JPH05164798A - Surface potential detection method - Google Patents

Abstract

PURPOSE:To obtain a surface potential detection method which can measure a potential distribution of a sample surface in the order of nm or less with a high resolution and at the same time can detect that of a section part of the sample. CONSTITUTION:A specified bias voltage is applied to a probe electrode 4 and a substrate region 2 of a sample 1 which are placed opposingly on a surface of the sample 1 by a bias circuit 5 and a first scanning is performed so that a tunnel current (It) which flows between them becomes constant, thus enabling a surface roughness of the sample 1 to be measured. Then, a surface roughness of the sample 1 is measured by a second scanning by applying a bias voltage which is equivalent to a band gap to an impurity-implantation region 3 of the sample by a bias circuit 6 and then a difference is obtained by subtracting the surface roughness data which is obtained by the first scanning from that obtained by the second scanning, thus enabling a potential distribution of the sample surface to be detected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting a potential on a surface of a sample, which is performed as a defect detection for reliability evaluation of an electronic device using a very fine circuit, a VLSI, a quantizing function device or the like. ..

[0002]

2. Description of the Related Art In recent years, a method of observing from the surface of an LSI circuit has been adopted, and a liquid crystal analysis technique for detecting a heat generation phenomenon due to carrier transportation, a recombination of carriers, and a light emission phenomenon due to hot electrons have been adopted. Captured emission analysis technology has become mainstream (for example, Uraoka, Tsutsumi, Tanaka, Akiyama SDM90-39 IEICE 19
June 26, 1990).

An example of the above-mentioned conventional emission analysis technique will be described below with reference to the drawings. FIG. 5 is a block diagram showing a configuration of an example of a hot carrier evaluation apparatus using a conventional emission analysis technique. Stage 2 of the optical microscope
The weak light emission 22 emitted from the test pattern or LSI 21 placed at 0 is amplified by the two-dimensional microchannel amplifier 23 and counted as the number of photons. The image processor 24 is capable of superimposing the photon count distribution and the image of the optical microscope, and has a light emission sensitivity of 400 to 800 nm and a spatial resolution of about 1 μm.
Is. The electrons injected into the gate oxide film emit energy when they obtain energy from the electric field and recombine with holes on the silicon surface. When the dielectric breakdown occurs, current is concentrated at the breakdown location and light emission occurs, and the light emission can be observed on the monitor screen 25.

FIG. 6 is a diagram showing the distribution on the light emitting surface (350 × 400 μm ² ). The steep projections in FIG. 6 show the light emission caused by the current density of 100 mA / cm ^2. The diameter shown is about 1 μm.

[0005]

However, in the above-mentioned structure, although the disconnection point and the withstand voltage defective section of the circuit can be roughly identified, the defective section of the ultra-miniaturized circuit pattern can be identified. There is a problem that the resolution that can be specified cannot be exhibited. Further, since the section of the circuit cannot be observed, there is a problem that the identification of the ion implantation profile can only be done by computer simulation, which becomes a decisive factor for manufacturing VLSI.

Therefore, the object of the present invention is to detect the surface potential on the surface of the sample with high resolution on the order of nm or less and to detect the potential distribution on the cross section of the sample. It is to provide a method.

[0007]

According to the surface potential detecting method of the present invention, a predetermined bias voltage is applied to the probe electrode and the substrate region of the sample which are arranged to face the sample surface, and the tunnel current flowing between the two is constant. As a result, the first scan is performed to measure the unevenness of the sample surface, and then a bias voltage corresponding to the band gap is applied to the impurity-implanted region of the sample to measure the unevenness of the sample surface by the second scan. The feature is that the potential distribution on the sample surface is detected by taking the difference between the unevenness data obtained by the second scanning and the unevenness data obtained by the first scanning.

[0008]

According to the structure of the present invention, the physical uneven shape of the sample surface is measured by the first scanning. In the second scan, a bias voltage corresponding to the band gap is further applied to the impurity-implanted region to measure the surface irregularities. Therefore, the irregularity data includes information indicating the physical shape of the sample surface and the electronic state of the surface. It also includes information on the spatial distribution of the surface potential, which is electrical unevenness based on the above. Therefore, the spatial distribution of the sample surface potential can be known by taking the difference between the unevenness data obtained by the second scan and the unevenness data obtained by the first scan. Since the surface potential distribution can be measured as a distribution in the real space by performing the scanning in the two-dimensional direction, it is possible to detect the electrical defect on the sample surface based on this data.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing the configuration of an apparatus for carrying out the surface potential detection method of the present invention.
Reference numeral 2 is a region of the N-type silicon substrate that is the sample 1 where ion implantation is not performed (hereinafter referred to as a substrate region).
A region in which arsenic ions are implanted as impurities (hereinafter referred to as an impurity implantation region), 4 is a probe electrode, 5 is a bias circuit applied to the substrate region 2, and 6 is a bias circuit applied to the impurity implantation region 3.

In order to detect the potential on the surface of the sample 1, first, the bias circuit 6 applied to the impurity-implanted region 3 is opened, and the bias circuit 5 applies 2 V to the substrate region 2. In this state, while maintaining the tunnel current flowing between the probe electrode 4 and the surface of the sample 1 constant, the distance between the probe electrode 4 and the surface of the sample 1 is controlled to set the probe electrode 4 to the sample. The first scanning is performed along the surface of No. 1 (hereinafter, such scanning is abbreviated as a constant current mode). T _{1 in} FIG. 2A is the sample 1 obtained by the first scan.
3 shows a trajectory of the probe electrode 4 when one-dimensionally scanned along the surface of the. This locus T ₁ represents physical irregularities that are the shape of the substrate surface.

Next, a bias circuit 6 further applies a bias to the impurity-implanted region 3, and the probe electrode 4 performs a second scan in the constant current mode. The bias applied to the impurity-implanted region 3 is a voltage corresponding to the band gap of the sample. The locus of the probe electrode 4 obtained by scanning in the one-dimensional direction in this manner is shown by T ₂ in FIG.
The locus T ₂ shows electrical unevenness based on the electronic state of the surface of the impurity-implanted region 3, that is, the spatial distribution of the surface potential, but the physical unevenness that is the shape of the surface of the sample 1 and that the impurities are injected. The surface roughness due to is also shown.

Next, the data showing the physical and electrical irregularities thus obtained is processed. That is, the difference between the unevenness data obtained by the second scanning and the unevenness data obtained by the first scanning is taken to obtain data showing the net potential distribution obtained by subtracting the physical unevenness of the surface. .. As a result, the locus shown by T ₃ in FIG. 2A is obtained.

By performing the above-described scanning and data processing in the two-dimensional direction, the potential distribution in the real space as shown in FIG. 2B can be known. By observing this potential distribution, defects existing on the surface can be detected. FIG. 3 is a band diagram at the time of potential detection, and FIGS. 3A and 3B are band diagrams when a positive bias is applied to the substrate side of the substrate region 2 and the impurity implantation region 3, respectively. Show. FIG. 3A is a band diagram in the substrate region 2 of N-type silicon and simultaneously shows the energy dependence of the electron distribution near the surface. FIG. 2B is a band diagram of the impurity-implanted region 3 in which arsenic is present as an impurity.
It is the area of ⁺ .

The tunnel current It9 tunnels through the space 10 between the probe electrode 4 and the sample surface and flows to the level of the sample 1 surface. In FIG. 5A, since the substrate region 2 is N-type silicon, the density of states of electrons is larger in the vicinity 12 of the valence band edge than in the vicinity 11 of the conduction band edge. N in the impurity implantation region 3
^In the case of the ⁺ type substrate, the density of states 14 at the valence band edge is higher than the density of states 13 near the conduction band edge. When the applied positive bias is set to a value equal to or larger than the band gap of the sample, a tunnel current reflecting the level in the gap and the electronic state near each band edge is transmitted from the probe electrode 4 to the silicon substrate 2.
Or, it flows to the vacant level of 3 and the difference between the N type of the substrate region 2 and the N ⁺ type of the impurity implantation region 3 is clearly detected. Therefore, when the second scan is performed, a voltage corresponding to the band gap of the sample is applied to perform the scan.

FIG. 4A shows the result when the bias is increased to 4 V by the bias circuit 6 and is applied to the impurity implantation region 3. Compared with FIG. 2B, the implantation region applied bias is increased. Therefore, it is recognized that the potential distribution starts to partially overlap with the potential distribution of the adjacent impurity implantation region. Such a dynamic phenomenon was impossible with the conventional evaluation means.

FIG. 4 shows the result when the same scanning as above is performed on the cross section of the sample 1 in which the probe electrode 4 is cleaved.
Shown in (a) and (b). FIG. 3A shows the impurity implantation region 3
Arrangement conditions of 80 keV, 1 × 1
0 ¹⁵ cm ^-2, and shows the locus of the probe electrode 4 when the bias circuit 6 to the impurity implantation region 3 is opened and two-dimensional scanning is performed in the constant current mode. In the same figure (b), arsenic implantation conditions are 160 keV, 1 × 10 ^{15 under the} same measurement conditions.
The electric potential distribution is shown for cm ^-2 . In this case, by changing the implantation energy from 80 keV to 160 keV, the distribution of the implanted arsenic becomes 0.15 in terms of the depth from the silicon surface.
The peak of the arsenic concentration distribution extends from μm to 0.20 μm
It can be seen that there is a shift from the surface of the silicon substrate to the deep part from around 0.5 μm to around 1.0 μm.

Since boron ions have a smaller mass and arsenic ions have a larger mass than silicon atoms, the boron ion implantation profile has an increased probability of wide-angle scattering and exhibits a biased distribution on the surface of the silicon substrate. On the contrary, it is considered that the implantation distribution of arsenic ions has a small wide-angle scattering and is biased toward the inside of the crystal of the silicon substrate. This result is in good agreement with the calculation result estimated by using the probability density function that gives the stationary position distribution of the incident particles, and the surface potential detection method of the present invention makes it possible to obtain the impurities in the substance which were conventionally difficult. It is now possible to directly measure the distribution of atoms displayed in two dimensions in real space.

As described above, according to the embodiment of the present invention, the spatial distribution of the potential can be expressed in a two-dimensional real space in distinction from the physical unevenness on the surface of the sample 1, and the cross section of the sample 1 can be obtained. Since it is possible to detect the potential distribution of a portion, it is possible to detect a micro electrical defect that could not be grasped conventionally.

[0019]

According to the surface potential detecting method of the present invention,
The spatial distribution of the potential can be expressed in a two-dimensional real space in distinction from the physical unevenness of the sample surface, and the potential distribution in the cross section of the sample can also be detected. It has become possible to detect various electrical defects, and more accurately evaluate the reliability of a sample ultrafine circuit or the like.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing the configuration of an apparatus for carrying out the surface potential detection method of the present invention.

FIG. 2A is a diagram showing a trajectory of a probe electrode when the probe electrode is scanned on a substrate, and FIG. 2B is a real space obtained by scanning the probe electrode in a two-dimensional direction. It is a figure which shows the electric potential distribution of.

3A is a band diagram in a substrate region at the time of potential detection, and FIG. 3B is a band diagram in the same impurity implantation region.

FIG. 4A is a diagram showing a trajectory of a probe electrode when scanning a cleaved cross section of a sample, and FIG. 4B is a diagram showing a trajectory of a probe electrode when an implantation condition of impurities is changed. Is.

FIG. 5 is a diagram showing a configuration example of a conventional device for performing hot carrier evaluation by light emission analysis.

6 is a diagram showing an in-plane distribution of light emission in the device shown in FIG.

[Explanation of symbols]

 1 sample 2 substrate region 3 impurity implantation region 4 probe electrode 5 bias circuit applied to substrate region 2 6 bias circuit applied to impurity implantation region 3

Claims (1)

[Claims] 1. A probe electrode is disposed opposite to a sample surface, a bias voltage is applied between the sample surface and the probe electrode, and the probe electrode is arranged along the sample surface so that a tunnel current flowing between the two becomes constant. It is a surface potential detection method that detects the surface potential by measuring the unevenness of the sample surface by scanning with a predetermined bias voltage to the substrate area of the sample and measuring the unevenness of the sample surface by the first scanning. Further, a bias voltage corresponding to the band gap was applied to the impurity-implanted region of the sample, the unevenness of the sample surface was measured by the second scanning, and the unevenness data obtained by the second scanning was obtained by the first scanning. A surface potential detection method, characterized in that the potential distribution on the sample surface is detected by taking the difference from the unevenness data.