Abstract

Deposition of poly(4-vinyl phenol) insulator layer is carried out by applying the spin coating technique onto p-type GaAs substrate so as to create Al/poly(4-vinyl phenol)/p-GaAs metal-oxide-semiconductor (MOS) structure. Temperature was set to 80–320 K while the current-voltage (I-V) characteristics of the structure were examined in the study. Ideality factor (n) and barrier height () values found in the experiment ranged from 3.13 and 0.616 eV (320 K) to 11.56 and 0.147 eV (80 K). Comparing the thermionic field emission theory and thermionic emission theory, the temperature dependent ideality factor behavior displayed that thermionic field emission theory is more valid than the latter. The calculated tunneling energy was 96 meV.

1. Introduction

Compound semiconductor materials can be used in such applications to achieve better results. Currently, many different compound semiconductors are available, but, among them, gallium arsenide (GaAs) has been the most studied one since the technologies used to process and fabricate this material have been highly developed [14]. Due to its direct energy bandgap, GaAs is ideally preferred for light emission and photovoltaic devices. It also has a wide bandgap (1.5 eV) and high electron mobility (8000 cm2 V−1 s−1), which has specific advantages in operations requiring high frequency and temperature [5].

The interest in and the efforts toward developing new organic-based electronic and optoelectronic devices have recently increased. Charge carrier injection from electrodes into organic material constitutes one important factor to determine the device function and performance for metal-semiconductor (MS) devices [6]. Modification of the electrical properties of (MS) Schottky structures can be achieved by using an organic semiconductor. In this technique, an organic interfacial layer is inserted between inorganic semiconductor and metal. As they can be applied to various areas like solar cells and Schottky diodes, there has been a growing interest in polymers such as poly(4-vinyl phenol), polyaniline, polyvinyl alcohol (PVA), poly(alkylthiophene) polypyrrole, polythiophene, and poly(3-hexylthiophene) [718]. In polymers, we find local free-volume holes or cavities of atomic and molecular dimensions, which may be the result of irregular packing of the molecules in amorphous phase (static and preexisting holes) and molecular relaxation of polymer chains and terminal ends (dynamic and transient holes). The density of the amorphous phase declines due to these holes approximately by 10%, when compared to that of the crystalline phase of the same polymeric material. Thus, such free-volume holes found in a polymeric system influence the polymer’s optic, thermal, dielectric, electrical, and relaxing properties. For this reason, poly(4-vinyl phenol) is much more preferred as an insulating layer since it has several advantages such as higher device performance, stability, and reliability over other kinds of insulator layers. Therefore, the use of poly(4-vinyl phenol) as a gate insulator material in organic field effect transistors (OFETs) has been common [18, 19]. The high performance and reliability of OFETs mainly stem from the insulator layer between metal and semiconductor, as well as from the interface states between semiconductor and insulator and series resistance. Poly(4-vinyl phenol) has been reported to be the best polymeric gate dielectric in terms of mobility [15]. Device performance is significantly affected by gate dielectric, which is an important component of organic thin film transistors (OTFT) [19]. Yet, certain electrical instabilities are found in an OTFT with poly(4-vinyl phenol) used as the gate dielectric, a case which could be exemplified by bias stress effect or hysteresis, leading to shifting threshold voltage depending on the amount of hydroxyl groups and thus to gate leakage current. The reason behind the occurrence of hysteresis is slow polarization as a result of remnant dipoles and charge injection as well as trapping mechanism [20, 21]. Several experimental studies have been carried out to examine the effect of organic materials used as an insulating layer in device applications [22, 23].

In the present study, the current-applied bias voltage-temperature (I-V-T) measurement in darkness was performed to clarify the current transport mechanism(s) and electrical features of Al/poly(4-vinyl phenol)/p-GaAs structures.

2. Design and Fabrication of Al/Poly(4-vinyl phenol)/p-GaAs Structures

A lot of Al/poly(4-vinyl phenol)/p-GaAs structures were fabricated on the 2-inch diameter float zone <111> p-type (zinc-doped) GaAs wafer with a thickness of 500 μm and a resistivity of 2 Ω cm. For the fabrication process, GaAs wafer was degreased through the RCA cleaning procedure. The RCA cleaning procedure has three major steps that are used sequentially: (I) organic cleaning: removal of insoluble organic contaminants with a 10-minute boiling in NH4OH + H2O2 + 6H2O solution, (II) oxide stripping: removal of a thin oxide layer where metallic contaminants might have accumulated as a result of (I), using a diluted (30 s) HF : H2O (1 : 10) solution, and (III) ionic cleaning: the procedure was followed by a 10-minute boiling in HCl + H2O2 + 6H2O solution [24, 25]. Next, drying was performed in N2 atmosphere for a prolonged time. Following the drying process, In-Ag (25%, 75%) with a thickness of 2000 Ǻ was thermally evaporated from the tungsten filament onto the whole back surface of the GaAs wafer under the pressure of 10−6 Torr. In order to obtain a low-resistivity ohmic back contact, GaAs wafer was sintered at 580°C for 3 min in N2 atmosphere. The poly(4-vinyl phenol) was purchased from Sigma-Aldrich (Sigma-Aldrich Cat. number 436216) with a molecular weight of 20.000. Spin coating process was applied to cover poly(4-vinyl phenol) on the front surface of the GaAs wafer from 2.5% solution in 2-propanol at 4.000 rpm for 45 s. The obtained film was cross-linked at 100°C under vacuum. In order to obtain a rectifying contact on the front surface of p-GaAs coated with poly(4-vinyl phenol) (Figure 1), a high-purity aluminum layer was coated on the surface in a high vacuum under the pressure of 10−7 Torr. The interfacial oxide layer thickness was estimated to be about 37 nm by Avantes spectrometer (AvaSpec-ULS2048).

The current-voltage (I-V) characteristics of the samples were measured in the temperature range of 80–320 K using a temperature controlled ARS -DMX-1SS high performance closed cycle cryostat, which allowed us to perform the measurements in the temperature range of 10–325 K, and using a Keithley 4200 programmable constant current source under dark conditions. The sample temperature was continually monitored using a GaAlAs sensor and a Lakeshore 330 autotuning temperature controller with a sensitivity better than ±0.1 K.

3. Measurement and Experimental Results

The current transport across a Schottky junction draws intense interest from both material physicists and device physicist. Usually, a wide range of temperatures are used in determining the Schottky barrier diode (SBD) parameters with the aim of acquiring a better understanding concerning the nature of the barrier and the conduction mechanism. Although, under normal conditions, SBD parameters are obtained through the use of thermionic emission (TE) theory, certain anomalies have been reported at lower temperatures resulting from this theory. The current through a SBD at a forward bias voltage () based on the TE theory is expressed as [26]where is the reverse saturation current and is described aswhere , , , , , and are the rectifier contact area, the effective Richardson constant (74 A/cm2 K2 for p-type GaAs), the temperature in Kelvin, the Boltzmann constant, the electron charge, and the barrier height, respectively.

The ideality values are derived from the slope of the linear region of the forward bias plots and can be calculated from (1) asIn Figure 2, a typical temperature dependence is displayed in semilogarithmic plots of I-V characteristics of Al/poly(4-vinyl phenol)/p-GaAs structure under higher voltages. The I-V plots shift toward the side with higher bias, but a decrease is observed in temperature. As can be seen in Figure 2, the forward bias I-V characteristics lose a significant amount of their linearity because of the series resistance effect of the Al/poly(4-vinyl phenol)/p-GaAs structure in high voltage region. The experimental values of and by using linear parts of the I-V characteristics were calculated through (2) and (3), respectively. Besides, and values of each temperature are displayed in Table 1. As it is obvious in Figure 3, these two parameters have a strong temperature dependency. Figure 3 reveals that values increased while values decreased together with a decreasing temperature. The current transport will be dominated by the current flowing through the lower BH and a larger ideality factor due to the temperature activated process [27]. In other words, a higher number of electrons have sufficient energy to overcome the higher barrier when temperature increases and, in turn, BH is increased by temperature and bias voltage.

The conventional Richardson plot of versus was obtained and is shown in Figure 4. The values of activation energy and Richardson constant were obtained from the slope of this straight line as 0.0178 eV and 2.71 × 10−3 A/cm2 K2, respectively. The Richardson constant value of 2.71 × 10−3 A/cm2 K2 is much lower than the known value of 74 A/cm2 K2 for p-type GaAs. As explained above, the deviation in the Richardson plots might be a result of the spatially inhomogeneous BHs and potential fluctuations at the interface, which consist of low and high barrier areas [2833]; in other words, the current through the barrier will flow preferentially through the lower barriers in the potential distributions.

The high ideality factor value (greater than unity) and its temperature dependence assume that the principally dominant factor of the current is the thermionic field emission (TFE). When the TFE theory suggested by Padovani [34, 35] is used to control the current transport, the relationship emerging between the current and voltage can be expressed with the following equation [26]:withwhere is the characteristic tunneling energy that is related to the tunnel effect transmission probability:where = 1.3 × 1019 cm−3, , and for p-type GaAs and  Js. The value of at room temperature was found to be 5.66. However, the experimental value of at room temperature was determined as 3.71. If , tunneling dominates since the Boltzmann distribution tail of thermionic emission reduces by a factor of every and it is much faster when compared to the decrease rate of the tunneling probability [36]. On the other hand, thermionic emission dominates whenever because the tunneling probability drops faster than thermionic emission in such a situation. In empirical and practical terms, it is observed that the effect caused by tunneling is small at room temperature for common semiconductors having a doping level of 1 × 1017 cm−3 or less ( meV), but it will be pretty significant for semiconductors with a doping level higher than 1 × 1018 cm−3 ( meV). In our study, as the doping level is 1.3 × 1019 cm−3, tunneling energy is found as 96 meV in our calculations as can be seen in Figure 5.

4. Conclusions

Measurements of the forward bias I-V characteristics of the Al/poly(4-vinyl phenol)/p-GaAs metal-oxide-semiconductor (MOS) structure were carried out in a temperature range of 80–320 K. Experimental forward bias I-V characteristics were analyzed and an increase was observed in barrier height as well as a decrease in the ideality factor with an increasing temperature. Comparing the thermionic field emission theory and thermionic emission theory, the temperature dependent ideality factor behavior displays that thermionic field emission theory is more valid than the latter. The calculated tunneling energy was 96 meV.

Competing Interests

The authors declare that there are no competing interests regarding the publication of this paper.