Reflection Type Ellipsometer for Measuring Thin Film Phase Difference and Thickness

1. Field of the Invention

This invention relates to a device for rapidly measuring the thickness of thin film deposited on the surface of a body by monitoring polarized light reflected from the surface of the body.

2. Description of the Prior Art

Ellipsometers have heretofore been used to analyse the surface of a body or thin film thereon. Most ellipsometers include a movable portion for mechanically rotating or sliding a polarizing plate or a phase plate. This approach has drawbacks, principally in that rapid measurement is difficult and inconvenient to achieve, particularly during so-called on-line measurement in which the thin film on a body is analysed in the course of a manufacturing process.

This invention provides an ellipsometer by means of which analysis of the surface of a body or of a thin film on a body is achieved under electrical control and without using any movable portion .

According to the present invention, a reflection type ellipsometer for measuring the phase difference possessed by a sample comprises a light source emitting monochromatic light of a particular wavelength; a polarizer for linearly polarizing the light from the light source and having the azimuth thereof disposed at -45.degree. or +45.degree. with respect to the incidence surface (reference surface) of the sample, the sample being disposed at a position for reflecting the linearly polarized light from the polarizer; an analyser disposed on the path of reflected light from the sample and at an azimuth angle of +45.degree. of -45.degree. with respect to the incidence surface or the sample; an adjustable phase difference compensating plate which is disposed on the light path between the polarizer and the analyser and at an azimuth angle of 0.degree. with respect to the reference surface, the phase difference introduced by the compensating plate being preset to a value which eliminates a predetermined or reference phase difference; a double-refractive element for producing a double-refraction phase difference proportional to a voltage applied thereto; a photoelectric converter element for photoelectrically converting the output light from the analyser; and control means for comparing the output voltage from the photoelectric converter element with a predetermined or reference voltage and for applying to the double-refractive element a voltage which causes the double-refractive element to introduce a phase difference corresponding to the amount of deviation of the phase difference of the sample from the predetermined or reference phase difference.

The phase difference compensating plate and the double-refractive element may be disposed at any desired positions between the polarizer and the analyser. Thus, for example, the phase difference compensating plate may be disposed on the light path between the polarizer and the sample, and the double-refractive element may be disposed on the light path between the sample and the analyser. Alternatively, the double-refractive element may be disposed on the light path between the polarizer and the sample, and the phase difference compensating plate may be disposed on the light path between the sample and the analyser. As a further alternative, the phase difference compensating plate and the double-refractive element may be disposed on the light path between the polarizer and the sample. Still alternatively, the phase difference compensating plate and the double-refractive element may be disposed on the light path between the sample and the analyser. The ellipsometer may further comprise means for indicating the value of the voltage applied to the double-refractive element by the control means.

The invention is more fully apparent from the following detailed description thereof taken in conjunction with the accompanying drawings.

FIG. 1 is a schematic view showing an embodiment of the present invention;

FIGS. 2 and 3 A. & B. explain details of the embodiment of the present invention; and

FIGS. 4 A, B, C, illustrate alternative embodiments of the present invention.

Assume that a predetermined or reference phase difference desired of a sample is .DELTA.o, that the actual phase difference of the sample is measured to be .DELTA.o + .DELTA.s, so that .DELTA.s is the deviation or error phase difference, that the phase difference resulting from a double-refractive plate such as KDP is .DELTA.k, and that the phase difference resulting from a phase plate is .DELTA.c. If a compensation method (null method) is used, then when the sum of all the phase differences, .DELTA., becomes 0, the quantity of light entering a light sensing element is 0. This can be formulated as follows:

where .DELTA.o is a reference phase difference such as the average value of the phase differences of samples to be measured and may correspond to a set or control value during a manufacturing process and to a reference value during inspection of products. .DELTA.s, then, represents an amount of variation or deviation from the set value or the reference value.

The foregoing equation may be transformed into:

Thus, the phase difference of the sample may be given by .DELTA.k + .DELTA.c. The above equation may in turn be transformed into:

In order that this equation may be satisfied, there should be these relations: .DELTA.o + .DELTA.c = 0 and .DELTA.s + .DELTA.k = 0. In other words, the phase difference .DELTA.o of the sample, as the set value or the reference value, should first be compensated for by the phase difference of a phase plate, and then the amount of deviation .DELTA.s should be compensated for by the phase difference of the double-refractive plate such as KDP. FIG. 1 shows an embodiment of the present invention which satisfies the above-described conditions.

In FIG. 1, light from a light source L is transformed into light of a particular wavelength by an interference filter F. Represented by P is a polarizer which, as illustrated in FIG. 2, is disposed at an azimuth of -45.degree. with respect to the incidence surface (the angles hereinafter mentioned refer to the azimuth angles of respective elements disposed with respect to the incidence surface as the reference). R designates an adjustable phase plate such as Soleil-Barinet's compensator disposed at 0.degree.. S designates a sample also disposed at 0.degree.. Q is a double-refractive substance such as KDP, disposed at 0.degree.. KDP, as is well known, is a uniaxial crystal cut perpendicularly to the Z-axis thereof, and if a voltage is applied thereto, the KDP acts as a biaxial crystal which creates a double-refraction phase difference proportional to the voltage applied. Designated by A is an analyser which, as depicted in FIG. 2, is disposed at an aximuth angle of 45.degree.. PM is a light-sensing element such as photomultiplier. Designated by AM is an amplifier for amplifying the output of the light-sensing element, and D is a detector. The detector D serves to compare, with a reference voltage, for example, 0 volts, the output from the light-sensing element PM which senses the light passed through the analyser A, and to apply to the KDP a voltage for nullifying the quantity of light entering the light-sensing element PM. The output of the amplifier AM is connected to the input of the detector D, and the output of detector D is connected to the KDP and to a recording meter IN.

In the above-described construction of an embodiment of the present invention, the monochromatic light passed through the interference filter F is transformed by the polarizer P into linearly polarized light having an azimuth angle of -45.degree.. Phase plate R introduces to this linearly polarized light a phase difference .DELTA.c, which is equal to the set value or the reference value .DELTA.o desired of the sample, and thus phase plate R provides elliptically polarized light corresponding to that phase difference. The light is then reflected by the sample S and undergoes the phase variation .DELTA.o + .DELTA.s actually introduced by the sample. This phase variation .DELTA.o + .DELTA.s is illustrated in FIG. 3A. The phase difference .DELTA.o, which is the set value or the reference value of the sample, is eliminated by virtue of the phase variation .DELTA.c introduced by the phase plate R (since .DELTA.o + .DELTA.c = 0) but the phase difference .DELTA.s, representing the amount of variation from the set value or the reference value, is not compensated for and so still is present, as illustrated in FIG. 3B. Such light passes through the KDP, which initially has 0 volts applied thereto, and further through the analyser A to the light-sensing element PM, which senses the light. Since the quantity of light from the analyser A is in proportion to sin.sup.2 (.DELTA.s/2), the output from the light-sensing element PM is likewise in proportion to sin.sup.2 (.DELTA.s/2). This output is amplified by the amplifier AM and enters the detector D. The output of the detector D provides a voltage which is used to nullify the quantity of light entering the light-sensing element PM. This output is recorded by the recording meter IN while, at the same time, it is applied to the KDP. Upon application of such voltage thereto, the KDP introduces a phase difference .DELTA.k which compensates for the phase difference .DELTA.s of the light which is then passing therethrough; i.e., the light from the KDP is made into completely linearly polarized light whose direction of vibration is orthogonal to the direction of vibration of the analyser A. Thus, .DELTA.s + .DELTA.k = 0, and the phase difference of the entire system is (.DELTA.o + .DELTA.c) + (.DELTA.s + .DELTA.k) = 0. From this relation, the phase difference .DELTA.o + .DELTA.s of the sample may be obtained as -(.DELTA.c + .DELTA.k).

If a different sample is to be measured, the phase difference .DELTA.c which is the set value or the reference value necessarily becomes different, but measurement can be carried out without any inconvenience by adjusting the phase plate R to that value prior to the measurement.

FIG. 4 shows another embodiment of the present invention. In the embodiment previously described with reference to FIG. 1, the adjustable phase plate R, the sample and the KDP are arranged in the named order, whereas any desired order of arrangement may be adopted between the polarizer P and the analyser A. Thus as illustrated in FIG. 4A, the double-refractive substance Q, which may be KDP, may be between the polarizer and the sample, while the variable phase plate R is between the sample and the analyser. Likewise, as shown in FIG. 4B, the double-refractive substance Q and the adjustable phase plate R may both be between the polarizer and the sample. Similarly, as illustrated in FIG. 4C, the double-refractive substance Q and the adjustable phase plate R may both be between the sample and the analyser.

Further, in the embodiment of FIG. 1, the polarizer is disposed at an azimuth of -45.degree. with respect to the incidence surface and the analyser is disposed at an azimuth of +45.degree. with respect to the incidence surface, but this relationship may be inverted. In other words, the polarizer may be at the azimuth of +45.degree. and the analyser at the azimuth of -45.degree., both with respect to the incidence surface.

According to the present invention, as has been described above, no mechanically movable portion is required, and this enables quick response and simplification of the device as well as enhanced reliability and lower cost. Also, compensation for .DELTA.o + .DELTA.s was heretofore effected by applying to the KDP a high voltage, e.g., several kilovolts, and this in turn involved not only the necessity for electrical parts capable of handling the high voltage but also the need to increase the thickness of the KDP, whereas such disadvantages are eliminated according to the present invention in which only a phase difference of a few degrees is compensated for by the KDP.