Effect of BCl3 concentration and process pressure on the GaN mesa sidewalls in BCl3/Cl2 based inductively coupled plasma etching

GaN mesa etching is investigated using BCl₃/Cl₂ based inductively coupled plasma at constant ICP/RF powers for HEMT fabrication. The effect of chamber process pressure (5-15 mTorr) and BCl₃/Cl₂ flow rate ratio >1 on mesa sidewall profile is studied in detail using less complex photoresist mask. Mesa sidewall sharpness varied strongly with chamber pressure and deteriorated at lower pressure ∼5 mTorr. The etched GaN mesas resulted in severely damaged sidewalls with significant sidewall erosion at BCl₃/Cl₂ ratio of <1, which reduced gradually as BCl₃/Cl₂ ratio was increased to values >1 mainly due to decreased Cl ion/neutral scattering at the edges. Finally, the smooth and sharp mesa sidewalls with angle of ∼80° and moderate GaN etch rate of ∼1254 Å/min are obtained at BCl₃/Cl₂ ratio of 2.5:1 and 10 mTorr pressure due to a better balance between physical and chemical components of ICP etching.     

Comparison of dry etching of PMMA and polycarbonate in diffusion pump-based O2 capacitively coupled plasma and inductively coupled plasma

We report a comparison of dry etching of polymethyl methacrylate (PMMA) and polycarbonate (PC) in O₂ capacitively coupled plasma (CCP) and inductively coupled plasma (ICP). A diffusion pump was used as high vacuum pump in both cases. Experimental variables were process pressure (30-180 mTorr), CCP power (25-150 W) and ICP power (0-350 W). Gas flow rate was fixed at 5 sccm. An optimized process pressure range of 40-60 mTorr was found for the maximum etch rate of PMMA and PC in both CCP and ICP etch modes. ICP etching produced the highest etch rate of 0.9 μm/min for PMMA at 40 mTorr, 100 W CCP and 300 W ICP power, while 100 W CCP only plasma produced 0.46 μm/min for PMMA at the same condition. For polycarbonate, the highest etch rates were 0.45 and 0.27 μm/min, respectively. RMS surface roughnesses of PMMA and PC were about 2-3 nm after etching. Etch selectivity of PMMA over photoresist was 1-2 and that of PC was less than 1. When ICP power increased from 0 to 350 W, etch rates of PMMA and PC increased linearly from 0.47 to 1.18 μm/min and from 0.18 to 0.6 μm/min, while the negative self bias slightly reduced from 364 to 352 V. Increase of CCP power raised both self bias and PMMA etch rate. PMMA etch rates were about 3 times higher than those of PC at the same CCP conditions. SEM data showed that there was some undercutting of PMMA and PC after etching at 300 W ICP, 100 W CCP and 40 mTorr. The results also showed that the etched surface of PMMA was rough and that of PC was relatively smooth.     

Etching characteristics and mechanism of Ga-doped ZnO thin films in inductively-coupled HBr/X (X = Ar, He, N2, O2) plasmas

We investigated the etch characteristics and mechanisms of Ga-doped ZnO (Ga-ZnO) thin films in HBr/X (X = Ar, He, N₂, O₂) inductively-coupled plasmas. The etch rates of Ga-ZnO thin films were measured as a function of the additive gas fraction in the range of 0-100% for Ar, He, N₂, and O₂ at a fixed gas pressure (6 mTorr), input power (700 W), bias power (200 W), and total gas flow rate (40 sccm). The plasma chemistry was analyzed using a combination of the global (zero-dimensional) plasma model and Langmuir probe diagnostics. By comparing the behavior of the etch rate and fluxes of plasma active species, we found that the Ga-ZnO etch process was not limited by ion-surface interaction kinetics and appeared in the reaction rate-limited etch regime. In the HBr/O₂ plasma, the etch kinetics were probably influenced by oxidation of the etched surface.     

GaN etch rate and surface roughness evolution in Cl2/Ar based inductively coupled plasma etching

Cl₂/Ar based inductively coupled plasma (ICP) etching of GaN is investigated using photoresist mask in a consequential restricted domain of pressure < 1.2 Pa and radio frequency (RF) sample power < 100 W, for selective mesa etching. The etch characteristics and root-mean-square (rms) surface roughness are studied as a function of process parameters viz. process pressure, Cl₂ percentage in total flow rate ratio, and RF sample power at a constant ICP power, to achieve moderate GaN etch rate with anisotropic profiles and smooth surface morphology. The etch rate and resultant surface roughness of etched surface increased with pressure mainly due to dominant reactant limited etch regime. The etch rate and surface roughness show strong dependence on RF sample power with the former increasing and the later decreasing with the applied RF sample power up to 80 W. The process etch yield variation with applied RF sample power is also reported. The studied etch parameters result in highly anisotropic mesa structures with Ga rich etched surface.     

HfO2 etching mechanism in inductively-coupled Cl2/Ar plasma

Etching characteristics and the mechanism of HfO₂ thin films in Cl₂/Ar inductively-coupled plasma were investigated. The etch rate of HfO₂ was measured as a function of the Cl₂/Ar mixing ratio in the range of 0 to 100% Ar at a fixed gas pressure (6 mTorr), input power (700 W), and bias power (300 W). We found that an increase in the Ar mixing ratio resulted in a monotonic decrease in the HfO₂ etch rate in the range of 10.3 to 0.7 nm/min while the etch rate of the photoresist increased from 152.1 to 375.0 nm/min for 0 to 100% Ar. To examine the etching mechanism of HfO₂ films, we combined plasma diagnostics using Langmuir probes and quadrupole mass spectrometry with global (zero-dimensional) plasma modeling. We found that the HfO₂ etching process was not controlled by ion-surface interaction kinetics and formally corresponds to the reaction rate-limited etch regime.     

Dry etching of TiN in N2/Cl2/Ar adaptively coupled plasma

We investigated the N₂ additive effect on the etch rates of TiN and SiO₂ and etch profile of TiN in N₂/Cl₂/Ar adaptively coupled plasma (ACP). The mixing ratio of Cl₂ and Ar was fixed at 75 and 25 sccm, respectively. The N₂ flow rate was increased from 0 to 9 sccm under the constant pressure of 10 mTorr. As N₂ flow rate was increased in N₂/Cl₂/Ar plasma, the etch rate of TiN was linearly increased, but that of SiO₂ was increased non-monotonically. The etch profile and the compositional changes of TiN was investigated with field emission-scanning electron microscope (FE-SEM), FE-Auger electron spectroscopy (FE-AES) and x-ray photoelectron spectroscopy (XPS). When 9 sccm N₂ was added into Cl₂/Ar, a steep etch profile and clean surface of TiN was obtained. In addition, the signals of TiN and Ti were disappeared in FE-AES and XPS when N₂ additive flow into Cl₂/Ar was above 6 sccm. From the experimental data, the increase in TiN etch rate was mainly caused by the increase of desorption and evacuation rate of etch by products because of the increased effective pumping speed. The etch mechanism of TiN in N₂/Cl₂/Ar ACP plasma can be concluded as the ion enhanced chemical etch.     

Profile simulation of high aspect ratio contact etch

A semi-empirical profile simulator was employed to better understand fundamental mechanisms of feature evolution in a high aspect ratio contact plasma etch process. Simulation results showed that the net deposition rate of polymer on sidewall defined the necking and surface scattering of ions from the secondary facet caused the formation of bowing. As neutral depositor flux was increased, the resulting profile showed a monotonic increase in necking. In contrast, the extent of bowing showed a maximum, such that minimal bowing was obtained at low and at high depositor fluxes. Primary faceting of photo resist showed only a small influence on the SiO₂ etch profile.     

The etching characteristics of Al2O3 thin films in an inductively coupled plasma

In this study, the etching characteristics of ALD deposited Al₂O₃ thin film in a BCl₃/N₂ plasma were investigated. The experiments were performed by comparing the etch rates and the selectivity of Al₂O₃ over SiO₂ as functions of the input plasma parameters, such as the gas mixing ratio, the DC-bias voltage, the RF power, and the process pressure. The maximum etch rate was obtained at 155.8 nm/min under a 15 mTorr process pressure, 700 W of RF power, and a BCl₃ (6 sccm)/N₂ (14 sccm) plasma. The highest etch selectivity was 1.9. We used X-ray photoelectron spectroscopy (XPS) to investigate the chemical reactions on the etched surface. Auger electron spectroscopy (AES) was used for the elemental analysis of the etched surfaces.     

锗表面二维亚波长结构的反应离子刻蚀制备

为了获得具有金字塔结构的二维亚波长结构表面,提高其高宽比,用掩模曝光光刻及反应离子刻蚀技术,以SF6和O2为反应气体,在Ge衬底上制备了二维亚波长结构.用扫描电镜对刻蚀图形的形貌进行了观察,研究了功率、气压、气体流量及掩模图形对刻蚀图形的影响.结果表明:刻蚀图形腰部被优先刻蚀,形成凹陷的侧壁轮廓;O2流量增大有利于在侧壁形成保护层,从而减小腰部刻蚀、增大顶部及根部刻蚀;功率及气压过大或过小均会使侧壁刻蚀较大;方形图案比圆形图案掩模更有利于刻蚀出深度较大的亚波长结构.     

Surface roughness of silicon carbide etching in a NF3 inductively coupled plasma

Silicon carbide was etched in a NF₃/CH₄ inductively coupled plasma. Surface roughness measured by atomic force microscopy was investigated as a function of process parameters. Both etch rate and dc bias were correlated to the surface roughness. To optimize the surface roughness, a 2⁴ full factorial experiment was conducted for 700-900 W source power, 50-150 W bias power, 0.80-1.60 Pa, and 20-100% NF₃ percentage. Main effect analysis revealed that the surface roughness is the most strongly affected by the bias power. For variations in the bias power or NF₃ percentage, decrease in the surface roughness was observed only as positive variations in the etch rate and dc bias are considerably large. The surface roughness with the pressure was chemically dominated as illustrated by its inverse relationship with the dc bias. For the variations in the NF₃ percentage, the radical variation was estimated to play a more dominant role. The smoothest surface roughness of 0.312 nm was obtained at 700 W source power, 150 W bias power, 1.60 Pa pressure, and 100% NF₃ percentage.     

Growth of crystallized Ge films from VHF inductively-coupled plasma of H2-diluted GeH4

We have studied the Ge crystalline nucleation and film growth on quartz substrate at 250 °C from inductively-coupled plasma (ICP) of GeH₄ diluted with H₂. The ICP was generated by supplying 60 MHz power to an external single-turn antenna which was placed on a quartz plate window of a stainless steel reactor and parallel to the substrate. We have found that the growth rate is significantly increased when the preferential growth of the (110) plane becomes pronounced after the formation of randomly-oriented crystalline network. The (110) oriented Ge films, of which average crystallinity is as high as 70%. The integrated intensity ratio of TO phonons in crystalline phase to those in disordered phase, were grown at a rate of ∼ 4.0 nm/s after the formation of amorphous incubation layer with a thickness of ∼ 0.1 μm on quartz.     

Dual etch processes of via and metal paste filling for through silicon via process

We investigated the etched slope control of silicon via and the filling of the through silicon via (TSV) with nano-scale Ag paste. Patterns for a 10 μm-diameter hole and line were etched using the Bosch process in a deep reactive ion etching system (the first etching process). The diameter of via_(Top) and the depth of the via were about 10.6 μm and 80 μm, respectively, with a nearly vertical profile. In sequence, the tapered via and the removal of the scallops were obtained using an inductively-coupled etching system (the second etching process). We investigated the effects of gas pressure and input power on the slope of the TSV during the second etching process. In the second etching process, as the process pressure increased from 10 to 80 mTorr, the diameter of via_(Top) decreased from 13 to 12.2 μm. Meanwhile, the expansion of via_(Top) increased with increasing source power from 200 to 600 W. We found that the expansions of the via_(Top) size were related to the desorption of by-product and the arrival of ion flux. We achieved a smooth surface and a slope angle of about 86° using the dual etch process. Finally, the depth of the 80 μm via was filled with nano-scale Ag paste using a vacuum-assisted filling method.     

Deep etch-induced damage during ion-assisted chemical etching of sputtered indium-zinc-oxide films in Ar/CH4/H2 plasmas

Plasma etch damage to sputtered indium-zinc-oxide (IZO) layers in the form of changes in the film stoichiometry was investigated using Auger Electron Spectroscopy (AES). While damage resulting from pure chemical etching processes is usually constrained to the surface vicinity, ion-assisted chemical etching of IZO in Ar/CH₄/H₂ plasmas produces a Zn-rich layer, whose thickness (∼ 50 nm) is well-above the expected stopping range of Ar ions in IZO (∼ 1.5 nm). Based on AES depth profiles as a function of plasma exposure time, it is concluded that the observed Zn enrichment and In depletion deep into the IZO film are driven by the implantation of hydrogen atoms.     

Deposition of amorphous carbon nitride films using Ar/N2 supermagnetron sputter

Amorphous carbon nitride (a-CN_x) films were formed by supermagnetron sputter deposition using N₂ and/or Ar gases. Supplying rf power with a substrate-holding electrode (bias sputter) and lowering the gas pressure were found to be effective at decreasing the optical band gap and increasing the hardness. Nitrogen concentrations of bias sputtered films were about 32-35 mass% (30-100 mTorr). The a-CN_x films deposited for electron field emission showed a low-threshold electric field (E_TH). With the decrease of gas pressure, admixture of Ar to N₂ or the use of pure Ar, and the use of bias sputter, the E_TH of a-CN_x films largely decreased to 11 V/μm (30 mTorr Ar/N₂ bias sputter).     

Dry etching process of GaAs in capacitively coupled BCl3-based plasmas

Dry etching of GaAs was investigated in BCl₃, BCl₃/N₂ and BCl₃/Ar discharges with a mechanical pump-based capacitively coupled plasma system. Etched GaAs samples were characterized using scanning electron microscopy and surface profilometry. Optical emission spectroscopy was used to monitor the BCl₃-based plasma during etching. Pure BCl₃ plasma was found to be suitable for GaAs etching at > 100 mTorr while producing a clean and smooth surface and vertical sidewall. Adding N₂ or Ar to the BCl₃ helped increase the etch rates of GaAs. For example, the GaAs etch rate was doubled with 20% N₂ composition in the BCl₃/N₂ plasma compared to the pure BCl₃ discharge at 150 W CCP power and 150 mTorr chamber pressure. The GaAs etch rate was ∼ 0.21 µm/min in the 20 sccm BCl₃ plasma. The BCl₃/Ar plasma also increased etch rates of GaAs with 20% of Ar in the discharge. However, the surface morphology of GaAs was strongly roughened with high percentage (> 30%) of N₂ and Ar in the BCl₃/N₂ and BCl₃/Ar plasma, respectively. Optical emission spectra showed that there was a broad BCl₃-related molecular peak at 450-700 nm wavelength in the pure BCl₃ plasma. When more than 50% N₂ was added to the BCl₃ plasma, an atomic N peak (367.05 nm) and molecular N₂ peaks (550-800 nm) were detected. Etch selectivity of GaAs to photoresist decreased with the increase of % N₂ and Ar in the BCl₃-based plasma.     

Comparative study on dry etching of polycrystalline diamond thin films

The reactive ion etching (RIE) technique was used to etch polycrystalline diamond thin films. In this study we investigate the influence of process parameters (total pressure, rf power, gas composition) of standard capacitively coupled plasma RIE system on the etching rate of diamond films. The surface morphology of etched diamond films was characterized by Scanning Electron Microscopy and the chemical composition of the etched film part was investigated by Raman Spectroscopy.We found that the gas composition had a crucial effect on the diamond film morphology. The use of CF₄ gas resulted in flatter surfaces and lateral-like etching, while the use of pure O₂ gas resulted in needle-like structures. Addition of argon to the reactant precursors increased the ion bombardment, which in turn increased the formation of non-diamond phases. Next, increasing the rf power from 100 to 500 W increased the etching rate from 5.4 to 8.6 μm/h. In contrast to this observation, the rise of process pressure from 80 to 150 mTorr lowered the etching rate from 5.6 down to 3.6 μm/h.     

High aspect ratio via etch development for Cu nails in 3-D-stacked ICs

In this study the etch development of high aspect ratio vias in Si for the fabrication of Cu nails is described. To enable subsequent metallisation, these vias need to meet strict requirements with respect to uniformity, slope, sidewall roughness and undercut. For aspect ratios up to 5 a SiO₂ hard mask based SF6/O₂ etch approach is used. For aspect ratios up to 10, a resist based passivation polymer type etch approach with C4F8/SF6 was used to successfully pattern vias in Si. Typical problems of this process and optimization to overcome the issues are described.     

Antireflective hydrophobic si subwavelength structures using thermally dewetted Ni/SiO2 nanomask patterns

We report the disordered silicon (Si) subwavelength structures (SWSs), which are fabricated with the use of inductively coupled plasma (ICP) etching in SiCl4 gas using nickel/silicon dioxide (Ni/SiO2) nanopattens as the etch mask, on Si substrates by varying the etching parameters for broadband antireflective and self-cleaning surfaces. For the fabricated Si SWSs, the antireflection characteristics are experimentally investigated and a theoretical analysis is made based on the rigorous coupled-wave analysis method. The desirable dot-like Ni nanoparticles on SiO2/Si substrates are formed by the thermal dewetting process of Ni films at 900 degrees C. The truncated cone shaped Si SWS with a high average height of 790 +/- 23 nm, which is fabricated by ICP etching with 5 sccm SiCl4 at 50 W RF power with additional 200 W ICP power under 10 mTorr process pressure, exhibits a low average reflectance of approximately 5% over a wide wavelength range of 450-1050 nm. The water contact angle of 110 degrees is obtained, indicating a hydrophobic surface. The calculated reflectance results are also reasonably consistent with the experimental data.     

Process optimization of CF4/Ar plasma etching of Au using I-optimal design

In this paper, reactive ion etching of Au is performed with CF₄/Ar gases, and process optimization method is suggested using a statistically established process model. The I-optimal design was employed to set up the etching experiment with operating parameters, namely, gas composition, RF power and chamber pressure. Its analysis was performed on individual parameters of the etch rate, selectivity, and profile. In addition, process optimization, including all three responses of interest, is provided simultaneously. We confirmed that a nonvolatile by-product AuF_x was re-deposited on the surface, but controlling the amount of carbon fluoride provided a good etch rate with a satisfactory sidewall profile by reducing by-products. Although RF power is closely related with etch rate, increased power gives poor selectivity due to increased physical etching. Pressure and gas flows strongly interact with each other, affecting sidewall characteristics. Suggested optimization simultaneously considers three responses of interests, which is crucial in process development and optimization for quickly ramping up high volume manufacturing.     

Field-emission characteristics of diamond-like amorphous carbon films deposited by mixed gas (N2 or H2) controlled i-C4H10 supermagnetron plasma

Diamond-like amorphous carbon (DAC) films were deposited for field-emission application using supermagnetron plasma by mixing N₂ or H₂ in i-C₄H₁₀ gas at the upper and lower electrode rf powers (UPRF/LORF) of 800 W/100-800 W. At an 800 W/800 W, the N₂ (0-80%) gas-mixed DAC films showed an emission threshold electric field (E_TH) of 19 V/μm. At the 800 W/100 W, the H₂ (20%) gas-mixed DAC film showed low E_TH's of 13 V/μm, respectively. The moderate reduction of CC and CN double bonds by the decrease of LORF from 800 W to 100 W was found to be effective to lower E_TH.