JP5439793B2 - Sample for depth calibration of secondary ion mass spectrometry, method for producing the same, and secondary ion mass spectrometry - Google Patents

Description

  The present invention relates to a depth calibration sample used for depth calibration of secondary ion mass spectrometry, a manufacturing method thereof, and a secondary ion mass spectrometry method using the calibration sample. The present invention relates to a depth calibration sample suitable for calibrating depth, a manufacturing method thereof, and a secondary ion mass spectrometry method using the calibration sample.

  Secondary ion mass spectrometry (SIMS) is a method for analyzing an element concentration distribution by irradiating a surface of a sample to be analyzed with a primary ion beam and mass-analyzing secondary ions sputtered from the sample surface. It is. In this secondary ion mass spectrometry, the secondary ion intensity is detected while sputtering is progressing. Therefore, by converting the transition time into depth for the data of the time transition of the ion intensity of the secondary ion, that is, the test element ion or the molecular ion bonded to the test element, the depth direction of the sample surface The concentration distribution of the test element can be known.

  Conventionally, the transition time is converted into depth by measuring the depth of the depression formed on the surface of the sample by irradiation with primary ions using a surface roughness meter, and calculating the average sputter rate from the depth of the depression and the transition time. And the irradiation time (that is, the transition time) is converted into the depth (sputtering amount) under the assumption that the sputtering rate is constant.

  However, today, in order to obtain high detection sensitivity, reactive ions such as oxygen or cesium ions are frequently used as primary ions. When such reactive primary ions are used, it is known that there is a period called a transition period in which the sputtering rate varies with time in a shallow surface layer of about several nm from the sample surface. Such fluctuations in the sputtering rate disturb the proportional relationship between the irradiation time and the depth and destroy the assumption that the sputtering rate is constant, making it difficult to accurately convert the irradiation time to the depth. As a result, accurate analysis of the concentration distribution in the depth direction by secondary ion mass spectrometry becomes difficult.

  In secondary ion mass spectrometry, the relationship between irradiation time and depth (sputtering amount) is usually calibrated using a standard sample in order to correct the depth error due to variations in the sputtering rate during the transition period. Yes.

  The standard sample has an impurity concentration distribution in the depth direction, and the concentration and depth (for example, the peak position or the depth of a constant concentration region) are known in advance. When this standard sample is subjected to secondary ion mass spectrometry, the time transition of the ion intensity of the ion containing the impurity is measured, and the time transition pattern of the measured ion intensity corresponds to the known concentration distribution of the standard sample. Assuming that, for example, the irradiation time at the peak position of the ion intensity and the ion intensity correspond to the known depth and impurity concentration at the peak position of the standard sample, respectively, the amount of spatter in secondary ion mass spectrometry (ie, sputter speed) And calibrate the impurity concentration.

  Conventionally, a silicon substrate in which impurities are ion-implanted on the surface has been used as the standard sample. However, in order to calibrate the sputtering amount in a shallow region of about several nm from the surface sputtered during the transition period, it is necessary to form an ion implantation layer having a peak concentration in the shallow region near the surface.

  By the way, in order to form a shallow ion implantation layer, the ion implantation energy must be lowered. However, low ion implantation energy increases the reflection of ions on the surface of the silicon substrate, making it difficult to implant ions with an accurate dose. For this reason, it is difficult to produce a standard sample having an accurate impurity concentration in a shallow region of several nm or less from the surface by ion implantation.

  In addition, a silicon substrate in which impurities are delta-doped is used as a standard sample. (For example, refer to Patent Document 1). This standard sample is formed on a silicon substrate by alternately depositing a plurality of delta-doped layers composed of silicon layers and one atomic layer of impurities (for example, Sb) using MBE (molecular beam epitaxy).

  However, in order to form a plurality of delta doped layers in a shallow region of about several nm, the thickness of the silicon layer inserted between them is extremely thin. For this reason, the deposition temperature must be lowered in order to suppress impurity diffusion into the thin silicon layer. On the other hand, when the deposition temperature is low, the surface roughness of the deposited silicon layer increases, and irregularities occur at the interface of the deposited layer. Such irregularities on the interface make it difficult to detect the exact depth of the delta dope. For this reason, it is difficult to produce a standard sample having a concentration peak at an accurate depth in a shallow region of several nm or less.

Further, a standard sample is used in which a peak concentration is formed in the vicinity of the surface by recrystallization after impurity ion implantation. In this standard sample, an impurity concentration distribution having a peak in a region of several nm or less from the surface can be easily formed. Therefore, the transition period can be calibrated using this standard sample. (For example, refer to Patent Document 2).
JP 2006-343244 A JP 2008-066664 A

  As described above, it is difficult to form an impurity concentration distribution having a peak in a shallow region of several nm or less in a conventional standard sample manufactured by ion implantation of impurities. For this reason, it is difficult to accurately calibrate the relationship between time and depth within the transition period.

  A conventional standard sample in which impurities are delta-doped is produced by an epitaxial deposition method. However, in order to delta-dope a shallow region that is sputtered during the transition period, it must be deposited at a low temperature in order to suppress impurity diffusion. At such a low temperature, the irregularity of the interface is likely to occur, and a standard sample having a steep interface is required. There is a problem that it is difficult to manufacture.

  A conventional standard sample produced by forming a recrystallized layer having a peak concentration near the surface by recrystallization of an impurity ion implanted layer has a shallow recrystallized layer of several nm or less. However, it must be heated to a high temperature for recrystallization and then cooled, and the preparation of the standard sample is complicated. Further, it is difficult to form an impurity concentration distribution having a sharp peak at a shallow position due to impurity diffusion at high temperature. Furthermore, in order to accurately control the impurity concentration distribution of the recrystallized layer, the crystallization temperature and time must be accurately controlled. For this reason, there is a problem that an expensive recrystallization apparatus capable of precise temperature control must be prepared.

  The present invention is a standard sample that can be easily and easily manufactured without using a special apparatus for preparing a standard sample, and is a shallow region (hereinafter referred to as “transition region”) that is sputtered during a transition period. It is an object of the present invention to provide a standard sample having a peak concentration of impurities, a production method thereof, and a secondary ion mass spectrometry method using the standard sample.

In order to solve the above-described problem, a first configuration of the present invention includes an oxygen-containing layer formed by irradiating a silicon layer having an impurity ion-implanted layer formed on its surface with oxygen ions, and the oxygen ions. It is provided as a sample for depth calibration of secondary ion mass spectrometry, characterized by having a redistribution layer formed by collecting the impurities in the vicinity of the surface of the oxygen-containing layer during irradiation .

  In the standard sample of this configuration, an impurity redistribution layer formed by irradiating a silicon substrate having an ion implantation layer formed by impurity ion implantation with oxygen ions is used as a reference for concentration distribution. This redistribution layer is formed in the vicinity of the surface of the oxygen-containing layer formed by oxygen ion irradiation and, as will be described later, a transition region (in the transition period) seen in secondary ion mass spectrometry in which oxygen ions are irradiated as primary ions. It is formed at substantially the same depth as the region to be sputtered. This depth can be precisely controlled by the energy and incident angle of the oxygen ions to be irradiated. Therefore, a standard sample having a precisely controlled depth-wise impurity concentration distribution in the shallow transition region can be easily produced.

  Further, irradiation with oxygen ions is performed at room temperature, for example, without heating. For this reason, the change in the concentration distribution due to the diffusion of impurities is small, and it is easy to manufacture by precisely controlling the concentration distribution.

  Furthermore, irradiation with oxygen ions can be performed by irradiation with oxygen ions by a secondary ion mass spectrometer. Therefore, it can be produced without using a special device for manufacturing, for example, a recrystallization device or an epitaxial deposition device, except for the ion implantation step.

  According to the present invention, a standard sample having a known impurity concentration distribution in the depth direction in a shallow transition region can be easily manufactured without using a special manufacturing apparatus. For this reason, since the relationship between the irradiation time and depth in secondary ion mass spectrometry can be calibrated easily and precisely, the concentration distribution in the depth direction of the test element can be measured accurately.

  The first embodiment of the present invention relates to a standard sample for depth calibration of secondary ion mass spectrometry manufactured by using In as an impurity and controlling the In concentration distribution by the irradiation energy of oxygen ions. In this specification, “impurity” is used to mean an impurity used for calibration included in a standard sample. The concentration distribution of the impurities in the standard sample is used as a reference concentration distribution for calibration of secondary ion mass spectrometry.

  First, the manufacturing process of the standard sample of 1st Embodiment is demonstrated.

  FIG. 1 is a cross-sectional view of a standard sample manufacturing process according to the first embodiment of the present invention, showing a cross section of the standard sample.

  First, referring to FIG. 1A, using an ion implantation apparatus, In is implanted as an impurity into a silicon substrate 1 to form an ion implantation layer 2 in which In is implanted on the surface of the silicon substrate 1. In the first embodiment, In ions are vertically incident on the silicon substrate 1.

  In this In ion implantation, the implantation energy is preferably 5 keV or more and 10 keV or less. When the implantation energy is less than 5 keV, there are many reflections of implanted ions on the surface of the silicon substrate 1, and it is difficult to precisely control the dose amount and concentration distribution of the ion implanted layer 2. On the other hand, if it exceeds 10 keV, the ion implantation layer 2 becomes deep and the In concentration near the surface is low, so that sufficient In ion intensity cannot be obtained at the time of calibration, and the calibration accuracy deteriorates.

The In dose (ion implantation amount) is preferably 1 × 10 ¹⁴ cm ⁻² or more and 1 × 10 ¹⁵ cm ⁻² or less. If the dose is less than 1 × 10 ¹⁴ cm ⁻² , the In concentration of the ion implantation layer 2 is low, and the In ion intensity necessary for calibration cannot be obtained. If the dose exceeds 1 × 10 ¹⁵ cm ⁻² , the concentration distribution of the ion implantation layer 2 is different from the expected distribution obtained by simulation. As a result, a standard sample having a planned concentration distribution is obtained. It becomes difficult to form.

In the first embodiment, the ion implantation layer 2 is formed with a dose amount of 1 × 10 ¹⁵ cm ⁻² and an ion implantation energy of 5 keV.

Next, referring to FIG. 1B, the silicon substrate 1 on which the ion implantation layer 2 is formed is placed in a secondary ion mass spectrometer together with a test sample. Next, at room temperature, the surface of the silicon substrate 1 was irradiated with oxygen ions 3 at a dose of 1 × 10 ¹⁷ cm ⁻² to 1 × 10 ¹⁸ cm ⁻² . With this irradiation of oxygen ions 3, referring to FIG. 1 (c), oxygen enters the silicon substrate 1 from the surface of the silicon substrate 1, and an oxygen-containing layer 5 containing oxygen combined with silicon is formed.

  The oxygen-containing layer 5 is formed by the irradiation of the oxygen ions 3 and at the same time, the redistribution of impurities (In) in the ion-implanted layer 2 occurs, and the redistribution layer 4 in which In is accumulated near the surface of the oxygen-containing layer 5. Is formed.

  The oxygen ions 3 used for irradiation are generated by the primary ion generator of the secondary ion mass spectrometer, and the irradiation energy and irradiation angle (angle formed by the irradiation ions and the normal of the surface of the silicon substrate 1) are set as necessary. can do. The oxygen ions 3 include various oxygen molecular ions in addition to oxygen atom ions.

In the first embodiment, the irradiation energy of the oxygen ions 3 is preferably set to 250 eV or more and 5 keV or less in terms of O ₂ ⁺ irradiation. As will be described later, in order to form the redistribution layer 4 in the transition region or in a shallower region, it is desirable to reduce the irradiation energy. However, if the irradiation energy is lower than 250 eV, surface reflection tends to occur, and it becomes difficult to precisely control the dose of oxygen ions. Further, when the irradiation energy exceeds 5 keV, the oxygen-containing layer 5 becomes deep and a shallow redistribution layer cannot be formed.

  In the first embodiment, a plurality of regions of one silicon substrate 1 are irradiated with oxygen ions 3 with different irradiation energies. Then, a silicon substrate having a redistribution layer 4 and an oxygen-containing layer 2 formed by irradiation with oxygen ions 3 having different irradiation energies in these regions was prepared as a standard sample for depth calibration.

  Next, the impurity concentration distribution of the standard sample according to the first embodiment manufactured by the above-described process will be described.

  FIG. 2 is a diagram illustrating the impurity concentration distribution of the standard sample according to the first embodiment of the present invention, and represents the depth distribution of the In concentration of the standard sample observed by secondary ion mass spectrometry. In FIG. 2, the curves that pass through ○, ●, Δ, ▲, □, ■, and ◇ indicate the irradiation energy of oxygen ions 3 in the standard sample preparation process at 250 eV, 500 eV, 1 keV, 1.5 keV, 2 keV, 3 keV, and 5 keV, respectively. Corresponds to the impurity concentration distribution in each region. The horizontal axis of FIG. 2 is calibrated using the above-described conventional standard sample prepared by recrystallization after ion implantation of impurities.

  Referring to FIG. 2, the In (impurity) concentration distribution in the depth direction of the standard sample of the first embodiment has a concentration peak in the vicinity of the surface, and after passing through a minimum occurring in a deep portion following the peak, It increases slowly with depth.

  This concentration peak is remarkable when the irradiation energy is low. For example, in the case of irradiation energy of 1 keV expressed by a curve passing through Δ, it has a peak concentration at a depth of about 0.5 nm from the surface and becomes a minimum at a depth of about 1 nm, and then gradually increases with the depth. It is clearly recognized that

  The peak concentration position of the In concentration and the concentration minimum position following the peak depend on the irradiation energy of the oxygen ions 3. For example, when the irradiation energy of oxygen ions 3 is as low as 250 eV, it has a peak concentration at a position shallower than 0.2 nm and a concentration minimum at a depth of approximately 0.7 nm. The peak concentration position and the minimum concentration position move to a deeper position as the irradiation energy increases, and the peak concentration position in the case of 1.5 keV irradiation energy is approximately 1.0 nm in depth, and the minimum concentration position is approximately the depth. A depth of 1.8 nm is reached.

  When the irradiation energy exceeds 1.5 keV, the In concentration distribution changes from a sharp peak to a plateau. This plateau formation region has a depth of approximately 2.0 nm from the surface when the irradiation energy is 2 keV, and has a concentration minimum at a depth of approximately 2.3 nm following the plateau. In a region deeper than the position where the concentration is minimum, the concentration gradually increases with the depth.

  The plateau formation region and the concentration minimum position become deeper as the irradiation energy increases, and reach a depth of 4.5 nm and a depth of 5.6 nm, respectively, at an irradiation energy of 5 keV.

  In the In concentration distribution having such a peak or plateau and the subsequent minimum concentration, In in the ion implantation layer 2 is redistributed and accumulated in the vicinity of the surface, and as a result, the redistribution layer 4 containing high concentration of In. Strongly suggests that That is, it is suggested that In atoms in the ion-implanted layer ion-implanted at a substantially uniform concentration from the surface to a depth of about 10 nm or more move from the vicinity of the minimum value to the vicinity of the surface and accumulate near the surface.

  The formation mechanism of the impurity redistribution layer 4 by the above-described oxygen ion irradiation has not been clarified yet. However, the inventor of the present invention shows that the formation of the redistribution layer 4 discharges impurities from the oxygen-containing layer 5 because of the relationship with the transition region observed when oxygen ions are irradiated as primary ions in secondary ion mass spectrometry. I guess it is due to being. Hereinafter, the relationship between the transition region and the redistribution layer 4 will be described.

FIG. 3 is a diagram showing a transition region of a silicon substrate, and shows the ionic strength of ⁴⁴ SiO ⁺ detected as secondary ions when the silicon substrate is irradiated with oxygen ions 3 as primary ions at different irradiation energies. . In FIG. 3, the curves passing through ◯, ●, △, ▲, □, ■, and ◇ are secondary detected when the irradiation energy is 250 eV, 500 eV, 1 keV, 1.5 keV, 2 keV, 3 keV, and 5 keV, respectively. Corresponds to ionic strength. In addition, the vertical axis | shaft of FIG. 3 is normalized by making ion intensity detected in depth 10nm into a reference value. The horizontal axis in FIG. 3 is calibrated using a conventional standard sample prepared by recrystallization after impurity ion implantation, as in FIG.

Referring to FIG. 3, the ionic strength of ⁴⁴ SiO ⁺ generated as secondary ions by irradiating a silicon substrate with oxygen ions has a minimum at a shallow position of 2 nm or less from the surface, and becomes deeper than the minimum position. It gradually increases and approaches a constant value that is the standard value for normalization.

The ion intensity distribution (that is, the time transition of the secondary ion intensity) of the secondary ions ( ⁴⁴ SiO ⁺ ) having a minimum near the surface is a reactive ion such as an oxygen ion or a cesium ion. Time periods and regions where secondary ion intensity decreases near this surface, often observed in secondary ion mass spectrometry, are widely known as transition periods and transition regions.

  The generation mechanism of the transition region is understood as follows. When the surface of the silicon substrate is irradiated with oxygen ions, the silicon in the vicinity of the surface is oxidized and the oxygen-containing layer 5 is formed. Since the oxygen-containing layer 5 maintains a stable steady state after a sufficient transition period, the secondary ion yield is maintained constant in a region deeper than the transition region, and the ionic strength (figure shown) A constant secondary ion intensity of 3).

However, since the amount of supplied oxygen is small immediately after irradiation, the steady-state oxygen-containing layer 5 is not formed, and an unstable unsteady-state oxygen-containing layer 5 with a low oxygen content is formed. For this reason, the intensity of secondary ions detected at the beginning of irradiation (ie, the transition region) is observed to be weaker than the steady state in the intensity of oxidized silicon ions (for example, ⁴⁴ SiO ⁺ ). This region, that is, the depth at which the secondary ion intensity recovers to the reference value is observed as a transition region. A transition region similarly occurs when other reactive ions are irradiated as primary ions.

2 and 3, the depth at which the ion intensity of the secondary ions ⁴⁴ SiO ⁺ shown in FIG. 3 recovers to the reference value (steady value) is almost at the minimum position of the In concentration distribution shown in FIG. It corresponds. For example, when the irradiation energy is 250 eV, 500 eV, 1 keV, 1.5 keV, 2 keV, 3 keV, and 5 keV, the depth at which the ion intensity recovers to the reference value in FIG. 3 is 0.1 nm or less, approximately 0.5 nm, respectively. It is approximately 1.0 nm, 1.7 nm, approximately 2.3 nm, approximately 3.8 nm, and approximately 5.5 nm, and the irradiation energy shown in FIG. 2 has the minimum depth in the same In concentration distribution, approximately 0.7 nm, approximately The values are in good agreement with 0.8 nm, approximately 1.0 nm, 1.8 nm, approximately 2.3 nm, approximately 3.8 nm, and approximately 5.6 nm except for 250 eV. In addition, it is speculated that the reason why the values do not agree well in the case of 250 eV is that the depth is too shallow and it is difficult to perform precise observation by secondary ion mass spectrometry.

^The depth at which the ionic strength of ⁴⁴ SiO ⁺ recovers corresponds to the depth of the transition region. On the other hand, it can be considered that the ion-implanted In atoms before the oxygen ion irradiation are distributed at a substantially constant concentration in a shallow region having a depth of about 20 nm. Therefore, the above-mentioned fact that the depth of the minimum concentration of the In concentration distribution shown in FIG. 2 and the depth of the transition region shown in FIG. 3 agree well is redistributed on the surface side at a high speed in the transition region. On the other hand, it is suggested that in a region deeper than the transition region, for example, a region having a depth of 6 nm or more where the oxygen-containing layer is in a steady state, it is redistributed toward the transition region at a low speed.

  That is, In atoms quickly accumulate in the vicinity of the surface in the transition region to form a concentration peak in the vicinity of the surface, and reduce the In concentration at a deep position in the transition region. In moves from a region deeper than the transition region to the transition region where the In concentration is reduced, In redistribution occurs. However, the redistribution speed in the region deeper than the transition region is smaller than in the transition region. For this reason, more In is supplied to the vicinity of the surface than In supplied from a deeper region near the bottom of the transition region. As a result, a minimum In concentration occurs near the bottom of the transition region. The inventor of the present invention estimates the redistribution of In in this way.

  The inventor of the present invention believes that the redistribution of impurities due to such oxygen ion irradiation, for example, the redistribution of In, is caused by the impurities being discharged from the oxygen-containing layer 5. When the silicon substrate 1 is irradiated with oxygen ions 3, an oxygen-containing layer 5 containing oxygen is formed on the surface of the silicon substrate 1. In the oxygen-containing layer 5, the irradiated oxygen and silicon are combined to form silicon oxide. In has a small solid solubility limit in silicon oxide, and is discharged from the oxygen-containing layer 5 together with the formation of silicon oxide. For this reason, redistribution of In near the surface with little oxygen occurs.

In the first embodiment of the present invention, the dose of the oxygen ions 3 to be irradiated is set to 1 × 10 ¹⁷ cm ⁻² to 1 × 10 ¹⁸ cm ⁻² . With this dose amount, the surface of the silicon substrate 1 reaches a steady state or a state close to a steady state. At this time, the oxygen-containing layer 5 formed by oxygen ion irradiation has an oxygen concentration sufficient to form a silicon oxide layer in a region deeper than the depth of the transition region, and the oxygen concentration toward the surface in the transition region. Decrease. As a result, In is discharged from the silicon oxide formed in the deep part and accumulates near the surface. The inventors of the present invention consider that a redistribution layer is formed in this way.

  As described above, the impurity concentration distribution of the redistribution layer 4 is strongly related to the transition region. This transition region is determined by the irradiation energy of oxygen ions if the irradiation angle and dose of oxygen ions are the same. In other words, the impurity concentration distribution of the redistribution layer 4 is determined by the irradiation energy of oxygen ions.

  Therefore, for example, by measuring the impurity concentration distribution of the redistribution layer 4 once using high resolution Rutherford backscattering spectroscopy, the relationship between the irradiation energy of oxygen ions and the impurity concentration distribution of the redistribution layer 4 is measured. Can be clarified. Once the relationship between the irradiation energy and the impurity concentration distribution is clarified, the impurity concentration distribution of the redistribution layer 4 can be known from the irradiation energy without actually measuring.

  Next, secondary ion mass spectrometry in which the depth is calibrated using the standard sample of the first embodiment will be described.

In this analysis, As ⁺ was ion-implanted into a silicon substrate with an implantation energy of 3 keV and a dose of 1 × 10 ¹⁵ cm ⁻² , and then a test sample subjected to heat treatment at 700 ° C. for 10 seconds was subjected to secondary ion mass spectrometry. The concentration distribution of As in the test sample in the depth direction was measured.

First, as described above, the test sample was placed in the secondary ion mass spectrometer together with the standard sample. Next, the test sample was irradiated with primary ions perpendicularly to the sample surface, and the time transition of the secondary ion intensity (AsSi ⁻ ) of the test sample was measured. Here, oxygen ions are used as primary ions, but other reactive ions such as cesium ions can also be used.

  The standard sample used was prepared through the above-described In ion implantation into the silicon substrate 1 and the oxygen ion irradiation process. Oxygen ion irradiation during the standard sample preparation process was performed in the apparatus used for secondary ion mass spectrometry of the test sample. In this case, the preparation of the standard sample (oxygen ion irradiation step) may be completed at the time of calibration, and may be performed before or after the secondary ion mass spectrometry of the test sample. Further, the standard sample may be prepared in advance using different oxygen ion irradiation apparatuses.

Next, after measuring the secondary ion (AsSi ⁻ ) intensity of the test sample, the same primary ion beam as the primary ion used to measure the test sample, for example, ion species, irradiation energy, irradiation angle (here, the sample surface) The standard sample was irradiated with the same primary ions (perpendicular to the standard), and the time course of the intensity of the impurity (In) ion that is the secondary ion of the standard sample was measured. The time course of the In ion intensity was measured for each of a plurality of regions formed on the surface of the standard sample by oxygen ion irradiation with different irradiation energies.

  Next, the time axis is converted into depth assuming that the time history of the In ion intensity of the standard sample corresponds to the In concentration distribution of the standard sample. As is well known, in the secondary ion analysis method, the surface of the test sample is etched by temporary ion irradiation, and the secondary ion intensity of the test element released from the test sample surface exposed by etching is measured. Observe the passage of time. Therefore, the secondary ion intensity observed at the elapse of an arbitrary time corresponds to the concentration of the test element at the depth etched up to that time. In other words, the time course of the secondary ion intensity represents the concentration distribution of the test element with the time axis as the depth. However, since the relationship between the time axis and the depth changes when there is a time variation in the etching rate, an accurate concentration distribution cannot be obtained unless this relationship is calibrated.

  In the first embodiment, when the etching time (that is, the time axis) is converted into depth, a standard sample having a known In concentration distribution is subjected to secondary ion analysis, and the concentration distribution (of secondary ion intensity) is calculated. The relationship between the time axis and the depth is determined so that (time passage) matches the known In concentration distribution. Specifically, since the peak of the In concentration or the depth of the plateau, or the depth at which the In concentration is minimized, is known, these depths are used as markers when converting the time axis into depth. That is, the time axis is converted into a depth, with the time when the marker located at the known depth appears during the lapse of time of the In ion intensity as the known depth.

  Next, secondary ion mass spectrometry of the test sample is performed by applying the converted value from the time axis to the depth calibrated by the standard sample to the conversion of the secondary ion intensity of the test sample from the time axis to the depth. The depth at is calibrated with a standard sample. Therefore, an accurate concentration distribution in the depth direction of the test element in the test sample is measured.

  FIG. 4 is a diagram for explaining the effect of the first embodiment of the present invention, and shows the result of secondary ion mass spectrometry of the As concentration distribution of the test sample described above. Curve B is obtained by the conventional method of converting the time axis into depth using the concentration distribution calibrated using the standard sample of the first embodiment, and curve C using the depth of the depression formed by sputtering. Represents the concentration distribution. Curve A is a concentration distribution measured using the high-resolution Rutherford backscattering method, and is shown together with curves B and C as representing a precise concentration distribution in a shallow region.

  Referring to curves A, B, and C in FIG. 4, the measured As concentration distributions all have a peak near the surface, and a second peak at a deeper position (5 to 6 nm depth). Have. With reference to the curves A and B, the secondary ion analysis result calibrated using the standard sample of the first embodiment shows that the peak position (depth) near the surface and the second peak position have high resolution. It is in good agreement with the results measured by Rutherford backscattering method. Also, the valley between the two peaks appears clearly, similar to the results of the high resolution Rutherford backscattering method.

  On the other hand, with reference to curve C, the concentration distribution measured by the conventional method is shifted to a position where the peak position near the surface is deeper than curves A and B. As a result, the valley is also unclear and the second peak position is also unclear. The movement of the peak position in the vicinity of the surface is caused by the fact that the time axis interval corresponding to the depth interval becomes longer in the transition region because the sputtering speed in the transition region is slower than that in the deep region.

  As described above, in the secondary ion mass spectrometry of the first embodiment, the movement of the peak is extremely small even in the shallow region near the surface as compared with the conventional method. As a result, the concentration distribution in the depth direction is also obtained in the shallow transition region. It can be measured accurately.

  The second embodiment of the present invention relates to a standard sample having different impurity concentration distributions produced by changing the irradiation angle of oxygen ions.

FIG. 5 is a diagram showing the oxygen ion irradiation angle dependence of the transition region of a silicon substrate. When the surface of the silicon substrate is irradiated with oxygen ions at different irradiation angles, the ionic strength of ⁴⁴ SiO ⁺ detected as secondary ions. Represents. The ionic strength on the vertical axis is normalized using the secondary ionic strength at a depth of 50 nm as a reference value. Further, the depth of the horizontal axis is calibrated by the same method as the horizontal axis of FIG. 3 of the first embodiment. Note that the irradiation energy of oxygen ions is constant at 500 eV in terms of the irradiation energy of O ₂ ⁺ .

Referring to FIG. 5, the secondary ion intensity ( ⁴⁴ SiO ⁺ ion intensity) observed when the silicon substrate is irradiated with oxygen ions at an irradiation angle is, for example, the secondary ion intensity at an irradiation angle of 60 °. Referring to (curve passing through ▲), it has a minimum at a depth of several nanometers and recovers to a reference value as it becomes deeper. The region where the secondary ion intensity decreases corresponds to the transition region where oxygen is insufficient as described with reference to FIG.

  The minimum depth of the secondary ion intensity and the depth at which the secondary ion intensity is restored to the reference value (steady state value) become deeper as the irradiation angle (angle formed between the normal of the silicon substrate and the irradiation ions) increases. This is the same as the case where the deep transition region is formed with the increase of the irradiation energy of oxygen ions in the first embodiment described with reference to FIG. 3 except that the irradiation energy is changed to the irradiation angle. . Therefore, by controlling the irradiation angle of oxygen ions, the characteristics of the transition region, for example, the depth at which the transition region is formed can be precisely controlled as in the first embodiment.

  The standard sample manufacturing process of the second embodiment will be described with reference to FIG. 1A until the process of forming an In ion implantation layer 2 by ion-implanting impurities such as In into the silicon substrate 1. This is the same as in the first embodiment.

Next, the silicon substrate 1 on which the In ion-implanted layer is formed is irradiated with oxygen ions at 500 eV in terms of O ₂ ⁺ irradiation energy. At this time, the plurality of regions of the silicon substrate 1 are irradiated with different irradiation angles, for example, irradiation angles of 0 °, 20 °, 25 °, and 30 °. Thereby, referring to FIG. 1C, as in the first embodiment, a redistribution layer 4 in which In is redistributed on the surface of the silicon substrate is formed.

  FIG. 6 is a diagram illustrating the impurity concentration distribution of the standard sample according to the second embodiment of the present invention, and represents the depth distribution of the In concentration of the standard sample observed by secondary ion mass spectrometry. In FIG. 6, the curves passing through ◯, △, ▲, and □ correspond to the impurity concentration distributions when the irradiation angle of oxygen ions 3 in the standard sample preparation process is 0 °, 20 °, 25 °, and 30 °, respectively. I'm. The horizontal axis in FIG. 2 is calibrated using the above-described conventional standard sample prepared by recrystallization after ion implantation of impurities.

  Referring to FIG. 6, the peak of the In concentration distribution is at a depth of approximately 1.2 nm when the irradiation angle is 0 °, and the depth is 1 as the irradiation angle increases to 20 °, 25 °, and 30 °, respectively. It moves to a deep position such as 0.7 nm, 2.1 nm, and 3.3 nm. Thus, the In concentration distribution of the redistribution layer 4 is controlled by controlling the irradiation angle of oxygen ions.

  As described in the description of the first embodiment, the In concentration distribution of the redistribution layer 4 is strongly related to the characteristics of the transition region determined by the irradiation energy of oxygen ions, particularly the depth thereof. On the other hand, the results of the second embodiment described above show that the characteristics of the transition region, particularly the depth, are precisely controlled by the irradiation angle of oxygen ions, and the In concentration distribution is precisely controlled by the irradiation angle. It is made clear.

  These facts indicate that the impurity concentration distribution of the redistribution layer 4 strongly depends on the irradiation energy of oxygen ions and the irradiation angle of oxygen ions. In the second embodiment, the oxygen ion irradiation energy and dose are kept constant, and the impurity concentration distribution of the recrystallized layer is controlled by the oxygen ion irradiation angle. Thereby, the In concentration distribution of the redistribution layer 4 can be controlled with good reproducibility. For this reason, since a standard sample having an impurity concentration distribution that can be regarded as the same in practical use is easily produced, depth calibration of secondary ion mass spectrometry can be calibrated easily and precisely.

  The third embodiment of the present invention relates to a standard sample using Ga as an impurity instead of In.

In the third embodiment, Ga is first ion-implanted into a silicon substrate at an implantation energy of 5 KeV and a dose of 5 × 10 ¹⁴ cm ⁻² to form an ion-implanted layer into which Ga is ion-implanted. Next, as in the first embodiment, oxygen ions are irradiated to the silicon substrate on which the ion-implanted layer is formed, thereby forming the redistribution layer 4 in which Ga is accumulated in the vicinity of the silicon substrate surface. As a result, a standard sample having a Ga redistribution layer 4 was produced.

  The standard sample of the third embodiment has a plurality of regions irradiated with oxygen ions with different irradiation energies. Accordingly, in this standard sample, redistribution layers having different Ga concentration distributions are formed in a plurality of regions, as in the first embodiment. Such a standard sample can be used as a concentration distribution reference for depth calibration in the same manner as the secondary ion analysis of the first embodiment.

  FIG. 7 is a diagram showing an impurity concentration distribution of a standard sample according to the third embodiment of the present invention, and shows a Ga concentration distribution of a recrystallized layer formed by irradiating oxygen ions with an irradiation energy of 2 keV.

  Referring to FIG. 7, the Ga concentration distribution agrees well with the In concentration distribution of the recrystallized layer formed by irradiating oxygen ions with the irradiation energy of 2 keV shown in FIG. For example, the minimum concentration can be seen near a depth of 2.3 nm. Thus, the impurity concentration distributions in the In and Ga redistribution layers are in good agreement if the irradiation energy, irradiation angle, and dose of the oxygen ions used to form the redistribution layer are the same.

  This is because Ga is discharged from an oxygen-containing layer having a high oxygen concentration and is accumulated in a transition region formed in the vicinity of the surface of the oxygen-containing layer (that is, the surface of the silicon substrate). This suggests that the redistribution layer 4 has been formed. Therefore, Ga can be used as an impurity of the standard sample in place of In in the first and second embodiments. At this time, the Ga distribution of the redistribution layer is controlled using one or both of the irradiation energy and the irradiation angle of oxygen ions.

  As described above, the impurity concentration distribution in the redistribution layer is controlled by the irradiation energy of oxygen ions in the first embodiment of the present invention and by the irradiation angle of oxygen ions in the second embodiment of the present invention. Further, the impurity concentration distribution can be controlled by controlling both the irradiation energy and the irradiation angle.

  In the first and second embodiments, In is used as the impurity of the standard sample, and Ga is used in the third embodiment. Other impurity elements discharged from the oxygen-containing layer 5 can also be used as impurities of these standard samples.

  By applying the present invention to secondary ion mass spectrometry using reactive primary ions, the concentration distribution in the vicinity of the surface layer of the shallow impurity doped layer of the semiconductor device can be accurately measured.

Sectional drawing of manufacturing process of standard sample of 1st Embodiment of this invention The figure showing the impurity concentration distribution of the standard sample of 1st Embodiment of this invention Diagram showing transition region of silicon substrate The figure explaining the effect of 1st Embodiment of this invention Diagram showing oxygen ion irradiation angle dependence of transition region of silicon substrate The figure showing the impurity concentration distribution of the standard sample of 2nd Embodiment of this invention The figure showing the impurity concentration distribution of the standard sample of 3rd Embodiment of this invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Silicon substrate 2 Ion implantation layer 3 Oxygen ion 4 Redistribution layer 5 Oxygen content layer

Claims (7)

An oxygen-containing layer formed by irradiating a silicon layer having an impurity ion-implanted layer formed thereon with oxygen ions;
During the irradiation with the oxygen ions, the impurities accumulate near the surface of the silicon layer shallower than the oxygen-containing layer, so that the maximum concentration and the heel portion of the concentration of the impurities from the surface of the silicon layer toward the depth direction. A sample for depth calibration of secondary ion mass spectrometry, characterized by having a redistribution layer formed so that.   The sample for depth calibration of secondary ion mass spectrometry according to claim 1, wherein the impurity is indium or gallium.   3. The depth calibration of secondary ion mass spectrometry according to claim 1, wherein the redistribution layer of the impurity is formed in a region shallower than a depth of 5 nm from the surface of the calibration sample. sample. In the method of manufacturing a sample for depth calibration of secondary ion mass spectrometry, comprising a silicon layer in which an impurity concentration distribution is formed in the depth direction,
Ion-implanting the impurities into the silicon layer;
A step of irradiating the silicon layer with oxygen ions, and accumulating the implanted ions in the vicinity of the surface of the silicon layer during the irradiation of the oxygen ions. Of manufacturing a depth calibration sample.   The method for producing a sample for depth calibration of secondary ion mass spectrometry according to claim 4, wherein the impurity is indium or gallium. The ion implantation energy of the impurity is 5 keV or more and 10 keV or less,
6. The sample for depth calibration of secondary ion mass spectrometry according to claim 5, wherein an ion implantation dose of the impurity is 1 × 10 ¹⁴ cm ⁻² or more and 1 × 10 ¹⁵ cm ⁻² or less. Manufacturing method. Storing a sample to be analyzed and a silicon substrate into which indium or gallium is ion-implanted in a secondary ion mass spectrometer;
Irradiating the sample to be analyzed with primary ions, and measuring a time change in ion intensity of ions including a test element released as secondary ions from the sample to be analyzed;
A step of irradiating the silicon substrate with oxygen ions at an irradiation energy of 250 eV or more and 5 keV or less in terms of O ₂ ⁺ irradiation;
Irradiating the silicon substrate irradiated with the oxygen ions with the primary ions, and measuring a temporal change in ion intensity of ions containing indium or gallium;
Depth calibration step for converting the time axis of the time change of the ion intensity of the ion containing the test element into the depth from the sample surface to be analyzed with reference to the time change of the ion intensity of the ion containing indium or gallium A secondary ion mass spectrometry method comprising:

Images