Metrology of gaseous air pollutants

The need for mutual recognition of accurate measurement results made by competent laboratories has been very widely accepted at the international level e.g., at the World Trade Organization. A partial solution to the problem was made by the International Committee for Weights and Measures (CIPM) in setting up the Mutual Recognition Arrangement (CIPM MRA), which was signed by National Metrology Institutes (NMI) around the world. The core idea of the CIPM MRA is to have global arrangements for the mutual acceptance of the calibration certificates of National Metrology Institutes. The CIPM MRA covers all the fields of science and technology for which NMIs have their national standards. The infrastructure for the metrology of the gaseous compounds carbon monoxide (CO), nitrogen monoxide (NO), nitrogen dioxide (NO2), sulphur dioxide (SO2) and ozone (O3) has been constructed at the national level at the Finnish Meteorological Institute (FMI). The calibration laboratory at the FMI was constructed for providing calibration services for air quality measurements and to fulfil the requirements of a metrology laboratory. The laboratory successfully participated, with good results, in the first comparison project, which was aimed at defining the state of the art in the preparation and analysis of the gas standards used by European metrology institutes and calibration laboratories in the field of air quality. To confirm the competence of the laboratory, the international external surveillance study was conducted at the laboratory. Based on the evidence, the Centre for Metrology and Accreditation (MIKES) designated the calibration laboratory at the Finnish Meteorological Institute (FMI) as a National Standard Laboratory in the field of air quality. With this designation, the MIKES-FMI Standards Laboratory became a member of CIPM MRA, and Finland was brought into the internationally-accepted forum in the field of gas metrology. The concept of ‘once measured - everywhere accepted’ is the leading theme of the CIPM MRA. The calibration service of the MIKES-FMI Standards Laboratory realizes the SI traceability system for the gas components, and is constructed to enable it to meet the requirements of the European air quality directives. In addition, all the relevant uncertainty sources that influence the measurement results have been evaluated, and the uncertainty budgets for the measurement results have been created.

Optical frequency metrology with an Rb-stabilized ring-cavity resonator-study of cavity-dispersion errors

We have developed a technique to measure the absolute frequencies of optical transitions by using an evacuated Rb-stabilized ring-cavity resonator as a transfer cavity. The absolute frequency of the Rb D-2 line (at 780 nm) used to stabilize the cavity is known and allows us to determine the absolute value of the unknown frequency. We study wavelength-dependent errors due to dispersion at the cavity mirrors by measuring the frequency of the same transition in the Cs D-2 line (at 852 nm) at three cavity lengths. The spread in the values shows that dispersion errors are below 30 kHz, corresponding to a relative precision of 10(-10). We give an explanation for reduced dispersion errors in the ring-cavity geometry by calculating errors due to the lateral shift and the phase shift at the mirrors, and show that they are roughly equal but occur with opposite signs. We have earlier shown that diffraction errors (due to Guoy phase) are negligible in the ring-cavity geometry compared to a linear cavity; the reduced dispersion error is another advantage. Our values are consistent with measurements of the same transition using the more expensive frequency-comb technique. Our simpler method is ideally suited for measuring hyperfine structure, fine structure, and isotope shifts, up to several hundreds of gigahertz.

Whole wafer magnetostriction metrology for magnetic films and multilayers

The requirements for metrology of magnetostriction in complex multilayers and on whole wafers present challenges. An elegant technique based on radius of curvature deformation of whole wafers in a commercial metrology tool is described. The method is based on the Villari effect through application of strain to a film by introducing a radius of curvature. Strain can be applied tensilely and compressively depending on the material. The design, while implemented on 3'' wafers, is scalable. The approach removes effects arising from any shape anisotropy that occurs with smaller samples, which can lead to a change in magnetic response. From the change in the magnetic anisotropy as a function of the radius, saturation magnetostriction ?s can be determined. Dependence on film composition and film thickness was studied to validate the radius of curvature approach with other techniques. ?s decreases from positive values to negative values through an increase in Ni concentration around the permalloy composition, and ?s also increases with a decrease in film thickness, in full agreement with previous reports. We extend the technique by demonstrating the technique applied to a multi-layered structure. These results verify the validity of the method and are an important step to facilitate further work in understanding how manipulation of multilayered films can offer tailored magnetostriction.

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.

MSC-clustering and forward stepwise regression for virtual metrology in highly correlated input spaces

Increasingly semiconductor manufacturers are exploring opportunities for virtual metrology (VM) enabled process monitoring and control as a means of reducing non-value added metrology and achieving ever more demanding wafer fabrication tolerances. However, developing robust, reliable and interpretable VM models can be very challenging due to the highly correlated input space often associated with the underpinning data sets. A particularly pertinent example is etch rate prediction of plasma etch processes from multichannel optical emission spectroscopy data. This paper proposes a novel input-clustering based forward stepwise regression methodology for VM model building in such highly correlated input spaces. Max Separation Clustering (MSC) is employed as a pre-processing step to identify a reduced srt of well-conditioned, representative variables that can then be used as inputs to state-of-the-art model building techniques such as Forward Selection Regression (FSR), Ridge regression, LASSO and Forward Selection Ridge Regression (FCRR). The methodology is validated on a benchmark semiconductor plasma etch dataset and the results obtained are compared with those achieved when the state-of-art approaches are applied directly to the data without the MSC pre-processing step. Significant performance improvements are observed when MSC is combined with FSR (13%) and FSRR (8.5%), but not with Ridge Regression (-1%) or LASSO (-32%). The optimal VM results are obtained using the MSC-FSR and MSC-FSRR generated models. © 2012 IEEE.

On Regression Methods for Virtual Metrology in Semiconductor Manufacturing

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. The prediction models for VM can be from a large variety of linear and nonlinear regression methods and the selection of a proper regression method for a specific VM problem is not straightforward, especially when the candidate predictor set is of high dimension, correlated and noisy. Using process data from a benchmark semiconductor manufacturing process, this paper evaluates the performance of four typical regression methods for VM: multiple linear regression (MLR), least absolute shrinkage and selection operator (LASSO), neural networks (NN) and Gaussian process regression (GPR). It is observed that GPR performs the best among the four methods and that, remarkably, the performance of linear regression approaches that of GPR as the subset of selected input variables is increased. The observed competitiveness of high-dimensional linear regression models, which does not hold true in general, is explained in the context of extreme learning machines and functional link neural networks.

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.