Resistive Switching Properties of Doped SrZrO 3 Films

In A. Beck’s study [16], [19], he declared that the MIM structure device manufactured with 0.2% Cr-doped SZO films as dielectric layer, SrRuO3 (SRO) and Au as bottom and top electrode, respectively has resistive switching behavior. The previous research of our lab also uses (100)-oriented LaNiO3 (LNO) which could enhance the preferred orientation of the SZO films to make 0.2% V-doped SZO films have resistive switching properties [26]. According to Robertsona simulation, Cr, Mn,

Fe, V, Co, Cu transition elements, which could provide over two valences in doped SZO thin films, the density of the defects could be modulated by the doping concentration. In addition, the defects are associated with different valences transformed by the applied voltage. Clearly, slight dopant concentration in SZO thin films could induce resistive switching phenomenon, and the V-doped SZO film had better resistive switching behavior. However, the effect of dopant concentration for V-doped SZO film is still unclear. This is because perovskite oxide consists of more than moments, and it’s hard to find optimal recipe to normal manufacturing process because their crystal structure and stoichiometry are hardly controllable.

From the previous research of our lab, V-doped SZO films deposited on LNO/SiO2/Si substrate have good performance in resistance switching. The resistive switching of Al/V-doped SZO/LNO device can be operated by bias voltage and voltage pulse. However, the drawbacks of Al/V-doped SZO/LNO device are high operation voltage more than 10V.

In this thesis, the physical and electrical properties of the SZO films deposited by sputter method are reported. The SZO film is deposited on the LNO buffer layer, which is also deposited by sputter method. Pt bottom electrode is deposited on Ti/SiO2/Si substrate by electron beam evaporation method. Ti deposited by E-Gun is acted as adhesion layer for Pt. Al is evaporated as top electrode by thermal coater.

The resistive switching phenomenon can be observed in electrode/resistive thin film/buffer layer/electrode (four-layer) structure.

Four-layer structure compared with electrode/resistive thin film/electrode (tri-layer) structure could enhance the forming voltage, the switching voltage, switching speed, and improving the resistance ratio of two current states. Moreover, based on the current-voltage (I-V) curves and resistive switching phenomena, it is proposed the resistive switching mechanism is local property of SZO films. The

conduction mechanisms, reliability, and retention time, are also investigated.

Fig. 1-1 Principle of operation of DRAM read (left) and write (right), for simple 4 by 4 array [5].

Fig. 1-2 A six-transistor CMOS SRAM cell [6].

Fig. 1-3 Typical floating gate memory structure.

Fig. 1-4 Schematic plot of a PCRAM cell. Depending on the state of the active region (crystalline or amorphous) the resistance of the cell changes by several orders

of magnitude.

Fig. 1-5 I-V curve of a PCRAM cell. SET and RESET denote the switching regions, while READ denotes the region of readout.

Fig. 1-6 ABO3 perovskite unit cell [4].

Fig. 1-7 Hysteresis loop of the ferroelectric material [4].

Fig. 1-8 Cross section schematic diagram of the RRAM. The transistor is fabricated in the front and the resistor in the back end [16].

Fig. 1-9 Equivalent circuit of an array for the write operation of a given bit resistor [17].

Fig. 1-10 Describing how leakage current paths make cell resistance misread [24].

Chapter 2

Experiment Details

2.1 Experiment Process Flow

There are two parts of the experimental flow, including sample preparation and device property analyses as shown in Fig. 2-1. At first, we prepare the samples and make sure these samples have switching characteristics. If not, we will modify the experimental parameters and investigate the switching characteristics again until the suitable experiment parameters are obtained.

There are many steps of sample preparation as shown in Fig. 2-2. First, 4 inch boron-doped (100) silicon substrates were cleaned by standard RCA clean, and then a 200-nm-thick SiO² layer was thermally grown on the substrates by wet oxidation process. The SiO2 layer is acted as the isolation layer to prevent the leakage current from the substrate. Ti deposited by electron beam evaporation is acted as adhesion layer for Pt. Then, Pt bottom electrode layer was deposited on the Ti layer also by electron beam evaporation. Synthesis of the LNO powder and the doped SZO powder were prepared previously. The powders were made to the disk-shaped target for sputtering. After that, the 100-nm-thick LNO films were deposited on the Pt bottom electrode layer to form a (100)-orientated buffer layer by radio frequency (RF) magnetron sputter. Then, the LNO buffer layer was treated by rapid temperature annealing (RTA). After that, the SZO films were deposited on the LNO buffer layer also by the RF magnetron sputter. Then, some of the SZO films were treated by rapid temperature annealing (RTA). Finally, 300-nm-thick Al top electrodes were evaporated on the SZO films by a thermal coater. The four-layer structure sample was

accomplished and the probe was detected on Al top electrode and Pt bottom electrode as shown in Fig. 2-3.

The scanning electron microscope system (SEM) and X-ray diffraction system (XRD) were used to obtain the micro-structure and the crystallization of the films, respectively. XRD analyses were helpful to confirm the orientation of the films.Field emission transmission electron microscope (FETEM) was used to analyze the interface between LNO and SZO films. Focus ion beam (FIB) was used to prepare FETEM sample. Energy dispersive X-ray analyzer (EDS) can analyze the components of LNO and SZO. An Agilent 4155C semiconductor parameter analyzer was used to record I-V curves. By the results of electrical measurement, the manufacturing process was modified to improve the performance of the device.

2.2 Radio-Frequency Magnetron Sputter Systems

In this study, a set of RF magnetron sputter system was utilized to deposit LNO and doped SZO films. The illustration of the sputter system is displayed in Fig. 2-4.

The components of the sputter system are indicated as follows.

2.2.1 Vacuum System

It includes a mechanism pump and a diffusion pump. The chamber base pressure was evacuated to 10^-5Torr before deposition process. There are several valves to control the atmosphere and the pressure in the chamber and tubes.

2.2.2 Pressure System

There are two digital gauges in the system to present the chamber pressure in different working condition. One is Granville-Phlips Co.’s product with a

display range from atmosphere to 0.1 mTorr to show the higher chamber pressure for rough vacuum or sputtering condition. The other is an ion gauge with an accurate display of a high vacuum from 10^-3 to 10^-7 Torr. Accordingly, the vacuum situation under sputtering can be precisely controlled and the diffusion pump is working in the safe pressure.

2.2.3 Temperature Controlling System

It contains two thermal couple sensors, a set of four quartz lamps used as heater, and a temperature controller. During the heating process, the lamps just located above the wafer holder could heat the sample directly by radiation in

Therefore, it could find out that the dependence on the mass ratio of the device performance by tuning recipe.

2.2.5 Plasma Controlling System

This system consists of a RF power generator, a network-matching box, and a 3-inch magnetron gun. The RF power generator has only one working

frequency at 13.56MHz, and the network-matching box has minimum reflection power by adjusting the capacitance of the whole circuit. It is able to gain the stable plasma by the controlling system.

2.2.6 Cooling System

There is cooling water which flows in the pipe welded on the chamber and in the magnetron gun. During the sputtering process, the heating lamps and plasma always produce a lot of redundant heat energy in the chamber, so the cooling water is to prevent from mechanical breakdown and maintain the sample uniformity.

2.3 Preparation of Devices

In the experiment, the four-layer structure device was fabricated. The preparation flow of device is shown in Fig. 2-2.

2.3.1 Preparation of Sputter Targets

Because the LNO and SZO thin films are deposited by the sputter system, it needs two kinds of disk-shaped sputter targets, including the LNO and the doped SZO powder targets.

I. Synthesis of the LaNiO³ Powder Target

The LNO and SZO targets are prepared by the conventional solid-state powder-mixing method. There are six steps in the synthesis processes. First, two kinds of the oxide powders, La2O3 and NiO, were mixed by the rule of stoichiometry. It should be especially careful of the equivalent mol because

1 mol of the LNO is composed of 0.5 mol of the La2O3 and 1 mol of the NiO. Second, the mixed powder was put into a jar with anhydrous alcohol and rolling glass balls, and then was mixed adequately by a grinder. Third, the mixture was dried by an 85^oC oven. The fourth step was the sintering step. It was the most critical process, because the sintering temperature and the heating time would affect on the LNO qualities including the resistance and orientation of the LNO sputtered films. The dried mixture was put into a furnace to execute a sequence of sintering, 600^oC for 2 hours and 1300^oC for 10 hours. In the fifth step, the mixed powder was put in the beaker and baked it in the oven at 150^oC for 2 hours. Finally, the mixed powder put in the disk-shaped target was squeezed by a high pressure of 2000 pounds for 60 seconds such that a compact target was produced for sputtering work.

The preparation flow of the LNO target was showed in Fig. 2-5.

II. Synthesis of the SrZrO³ Powder Target

The SZO powder was synthesized from two kinds of oxide powder, SrCO3 and ZrO2. In order to substitute Zr atom, it was considered the suitable ionicradius compared with Zr atom. Considering all the conditions, transition metaloxide V2O5 was added to form the doped SZO powder.

Because V has freakyoxidation number, it could show more effect on the electric properties of our memory thin films. For example, when it is expected to synthesize 0.2 mol% V-dopedSZO powder, it should use 1 mol of SrCO3, 0.997 mol of ZrO2, and 0.00075mol of V2O5. After mixing above elements of the doped SZO powder, the same steps were followed as synthesis of LNO powder. The mixed powder was put into a furnace of a sequence of sintering process. In the last step, a disk-shaped target was

made by a high pressure of 2000 pounds for 1 min.The manufacturing process was showed in Fig. 2-5. In this study, we prepared pure (undoped), 0.1%, 0.2%, 0.3% and 0.4% V-doped SZO powder.

2.3.2 Thin Film Depositions

The LNO bottom electrode and the doped SZO films were deposited by RF magnetron sputter sequentially. To meet our demands for different process recipe, several parameters were controlled to deposit the films based on the plasma theorem and the models of the thin film growth. There were many parameters including the chamber pressure, the RF power, the working temperature, the ambient conditions, and the deposition time. In general, chamber pressure affected the mean free path (MFP) of plasma which is relative to the deposition rate. The lower pressure was choice to create the larger MFP in the chamber, which leads to the higher deposition rate. Moreover, the deposition rate is dependent on the RF power as well. In the experiment, while depositing both the LNO and doped SZO films, the RF power was set 100 or 150W and the chamber pressure at 10 mTorr. In addition, the temperature and the ambient condition could have influence on the density of the defects, the crystallization, the conductivity, the stoichiometry, and the dielectric constant of thin films.

For the accuracy of the atmosphere, it needs the base pressure about 3×10^-5 Torr before sputtering. Next, to maintain the ambient condition, the flow rate of Ar and O2 by MFC was controlled, and the working pressure was kept by the valves among low pressure where the plasma was generated.

2.3.3 Heat Treatment after Thin Film Depositions

There were two purposes for the experiment using the RTA systems. One is in order to get stronger crystallization orientation or better conductivity of the LNO buffer layer. The other is to control the properties of our sample by changing the heating profile of RTA temperature. The RTA model was FE-004A made by JETFIRST.

2.3.4 Deposition of the Top Electrode

Before the Al top electrodes were deposited on the doped SZO films, the sample had been adhered to a metal mask. The metal mask had different hole with three kinds of diameters that are 150, 250, and 350µm. So the different areas are defined for the top electrodes, which are 1.767×10^-4, 4.908×10^-4, and 9.612×10^-4 cm².

Al used as the top electrode was deposition by a thermal evaporation coater (EBX-6D) manufactured by ULVAC. The samples were loaded with metal masks on the spinning holder, which made the deposition rate more uniform.

Then, the rough pump and the turbo pump would work in term in order that the base pressure before deposition reached 5×10^-6 Torr.

2.4 Measurements and Analyses

2.4.1 X-Ray Diffraction (XRD)

Generally, thin films are classified according to its crystallization. There are three types of crystallization, including amorphous, polycrystalline, and crystalline. XRD analysis was used to investigate the crystal structure and orientation of our sample. Furthermore, the crystallization dependence of the

samples could be identified for heat treatment. In the experiment, the thin films and θ is the diffraction angle. In this analysis, X-ray was made with 0.02 degree beam divergence and operation configuration at 30KV, 20mA.

2.4.2 Scanning Electron Microscope (SEM)

Comprehensively, the surface morphology issue is also a quite important character compared with the character of bulk for the thin films. The surface micro-morphology and cross section of our sample could be observed by SEM analysis. Besides, the crystallization of the thin films needed to be investigated directly by XRD analysis. So, SEM analysis is helpful to get enough information to support our illustration. The SEM model is S4700I with high resolution of 15Å made by Hitachi.

2.4.3 Focus Ion Beam (FIB)

FIB is a scientific instrument that resembles a scanning electron microscope.

However, the SEM uses a focused beam of electrons to image the sample in the chamber, whereas a FIB instead uses a focused beam of gallium ions. Gallium is chosen because it is easy to build a gallium liquid metal ion source (LMIS). In a Gallium LMIS, gallium metal is placed in contact with a tungsten needle and heated. Gallium wets the tungsten, and a huge electric field (greater than 10⁸V per centimeter) causes ionization and field emission of the gallium atoms.

Unlike an electron microscope, the FIB is inherently destructive to the specimen. When the high-energy gallium ions strike the sample, they will sputter atoms from the surface. Gallium atoms will also be implanted into the top few nanometers of the surface, and the surface will be made amorphous.

Because of the sputtering capability, the FIB is used as a micro-machining tool, to modify or machine materials at the micro- and nano-scale.

The FIB is commonly used to prepare samples for the transmission electron microscope. The TEM requires very thin samples, typically ~100 nanometers.

The nanometer-scale resolution of the FIB allows the exact thin region to be chosen.

2.4.4 Field Emission Transmission Electron Microscope and Energy Dispersive X-Ray Spectrometer

The state-of-the-art JEOL JEM-2100F field emission transmission electron microscope is equipped with an Oxford INCA Energy TEM 200 EDS (energy dispersive X-ray spectrometer) system, a Gatan GIF Tridiem EELS (electron energy loss spectrometer) system and a Fischione high-angle annular dark field detector. Features of the JEM-2100F include a high-brightness Schottky field emission electron gun producing a probe size of less than 0.2 nm. Ultra-high point-to-point TEM resolution is 0.19 nm; atomic scale resolution of 0.136 nm can be achieved using high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) imaging. The facilities are ideally suited for crystallographic and chemical analyses at a sub-nanometer scale, including high-sensitivity EDS and EELS.

Both EDS and EELS are analytical TEM (ATEM) techniques and can provide elemental composition and distribution information. The Oxford INCA Energy TEM 200 EDS system has the following features: automatic peak ID and labeling; element maps and linescans using SmartMap data acquisition; ability to define a line or grid of points for automatic data acquisition; absorption correction for samples of finite thickness and Sitelock image drift correction.

2.4.5 Auger Electron Microscopy (AEM)

Auger Electron Spectroscopy (Auger spectroscopy or AES) was developed in the late 1960's , deriving its name from the effect first observed by Pierre Auger, a French Physicist, in the mid-1920's. It is a surface specific technique utilizing the emission of low energy electrons in the Auger process and is one of the most commonly employed surface analytical techniques for determining the composition of the surface layers of a sample.

2.4.6 Secondary Ion Mass Spectrometer (SIMS)

SIMS is a technique used to analyze the composition of solid surfaces and thin films by sputtering the surface of the specimen with a focused primary ion beam and collecting and analyzing ejected secondary ions. While only charged secondary ions emitted from the material surface through the sputtering process are used to analyze the chemical composition of the material, these represent a small fraction of the particles emitted from the sample. These secondary ions are measured with a mass spectrometer to determine the elemental, isotopic, or molecular composition of the surface. SIMS is the most sensitive surface analysis technique, being able to detect elements present in the parts per billion range.

2.4.7 Current-Voltage Measurements

The most important part of all is current-voltage measurement. It could understand the electrical properties of the device from current-voltage curve. The electrical measurement system consisted of a probe station, an Agilent 4155C semiconductor parameter analyzer, an Agilent E5250A low leakage switch which are controlled by personal computer with the Agilent VEE software, and GPIB controller.

Our electrical measurements were sorted into five items, static conductivity switching measurement, retention test, stress test, endurance test, and other electrical phenomenon measurement. The aforementioned four items were tested for criteria of our memory device and the last item was executed to understand the fundamental mechanism of our samples.

I. Static Resistive Switching Measurement

The measurement was performed by Agilent 4155C which applied a dc voltage sweeping between two specified voltages to observe the resistive switching of the sample. The measured results could observe the relation of the switching voltage and the H-state or L-state current. Use Agilent 4155C to execute the double voltage sweep function, current-voltage curve was determined with two different-current states associated with the positive applied voltage or the negative one.

II. Retention Test

Retention time is the time of information keeping. The data (1 or 0) is

not able to be distinguished beyond retention time. The current of the sample in the H-state or L-state was measured after fixed period. The retention time of the V-doped SZO film was very long. By applying the higher temperature on the device, the retention test is accelerative.

III. Non-Destructive Readout Test

The sample stressed smaller voltage than the switching voltage was able to stay in the same conductivity state. The smaller positive and the smaller negative sweep voltage were applied on the sample all the time and

The sample stressed smaller voltage than the switching voltage was able to stay in the same conductivity state. The smaller positive and the smaller negative sweep voltage were applied on the sample all the time and