JP2015185811A - Evaluation method of semiconductor substrate - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an evaluation method of a semiconductor substrate capable of evaluating a state of a crystal defect of a semiconductor substrate even when there is no characteristic light emission.SOLUTION: The evaluation method of a semiconductor substrate having a crystal defect to which a defect recovery heat treatment is applied for recovering the crystal defect includes steps of: measuring intensity of band end light emission of the semiconductor substrate by using a luminescence method; and evaluating a degree of recovery of crystallinity of the semiconductor substrate on the basis of a result of the measurement.

Description

  The present invention relates to a method for evaluating a semiconductor substrate.

  Miniaturization has been advanced for higher performance of LSI, and the gate length has been shortened. Since the gate length is shortened, it is necessary to reduce the diffusion depth of the source / drain regions. For example, in the case of a device (transistor) having a gate length of about 30 nm, the diffusion depth of the source / drain portion is about 15 nm, and very shallow diffusion is required.

Conventionally, ion implantation is used to form such a diffusion layer. For example, a method of implanting B ⁺ or BF ₂ ⁺⁺ at a very low acceleration of 0.2 to 0.5 keV is used. However, the resistance of the ion-implanted atoms cannot be lowered as it is. In the ion-implanted region, point defects such as interstitial silicon and atomic vacancies are generated in the silicon substrate.

  For this reason, after ion implantation, annealing is performed to activate the atoms (reduce the resistance) and recover defects, but this annealing diffuses the implanted ions and expands the impurity distribution. Furthermore, a phenomenon is known in which impurity diffusion is accelerated not only by annealing but also by point defects caused by ion implantation.

  Even in consideration of the diffusion spread described above, in order to be able to form a shallow pn junction having a depth of 15 nm or less and a width of 10 nm or less in the lateral direction directly under the ion implantation mask, high energy can be applied in a very short time. An annealing method for irradiation has been studied and adopted (for example, see Patent Document 1).

Examples of the annealing method include annealing using a flash lamp in which a rare gas such as xenon is sealed. This lamp irradiates high energy of several tens of J / cm ² or more as pulsed light of 0.1 to 100 milliseconds. Therefore, it is possible to activate the impurity distribution formed by ion implantation with almost no change.

  However, since this high energy is used, it is considered that the thermal stress in the silicon substrate increases and damage such as cracking or slipping of the silicon substrate occurs, and this is actually being studied.

  For example, in Patent Document 2, in order to form a shallow impurity diffusion region without causing damage in a semiconductor substrate, a substance that becomes an acceptor or a donor with respect to the semiconductor substrate and an acceptor or a donor with respect to the semiconductor substrate are disclosed. It is described that a substance having a substance that does not have to be implanted into a semiconductor substrate.

  As described above, crystal defects generated by ion implantation or the like in the semiconductor substrate are recovered by heat treatment. At this time, the recovery state of the crystal defects is evaluated using, for example, a luminescence method. Examples of the luminescence method include cathodoluminescence using an electron beam and photoluminescence method using a laser beam. Both of these excite carriers using an electron beam or laser light, and capture the process in which these carriers lose their energy as phonons as light emission. High sensitivity. If a level exists in the band during the energy disappearance process, light emission corresponding to the energy level can be obtained, and the presence of a defect can be detected with high sensitivity.

JP 2005-347704 A JP 2009-027027 A

  However, as described above, the defect density to be introduced is reduced by improving the ion implantation technique and the annealing technique, and light emission lines (characteristic light emission) caused by crystal defects obtained by using a conventional luminescence method are observed. With the method, it has become difficult to evaluate the state of crystal defects in the semiconductor substrate.

  The present invention has been made in view of the above problems, and an object of the present invention is to provide a semiconductor substrate evaluation method capable of evaluating the state of crystal defects in a semiconductor substrate without characteristic light emission.

In order to achieve the above object, the present invention provides a method for evaluating a semiconductor substrate that has been subjected to a defect recovery heat treatment for recovering the crystal defect with respect to a semiconductor substrate having a crystal defect,
A step of measuring the intensity of the band edge emission of the semiconductor substrate using a luminescence method;
And a step of evaluating the degree of crystallinity recovery of the semiconductor substrate based on the result of the measurement.

  With such a method for evaluating a semiconductor substrate, the state of crystal defects in the semiconductor substrate can be evaluated without characteristic light emission.

  The luminescence method is preferably a cathodoluminescence method.

  In the present invention, for example, such a method is preferably used when evaluating a silicon semiconductor substrate.

  The crystal defects are preferably ion implantation defects generated by ion implantation into the semiconductor substrate.

  The present invention is particularly suitable for evaluating the degree of recovery of ion implantation defects such as point defects generated by ion implantation into a semiconductor substrate.

  Moreover, it is preferable to have the process of measuring the intensity | strength of the band edge light emission of the semiconductor substrate before the said ion implantation further using the luminescence method before the said process to measure.

  With such a method for evaluating a semiconductor substrate, the state of crystal defects in the semiconductor substrate after the defect recovery heat treatment can be more accurately evaluated in the step of evaluating the degree of crystallinity recovery.

  In this case, in the step of evaluating, the intensity of the band edge emission is a value obtained by dividing the intensity of the band edge emission of the semiconductor substrate after the defect recovery heat treatment by the intensity of the band edge emission of the semiconductor substrate before the ion implantation. Based on this, it is preferable to evaluate the degree of crystallinity recovery of the semiconductor substrate after the defect recovery heat treatment.

  If the semiconductor substrate evaluation method has such a process, the state of the semiconductor substrate after the defect recovery heat treatment and the state of the semiconductor substrate before the ion implantation can be compared and evaluated. Can be evaluated.

  In this case, in the step of evaluating, when the intensity ratio of the band edge emission is 0.8 or more, it is preferable to evaluate the semiconductor substrate after the defect recovery heat treatment as having recovered the defects.

  If it evaluates in this way, it is possible to distinguish easily the semiconductor substrate from which the defect was recovered.

  The defect recovery heat treatment can be a rapid heating / cooling heat treatment.

  In the present invention, such heat treatment can be employed as defect recovery heat treatment.

  The semiconductor substrate is preferably a silicon semiconductor substrate.

  The present invention is particularly suitable for evaluating the degree of crystallinity recovery of a silicon semiconductor substrate having crystal defects.

  In this case, it is preferable that the band edge emission is a TO line.

  When evaluating the degree of crystallinity recovery of the silicon semiconductor substrate according to the present invention, it is preferable that the band edge emission is the TO line.

  With the method for evaluating a semiconductor substrate of the present invention, the state of crystal defects in the semiconductor substrate can be evaluated without characteristic light emission. In addition, by measuring the intensity ratio of the band edge emission, the state of the semiconductor substrate after the defect recovery heat treatment and the state of the semiconductor substrate before the ion implantation can be compared and evaluated. Can be evaluated. Further, according to the present invention, it is possible to grasp the degree of recovery of defects caused by ion implantation, which is very effective from the viewpoint of defect control.

It is a graph which shows the relationship between TO line intensity ratio obtained in the Example, and junction leakage current. In the graph of FIG. 1, it is a CL spectrum when the TO line intensity ratio is 0.3.

Hereinafter, the present invention will be described in more detail.
As described above, there is a need for a method for evaluating a semiconductor substrate that can evaluate the state of crystal defects in the semiconductor substrate without characteristic light emission.

As a result of intensive studies to achieve the above object, the present inventors have found a semiconductor substrate evaluation method in which a semiconductor substrate having crystal defects is subjected to a defect recovery heat treatment for recovering the crystal defects. And
A step of measuring the intensity of the band edge emission of the semiconductor substrate using a luminescence method;
Based on the result of the measurement, the method for evaluating a semiconductor substrate having a step of evaluating the degree of crystallinity recovery of the semiconductor substrate has been found to solve the above problems, and the method for evaluating a semiconductor substrate of the present invention has been completed. It was.

  Hereinafter, embodiments of the present invention will be described in detail, but the present invention is not limited thereto.

[Step of preparing a semiconductor substrate having crystal defects]
First, a silicon semiconductor substrate, for example, a P-type silicon wafer doped with a dopant such as boron is prepared. Next, an impurity diffusion layer is formed on the wafer surface. The impurity diffusion layer can be formed, for example, by ion implantation of a dopant such as boron. By this ion implantation, ion implantation defects such as point defects are formed in the silicon semiconductor substrate.

  Before forming the impurity diffusion layer, for example, before ion implantation, it is preferable to measure the intensity of the band edge emission of the semiconductor substrate using a luminescence method described later. Thereby, in the step of evaluating the degree of recovery of crystallinity, the state of crystal defects of the semiconductor substrate after the defect recovery heat treatment can be evaluated in more detail.

[Step of measuring the intensity of the band edge emission of the semiconductor substrate]
Next, the intensity of the band edge emission of the semiconductor substrate subjected to the defect recovery heat treatment is measured using a luminescence method. The method of defect recovery heat treatment is not particularly limited, and examples thereof include rapid heating / cooling heat treatment (RTA treatment) at 1050 ° C. for 30 seconds.

  The intensity of the band edge emission may be measured at at least one location (for example, one location on the substrate surface) of the semiconductor substrate, but the surface of the same semiconductor substrate may be measured at two or more locations. Thereby, the in-plane distribution of the degree of recovery of crystallinity on the surface of one semiconductor substrate can be evaluated, and more accurate evaluation can be performed.

  Alternatively, a plurality of substrates cut from the same ingot may be prepared, defect recovery heat treatment may be performed under a plurality of heat treatment conditions, and the intensity of each band edge emission may be measured. Thereby, it is possible to evaluate how the crystallinity of the semiconductor substrate is restored by performing heat treatment in the next step.

  Among the luminescence methods, it is particularly preferable to use the cathodoluminescence method (CL method) when measuring a silicon semiconductor substrate. The CL method can evaluate the stress / strain distribution, defect distribution, and carrier distribution of a sample with high spatial resolution using an electron beam as a probe. Cathodoluminescence is light emission in the ultraviolet / visible / near infrared region emitted when a sample is irradiated with an electron beam.

  When the CL method is used, the intensity of the band edge emission of the semiconductor substrate can be measured by irradiating the semiconductor substrate with an electron beam and obtaining an emission spectrum. Thus, by using the CL method, it becomes possible to more accurately investigate the degree of defect recovery.

  The light emission mechanism in this CL method differs depending on the material, but in the case of an indirect transition type semiconductor such as silicon, (1) generation of electron / hole pairs, (2) carrier diffusion, (3) light emission recombination There are three. In the case of silicon, a TO phonon line (TO line) corresponding to a band gap (about 1.1 eV) is strongly observed. This is an interband transition with phonon emission because silicon is an indirect transition semiconductor.

  As an apparatus, a scanning electron microscope (SEM) is generally used as an electron beam source, and a detector / spectrometer for detecting light emission from a sample is used for this. Further, stage cooling for obtaining light emission intensity by suppressing lattice vibration. It is preferable to use one provided with a mechanism such as. As can be seen from the outline of an apparatus using an SEM as an electron beam source, the CL method can be compared with an SEM image, an emission spectrum of a wide range of wavelengths can be obtained, high resolution, and deep by changing the acceleration voltage. There is a point that can be analyzed.

  When crystal defects and impurities form energy levels in the band gap, light emission (characteristic light emission) through crystal defects and impurities occurs in addition to interband transition light emission (band edge light emission). However, when the defect density introduced into the semiconductor substrate is small as described above, characteristic luminescence may not be observed even if crystal defects exist. That is, a non-luminescent center may exist. According to the present invention, even if characteristic luminescence is not observed (even if a non-luminescent center is present), the crystal of the semiconductor substrate can be measured by measuring the intensity of band edge emission (for example, the intensity ratio of band edge emission described later). The presence of defects can be confirmed.

  The semiconductor substrate that can be used in the present invention is not particularly limited. Although the present invention can also evaluate a direct transition type semiconductor that emits light without phonon intervention, as described above, a silicon semiconductor substrate or the like in which band edge emission (TO line) accompanied by phonon emission is strongly observed, etc. It is suitable for evaluating the indirect transition type semiconductor. The TO line is not caused by defects (defects existing in the band) but is emitted from the band edge in the complete crystal. Hereinafter, the case where the intensity of the TO wire of the silicon semiconductor substrate is measured will be mainly described.

[Step of evaluating crystallinity recovery degree of semiconductor substrate]
Next, the degree of crystallinity recovery of the semiconductor substrate is evaluated based on the measurement result of the intensity of the band edge emission.

  As an evaluation method of the degree of recovery of crystallinity, for example, the surface of the semiconductor substrate is measured in one place in the previous step, and the degree of crystallinity recovery of the surface of the semiconductor substrate is evaluated in this step, the same semiconductor substrate in the previous step This method measures the surface of two or more locations, evaluates the difference in the degree of recovery of crystallinity on the surface of one semiconductor substrate in this step, and prepares multiple substrates in the previous step, and performs defect recovery heat treatment under multiple heat treatment conditions. The intensity | strength of each band edge light emission is measured, The method etc. which evaluate the crystallinity recovery | restoration degree of the semiconductor substrate according to heat processing conditions at this process can be mentioned.

  In these cases, it is possible to evaluate the degree of crystallinity recovery by measuring only the intensity of the band edge emission of the semiconductor substrate after the defect recovery heat treatment. It is preferable to evaluate the crystallinity recovery degree of the semiconductor substrate after the defect recovery heat treatment based on the intensity ratio of the band edge emission that is a value divided by the intensity of the band edge emission of the semiconductor substrate before ion implantation. In this case, as described above, before forming the impurity diffusion layer, the intensity of the band edge emission of the semiconductor substrate is measured using a luminescence method. Hereinafter, a method for evaluating the degree of crystallinity recovery based on the intensity ratio of band edge emission will be mainly described.

  Since the semiconductor substrate before ion implantation does not have ion implantation defects, it has fewer crystal defects than the semiconductor substrate after ion implantation. Therefore, the TO line that is the band edge emission in the complete crystal is strongly observed. On the other hand, the TO line of the semiconductor substrate after the defect recovery heat treatment is observed relatively weakly. This is because the crystallinity is relatively lowered as compared with the semiconductor substrate before ion implantation.

  The intensity of the band edge emission of the semiconductor substrate after the defect recovery heat treatment varies depending on the heat treatment conditions. For example, if heat treatment is performed under conditions where crystal defects are not sufficiently recovered, the intensity ratio of band edge emission becomes small. In contrast, when the heat treatment is performed under the condition that the crystal defects are sufficiently recovered, the intensity ratio of the band edge emission increases. Therefore, by calculating the intensity ratio of the band edge emission (confirming the degree of recovery of the intensity of the band edge emission), the degree of recovery of the crystallinity of the semiconductor substrate after the defect recovery heat treatment can be more accurately evaluated.

  Further, when the intensity of the band edge emission is measured in two or more places on the surface of the same semiconductor substrate in the previous process, the intensity ratio of a plurality of band edge emission can be calculated in one semiconductor substrate in this process. This makes it possible to more accurately evaluate the difference in the degree of crystallinity recovery on the surface of one semiconductor substrate.

  In addition, when multiple substrates are prepared in the previous process, defect recovery heat treatment is performed under multiple heat treatment conditions, and the intensity of each band edge emission is measured, the intensity ratio of the band edge emission is calculated for each heat treatment condition in this process. can do. This makes it possible to more accurately evaluate the degree of crystallinity recovery of the semiconductor substrate in accordance with the heat treatment conditions.

  FIG. 1 is a graph showing the relationship between the TO line intensity ratio and the junction leakage current obtained in the examples described later. In FIG. 1, the vertical axis represents the TO line intensity ratio, and the horizontal axis represents the value of the junction leakage current. Defect recovery heat treatment is performed at a plurality of processing temperatures, and the TO line intensity ratio and the junction leakage current are measured at each temperature. At this time, the junction leakage current tends to increase as the TO line intensity ratio decreases.

  In general, the junction leakage current increases as the amount of crystal defects on the surface of the semiconductor substrate increases. Therefore, it can be seen from this tendency that when the TO line intensity ratio is small (for example, 0.5 or less), it is possible to evaluate that the degree of recovery of crystallinity of the semiconductor substrate is not sufficient.

  It can be confirmed from FIG. 1 that there is a correlation between the intensity ratio of the band edge emission and the abundance of crystal defects. Therefore, in order to improve the junction leakage current (electrical characteristics), it is of course possible to evaluate the degree of crystallinity recovery of the semiconductor substrate after the defect recovery heat treatment by simply calculating the intensity (intensity ratio) of the band edge emission. it can.

  Further, as shown in FIG. 1, particularly when the TO line intensity ratio is 0.8 or more, the junction leakage current is also suppressed. Therefore, when the TO line intensity ratio is 0.8 or more, the semiconductor substrate after the defect recovery heat treatment can be evaluated as having recovered the defects. For example, if a heat treatment condition in which the TO line intensity ratio is 0.8 or more is obtained in advance according to the present invention and defect recovery heat treatment is performed under this condition, a semiconductor substrate with few crystal defects can be manufactured.

  FIG. 2 is a CL spectrum when the TO line intensity ratio is 0.3 in the graph of FIG. As shown in FIG. 2, characteristic light emission may not be observed even in a state where the leakage current is high (a state where crystal defects exist). In this case, since the non-light emission center exists, the TO line intensity ratio is small.

  Thus, according to the present invention, it is possible to detect crystal defects (such as non-emission centers) that could not be detected by the conventional method of observing characteristic luminescence, and to more accurately evaluate a semiconductor substrate. Can do. In addition, since it is not necessary to perform a special process (for example, formation of a PN junction) in order to measure the intensity of the band edge emission, the semiconductor substrate can be more easily evaluated.

[Use of the present invention]
The present invention is suitable for evaluating the degree of recovery of crystal defects in a semiconductor substrate, in particular, ion implantation defects generated when forming a PN junction. Therefore, the present invention can be applied when manufacturing a semiconductor substrate in which an impurity diffusion layer is formed on the surface.

  EXAMPLES Hereinafter, although an Example is shown and this invention is demonstrated more concretely, this invention is not limited to this Example.

(Example)
As a sample, a P-type silicon wafer having a diameter of 200 mm doped with boron was used. The resistivity of this silicon wafer is 10 Ω · cm. The wafer was ion-implanted with boron at 55 keV at 5 × 10 ¹² atoms / cm ² and preliminarily annealed with RTA. At this time, the temperature was changed from 800 to 1100 ° C. In this way, wafers actually evaluated by CL were obtained.

Next, after preparing the same substrate separately, forming an oxide film and opening an oxide film, boron ions were implanted under the same conditions. Next, ion implantation for forming a PN junction was performed on the surface of the substrate, and recovery annealing was performed by RTA at each temperature. Next, the leakage current of the substrate on which the PN junction was formed was measured at each temperature (measurement conditions: applied voltage 3 V, room temperature, junction area 1 mm ² ). This measurement was performed at 900 locations, and the median value of the measured values obtained at 900 locations was used as a representative value.

  Thereafter, in the wafer to be evaluated, the degree of recovery of ion implantation defects (the degree of crystallinity recovery) at the same locations as the joint locations as the representative values was evaluated by the CL method.

  Under all temperature conditions, no emission other than the TO line due to silicon band edge emission was observed. Moreover, the TO line intensity ratio tended to decrease as the annealing temperature after ion implantation decreased. In the transmission electron microscope (TEM), no defects were observed in the ion-implanted region.

  FIG. 1 shows the result of examining the relationship between the TO line intensity ratio and the junction leakage current. As the leakage current decreases, the TO line intensity ratio tends to increase. FIG. 2 is a CL spectrum when the TO line intensity ratio is 0.3 in the graph of FIG. As shown in FIG. 2, when the TO line intensity ratio is 0.3, no characteristic light emission is observed other than the TO line. However, at this time, the leakage current is in a high state (about 100 pA). Therefore, in this case, it can be seen that the TO line intensity is reduced due to the presence of the non-luminescent center. Therefore, it can be evaluated that a semiconductor substrate having a TO line intensity ratio of 0.3 has a defect (non-emission center).

  As described above, according to the present invention, the state of crystal defects in the semiconductor substrate can be evaluated without characteristic light emission. In particular, crystal defects that could not be detected by the conventional method of observing characteristic light emission can be detected, and the semiconductor substrate can be evaluated more precisely.

  The present invention is not limited to the above embodiment. The above-described embodiment is an exemplification, and the present invention has substantially the same configuration as the technical idea described in the claims of the present invention, and any device that exhibits the same function and effect is the present invention. It is included in the technical scope of the invention.

Claims (9)

A method for evaluating a semiconductor substrate that has been subjected to a defect recovery heat treatment for recovering the crystal defects with respect to a semiconductor substrate having crystal defects,
A step of measuring the intensity of the band edge emission of the semiconductor substrate using a luminescence method;
And a step of evaluating a degree of crystallinity recovery of the semiconductor substrate based on a result of the measurement.   The method for evaluating a semiconductor substrate according to claim 1, wherein the luminescence method is a cathodoluminescence method.   3. The semiconductor substrate evaluation method according to claim 1, wherein the crystal defect is an ion implantation defect generated by ion implantation into the semiconductor substrate.   4. The semiconductor substrate evaluation according to claim 3, further comprising a step of measuring the intensity of light emission at a band edge of the semiconductor substrate before the ion implantation by using a luminescence method before the measuring step. Method.   In the step of evaluating, based on the intensity ratio of the band edge emission that is a value obtained by dividing the intensity of the band edge emission of the semiconductor substrate after the defect recovery heat treatment by the intensity of the band edge emission of the semiconductor substrate before the ion implantation, The semiconductor substrate evaluation method according to claim 4, wherein the degree of crystallinity recovery of the semiconductor substrate after the defect recovery heat treatment is evaluated.   6. The step of evaluating, wherein, when the intensity ratio of the band edge emission is 0.8 or more, the semiconductor substrate after the defect recovery heat treatment is evaluated as having recovered the defects. The evaluation method of the semiconductor substrate of description.   The semiconductor substrate evaluation method according to claim 1, wherein the defect recovery heat treatment is a rapid heating / cooling heat treatment.   The semiconductor substrate evaluation method according to claim 1, wherein the semiconductor substrate is a silicon semiconductor substrate.   The method for evaluating a semiconductor substrate according to claim 8, wherein the band edge emission is a TO line.

Images