Chapter 1 Introduction
1.1 Brief history of ISFET
In general, the application of sensors contains a widespread range including electrical, magnetic, physical, optical, thermal, and chemical…etc. The pH sensors belong to the category of ‘chemical sensors’. The concept of the chemical sensor was first introduced in 1962 by the late Professor Kiyoyama of Kyushu University. Work in the field of FET chemical sensors began ten years after the discovery of the gas sensor.
Ion-sensitive field-effect transistors (ISFETS), first described by Bergveld in the early 1970s[1], have experienced a strong development. Bergveld describe the details of measurement of ion density with an ISFET-only configuration without a reference electrode. The operational mechanism of the ISFET with ID, as an expression for the drain current in the linear region, for this reason as changes in the drain current are attributed to changes in the electrostatic potential only. In 1971, Professor Matsuo [2-3] conducted research on a high-impedance circuit using an organic microelectrode with an FET which proposed a measurement system employing the reference electrode. In 1978, an ISFET on a silicon island isolated by a P-N junction and insulator was proposed [1]. The discovery of the planar ISFET has been a revolutionary development for researchers previously restricted by the need for an insulating coating on the silicon substrate.
1.2 The evolution of the double layer model to ISFET
Briefly after description the history of ISFET it was recognized that there is a direct relation between the sensitivity of the ISFET and the charging behavior of metal oxides, the metal oxides always are used to be the sensing films of the ISFET.
For a long period, the site-binding model (also called site-dissociation model) developed by Yates [4] was used to describe the ISFET pH sensitivity [5]. The site-binding model, illustrated in Figure 1-1, is indicated that reactions can happen between protons (H+) in the solution and the hydroxyl groups formed at the oxide-solution interface. At chapter 2 will briefly derived the intrinsic buffer capacity,
βS at the site-binding model which is the important factor of the ISFET sensitivity.
In colloid chemistry, there is no consensus about the correct physical interpretation of the experimental observation on metal oxide. However, the most supported model for the ISFET pH sensitivity, a combination of a double layer model with a model describes the adsorption of protons. This is approach will be used to develop a new, more general model for the ISFET sensitivity. This new model can in corporate any combination of a double layer model and a charging mechanism described by surface reaction. The theoretical calculations are verified with some experimental results. At the double layer model there is another important factor to the ISFET sensitivity, the double layer capacitance C , also been briefly derived at S chapter 2.
1.2.1 The Helmholtz theory
There are several models to describe the double layer capacitance. In 1850’s, the double layer structure for the metal-electrolyte interface was first supported by Helmholtz. He took the double layer as a parallel capacitance, illustrated in Figure 1-2.(a), The Helmholtz layer is divided into two plane, one is inner Helmholtz plane
which is dehydrated ions immediately next to a surface and the other is outer Helmholtz plane at the center of a next layer of hydrated. The capacitance between the inner and outer Helmholtz plane is the double layer capacitance Cs which depend on the surface potential which is contributed by the ions of the aqueous solution. The double layer capacitance is not a constant which follow the concentration of electrolyte. In Fig. 1-2(b), we can obtain the relation between the surface potential and the bulk potential is linear. As the concentration of electrolyte is very high, the Helmholtz double layer model is suitable for use, otherwise it can not been applied.
Consequently, the Helmholtz double layer model need to be amended.
1.2.2 The Gouy – Chapman theory
The Helmholtz double layer model only considered the electrostatics force but ignore the thermal motion among the ions, so that it can not explain the value of the double layer capacitance, and the relation between the surface potential and the concentration of electrolyte which depend on each other. In the beginning of twenty century Gouy and Chapman proposed independently the idea layer to interpret the capacitive behaviour of an electrode-electroloyte solution interface.
Gouy and Chapman thought it was impossible that the ions were fixed at the metal-electrolyte interface regularly. The electrostatics force among electrode and ions, besides there is also an molecular thermal motion effect on the ions that makes a part of ions disperse near the bulk solution. As a result, the Gouy and Chapman model [6] was proposed to adjust the Helmholtz double layer model, illustrated in Figure 1-3.
The Gouy and Chapman model has a major drawback. The ions are considered as a point charges that can approach the surface arbitrarily close. This assumption causes unrealistic high concentrations of ions near the surface at high value of surface
potential. An adjustment to solve this problem was first suggested by Stern [7].
1.2.3 The Gouy – Chapman - Stern theory
In 1924, Stern obtain if the concentration of electrolyte was high that the Helmholtz double layer model could match the experiment, but the concentration of electrolyte was not high enough the Gouy and Chapman model was suitable to use.
For this reason, Stern thought the double layer model should combine the Helmholtz double layer model with the Gouy and Chapman model, so that the new model could explain any concentration of electrolyte for the double layer. This new model is called the Gouy-Chapman-Stern model, shown in Figure 1-4 illustrates. The Gouy-Chapman-Stern model involves two parts:
(1) The front part is stern layer which obey the Helmholtz double layer, the thickness d from the metal-electrolyte interface is nearly to the electrolyte ions radius, the great part of ions are included in the Stern layer.
(2) The later part is diffuse layer which contain remnant electrolyte ions, the ions decay according to the Gouy and Chapman model in this layer. Like Gouy, Stern also ignore the dimension of the ions in the diffuse layer. The surface potential to the bulk solution is the sun of the stern layer and the diffuse layer, and the double layer capacitance is series connection by the stern capacitance and the diffuse capacitance.
1.3 Introduction to ISFET
ISFET was proposed by Bergveld more than 30 years ago, more than 600 papers appeared in these 30 years devoted on ISFETs and another 150 on related devices, such as EnzymeFETs (ENFETs), ImmunoFETs (IMFETs)…etc [8]. By the way , the
pH-ISFTE can be not only a chemical sensor but also a physical sensor [9]. The pH-ISFET can measure the flow-velocity,flow-direction and diffusion-coefficient, it is showed that ISFET can be a multi-senor.
The ISFET has advantages over than ion selective electrode (ISE), such as small size, low coat and robustness. Traditional chemical analysis can not be the usual instrument, because of the size and even the coat that can not be the personal chemical analysis. For this reason, some people think the the traditional chemical analysis will be replace with ISFET or related devices of ISFET. In spite of ISFET have proposed more 30 years, there are less product in the market, the main problem is that ISFET has only sensing layer in its gate region which contact the aqueous solution. By contrast, in a CMOS process poly-si electrode in the gate region is required to define the self-aligned source and drain regions for the MOS transistors. This means that specific processes or design structures must be used to fabricated the ISFETs in a CMOS process.
The fabrication of pH-sensitivity ISFET devices in an undefined two-metal commercial CMOS technology is reported by J.Bausells [10]. Even though, when the pH-sensitivity ISFET operate for a long time, the problem “drift” is always be there.
1.4 Motivation of this work
Some people find out the low-drift material of sensing layer or coat a pasaivation layer onto sensing layer can reduce the drift phenomenon. Even if the two ways at process that can reduce the drift phenomenon, but as immerse the different pH value of aqueous solution, the degree of drift is distinct. The drift phenomenon according to any pH aqueous solution can not be the same, as ISFET operate at any pH aqueous solution for a long time, the value of ΔVG is distinct from any different aqueous
solution will make ISFET detect the pH value inaccurate. This phenomenon can not avoid, whether use low-drift sensing layer or pasaivation layer. It becomes another problem, in order to improve this problem we will suppose a method to make the ΔVG
to be a constant or nearly zero as immerse the different pH value of aqueous solution, so it is more easily to compensate the ΔVG.
The p-type ISFET use sol-gel process to prepare the tin oxide (SnO2) sensing membrane, it show that there is a good linear characteristic of drift from pH 1 to pH 9 [11]. I suppose that the n-type ISFET probably has the opposite drift characteristic from different pH aqueous solution to p-type ISFET. According to the dispersive transport model for drift [12], we can discover that the degree of ΔVG is dependent on the total charge in the sensing film, the depletion charge and the inversion charge which are different from the silicon substrate. Therefore, we can integrate the n-type ISFET and p-type to into a complementary ISFET to eliminate the drift. In case the absolute slope value of drift to different aqueous solution for p-type ISFET and n-type ISFET is the same, we can through the common mode circuit to hold the ΔVG to be a constant. Consequently, it is more easily to compensate for the drift through the circuit. We will use ZrO2 for the sensing film to n-type and p-type ISFET which is grown by sputter. In order to compare with the drift to different aqueous solution, the process to the n-type and p-type ISFET almost similar to each other.Detailed experiment steps will be presented in chapter 3, and the results will be discussed in chapter 4.
1.5 References
[1] P. Bergveld, “Development of an ion sensitive solid-state device for neurophysiological measurements”, IEEE Trans.Biomed. Eng.,vol. BME-17, p.70,
1970.
[2] T. Matsuo and K.D. Wise, An integrated field-effect electrode for biopotential recording, IEEE Transactions on Bio-Medical Engineering 21 (1974) 485.
[3] T. Matsuo and M.Esashi, Method of ISFET fabrication, Sensors and Actuators 1 (1982) 77.
[4] D.E. Yates,S. Levine and T.W. Healy, Site-binding model of the electrical double layer at the oxide/water interface. J. Chen. Soc. Faraday Trans. I, 70 (1974).
1807-1818
[5] L.Bousse,N.F. De Rooij and P. Bergveld, IEEE Trans. Electron.. Dev., ED-30 (1983) 1263-1270
[6] R.E.G. van Hal, J.C.T. Eijkel and P. Bergveld, A general model to describe the electrostatic potential at electrolyte oxide interfaces, Adv. Coll. Interf. Sci. 69 (1996) 31-62.
[7] O.Stern, Z. Elektrochem., 30 (1924) 508.
[8] P. Bergveld, “Thirty years of ISFETOLOGY What happened in the past 30 years and what may happen in the next 30 years” Sensors and Actautors B 88 (2003)1-20 [9] Arshak Poghossian, Lars Berndsen , Michael J. Schöning, Chemical sensor as physical sensor: ISFET-based flow-velocity, Sensors and Actuators B 95 (2003) 384–390
[10] J. Bausells, J. Carrabina, A. Errachid, A. Merlos, Ion-sensitive field-effect transistors fabricated in a commercial MOS technology, Sensors and Actuators B 57 (1999) 56–62
[11] Jung Chuan Chou, Yii Fang Wang, Preparation and study on the drift and hysteresis properties of the tin oxide gate ISFET by the sol–gel method, Sensors and Actuators B 86 (2002) 58–62
[12] S. Jamasb, S.D. Collins and R.L. Smith, A physical model for threshold voltage
instability in Si3N4-gate H+-sensitive FET’s (pH ISFET’s), IEEE Trans. Electron Devices 45 (1998) 1239-1245.
Chapter 2
Theory & Drift Mechanism Description
2.1 Introduction
In this chapter, pH, the theories of ISFET which are relevant to metal oxide semiconductor field effect transistor (MOSFET) and drift mechanism will be interpreted in turn. In the first section will describe what pH is briefly. Subsequently, the theory of ISFET include both ISFET concept and the relation between oxide to electrolyte interface. In the final section a physical model for drift developed by Jamasb [1] will be presented. This model can help us to understanding of the drift mechanism which is caused by the hydration effect and the ions transport in the insulater for the instability of ISFET under long-term operating.
2.2 Why is pH important?
Measuring pH is essential in finding the chemical characteristics of a substance.
Thereinafter, these two examples can make us realize how important it is [2].
(1)Both the solubility of many chemicals or biomolecules in solution and the speed or rate of (bio-)chemical reactions are dependent on pH.
(2)The body fluid of living organisms usually has specific pH range. If the pH of the human blood changes by as little as 0.03 pH units or less the functioning of the body will be greatly impaired. The pH values of lakes, rivers, and oceans differ and depend on the kinds of animals and plants living there. All the industries that deal with water: from the drinking water, the food and the drugs to the paper, plastics,
semiconductors, cements, glass or textiles. In a word, pH is to be closely linked to us.
2.3 Definition of pH
The term pH is derived from a combination of p for the word power and H for the symbol of the element hydrogen [3]. In aqueous solution, the following equilibrium exists between the water (H2O), the acid (H+) and the alkali (OH-):
H2O ＜＝＞ H+
+
OH- (2-1) pH is one of the most common chemical and biomedical measurements. The degree of the pH is the solution of ionization which can supply how much hydrogen ions (H+), not the concentration of the solution itself. The definition in pH is expressed aslog H log [ ]
pH = − a + = − γ H+ (2-2)
where aH+ is the hydrogen ion activity, γ is the activity coefficient which equals to 1 when diluted solution, and [H+] is the molar concentration of solvated protons in units of moles per liter. In practice, pH depends on a number of factors, such as the concentration of the added acid and its dissociation constant [4].
2.4 The method for pH detection
Traditionally, the methods for the measurement of pH values include indicator reagent, pH test strips, metal electrode and glass electrode. There are some drawbacks on the other methods, except for glass electrode. Such as, indicator reagent can show different colors at different solvent, but it only exhibit a range of pH not the accuracy value; As pH test strips immersed in the test liquid, they show a particular color
corresponding to the pH of the solution. These are similar to indicator reagent; The hydrogen electrode method is a golden standard for all methods of pH measure. The activity of the hydrogen ions is determined by potentiometric measurement using a standard hydrogen electrode and a reference electrode. In order to ensure a saturated layer of hydrogen adsorbed at the platinum surface, hydrogen gas is continuously bubbled around the platinum electrode. However, this method is not suitable for daily use due to the inconvenience of handling hydrogen gas [2]. Because of some limitations in practical applications of the first three methods, the glass electrode becomes the most widely used method for the pH measurement, and it is considered to be the standard measuring method.
The glass electrode is most widely used for pH measurement due to idea Nernstian response independent of redox interferences, short balancing time of electrical potential, high reproducibility and long lifetime. However, glass electrode has several drawbacks for many industrial applications. Firstly, they are unstable in alkaline or HF solutions or at temperatures higher than 100°C. Also, they exhibit a sluggish response and are difficult to miniaturize. Moreover, they cannot be used in food or in in vivo applications due to their brittle nature [1]. There is an increasing need for alternative pH electrodes [2].
New trends of pH measurements such as optical-fiber-based pH sensor, mass-sensitive pH sensor, metal oxide sensor, conducting polymer pH sensor, nano-constructed cantilever-based pH sensor, ISFET-based pH sensor and pH-imaging sensor. In this study, we will discuss what problems in practical applications of ISFETs and how to improve them.
2.5 The theory of ISFET
Since ISFET was the first reported by Bergveld, research on new material of sensing thin and fabrication process to improve the sensitivity and stability has been continuously proposed [5-7]. At the same time, the mechanism of the pH response of pH ISFET has also been studied extensively [6-12]. Electrochemical measurement of pH utilizes devices that transduce the chemical activity of the hydrogen ion into an electronic signal, such as an electrical potential difference or a change in electrical conductance. The followings are the theoretical foundations which are mostly adopted to characterize the ISFET.
2.5.1 From MOSFET to ISFET
The ISFET is a new approach of electrochemical measurement of pH, which is similar to MOSFET except that the metal/poly gate is replaced by sensing layers, and the sensing layer is immersed in aqueous solution. Because of it can not directly supply on the aqueous, therefore the reference electrode is adopted to connect with sensing layer. The reference electrode not only supply stable voltage but also can connect the circuit with sensing layer to make a loop. It can trace back to the history of the development of ISFET, it is not difficult to find out the similarities between ISFET and MOSFET. In general MOSFET is Metal-Insulator-Semiconductor structure, ISEFT is Electrolyte-Insulator-Semiconductor structure. The most obvious characteristic is the similarity between their structures. For this reason, the best way to comprehend the ISFET is to understand the operating principle of a MOSFET first.
When MOSFET is operated in the so-called ohmic or non-saturated region, the drain current ID is given by:
where COX is the gate insulator capacitance per unit area; μ is the electron mobility in the channel; W/L is the width-to-length ratio of the channel; VGS is gate to source voltage; VDS is drain to source voltage and VT is the threshold voltage. VT can be described by following expression:
F where VFB is the flat-band voltage; QB is the depletion charge in the silicon substrate, andψF is the potential difference between the Fermi level and intrinsic Fermi level.
The degree ofψF is dependent on the doped concentration. VFB can be described by following expression:
QOX is the charge in the oxide and QSS is the surface state density at the oxide-silicon interface. Substitution of Eq. (2-4) in Eq (2-5), the general form of the threshold voltage of a MOSFET can be described by following expression:
F
are no longer to considered on the ISFET. At Figure 2-1 illustrates, it can observe the similarities and differences between MOSFET and ISFET. When immersed in a aqueous solution, it must occur surface potential at the oxide-solution interface. The surface potential must take in into account. Hence the threshold voltage become the following expression: where Eref is the constant potential of the reference electrode; χsol is the surface dipole potential of the solution which also has a constant value. The surface dipole
potential is different from aqueous solution, even though a little variation of surface dipole potential at disparity aqueous solution. The value compare to other term is too small to take as a constant. All terms are constant except Ψ0, it is the kernel of ISFET sensitivity to the electrolyte pH which is controlling the dissociation of the oxide surface. In order to obtain an accuracy pH value, to investigate a high pH sensitivity
potential is different from aqueous solution, even though a little variation of surface dipole potential at disparity aqueous solution. The value compare to other term is too small to take as a constant. All terms are constant except Ψ0, it is the kernel of ISFET sensitivity to the electrolyte pH which is controlling the dissociation of the oxide surface. In order to obtain an accuracy pH value, to investigate a high pH sensitivity