High Bias Power Research Articles

Effect of radiofrequency bias power on transmission spectrum of flat-cutoff sensor in inductively coupled plasma

Real-time monitoring of plasma parameters at the wafer plane is important because it significantly affects the processing results, yield enhancement, and device integrity of plasma processing. Various plasma diagnostic sensors, including those embedded in a chamber wall and on-wafer sensors, such as flat-cutoff sensors, have been developed for plasma measurements. However, to measure the plasma density on the wafer surface in real-time when processing plasma with bias power, such as in the semiconductor etching process, one must analyze the transmission spectrum of the flat-cutoff sensor in an environment with bias power applied. In this study, the transmission-spectrum and measured plasma-density characteristics of an electrode-embedded flat-cutoff sensor are analyzed via electromagnetic simulations and experiments under applied bias power. Our findings indicate that the flat-cutoff sensor accurately measures the plasma density, which is equivalent to the input plasma density under low bias power. Conversely, under high bias power, the plasma density measured by the sensor is lower than the input plasma density. Also, a thick-sheath layer is formed owing to the high bias power, which may complicate the measurement of plasma parameters using the flat-cutoff sensor. Plasma diagnostics using a flat-cutoff sensor in thick-sheath environments can be achieved by optimizing the flat-cutoff sensor structure. Our findings can enhance the analysis of plasma parameters on-wafer surfaces in processing environments with bias power applied.

Silver film etching using halogen gas plasma

Silver (Ag) film etching was studied with a focus on suppressing the surface roughness induced by Cl2 and CF4 plasmas. After Cl2 plasma etching, roughening of the Ag surface was observed. From in situ x-ray photoelectron spectroscopy and atomic force microscopy analyses using a plasma beam system, the Ag surface was roughened with AgCl formation after Cl2 plasma treatment before exposure to air. In capacitively coupled Cl2 plasma, it seemed that many voids were formed on the Ag surface at a high bias power and cathode temperature. This was considered to be characteristic of agglomeration. In contrast, severe surface roughness was not observed after CF4 plasma etching, even at a high bias power and cathode temperature. Secondary ion mass spectrometry analysis showed high chlorine accumulation near the Ag film surface after Cl2 plasma etching. Possible agglomeration of the Ag film during Cl2 plasma etching was volume expansion caused by chlorine accumulation assumed to enhance the compressive stress of the Ag film, and this resulted in increased boundary grooving and, finally, agglomeration. In contrast, fluorine accumulation was unlikely during CF4 plasma etching, resulting in less Ag film stress, which suppressed grain boundary grooving and agglomeration.

Formation of high aspect ratio fused silica nanowalls by fluorine-based deep reactive ion etching

Even though fused silica was considered as a good mold material for nanoimprint lithography due to the properties of UV-transparent and high strength, fabrication of high aspect ratio nanometer-scale mold was challenging. A fabrication process for 120 nm-wide fused silica ‘nano-walls’ with high aspect ratio has been developed by using fluorine-based deep reactive ion etching (DRIE). The optimum conditions of the DRIE process to result in anisotropic vertical sidewalls with high-aspect ratio and good etch rate control were demonstrated as a function of bias power, process pressure, and argon percentage of the gas mixture. One of the vast applications of this process is the fabrication of the mold for UV nanoimprint lithography (NIL). Due to the fact that the resist is cured using UV-light, it is critical that the mold is transparent. A thin aluminum layer is used as a conducting layer for e-beam lithography, then used as an etch mask for fused silica etch. In the presented study, the etch rate increases with higher bias power and lower gas pressure. The existence of Ar makes the vertical surface smoother.

Improvement of Power Characteristics of All-Solid-State Thin-Film Rechargeable Lithium Batteries

1. Introduction Recently, all-solid-state thin-film rechargeable lithium batteries (TFBs) has expected in various applications such as wireless sensor, smart device and other small devices for an auxiliary power supply. The TFBs are fabricated by forming each layers (i.e. cathode, electrolyte, anode and current collector) using physical vapor deposition process1,2). As a solid electrolyte, lithium phosphorus oxynitride (LiPON) thin film has been used commonly due to no grain-boundary, isotropic property, less electronic conductivity and electrochemical stability. Ionic conductivity of LiPON thin film has been reported ~3.3 x 10-6 S/cm at 25 deg.C3), but it is lower compared with other solid electrolyte4). It is necessary to reduce the internal resistance of TFBs for applying to the device as described above. In order to improve this, it has to be thinner LiPON film or change the electrolyte material having higher ionic conductivity. However, there is a limited to the thickness of electrolyte film without short- circuit because the surface of cathode film is rougher with thicker cathode film5). In this study, we investigated the planarization of cathode film for thinner electrolyte film. 2. Experimental Lithium cobalt oxide (LCO) film was prepared by RF and DC hybrid power magnetron sputtering method on platinum film as current collector. Ar was used as sputtering gas, and process pressure was kept at 1.6 Pa. Furthermore, LCO was deposited with applying each bias power to the substrate for investigating the planarization. After deposition, LCO was annealed at 600 deg.C under atmospheric pressure by lamp heating system. Solid electrolyte (LiPON) film was prepared by RF magnetron reactive sputtering method with only N2gas. Lithium film as an anode was prepared by vacuum evaporation method. Acrylic monomer as a sealant was coating on the anode surface, covered with barrier lid, and cured by ultraviolet light for encapsulation. The fabricated TFB cells were investigated by electrical and electrochemical properties at room temperature. 3. Results and discussion Figure 1 shows the scanning electron microscopy (SEM) image of 3-μm-thick LCO film under different bias power condition. As the result of this, it was confirmed that surface of LCO becomes smoother with higher bias power and surface morphology changes from pyramid-like to dome-like by re-sputtering. TFB cell was fabricated with 2-μm-thick LiPON and 2-μm-thick lithium and discharge capacity was also checked. The charge method is CCCVmode (current density: 0.21 mA/cm2, cut-off: >4.2 V and <0.021 mA/cm2) and the discharge method is CC mode (current density: 0.21 mA/cm2, cut-off: <3.0 V). Figure 2 shows the discharge curve of TFB cells. It was confirmed that discharge capacity of TFB cell is higher with increasing bias power. It is thought that film density of LCO becomes high due to ion peening effect. The improvement of thinner LiPON film and the other are on evaluating. Reference 1) J. B. Bates, N. J. Dudney, G. R. Gruzalski, R. A. Zuhr, A. Choudhury, C. F. Luck and J. D. Robertson, J. Power Sources, 43, 103 (1993).2) T. Jimbo, P. Kim and K. Suu, Energy Procedia, 14, 1574 (2012).3) Xiaohua Yu, J. B. Bates, G. E. Jellison, Jr., and F. X. Hart, J. Electrochem. Soc., 144, 524 (1997).4) N. Kamaya, K. Homma, Y. Yamakawa, M. Hirayama, R. Kanno, M. Yonemura, T. Kamiyama, Y. Kato, S. Hama, K. Kawamoto and A. Mitsui, Nature Materials, 10, 682 (2011).5) T. Jimbo, S. Sasaki, A. Suzuki, I. Kimura and K. Suu, #1235, 224th ECS Meeting (2013) Figure 1

Release etching and characterization of MEMS capacitive pressure sensors integrated on a standard 8-metal 130 nm CMOS process

Monolithic post-processing and characterization of CMOS MEMS capacitive absolute pressure sensors co-integrated on an 8-metal BEOL (back-end-of-line) 130nm CMOS device is reported in this paper. An optimized foundry compatible etch process for an IBM CMOS fabricated top triple layered passivation is discussed. A mixture of wet and plasma dry etch process is proposed for both an elliptic and a rectangular structured pressure sensor capacitor. Lateral 125μm stiction free etch from opposite sides was performed successfully for the monolithically integrated diaphragms on the 130nm CMOS platform. Low power inductive coupled plasma using CHF3 gas along with high RF bias power is utilized to increase lateral etch rate compared to vertical etch rate. Mechanical and electrical characterization results indicate a successful etch of the triple layer passivation and the sacrificial oxide. Sensitivities of the fluorosilicate sealed absolute pressure sensors were measured to be 0.07mV/Pa and 0.05mV/Pa for the elliptic and rectangular element, respectively. In addition, the linear capacitive transduction dynamic range was found to be 0.32pF and 0.23pF, respectively, for the elliptic and rectangular element (for 80hPa pressure variation).

Post-processing and performance analysis of BEOL integrated MEMS pressure sensor capacitors in 8-metal 130 nm CMOS

Customized post-processing and characterization of MEMS capacitive pressure sensors in an 8-metal back-end-of-line (BEOL) standard 130nm “non-MEMS” CMOS technology is reported in this paper. An optimized foundry compatible etch process customized for an IBM CMOS fabricated top triple layered passivation is discussed. A mixture of wet and plasma dry etch process is proposed for both an elliptic and a rectangular structured pressure sensor capacitor. Lateral 125μm stiction free etch from opposite sides was performed successfully for the integrated diaphragms. Low power inductive coupled plasma using CHF3 gas along with high RF bias power is utilized to increase lateral etch rate compared to vertical etch rate. Mechanical and electrical characterization indicate a successful etch of the triple layer passivation and the sacrificial silicon dioxide. Sensitivities of the sealed absolute pressure sensors were measured to be 0.07mV/Pa and 0.05mV/Pa for the elliptic and rectangular element respectively. In addition, the linear capacitive transduction dynamic range was found to be 32fF and 23fF respectively for the elliptic and rectangular element (for 80hPa pressure variation).

J161041 (K,Na)NbO_3圧電薄膜の微細加工技術([J16104]マイクロナノメカト口ニクス(4))

It was investigated that the etching characteristics of lead-free (K,Na)Nb03 (KNN) thin films by the ICP-RIE (Inductive Coupled Plasma ・Reactive Ion Etching) with Ar-C4F8 mixture for the reactive gas. The Ar-C4F8 mixing ratio fixed 10:1, and the chamber pressure was 0.5Pa during the etching. The KNN etching depth has linear relationship as the function of the etching time. The etching rate increased with increasing the antenna power and the bias power. The antenna and the bias dependence of the etching rate show good linearity. These linear relationships indicate the good controllability of the etching. The KNN/Pt selectivity increased with increasing antenna power and decreasing bias power. Thus, the high KNN etching rate was obtained at high antenna power and high bias power condition. The high KNN/Pt selectivity was obtained at high antenna power and low bias power condition. The maximum KNN etching rate and KNN/Pt selectivity were 371.6nm/min. and 85.8, respectively. These results are comparable to that of the conventional piezoelectric PZT films.

Investigating the plasma chemistry for the synthesis of carbon nanotubes/nanofibres in an inductively coupled plasma-enhanced CVD system: the effect of processing parameters

A parameter study is carried out for an inductively coupled plasma used for the synthesis of carbon nanotubes or carbon nanofibres (CNTs/CNFs), by means of the Hybrid Plasma Equipment Model. The influence of processing parameters including gas ratio for four different gas mixtures typically used for CNT/CNF growth (i.e. CH4/H2, CH4/NH3, C2H2/H2 and C2H2/NH3), inductively coupled plasma (ICP) power (50–1000 W), operating pressure (10 mTorr–1 Torr), bias power (0–1000 W) and temperature of the substrate (0–1000 °C) on the plasma chemistry is investigated and the optimized conditions for CNT/CNF growth are analysed. Summarized, our calculations suggest that a lower fraction of hydrocarbon gases (CH4 or C2H2, i.e. below 20%) and hence a higher fraction of etchant gases (H2 or NH3) in the gas mixture result in more ‘clean’ conditions for controlled CNT/CNF growth. The same applies to a higher ICP power, a moderate ICP gas pressure above 100 mTorr (at least for single-walled carbon nanotubes), a high bias power (for aligned CNTs) and an intermediate substrate temperature.

Ni-P, Ni-B and SiO2 as Materials for Hard Mask in Deep Silicon Etching for MEMS Fabrication using ICP reactor

Ni-P, Ni-B and SiO2 films were used as hard mask materials in Si etching using a high-density inductively coupled plasma (ICP) reactor for MEMS fabrication. The Ni-P and Ni-B films were deposited using an electroless method, and the SiO2 film was thermally grown in a conventional furnace. Two etching processes were used to characterize the masks. The first uses SF6/Ar gas mixture varying bias power and process time, and the second is a Bosch like process, using C4F8 as a passivation gas. The Ni-P mask showed the highest resistance to etching, being applicable in Si deep etching (>100m); while the SiO2 mask was found to be less resistive, especially under strong ion bombardment (high bias power). The Ni-B mask was found to be highly porous, resulting in formation of micropillars during ecthing, which may be interesting for some apllications such as sensors.

Inductively coupled plasma etching of GaAs suspended photonic crystal cavities

The authors studied the dry-etching process by chlorine-based inductively coupled plasma for fabricating GaAs suspended photonic crystal cavities. To achieve low optical loss in such resonators, the photonic crystal holes must demonstrate simultaneously smooth sidewalls and good verticality. The influence of various parameters of the inductively coupled plasma process was investigated: a systematic analysis is provided on the dependency of hole sidewall roughness and shape on the process parameters such as gas mixture, etching power, pressure, and rf bias. The authors show that a combination of high pressure, high bias, and high etching power in an inductively coupled plasma with a Cl2/N2 chemistry is beneficial for achieving straight and smooth sidewalls, and wall tilt of less than 4° was obtained.

Characterization of plasma etching damage on p-type GaN using Schottky diodes

The plasma etching damage in p-type GaN has been characterized. From current-voltage and capacitance-voltage characteristics of Schottky diodes, it was revealed that inductively coupled plasma (ICP) etching causes an increase in series resistance of the Schottky diodes and compensation of acceptors in p-type GaN. We investigated deep levels near the valence band of p-type GaN using current deep level transient spectroscopy (DLTS), and no deep level originating from the ICP etching damage was observed. On the other hand, by capacitance DLTS measurements for n-type GaN, we observed an increase in concentration of a donor-type defect with an activation energy of 0.25eV after the ICP etching. The origin of this defect would be due to nitrogen vacancies. We also observed this defect by photocapacitance measurements for ICP-etched p-type GaN. For both n- and p-type GaN, we found that the low bias power ICP etching is effective to reduce the concentration of this defect introduced by the high bias power ICP etching.

Low energy N2 ion bombardment for removal of (HfO2)x(SiON)1−x in dilute HF

The ion assisted wet removal of (HfO2)x(SiON)1−x high dielectric constant (k) materials and its effect on electrical properties were investigated. Crystallization temperature of (HfO2)x(SiON)1−x increased as the percentage of SiON increased. The crystallized (HfO2)0.6(SiON)0.4 was damaged and turned to an amorphous film via incorporation of N species into the film by N2 plasma treatment. In addition, the structure of (HfO2)0.6(SiON)0.4 was disintegrated into HfO2, SiO(N), and ON after N2 plasma treatment. N2 plasmas using low bias power were applied for wet removal of high-k films and the mechanism of the ion assisted wet removal process was explored. When high bias power was applied, the surface of source and drain regions was nitrided via the reaction between N and Si substrates. Feasibility of the low bias power assisted wet removal process was demonstrated for short channel high-k metal oxide semiconductor device fabrication by the smaller shift of threshold voltage, compared to the high bias power as...

Effect of DC Bias Voltage on the Optical Properties of TiO&lt;sub&gt;2&lt;/sub&gt; Thin Film Deposited by Plasma Assisted Electron Beam Evaporation

It has been well known the effect of ion-bombardment is controlled by self-bias voltage around a substrate, and it is increased by high RF power and vacuum, but decreased with a substrate size. However, RF matching is unstable under high vacuum state and high RF power. Thus in this study, in order to overcome those problem and to increase the ion collision effect, we supplied various DC bias voltages(self bias, -150, -200, -250, -300, -350 Voltage) to substrate at lowRF power(100W). It was investigated by XRD, AFM, SEM, UV spectrometer and Ellipsometer that effect of DC bias voltage on the optical properties of TiO 2 single layer film deposited on p-type Si wafer(100) and glasses by RF plasma assisted electron beam evaporation(PAEBE) without substrate heating. It was measured the self bias is -110V at condition of 100W RF power, 3x10 -3 torr. The micro crack on surface was not found until -200V DC bias power, but large grain (cluster) is occurred at high bias power (-250, -300, -350V). RMS values were increased with increasing the bias power but refractive index (n-value) was decreased with one.

Characterization of the NiFe sputter etch process in a rf plasma

The sputter etching of NiFe thin films by Ar ions in a rf plasma has been studied and characterized with the use of a Langmuir probe. The NiFe sputter etch rate was found to depend strongly on incident ion energy, with the highest NiFe etch rates occurring at high rf bias power, low pressure, and moderate rf source power. NiFe etch rates initially increased with increasing rf source power, then saturated at higher rf source powers. Pressure had the weakest effect on NiFe etch rates. Empirically determined sputter yields based on the NiFe etch rates and ion current densities were calculated, and these compared favorably to sputter yields determined using the sputtering model proposed by Sigmund [Phys. Rev. 184, 383 (1969)].

Deep reactive ion etching of borosilicate glass using an anodically bonded silicon wafer as an etching mask

A new dry-etching method for fabricating anisotropic deep grooves on a borosilicate glass wafer is reported. The method uses a 200 µm thick bulk silicon wafer bonded to a borosilicate glass wafer by anodic bonding as an etching mask and inductively coupled plasma generated by C4F8 or CHF3 gases for etching. The measured etching rate showed that the deep reactive ion etching conditions for achieving a high etching rate are low gas pressure, high gas-flow rate, high antenna power and high bias power. The measured groove profile revealed that the sidewall angles of the etched grooves were less than 80° and C4F8 plasma provided a slight difference in width between the mask opening and etched groove. These results indicate that C4F8 plasma is suitable for precise groove fabrication. Even after fabrication of a 430 µm deep groove, enough silicon mask (135 µm thickness) remained to fabricate a deeper groove. Consequently, our etching method using an anodically bonded silicon mask and C4F8 plasma enables the fabrication of very deep grooves in borosilicate glass.

Etching of Copper Films for Thin Film Transistor Liquid Crystal Display using Inductively Coupled Chlorine-Based Plasmas

In this study, using inductively coupled Cl2/Ar plasmas (ICP), the effects of various process conditions such as source power, bias power, Cl2/Ar gas ratio, and ultraviolet (UV) ray were investigated to obtain high Cu etch rates without remaining any nonvolatile etch products. Due to the formation of nonvolatile copper chloride, copper film was not etched and, instead, a thick copper chloride residue was formed on the copper surface when Cl2/Ar ICP plasma was used. However, the residue could be removed and the copper film could be etched when high ICP source power was used with high intensity of ultraviolet rays and high bias power. Especially, when the ICP source power higher than 300 W and the bias power higher than 70 W were used, no UV irradiation was required to etch copper films, and which is more desirable for the etching of thin film transistor liquid crystal display (TFT-LCD) substrates. The maximum copper etch rate obtained was 300 nm/min with inductive power of 600 W, bias power of 75 W, and 0.5 of Cl2/Ar gas ratio at the pressure of 7 mTorr without applying UV.

Pressure Controlled Surface Roughness of SiC Plasma Etching

The effects of pressure on the surface roughness of silicon carbide (SiC) film were investigated with variations in other process parameters, including radio frequency source power, bias power, O2 fraction and gap. The SiC films were etched in a C2F6 inductively coupled plasma. The surface roughness, measured by atomic force microscopy, was examined in three particular situations: generated at minimum parameter level (case 1), at high source power (case 2); and at high bias power (case 3). For all variations in the process parameters except the O2 variation, smoother surfaces were observed at lower pressures in all cases. Both surface roughness and dc bias due to the pressure variation were highly correlated in case 1. In case 3, they were inversely related. It is noticeable that in cases 1 and 2 the surface roughness was little affected by the pressure, particularly at high bias power. This feature can be used not only for tight control of surface roughness, but for optimising other etch outputs.

Optimization of post-N2 treatment and undoped-Si-glass cap to improve metal wring delamination in deep submicron high-density plasma-fluorinated silica glass intermetal dielectric application

Integration issues of metal line delamination from fluorinated silica glass (FSG) in deep submicron intermetal dielectric applications were investigated in this study. A metal line peeled off after a nonoptimized in situ deposition of undoped-silicon-glass (USG; SiO2) capping layer followed the post-FSG-chemical mechanical polishing N2 treatment. It was found that higher bias power and longer process time of N2 treatment led to more active fluorine species diffusing from the FSG films to the USG surface, which might react with subsequent Ti/TiN/W metal layer and result in metal delamination. Using plasma-enhanced N2 treatment and ex situ USG capping with lower initial deposition temperature by extra cooling step, the stability of the FSG films was improved and resulted in a robust structure without metal peeling.

Vertical Etching of Ru Electrodes Using the O&lt;sub&gt;2&lt;/sub&gt;/Cl&lt;sub&gt;2&lt;/sub&gt; System

In this study, the etching characteristics of ruthenium (Ru) electrode using O2/Cl2 helicon plasmas were investigated. The variation of Ru etch rate, Ru to SiO2 etch selectivity and Ru etching slope with varied process variables were scrutinized. Not only the gas chemistry affected the Ru etching, but also the higher bias power and argon (Ar) bombardment helped to faciliate the Ru etching.

Prediction of profile surface roughness in CHF3/CF4 plasma using neural network

Using a neural network, a profile roughness of plasma etching is characterized. The etching was conducted in a CHF3/CF4 inductively coupled plasma. The etch process was characterized by a 23 full factorial experiment. The process parameters that were varied in the design include radio frequency source and bias powers, and gas ratio. Relationships between the parameters and profile roughness were captured by training neural network with eight experiments plus one center experiment. Model appropriateness was tested with six experiments not pertaining to the training data. Model prediction capability was optimized by means of a genetic algorithm (GA). Compared to a conventional model, GA-optimized model demonstrated a drastic improvement of about 54% in predicting profile roughness. From the optimized model, several plots were generated to examine parameter effects on the profile roughness. Increasing the source power (or bias power) under high bias power (or source power) increased the profile roughness. More significant effect of the bias power was revealed. The profile roughness decreased with increasing the gas ratio was strongly correlated to the dc bias. The little variation in the profile roughness was ascribed to chamber plasma condition maintained at relatively low dc bias.