INSPECTION APPARATUS AND METHOD

Description

The invention relates to an inspection apparatus as defined in the preamble of claim 1 and an inspection method as defined in the preamble of claim 19 for detection of features in a work object. While the invented subject matter may be used in various applications, it will be illustrated herein in the context of the fabrication of silicon wafers. Wafer fabrication involves the drawing of a single continuous ingot which then is sliced into wafers of standard diameters and thickness. These standard wafers are then shipped to a semiconductor processing company for use in creating semiconductor devices, namely, integrated circuits .

In some instances, a semiconductor company will further process the wafer using specialized equipment, in particular grinding equipment. Grinding the unpatterned silicon (e.g. bare), wafers may result in the appearance of structural voids, inclusions and/or pinhole defects that were previously internal to the wafer. These voids, inclusions and/or pinholes may be considered defects rendering the wafer unusable for further processing.

Inspection apparatus of the relevant kind for detection of features in a work object are known and include a radiation source directing radiation to the work object in use. Furthermore, the known inspection apparatus include an image capturing system including

at least one sensor.

It is an object of the present invention to provide an inspection apparatus which may be manufactured at relatively low costs and which allows for an inspection of a work object with high resolution.

This object is achieved by the invention defined in claim 1.

The invention is based on the idea to use near- infrared light for imaging the work object. Within the terms of the invention near-infrared (NIR) light includes light in a spectrum between 760 nm and 2,500 nm, in particular between 950 nm and 1,300 nm. Particularly preferred is the spectrum between 1,000 and 1,200 nm. Surprisingly, it has been found that particularly in the inspection of silicon wafers, using NIR light results in an excellent image quality.

It is a further advantage of the invention that using NIR light allows for standard equipment in particular standard cameras to be used.

It is a further advantage of the invention that NIR light results in a resolution that is sufficient to image some finer wafer structures .

Preferably, the inventive inspection apparatus includes at least one NIR responsive camera.

According to a further preferred embodiment, the image capturing system includes at least one NIR transmissive lens or lens assembly.

According to the respective application of the inspection apparatus it may be sufficient if the work object and the sensor are mounted in fixed relationship to each other. However, according to a preferred embodiment positioning means are provided for positioning the work object and the sensor relative to

each other. In this embodiment, the work object may be inspected at various locations. In accordance with this embodiment, one of the work object and the sensor may ^¬ be fixed while the other components may be movable. However, in accordance with the embodiment both the work object and the sensor may be movable.

In accordance with a further development of the aforementioned embodiment the positioning means are adapted for positioning the work object and the sensor substantially parallel to the surface of the work object. If e. g. the surface of the work object is located within a X-Y-plane, the work object and the sensor are positionable along the X- and the Y-axis. According to a further preferred embodiment, the positioning means are adapted for positioning the work object and the sensor substantially perpendicular to the surface of the work object. If again e. g. the surface of the work object is located within a X-Y plane, in this embodiment the work object and the sensor are positionable relative to each other along the Z axis.

Preferably, control means for controlling the positioning means during an inspection of the work objects are provided wherein according to a further preferred embodiment the work object and the sensor are positioned relative to each other automatically during an inspection of the work object. In this embodiment, inspection of the work object may be carried out automatically without human intervention. According to a further preferred embodiment, the image capturing system includes image processing and/or storing means. Said image processing and/or storing means may for example be embodied as a computer. According to a further development of the

beforetnentioned embodiment the image processing and/or storing means are adapted to compare acquired image data to a known reference, image characteristics or image representation. In this embodiment, evaluation of the acquired image data may be carried out in a particularly time-saving manner.

According to a further preferred embodiment the work object at least partially is translucent for the near- infrared light, such that the work object is located between the light source and the sensor.

However, it is possible that the work object at least partially is reflective for the near infrared light such that the light source and the sensor are located at the same side of the work object. In summary, the inventive inspection apparatus may be used for inspection of work objects which are translucent for the NIR light as well as for inspection of work objects which are reflective for the NIR light.

The inventive inspection apparatus may be used for inspecting various work objects. However, the apparatus is particularly suitable for inspection of wafers. Accordingly, further preferred embodiments of the invention provide that the work object is at least partially manufactured from semiconductor material, in particular silicon, wherein the work object preferably is a wafer.

However, according to further preferred embodiments the work object may be microstructure, in particular a micromechanical structure, wherein the micromechanical structure preferably is a MEMS (Microelectromechanical System) .

Further preferred embodiments of the invention provide that at least one feature to be detected is a defect and/or that at least one feature to be detected

is an intended structure in the work object.

An inventive method is claimed in claim 19. Preferred embodiments of the inventive method are claimed in claims 20 to 25. The invention will now be explained in greater detail with reference to the accompanying drawings which illustrate an embodiment of an inspection apparatus according to the invention for carrying out and embodiment of the inventive method, wherein all features described in the specification, shown in the drawings or claimed in the claims constitute the subject matter of the invention, either taken alone or in arbitrary combination with each other, regardless of their combination in the claims and the reference of the claims as well as regardless of their description in the specification and of the illustration in the drawings .

In the drawings Fig. 1 depicts a system diagram of a first embodiment of an inspection apparatus according to the invention,

Fig. 2 depicts a system diagram of a second embodiment of an inspection apparatus according to the invention,

Fig. 3 depicts a system diagram of a third embodiment of an inspection apparatus according to the invention,

Fig. 4 is a side view of a wafer and a camera subsection of the inspection apparatus shown in fig. 2,

Fig. 5 is a view of the X-Y-Z inspection apparatus shown in fig. 2 operating in conjunction with an X-Y defect

inspection apparatus,

Fig. 6 is a view of the inspection apparatus shown in fig. 2 configured to operate in the same work space as an X-Y inspection apparatus,

Fig. 7 is a view of an X-Y-Z inspection apparatus integrated with an X-Y inspection apparatus using adjustable optics and Fig. 8 shows an example of an NIR image captured by an inspection apparatus according to the invention.

Fig. 1 illustrates a first embodiment of an inspection apparatus 2 according to the invention.

The inspection apparatus 2 includes a radiation source 4 directing radiation to a work object which in the present embodiment is a wafer 6. Furthermore, the inspection apparatus 2 includes an image capturing system 8 including a camera 10, lens 12, wherein the camera 10 is in data transfer connection with an imaging processing and/or storing means which in the present embodiment is embodied as a computer 14.

For positioning the work object 6 and the camera 10 relative to each other positioning means 15 are provided which in the present embodiment are adapted for positioning the wafer 6 and the camera 10 parallel to the surface of the wafer 6 along the Y and the X axis which in figure 1 are symbolized by arrows 16 and 18 respectively.

According to the invention the radiation source 4 is a NIR light source outputting near infrared light in use. In order to capture images of the wafer 6 the camera 10 is a NIR responsive camera and the lens 12 is

a NIR transmissive lens.

Furthermore, the apparatus 2 includes control means for controlling the positioning means 15 during an inspection of the wafer 6, wherein in the present embodiment the control means are constituted by the computer 14.

In use, the radiation source 4 irradiates the work object 6 with NIR light, wherein the camera 10 is used to capture images of the wafer 6 as the control means included in the computer 14 positions the camera 10 and the wafer 6 relative to each other.

As the control means positions the camera 10 and the wafer 6 relative to each other, the camera 10 takes incremental snapshots of the wafer 6, wherein the entire thickness of the wafer 6 is scanned in one image. Defects 24, 24' and 24 '' are detected at various points in the wafer 6. These physical defects 24, 24' and 24'' are stored by the image processing and storing means included in the computer 14. A software system included in the computer 14 can then determine, via image processing or through human intervention, whether the wafer 6 is acceptable or is to be rejected.

Due to the use of NIR light for inspection, images are captured quickly and with high resolution, wherein a relatively inexpensive camera 10 and lens 12 can be used. Therefore, the inventive inspection apparatus 2 may be manufactured at relatively low costs.

Fig. 2 illustrates a second embodiment of the inventive inspection apparatus 2' which differs from the embodiment illustrated in figure 1 mainly in that the positioning means are adapted for positioning the wafer 6 and the camera 10 substantially perpendicular to the surface of the wafer 6, i. e. in the illustrated embodiment along the Z axis. Furthermore, the camera 10

/ lens 12 assembly is constructed such that it has a shallow depth of field.

In use, the wafer 6 and the camera 10 are positionable relative to each other along the X, Y and Z axis. Consequently, images of the wafer 6 may be captured at arbitrary positions within the volume of the wafer 6 such that the X, Y and Z coordinate values of defects 24, 24', 24'' are determined by means of the inventive inspection apparatus 2. However, for X, Y and Z defect inspection the inspection apparatus 2 may also use known X-Y coordinate values. For example, the X-Y coordinate values may be determined using the apparatus 2 illustrated in figure 1. For further inspection by the apparatus 2' illustrated in figure 2, the known values could be obtained from a stored defect table loaded on an X-Y-Z image processing system of the computer 14. Alternatively, this known data could be determined in real time from the same or different system acquiring the Z image data, or it could be manually input. The defect table may consist of a list of X-Y positions on the wafer 6 where defects have been predetermined.

The positioning means 15 positions the camera 10, the light source 4 and the lens 12 to a predetermined location at which the positioning means incrementally adjust the lens 12 to scan the wafer from one side to the other along the Z axis. At each incremental adjustment along the Z axis, an image is taken by the camera 10 and processed by the X-Y-Z image processing system of the computer 14. Based on the captured images, the image processing and/or storing means can then create an X-Y-Z defect table. The X-Y-Z defect table being a list of defects 24, 24', 24'' not only including the X-Y coordinates of the defects but also

the depth information (Z axis) , the accuracy of the X- Y-Z defect table being dependent on the positioning means 15, the camera 10 and the image processing means.

Fig. 3 illustrates a third embodiment of the inventive inspection apparatus 2' which differs from the embodiment of figure 2 in that the camera 10 and the NIR light source 4 are located at different sides of the wafers 6 such that the NIR light source 4 is configured as a backlight source. Fig. 4 depicts a side view of the wafer 6 with a representative width of 700 μm. A portion of the inspection apparatus 2 is shown with the camera 10 and the lens 12. As explained above, the camera 10 / lens 12 assembly is positionable along the Z axis relative to the wafer 6. The camera 10 / lens 12 assembly has a shallow depth of field that is approximately 50 μm, for example. Three reprentative defects 24, 24' and 24'' are shown in the wafer at depth locations of 0 μm, 350 μm and 650 μm. An inspection of the wafer 6 is carried out using a defect table including X-Y coordinate values of the defects 24, 24', 24''. The first defect 24'' is located without adjusting the camera 10 / lens 12 assembly along the X and Y axis. The second defect 24' is located after the camera 10 adjust seven steps (7 x 50 μm = 350 μm) . Likewise, the third defect 24 is located after the camera 10 / lens 12 assembly adjust 13 steps (13 x 50 μm = 650 μm) .

The inspection apparatus 2' is not limited to storing binary values of whether a defect is located at a particular location along an axis. The apparatus 2' may store the actual image itself and with image postprocessing and interpolation being able to determine Z values that lie between incremental Z steps. The

positioning means 15 may position the camera 10 / lens 12 assembly relative to the wafer 6 in a stepwise manner as explained above. However, the positioning means 15 may position the camera 10 / lens 12 assembly relative to the wafer 6 continuously.

While the operation of the inspection apparatus 2 has been described with respect to the detection of defects, the apparatus 2 may also be used to detect intended structures within the wafer 6. The embodiment of figure 2 may also be configured to work with other inspection systems as shown in figure 5. A non-infrared inspection system 26 inspects a wafer 6 and records a defect table. The wafer 6 is then transferred by conveyor or a similar mechanism to the apparatus illustrated in figure 2 for X-Y-Z inspection as described above. The X-Y-Z inspection apparatus 2' then views the wafer 6 referencing the X-Y table to determine the Z values of the defects 24, 24', 24'' detected by the inspection system 26. The embodiment of figure 2 may also be configured to work in tandem with the embodiment of figure 1. As shown in figure 6, a wafer 6 is placed in a mount 28. The apparatus 2 of figure 1 (X-Y defect inspection) is mounted on one plane above the wafer 6 and the apparatus 2' of figure 2 (X-Y-Z defect inspection) is mounted on a separate, non- intersecting plane. This configuration allows both systems to potentially work in tandem and sharing a common work space . Alternatively, a single system may be configured with all functions to determining X-Y-Z data.

As shown in figure 7, instead of a single lens 6 a lens assembly 30 having lenses with different depth of field may be used. In one configuration, a wider view length 32 is selected by rotating the lens 32 into the

position to view the wafer 6. The camera 10 is adjustable on the Z axis. By rotating the lens assembly 30 a lens 34 with a smaller depth of field or a lens 36 with an even smaller depth of field may be used. Fig. 8 shows an example of a NIR image of the wafer captured by the inventive inspection apparatus using NIR light.

The process of detecting defects is performed by the application of contrast algorithms to the captured images. Contrast algorithms consist of taking a known reference image or characteristics of an image that represents a defect- free area in the wafer and comparing it to the captured image or characteristics or a representation of the captured image. The two images are compared and a determination is made whether there are defects present.

Alternatively, instead of storing the characteristics of an actual image, the defect-free image is processed with an edge-detection algorithm, followed by histogram that computes a single number

(K) . Subsequent images can be processed by the same algorithms, where a higher K value would indicate a defect. The Z slice with the highest "K" would then be the Z location of the defect. Therefore, in the X-Y defect inspection apparatus a defect can be located by comparing a reference value or a computed value of a good wafer with the image recorded for the entire depth of the wafer.

In the inspection apparatus 2' shown in figure 2 modified contrast algorithms are used to determine if there are defects along the Z axis. An image is captured at each depth using the shallow depth of view camera. Each captured image is contrasted with a known reference value. If the algorithm detects sufficient

contrast, then the defect is noted in the X-Y-Z defect table.

The NIR light emitted by the light source 4 is directed to structures via an internal beam splitter in the lens system. The light, so directed, generally is reflected by structures at various intensities (e. g. depending on the bond characteristics and other features and defects of the semiconductor structure) , so as to travel back up through the lens system, to a camera, such camera being based on one or using one or more solid state imaging devices, e. g. CCD or CMOS detectors. The camera detects reflected radiation in the NIR spectrum. Via such detection, an image of the structures is captured. The image, so captured, may be provided for further processing via e. g. a computer. The captured image, so processed or otherwise, may be employed for test and quality control toward the identifying relevant features of such structures, e. g. where such relevant features are associated with bonded or stacked layers or with other bonded or stacked materials .