Rf Bias Power Research Articles

Solid Source Dry Etching Process for GaAs and InP

In this study, we have demonstrated vertical and smooth inductively coupled plasma (ICP) dry etching of InP and GaAs using either solid iodine or ICl3 powder as gas sources at a low process temperature of 90 °C. We prepared solid iodine crystal and ICl3 powder in the ICP etching chamber as etching gas sources in a process chamber. Iodine or chlorine gases were supplied with a high vapor pressure from the solid source into the process chamber. Vertical and smooth etching was realized under a condition of low temperature and low bias RF power. We believe that solid iodine- and ICl3 powder-ICP etching are very simple and useful processes for InP- and GaAs-based device fabrication with a resist mask.

Tantalum carbide etch characterization in inductively coupled Ar∕Cl2∕HBr plasmas

The etching properties of tantalum carbide (TaC) in inductively coupled Ar∕HBr∕Cl2 plasmas are investigated in this article. Both etching experiments on patterned and blanket wafers and an integrated plasma equipment-feature scale computational model are utilized in this investigation. Results show that TaC etching is adequately described by the classical reactive ion etching mechanism, whereby etching occurs due to the synergistic effect of Cl or Br atoms and energetic ions. TaC etches faster in the presence of Cl relative to Br. The TaC etch rate is small in gas mixtures containing 5% of Cl2 or HBr and 95% of Ar, and it increases considerably as Cl2 or HBr concentration is increased. Although this etch rate increase is partially due to the availability of more Cl or Br, the chemical nature of chlorine (Cl2+,Cl+) or bromine (Br+) ions also plays a strong role. The TaC etch rate increases little if Cl2 or HBr fraction in Ar∕Cl2 or Ar∕HBr gas mixture, respectively, is increased beyond 25%. The TaC etch rate increases with rf bias power under all conditions. Scanning electron micrographs of TaC films etched using a patterned mask show that TaC sidewalls are tapered at about 77°±3° and the angle does not change appreciably with gas mixture or rf bias power. It is determined that an angle dependent ion etching yield captures well the observed trends in TaC sidewall slope.

Repairing of Etching-Induced Damage of High-k Ba0.5Sr0.5TiO3 Thin Films by Oxygen Surface Plasma Treatment

Ba0.5Sr0.5TiO3 (BST) thin films were patterned for fabricating BST capacitors in a helicon-wave plasma system. The optimization conditions were an Ar (80%)/Cl2 (20%) gas mixture with a helicon-wave plasma power and a substrate bias rf power of 1500 and 90 W, respectively. From results of X-ray photoelectron spectroscopy, physical ion bombardment is more effective than chemical reaction for removing Sr, while Ti can be removed by the formation of volatile TiClx. Ba was primarily removed by chemically assisted physical etching (such as that using BaClx). Some etching residues consisting of Ba and Sr were found after the BST films were etched and increased leakage current density. Oxygen surface plasma treatment can effectively repair surface damage caused by etching, and it reduced the leakage current density of the BST capacitor from 4.0×10-7 to 3.0×10-8 A/cm2 at 1 MV/cm and increased the breakdown field to ∼2 MV/cm at 1.0×10-6 A/cm2.

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)].

Silicon columnar microstructures induced by an SF6/O2 plasma

An inductively coupled SF6/O2 plasma is used to form a columnar microstructure (CMS) on silicon samples cooled at very low temperature (∼ −100 °C). The formation of this CMS is studied as a function of bias voltage, temperature, RF power and gas pressure. The characteristic mean diameter and mean height of the microstructure are evaluated by image processing tools from SEM micrographs. A crystallographic effect is also observed at very low temperature, which induces a needle-shaped structure. A physical mechanism is proposed to explain the formation of this CMS.

Simulation of ion transport characteristics in a collisionless rf sheath

With a self-consistent fluid model, we study the transport characteristics of the collisionless radio frequency (rf) sheath in a high density plasma and calculate the spatiotemporal ion density and the potential drop across the sheath in one rf cycle for an rf-bias frequency of 13.56 MHz. We also study the effects of the rf frequency on the voltage at the electrode, the sheath width, the ion energy distribution (IED) and the ion angular distribution (IAD) at the electrode. The simulated results indicate that the incidence angles of the ions impinging on the electrode are less than 8° when the ion velocity in the direction perpendicular to the sheath field is assumed to be a Boltzmann distribution. It is concluded that the rf frequency plays a crucial role for the energy and angular distributions of ions impinging on the electrode surface, and the IED and IAD are affected by the plasma density and the rf-bias power.

Study of neutral-beam etching conditions for the fabrication of 7-nm-diameter nanocolumn structures using ferritin iron-core masks

We fabricated nanocolumn structure by using a low energy neutral beam and a ferritin iron-core mask. By using Cl2 gas plasma for generating neutral beam, we obtained a better etching profile than with SF6 gas plasma. Though Cl2:SF6=90:10 enabled faster etching than Cl2 gas without degrading the etching profile when the etching depth was 25 nm, a mixture with any quantity of SF6 gas resulted in a poor etching profile when the etching depth was 50 nm. The beam energy was optimized for the 50-nm depth using Cl2 gas by changing the rf bias power to the bottom electrode of the neutral-beam source. Using the optimum beam energy, extremely high etching selectivity of the Si to ferritin iron-core masks (about 80) as well as highly anisotropic etching profile could be realized. As a result, the diameter of the top of the Si nanocolumn structure was 7 nm, which was identical to that of the iron core in the ferritin. Additionally, the etching profile was almost vertical. We were also able to achieve a high aspect ratio of about 4.6. It is very difficult for conventional plasma etching processes to fabricate such fine structure, because the high-energy photons enhanced the etching of the iron-core mask in the conventional plasma processes.

Memory operation of Pt–SrBi 2Ta 2O 9–Y 2O 3–Si field-effect transistor with damage-free selective dry etching process

Metal–ferroelectric–insulator–semiconductor field-effect transistor (MFISFET) with Pt–SrBi 2Ta 2O 9 (SBT)–Y 2O 3–Si gate structure was fabricated with a selective dry etching of SBT and Y 2O 3. The etching characteristics of SBT, Y 2O 3, and silicon were investigated with various Ar/Cl 2 gas mixing ratios. Inductively coupled plasma (ICP) Powers, and RF bias powers and the surface condition after the etching was analyzed by using the scanning electron microscopy (SEM), atomic force microscopy (AFM), X-ray photoemission spectroscopy (XPS), and Rutherford back scattering (RBS) measurement. The etch-stop process was successfully achieved at the condition of 50% Cl 2 concentration with the ICP power of 900 W and the RF bias power of 100 W where the selectivity of SBT to Y 2O 3 was 4. The fabricated MFISFET with the etch-stop process showed good memory operation with 1.2 V threshold voltage difference and more than three orders of drain current difference in the on/off drain current ratio at the operating voltage of 7 V. These results meant that the etch-stop process was successfully carried out and applied to MFISFET fabrication without degradation of the ferroelectric characteristics and damage on silicon surface.

Phase transition and microstructure change in Ta–Zr alloy films by co-sputtering

Ta–Zr Thin films with wide range of compositions were deposited on Si (100) substrate by co-sputtering. RF bias power from 0 to 80 W was applied during deposition. Films were then characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), X-ray photoelectron spectroscopy (XPS) and four-point probe. Result showed the incorporation of Zr caused the formation of low-resistivity BCC Ta, which made the resistivity of Ta–Zr films decrease. With substrate bias increasing, the compositional range for forming amorphous structure shrank. The effects of both thermodynamics and kinetics on Ta–Zr amorphization tendency were discussed.

Production and loss mechanisms of SiClX etch products during silicon etching in a high density HBr∕Cl2∕O2 plasma

Si Cl X (X=0–2) radicals’ concentrations have been measured by broadband ultraviolet absorption spectroscopy during the etching of 200mm diameter silicon wafers in HBr∕Cl2∕O2 plasmas. We report the variations of the concentrations of these radicals as a function of the radio frequency (rf) source power and rf-bias power. The silicon wafer etch rate is measured simultaneously. From the measured radicals densities and using electron impact ionization cross sections found in the literature, the densities of SiClX+ ions are calculated and are found to be in good agreement with ion densities measured by mass spectrometry. The upper limit for the SiCl2 radical concentration is calculated from the wafer etch rate. By comparison with the measured SiCl2 radical concentration it is concluded that SiCl2 radicals should also be produced by the reactor walls due to the etching of silicon containing species adsorbed on the reactor walls. Finally, using electron impact dissociation cross sections, the densities of SiCl and Si are calculated from the measured densities of SiCl2 and SiCl, respectively. The comparison between the calculated and measured values of SiClX densities allowed us to conclude that SiCl (and Si) are produced both in the gas phase by electron impact dissociation of SiCl2 (SiCl) radicals and at the reactor wall surfaces by the neutralization and reflection of ≈50% of the flux of SiCl+ (Si+) ions impinging on these surfaces. At the same time SiCl and Si are estimated to be lost (adsorption and abstraction reactions) on the reactor walls with a probability ranging between 0.2 and 1.

Study of the 300-mm CoSi 2 defects induced by Soft Sputter Etch process before cobalt deposition—characterization, design of experiment and 200/300 mm comparison

Cobalt salicide has been the reference salicide for sub −0.25 μm CMOS technology nodes to lower the sheet resistances on source, drain and poly areas. A critical step in the cobalt silicide formation is the Soft Sputter Etch (SSE) before Co deposition [A.H.M. Kamal, N.S. Argenti, C.S. Blair, IEEE Transaction on Semiconductor Manufacturing, vol. 15, No. 3, August 2002.]. It has to remove native oxide and impurities to ensure a clean surface, but it should also be soft enough to avoid leakages and resputtering. In this paper, the 300-mm SSE chamber behavior is analyzed on blanket and 90-nm technology node patterned wafers to explain the formation mechanism of defects observed in salicided areas. Defectivity has been observed along the spacers, especially in pinched active structures. A nitrogen analysis performed by μ-Auger spectroscopy showed that nitride from the spacers was re-sputtered on the active areas during the SSE process, subsequently blocking the salicidation. DOE on blanket wafers were run characterizing the 300-mm chamber behavior. No major differences were seen between 200- and 300-mm chambers on blanket wafers. Etch rate was found to be mainly dependent on the RF bias power, whereas both RF bias and RF coil powers influenced the DC bias voltage on the wafer during process. Furthermore, a DOE on 90-nm technology node patterned wafers was performed in order to reduce nitride resputtering and thereby improve defectivity and device performance. Results showed that optimizing the ratio between RF bias and RF coil and changing both RF values was a way to improve the salicidation process. Indeed, salicidation improvement can be achieved by rising the DC bias on the wafer.

PECVD SiO2 and SiON films dependant on the rf bias power for low-loss silica waveguide

SiO2 and SiON films were deposited by radiofrequency plasma-enhanced chemical vapor deposition (rf PECVD) technique using SiH4 and N2O as precursor gases. The refractive index (n) decreases to 1.4480 with the increase of the rf bias power from 0 to 75 W and again increases to 1.4486 at the rf bias power of 100 W. A uniform accumulation of fine grains with the root-mean-square (RMS) surface roughness within the area ∼1.5 nm was detected and the grains pack together very densely. The thickness, refractive index, and surface morphology of the films were characterized by prism coupler, scanning electron microscopy (SEM) and atomic force microscopy (AFM).

Selective etching process of SrBi2Ta2O9 and CeO2 for self-aligned ferroelectric gate structure

We have investigated an etch-stop process and fabricated the ferroelectric gate transistor without damage of source and drain regions using etching selectivity between etch rates of SrBi2Ta2O9 (SBT) and CeO2. The SBT and CeO2 were etched with inductively coupled plasma (ICP) reactive ion etching at various Ar/Cl2 gas mixing ratios, ICP powers, and rf bias powers. The etching selectivity of SBT/CeO2 was 6.8 and the vertical etching angle of SBT was 82° at the condition of 50% Cl2 concentration with the ICP power of 900 W and the rf bias power of 100 W. The characteristics of devices fabricated with an etch-stop process showed that there was no degradation of the ferroelectric characteristics and on/off ratio of programmable operation was 104.

Direct trim etching process of Si/SiO2 gate stacks using 193 nm ArF patterns

A polysilicon gate of 30 nm length is successfully fabricated by direct trimming of a gate stack that consists of organic BARC, polycrystalline silicon (poly-Si) and SiO2. It is achieved by patterning the gate stack using a 193 nm photoresist process and trimming the poly-Si gate with HBr/Cl2 plasma chemistry in an inductively coupled plasma (ICP) etcher. HBr is found to be more effective than Cl2 as a trimming etchant since it can achieve a higher trimming rate. A maximum trimming rate of 32 nm/min is obtained with gas chemistry of 80% HBr and 20% Cl2 at 40 mTorr, ICP power of 700 W, and rf bias power of 0 W. In general, the trimming rate increases with an increase in ICP power from 200 to 800 W and decrease in pressure from 70 to 10 mTorr. The inclusion of SF6 and O2 to the plasma and the longer trimming time are found to be very effective in reducing the poly-Si footprint which is observed after the trimming process.

Formation of polycrystalline silicon germanium/HfO2 gate stack structure using inductively coupled plasma etching

A gate stack structure consisting of a polycrystalline silicon germanium (poly-SiGe) conductor and a HfO2 dielectric on a silicon substrate was formed by inductively coupled plasma etching using HBr/Cl2/O2. Etch rates of poly-SiGe with 46% Ge were 1.7–2.0 times higher than those of polycrystalline silicon, depending on processing conditions. In the small feature sized devices of 100 nm gates, the notching at the sidewall of poly-SiGe was pronounced up to the depth of 50 nm. The amount of notching increased with increasing inductive power and pressure, and decreasing rf bias power. A HfO2 etch rate of 950 Å/min was obtained at the condition of 550 W inductive power, 360 W rf bias power, and 10 mTorr pressure. Etch rates of HfO2 increased with increasing inductive power and rf bias power. Etching selectivity of poly-SiGe with respect to HfO2 increased significantly with the addition of 3.8% O2 to HBr, and it was possible to control the selectivity in the range of 15–70 by changing the rf bias power. The change in etching selectivity was considered mainly due to the change in HfO2 film property, originating from the incorporation of O into the remaining nonvolatile Hf and the reformation of HfOx during etching.

Etching silicon by SF6 in a continuous and pulsed power helicon reactor

The etch rate of silicon by SF6 in a helicon reactor has been measured along with simultaneous actinometric measurements of the concentration of atomic fluorine in the gas phase for a variety of gas flow rates resulting in pressures in the mTorr range. A bias rf power was applied to the substrate to investigate the effect of ion energy on the etch rate. The etch rate was found to be proportional to the fluorine concentration and independent of the bias for the higher gas flow rates. However, at lower flow rates, the situation was more complicated and no simple model can explain the measurements. Measurements of the etch rate were also made in the afterglow of a repetitively pulsed discharge so that the directed ion energy would be reduced to the thermal motion after the rapid collapse of the plasma potential. A simple model was developed to explain the temporal etching phenomena in terms of the lifetime of the atomic fluorine.

Study on self-bias voltage induced on the substrate by r.f. bias power in a high density plasma

Characteristics of the self-bias voltage induced on the substrate by a r.f. bias power in a high density helicon wave plasma were studied. The self-bias voltage induced on the wafer by r.f. bias of 300 kHz (less than the ion plasma frequency value) was directly measured with a Pt wafer using a high voltage probe. The self-bias voltage and the peak-to-peak voltage were influenced by both bias and source RF power in the high-density plasma. A new probe system was designed to trace these phenomena. It was found that these phenomena were related to the charging time of the blocking capacitor by the ion current. The role and effect of the negative ions on the etching process at low RF frequency are discussed. A method to apply the RF bias effectively without the influence of the source power was suggested for application in a real etching system.

Effects of Ar and O2 additives on SiO2 etching in C4F8-based plasmas

Gas mixtures based on C4F8 are promising for the development of high-performance SiO2 plasma etching processes. Measurements of important gas phase species, thin film etching rates and surface chemistry changes were performed for inductively coupled plasmas fed with C4F8/Ar and C4F8/O2 gas mixtures. The addition of Ar to C4F8 causes a strong increase of the plasma density relative to that of pure C4F8 (by up to a factor of 4× at 90% Ar). For O2 addition the changes in plasma density are small up to 90% O2 relative to pure C4F8. Infrared laser absorption spectroscopy was used to determine the absolute densities of neutral CF, CF2 and COF2 radical species as a function of the gas composition. The densities of CF and CF2 were enhanced for certain operating conditions when Ar was added to C4F8 as long as the amount of Ar remained below 20%. For instance, the partial pressure of CF was 0.1 mTorr for a 20 mTorr 1400 W source power discharge for pure C4F8, and increased to 0.13 mTorr at 20% Ar. Above 20% Ar it decreased, roughly following the gas dilution. The CF2 partial pressure was about 5 mTorr for the same conditions, and increased by about 10% at 20% Ar. Above 20% Ar the CF2 partial pressure decreased roughly linearly with the amount of Ar added, to about 2 mTorr at 50% Ar. Of particular interest was the analysis of the difference in behavior of CF, CF2 and COF2 partial pressures over SiO2 and Si surfaces, with and without rf bias power (in the latter case a self-bias voltage of −100 V was used). For pure C4F8 discharges at 20 mTorr and 1400 W inductive power without rf bias the partial pressures of CF, CF2 and COF2 radicals are comparable over SiO2 and Si surfaces. Upon applying a rf bias, the CF2 partial pressure over a SiO2 surface is reduced much more strongly than for a Si surface. The overall reduction appears to be consistent with the relative SiO2/Si etch rate ratios observed for these conditions. These results indicate that CF2 is consumed during the etching of SiO2 and Si. We also measured fluorocarbon deposition rates without rf bias and etching rates of blanket SiO2, silicon, resist and deposited fluorocarbon films as a function of the rf bias and feed gas composition. Important differences in the response of the etching rates of those materials upon the addition of O2 and Ar to C4F8 were observed. In particular, we show that the SiO2/Si and SiO2/resist etching selectivities can be doubled by adding up to 90% Ar to C4F8, without inducing an unacceptably large reduction of the SiO2 etching rate. The change in etch rate ratios is at least in part due to strong surface chemical changes seen for Ar-rich fluorocarbon gas mixtures. The surface chemical changes of Si and SiO2 surfaces were investigated by real-time ellipsometry and x-ray photoelectron spectroscopy. A strong reduction of the fluorine content of the fluorocarbon steady-state layer and an increase in thickness is seen when up to 90% Ar was added to C4F8, and this coincides with an increase of the SiO2/Si etching selectivity. The change in fluorocarbon surface chemistry can be explained by the strongly increased ion/neutral flux ratio that is characteristic of Ar-rich C4F8/Ar gas mixtures.

Charged species dynamics in an inductively coupled Ar/SF6 plasma discharge

The chemistry of high-density SF6 plasma discharges is not well characterized. In this article, a combination of computational modeling and experimental diagnostics has been utilized to understand charged species dynamics in an inductively coupled Ar/SF6 plasma discharge. The model is based on the two-dimensional Hybrid Plasma Equipment Model with a detailed plasma chemical mechanism for Ar/SF6. In the experiments, absolute electron density and total negative ion density have been measured using microwave interferometry and laser photodetachment, respectively. In addition, we have also utilized prior measurements of mass and energy resolved ion fluxes by Goyette et al. [J. Vac. Sci. Technol. A 19, 1294 (2001)]. Computational results show that all SFx+(x=0–5) ions are present in the plasma discharge. Important negative ions include SF6−, SF5−, and F−. Electron and positive ion densities increase with coil power due to enhanced ionization. However, negative ion densities decrease with coil power as the main negative ion precursor, SF6, is lost through neutral dissociation. An increase in SF6 concentration in the Ar/SF6 gas mixture decreases electron density due to enhanced electron loss through (dissociative) attachment, which enhances negative ion densities. RF bias power does not have an appreciable impact on the ion and electron densities for the parameter range of interest. Experiments show that electron density decreases with gas pressure while the total negative ion density increases up to 25 mTorr. This is due to a decrease in electron temperature, which enhances electron loss through (dissociative) attachment. Although the model is able to capture most of the experimentally observed trends, there are discrepancies regarding the impact of gas pressure on electron density and relative flux of large positive ions.

Texture enhancement of Al films on Ti underlayers by radio-frequency bias sputtering

The influence of sputtering pressure and radio-frequency (RF) bias power on the texture of Al/Ti thin films has been investigated. The Al/Ti thin films were deposited sequentially onto thermally oxidized Si wafers in a direct-current (DC) magnetron system. The RF bias was applied during Ti deposition. The texture of Al thin films was quantified by θ-2θ scans and rocking curves of x-ray diffraction (XRD). The Al thin films deposited on bias-sputtered Ti underlayers showed an epitaxial growth and strong (111) texture. The Al (111) texture improved with decreasing sputtering pressure and increasing RF-bias power. The Al/Ti texture was also enhanced when the SiO2/Si substrate surface was RF plasma cleaned prior to Ti deposition. The Al (111) texture was closely related to Ti (0002) texture. The mechanism of Ti-texture improvement by applying bias sputtering was explained based on the ion-bombardment effect.