Predictive Maintenance for Improved Sustainability - An Ion Beam Etch Endpoint Detection System Use Case

In modern semiconductor manufacturing facilities maintenance strategies are increasingly shifting from traditional preventive maintenance (PM) based approaches to more efficient and sustainable predictive maintenance (PdM) approaches. This paper describes the development of such an online PdM module for the endpoint detection system of an ion beam etch tool in semiconductor manufacturing. The developed system uses optical emission spectroscopy (OES) data from the endpoint detection system to estimate the RUL of lenses, a key detector component that degrades over time. Simulation studies for historical data for the use case demonstrate the effectiveness of the proposed PdM solution and the potential for improved sustainability that it affords.

Extreme learning machines for virtual metrology and etch rate prediction

Virtual metrology (VM) aims to predict metrology values using sensor data from production equipment and physical metrology values of preceding samples. VM is a promising technology for the semiconductor manufacturing industry as it can reduce the frequency of in-line metrology operations and provide supportive information for other operations such as fault detection, predictive maintenance and run-to-run control. Methods with minimal user intervention are required to perform VM in a real-time industrial process. In this paper we propose extreme learning machines (ELM) as a competitive alternative to popular methods like lasso and ridge regression for developing VM models. In addition, we propose a new way to choose the hidden layer weights of ELMs that leads to an improvement in its prediction performance.

Frequency coupling in dual frequency capacitively coupled radio-frequency plasmas

An industrial, confined, dual frequency, capacitively coupled, radio-frequency plasma etch reactor Exelan®, Lam Research has been modified for spatially resolved optical measurements. Space and phase resolved optical emission spectroscopy yields insight into the dynamics of the discharge. A strong coupling of the two frequencies is observed in the emission profiles. Consequently, the ionization dynamics, probed through excitation, is determined by both frequencies. The control of plasma density by the high frequency is, therefore, also influenced by the low frequency. Hence, separate control of plasma density and ion energy is rather complex.

Fabrication of Sub-micron Active Layer SSOI Substrates using Ion Splitting and Wafer Bonding Technologies

This paper reports the fabrication of SSOI (Silicon on Silicide On Insulator) substrates with active silicon regions only 0.5mum thick, incorporating LPCVD low resistivity tungsten silicide (WSix) as the buried layer. The substrates were produced using ion splitting and two stages of wafer bonding. Scanning acoustic microscope imaging confirmed that the bond interfaces are essentially void-free. These SSOI wafers are designed to be employed as substrates for mm-wave reflect-array diodes, and the required selective etch technology is described together with details of a suitable device.

Chemical etching of optical fibre tips - experiment and model

The formation of chemically etched fibre tips for use in optical scanning probe microscopy is addressed. For tips formed at a cleaved fibre end in the bulk of a buffered HF acid solution the morphological features (tip height, cone angle) are found to depend strongly on the temperature and etchant composition. The tip formation process is analysed and explained in terms of a simple model in which the only pertinent physical parameters are the fibre core diameter and etch rates of the fibre core and cladding. The etch rates are determined in separate experiments as a function of temperature (in the range 24-50 degreesC) for etchant solutions of de ionised water: 50% HF acid: 40% NH4F in the volume ratio 1 : 1 : X for X=2, 4 and 6, and used in the model to yield a correct description of the experimental tip cone angles. The model is successfully extended to the intriguing case of negative tip formation which initiates in a normal, positive tip structure. By contrast, tip formation in the meniscus region of a bare fibre/etchant/organic solvent system is found to be independent of etchant composition and temperature. (C) 2000 Elsevier Science B.V. All rights reserved.

Sub-mm wet etched linear to circular polarization fss based polarization converters

This communication investigates the potential for fabrication of micromachined silicon sub-millimeter wave periodic arrays of freestanding slot frequency selective surfaces (FSS) using wet etch KOH technology. The vehicle for this is an FSS for generating circularly polarized signals from an incident linearly polarized signal at normal incidence to the structure. Principal issues and fabrication processes involved from the initial design of the core FSS structures to be made and tested through to their final testing are addressed. Measured and simulated results for crossed and ring slot element shapes in single and double layer polarization convertor structures are presented for sub-mm wave operation. It is shown that 3 dB axial ratio (AR) bandwidths of 21% can be achieved with the one layer perforated screen design and that the rate of change is lower than the double layer structures. An insertion loss of 1.1 dB can be achieved for the split circular ring double layer periodic array. These results are shown to be compatible with the more specialized fabrication equipment dry reactive ion etching approach previously used for the construction of this type of structure. © 2011 IEEE.

Max separation clustering for feature extraction from Optical Emission Spectroscopy Data

This paper proposes max separation clustering (MSC), a new non-hierarchical clustering method used for feature extraction from optical emission spectroscopy (OES) data for plasma etch process control applications. OES data is high dimensional and inherently highly redundant with the result that it is difficult if not impossible to recognize useful features and key variables by direct visualization. MSC is developed for clustering variables with distinctive patterns and providing effective pattern representation by a small number of representative variables. The relationship between signal-to-noise ratio (SNR) and clustering performance is highlighted, leading to a requirement that low SNR signals be removed before applying MSC. Experimental results on industrial OES data show that MSC with low SNR signal removal produces effective summarization of the dominant patterns in the data.

Glued-in Rods

Glued-in rods (GiR) have been successfully used for both constructing new and strengthening existing timber structures. The research and development of connecting and strengthening timber structural elements with GiR has been going on since the 1980s. However, agreement regarding design criteria for these applications has not been reached. Today, some few technical approvals for specific adhesives suitable to GiR exist, but an approach for the design of connections or reinforcement with GiR has not been included in the European design code EN 1995 so far. Therefore, it is desired to gather the current state of knowledge to enable application in practice of the existing and documented knowledge and experience. This state-of-the-art review (STAR) summarises results from research done regarding connections and reinforcement with GiR. The review considers manufacturing methods, mechanisms and parameters governing the performance and strength of GiR, theoretical approaches and existing design recommendations. For GiR applied as reinforcement similar rules and requirements apply as for GiR being used as connectors.

Virtual metrology enabled early stage prediction for enhanced control of multi-stage fabrication processes

Semiconductor fabrication involves several sequential processing steps with the result that critical production variables are often affected by a superposition of affects over multiple steps. In this paper a Virtual Metrology (VM) system for early stage measurement of such variables is presented; the VM system seeks to express the contribution to the output variability that is due to a defined observable part of the production line. The outputs of the processed system may be used for process monitoring and control purposes. A second contribution of this work is the introduction of Elastic Nets, a regularization and variable selection technique for the modelling of highly-correlated datasets, as a technique for the development of VM models. Elastic Nets and the proposed VM system are illustrated using real data from a multi-stage etch process used in the fabrication of disk drive read/write heads. © 2013 IEEE.