JP4059345B2 - Crystal structure analysis method - Google Patents

Description

  The present invention relates to a structure analysis method for a semiconductor thin film or a multiple quantum well structure.

  Conventionally, the importance of the quantity “critical film thickness” (or “critical strain”) is well known in the production of strained epitaxial layers and strained superlattice structures using materials such as InGaAs / GaAs and Si / SiGe. ing. The evaluation of the critical film thickness and the evaluation of the lattice relaxation process in the region exceeding the critical film thickness were important evaluation objects. X-ray diffraction has been frequently used for evaluation of critical film thickness and lattice relaxation process as well as PL (photoluminescence) measurement and TEM (transmission electron microscope) analysis.

  The critical film thickness values evaluated by these various evaluation methods have been discussed by comparing with predicted values from theories such as the Matthews-Blakeslee (MB) model and the People-Bean (PB) model. However, the degree of coincidence has conventionally varied according to experimental examples, and the cause has been a problem. Furthermore, it cannot be said that the correlation between the critical film thickness values obtained from the above evaluation methods is sufficient, and the evaluation detection sensitivity has been regarded as a problem.

  Under such circumstances, an experimental fact has been reported that the detection sensitivity of dislocations by X-ray diffraction is inferior to the detection sensitivity when other evaluation methods such as PL measurement are used (Non-Patent Document 1, 2). From this, it was pointed out that the critical film thickness value obtained by X-ray diffraction is larger than other critical film thickness values obtained from PL measurement and the like, and a correct value cannot be obtained. . And it has been pointed out that X-ray diffraction is not suitable for detection of initial structural deterioration when the density of defects and the like is small.

  However, the problem of the evaluation method by X-ray diffraction is not widely recognized. In particular, in the evaluation of critical film thicknesses and detection of structural deterioration in other material systems, there have been some reports recently evaluated by applying X-ray diffraction without fully recognizing the above problems.

  Furthermore, even when the above problems are fully recognized, X-ray diffraction is not used for evaluation, and other evaluation methods with higher detection sensitivity are used to evaluate critical film thickness and detect structural deterioration. Was only taken. In other words, no further examination has been made on the point of "why X-ray diffraction cannot provide detection sensitivity", and methods for improving detection sensitivity from the standpoint of X-ray diffraction have been completely studied so far. There wasn't.

In the X-ray diffraction evaluation described above, evaluation was mainly performed using the rocking curve measurement method, but methods such as reciprocal lattice mapping were later developed. However, even in these newly developed methods, no remarkable improvement effect was observed with respect to the above-mentioned problem of “detection sensitivity of structural deterioration compared to other measurement methods”.
IJ Fritz, PL Gourley, and LR Dawson, Appl.Phys. Lett. 51, 1004-1006 (1987) PL Gourley, IJ Fritz, and LR Dawson, Appl.Phys. Lett. 52, 377-379 (1988)

  As described above, structural degradation such as evaluation of critical film thickness and evaluation of lattice relaxation process in regions beyond the critical film thickness are important for the production of strained epitaxial layers and strained superlattice structures. When evaluating using the diffraction method, there is a problem that the detection sensitivity is poor compared to the case of using other PL measurement methods.

  An object of the present invention is to provide a crystal structure analysis method capable of improving detection sensitivity as compared with the prior art in solving the above-described problems and evaluating structural deterioration using an X-ray diffraction method. .

In order to solve the above-mentioned problems, the invention of claim 1 is directed to a crystal structure analysis method for analyzing a structure of a multilayer structure crystal produced by epitaxial growth on an arbitrary substrate by X-ray diffraction. Obtained by the first step of measuring a plurality of diffraction profiles having the same reflection index (hkl), with the other measurement conditions being the same under the condition that only the aperture angle condition is changed, and the first step. A comparison profile in which the plurality of (hkl) profiles are displayed in an overlapping manner by aligning the diffraction intensities and diffraction angles of the peak tops of the respective substrate diffraction peaks in the plurality of (hkl) profiles from the plurality of (hkl) profiles. The diffraction peak from the epitaxial multilayer structure crystal is obtained by using the second step for fabricating the comparative profile and the comparative profile obtained in the second step. By measuring the difference in diffraction intensity between serial plurality of (hkl) profile, a third step of determining the degree of structural deterioration of the multilayered crystal based on the greater the degree of structural deterioration larger the difference It is a structure analysis method of the crystal | crystallization characterized by including.
The invention of claim 2 is the crystal structure analysis method according to claim 1, wherein the substrate has a plane index (hkl) and the reflection index is the same as the plane index (hkl), or (ah, ak, al) is a reflection index (a is an integer).
According to a third aspect of the present invention, in the crystal structure analysis method according to the first or second aspect, in the first step, at least “aperture angle conditions of a plurality of light receiving systems used for measuring a plurality of diffraction profiles” Includes a “wide aperture angle condition” and a “narrow aperture angle condition”. The “wide aperture angle condition” refers to whether a light receiving slit having an aperture angle of 0.25 degrees or more is used immediately before the X-ray receiver. Alternatively, the “narrow aperture angle condition” is a measurement condition in an arrangement without using a light receiving slit. The “narrow aperture angle condition” is a three-axis arrangement condition by inserting an analyzer crystal immediately before the X-ray receiver, and an aperture angle of 0. It is characterized by the measurement condition of 01 degrees or less.

  According to the present invention, each of the steps described above causes the diffraction X-ray spread reflecting the imperfection of the multilayer structure crystal accompanying the generation of defects to be the intensity of the two X-ray diffraction profiles with different aperture angle conditions of the light receiving system. The difference can be measured easily.

  According to the present invention, X-ray diffraction analyzes structural deterioration such as evaluation of critical film thickness and evaluation of lattice relaxation process in a region exceeding the critical film thickness, which are important in the production of strained epitaxial layers and strained superlattice structures. When the evaluation is performed using the method, it is possible to solve the problem that the detection sensitivity is poor as compared with the case of using another PL measurement method that has been a problem in the past. In evaluating the structural deterioration using the X-ray diffraction method, the detection sensitivity can be improved as compared with the conventional method. In particular, it is useful for investigating the presence and extent of structural degradation in strained epitaxial thin films and strained superlattice structures.

  In this example, a GaInNAs / GaAs double quantum well (DQW) structure sample produced by a MOVPE growth method (metal organic vapor phase epitaxy) on a GaAs substrate having a plane index (001) is used in the present invention. An example of structural evaluation by applying is described below.

Four GaInNAs / GaAs DQW samples were used in this example. All four samples are on a (001) GaAs substrate.
GaAs buffer layer (100nm thickness)
GaInNAs / GaAs DQW structure GaAs clad layer (film thickness 100nm)
It has a layer structure that grows sequentially. The composition of the GaInNAs layer, which is the well layer used for the DQW structure portion, is common, and the composition Ga _0.7 In _0.3 N _0.02 As _0.0 having a compressive strain of 2.2% in terms of strain _{. 98} was used. The GaAs barrier layer has a common film thickness of 11 nm.

In the four samples, the film thickness d of the GaInNAs layer, which is a well layer, was sequentially changed. The thickness d of the GaInNAs layer in each sample is
Sample A: d = 3.2 nm
Sample B: d = 4.5 nm
Sample C: d = 6.0 nm
Sample D: d = 9.0 nm
It is.

  First, as a representative example of other measurement methods, FIG. 4 shows PL measurement results of the above four samples. As the film thickness of the GaInNAs layer, which is a well layer, increases, the PL peak wavelength shifts to the longer wavelength side. This is because the quantum confinement effect of the DQW structure becomes weak, and is a qualitatively valid result. However, it can be seen that as the thickness of the GaInNAs layer is increased, the PL intensity is remarkably lowered, and the PL half-width is also significantly increased accordingly. This result shows that as the thickness of the GaInNAs layer increases between the four samples, the critical thickness is exceeded and defects are introduced into the DQW structure, and the structural deterioration progresses. From this, it can be seen that structural degradation is clearly detected in the evaluation by PL measurement.

  In contrast, FIG. 5 and FIG. 6 show the results of measuring the above four samples by the 004 X-ray diffraction rocking curve measurement method that has been often used conventionally. The measurement was performed by a θ-2θ method scan in a measurement arrangement that does not use a light receiving slit immediately before the X-ray receiver. 5 and 6 also show calculation profiles fitted by simulation using kinematic theory. Further, structural parameters obtained by fitting, that is, structural parameters of the DQW structure obtained by fitting analysis by simulation of the diffraction profiles shown in FIGS. 5 and 6 (the strain amount ε and the thickness d of the well layer and the barrier layer) Values) are summarized in Table 1 below.

  From the results in Table 1, it can be seen that the amount of distortion almost as designed is obtained. In addition, the diffraction intensity of the profiles in FIGS. 5 and 6 and the broadening of the peak shape are in good agreement with the calculated profiles, so that it can be seen that no significant signs indicating structural deterioration are observed.

  Next, evaluation results by reciprocal lattice mapping measurement using 224 reflection, which is often used in X-ray diffraction evaluation in recent years, are shown in FIGS. In the reciprocal lattice mapping measurement, measurement was performed in a measurement configuration in which a 022 reflection Ge analyzer crystal is inserted in front of an X-ray receiver, which is usually used. As can be seen from FIG. 7 and FIG. 8, in all four samples, the position of the satellite spot from the DQW structure coincides with the position of the spot of the substrate peak and the coordinate on the horizontal axis. From this, it can be concluded that the DQW structure is elastically distorted with respect to the GaAs substrate in all four samples, and no lattice relaxation occurs. In conventional reciprocal lattice mapping measurement evaluation, it has been concluded that when such a result is obtained, there is usually no significant structural deterioration.

  As described above, as a result of applying two typical evaluation analysis methods normally used in the conventional X-ray diffraction evaluation to the four samples described above, remarkable data indicating structural deterioration is obtained in any case. I found that there was no. However, in PL measurement, as described above, structural deterioration has been confirmed. From this, it can be seen that conventional X-ray diffraction evaluation is less sensitive to structural deterioration than evaluation by PL measurement, and the problems that have been pointed out in the past can be reconfirmed.

  On the other hand, the result of applying the present invention to the above four samples is shown below. FIG. 1 and FIG. 2 show two types of reflection X-ray diffraction profiles of reflection index 004 measured using the aperture angle conditions of different X-ray receivers, and the peak tops of the substrate peaks observed in both profiles. Each sample is shown in an overlapping manner by matching the positions so that they match.

  Of the two types of 004 reflection X-ray diffraction profiles used here, one is a profile measured under an extreme wide aperture angle condition in which no light receiving slit is placed immediately in front of the light receiver. It is the same as the profile shown in FIGS. Another profile is a profile measured in an arrangement that is an extremely narrow aperture angle condition in which an analyzer crystal for reciprocal lattice mapping measurement is inserted immediately before the light receiver.

  In sample A, these two types of profiles are almost the same, and there is no difference in diffraction intensity, profile shape, or the like. On the other hand, in Samples B to D, there is no difference in the shape of the profile from the DQW structure, but there is a significant difference in the diffraction intensity. Furthermore, it can be seen that the magnitude of this difference increases as the well layer thickness increases.

  In order to quantify the difference in diffraction intensity, the point where the diffraction intensity in the profile shape from the DQW structure is maximized is taken as a representative point, and data quantified as the intensity ratio of the diffraction intensity at this point is constructed. FIG. 3 shows the sample dependency of the constructed data. FIG. 3 also shows the sample dependence of the PL intensity obtained from FIG.

  As can be seen from FIG. 3, the sample dependency shown in the X-ray diffraction evaluation results (invention) of FIGS. 1 and 2 is very similar to the sample dependency obtained from the PL evaluation of FIG. It can be seen that the sample dependence has a strong correlation. This indicates that both the X-ray diffraction evaluation results of FIGS. 1 and 2 and the PL evaluation of FIG. 4 detect the same structural deterioration as a prominent signal.

  In the single analysis for only the measurement profile under one opening angle condition as used in the conventional method shown in FIGS. Combined with the fact that it could not be detected as a prominent signal, the new analysis method for comparing and examining the difference between a plurality of measurement results measured under a plurality of receiver aperture angle conditions as shown in the present invention is an X-ray diffraction method. It can be concluded that the detection sensitivity of the structural deterioration in is improved compared to the conventional method, and has the same detection sensitivity as the PL evaluation.

  From the above Example, it was experimentally confirmed that the present invention works very effectively in improving the detection sensitivity for structural deterioration in X-ray diffraction evaluation.

It is a figure which shows the result of applying this invention with respect to the GaInNAs / GaAs DQW sample 4 sample, (a) shows the sample A and (b) shows the sample B. It is a figure which shows the result of applying this invention with respect to the GaInNAs / GaAs DQW sample 4 sample, (a) shows the sample C, (b) shows the sample D. It is a figure which shows the sample dependence of the diffraction intensity difference between the satellite patterns from the DQW structure between the two types of X-ray diffraction profiles in FIGS. 1 and 2, and the PL dependence of FIG. It is a figure which shows the PL measurement result of a GaInNAs / GaAs DQW sample 4 sample. It is a figure which shows the 004 X-ray-diffraction rocking curve measurement result and simulation analysis result of a GaInNAs / GaAs DQW sample 4 sample, (a) shows the sample A, (b) shows the sample B. It is a figure which shows the 004 X-ray-diffraction rocking curve measurement result and simulation analysis result of a GaInNAs / GaAs DQW sample 4 sample, (a) shows the sample C, (b) shows the sample D. It is a figure which shows the 224 X-ray-diffraction reciprocal lattice mapping measurement result of a GaInNAs / GaAs DQW sample 4 samples, (a) shows the sample A and (b) shows the sample B. It is a figure which shows the 224 X-ray-diffraction reciprocal lattice mapping measurement result of the GaInNAs / GaAs DQW sample 4 sample, (a) shows the sample C, (b) shows the sample D.

Claims (3)

In a crystal structure analysis method for analyzing a multilayer structure crystal produced by epitaxial growth on an arbitrary substrate by X-ray diffraction,
A first step of measuring a plurality of diffraction profiles having the same reflection index (hk1), with the other measurement conditions being the same, with only the aperture angle condition of the diffracted X-ray light receiving system being changed;
By aligning the diffraction intensity and diffraction angle of the peak top of each substrate diffraction peak in the plurality of (hk1) profiles from the plurality of (hk1) profiles obtained in the first step, the plurality of (hk1) A second step of creating a comparative profile in which the profiles are superimposed and displayed,
Using comparative profile obtained by the second step, by measuring the difference between the diffraction intensity among the plurality of (hk1) profile of the diffraction peak from the epitaxial multilayer structure crystal, the greater the difference structure A third step of determining the degree of structural deterioration of the multilayer crystal based on the degree of deterioration;
A structure analysis method for a crystal characterized by comprising: The substrate has a surface index (hkl);
2. The crystal structure analysis method according to claim 1, wherein the reflection index is the same (hkl) as the surface index, or a reflection index (a is an integer) of (ah, ak, al). The first step includes at least a “wide aperture angle condition” and a “narrow aperture angle condition” as the aperture angle conditions of the plurality of light receiving systems used when measuring the plurality of diffraction profiles,
The “wide aperture angle condition” is a measurement condition in which a light receiving slit having an opening angle of 0.25 degrees or more is used immediately before the X-ray receiver or an arrangement in which no light receiving slit is used.
2. The “narrow aperture angle condition” is a measurement condition in which a triaxial arrangement condition is obtained by inserting an analyzer crystal immediately before the X-ray receiver, and an aperture angle is 0.01 degrees or less. Or the crystal structure analysis method of 2 description.

Images