WO2009028956A1

WO2009028956A1 - Method and device for inspection of object surfaces

Info

Publication number: WO2009028956A1
Application number: PCT/NO2008/000307
Authority: WO
Inventors: Arne Sommerfelt; Bernt Ribbum; Torleif Ringsaker; Thor Vollset
Original assignee: Tordivel Solar As
Priority date: 2007-08-31
Filing date: 2008-09-01
Publication date: 2009-03-05
Also published as: EP2191230A1; KR20100052546A; CN101821581A; NO20074444L; NO328737B1

Abstract

A device for inspection of an object comprises a light source for illuminating the object, a shadowing device arranged between the light source and the object for generating a shadow on the surface of the object, an image capturing device for capturing an image of the surface of the object, and an processing unit for processing the image data to provide an indication of the quality of the surface of the object. There is also provided a method for inspection of objects which provides a shadow on at least one surface of the object, captures an image of the surface of the object including the shadow, and analyses at least one edge of the shadow in the image to provide an indication of the quality of the surface of the object.

Description

METHOD AND DEVICE FOR INSPECTION OF OBJECT SURFACES

The invention regards a method and device for inspection of objects.

In a number of applications there exists a need of being able to inspect the surface of objects, for wafers such as semiconductor substrates for use in electronics and/or solar cells.

Various solutions have been used for this purpose. JP 2006292617 describes an optical approach, where a scratch on a semiconductor substrate is irradiated from a surface with illumination light emitted from a light source and transmitted through an annular filter. The scratch is detected based on the electric signal from a photoelectric converter derived from interference.

JP 2005221288 describes a method for measuring surface properties of a material by projecting a shadow onto the material by an angled light source. The image of the shadow is captured by a camera and the shape/extent of the shadow is used to calculate the surface state of the material. The material may for example be timber, metal or plastic. The described method is thus suited for measuring properties of relatively rough surfaces, and has not the accuracy and resolution needed for inspecting small irregularities in the surface, for example due to sawing a wafer from a semiconductor material/substrate.

The object of the invention is to provide a method and device for inspection of the surface of semiconductor substrates, which has very high accuracy and high reliability and which is easy to use, is compact and has low maintenance costs. The semiconductor substrates are for example silicon wafers for solar cells.

The object of the invention is achieved by means of the features of the patent claims. The device for inspection of the surface of semiconductor objects, comprises in one embodiment:

- a light source for illuminating the object,

- a shadowing device arranged between the light source and the object for generating a shadow on the surface of the object, - an image capturing device for capturing an image of the surface of the object,

- and an processing unit for processing the image data to provide an indication of the quality of the surface of the object.

The object to be inspected may be an object where it is important to inspect the surface. The objects may have any shape and size. In one embodiment, the object to be inspected is a wafer, for example a semiconductor wafer or a photovoltaic waver for solar cells. The quality of the surface may for example comprise information of height variations of the surface, for example due to sawing wafers from larger blocks/ingots.

The light source may be any suitable light source such as a halogen lamp, one or more light emitting diodes, etc. The light source is illuminating the shadowing device in order to generate a shadow on the surface of the object. The distance between the light source and the shadowing device and/or the object may be optimized to proved sharp edges of the shadow. For example will a halogen lamp arranged a relatively large distance from the shadowing device imitate an ideal point source of light. There may also be a number of light sources.

The shadowing device may be any device providing a suitable shadow on at least one surface of the object. The shadowing may have any desirable shape and size and there may be any number of shadowing devices. Using several shadowing devices provides increased redundancy in the measurements and thus increased robustness of the measurements. In one embodiment, the shadowing device has a longitudinal shape, for example a string or a rod.

The image capturing device may be any imaging device such as a camera, an image sensor such as CCD or CMOS device, etc. The image capturing device may also comprise lens assembly or other focusing optics in order to obtain a sharp image of the object surface and/or the shadow(s). There may be one or several image capturing devices.

The processing unit is adapted for processing the image data and analysing at least one edge of the shadow in the image to provide an indication of the quality of the surface of the object. The shape of the shadow will vary with the height variations of the surface of the object, and thus height information of the surface of the object can be derived from the image data of the edges of the shadow.

In one embodiment the correspondence between corresponding points on the edges of the shadows is calculated and this correspondence is used as a measure of quality of the shadow in that point. If the correspondence is poor, ie. the difference between the value in the points is above a threshold, the points are rejected in the further calculations.

In one embodiment analysing the edge of the shadow in the image comprises calculating variations in the shape of the edge.

In one embodiment, the location of the object is identified in a reference system. In one embodiment the location of the shadow on the surface of the object is identified. Also, the location of the edges of the shadow can be identified on the surface of the object. In one embodiment, the image is normalized. This is particularly interesting when the object is of a multi-crystalline material, as the difference in reflectivity of the surface caused by the different crystal properties may generate noise in the measurements. The normalizing may for example be performed by additionally capturing an image of the surface of the object without shadow, identifying the location of the object without shadow in the reference system and combine the image of the object without shadow from the image of the object with shadow. The combination of images may be performed by subtracting or dividing the image of the object without shadow from/of the image of the object with shadow. Normalization by division leads to improved contrast and signal strength in the resulting image, thus bringing out the differences in the image between the dark (shadows) and light portions.

Identification of the location of the object enables the subtraction and/or division to be done for the same sections of the surface of the objects in the two images, thus ensuring correct normalization and little noise in the measurement.

In one embodiment the analysis of the at least one edge of the shadow in the image provides a measure of height variations of the surface of the object.

In one embodiment the above described steps are performed for two opposite surfaces of the object. This provides a height profile/curve for each of the surfaces. Subtracting the results can provide measure of the thickness of the object. In one embodiment, an object with known thickness is also measured in the same way as described above, and this known thickness is combined with the measured thickness from the height profiles to give the correct actual thickness of the object to be inspected. In one embodiment the object is a wafer, for example a semiconductor wafer such as for use in solar cells.

The invention will now be described in more detail by means of examples and with reference to the accompanying figures.

Fig. 1 illustrates schematically the principles of the invention. Fig. 2 illustrates one possible layout for one embodiment of the invention. Fig. 3 shows an example of an image with shadows.

Fig. 4 - 8 shows different stages of embodiments of the method according to the invention.

Fig. 9 illustrates a normalization process according to one embodiment of the invention.

Fig. 10 illustrates the principles of one embodiment of the invention for providing an indication of the thickness of the object. In figure 1 a surface 10 of an object 14 is inspected. A light source 11 illuminates the object 14 as well as the shadowing device 13 which is arranged between the light source 11 and the object 14. The shadowing device 13 thus generates a shadow 15 on the surface 10 of the object. An image capturing device 12 is arranged to capture an image of the surface 10 of the object including the shadow 15. The image data is then processed/analysed in order to provide an indication of the quality of the surface of the object.

The light source 11 may be any suitable light source such as a halogen lamp, one or more light emitting diodes, etc. For illustration purposes the light source 11 is in this example depicted as a single light bulb, for example a halogen bulb. The light source 11 is arranged laterally shifted with respect to the object 14, which leads to the light path from the light source 11 via the shadowing device 13 to the surface 10 of the object being angled with an angle with respect to the surface 10 of the object. The angle may for example be in the range 20 - 40°. The distance between the light source 11 and the shadowing device 13 and/or the object may be optimized to provide sharp edges of the shadow. In the figure the distance between the shadowing device 13 and the object 14 is significantly smaller than the distance between the light source 11 and the shadowing device 13. It may be advantageous to have a small distance from the shadowing device 13 to the object 14 to achieve a sharp shadow on the surface 10 of the object 14. The distance from the shadowing device 13 to the object 14 may be in the range 1-5 mm, for example in the range 1.5-2.5 mm while the distance from the light source to the object may be in the magnitude range 50-250 mm.

The shadowing device may be any device providing a suitable shadow on at least one surface of the object. The shadowing device may have any desirable shape and size and there may be any number of shadowing devices. In figure 1 there are two shadowing devices 13 arranged on opposite side sections of the same surface of the object 14. Other embodiments may employ other configurations of the shadowing device, for example over the central section of the object. The image capturing device 12 is in figure 1 depicted as a camera, but may be any imaging device such as an image sensor such as CCD or CMOS device, etc. The image capturing device may also comprise lens assembly or other focusing optics in order to obtain a sharp image of the object surface and/or the shadow(s). In figure 1 there is only one image capturing device 12, but in other embodiments there may be several image capturing devices.

The shape of the shadow 15 on the surface 10 will vary with the height variations of the surface 10 of the object 14. This shape variation can be detected from the image captured by the image capturing device, and thus height information of the surface of the object can be derived. This may for example be derived by calculating the position of the edge of the shadow in a number of points, a sequence of edge shadow positions constituting a shadow edge profile corresponding to a height curve/profile representing the surface 10 of the object. In order to obtain calculations in a number of points/locations, the shadowing device and the object may be moved with respect to each other. In the example of figure 1, the object is not moving during the image capturing, but is moved laterally for inspection of the underside of the object or for further processing of the object. The height of the surface variations may be calculated from the shadow edge profile by means of geometric calculations when the distance from the object to the shadowing device and the light source is known. Calculation of a number of height curves/profiles along the surface may be combined to form a 3-dimensional model of the surface of the object.

In order to provide an accurate calculation of the shadow positions, the image should be of good quality. This may be ensured by using an image capturing device with high resolution and/or calibrating the image capturing device, for example by using a n-th order lens calibration or other known calibration technique. The image capturing device may also be adapted to control the exposure and/or contrast of the image.

In figure 2 an example of a layout for embodying the invention is shown. An object 20 which is to be inspected is arranged on a transport device 21. The transport device 21 enables continuous measurement of the object 20 which is moved in the direction of the arrow. The transport device 21 in this embodiment has support elements 24, 25 for supporting and moving the object 20. The support elements are alternately arranged to support different sections of the surface of the object. This enables measurement of two opposite surfaces of the object, for example the under and upper surface. If both the under surface and the upper surface of the object

(when arranged on the transport device) is to be inspected, there should be arranged light sources and shadowing devices over and under the transport device 21. Measuring both sides of the object eliminates the need for accurate height positioning/alignment of the object, as the two measurements will cancel out the height and/or positioning inaccuracies. In figure 2 first support elements 24 support the object's outer edge sections in the first part 26 of the transport path, then the object is transferred to the second part 27 of the transport path where second support elements 25 support the object's central section. In this configuration, a first shadowing device 22 is arranged between the first support elements and measures the properties of the central section of the under surface of the object, while a second shadowing device 23 is arranged outside the second support elements 25 and measures the properties of the edge sections of the under surface of the object. In one embodiment there is arranged several light imaging capturing devices and light sources, for example the number of light sources corresponding to the number of image capturing devices. The measurements generated from the different images captured in different positions of the object may be added to provide an indication of the surface over the full width of the object surface.

Fig. 3 shows an example of an image 30 of an object 31 with shadows 32. The shadows 32 are provided by an elongated shadowing object 33 arranged above the object. The elongated shadowing object may for example be a stretched string, a rod, or metal strip etched on a glass plate. In the example of figure 3, each shadowing element comprises three elongated elements, each having two longitudinal edges, thus each shadowing element having six longitudinal edges. In other configurations, there may be other numbers of elongated elements, for example 4 or 5. This provides good redundancy in the calculations and thus robustness of the system. The longitudinal elements are arranged close to each other, with a typical center-to-center distance of 2 mm.

Fig. 4 - 8 shows different stages of embodiments of the method according to the invention where an image is analysed my means of image processing/analysing. In figure 4, the edges 40, 41 of the object 31 are identified and their location is calculated in a reference system by means of image processing/analysis. The location of the edges in the reference system may then be used to calculate the dimensions and shape of the object. The reference system may be related to the image area of the image capturing device. The location of the edges is used in the next step, shown in figure 5, where the location of the object 31 is identified in the reference system. In figure 6, the shadows 32 are identified in the image and the location of the shadows 32 on the surface of the object 31 is identified. In figure 7, the image processing uses the location information of the shadows 32 to identify the location of the shadows' edges 70-75. The quality of the shadow edges is evaluated, in order to identify the shadows suitable for making the calculations. The quality of the shadow's edges may be deteriorated due to artefacts in the images, for example due to particles of dust on the surface of the object. The position of the edges is calculated in a number of points, a sequence of points constituting a shadow edge profile. The quality of the shadow's edges is evaluated for example by statistical analysis. Corresponding points for both edges of all shadows (in figure 7 there are three shadows) are compared and the correlation between them is calculated. If a satisfactory, predetermined, number of points correlate (i.e. differ with less than a predetermined threshold), the points are accepted, if not, the points are rejected. In case of four shadows (eight edges) in each shadowing element, the acceptable number of points corresponding may e.g. be four, while for three shadows (six edges) the acceptable number may be three or four. The accepted points are assigned a common value, which for example is the median of the correlating points. In the case where not sufficient points correlate and the points are rejected, the value of these points is interpolated by using adjacent points. If too many points must be interpolated, the whole shadow is rejected and that area of the surface of the object is not evaluated. From the shadow edge profile, a height profile of the surface of the object may be calculated, for example by means of trigonometric calculations. Figure 8 shows a calculated height profile 80 superimposed on the image 30 of the object 31 including shadow 32. Fig. 9 illustrates the effect of a normalization process according to one embodiment of the invention. The normalizing is performed by capturing an image of the surface of the object without shadow, identifying the location of the object without shadow in the reference system and using the two images of the surface of the object, together with the location information, to normalize the image, for example by means of subtracting or dividing the image of the object without shadow from/with the image of the object with shadow. The result of this normalisation can be seen in the figure as a white section 90. The normalisation eliminates image and signal noise such as vibrations and noise generated from large differences in reflectivity of the surface of the object. The latter is particularly a problem when the method and device according to the invention is used for inspecting multi-crystalline wafers for solar cells. The multi crystalline structure causes noise in the measurements which is cancelled out by the normalization process.

Fig. 10 illustrates the principles of one embodiment of the invention for providing an indication of the thickness of the object. The surface of the object is measured as described above on two opposite outside surfaces 101 and 102. This is performed both for the object to be inspected 104 and for a reference object 103 with known thickness. The measurement of the reference object 103 is shown in figure 10a, while figure 10b shows the measurement of the object to be inspected 104. Height profiles are obtained from each surface 101, 102 as described above. The height profiles are then subtracted in a number of positions/location to calculate a thickness of the object for each calculated height profile position/point. In one embodiment, a number of height profiles are calculated, and an average of these are used to calculate the thickness profile. The averaging may be performed for each height location in a series of locations forming a height profile, or for the thickness profile calculated from un-averaged height profiles. The calculated thickness profile of the reference object 103 is used to calculate gain and offset to achieve a correct thickness of the object.

Claims

1. Method for inspection of the surface of semiconductor objects, characterised in that the method comprises the following steps:

- providing a number of shadows on at least one surface of the object, - capturing an image of the surface of the object including the shadow,

- analysing at least two edges of the shadow(s) in the image to provide an indication of the quality of the surface of the object.

2. Method according to claim 1, characterised in that analysing the edge of the shadow in the image comprises calculating variations in the shape of the edge.

3. Method according to claim 1 or 2, characterised in that it further comprises:

- identifying the location of the object in a reference system.

4. Method according to claim 1 -3, characterised in identifying the location of the shadow on the surface of the object.

5. Method according to claim 4, characterised in identifying the location of the edges of the shadow on the surface of the object.

6. Method according to claim 1-3, characterised in calculating the correspondence between corresponding points on the edges of the shadows and use this correspondence as a measure of quality of the shadow in that point.

7. Method according to claim 5, chacterised in calculating the dimensions of the object based on the location of the edges.

8. Method according to claim 1, characterised in normalizing the image.

9. Method according to claim 8, characterised in that the normalizing is performed by additionally capturing an image of the surface of the object without shadow, identifying the location of the object without shadow in the reference system and subtracting the image of the object without shadow from the image of the object with shadow.

10. Method according to claim 8, characterised in that the normalizing is performed by additionally capturing an image of the surface of the object without shadow, identifying the location of the object without shadow in the reference system and dividing the image of the object with shadow with the image of the object without shadow.

11. Method according to claim 1, characterised in that the analysis of the at least one edge of the shadow in the image provides a measure of height variations of the surface of the object.

12. Method according to claim 11, characterised in that the steps are performed for two opposite surfaces of the object, and the results are subtracted in order to achieve a measure of the thickness of the object.

13. Method according to one of the previous claims, characterised in that the object is a wafer.

14. Method according to claim 1, characterised in that the object is a semiconductor wafer.

15. Device for inspection of the surface of semiconductor objects, characteris e d in that it comprises: - a light source for illuminating the object,

- a shadowing device arranged between the light source and the object for generating a number of shadows on the surface of the object,

- an image capturing device for capturing an image of the surface of the object including shadow, - and an processing unit for processing the image data to provide an indication of the quality of the surface of the object.

16. Device according to claim 15, characterised in that the processing unit is adapted for calculating variations in the shape of the shadow.

17. Device according to claim 15, characterised in that the processing unit is adapted for identifying the location of the object in a reference system.

18. Device according to claim 15- 17, characterised in that the processing unit is adapted for identifying the location of the shadow.

19. Device according to claim 18, characterised in that the processing unit is adapted for identifying the location of the edges of the shadow.

20. Device according to claim 15, characterised in that the processing unit is adapted for normalizing the image.

21. Device according to claim 20, characterised in that the normalizing is performed by capturing an image of the surface of the object without shadow, identifying the location of the object without shadow and subtracting the image of the object without shadow from the image of the object with shadow.

22. Device according to claim 20, characterised in that the processing unit is adapted for providing a measure of height variations of the surface of the object.

23. Device according to one of the claims 15-22, characterised in that the object is a wafer.

24. Device according to claim 23, characterised in that the object is a semiconductor wafer.