WO1993011440A1

WO1993011440A1 - Process and device for the optical determination of charge carrier density variations in semiconductor components

Info

Publication number: WO1993011440A1
Application number: PCT/DE1992/000771
Authority: WO
Inventors: Gerald SÖLKNER
Original assignee: Siemens Aktiengesellschaft
Priority date: 1991-11-29
Filing date: 1992-09-11
Publication date: 1993-06-10
Also published as: DE4139438A1

Abstract

The invention relates to a process in which a laser beam (TB1) from a laser source (LS) is divided into a test beam (TB2) and a reference beam (RBI). The splitting angle (A) and hence a distance (V) between the test and reference beams is matched by altering the acoustic frequency (Δf) of a control signal (S) at an acoutic-optical Bragg cell (BC) to the distance (W) between a target region (D1) and a reference region (D2) allocated to the target region in a semiconductor body (SUB). Reflected portions of the test and reference beams with their phase relation affected by the charge carrier density in the target and reference regions are brought together to form a measuring beam (MB). A measuring signal (M) is detected from the measuring beam with the aid of a photodetector device (PD) and the measuring signal is processed with a reference signal (REF) into an output signal (OUT) which is a measure of the difference in charge carrier density between the target and reference regions.

Description

Method for optically determining charge carrier density differences in semiconductor components and arrangement for carrying it out.

The present invention relates to a method in which a laser beam coming from a laser source is split into a probe beam and a reference beam, in which a probe beam is directed at a target area in a semiconductor body and a reference beam is directed at a reference area in a semiconductor body. in which portions of the probe and reference beam that are reflected and influenced by the charge carrier density in the target and reference area in the phase position are brought together to form a measuring beam and in which a measuring signal is obtained with the aid of a photodetector device, which is a measure of the difference in charge carrier density between Represents target and reference area, and an arrangement with a laser source, a means for splitting the laser beam coming from the laser source into a probe beam and a reference beam, a lens device for aligning the probe beam with a target area and the reference beam a uf a reference area, a means for bringing together the portions of the probe and reference beam which are reflected and influenced in phase by the charge carrier density in the target and reference area to form a measuring beam and a photo detector device for detecting the measuring beam and for forming a measuring signal, the represents a measure of the charge carrier density difference between target and reference area.

A method and an arrangement of this type are known from US Pat. No. 4,758,092 (Heinrich et al.).

The invention has for its object to provide a method and an arrangement of the type mentioned that a rapid adaptation of the distance between a probe and a reference beam to the distance between a target area and a reference area in a semiconductor body, for example the channel width of a field effect transistor, enables.

This object is achieved according to the invention by the features specified in patent claim 1 for the method and by the features specified in patent claim 5 for the associated arrangement.

The advantage of the method according to the invention and the associated arrangement is, in particular, that it is possible to dispense with polarizing the laser beam coming from the laser source.

Claims 2 to 4 are directed to preferred further developments of the method and claims 6 and 7 are directed to preferred configurations of the associated arrangement.

The invention is explained below with reference to the drawing.

The drawing shows an arrangement for carrying out the method according to the invention, wherein a laser beam TB1 coming from a laser source LS can be split into a probe beam TB2 and a reference beam RBI with the aid of an acousto-optical Bragg cell. The Bragg cell BC is advantageously arranged such that the emerging probe beam TB2 is colinear with the laser beam TB1 coming from the laser source and the emerging reference beam RBI includes a splitting angle A with the probe beam TB2. A control signal S is applied to the acousto-optic Bragg cell BC, which comes from a signal generator SG and generates acoustic waves of the frequency Δf in the Bragg cell BC. A Bragg cell consists, for example, of tellurium oxide, a layer for generating acoustic waves being generally provided at one end of the cell and a layer for damping the acoustic waves at the other end. The probe beam TB2 is bundled by a lens device L and as a bundled one Probe beam TB3 directed at a target area D1 at a structuring surface SO of a semiconductor body SUB, the bundled probe beam TB3 from a surface RO facing away from the structuring surface 5 into the semiconductors

~ \ body SUB enters. The reference beam RBI can be bundled with the aid of the lens device L to form a bundled reference beam RB2 and from the surface RO of the semiconductor body SUB facing away from the structuring surface SO

10 reference area D2 belonging to the target area. The lens device can preferably be designed or arranged such that the axis of the bundled probe beam TB3 runs at a distance V parallel to the axis of the bundled reference beam RB2. The target area Dl can, for example, how

15 shown here, the source (drain) region and the reference region D2 can, for example, as shown here, represent the drain (source) region of a field effect transistor T, which has a channel width W and on the structuring surface SO of the semiconductor body SUB is provided. Colinear to the reference

A beam RB can be formed from a reflected part of the reference beam RBI and a reflected part of the probe beam TB2 which is diffracted by the Bragg cell BC by the splitting angle A. The measuring beam MB arrives at a photodetector device PD depending on the measuring beam

25 measurement signal M can be generated. In a commercially available lock-in receiver LIR, the measurement signal M is combined with a reference signal REF, which, for example, also originates from the signal generator SG and always has the double frequency 2Λf of the control signal S of the Bragg cell Output signal OUT

30 processable. The laser beam TB1 coming from the laser source LS has the fixed light frequency f _n and, due to the arrangement of the Bragg cell BC, the probe beam TB2 leaving the Bragg cell and the bundled probe beam TB3 likewise have the fixed light frequency f _n . With the reference beam RBI

35, however, as shown in the drawing, the light frequency f _Q is reduced by the acoustic frequency Δf. This also applies to the bundled reference beam RB2. The measuring beam MB is composed of a combination of the light frequencies f _Q + Δf and f _Q - Δf, since the reflected and diffracted part of the probe Beam TB2 is raised in its light frequency by jf. Furthermore, the phase position of the reflected portion of the probe beam TB2 is dependent on the charge carrier density in the target area D1 and the phase position of the reflected portion of the reference beam RBI is dependent on the charge carrier density in the reference area D2. A change in the phase position of the probe beam or the reference beam causes a change in intensity due to interference, which change can be detected by the photodetector device PD.

In the method according to the invention for the optical determination of charge carrier density differences in semiconductor components, the laser beam TB1 coming from the laser source LS, which should have a light frequency fg that is as constant as possible, is directed onto the acousto-optical Bragg cell BC and diffracted at the acoustic waves propagating in the Bragg cell. The acoustic waves are generated by thickness vibrations of a layer, for example a zinc oxide layer, by applying the control signal S. The splitting angle A between the emerging probe beam TB2 and the emerging reference beam RBI depends on the acoustic frequency of the control signal S as follows:

ak where A is the splitting angle, Ä is the wavelength of the laser light, Δf is the acoustic frequency and v. represents the speed of propagation of the acoustic waves. If, for example, the propagation speed is v _k = 4,260 m / s, the wavelength = 1,300 n, the deflection angle for an acoustic frequency of 10 MHz is 2.5 mrad and 25 mrad at 100 MHz. The splitting angle A can thus be quickly changed within relatively wide limits by the acoustic frequency f of the control signal S, which is not readily possible with purely optical means. A change in the splitting angle A is converted by the lens device L into a corresponding change in the distance V between the bundled probe beam TB3 and the bundled reference beam RB2. This is particularly Another advantage is that the channel width of MOS transistors can vary widely within a semiconductor component, for example, and the distance V can be adapted quickly and easily to the respective channel width W. A part of the bundled probe beam TB3 is modulated in phase by the charge density in the target area D1, reflected back to the Bragg cell BC and diffracted there to the photodetector device PD in the direction of the measuring beam MB. The light frequency of the diffracted reflected probe beam is raised from f _n to f _Q + Λf. Correspondingly, the bundled reference beam RB2 with the light frequency f ₀ - Λf is modulated in phase by the charge carrier density in the reference area D2 and a part of it via the lens device L and the Bragg cell BC from the Bragg cell undeflected in the direction of the measuring beam MB reflects back where it interferes with the reflected portion of the probe beam and collinearly strikes the photodetector device. The photodetector device PD then detects fluctuations in intensity of the beat frequency

f _rt + Λf - (f _n - Δf) = 2Δf

and forms the measurement signal M with the beat frequency 2 / lf. Advantageously, the signal generator SG not only generates the control signal S with the frequency .DELTA.f, but also a reference signal REF with the double frequency .DELTA.f, that is to say the beat frequency, and in addition to the measurement signal M the lock-in-E receiver LIR fed. By means of the lock-in receiver LIR, the reference signal REF of the frequency 2Δf is multiplied by the measurement signal M of the frequency 24f and integrated in time, accordingly a reference signal REF of the frequency 2Δf which is phase-shifted by 90 ^{* is} multiplied by the measurement signal M of the frequency 24f plicated and integrated in time. The lock-in receiver LIR forms two orthogonal components which can be superimposed to form an overall amplitude and which form the output signal OUT. The beat frequency 2Δf is now preferably selected such that the 1 / f noise of the laser source and other components is negligible, as a result of which the sensitivity of the arrangement can be significantly increased. The more target areas the probe beam TB3 scans per time or the more amplitude samples of a time-varying charge carrier density of the same target area are taken, the shorter the integration time of the lock-in receiver must be, whereby the signal / noise ratio decreases. For measurements of high-frequency changes in charge carrier density, pulsed laser sources are to be used according to the invention in order to sample the temporal signal curves with the aid of a sampling method. Sampling methods such as multisampling or coincidence sampling are part of the prior art. The measuring method is not limited to silicon components, but can also be used for components made of other semiconductor materials, such as GaAs. The main pulsed light sources are 15 lasers in the infrared wavelength range: for silicon, for example, gain-switched laser diodes with wavelengths of 1,315 nm and 1,500 nm and mode-coupled Nd: Yag and Nd: YLF solid-state lasers with a wavelength of 1,300 nm. 20

Arrangements are conceivable in which, for example, the probe beam TB2 is not colinear with the laser beam TB1 coming from the laser source LS or in which the reference beam RBI does not have the light frequency f _Q -.Δf but the light frequency 25 f ₀ +.

The method according to the invention is particularly suitable for testing electronic components with contact pads arranged in the middle, because here, due to the shadowing by

30 bond wires, Spider and needles on the block control, elec- ^* tronenstrahl measuring method largely fail. As a rule, a measurement will have to be made from the surface RO facing away from the structuring surface, since the target and reference area mostly from the structuring surface

35. Metallization areas lying above are covered.

Claims

1. Method for determining the charge carrier density of a target area (DI) in a semiconductor body (SUB) with the following method steps:

a) splitting a laser beam (TB1) coming from a laser source (LS) into a probe beam (TB2) and a reference beam (RBI), the laser beam (TB1) coming from the laser source having a fixed light frequency (f _n ) ;

b) aligning the probe beam (TB2, TB3) to the target area (Dl),

c) Aligning the reference beam (RBI, RB2) with a reference area (D2) belonging to the target area by changing the splitting angle (A) between the probe beam (TB2) and the reference beam (RBI) as a function of a control signal ( S);

d) bringing together a reflected portion of the probe beam (TB2) with a reflected portion of the reference beam (RBI) to form a measuring beam (MB), the reflected portion of the probe beam depending on the charge carrier density in the target area and the reflected portion of the reference beam its phase position changes within the semiconductor body (SUB) from the charge carrier density of the reference region;

e) detecting the measuring beam (MB) and forming a measuring signal (M) with the aid of a photodetector device (PD); and

f) Processing the measurement signal (M) in the presence of a reference signal (REF) dependent on the control signal (S) to form an output signal (OUT), the amplitude of which is a measure of the charge carrier density difference between the charge carrier density of the target area and the charge carrier density of the target area represents the territory.

2. The method according to claim 1, characterized in that the laser beam (TB1) coming from the laser source is split in an acousto-optical Bragg cell by diffraction into the probe beam (TB2) and the reference beam (RBI), that by , are generated in the er¬ _r Bragg cell, the control signal (S), acoustic waves, on which the diffraction occurs, and that the split angle (A) by the acoustic frequency (.DELTA.f) is changed the control signal.

3. The method according to claim 1 or 2, dadurchge ¬ indicates that in the case of rapid changes in the charge carrier density difference over time, the laser beam (TB1) coming from the laser source consists of short laser pulses and the time profile of the charge carrier density difference is obtained by sampling -Procedure is scanned.

4. The method according to any one of claims 1 to 3, characterized in that the target region (Dl) represents the drain (source) region of a field effect transistor (T) and the reference region represents the source (drain) region of the field effect transistor (T), the field effect transistor (T) being on the structured surface (SO) of the semiconductor body (SUB), the probe beam (TB3) and the reference beam (RB2) on the surface (RO) facing away from the structured surface (SO) of the semiconductor body (SUB) is directed and that depending on the control signal (S) the reference beam is aligned so that the distance (V) between probe beam (TB3) and reference beam (RB2) is adjusted to the channel width (W) of the field effect transistor (T) becomes.

5. Arrangement for carrying out a method for determining the charge carrier density of a target area (DI) in a semiconductor body (SUB), consisting of

- a laser source with a fixed light frequency (f ** - *.) _»

a means for splitting a light beam coming from the laser source into a probe beam and into a radiation beam. reference beam, in which the splitting angle (A) between the probe beam and the reference beam can be controlled as a function of a control signal (S),

a lens device (L) for aligning and focusing the probe beam (TB2) onto the target area (D1) and for aligning and focusing the reference beam (RBI) onto the reference area (D2),

- A means for bringing together a reflected portion of the probe beam with a reflected portion of the reference beam and their alignment with the photodetector device (PD) and

- A means for processing a measurement signal (M) generated by the photodetector device (PD) in the presence of a reference signal (REF) dependent on the control signal (S).

6. Arrangement according to claim 5, characterized in that the means for splitting the laser beam coming from the laser source includes an acousto-optical Bragg cell (BC) and that the splitting angle (A) between the probe beam and reference beam by the acoustic frequency (f ) of the control signal (S) is controllable.

7. Arrangement according to claim 5 or 6, so that the Bragg cell (BC) is arranged so that the laser beam (TB1) coming from the laser source is colinear with the probe beam (TB2).