Overlay Measurement Research Articles

Qualification of tools for overlay measurement on processed wafers

In this paper we present two new methodologies to verify the accuracy of a measurement tool. The dependence of the measurements on substrates and the correlation with the same patterns with the same misalignment is shown. Theusefulness and reliability of the methods is proved by examples on commercially available equipment.

Overlay measurement using the low voltage scanning electron microscope

Accurate and precise overlay metrology is essential for the successful fabrication of integrated circuits with sub-micron critical dimensions. We have investigated the use of the low voltage scanning electron microscope (SEM) to measure overlay at the device level and to calibrate the measurements made with optics-based systems. A study of the fundamental aspects of SEM overlay metrology was made on resist/etched silicon as well as resist/silicon test wafers. It was found to be critical to optimize the SEM accelerating voltage, scanning technique, mark design, detector design, and measurement algorithm. 3σ measurement precisions better than 0.010 μm were easily obtained on a variety of samples. SEM tool-induced shift 1 (TIS), when present, was found to be of the same order as the 3σ precision and was negligible in the absence of charging. Finally, a comparison of automated optically-based measurements to SEM overlay measurements was made on advanced integrated circuit production wafers. Agreement was quite good with the average difference less than 0.01 μm and 3σ variation of the difference better than 0.04 μm for marks in resist.

Submicron metrology in the semiconductor industry

Techniques used to measure critical dimensions and overlay misregistration on integrated circuits during manufacturing are reviewed. The current needs for accuracy and reproducibility of these measurements in the semiconductor industry are discussed. The measurement capabilities of three types of instruments: electrical testers, scanning electron microscopes and sectioning optical microscopes are compared. Measurements made on sectioning optical microscopes are emphasized because of their ability to independently image different focus levels when making critical dimension measurements. A comparison of overlay measurements made with an electrical tester and a sectioning optical microscope is also presented.

Accuracy of overlay measurements: tool and mark asymmetry effects

Definitions of centerline and centerline overlay (O/L) are built on the implicit assumptions of certain symmetries. Verification of these symmetries, and the use of redundancies built into the design of O/L measurement marks, can estimate the relative accuracy of O/L measurements, even in the absence of certified reference materials (standards). Verification carried out as of paired comparison is shown to be very effective; in addition, it constitutes the basis for data culling. Several sources of O/L measurement inaccuracy, associated with O/L measurement tools and with O/L measurement marks, are pointed out and some are illustrated using modeling. A simple measure of tool-related inaccuracy, tool-induced shift (TIS), is proposed and utilized as a performance estimate of O/L measurement tools relative to accuracy. Inaccuracies of O/L measurements, caused by O/L measurement tools and marks, on the order of 100 nm, are observed and their causes identified. Examples are given of significant improvements in accuracy. A roadmap toward accurate O/L measurement is outlined. This approach is also applicable to alignment systems.

Multilevel CD/overlay metrology using a real-time confocal scanning optical microscope

Linearity and short-term repeatability have been studied for critical dimension (CD) and overlay measurements using the Real-Time Confocal Scanning Optical Microscope (RSOM). Short-term repeatability of 15nm (three-sigma) has been demonstrated for submicron CD and overlay. CD linearity was measured in the 0.5μm to 1.3μm region using a top-down SEM for reference. Overlay linearity was measured using electrical overlay structures with known offsets in the range -0.2μm to +0.2μm. Response surface methodology was used to determine the optimum focus for achieving repeatable overlay measurements. The results were used to compare the performance of various overlay structures.

A new signal processing method for overlay and grid characterization measurements

A recursive method is described for the automatic computation of the centroid of a one-dimensional, quasi-symmetrical sampled signal. This method, which applies properties of the phase of the signal spectrum, is shown to present some advantages when compared with equivalent techniques implemented in the signal space (such as correlation or convolution): rapidity and resolution (far better than the sampling step in the signal space). This method is also shown to apply to the computation of the temporal shift between two identical signals at low noise level.Beschrieben wird ein rekursiver Algorithmus zur Bestimmung des Zentrums eines abgestasteten eindimensionalen quasisymmetrischen Signals. Der Algorithmus verwendet die Eigenschaften des Phasenspektrums. Gegenüber Methoden, die direkt im Signalbereich arbeiten, wie Korrelations- oder Faltungsmethoden, bietet dieser Algorithmus einige Vorteile bezüglich der Schnelligkeit und des Auflösungsvermögens, das wesentlich feiner ist als die Größe des Abtastintervalls im Signalbereich. Wie weiterhin gezeigt wird, ist das Verfahren auch dazu geeignet, bei geringen Rauschpegeln die Zeitverzögerung zwischen zwei identischen Signalen zu berechnen.On présente un algorithme recursif permettant la déteŕmination du centre d'un signal échantillonné quasi-symétrique, monodimensionnel. Cet algorithme utilise les propriétés de la phase du spectre du signal. Il présente plusieurs avantages par rapport aux méthodes travaillant dans l'espace du signal (corrélation, convolution): rapidité et précision (la résolution de l'algorithme est trés supérieure au pas d'échantillonnage du signal). On montre ensuite que le même algorithme peut être appliqué au calcul du décalage temporel entre deux signaux de forme identique, au bruit près.

Recent developments in X-ray lithography systems

High resolution, high throughput x-ray lithography systems are being developed to meet future needs for submicron integrated circuit fabrication. An advanced step-and-repeat x-ray system, under development at the Perkin-Elmer Corporation, is described. The critical technologies, precision alignment and high-power x-ray source, are explored. To achieve half-micron or better resolution, overlay accuracy of at least 0.1 micron is required. This is achieved by a proximity system that controls alignment in six degrees of freedom on a field-by-field, real-time basis. Alignment techniques are described and overlay measurements presented for Boron Nitride mask-to-wafer levels. The demands placed on the x-ray source are great. High power is required to achieve practical levels of throughput, and a finite x-ray emitting area is required to achieve a small penumbra blur. The source developed achieves a system resolution of 0.2 micron or less, with a throughput of 40 wafer levels per hour. Experimental results on this source are shown.

Overlay measurements for x-ray lithography

The metrology in submicron lithography is an increasingly difficult task. Especially the control of the stability of the x-ray masks, which needs resolution better than 0.1 μm over a large area, supplies a lot of problems. Some of them can be investigated by double exposures evaluated with overlay measurement methods. We use a method that needs a minimum of process steps. Both exposures are made within the same resist. The resist patterns are measured directly. No additional processing step other than lithography and resist developing are necessary. We built an experimental alignment system for x-ray lithography in a vacuum chamber, so we can expose in vacuum or in a He atmosphere. A swivel-mounted microscope allows the observation of the alignment marks on the mask and on the wafer. With six motordriven screws we can adjust wafer to mask. With another setup it is possible to make tilted exposures. After the double exposure we measure the relative position of special metrology marks with light optics on an accuracy of 30 nm (3δ value). The assembly consists of a LEITZ linewidth measuring microscope, a high precision motordriven x-y-z table, and a microcomputer. The automatic measuring procedure is performed under the control of the computer. With these measuring and evaluation methods we could measure the alignment accuracy, and changes of the x-ray masks due to heating by the synchrotron radiation. It is also possible to compare the accuracy of two x-ray masks and to control the stability of the exposure system.

Overlay in Lithography

Advances in lithography rely largely on the capability of reducing overlay errors, which in turn depends on the capability to make two-dimensional overlay measurements. This paper describes a simple and accurate method of determining singular overlay errors of step-and-repeat exposure systems with a precision of ±0.01 µm (standard deviation).

Interferometric surface mapping with variable sensitivity

In the photolithographic process, presently employed for the production of integrated circuits, sets of correlated masks are used for exposing the photoresist on silicon wafers. Various sets of masks which are printed in different printing tools must be aligned correctly with respect to the structures produced on the wafer in previous process steps. Even when perfect alignment is considered, displacements and distortions of the printed wafer patterns occur. They are caused by imperfections of the printing tools or/and wafer deformations resulting from high temperature processes. Since the electrical properties of the final integrated circuits and therefore the manufacturing yield depend to a great extent on the precision at which such patterns are superimposed, simple and fast overlay measurements and flatness measurements as well are very important in IC-manufacturing. A simple optical interference method for flatness measurements will be described which can be used under manufacturing conditions. This method permits testing of surface height variations by nearly grazing light incidence by absence of a physical reference plane. It can be applied to polished surfaces and rough surfaces as well.

Urban Growth Modelling And GIS

The purpose of this paper is to explore the application of mathematical models and numerical modelling concerning the growth of urban expansion areas. The models are tested on four case studies located in the 2nd Metropolitan Area of Portugal around the town of Oporto. An introduction to the approach taken is presented which can be summarised very briefly in three phases. The 1st phase deals with the selection of areas in terms of planning criteria and objectives of land-use plans evaluation as well as data availability. The 2nd phase concerns the delimitation of boundaries, preparation of aerial photographs, base maps, plans and finally, the actual computer mapping of the urban growth. The 3rd phase consists in overlay analysis and measurements of the aerial photography computer mapping performed in digital form associated to alpha-numerical databases. The results show that for the study areas of Metropolitan Area of Oporto, of around 20 Km each, the growth parameters for exponential and logistic curves are more similar according to the time period than depending on other factors, such as: i) existence of land-use plans for urban expansion; ii) classical mathematical models applied to growth phenomena, namely population. A second set of results allowed the conclusion that recent levels of external accessibility seem more determinant than levels of past urban growth. Nevertheless, the above mentioned results cannot be derived for other areas or for other periods as the available statistical data does not allow such extrapolations. But the importance of the adopted approach relies on the relatively low-cost, speed of results production as well as possibilities for generalisation and user-application. Transactions on Information and Communications Technologies vol 18, © 1998 WIT Press, www.witpress.com, ISSN 1743-3517