JP4841458B2 - Crystal sample shape evaluation method, shape evaluation apparatus, and program - Google Patents

Description

  The present invention relates to a crystal sample shape evaluation method and a shape evaluation device, and more particularly to an evaluation method and an evaluation device that can also evaluate the shape of a crystal sample that is enclosed in a package and cannot be observed from the outside.

  Conventionally, a surface roughness meter has been widely used as a method for measuring the shape of a crystal with high accuracy. Japanese Patent No. 2964317 discloses a stylus for detecting irregularities on a sample surface, a displacement detection means including a semiconductor laser, a lens, and a light detection element for optically enlarging and detecting displacement of the stylus, and a sample and a stylus. We propose a surface roughness meter that has a fine movement mechanism that controls the distance to a certain distance, a motor-driven Z stage that controls the relative position of the stylus and the sample, a motor-driven XY stage, and a computer that controls the whole. ing.

In the manufacturing process of a semiconductor device such as an LSI, the surface shape of a wafer can be measured with a surface roughness meter. After a semiconductor chip is sealed in a package such as an epoxy mold, a stylus is placed on the crystal surface. The contact cannot be made, and the surface shape of the crystal cannot be measured. An element defect may occur after the LSI chip is sealed in the LSI package. One of the causes is a phenomenon that the semiconductor chip is deformed by the generation of stress due to packaging. However, if the package is removed, the deformation is likely to return to its original state. There is an increasing need to measure the shape of a packaged semiconductor chip as it is.

  An object of the present invention is to provide a shape evaluation technique that can also evaluate the shape of a crystal sample enclosed in a package.

  Another object of the present invention is to provide a shape evaluation technique capable of nondestructively evaluating the shape of a crystal sample enclosed in a package.

According to one aspect of the present invention,
(A) A crystal sample is arranged on an xy plane in a space having a virtual xyz orthogonal coordinate system, X-rays are irradiated on the crystal sample in parallel to the xz plane, and parallel to the xz plane from a predetermined crystal plane of the crystal sample. An X-ray source and an X-ray detector are arranged so as to detect X-rays diffracted in the same direction, and an ω scan around the y-axis is performed to determine an angle ω that maximizes the diffracted X-ray intensity in the crystal sample plane. Detecting as a function of the position of
(B) integrating the angle ω at which the diffracted X-ray intensity is maximum in the x-axis direction of the crystal sample, and evaluating the shape of the crystal sample in the x-axis direction;
A method for evaluating the shape of a crystal sample containing

According to another aspect of the invention,
A program for executing the above-described crystal sample shape evaluation method via a computer is provided.

According to yet another aspect of the invention,
A crystal sample shape evaluation apparatus including an X-ray diffractometer equipped with the above program is provided.

  Even if the crystal sample is enclosed in a package, the shape of the crystal sample can be evaluated nondestructively if the package is a material that transmits X-rays.

  Embodiments of the present invention will be described below with reference to the drawings.

  1A and 1B are cross-sectional views schematically showing the configuration of an X-ray diffractometer, FIGS. 1C and 1D are diagrams showing ω scan and χ scan, and FIG. 1E is a diagram showing X-ray diffraction by a crystal plane of a crystal sample. is there.

As shown in FIG. 1A, the sample stage 10 of the X-ray diffractometer has a structure in which in-plane translation stages 12 and 13 and a vertical translation stage 14 are mounted on a rotating shaft 11. As shown in the figure, virtual orthogonal coordinates x ₀ , y ₀ , z ₀ are taken in the space, and orthogonal coordinates x ₁ , y ₁ , z ₁ are taken in the sample stage. Rotary shaft 11 performs a rotation φ to z ₁ axis, translation stage 12 constitutes a stage for translating the x ₁ direction, the translation stage 13 constitutes a stage for translating the y ₁ direction, vertical translation stage 14 z A stage that translates in _one direction is formed. In the illustrated arrangement, x ₁ , y ₁ . z ₁ is parallel to the x ₀ , y ₀ , and z ₀ axes, respectively, but if φ is rotated by 90 degrees, x ₁ , y ₁ . z ₁ is parallel to the y ₀ , −x ₀ , and z ₀ axes, respectively. In addition, if the sample stage 10 has a φ rotation and a three-dimensional translation function, the vertical relationship between the stages 11 to 14 may be changed.

The stage 10 mounts the crystal sample 18 thereon, and can translate the crystal sample 18 in φ rotation and in the x ₁ , y ₁ , and z ₁ directions. The crystal sample 18 is irradiated with X-rays from an X-ray source, and diffracted X-rays are detected with an X-ray detector.

As shown in FIG. 1B, the X-ray source 15 and the X-ray detector 16 are arranged in the x ₀ z ₀ plane, and the diffraction angle 2θ can be adjusted. When the diffraction angle is 0, the X-ray source 15 and the X-ray detector 16 face each other on the horizontal plane. Furthermore, X-rays source 15, X-rays detector 16 is in a state of fixing the diffraction angle 2 _[Theta], x 0 _{z 0} plane at _{y 0} around the axis ω _rotation, with y 0 _{z 0} plane _{x 0} axis about χ rotation is possible.

Figure 1C, 1D _are x 0 _{z 0} plane of ω _scan, y 0 _{z 0} diagram schematically showing the χ scan in plane.

  The ω scan and χ scan are scans of an angle between the X-ray and the crystal sample, and the stage 10 may be rotated instead of rotating the X-ray source 15 and the X-ray detector 16.

FIG. 1E is a diagram showing diffraction by a crystal plane of a crystal sample. For example, in the case of MOS LSI using a Si substrate, the Si substrate is usually a (001) substrate 18, and a (00n) crystal plane 19 exists in parallel to the substrate surface. Assuming that the wavelength of the X-ray is λ and the interval between the predetermined crystal planes 19 is d, X-ray diffraction occurs when the relationship of λ = 2d * sin θ is satisfied. When incident X-rays travel in the x ₀ z ₀ plane, the normal of the crystal plane must also be in the x ₀ z ₀ plane. This angle can be adjusted by χ scan. Also. Incident X-rays and diffracted X-rays must be symmetric with respect to the normal of the crystal plane. This angle can be adjusted by ω scanning. When the relationship of λ = 2d * sin θ is satisfied, the diffracted X-ray intensity is maximized. When the crystal plane is tilted, the diffracted X-ray intensity decreases.

  In other words, by performing ω scan and χ scan to obtain ω and χ that maximize the diffracted X-ray intensity, the inclination of the crystal plane of the surface irradiated with the X-ray can be known.

  2A and 2B are cross-sectional views and graphs showing ω scan when the surface of the crystal sample is flat, and FIGS. 2C and 2D are cross-sectional views and graphs showing ω scan when the crystal surface is convex upward. .

As shown in FIG. 2A, when the surface of the crystal sample 18 is flat, the crystal plane parallel to the surface is also flat. Performing ω scan while changing the position slightly along the x ₀ direction, the intensity of the diffracted X-rays at any x ₀ position is constant ω angle becomes maximum. This angle is defined as angle 0.

Figure 2B is a graph showing changes in the angle ω with respect to x ₀ direction position of the crystal sample. The angle ω is 0 at all positions. X ₀ direction shape of the crystal sample surface is given by the integral of the angle omega. If omega = 0 (or a predetermined value), x ₀ direction shape of the crystal sample surface is found to be flat.

  As shown in FIG. 2C, when the surface of the crystal sample 18 is bent, when the ω scan is performed at different positions, the angle ω at which the diffracted X-ray intensity becomes maximum varies depending on the position. As shown in the figure, when the crystal surface is convex upward, the angle ω at which the diffraction X-ray intensity becomes maximum gradually decreases from left to right.

Figure 2D is a graph showing changes in the angle ω with respect to x ₀ direction position of the crystal sample. The angle ω decreases from positive to negative with the increase of x _0. When this angle ω is integrated, the rate of increase gradually decreases, and after saturation, the rate of decrease gradually increases, and an upwardly convex crystal shape can be evaluated.

  As described above, if the ω scan is performed while changing the position along the X-ray traveling direction on the crystal plane, and the angle ω at which the diffracted X-ray intensity becomes maximum is detected and sequentially integrated, the X-ray traveling direction is obtained. The shape of the crystal surface along the line can be evaluated.

  In the case of an epoxy-molded LSI device, the Si crystal as a semiconductor substrate is covered with the epoxy mold and cannot be visually observed. In order to appropriately arrange the Si crystal, it is better to perform a preparation work.

FIG. 3 is a flowchart showing the preparation process. In step ST11, the package 1 in which the crystal sample 18 is sealed is placed (fixed) near the center on the sample stage 10. In step ST12, the X-ray diffraction angle 2θ is 0, lower the z ₁ stage, the positional relationship in which the X-ray is incident on the X-ray detector as it is. The z ₁ stage is scanned upward to detect a position where the X-ray intensity becomes approximately half. At this position, about half of the X-rays are shielded by the package (half of the X-rays). The crystal sample 18 in the package 1 is below a predetermined distance determined by design from the package surface. In step ST13, increasing the predetermined distance _{z 1} stage. The crystal surface will be flush with the center of the X-ray (half of the X-ray by the crystal). The crystal height has been adjusted.

  In step ST14, 2θ is set in accordance with the diffraction angle determined by the surface spacing of the crystal plane to be measured. Here, the angles of the X-ray source 15 and the X-ray detector 16 were set so that the (0020) plane of Si was measured and 2θ = 131.93 °. If the crystal plane is horizontal, diffracted X-rays are generated, but if the crystal plane is tilted, no diffracted X-rays are generated. Ω scan at step ST15. Perform χ scan to find the diffracted X-rays from the crystal. If a diffracted X-ray is found in step ST16, ω scan and χ scan are further performed to obtain an angle at which the diffracted X-ray intensity becomes maximum. This means that the crystal surface can be arranged horizontally. If X-ray halving is performed while the surface of the package or crystal is initially tilted, the X-ray halving conditions depending on the crystal surface may be shifted if the tilt of the crystal plane is adjusted. If necessary, the process may return to step ST12 and perform steps ST12 to ST16 again. In this way, the preparation process is completed.

  FIG. 4 is a flowchart showing a crystal sample shape evaluation step. A scintillation counter was used as the X-ray detector 16 shown in FIG. 1B. The energy of X-rays emitted from the X-ray source 15 was 25 keV so that diffraction measurement could be performed through the epoxy mold package. The size of the crystal sample to be measured is about several mm square. In order to scan the X-ray on the crystal surface and measure the shape of the crystal surface depending on the position, the X-ray beam size was set to about 100 μm square which is sufficiently smaller than the size of the crystal sample.

In step ST21, a ω scan is performed (near the center of the crystal sample), and the angle ω at which the X-ray diffraction intensity is maximum is measured. This angle ω is used as a reference. Performing _{x 0} movement of the package in step ST22. Even if the crystal sample is deformed, the moving distance is set such that the inclination of the crystal plane changes only slightly. In step ST23, a ω scan at a new position is performed, and the angle ω at which the X-ray diffraction intensity is maximized is measured. From the change in the angle ω, the change in the tilt of the crystal plane due to the change in position can be seen. If the diffraction intensity of X-rays to some extent, from the step ST24, the process returns to step ST22, _{x 0} movement of the step ST22, and repeats the ω scan step ST23. When the X-ray irradiation position reaches the end of the crystal or deviates from the crystal, the X-ray diffraction intensity greatly decreases or disappears. At this time, in accordance with a NO arrow from step ST24, the process proceeds to step ST25, returns the _{x 0} direction position of the package in the first position. Since it is unclear from which position on the crystal the measurement was started, it is returned to its original position in order to scan in the reverse direction. In step ST26, the ω scan is performed, and the angle ω at which the X-ray diffraction intensity is maximized is measured. Although the reference angle ω is confirmed, this step may be omitted. Performing _{x 0} direction movement backward of the package in step ST27. In step ST28, a ω scan at a new position is performed, and the angle ω at which the X-ray diffraction intensity is maximized is measured. From the change in the angle ω, the change in the tilt of the crystal plane due to the change in the position in the reverse direction can be seen. If the diffraction intensity of X-rays to some extent, from the step ST29, the process returns to step ST27, reverse _{x 0} movement of the step ST27, and repeats the ω scan step ST28. When the X-ray irradiation position reaches the end of the crystal or deviates from the crystal, the X-ray diffraction intensity greatly decreases or disappears. In this case, so that the x ₀ direction of the scan is complete. The measurement may be terminated at step ST30. In step ST40, the obtained angle ω is integrated to evaluate the surface shape of the crystal.

Figure 5A-5B, 5C-5D, 5E-5F, 5G-5H is a graph diffracted X-ray intensity of four sample indicates the _{x 0} direction position dependent angle ω which is a maximum, integrates the angle ω It is a perspective view which shows the crystal surface shape obtained in this way.

5A, 5C, 5E, 5G are samples A, B, C.I. It shows the x ₀ direction position dependence of the angle ω of D. 5B, 5D, 5F, and 5H show Samples A, B, C.I. It shows the shape changes obtained by integrating the x ₀ direction position dependence of the angle ω of D. It was possible to measure and evaluate the shape of the crystal molded and hidden in the epoxy in a non-destructive manner. When the deformation is large, there is a possibility that a large distortion has occurred, which is considered to be a cause of performance degradation of the LSI.

In the above steps, the shape along one direction of the crystal was measured. In many cases, it is desirable to perform orthogonal measurement and check the deformation of the entire crystal plane. In that case, in step ST30, the sample stage is rotated 90 degrees phi, to collimate the _{y 1} direction of the stage in the _{x 0} direction. Y ₁ direction of the scanning and non crystals, so along the x ₀ direction which is the traveling direction of the X-ray. Thereafter, the process returns to step ST21, and the angle ω along the y1 direction of the crystal is measured. If NO in step ST30, the measurement ends. X ₁ direction of the crystal, because y ₁ direction ω distributions were measured and by integrating ω in step ST40 to simulate the shape of the crystal plane can be evaluated crystal shape in the orthogonal direction. Incidentally, x ₁ direction of the crystal, along the y ₁ direction, has been evaluated the shape of the stripe-shaped target regions, respectively, may be scanned a crystal sample entire respectively. A second embodiment suitable for scanning the entire surface will be described below.

6A is irradiated with a long stripe region y ₀ direction, schematic plan view illustrating the same time the X-ray source 25, the X-ray detector 26 for measuring the slope at each y ₀ direction position of the X-ray irradiated crystal surface FIG 6B, 6C has a _{x 1} direction of the crystal sample 18 scans were parallel to the _{x 0} direction is an X-ray traveling direction, parallel to _{y 1} direction of the crystal sample 18 in the _{x 0} direction is an X-ray traveling direction 2 is a diagram schematically illustrating the scan performed.

As shown in FIG. 6A, X-ray detector 26 are arranged along a number of measurement channels in the y ₀ direction. X-ray source 25, as can be supplied diffracted X-ray in each measurement channel of the X-ray detector 26, and supplies the long X-ray beam in the y ₀ direction.

As shown in FIG. 6B, irradiated with long X-ray beam 28 in the y ₀ direction crystal sample 18 is moved in the x ₀ direction, perform ω scanning in the x ₀ position. For comparison, the spot-like X-ray of the first embodiment is indicated by a broken line. In the second embodiment, along the y ₀ direction, by irradiating the stripes X-ray beam 28 that traverses the crystal sample 18, by scanning in the x ₀ direction, to facilitate the measurement of the crystal of the entire surface. Each measurement channel has the maximum diffracted X-ray intensity when the diffraction conditions are met. long stripe region y ₁ direction can be measured in a single ω scan.

As shown in FIG. 6C, when finished the x ₀ direction of the scanning, the y ₁ direction of the crystal is parallel to the x ₀ direction, it performs y ₁ direction of the scan of the crystals. The ω scan itself is the same as in FIG. 6B. The measurement procedure is the same as the measurement procedure of FIG. 4 except that the stripe region is irradiated with X-rays and each point of the stripe region is measured in parallel using a large number of measurement channels. Note that the X-ray beam 28 does not necessarily have a length that crosses the crystal sample 18. By performing simultaneous measurement on multiple channels, the measurement procedure is simplified. Further, a third embodiment in which the measurement procedure is simplified will be described below.

  FIG. 7A is a schematic perspective view showing the X-ray source 35 corresponding to the two-dimensional X-ray detector 36 and the crystal sample 18, and FIGS. 7B and 7C are diagrams showing the measured surface shapes.

  As shown in FIG. 7A, the X-ray source 35 irradiates the entire surface of the crystal sample 18 with a planar X-ray beam 38. The two-dimensional X-ray detector 36 is composed of, for example, an X-ray CCD camera and has a size capable of detecting X-rays diffracted from the entire surface of the crystal sample 18. X-rays diffracted from each position of the crystal are measured at corresponding different positions of the two-dimensional X-ray detector. When one ω scan is performed, the entire crystal surface can be measured. If the crystal is rotated 90 degrees in the plane and the orthogonal ω scan is performed, the measurement in the orthogonal direction on the entire crystal surface can be completed.

  7B and 7C show the shape of the crystal plane obtained by integrating the ω distribution obtained in this way in the orthogonal direction.

  Note that the measurement of the entire crystal surface is not necessarily performed once. By using a two-dimensional detector, the surface can be measured at a time.

  As mentioned above, although this invention was demonstrated along the Example, this invention is not limited to these. It will be apparent to those skilled in the art that various modifications, substitutions, improvements, combinations, and the like can be made.

1A and 1B are cross-sectional views schematically showing the configuration of an X-ray diffractometer, FIGS. 1C and 1D are diagrams showing ω scan and χ scan, and FIG. 1E is a diagram showing X-ray diffraction by a crystal plane of a crystal sample. is there. 2A and 2B are cross-sectional views and graphs showing ω scan when the surface of the crystal sample is flat, and FIGS. 2C and 2D are cross-sectional views and graphs showing ω scan when the crystal surface is convex upward. . FIG. 3 is a flowchart showing the preparation process. FIG. 4 is a flowchart showing a crystal sample shape evaluation step. Figure 5A-5B, 5C-5D, 5E-5F, 5G-5H is a graph diffracted X-ray intensity of four sample indicates the _{x 0} direction position dependent angle ω which is a maximum, integrates the angle ω It is a perspective view which shows the crystal surface shape obtained in this way. Figure 6A, y ₀ irradiating a stripe region along the direction, schematic plan view showing an X-ray source 25, the X-ray detector 26 for simultaneously measuring the inclination of the crystal planes in the respective y ₀ direction position, FIG. 6B, 6C the scan was parallel to x ₁ direction of the crystal in the x ₀ direction is an X-ray traveling direction, the scanning which is parallel to y ₁ direction of the crystal in the x ₀ direction is an X-ray traveling direction in the diagram schematically illustrating is there. FIG. 7A is a schematic perspective view showing the X-ray source 35 corresponding to the two-dimensional X-ray detector 36 and the crystal sample 18, and FIGS. 7B and 7C are diagrams showing the measured surface shapes.

Explanation of symbols

1 package,
10 Sample stage,
11 Rotation axis (φ axis),
12, 13, 14 Translation stage,
15, 25, 35 X-ray source 16, 26, 36 X-ray detector
18 Crystal sample

Claims (7)

(A) A crystal sample is arranged on an xy plane in a space having a virtual xyz orthogonal coordinate system, X-rays are irradiated on the crystal sample in parallel to the xz plane, and parallel to the xz plane from a predetermined crystal plane of the crystal sample. An X-ray source and an X-ray detector are arranged so as to detect X-rays diffracted in the same direction, and an ω scan around the y-axis is performed to determine an angle ω that maximizes the diffracted X-ray intensity in the crystal sample plane. Detecting as a function of the position of
(B) integrating the angle ω at which the diffracted X-ray intensity is maximum in the x-axis direction of the crystal sample, and evaluating the shape of the crystal sample in the x-axis direction;
Method for evaluating shape of crystal sample including The step (a)
(A-1) performing a ω scan at an x position of the crystal sample to detect an angle ω at which the diffracted X-ray intensity is maximum;
(A-2) translating the crystal sample in the x direction;
(A-3) repeating steps (a-1) and (a-2);
The method for evaluating the shape of a crystal sample according to claim 1, comprising:   In the step (a), the X-ray source irradiates a striped X-ray beam long in the y direction on the crystal sample, and the X-ray detector has a large number of detection channels arranged in the y direction. 2. The method for evaluating the shape of a crystal sample according to 2.   The shape of the crystal sample according to claim 1, wherein, in the step (a), the X-ray source irradiates a planar X-ray beam onto the crystal sample, and the X-ray detector is a two-dimensional X-ray detector. Evaluation methods.   In the step (a), the crystal sample is further rotated 90 degrees in the xy plane, ω scan is performed, and the angle ω at which the diffracted X-ray intensity is maximum is detected as a function of the position in the crystal sample surface. The method for evaluating the shape of a crystal sample according to claim 1, comprising a step.   The program which performs the shape evaluation method of the crystal sample of any one of Claims 1-5 via a computer.   A crystal sample shape evaluation apparatus including an X-ray diffractometer equipped with the program according to claim 6.

Images