JP5504634B2 - Lifetime evaluation method - Google Patents

Description

  The present invention relates to a method for evaluating the lifetime of a silicon single crystal wafer. Specifically, the present invention relates to a method for evaluating contamination of a silicon single crystal wafer having a dopant impurity diffusion layer formed on one surface due to metal impurities.

Conventionally, the measurement of the lifetime of a silicon single crystal wafer having a low resistance layer on the surface where a high concentration of dopant impurities is diffused on the wafer surface has been unreliable and has hardly been performed.
When the lifetime of a silicon single crystal wafer in which a diffusion layer was actually formed by ion implantation or the like was measured, it often showed a value of about one-tenth of the original lifetime value.

  Since the original lifetime can be obtained by measuring the wafer lifetime after removing the ion-implanted layer or diffusion layer by etching, etc., and performing chemical passivation, the ion-implanted layer or diffusion layer affects the lifetime measurement and is accurate. Many people thought that it was impossible to measure the lifetime.

  Therefore, contamination based on indirect evaluation data, such as estimating contamination of product wafers by managing contamination of equipment / equipment using monitor wafers, rather than contamination assessment of product wafers with diffusion layers formed by ion implantation, etc. Management methods are widely used.

Here, in a semiconductor device such as an image sensor, a redundant circuit cannot be used, and since its electrical characteristics are very sensitive to impurity contamination, thorough contamination control has been performed by a monitor wafer.
Even so, the deterioration of the element characteristics presumed to be caused by metal contamination often occurs, making it difficult to stably manufacture the image pickup element. And it is often found that the assumption that the contamination of the monitor and the contamination of the product wafer itself is the same is broken.

However, evaluation and inspection methods that can detect trace metal contamination are limited. In particular, methods that can measure in a short time and can feed back the evaluation results to the process are almost limited to two methods of wafer lifetime measurement by the μPCD method and surface photovoltage (SPV) method (for example, non-patent document). 1).
And only these two methods are standardized by SEMI (Semiconductor Equipment and Materials International).

  However, these methods are also considered to be not directly applicable to, for example, a low-resistance silicon single crystal wafer, and evaluation of the silicon single crystal wafer itself after the ion implantation process has hardly been performed. It was.

"Issues with Silicon Crystal / Wafer Technology" (Realize Inc., issued on January 31, 1994) pp. 265-269

Furthermore, in recent years, improvement in characteristics has progressed in the field of image pickup devices and the like, and it has become more sensitive to metal contamination in the substrate and process than in the past. In some cases, it may be required that the metal impurities be 10 ⁹ / cm ³ or less. However, there is an aspect that the evaluation technology for that has not kept up with the demand.

  Since the evaluation of the contamination of the product wafer itself cannot be performed due to the limitations of the evaluation technique as described above, the contamination of the process apparatus is controlled using the monitor wafer. For this reason, sometimes the device process is completed, and the defect that seems to be caused by contamination occurs only at the stage of confirming the characteristics of the element at the end. Deterioration has occurred.

  Therefore, a method for evaluating the lifetime of a silicon single crystal wafer that provides a measurement result with high sensitivity and speed is desired. The method is not a management of the contamination state of the apparatus by the monitor, but is desired to be a technique that can be evaluated non-destructively if the product wafer itself can be contaminated.

  The present invention has been made in view of the above problems, and is an evaluation for accurately and simply measuring the lifetime of a silicon single crystal wafer on which an ion-implanted diffusion layer, which has been conventionally considered impossible, is formed. It aims to provide a method.

  In order to solve the above-described problems, the present invention is a method for evaluating the lifetime of a silicon single crystal wafer in which dopant impurities are diffused on one side, and the diffused silicon single crystal wafer is evaluated when the lifetime is evaluated. Passivation is performed on the entire surface, and then the lifetime is evaluated by irradiating the surface opposite to the diffusion surface with excitation light, incident high frequency, and detecting reflected waves. The lifetime evaluation method characterized by the above is provided.

Thus, when forming a diffusion layer by ion implantation, recovery heat treatment is often performed in an inert gas, and a passivation film (oxide film) is often not formed on the wafer surface. In that case, passivation by oxide film formation or chemical passivation is performed. Thereafter, the entire surface on the side opposite to the surface where the ion implantation diffusion layer is formed is irradiated with excitation light for exciting carriers. Then, high-frequency incidence and reflected wave detection for evaluating the lifetime of the injected carriers are performed on the surface opposite to the ion-implanted surface.
This makes it possible to accurately evaluate the lifetime of a silicon single crystal wafer on which a diffusion layer that cannot be measured conventionally is formed. And by using such an evaluation method for the evaluation of the presence or absence of metal impurity contamination after ion implantation when manufacturing a silicon single crystal wafer with an ion implantation diffusion layer formed, an ion implantation diffusion layer having a low metal impurity concentration It is effective to manufacture a silicon single crystal wafer having a low cost and high yield.

Here, it is preferable that at least one element of boron, phosphorus, arsenic, antimony, and carbon is ion-implanted in the diffusion silicon single crystal wafer.
As described above, a silicon single crystal wafer into which at least one element selected from boron, phosphorus, arsenic, antimony, and carbon is ion-implanted is effective in preventing carrier diffusion from the substrate and gettering, but forms a diffusion layer. In this process, there is a problem that metal contamination is likely to occur, but conventionally, it has been difficult to evaluate the lifetime of the wafer itself on which the diffusion layer is formed. In general, the cleanliness of the heat treatment furnace is regularly managed by the lifetime of the monitor wafer (mirror wafer). However, according to the lifetime evaluation method of the present invention, it is also possible to evaluate a silicon single crystal wafer in which such an ion implantation diffusion layer is formed.

The ion implantation is preferably performed at a dose in the range of 1 × 10 ¹⁴ to 1 × 10 ¹⁶ atoms / cm ² .
Thus, ions are implanted at a dose of 1 × 10 ¹⁴ to 1 × 10 ¹⁶ atoms / cm ² to form a diffusion layer having a sufficiently low resistivity on the surface layer. In this case, it plays a role of a passivation film. Also, many layers in which dopant impurities are diffused at a high concentration often have high gettering ability for metal impurities. With the lifetime evaluation method of the present invention, it is possible to grasp the comprehensive metal contamination state including the gettering effect by such an ion implantation diffusion layer. For this reason, when manufacturing the board | substrate for devices sensitive to metal contamination, the evaluation method of this invention is suitable.

Further, the diffusion silicon single crystal wafer may be one whose crystallinity has been recovered by diffusion heat treatment after ion implantation, and in which a diffusion layer is formed at a depth of at least 5 μm or less from the surface.
Thus, the lifetime evaluation method of the present invention is effective for a silicon single crystal wafer having a sheet resistance of 10Ω / □ or more in which a diffusion layer is formed at a depth of at least 5 μm or less from the surface. The passivation effect increases when a low-resistance region is formed in a thin surface layer region. If the sheet resistance is 10 Ω / □ or more, the change in bulk resistivity can be measured, resulting in a high-concentration lifetime evaluation. Can be implemented. For example, the low resistance diffusion layer immediately below the epitaxial layer for the imaging device is often required to be thin and have a low resistivity, so that the evaluation method of the present invention is suitable.

Further, the present invention is a method for evaluating the lifetime of an epitaxial wafer in which an epitaxial layer is formed on the diffusion surface side of an ion-implanted single-sided diffusion wafer, and when the lifetime is evaluated, Passivation is performed on the entire surface, and then the lifetime is evaluated by performing excitation light irradiation, high frequency incidence and detection of the reflected wave on the surface opposite to the diffusion surface. Provided is a method for evaluating the lifetime of a featured epitaxial wafer.
In this case as well, after performing passivation treatment by chemical passivation, etc., irradiation with excitation light and high frequency is performed from the surface opposite to the surface on which the diffusion layer and the epitaxial layer are formed, and the lifetime is measured. The lifetime value of the wafer bulk can be measured.
With such an evaluation method of the present invention, metal impurity contamination that occurs from the diffusion process through the epitaxial process can be comprehensively evaluated based on the lifetime of the product wafer. Highly accurate management is possible.

Chemical analysis methods are also used for highly sensitive evaluation of metal contamination, which is also applied to the evaluation of epitaxial wafer contamination, but generally requires 1 to 2 days for measurement and evaluation. It is. However, in the evaluation method of the present invention, for example, if chemical passivation is used, highly accurate metal impurity contamination of the epitaxial wafer can be evaluated in about 30 minutes.
When starting the epitaxial process, the lifetime can be measured in a time that is not much different from the time for checking whether the resistivity is correctly controlled. Since it is not necessary, it can be applied to the mass production process.

As described above, according to the present invention, a silicon single crystal wafer formed with an ion implantation layer or a diffusion layer used for a semiconductor device such as an imaging device whose electrical characteristics are sensitive to impurity contamination or epitaxially formed on the surface thereof. It becomes possible to evaluate impurity contamination with high accuracy in the epitaxial wafer in which the layer is formed.
In addition, with the evaluation method of the present invention capable of performing such simple and highly accurate evaluation, the lifetime of the product wafer itself can be measured, and the cause and origin of contamination can be ascertained and highly accurate contamination evaluation can be performed. Therefore, it becomes possible to stably manufacture a high-quality silicon single crystal wafer at a low cost.

It is the figure which showed the outline | summary of the lifetime evaluation method of this invention and the conventional diffused silicon single crystal wafer. It is the figure which showed an example of the lifetime map of the diffusion silicon single crystal wafer of the Example and comparative example of this invention. It is the graph which showed the time dependence of the reflective intensity of a μ wave at the time of evaluating the lifetime of the silicon single crystal wafer of the example and comparative example of the present invention by voltage. It is the figure which showed an example of the measuring device used in order to measure the lifetime of a silicon single crystal wafer.

Hereinafter, the present invention will be described more specifically.
As described above, development of an evaluation method for accurately and simply measuring the lifetime of a silicon single crystal wafer formed with an ion-implanted layer and a diffusion layer, which has been conventionally considered impossible, has been awaited. .

  In the device process, various evaluation methods are used to eliminate contamination by heavy metal impurities. Among them, the μPCD method and the SPV method, which are simple and have good sensitivity, are often used. However, in both cases, there are various restrictions on the object to be measured, and the actual situation is that the contamination status of the apparatus using the monitor is indirectly managed rather than the evaluation of the product itself.

In the contamination control of the equipment itself, the contamination of the main equipment can be grasped to a certain extent, but the product wafer is handled during that time, handling, cleaning, inspection, etc., but it is extremely difficult to grasp the contamination in these processes. is there.
If the product can be extracted and evaluated for contamination, it will be possible to manage and monitor the product wafer handling, cleaning and inspection processes. By comparing this result with the result of the device management monitor, knowledge about contamination in many processes can be obtained.

  Furthermore, if possible, it is desirable to establish a technique that enables nondestructive measurement of products. In addition to the great merit in terms of cost, the cleanliness of the product can be assured with high accuracy. And by continuous data acquisition, when something abnormal occurs, it can be grasped immediately, and the cause can be investigated and countermeasures can be taken.

  However, since it was almost impossible to directly measure the lifetime of a silicon single crystal wafer on which an ion-implanted layer or a diffusion layer was formed, many devices in which actual wafers are not monitored for contamination It is also processed by. For this reason, although there is no problem in monitoring device contamination, defects that are presumed to be caused by metal contamination such as leaks often occur in actual devices (particularly imaging devices).

Here, when the wafer is irradiated with excitation light, electron-hole pair generation occurs, and carriers increase beyond the thermal equilibrium state. That is, the resistance is lowered. When the light irradiation is stopped, the electron and hole concentrations return to an equilibrium state with a certain time constant.
This time constant is generally called lifetime, and greatly depends on the concentration of heavy metal impurities such as iron dissolved in silicon. A problem in measuring the wafer lifetime is surface recombination. Because τ _obs obtained from the attenuation curve substantially obtained is obtained by using bulk lifetime (recombination lifetime) τ _bulk and surface lifetime τ _surf .
1 / τ _obs = 1 / τ _bulk + 1 / τ _surf
It is because it is expressed.
Then, the impurity concentration in the wafer is evaluated from the bulk lifetime τ _bulk . The bulk lifetime τ _bulk is mainly determined by the density of impurity levels (deep levels) produced by metal impurities in an indirect transition type semiconductor such as silicon.

And, at the metal contamination level of the present silicon single crystal wafer, τ _bulk is large (1 to 2 msec), and τ _surf greatly affects τ _obs when the wafer thickness is 1 mm or less. For this reason, it is necessary to increase τ _surf by some means (decrease the surface recombination rate) to _satisfy τ _obs ≈τ _bulk, and therefore, an oxide film has generally been formed. Recently, chemical passivation methods that are simpler and can increase τ _bulk have been widely used.

  Here, the change in resistivity is obtained from the reflectance of the reflected wave by making a high frequency incident on the substrate on which the excitation light is incident, but if a large number of carriers are present in the wafer in advance, the thermal balance due to light irradiation The amount of change in the resistance becomes relatively small, the S / N deteriorates, and the resistance cannot be accurately attenuated. Therefore, in the wafer lifetime measurement, it is necessary to use a wafer having a resistivity of 1 to 3 Ωcm or more.

However, a silicon single crystal wafer in which an ion-implanted layer and a diffusion layer are formed has a layer in which a dopant impurity having a concentration of about 10 ¹⁸ atoms / cm ³ is implanted in the range of the surface layer to 5 μm. Therefore, in the lifetime measurement, the irradiation light is absorbed by the diffusion layer and hardly enters the bulk, and the carrier concentration in an equilibrium state with respect to the carrier concentration by photoexcitation increases by the dopant concentration of the diffusion layer. There was a problem of deteriorating N. Further, the reflection component of the μ wave from the diffusion layer increases, and the change in the carrier concentration of the surface layer is detected.
For this reason, when the lifetime is measured by a normal method, the diffusion wafer is mostly capable of obtaining a lifetime value that is 1 to 2 digits smaller.

  Therefore, the present inventors have grasped the cause that prevents stable measurement, avoided it, and accurately and easily measured the lifetime of a silicon single crystal wafer in which an ion implantation layer and a diffusion layer are formed. We studied hard to see if it could be done.

As a result, paying attention to the fact that the ion-implanted layer and the diffusion layer in which the implanted ions are diffused are formed only on one side of the silicon single crystal wafer, the excitation light is irradiated from the back side where it is not formed. In addition, the inventors have conceived that the lifetime can be measured while ensuring S / N by performing high-frequency incidence and reflected waves on the back side.
As a result, it has been found that the influence of the ion-implanted layer and the diffusion layer can be substantially avoided, so that the wafer lifetime can be evaluated, and the present invention has been completed.

  Hereinafter, the present invention will be described in detail with reference to the drawings, but the present invention is not limited thereto.

First, the μPCD method will be briefly described.
Wafer lifetime measurement by the μPCD method evaluates metal impurities in a sample by irradiating the sample (wafer) with light and detecting the lifetime of the minority carriers generated by changes in the reflectance of the microwave. It is.

Next, an outline of a wafer lifetime measuring apparatus that is generally widely used will be described with reference to FIG.
The wafer lifetime measuring apparatus 10 irradiates, for example, at least a high frequency (hereinafter also referred to as μ wave or microwave (wavelength: 1.0 to 10.0 GHz)) to a silicon single crystal wafer to be measured using a waveguide. To the silicon single crystal wafer 16 and the excitation light irradiation apparatus 12 for irradiating the wafer with pulsed light using a μ-wave oscillator 13 and a laser diode of 10 to 100 W to increase the carrier concentration of the wafer by about two digits The circulator 15 for isolating the incident high frequency and the reflected wave from it, and the silicon single crystal wafer 16 are moved to measure the photoconductivity decay curve of a specific region in the wafer at a few millimeter pitch. Has detector 14 for calculating carrier lifetime by arithmetic processing That.

  A method for evaluating the lifetime of the silicon single crystal wafer of the present invention using such a wafer lifetime measuring apparatus 10 will be specifically described with reference to FIG. 1, but is not limited to this.

First, a silicon single crystal wafer is prepared.
The silicon single crystal wafer prepared at this time may be a commonly used one, for example, a slice produced from a silicon single crystal rod grown by the CZ method may be used. In addition, the conductivity type, crystal orientation, crystal diameter, and the like can be appropriately selected so as to be suitable for the target semiconductor device. However, the resistivity is preferably 1 Ω · cm or more, more preferably 5 Ω · cm or more.

Next, an ion implantation layer is formed by performing ion implantation on the prepared silicon single crystal wafer. For example, a large current ion implantation apparatus may be used for this ion implantation.
Then, after the ion implantation, diffusion heat treatment is performed so that the ion implantation layer becomes an ion implantation diffusion layer (diffusion layer) 11a, and a diffusion silicon single crystal wafer (diffused silicon single crystal wafer) 11 is obtained.

Here, the ion species to be implanted is at least one of boron, phosphorus, arsenic, antimony, and carbon, and the dose amount of ions to be implanted is in the range of 1 × 10 ¹⁴ to 1 × 10 ¹⁶ atoms / cm ² . If the specification of the substrate for the element to be manufactured is included, the product wafer itself can be evaluated.

A silicon single crystal wafer into which boron is implanted at a dose of 1 × 10 ¹⁴ to 1 × 10 ¹⁶ atoms / cm ² is a wafer having a high gettering capability against metal impurities. For the purpose, these diffusion layers may be formed by ion implantation.
As described above, the lifetime evaluation method of the present invention can also evaluate the lifetime of such a wafer, so that a comprehensive metal contamination state including gettering ability can be achieved in a short time and at a low cost. Can be evaluated.
The lifetime evaluation method of the present invention can be applied to a so-called patterned buried diffusion wafer in which a buried diffusion layer is formed in an island-shaped region for bipolar IC and bi-CMOS IC as long as it is single-sided diffusion. Of course you can. The lifetime evaluation method of the present invention can be applied not only to an n-type low resistance buried layer but also to a substrate on which a p-type low resistance layer is formed.

Also, it is common to form an epitaxial layer on the surface on the ion implanted side.
Thus, even if an epitaxial layer is formed on the surface on the side where the diffusion layer is formed, if the epitaxial wafer is passivated, carriers hardly recombine at the boundary with the diffusion layer. It can be performed.
A silicon single crystal wafer in which an epitaxial layer is formed on the surface on which such a diffusion layer is formed is often used as a substrate for manufacturing a semiconductor device such as an image sensor. In bipolar IC and bi-CMOS IC, the pattern is formed on the epitaxial layer on the diffusion layer. It is very important to evaluate the metal impurity concentration even when manufacturing such a silicon single crystal wafer, and the evaluation method of the present invention is also suitable in such a case, whereby an epitaxial with a low impurity concentration is formed. You can get a wafer.

  Thereafter, passivation is performed on the entire surface of the diffusion silicon single crystal wafer 11 in order to form a passivation film 11b for suppressing surface recombination of carriers.

Here, it is desirable to perform thermal oxidation or chemical passivation as this passivation.
In the case of chemical passivation, a passivation film can be easily formed in a short time, and the lifetime of the wafer can be evaluated more quickly. Further, in the case of thermal oxidation, an oxide film having a low interface state can be formed on the surface. For this reason, surface recombination can be suppressed and wafer lifetime can be evaluated accurately with high sensitivity.

Thereafter, lifetime measurement is performed on the diffused silicon single crystal wafer 11 on which the passivation film 11b is formed, and the metal contamination state in the diffused silicon single crystal wafer 11 is evaluated.
The lifetime measurement method at this time includes the μPCD method.
At this time, as shown in FIG. 1 (a), the surface opposite to the surface on which the ion implantation diffusion layer 11a is formed is irradiated with excitation light, and high-frequency incidence and detection of reflected waves are detected. I do.

In this way, by irradiating the surface opposite to the surface on which the ion-implanted diffusion layer is formed by the ion implantation, the excitation light is irradiated, the high frequency is incident, and the reflected wave is detected. The reflected wave is prevented from being affected by the diffusion layer, and the incident intensity and signal intensity are prevented from being attenuated. Thereby, even a silicon single crystal wafer in which a diffusion layer is formed, the lifetime can be evaluated with high accuracy (that is, even for a wafer having a long lifetime).
In addition, since the wafer lifetime can be measured in a short time, the wafer lifetime distribution can be measured with high accuracy. From the pattern, the cause of contamination is the chuck of the transfer robot, or the contact part of the carrier. It is effective for pollution control that it can be inferred relatively easily.

  On the other hand, as shown in FIG. 1B, conventionally, irradiation of excitation light, incidence of high frequency, and detection of reflected waves have been performed on the surface on which ion implantation has been performed. In this case, since the excitation light incident on the ion implantation diffusion layer 11a formed by ion implantation is absorbed and only a part of carriers are injected into the bulk portion of the silicon single crystal wafer, the lifetime can hardly be evaluated. It was.

  Here, in the present invention, passivation on the back surface side (excitation light irradiation side) is necessary. On the other hand, with respect to passivation on the surface (ion implantation surface) side, since the diffusion layer is in contact with the bulk, the effect of the conventional oxide film and chemical passivation cannot be expected. However, in many cases, the diffusion layer itself serves as a passivation film regardless of the conductivity type, depending on the concentration and the like. By utilizing this fact, it becomes possible to measure the wafer lifetime of a wafer having an ion implantation diffusion layer with relatively high accuracy.

  The lifetime evaluation method of the present invention can also be applied to a wafer having a buried diffusion structure such as a bipolar IC. This is because, when a technique such as ion implantation that does not diffuse on the back surface is used, if the excitation light for lifetime measurement is irradiated from the back surface side, the ion-implanted layer or diffusion layer on the opposite side of the irradiated surface is the passivation film. Since the oxide film is formed on the back surface, which plays a role and is an implanted region or an undiffused region, the bulk lifetime can be similarly measured by applying the present invention.

EXAMPLES Hereinafter, although an Example and a comparative example are shown and this invention is demonstrated more concretely, of course, this invention is not limited to these.
(Examples and comparative examples)
First, a p-type mirror wafer having a diameter of about 200 Ωcm and a diameter of 200 mm (thickness: 725 μm) was prepared as a silicon single crystal wafer.
Thereafter, boron was ion-implanted under the conditions of a dose of 1 × 10 ¹⁵ atoms / cm ² and an acceleration energy of 90 keV, and cleaning was performed.
Then, the crystallinity was recovered by annealing in a nitrogen atmosphere at 1000 ° C. for 10 minutes, and thereafter, the oxide film was switched to an oxygen gas atmosphere to form an oxide film of about 300 mm (30 nm).

With respect to the produced diffusion silicon single crystal wafer, the wafer lifetime was measured by a conventional wafer lifetime evaluation method as shown in FIG. 1B (comparative example). Thereafter, the wafer is inverted and, as shown in FIG. 1A, the excitation light is irradiated from the back side opposite to the surface on which the ion-implanted diffusion layer is formed. Time was measured (Example). The results are shown in FIGS. 2 (a) and 2 (b). This FIG.2 (b) is a measurement result by the evaluation method of this invention.
The measurement conditions in FIGS. 2A and 2B were standard conditions used when measuring the monitor wafer.

As a result, in the case of FIG. 2A performed by the conventional evaluation method, the average lifetime is 18 μsec (maximum 23 μsec · minimum 4 μsec). In the case of FIG. The average was 286 μsec (maximum 458 μsec, minimum 122 μsec).
As shown in FIGS. 2 (a) and 2 (b), the absolute value of the lifetime is obtained even though the same diffusion silicon single crystal wafer is subjected to the same conditions except the excitation light irradiation and the high frequency incident / reflection. Are very different.

  A PCD (Photo Conductive Decay) waveform as a measurement result of the two evaluation methods was output as a voltage with respect to time. The result is shown in FIG. FIG. 3A shows a conventional evaluation method, and FIG. 3B shows an evaluation method of the present invention.

As shown in FIG. 3A, when the lifetime is measured by a conventional method on a silicon single crystal wafer on which an ion implantation diffusion layer is formed, noise increases and the amount of reflected microwaves increases. A steep decay is observed at the beginning of the decay. As a result, it was found that the time and lifetime that the reflection amount of the microwave becomes 1 / e is shortened.
On the other hand, as shown in FIG. 3B, in the evaluation method of the present invention, an abrupt decrease caused by the influence of the surface (surface recombination) at the initial stage of attenuation is not observed, and an ion implantation diffusion layer is formed. Attenuation similar to the PCD waveform of lifetime measurement on a silicon single crystal wafer that has not been obtained can be obtained.

For confirmation, a new mirror-surface wafer is prepared, ion implantation is performed, recovery annealing after ion implantation is performed only in a nitrogen atmosphere, and then a lifetime map of a diffusion silicon single crystal wafer without passivation is shown. This is shown in 2 (c). The measurement conditions were the same as in FIG. The lifetime at this time was an average of 27 μsec (maximum 42 μsec, minimum 6 μsec).
In this case, since an oxide film as a passivation film is not formed on the surface irradiated with excitation light, the contribution of surface recombination is increased, and the absolute value of lifetime is decreased. As described above, when only such a low lifetime value is obtained, in general, local contamination cannot be confirmed due to the deterioration of S / N. As in the conventional evaluation method, the impurity concentration cannot be evaluated from the lifetime.

Further, in order to confirm whether or not the lifetime evaluation method of the present invention can be appropriately evaluated, an ion-implanted diffusion layer of the diffusion silicon single crystal wafer used for the evaluation of FIGS. 2 (a) and (b) is formed. The polished side was mirror-polished to remove about 5 μm from the surface and remove the ion implantation diffusion layer. Then, the removed wafer is washed, the natural oxide film is removed with an HF aqueous solution, and the wafer surface is wetted with a 5% iodine-ethanol solution using a plastic bag. ) To (c), the wafer lifetime was measured and shown in FIG.
The average lifetime was 769 μsec (maximum 1027 μsec · minimum 188 μsec).

As shown in FIGS. 2B and 2D, when the lifetime maps of the same diffusion silicon single crystal wafer are compared, there is a difference of about twice as much as the absolute value, but the distribution pattern is very similar. The local contamination status was clearly detected in all cases. The difference in the lifetime absolute value in FIG. 2 is mainly due to the difference in passivation. In general, a lifetime value of about 1 to 2 [mu] sec is obtained for a wafer free from metal contamination when passivation is not performed, 400 to 600 [mu] sec for oxide film passivation, and about 800 to 1500 [mu] sec for chemical passivation.
From the above, it was confirmed that the contamination on the wafer can be evaluated by performing lifetime measurement by the evaluation method of the present invention on the silicon single crystal wafer on which the ion implantation diffusion layer is formed.

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

10: Wafer lifetime measuring device,
11 ... Diffusion silicon single crystal wafer, 11a ... Diffusion layer, 11b ... Passivation film,
12 ... Excitation light irradiation device,
13 ... μ wave oscillator,
14 ... detector,
15 ... circulator,
16: Silicon single crystal wafer.

Claims (5)

A method for evaluating the lifetime of a silicon single crystal wafer in which dopant impurities are diffused on one side,
When evaluating the lifetime, the entire surface of the diffusion silicon single crystal wafer is passivated, and then the surface opposite to the diffusion surface is irradiated with excitation light, high frequency incident and A lifetime evaluation method characterized in that a lifetime is evaluated by detecting a reflected wave.   2. The lifetime evaluation method according to claim 1, wherein the diffusion silicon single crystal wafer is formed by ion implantation of at least one element selected from boron, phosphorus, arsenic, antimony, and carbon. The lifetime evaluation method according to claim 2, wherein the ion implantation is performed at a dose in a range of 1 × 10 ¹⁴ to 1 × 10 ¹⁶ atoms / cm ² .   The diffusion silicon single crystal wafer is one in which crystallinity is restored by diffusion heat treatment after ion implantation, and a diffusion layer is formed at a depth of at least 5 μm or less from the surface. The lifetime evaluation method according to any one of claims 1 to 3. A method for evaluating the lifetime of an epitaxial wafer in which an epitaxial layer is formed on the diffusion surface side of an ion-implanted single-sided diffusion wafer,
When evaluating the lifetime, the entire surface of the epitaxial wafer is passivated, and then the surface opposite to the diffusion surface is irradiated with excitation light, high-frequency incidents and reflected waves thereof. A method for evaluating the lifetime of an epitaxial wafer, characterized in that the lifetime is evaluated by performing detection.

Images