Method for classifying unknown particles on a surface of a semi-conductor wafer

Abstract

Unknown particles on a surface of a semiconductor wafer are classified by applying a range of particles of known chemical composition and different sizes onto a test wafer, measuring the sizes of a plurality of the particles and spectrally analyzing a makeup of the particles by energy-dispersive x-ray spectroscopy, followed by ascertaining a substantive content therefrom; creating a best-fit curve to the size and substantive content of the particles; measuring the particle size of an unknown particle and recording its spectrum by energy-dispersive x-ray spectroscopy and classifying the unknown particle as the result of a comparison of the size and the substantive content of the unknown particle with the best-fit curve.

Claims

1. A method for classifying unknown particles on a surface of a semiconductor wafer, comprising: applying SiO.sub.2 particles of different sizes in the form of a suspension of the SiO.sub.2 particles in a liquid to a test wafer, ascertaining a size of a plurality of the SiO.sub.2 particles, and recording a spectrum of an energy-dispersive x-ray spectroscopy of the plurality of the SiO.sub.2 particles; subsequently respectively ascertaining a substantive content of the plurality of the SiO.sub.2 particles therefrom, and constructing a best-fit curve of size and substantive content of the plurality of the SiO.sub.2 particles; and ascertaining a particle size of an unknown particle by means of an electron microscope, recording a spectrum of an energy-dispersive x-ray spectroscopy of the unknown particle, and determining therefrom the substantive content of the unknown particle on the semiconductor wafer, and classifying the unknown particle as the result of the comparison of the size and the substantive content of the unknown particle with the best-fit curve; wherein the semiconductor wafer comprises silicon; and wherein the particles are in a size range of 20 nm-200 nm.

2. The method of claim 1, wherein the best-fit curve is a straight line.

3. The method of claim 1, wherein the test wafer comprises silicon.

4. The method of claim 1, wherein the size of the plurality of SiO.sub.2 particles is determined by means of an electron microscope.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a plot of oxygen content versus size of particles on a semiconductor surface.

(2) In the invention, particles of known chemical composition but of different sizes are applied to a surface of a substrate (test wafer). This substrate preferably comprises silicon.

(3) Particles are applied to a test wafer preferably by means of a suspension containing the particles. The particles contained in the suspension preferably have a very limited size range between 15 nm and 3000 nm. It should also be ensured that the extraneous substance content (unwanted impurities) is minimal.

BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENTS

(4) The particle suspension may be applied in analogy to the production of calibrating wafers for scattered laser light systems. The particle suspension in this case may be implemented over the whole area or, preferably, in spots (partial areas). The use of a particle counter may be helpful in this case.

(5) An appropriate basis for correct classification of particles are standard particles of known size and chemical composition. Standardized particles are available on the market with a very wide variety of different chemical compositions, such as aluminum, copper or even yttrium.

(6) The particles used preferably have the same chemical composition but different sizes, with the size distribution of the particles used being very preferably between 20 nm and 100 nm.

(7) After the particles have been applied to a test wafer, the size of a number of particles is ascertained, and a spectrum of an energy-dispersive x-ray spectroscopy is recorded. Both items of information are stored.

(8) The energy-dispersive x-ray spectroscopy may be measured using, for example, a Zeiss Auriga 60 SEM-EDX instrument, 70 mm.sup.2 with EDAX Elite Super EDX Detector, with the following settings: 5 keV, working distance: 6.1 mm, orifice 30 m, EDX integration time: 10 s. However, any other equivalent instrument may also be used.

(9) The sizes of the particles of known chemical composition are preferably determined using an electron microscope (Scanning Electron Microscope).

(10) The size in this context refers to a diameter determined for the particle; equivalently, it may also refer to an area of the particle or to magnitudes derived directly therefrom.

(11) The spectra obtained are preferably analyzed so that they produce the substantive content of the desired substance. This may be done by analyzing the corresponding peak in the spectrum that is characteristic of the chemical substance.

(12) Particular preference is given to using SiO.sub.2 particles, in which case the oxygen content and the size are determined. However, other particles with other, known chemical compositions may also be used for the same purpose.

(13) The substantive content is plotted preferably as a function of the size (diameter, for example), and this plot is used to determine a best-fit curve, which produces a mathematical relationship between substantive content and size in the form of the best-fit curve.

(14) Both the size and the chemical composition of particles of unknown chemical composition are ascertained on the surface of a semiconductor wafer which preferably comprises silicon. In this case, an energy-dispersive x-ray spectroscopy is preferably used. The size of the unknown particles is ascertained preferably using an electron microscope (SEM).

(15) FIG. 1 shows an oxygen measurement (EDX, number of photons in units standardized to 1) obtained from a spectrum of an energy-dispersive x-ray spectroscopy as a function of the measured diameter of spherical particles which are located on a substrate (in this specific case, a silicon semiconductor wafer).

(16) The data points represented by empty circles are values which originate from deliberately applied SiO.sub.2 particles. The shaded region indicates the confidence region in which SiO.sub.2 particles ought to be situated with 95% probability, on the assumption that for these particles the oxygen content shows a linear dependence on the diameter.

(17) The solid circles relate to particles which exhibit no significant dependence between the oxygen content and the particle diameter. The region in which these particles lie in the FIGURE is significantly different from the aforementioned confidence region.

(18) The data point represented by an empty triangle originates from a particle which apparently exhibits a certain dependence between the oxygen content and its diameter; nevertheless, the data point is outside the above-defined confidence region for SiO.sub.2 particles.

(19) FIG. 1 shows, for example, the oxygen content of SiO.sub.2 particles, standardized by means of energy-dispersive x-ray spectroscopy, as a function of the diameter (hollow circles). A linear best-fit curve has been fitted so as to illustrate a proportionality between oxygen content and diameter. It is also possible to ascertain a confidence region, as depicted by the shaded portion, so that 95% of the data points are situated within this region.

(20) The position of the measurement points for particles of unknown chemical composition, shown in FIG. 1 (solid circles, hollow triangle) is able to provide information about the chemical composition, without data regarding the structure of the particles.

(21) The data, indicated for example by solid circles, shows that the oxygen content of these particles is evidently largely independent of their size. The inventors have recognized here that an SiO.sub.2 particle can with high probability be ruled out as the origin of these particles. These particles would therefore be classified as non-SiO.sub.2 particles.

(22) For data points which fall within a confidence region for SiO.sub.2 particles, it can be assumed with a high degree of probability that they are also SiO.sub.2 particles. The classification therefore reads SiO.sub.2 particles.

(23) For data points which apparently neither fall within the confidence range for SiO.sub.2 nor have an oxygen content independent of the size (e.g., the hollow triangle in FIG. 1), it can be assumed either that these are particles with a certain oxygen content that are not spherical, or that these are particles with a possibly intrinsic oxide layer. This information too may be useful if the aim is to determine the origin for the incidence of particles. Particles of this kind are also classified as non-SiO.sub.2 particles.

(24) The inventors have recognized that the procedure described is especially advantageous when it can be assumed that the semiconductor wafer on which the unknown particles are located carries a certain oxide layerfor example, a native oxide layer. This oxide layer influences the energy-dispersive x-ray spectroscopy. Through the additional information on the size of the particle and through the comparison with a best-fit curve for known particles, this measurement error shows only a small negative effect.

(25) A similar approach can be taken with particles of different chemical composition as well, in order, for example, to determine their probable origin.

(26) The above description of illustrative embodiments should be understood as exemplary. The disclosure made therewith enables the skilled person on the one hand to understand the present invention and its attendant advantages, and on the other hand the understanding of the skilled person extends to obvious modifications and alterations of the structures and methods described. The intention is therefore that the scope of protection of the claims is to cover all such modifications and alterations, and also equivalents.