Modelling of plasma etching using a generalized regression neural network

Plasma etching was modelled by using a generalized regression neural network (GRNN). The etching process was characterized with a statistical experimental design. Three etch responses were modelled, which include two etch rates of aluminium and silica and etching profile. GRNN prediction ability was optimized as a function of training factor. Three types of models were constructed depending on the type of prepared data. Type I model corresponds to the model constructed with the original, non-classified data. Type II and III models were built for the classified data without and with the control of data interface, respectively. Compared to type I models, type II models for two etch rates demonstrated more than 25% improvement. By the control of data interface, type III models exhibited more than 15% improvement over type II models. Classification-based models in conjunction with data control thus illustrated much improved prediction of GRNN over those for non-classified models.     

In-situ monitoring of plasma equipment using spectrophotometric colorimetry and neural network

Plasma plays a critical role in fabricating thin films for manufacturing electronic devices. To improve equipment throughput and device yield, plasma should be strictly monitored. A new technique is presented to monitor process plasma. This is accomplished by using a spectrophotometer and a neural network. The spectrometer was used to calculate color chromaticity coordinates and color temperature. Their sensitivity was examined as a function of radio frequency source and bias powers. All the variables linearly increased or decreased for process-induced faults. For an in-situ monitoring, the color variables were modeled by using an auto-correlated type of a time-series neural network and model prediction was applied to a CUSUM control chart. Unlike conventional models, the neural network trained with a reference pattern was utilized for fault detection. The constructed models demonstrated prediction errors ranging from 0.018 to 0.027%. A presented model-based CUSUM yielded improved fault sensitivity over a raw sensor data-based one. The improvement was more drastic for the bias power-induced faults. The proven high sensitivity can be effectively applied to monitor plasma equipment in real-time.     

The application of pulsed HTHD plasma to the synthesis of ultra-thin DLC films

The pulsed high-temperature and high-density (HTHD) plasma, which was mainly used for nuclear fusion research, was utilized to synthesize DLC films. X-ray photo-electron spectra (XPS), nano-indention, and atomic force microscopy (AFM) were employed to investigate the films. It was found that the films were quite smooth and can be managed to be extremely thin. The characteristics of HTHD plasma in DLC deposition were also discussed by comparing with other techniques.     

Studies using AFM and STM of the correlated effects of the deposition parameters on the topography of gold on mica

A systematic study of the correlated effects of deposition temperature, film thickness and deposition rate on the morphology of gold films on mica was carried out using atomic force microscopy and scanning tunnelling microscopy. For the range of thicknesses, rates and temperatures concerned, a variety of surface structures, mainly in the form of rounded mounds, islands, long channel and short channel plateaux topographical features were formed under various combinations of these three parameters. The rounded mounds and islands formed, respectively, on the mica substrate at room temperature and at 150°C were found to be essentially independent of the film thickness and deposition rate selected. When deposited at higher temperatures (300°C and 440°C), a change from islands to channelled features, via the growth and coalescence of the islands, was observed either on increasing the film thickness for a given deposition rate (≈ 1.0 Å s⁻¹) or on increasing the deposition rate to a given film thickness (400 Å). It is evident from the results presented that, whereas the film thickness and the deposition rate at a given temperature determine to what extent the film has coalesced, the growth temperature influences the lateral size of the surface features formed. In consequence, conducting films were found when the gold coverage was ≥ ≈0.9. Investigation of the vertical characteristics of the films was also conducted. The origin of all the phenomena observed can be attributed to competition between secondary nucleation and thermally enhanced diffusion processes occurring on the gold surface during deposition and film formation.     

Controlled modification of octadecyltrichlorosilane self-assembled monolayer by CO2 plasma

CO₂-plasma is used to introduce functional groups on the uppermost surface of an alkoxysilane self-assembled monolayer (SAM). The structural and chemical modifications of the material surface were monitored by X-ray reflectometry, atomic force microscopy, X-ray photoelectrons spectroscopy and water contact angle measurements. Optimization of the plasma parameters is performed in order to achieve a maximum functionalization and to prevent degradation of the SAM. Finally, the ability of grafting organic compounds onto the plasma modified SAMS was demonstrated by the formation of an alkoxysilane bilayer.     

Investigation of sub-nm ALD aluminum oxide films by plasma assisted etch-through

A new technique, called “plasma defect etching” (PDE), is proposed for studying the continuity of ultra-thin layers. The PDE technique utilizes the extremely high selectivity in the deep reactive ion etching (DRIE) process, thus achieving visualization of the defects in the layer, because etching of substrate happens only through voids and microholes of the layer. The etch profile generally reproduces the non-continuous structure of the layer. This PDE technique was applied for the investigation of thin, sub-nm aluminum oxide films grown on silicon wafers by atomic layer deposition (ALD) technique. Silicon substrate was etched by SF₆ at cryogenic temperatures in an inductively coupled plasma (ICP) reactor, exploiting the extremely high ratio of silicon/aluminum oxide etch rates in fluorine plasmas. The surface morphology was characterized by scanning electron microscopy (SEM) and atomic force microscopy (AFM). The PDE method shows that in the case of water as an oxidation precursor, separate islands of aluminum oxide form during the five first ALD cycles. On the other hand, the use of ozone precursor helps to oxidize silicon surface and facilitates growth of a uniform layer.     

Prediction of plasma etching using a randomized generalized regression neural network

A new empirical technique to construct predictive models of plasma etch processes is presented. This was accomplished by combining a generalized regression neural network (GRNN) and a random generator (RG). The RG played a critical role to control neuron spreads in the pattern layer. The proposed R-GRNN was evaluated with experimental plasma etch data. The etching of silica thin films was characterized by a 2³ full factorial experiment. The etch responses examined include aluminium etch rate, silica etch rate, profile angle, and DC bias. Additional test data were prepared to evaluate model appropriateness. Compared to conventional GRNN, the R-GRNN demonstrated much improved predictions of more than 40% for all etch responses. This was illustrated over statistical regression models. As a result, the proposed R-GRNN is an effective way to considerably improve the predictive ability of conventional GRNN.     

Tailoring of the morphology and chemical composition of thin organosilane microwave plasma polymer layers on metal substrates

The growth of thin microwave organosilicon plasma polymers on model zinc surfaces was investigated as a function of the film thickness and the oxygen partial pressure during film deposition. The evolution of the topology of the film was studied by atomic force microscopy (AFM). The nano- and micro-roughness was investigated at the inner and the outer surfaces of the plasma polymers. A special etching procedure was developed to reveal the underside of the plasma polymer and thereby its inner surface. Rough films contained voids at the interface, which reduced the polymer/metal contact area. The increase in oxygen partial pressure led to a smoother film growth with a perfect imitation of the substrate topography at the interface. The chemical structure of the films was determined by infrared reflection absorption spectroscopy (IRRAS), X-ray photoelectron spectroscopy and time-of-flight secondary ion mass spectroscopy (ToF-SIMS). ToF-SIMS at the outer and the inner surface of the plasma polymers showed that the density of methylsilyl groups increases in the outer surface layer of the plasma polymer and depends on the oxygen partial pressure. The chemical composition of the films could be altered to pure SiO₂ without changing the morphology by using oxygen-plasma post-treatment. This was proved by means of IRRAS and AFM. Chemistry and topology of the films were correlated with the apparent water contact angle. It was found that a linear relationship exists between the nanoscopic roughness of the plasma polymer and the static contact angle of water. Superposition of a nanoscopic roughness of the metal surface and the nanoscopic roughness of methylsilyl-rich films led to ultra-hydrophobic films with water contact angles up to 160°.     

AFM studies of microindented GaN and InGaN

The microhardness characteristics of GaN and GaN/InGaN films epitaxially grown on (0001) sapphire have been investigated using Vickers and Knoop indenters. The variation of H_V and H_K follows a reverse type of indentation size effect (reverse ISE). The microhardness results have also been analyzed using Meyer's law, Hays-Kendall approach and Proportional specimen resistance (PSR) model. The effect of N⁺ implantation on the microhardness of GaN has also been studied. The implanted sample is more resistant to plastic penetration than the unimplanted one and it is found that implantation enhances the surface hardness. Detailed AFM studies around the indented regions of the GaN and GaN/InGaN give the nature and behavior of the deformation on the surface.     

Single layer and multilayered films of plasma polymers analyzed by nanoindentation and spectroscopic ellipsometry

Well-defined single layer and multilayered a-SiC:H films, deposited from tetravinylsilane at different powers by plasma-enhanced chemical vapor deposition on silicon, were intensively studied by in situ spectroscopic ellipsometry, nanoindentation, and atomic force microscopy. A realistic model of the sample structure was used to analyze ellipsometric data and distinguish individual layers in the multilayered film, evaluate their thickness and optical constants. Dispersion dependences for the refractive index were well separated for each type of individual layer, if the thickness was decreased from 315 to 25 nm, and corresponded to those of the single layer. A beveled section of the multilayered film revealed the individual layers that were investigated by atomic force microscopy and nanoindentation to confirm that mechanical properties in multilayered and single layer films are similar.     

AFM线宽测量不确定度的分析与评定

详细分析了P47型原子力显微镜线宽测量不确定度的来源,给出了基于几何形状的线宽测量模型,提出了线宽测量不确定度的评定路线和方法。确定了探针针尖引起的测量不确定度是AFM线宽测量不确定度的主要来源,并对其进行了定量分析。普通Si3N4探针针尖引起的不确定度分量约占线宽总量的5%。     

Fabrication of Periodic Nanostructures Using AFM Tip-Based Nanomachining: Combining Groove and Material Pile-Up Topographies

This paper presents an atomic force microscopy (AFM) tip-based nanomachining method to fabricate periodic nanostructures. This method relies on combining the topography generated by machined grooves with the topography resulting from accumulated pile-up material on the side of these grooves. It is shown that controlling the distance between adjacent and parallel grooves is the key factor in ensuring the quality of the resulting nanostructures. The presented experimental data show that periodic patterns with good quality can be achieved when the feed value between adjacent scratching paths is equal to the width between the two peaks of material pile-up on the sides of a single groove. The quality of the periodicity of the obtained nanostructures is evaluated by applying one- and two-dimensional fast Fourier transform (FFT) algorithms. The ratio of the area of the peak part to the total area in the normalized amplitude–frequency characteristic diagram of the cross-section of the measured AFM image is employed to quantitatively analyze the periodic nanostructures. Finally, the optical effect induced by the use of machined periodic nanostructures for surface colorization is investigated for potential applications in the fields of anti-counterfeiting and metal sensing.     

Contribution of material properties of cellular components on the viscoelastic,stress-relaxation response of a cell during AFM indentation

The close relationship between the mechanical properties of biological cells, namely, elasticity, viscosity, and the state of its disease condition has been widely investigated using atomic force microscopy (AFM). In this study, computational simulation of the AFM indentation is carried out using a finite element (FE) model of an adherent cell. A parametric evaluation of the material properties of the cellular components on the viscoelastic, stress-relaxation response during AFM indentation is performed. In addition, the loading rate, the size of the nucleus, and the geometry of the cell are varied. From the present study, it is found that when comparing the material properties derived from experimental force-deflection curves, the influence of loading rates should be accommodated. It also provides a framework that can quantify the variation of the mechanical property with various stages of malignancy of the cancer cell, a potential procedure for cancer diagnosis.     

Calibration and verification of DEM parameters for dynamic particle flow conditions using a backpropagation neural network

The Discrete Element Method (DEM) requires input parameters to be calibrated and validated in order to accurately model the physical process being simulated. This is typically achieved through experiments that examine the macroscopic behavior of particles, however, it is often difficult to efficiently and accurately obtain a representative parameter set. In this study, a method is presented to identify and select a set of DEM input parameters by applying a backpropagation (BP) neural network to establish the non-linear relationship between dynamic macroscopic particle properties and DEM parameters. Once developed and trained, the BP neural network provides an efficient and accurate method to select the DEM parameter set. The BP neural network can be developed and trained for one or more laboratory calibration experiments, and be applied to a wide range of bulk materials under dynamic flow conditions.     

原子力显微镜探针原位有效参数对线宽测量的修正

针对原子力显微镜（AFM）的线宽和轮廓的精确测量，对AFM探针的原位有效参数进行了定义和表征，提出使用AFM探针的原位有效参数对AFM的线宽测量结果进行修正的模型。采用有效半径和半内角表征AFM探针的复合形状，悬臂轴倾角表征探针的安装状态，设计了具有不同梯形截面的两个表征样板，通过对表征样板进行AFM和扫描电子显微镜（SEM）的比对测量获得了探针的原位有效参数。提出了在线宽测量中，当AFM的扫描轮廓线具有不同的斜度时分别采用的不同的修正公式。采用此公式和探针的原位参数对掩膜板的AFM线宽测量结果进行了修正。     

AFM study of the surface morphology of L-CVD SnO2 thin films

In this paper we present the results of Atomic Force Microscopy (AFM) characterisation of the surface morphology of the L-CVD SnO₂ thin films prepared by L-CVD technology and studied after exposure to air, dry air oxidation, and ion beam profiling. The L-CVD SnO₂ thin films after air exposure have a very smooth surface morphology with an average surface roughness (RMS) smaller than 0.5 nm, and average and maximal grain heights of about 1 and 2 nm, respectively. After dry air oxidation the L-CVD SnO₂ thin films exhibit an average surface roughness (RMS), as well as the average and maximal grain height, increased by one order of magnitude. Finally, after the ion beam profiling the L-CVD SnO₂ thin films exhibit an evidently disordered structure with a lot of craters. These experiments showed that the L-CVD SnO₂ thin films exhibit a very high quality surface morphology, what can be useful for solar cells and gas sensors application.     

Effect of thermal stress on magnetic domain patterns of multiferroic Bi1-xDyxFeO3 thin films

Effect of temperature on magnetic domain structure of Bi_0.7Dy_0.3FeO₃ (BDFO) multiferroic thin films is studied in situ using magnetic force microscopy (MFM). Initially, as the temperature increases the domains start aligning from irregular to more distinct stripe pattern. However, above 250 °C, the domain alignment is disturbed. The systematic change in the domain configuration with temperature, suggests a strong thermal history of the system. The randomness in domain alignment caused above 250 °C is correlated to internal stress developed during ferromagnetic to paramagnetic phase transition occurring in BDFO. Indirect experimental evidence is given to support the explanation based on stress.     

Wavelet uniformity of plasma-processed surface roughness

A technique to characterize the nonuniformity of surface roughness (NSR) is presented. A discrete wavelet transformation (DWT) was used to quantitatively differentiate surface patterns. The technique was evaluated with the data collected from the etching of silicon oxynitride films in a C₂F₆ inductively coupled plasma. 3-D surface images were obtained by using atomic force microscopy. Vertical and lateral NSRs were investigated as a function of process parameters, including radio frequency source power, bias power, and pressure. The NSR data were correlated to experimental measurements of the surface roughness. It is noticeable that for any parameter variations there exist nearly identical NSRs. For each parameter variation, there was at least one specific NSR consistent with the surface roughness measurement. Selected NSRs can be utilized to monitor a variation in NSR and surface roughness simultaneously. Also, NSR may be more stringently optimized by controlling NSRs in a directional fashion.     

Heterogeneity of spiral wear patterns produced by local heating on amorphous polymers

We report on spiral wear patterns produced at constant angular velocity by hot tip atomic force microscopy (HT-AFM) on surfaces of two common amorphous polymers: polystyrene (PS) and polymethylmethacrylate (PMMA). Topography of these patterns is obtained with regular AFM cantilevers. Topography cross-sections taken from a center of each spiral at a given azimuthal angle Θ relate changes of surface corrugation h_corr with tangential velocity v of a thermal cantilever. Polymer wear is characterized by a power law h_corr(v) = α(v/v_max)^−β, which yields a pre-factor α and an exponent β. Below the glass transition temperature T_g, α is polymer specific and β varies weakly between similar conditions and samples. Variations of β are hypothesized to reflect polymer relaxation processes, which are expected to vary only weakly between amorphous polymers. At and above T_g, α approaches initial thermal tip indentation depth within a polymer, β plummets, and a power law relation of h_corr with v diverges. These results are explained by heterogeneous wear around T_g due to a local nature of glass transition. At all studied temperatures, additional wear heterogeneities are found as due to position on the polymer and Θ. Variations of α and β with position on the polymer are found to be only marginally larger then uncertainties of the thermal tip–polymer interface temperature. Variations of α and β with Θ are found to be largely influenced by buckling of thermal cantilevers traveling in a spiral pattern.     

Effects of argon and oxygen flow rate on water vapor barrier properties of silicon oxide coatings deposited on polyethylene terephthalate by plasma enhanced chemical vapor deposition

Plasma polymer coatings were deposited from hexamethyldisiloxane on polyethylene terephthalate (PET) substrates while varying the operating conditions, such as the Ar and O₂ flow rates, at a fixed radio frequency power of 300 W. The water vapor transmission rate (WVTR) of the untreated PET was 54.56 g/m²/day and was decreased after depositing the silicon oxide (SiO_x) coatings. The minimum WVTR, 0.47 g/m²/day, was observed at Ar and O₂ flow rates of 4 and 20 sccm, respectively, with a coating thickness of 415.44 nm. The intensity of the peaks for the Si-O-Si bending at 800-820 cm^− 1 and Si-O-Si stretching at 1000-1150 cm^− 1 varied depending on the Ar and O₂ flow rates. The contact angle of the SiO_x coated PET increased as the Ar flow rate was increased from 2 to 8 sccm at a fixed O₂ flow rate of 20 sccm. It decreased gradually as the oxygen flow rate increased from 12 to 28 sccm at a fixed Ar carrier gas flow rate. The examination by atomic force microscopy revealed a correlation of the SiO_x morphology and the water vapor barrier performance with the Ar and O₂ flow rates. The roughness of the deposited coatings increased when either the O₂ or Ar flow rate was increased.