Non-destructive method and device for detecting bulk defects and determining their depth distribution in semiconductor wafers

Abstract

The device and method according to the invention serve for the non- destructive detection of bulk defects in semiconductor wafers (W) and for determining their depth distribution. In the solution according to the invention, illumination is performed from a polished surface of the semiconductor wafer (W) without breaking the wafer (W), by focussing an illumination beam (11) into the vicinity of a focal line (13) and detecting the light (14) scattered by the illuminated bulk defects. Satisfactory accuracy in determining the depth distribution of the bulk defects is achieved by forming an intermediate image from the scattered light (14) collected by an objective (21 ) using a bi-telecentric optical system (43), and masking part of the intermediate image with an aperture (42), preferably a slit, so that at any given time only the scattered light (14) arriving from a selected small spatial region reaches a detector (44). By coordinated movement — at least in the depth direction — of the systems performing illumination and imaging relative to the semiconductor wafer (W), the depth distribution of the bulk defects is also determined.

Description

[0001] NON-DESTRUCTIVE METHOD AND DEVICE FOR DETECTING BULK DEFECTS AND DETERMINING THEIR DEPTH DISTRIBUTION IN SEMICONDUCTOR WAFERS

[0002] The present invention relates to inspection methods for detecting bulk defects in semiconductor wafers used in the semiconductor industry. More specifically, the invention pertains to a non-destructive method based on light scattering and an apparatus associated with the method, which is suitable for detecting bulk defects in semiconductor wafers, in particular silicon wafers, and for determining their depth distribution.

[0003] During the manufacturing of various semiconductor devices, the defect- free quality of the semiconductor wafers used in production and the detection of defects during the manufacturing process are of critical importance for the operability of the devices. Both the so-called surface defects located on the surface of the semiconductor wafers and the so-called bulk micro defects (BMD) located inside the semiconductor wafers play a significant role. A bulk micro defect may be any kind of imperfection, including voids, inclusions, etc.; therefore, these micro defects are referred to by many different names, depending on the problems they cause, the way they are formed, or their physical characteristics. Micro defects may form during crystal growth, thermal processing of semiconductor wafers, or during epitaxial layer growth. Most semiconductor materials are transparent to infrared light of suitable wavelength; however, bulk micro defects alter the local optical properties of the semiconductor, which enables their detection through the scattering of infrared light by means of a suitable camera.

[0004] Light scattering-based destructive material testing methods are nowadays widely used in industry and in the field of materials science research. The industrial standard method for detecting bulk defects in silicon wafers is so- called Light Scattering Tomography (LST), in which a broken silicon wafer is illuminated with infrared laser light in a direction perpendicular to the plane of the wafer, and the light scattered by the bulk defects is detected through the fractured surface by means of a microscope unit and a CCD (Charge-Coupled Device) camera. A disadvantage of the method is that the sample must be broken for the procedure, meaning that the method is not non-destructive. There is an increasing demand in the industry for measurement equipment that can be integrated into production lines and allows each wafer to be measured individually, which is only possible by using non-destructive methods.

[0005] Japanese patent application JP2001338959A discloses a dark-field method based on light scattering for the non-destructive inspection of semiconductor wafers. In the solution presented in the patent application, a semiconductor wafer is illuminated throughout its full depth from a direction forming an acute angle with the surface of the wafer, and the light scattered by surface irregularities and bulk defects is detected by a CCD camera in a direction perpendicular to the surface. Due to the full-depth illumination of the semiconductor wafer, the light scattered by defects present at all depths of the wafer appears simultaneously in the image; however, only a small portion of the defects are in focus, while the majority merely appear as noise. This noise is used to estimate the areal density of the bulk defects. Therefore, the method disclosed in the Japanese patent document is suitable for the non-destructive detection of bulk defects, but it only provides information regarding their spatial (areal) distribution, and not about their depth distribution.

[0006] None of the above-described technical solutions provide a method for the non-destructive detection of bulk micro defects present in semiconductor wafers and for determining their depth distribution.

[0007] In light of the above, the present invention aims to mitigate and / or eliminate the disadvantages of the known solutions, in particular by providing an optical inspection method and inspection apparatus that are suitable for determining the depth distribution of bulk defects in semiconductor wafers.

[0008] Specifically, the objective of the present invention is to develop a light scattering-based method and corresponding inspection apparatus that allow high-resolution and high-sensitivity detection of bulk defects in semiconductor wafers and the determination of their depth distribution in a non-destructive manner.

[0009] The objective of developing an inspection method according to the above is achieved by the method defined in claim 1 . Preferred embodiments of the method according to the invention are defined in claims 2 to 14.

[0010] The objective of developing an inspection apparatus according to the above is achieved by the device defined in claim 15, preferred embodiments of which are defined in claims 16 to 23.

[0011] The invention is based in part on the recognition that not only the detection of bulk defects in semiconductor wafers, but also the determination of their depth distribution can be accomplished in a non-destructive manner, without breaking the semiconductor wafer, provided that illumination and detection are carried out through the upper (polished) surface of the sem iconductor wafer along specifically selected directions, and that spatial filtering is applied to ensure that, at a given moment, only the image of defects located within a narrow spatial region appears in focus on the detector surface.

[0012] The invention is further based in part on the recognition that by selecting an appropriate scanning direction, it is possible to obtain depth-related information on the bulk defects from a single captured image.

[0013] The invention is further based in part on the recognition that, for certain scanning directions, it is possible to simultaneously obtain depth-related information on the bulk defects and to increase the sensitivity of the measurement by using the detector in TDI (Time Delay Integration) mode.

[0014] The solution according to the invention will be described in detail below with reference to the accompanying drawings, in which:

[0015] Fig. 1 is a simplified schematic view of an exemplary embodiment of the method according to the invention;

[0016] Fig. 2 illustrates a preferred embodiment of the method according to the invention in a direction perpendicular to the plane of the semiconductor wafer and the scanning direction;

[0017] Fig. 3 schematically illustrates a preferred embodiment of the imaging system of the device according to the invention; and

[0018] Fig. 4 illustrates the imaged spatial region of the semiconductor wafer W in an exemplary embodiment of the method according to the invention.

[0019] Fig. 1 is a simplified schematic illustration of the method according to the invention. In the method according to the invention, a semiconductor wafer W, for example a silicon wafer, is illuminated through one of its polished surfaces during an illumination step. In the illumination step, illumination light is generated by means of an illumination system 10 and is directed onto the semiconductor wafer as an illumination beam 11 from an illumination direction I. Essentially simultaneously with the illumination step, the scattered light 14 resulting from bulk defects present within the semiconductor wafer W is detected in a detection step by means of an imaging system 40 from a detection direction D. The upper surface of semiconductor wafers W — which by definition is the surface on which semiconductor devices are formed during the manufacturing process — is always polished, meaning that both illumination and detection can always be performed from above. If the lower surface of the semiconductor wafer W is also polished, illumination and detection may also be performed from below. This is particularly advantageous when a patterned structure has already been formed on the upper surface of the semiconductor wafer, which would otherwise complicate measurement from above.

[0020] In order to detect light scattered from defects located at a depth of at least 100 micrometres from the illuminated surface of the semiconductor wafer W, the wavelength of the illumination light is selected such that the material of the semiconductor wafer W is at least partially transparent at the given wavelength. However, during the development of the invention it was recognized that wavelengths at which the semiconductor material is overly transparent are disadvantageous, since in such cases light reflected within the semiconductor wafer W and entering the imaging system 40 degrades measurement accuracy. In a preferred embodiment of the invention, the semiconductor wafer W is made of silicon, and illumination is performed using a wavelength of preferably 1020 nm. In a further preferred embodiment of the method, for inspecting silicon wafers, a non-silicon-based sensor is preferably used, in particular an indium gallium arsenide (InGaAs)-based sensor, and most preferably an InGaAs TDI camera. Using a camera made of a material different from that of the semiconductor wafer W provides improved sensitivity. In another preferred embodiment of the method, a silicon-sensor CCD camera is used as the detector, particularly preferably a silicon-sensor TDI camera, which has the advantage of being easily and relatively inexpensively available.

[0021] The illumination beam 11 is focused within the semiconductor wafer W into a small region around a focal line 13 located at a given depth, preferably within a range of 5-8 micrometres. For ease of understanding, the mutually orthogonal directions X, Y, and Z indicated in the figure are used below to describe the respective directions and angles, though it is evident to a person skilled in the art that the method can be implemented in any other coordinate system as well. Accordingly, any characteristics associated with the directions X, Y, and Z should not be interpreted in a limiting manner. In the figure, the Z direction indicates the depth direction of the semiconductor wafer, while the surface of the semiconductor wafer W is parallel to the plane spanned by the X and Y directions. In the embodiment shown in Fig. 1 , the scanning direction S is parallel to the X direction and thus parallel to the surface of the semiconductor wafer W. The focal line 13 is preferably parallel to the Y direction, i.e., perpendicular to the scanning direction S.

[0022] The illumination step and the imaging step are repeated multiple times, and between each repetition the focal line 13 is displaced within the semiconductor wafer W, thereby scanning a spatial region of the sample with the illumination beam 11 while capturing images of the same region by repeating the imaging step. The displacement may be performed either in a stepwise or continuous manner. By moving the imaging system 40 and the illumination system 10 relative to the semiconductor wafer W in the scanning direction S (or by moving the semiconductor wafer W in the opposite direction to the scanning direction S), an image is captured in a plane parallel to the scanning direction S.

[0023] In the figures, identical reference signs indicate identical elements.

[0024] Fig. 2 illustrates a preferred embodiment of the method according to the invention, viewed from a plane of the semiconductor wafer W in a direction perpendicular to the scanning direction S. The illumination direction I and the detection direction D are spatially arranged relative to one another such that the portion of the illumination beam 11 that is specularly reflected from the surface of the semiconductor wafer W does not directly enter the objective 21 placed in the detection direction D.

[0025] In order to determine the depth distribution of bulk defects, it is desired to detect only the light scattered from defects located in the immediate vicinity of the focal line 13 at a given moment. For this purpose, the objective 21 is preferably positioned such that its object plane 22 coincides with the focal line 13. However, a problem arises due to the relatively high refractive index of the material of the semiconductor wafer W — approximately 3.5 for silicon in the nearinfrared range — whereby the rays of the illumination beam 11 introduced into the semiconductor wafer W and the rays of the scattered light 14 detectable from within the semiconductor wafer W form a small angle with each other inside the wafer, regardless of how large the angle of incidence is for illumination and detection. As a result, defects located both above and below the focal line 13 are also illuminated, and the light scattered from these defects is collected by the objective 21. Some of these defects are located outside the depth of field 23 characteristic of the imaging system relative to the object plane 22 of the objective 21 , and thus form relatively large, defocused spots in the image, which hinder or possibly even prevent the identification of defects at the intended depth.

[0026] This problem is addressed in the method according to the invention by applying spatial filtering, wherein only the rays of light originating from defects located within a selected spatial region 24 of the semiconductor wafer Ware used for imaging, while the propagation of rays from points outside the selected spatial region 24 is suppressed. Preferably, spatial filtering is implemented by generating an intermediate real image after the objective 21 , a portion of which is masked, thereby creating a modified real image, which is then used for imaging. In a particularly preferred embodiment of the method according to the invention, spatial filtering is achieved using a slit placed at the location of the intermediate image, which allows rays originating from the vicinity of the focal line 13 to pass through unobstructed, while partially blocking rays from other regions — meaning that a smaller portion of the light scattered by defects outside the selected spatial region 24 contributes to image formation, making those defects appear dimmer in the image, thereby enabling distinction between defects within and outside the selected spatial region 24. Particularly preferably, rays originating from outside the selected spatial region 24 in the direction parallel to the object plane 22 — defined by the magnification of the imaging system 40 and the width of the slit — are blocked by the slit, so that these illuminated defects do not participate in image formation. In other words, in the direction parallel to the object plane, the selected spatial region 24 is bounded by the projection of the image of the slit into the semiconductor wafer W, which is indicated in the figure by dashed lines. In the direction perpendicular to the object plane 22, the depth of field 23 of the imaging system 40 defines the range within which defect images will appear sharp at the location of the slit. For defects whose images are defocused at the slit, some of the converging or diverging rays are blocked by the slit; accordingly, these defects appear dimmer in the resulting image. That is, in the direction perpendicular to the object plane 22, the selected spatial region 24 is bounded by the depth of field 23, also indicated in the figure by dashed lines.

[0027] In the method according to the invention, scanning may be performed in various ways depending on the type of information to be extracted. In an exemplary embodiment of the method according to the invention, the scanning direction S is chosen to be parallel to the surface of the semiconductor wafer W, and during image acquisition, the semiconductor wafer W and the optics used for illumination and imaging are moved relative to one another in the X direction. Image acquisition is preferably carried out in Time Delay Integration (TDI) mode using a TDI camera, thereby increasing the sensitivity of the image capture. In this case, an image can be captured of an entire plane of the semiconductor wafer W, and information can be obtained about the spatial distribution of defects at a given depth. To determine the depth distribution of the defects, several such scans can be performed consecutively, from which the volumetric distribution of defects within the scanned volume can be determined. However, scanning a large volume in this manner requires a relatively long time; each individual scan yields an image lying in the XY plane as indicated in Fig. 3.

[0028] In a preferred exemplary embodiment of the method according to the invention, the depth distribution of defects within a relatively small region (e.g., 1-2 millimetres) of the semiconductor wafer W can be determined from a single captured image, provided that the scanning direction S is chosen to be perpendicular to the surface of the semiconductor wafer W. During image capture, the semiconductor wafer W is moved in the Z direction relative to the illumination system 10 and the imaging system 40, and the illumination system 10 and the imaging system 40 are moved relative to each other in the X direction such that the focal line 13 continuously coincides with the object plane 22. In addition, the illumination intensity is adjusted according to the instantaneous depth of the focal line 13 and the light absorption of the semiconductor wafer W at the given illumination wavelength, so that the scattering centres (defects) located on the focal line 13 appear with equal intensity in the captured image. The advantage of this method is its speed; however, its drawback is that it cannot be performed in TDI mode, and thus its sensitivity is lower. Each individual scan produces an image lying directly in the YZ plane as indicated in Fig. 3, directly showing the depth distribution of defects.

[0029] In a particularly preferred exemplary embodiment of the method according to the invention, the shortcomings of the above two measurement methods are eliminated, and their advantages are combined in the following manner. The depth distribution of defects within a relatively small region of the semiconductor wafer W can be determined from a single captured image, provided that the scanning direction S is chosen to form an acute angle with the surface of the semiconductor wafer W. The angle enclosed by the scanning direction S and the surface of the semiconductor wafer W is preferably less than 45 degrees, more preferably less than 30 degrees, even more preferably less than 20 degrees, still more preferably between 5 degrees and 12 degrees, and particularly preferably approximately 8.5 degrees. During image capture, the semiconductor wafer W and the illumination system 10 and imaging system 40 are moved relative to one another such that the focal line 13 progresses in the scanning direction S relative to the semiconductor wafer W, and the illumination system 10 and the imaging system 40 are moved relative to each other in the X direction such that the focal line 13 continuously coincides with the object plane 22. Furthermore, the illumination intensity is adjusted according to the instantaneous depth of the focal line 13 and the light absorption of the given semiconductor wafer W at the selected illumination wavelength, so that the scattering centres (defects) located on the focal line 13 appear with equal intensity in the captured image. During testing of this method, it was surprisingly found that a plane scanned at a relatively small angle (e.g., 8.5 degrees) to the surface of the semiconductor wafer Wstill provides relatively accurate information about the depth distribution of defects in a given region of the semiconductor wafer W, even though the ends of the oblique image may be several millimetres apart and thus represent defects located at different depths and lateral XY positions. The reason for this is that, in most semiconductor wafers W, the density of bulk defects varies much less along the XY plane over a range of a few millimetres than it does in the Z direction over a range of only 100 micrometres. This recognition enables the exploitation of the particular advantages of this method, namely that information on the depth distribution of defects can be obtained from a single captured image, and that image acquisition can be performed in TDI mode, thereby ensuring particularly high sensitivity.

[0030] In a preferred embodiment of the method according to the invention, the angle enclosed by the illumination direction I and the surface of the semiconductor wafer W is 20 degrees to 60 degrees, preferably 30 degrees to 50 degrees, and particularly preferably approximately 40 degrees.

[0031] In a particularly preferred embodiment of the method according to the invention, the marginal rays of the focused illumination beam 11 preferably enclose a maximum angle of 10 degrees with the optical axis 12 of the illumination beam 11 .

[0032] In a preferred embodiment of the method according to the invention, the optical axis 15 of the objective 21 used for detecting the scattered light 14 encloses an angle of 78 degrees with the surface of the semiconductor wafer W.

[0033] Fig. 3 illustrates the spatial regions corresponding to images that can be captured using the method according to the invention. If the scanning direction S is chosen to be parallel to the surface of the semiconductor wafer W, and the illumination beam is focused into a line perpendicular to the scanning direction S, then with a single continuous scan, defects located along a rectangle lying in the XY plane — having a width equal to the length of the focal line (e.g., 0.5 mm or 1 .2 mm) and arbitrary length — are detected. If scanning is performed along the thickness of the semiconductor wafer W in the Z direction, then defects located along a rectangle lying in the YZ plane are detected, where the width of the rectangle in the Y direction is equal to the length of the focal line (e.g., 0.5 mm or 1 .2 mm), and the height in the Z direction is at most equal to the thickness of the sem iconductor wafer W. If oblique scanning is applied, then defects located along an oblique rectangle 45 are detected in the image, where the width of the rectangle in the Y direction is equal to the length of the focal line (e.g., 0.5 mm or 1 .2 mm), and its length in the X direction is determined by the scanning depth in the Z direction and the tilt angle — for example, scanning to a depth of 300 pm at a tilt angle of 8.5 degrees results in a rectangle of 2 mm in length. In a preferred embodiment of the method according to the invention, defects are detected in the vicinity of these rectangles within a thickness of 20-25 pm.

[0034] In a preferred exemplary embodiment of the method according to the invention, image acquisition is performed in TDI mode, meaning that the semiconductor wafer W and the imaging system 40 are moved relative to each other while the pixel lines of a camera are shifted, such that the same point of the semiconductor wafer W — i.e. , for example the image of a defect — moves across the detector surface of the camera in synchronicity with the pixel shifting, so that the signal from the defect is always summed with itself. By applying the TDI method, the image of the detected bulk defects becomes sharp and receives the full integrated intensity of the illumination, thereby improving both the sensitivity and resolution of the image acquisition. In the resulting image, the longitudinal (i.e., X direction) position of the bulk defects is determined, and using a simple coordinate transformation based on the known tilt angle, the position of the defects in the Z direction is calculated, thereby determining the depth distribution of the bulk defects.

[0035] Fig. 4 illustrates the imaging system 40 used in a preferred embodiment of the apparatus implementing the method according to the invention. In the direction of propagation of the light scattered by the bulk defects, the imaging system 40 sequentially includes: an objective 21 , which collects the light rays originating from its focal plane and transmits them in a collimated manner; a first lens system 41 for forming an intermediate image, which creates a real image (hereinafter: intermediate image) in its rear focal plane from the collimated rays coming from the objective; an aperture 42, preferably a slit, placed in the rear focal plane of the first lens system 41 for masking the intermediate image; and a second lens system 41’, which focuses the rays originating from the masked intermediate image, together with one or more subsequent optical elements, onto the detector 44. The aperture 42 completely blocks rays originating from regions outside the desired imaging region in the direction parallel to the object plane 22, and partially blocks rays originating from points outside the desired imaging region in the direction perpendicular to the object plane 22, thus dimming the defocused spots of defects located in front of or behind the object plane 22 (i.e. , outside the depth of field) in the image. Overall, the bi-telecentric optical system 43 performs the spatial filtering in the method according to the invention. In the preferred exemplary embodiment shown in Fig. 4, the bi-telecentric optical system 43 is formed by the combination of the first lens system 41 , the aperture 42, and the second lens system 4T. The second lens system 4T preferably consists of the same elements as the first lens system 41 , arranged in mirror symmetry with respect to the plane of the aperture 42, i.e., as a mirror image of the first lens system 41 .

[0036] The sole purpose of generating the intermediate image is to perform spatial filtering; therefore, the imaging system 40 would be fundamentally functional even without the bi-telecentric optical system 43, as is generally the case in prior art solutions. The imaging system 40 may additionally include further elements, such as a field lens, mirrors to deflect the light path once or multiple times in order to achieve a more compact optical layout, or one or more spectral filters to select light of a specific wavelength (especially in the case of a broadband light source). The application of these components is evident to a person skilled in the art and will therefore not be described in further detail. In Fig. 4, the direction labelled Y’ corresponds to the Y direction used in the other figures, but does not necessarily coincide with it, as the optical system illustrated in Fig. 4 may include mirrors at any point in order to achieve a more compact layout of the optical system. Thus, the optical path from the objective 21 to the detector 44 may be deflected one or more times.

[0037] In an exemplary embodiment of the device according to the invention, the sample is examined using line scanning; in this case, the aperture 42 is preferably a slit, having a size of, for example, 50 micrometres x 8 millimetres.

[0038] In a preferred exemplary embodiment of the device according to the invention, the objective 21 has a magnification of 50x, enabling diffraction-limited imaging with a resolution of 0.5 micrometres / pixel over a 500-micrometre object field. In another preferred embodiment, the objective 21 has a magnification of 20x, with a resolution of 1 .2 micrometres / pixel over a 1 .2-millimetre object field.

[0039] The device according to the invention comprises a sample holder (not shown in the figure) for holding the semiconductor wafer W, which is preferably movable by means of a motion system (also not shown in the figure). The sample holder is preferably suitable for moving the semiconductor wafer W in at least the Z direction, and particularly preferably suitable for moving the semiconductor wafer W in the X, Y, and Z directions. The X-Y movement of the sample holder may be performed purely by translation or by a combination of translation and rotation. Such sample holders are known in the industry, and therefore will not be described in further detail. The device according to the invention further comprises an illumination system 10 suitable for illuminating the semiconductor wafer W from the illumination direction I, and an imaging system 40 suitable for detecting the light scattered on bulk micro defects present within the semiconductor wafer W from the detection direction D. The illumination system 10 and the imaging system 40 are preferably movable by means of a motion system.

[0040] For the operation of the invention, it is necessary that the focal line 13 and the object plane 22 within the semiconductor wafer W be positionable at different depths. Accordingly, the motion system is configured to move the sample holder, or the illumination system 10 and the imaging system 40, or all of these, in the Z direction.

[0041] It is also necessary for the operation of the invention that the focal line 13 and the selected spatial region 24 within the semiconductor wafer W be positionable at different positions in the X and Y directions. Accordingly, the motion system is configured to move the sample holder, or the illumination system 10 and the imaging system 40, or all of these, in the X and Y directions. The motion system is connected to a control unit.

[0042] In a preferred embodiment of the invention, in which the device is configured to be suitable for the inspection of a relatively large depth range (several hundred micrometres), the illumination system 10 and the imaging system 40 are configured to be movable relative to each other in the X direction. This ensures that the focal line 13 and the object plane 22 coincide at different depths. The motion mechanisms of the sample holder, the illumination system 10 and the imaging system 40 — such as stepper motors and piezoelectric actuators — are connected to a control unit, which is preferably configured to initiate the relative movement of the illumination system 10 and the imaging system 40 in the X direction simultaneously with the depthwise movement, such that the focal line 13 and the object plane 22 remain aligned continuously at varying depths.

[0043] In a particularly preferred embodiment of the device according to the invention, the motion system comprises a sample holder actuator configured to perform translational movement of the sample holder in the Z direction, as well as to perform translational movement of the sample holder in the X and Y directions or rotational and radial translational movement of the semiconductor wafer W. Furthermore, the motion system comprises an actuator configured to displace the illumination system 10 and / or the imaging system 40 in the X direction.

[0044] The device according to the invention further comprises a control unit operatively connected to the illumination system 10, the imaging system 40 and the motion system, configured for controlling these components.

[0045] In a preferred embodiment of the device according to the invention, the control unit is configured to control the motion system such that during the relative movement of the illumination system 10, the imaging system 40 and the semiconductor wafer W, the focal line 13 and the object plane 22 move in a scanning direction S relative to the sem iconductor wafer W, wherein the scanning direction S forms an angle of 5-45 degrees, preferably 5-20 degrees, and particularly preferably 5-10 degrees with the surface of the semiconductor wafer W.

[0046] In a preferred embodiment of the device according to the invention, the control unit is configured such that, during the displacement of the position of the illumination system 10 and the imaging system 40 relative to the semiconductor wafer W in a direction perpendicular to the surface of the semiconductor wafer W, it moves the illumination system 10 and / or the imaging system 40 relative to each other in a direction parallel to the surface of the semiconductor wafer W such that the focal line 13 remains within the depth of field 23 of the object plane 22 of the imaging system 40.

[0047] In a preferred embodiment of the device according to the invention, the control unit is configured such that, during the displacement of the position of the illumination system 10 and the imaging system 40 relative to the semiconductor wafer W in a direction perpendicular to the surface of the semiconductor wafer W, it controls the illumination system 10 in such a manner that the intensity of the illumination beam 11 emitted by the illumination system 10 is varied as a function of the instantaneous distance of the focal line 13 from the surface of the semiconductor wafer W.

[0048] In light of the above teaching, it will be readily apparent to a person skilled in the art that the implementation of the motion mechanisms and the control unit may be realised in numerous different ways. The selection of motion mechanisms in accordance with the specific requirements of a given device, the determination of the relative movement of the illumination system 10 and the imaging system 40 in the X direction based on the refractive index of the semiconductor wafer W, and the adjustment of illumination intensity as a function of depth according to the absorption characteristics of the semiconductor material at the given wavelength, all fall within the routine knowledge of the skilled person and are therefore not described in further detail.

[0049] The device and method according to the invention serve for the nondestructive identification of bulk defects in semiconductor wafers and for determining the depth distribution of such bulk defects.

Claims

CLAIMS1 . A non-destructive method for detecting bulk defects and determining their depth distribution in a semiconductor wafer (W), the method comprising the steps of: performing an illumination step, in which a polished surface of the semiconductor wafer (W) is illuminated in an illumination direction (I) by means of an illumination beam (11 ) generated by an illumination system (10), the illumination beam (11 ) being focussed inside the semiconductor wafer (W) onto a focal line (13) located at a given depth; performing an imaging step, in which a portion of scattered light (14) reflected from bulk defects present within the semiconductor wafer (W) is detected by means of an imaging system (40) comprising an objective (21 ) and a detector (44), wherein the imaging system (40) is arranged such that the focal line (13) lies within the depth of field of the object plane (22) of the imaging system (40); performing a movement step, in which the focal line (13) is moved relative to the semiconductor wafer (W) in a scanning direction (S) by moving the semiconductor wafer (W) and the illumination system (10) relative to each other, and the object plane (22) is moved relative to the semiconductor wafer (W) in the scanning direction (S) by moving the semiconductor wafer (W) and the imaging system (40) relative to each other; wherein the illumination step, the imaging step and the movement step are repeated to perform scanning and image acquisition; characterised in that a real intermediate image is generated from the scattered light (14) collected by the objective (21 ), said real image being masked by an aperture (42), and the masked real image is imaged onto the detector (44), wherein in order to determine the depth distribution of the bulk defects, the illumination step and the imaging step are repeated at multiple different depthsduring one or more scans and image acquisitions.

2. The method according to claim 1 , characterised in that the scanning direction (S) is parallel to the surface of the semiconductor wafer (W), and the illumination and imaging are repeated during scanning and image acquisition at different depths to determine the depth distribution of the bulk defects.

3. The method according to claim 1 , characterised in that the scanning direction (S) encloses an acute angle with the surface of the semiconductor wafer (W), and the illumination and imaging are repeated during a single scan and image acquisition at multiple different depths to determine the depth distribution of the bulk defects.

4. The method according to claim 3, characterised in that the scanning direction (S) encloses an angle of 5-45 degrees with the surface of the semiconductor wafer (W).

5. The method according to claim 4, characterised in that the scanning direction (S) encloses an angle of 5-10 degrees with the surface of the semiconductor wafer (W).

6. The method according to any one of claims 1 to 5, characterised in that the image acquisition is performed in TDI mode.

7. The method according to any one of claims 1 to 6, characterised in that the semiconductor wafer (W) examined during the method is a silicon wafer.

8. The method according to claim 7, characterised in that a CCD camera with a silicon sensor is used as the detector (44).

9. The method according to any one of claims 1 to 8, characterised in that monochromatic light with a wavelength of 1020 nm is used as the illumination beam (11 ).

10. The method according to any one of claims 1 to 9, characterised in that the thickness of the illumination beam (11 ) at the position of the focal line(13) is 5-8 micrometres.

11. The method according to any one of claims 1 to 10, characterised in that the illumination direction (I) encloses an angle of 35-45 degrees with the surface of the semiconductor wafer (W).

12. The method according to any one of claims 1 to 11 , characterised in that the optical axis (15) of the objective (21 ) collecting the scattered light (14) encloses an angle of approximately 78 degrees with the surface of the semiconductor wafer (W).

13. The method according to any one of claims 3 to 12, characterised in that the focal line (13) is scanned within the semiconductor wafer (W) in a depth range of 0-300 micrometres measured from the surface.

14. The method according to any one of claims 3 to 13, characterised in that the intensity of the illumination beam (11 ) is varied as a function of the distance of the focal line (13) from the surface of the semiconductor wafer (W).

15. A device for the non-destructive detection of bulk defects and for determining their depth distribution in a semiconductor wafer (W), comprising: an illumination system (10) configured to direct a converging illumination beam (11 ) in an illumination direction (I) onto a polished surface of the semiconductor wafer (W), such that the illumination beam (11 ) at least partially penetrates into the semiconductor wafer (W) and is focussed within the semiconductor wafer (W) onto a focal line (13) located at a given depth; an imaging system (40) configured to generate a digital image from the light (14) scattered by bulk micro defects located in the vicinity of the focal line (13), wherein the imaging system (40) is arranged such that the focal line (13) lies within the depth of field of the object plane (22) of the imaging system (40); wherein the imaging system (40) comprises an objective (21 ) and a detector (44), and the illumination system (10) and the imaging system (40) are connectedto a control unit; characterised in that the device further comprises a motion system connected to the control unit, said motion system being configured: to change the position of the illumination system (10) and the imaging system (40) relative to the semiconductor wafer (W) in the direction perpendicular to and parallel with the surface of the semiconductor wafer (W), by moving the semiconductor wafer (W) and / or by moving the illumination system (10) and the imaging system (40), and to move the illumination system (10) and the imaging system (40) relative to each other in a direction parallel to the surface of the semiconductor wafer (W), and wherein the imaging system (40) comprises a bi-telecentric optical system (43) arranged between the objective (21 ) and the detector (44).

16. The device according to claim 15, characterised in that the bi- telecentric optical system (43) comprises a first lens system (41 ), an aperture (42), and a second lens system (41 ’ ) , wherein a focal plane of the first lens system (41 ), a focal plane of the second lens system (41 ’), and the aperture (42) lie in a common plane.

17. The device according to claim 15, characterised in that the motion system comprises a sample holder actuator for moving the semiconductor wafer (W) both in a direction perpendicular and parallel to the surface of the semiconductor wafer (W), and an actuator configured to displace the illumination system (10) and / or the imaging system (40) in the X direction.

18. The device according to claim 15, characterised in that the control unit is configured to control the motion system such that, during relative movement of the illumination system (10), the imaging system (40) and the semiconductor wafer (W), the focal line (13) and the object plane (22) move relative to the semiconductor wafer (W) in a scanning direction (S), wherein the scanning direction (S) encloses an angle of 5-45 degrees with the surface of thesem iconductor wafer (W).

19. The device according to claim 18, characterised in that the scanning direction (S) encloses an angle of 5-10 degrees with the surface of the semiconductor wafer (W).

20. The device according to claim 15, characterised in that the control unit is configured such that, during the change of the position of the illumination system (10) and the imaging system (40) relative to the semiconductor wafer (W) in a direction perpendicular to the surface of the semiconductor wafer (W), the illumination system (10) and / or the imaging system (40) are moved relative to each other in a direction parallel to the surface of the semiconductor wafer (W) in such a way that the focal line (13) remains within the depth of field of the object plane (22) of the imaging system (40).

21. The device according to claim 15, characterised in that the control unit is configured such that, during the change of the position of the illumination system (10) and the imaging system (40) relative to the semiconductor wafer (W) in a direction perpendicular to the surface of the semiconductor wafer (W), it controls the illumination system (10) so that the intensity of the illumination beam (11 ) emitted by the illumination system (10) is varied as a function of the instantaneous distance of the focal line (13) from the surface of the semiconductor wafer (W).

22. The device according to claim 15, characterised in that the detector (44) is a TDI camera.

23. The device according to claim 15, characterised in that the illumination system (10) is configured to emit monochromatic light with a wavelength of 1020 nm.

Images