time of standard OTFTs and UV-treated PMMA OTFTs

0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

Standard

Id/I d0 NH3 concentration (ppm) 0 (second) 1000 (second) 2000 (second) 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

UV-treated

Id/Id0 NH3 concentration (ppm) 0 (second) 1000 (second) 2000 (second)

Figure 3-13. Threshold voltage shift (Vth) and mobility variation ((0-)/0)

percentage for standard and UV-treated OTFTs when devices were exposed to different kinds of gas molecules.

(a) (b)

0

1

2

CH₄ CH₃COCH₃ C₂H₅OH CO₂ NH₃

|



V

th

(V

)|

Standard Devices UV-treated Devices N₂

0

5

10

15

20

25

(



-



)

/



(%

)

Standard Devices UV-treated Devices CH₄ CH₃COCH₃ C₂H₅OH CO₂ NH₃ N₂

Figure 3-14. Mobility variation versus different time of (a) standard OTFTs and (b) UV-treated PMMA OTFTs under different ammonia concentration from 0

second to 3000 seconds

(a) (b)

Figure 3-15. Threshold voltage shift versus different time of (a) standard OTFTs and (b) UV-treated PMMA OTFTs under different ammonia concentration

0 1500 3000 4500 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Standard

 /  0 time (second) 0.5 ppm NH3 1 ppm NH3 3 ppm NH3 5 ppm NH3 0 1500 3000 4500 0.4 0.5 0.6 0.7 0.8 0.9 1.0

UV-treated

 /  0 time (second) 0 1500 3000 4500 -5 -4 -3 -2 -1 0

Standard

 Vth ( V ) time (second) 0.5 ppm NH3 1 ppm NH3 3 ppm NH3 5 ppm NH3 0 1500 3000 4500 -5 -4 -3 -2 -1 0

UV-treated

 Vth ( V ) time (second)

(a) (b)

Figure 3-16. Threshold voltage shift versus different NH3 concentrations in the dry

ambient (RH=0%) and in the wet ambient (RH=50%) of (a) standard OTFTs and (b) UV-treated PMMA OTFTs

0 1 2 3 4 5 -7 -6 -5 -4 -3 -2 -1 0 1

Standard

 Vth ( V ) NH₃ concentration (ppm) RH = 0% RH = 50% 0 1 2 3 4 5 -7 -6 -5 -4 -3 -2 -1 0 1  Vth (V) NH 3 concentration (ppm) RH = 0% RH = 50%

UV-treated

(a)

(b)

Figure 3-17. (a) Mobility variation and (b) drain current variation versus different NH3 concentrations in the dry ambient (RH=0%) and in the wet ambient

(RH=50%) of standard OTFTs or UV-treated PMMA OTFTs

0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

Standard

 NH₃ concentration (ppm) RH = 0% RH = 50% 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

Standard

Id /I d0 NH₃ concentration (ppm) RH = 0% RH = 50% 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

UV-treated

  NH 3 concentration (ppm) RH = 0% RH = 50% 0 1 2 3 4 5 0.00 0.25 0.50 0.75 1.00 1.25

UV-treated

Id /Id 0 NH 3 concentration (ppm) RH = 0% RH = 50%

Chapter 4

Conclusion

A pentacene-based OTFT was shown to be highly sensitive for ammonia sensing

from 0.5 to 5 ppm, a critical range for the diagnosis of patients with chronic liver

diseases and renal failure. This demonstrated that OTFT devices, which can be

fabricated by simple and cheap process and exhibited channel length and width as

large as several hundreds of microns, are useful as non-invasive biomedical sensors.

This is on the contrary to inorganic MOSFET devices that require high fabrication

cost and complicated fabrication process to scale down its dimension to the range of

nanometers to increase the gas sensing sensitivity. The sensitivity and selectivity of

OTFTs as gas sensor can be further improved by the modification of the PMMA

dielectric layer, selecting suitable measuring parameters and providing additional

local electric field. Due to the simple fabrication processes of the devices, OTFTs are

Reference

[1] Y. Y. Lin, D. J. Gundlach, S. F. Nelson, and T. N. Jackson, IEEE Trans. Electron

Devices 44, 1325 (1997)

[2] G. Horowitz, M. E. Hajlaoui, and R. Hajlaoui, J. Appl. Phys. 87, 4456 (2000)

[3] D. Knipp, R. A. Street, A. Volkel, and J. Ho, J. Appl. Phys. 93, 347 (2003)

[4] A. Volkel, R. A. Street, and D. Knipp, Phys. Rev. B 66, 195336 (2002)

[5] G. Horowitz, Adv. Mater. (Weinheim, Ger.) 10, 365 (1998)

[6] H. E. Katz and Z. Bao, J. Phys. Chem. B 104, 671 (2000)

[7] C. D. Dimitrakopoulos, S. Purushothanman, J. Kymissis, A. Callegari, and J. M.

Shaw, Science 283, 822 (1999)

[8] I. Manunza, A. Sulis, and A. Bonfiglio, Appl. Phys. Lett. 89. 143502 (2006)

[9] K. Nomoto, N. Hirai, N. Yoneya, N. Kawashima, M. Noda, M. Wada, and J.

Kasahara, IEEE trans. Electron Devices 52, 1519 (2005)

[10] R. Rotzoll, S. Mohapatra, V. Olariu, R. Wenz, M. Grigas, K. Dimmler, O.

Shchekin, and A. Dodabalapur, Appl. Phys. Lett. 88, 123502 (2006)

[11] Y. H. Noh, S. Y. Park, S,-M. Seo, and H. H. Lee, Org. Electron, 7, 271 (2006)

[12] D. K. Hwang, J. H. Park, J. Lee, J.-M. Choi, J. H. Kim, E. Kim, and S. Im, J.

[13] C. R. Kagan, A. Afzali, and T. O. Graham, Appl. Phys. Lett. 86, 193505 (2005)

[14] W. J. Kim, W. H. Koo, S. J. Jo, C. S. Kim, H. K. Baik, D. K. Hwang, K. Lee, J. H.

Kim, and S. Im, Electrochem. Solid-state Lett. 9, G251 (2006)

[15] A. Kahn, N. Koch, W. Gao, J. Polym. Sci. B, Polym. Phys. 41, 2529-2548 (2003)

[16] G. S. Trivikrama Rao, and D. Tarakarama Rao, Sens. Actuators B 55, 166-169

(1999)

[17] A. Karthigeyan, R. P. Gupta, K. Scharnagl, M. Burgmair, S. K. Sharma, I. Eisele,

Sens. Actuators B 85, 145-153 (2002)

[18] D. Mutschall, K. Holzner, E. Obermeier, Sens. Actuators B 36, 320-324 (1996)

[19] K. P. Kakde, D. J. Shirale, H. J. Kharat, P. D. Gaikwad, P. A. Savale, V. K. Gade,

M. D. Shirsat, Proceedings of NSPTS-11, C 17 1-5 (2006)

[20] A. L. Kukla, Y. M. Shirshov, S. A. Piletsky, Sens. Actuators B 37, 135-140 (1996)

[21] V. V. Chabukswar, S. Pethkar, A. A. Athawale, Sens. Actuators B 77, 657-663

(2001)

[22] G. Lahdesmaki, A. Lewenstam, A. Ivaska, Talanta 43, 125-134 (1996)

[23] X. Wang, N. Miura, N. Yamazoe, Sens. Actuators B 66, 74-76 (2000)

[24] C. N. Xu, N. Miura, Y. Ishida, K. Matuda, N. Yamazoe, Sens. Actuators B 65,

163-165 (2000)

[26] F. Winquist, A. Spetz, I. Lundstrom, Anal. Chim. Acta 164, 127-138 (1984)

[27] I. Lundstrom, A. Spetz, F. Winquist, U. Ackelid, H. Sundgren, Sens. Actuators B

1 (1-6),15-20 (1990)

[28] E. Palmqvist, C. Berggren Kriz, K. Svanberg, M. Khayyami, D. Kriz, Biosens.

Bioelectron. 10, 283-287 (1995)

[29] I. Lahdesmaki, A. Lewenstam, A. Ivaska, Talanta 43, 125-134 (1996)

[30] M. P. E. Berthelot, Repertoire Chimique Appliqueel 284 (1859)

[31] P. L. Saerle, Analyst 109, 549-568 (1984)

[32] G. H. Mount, B. Rumberg, J. Havig, B. Lamb, H. Westberg, D. Yonge, K. Johson,

R. Kincaid, Atmos. Environ. 36 (11), 1799-1810 (2002)

[33] B. Crone, A. Dodabalapur, A. Gelperin, L. Torsi, H. E. Katz, A. J. Lovinger, Z.

Bao, Appl. Phys. Lett. 78, 2229-2231 (2001)

[34] M. C. Tanese, L. Torsi, N. Cioffi, D. Colangiuli, G. M. Farinola, F. Babudri, F.

Naso, M. M. Giangregorio, L. A. Zotti, L. Sabbatini, P. G. Zambonin, Sens. Actuators

B 100, 17-21 (2004)

[35] D. Fine, D. Cauble, T. Jung, H. von Seggern, M. Krische, A. Dodabalapur, APS

Meeting (2003)

[36] T. Someya, H. E. Katz, A. Gelperin, A. J. Lovinger, A. Dodabalapur, Appl. Phys.

[37] L. Torsi, M. C. Tanese, N. Cioffi, M. C. Gallazzi, L. Sabbatini, P. G. Zambonin,

G. Raos, S. V. Meille, M. M. Giangregorio, J. Phys. Chem. B 107, 7589-7594 (2003)

[38] Z. –T. Zhu, J. T. Mason, R. Diechkermann, G. Malliaras, Appl. Phys. Lett. 81,

4643-4645 (2002)

[39] T. Someya, T. Sekitani, S. Iba, Y. Kato, H. Kawaguchi, T. Sakurai, Proc. Natl. Sci.

USA 101, 9966-9970 (2004)

[40] L. Torsi, A. Dodabalapur, L. Sabbatini, P. G. Zambonin, Sens. Actuators B 67,

312-316 (2000)

[41] L. Torsi, A. J. Lovinger, B. Crone, T. Someya, A. Dodabalapur, H. E. Katz, A.

Gelperin, J. Phys. Chem. B 106, 12.563-12.568 (2002)

[42] L. Wang, D. Fine, A. Dodabalapur, Appl. Phys. Lett. 85 (26), 6386-6388 (2004)

[43] M. J. Powell, Appl. Phys. Lett. 43, 597-599 (1983)

[44] H. L. Cheng, Y. S. Mai, W. Y. Chou, and L. R. Chang, Appl. Phys. Lett. 90,

171926 (2007)

[45] C. D. Dimitrakopoulos, A. R. Brown, A. Pomp, J. Appl. Phys. 80, 2501 (1996).

[46] L. Takeya, T. Nishikawa, T. Takenobu, S. Kobayashi, Y. Iwaea, T. Mitani, Appl.

Phys. Lett. 85, 5078 (2004)

[47] D. J. Gundlach, T. N. Jackson, D. G. Schlon, and S. F. Nelson, Appl. Phys.

[48] S. V. Novikov, D. H.Dunlap, V. M. Kenkre, P. E. Parris, and A. V. Vannikov, Phys.

Rev. Lett. 81, 4472 (1998)

[49] K. P. Pernstich, S. Haas, D. Oberhoff, C. Goldmann, D. J. Gundlach, B. Batlogg,

A. N. Rashid, and G. Schitter, J. Appl. Phys. 96. 11 (2004)

[50] C. Shimamoto, I. Hirata, and K. Katsu, Hepato-Gastroenteral, 47, 443 (2000)

[51] A. Manolis, Clin. Chem. 29, 5 (1983)

[52] H. W. Zan and K. H. Yen, Electrochem. Solid-state Lett. 118 H222-H225 (2008)

[53] A. Bolognesi, M. Berliocchi, M. Manenti, A. D. Carlo, P. Lugli, K. Lmimouni, C.

Dufour, IEEE Trans. Electron Device 51, 1997 (2004)

[54] S. Scheinert, G. Paasch, M. Schrodner, H. -K. Roth, S. Sensfuss, Th. Doll, J.

Appl. Phys. 92, 330 (2002)

[55] K. P. Pernstich, S. Haas, D. Oberhoff, C. Goldmann, D. J. Gundlach, B. Batlogg,

A. N. Rashid, G. Schitter, J. Appl. Phys. 96, 6431 (2004)

[56] J. Takeya, T. Nishikawa, T. Takenobu, S. Kobayashi, Y. Iwasa, T. Mitani, C.

Goldmann, C. Krellner, B. Batlogg, Appl. Phys. Lett. 85, 5078 (2004)

[57] Y. S. Yang, S. H. Kim, S. C. Lim, J. -I. Lee, J. H. Lee, L. -M. Do, T. Zyung, Appl.

Phys. Lett. 83, 3939 (2003)

[58] D. Li, E. J. Borkent, R. Nortrup, H. Moon, H. Katz, and Z. Bao, Appl. Phys. Lett.

[59] K. N. U. Narayanan, D. S. Sylvie, J. M. Nunzi, J. Phys. D: Appl. Phys. 38, 1148

(2005)

[60] S. H. Kim, H. Yang, S. Y. Yang, K. Hong, D. Choi, C. Yang, D. S. Chung, and C.

E. Park, Organic electronics 9, 673-677 (2008)

[61] D. H. Dunlap, P. E. Parris, and V. M. Kenkre, Phys. Rev. Lett. 77, 542 (1996)