U.S. patent application number 12/170848 was filed with the patent office on 2008-12-04 for method for increasing the accuracy of the positioning of a first object relative to a second object.
This patent application is currently assigned to SUSS MicroTec Testsystems (GmbH). Invention is credited to Claus Dietrich, Jorg Kiesswetter, Stefan Schneidewind, Michael Teich, Thomas Tharigen.
Application Number | 20080298671 12/170848 |
Document ID | / |
Family ID | 33103491 |
Filed Date | 2008-12-04 |
United States Patent
Application |
20080298671 |
Kind Code |
A1 |
Schneidewind; Stefan ; et
al. |
December 4, 2008 |
Method For Increasing The Accuracy Of The Positioning Of A First
Object Relative To A Second Object
Abstract
A method is provided for increasing the accuracy of the
positioning of a first object relative to a second object. The
method overcomes the disadvantageous influence of thermal drift
between a first and a second object during a positioning of a first
object on a second object. The method finds applications in
manufacturing, for example, in the manufacturing of semiconductor
components. The method utilizes recognition of structures on the
second object which have a minimum structure width. At a first
instant, using one recognition procedure, the first object is
positioned on the second object in a desired position. The relative
displacement of the two objects is determined at the first instant
and on at least one subsequent instant. A second recognition
procedure may be used for this purpose. The second recognition
procedure may have a resolution accuracy which is different than
the resolution accuracy of the first resolution procedure. The
second recognition procedure may be a pattern recognition method.
The relative displacement determined at the second instant is used
to correct the positioning of the first and second objects as
necessary to maintain a desired position of the two objects.
Inventors: |
Schneidewind; Stefan;
(Reichenberg, DE) ; Dietrich; Claus; (Sacka,
DE) ; Kiesswetter; Jorg; (Dresden, DE) ;
Teich; Michael; (Riedewald, DE) ; Tharigen;
Thomas; (Dresden, DE) |
Correspondence
Address: |
BAKER BOTTS L.L.P.
30 ROCKEFELLER PLAZA, 44TH FLOOR
NEW YORK
NY
10112-4498
US
|
Assignee: |
SUSS MicroTec Testsystems
(GmbH)
|
Family ID: |
33103491 |
Appl. No.: |
12/170848 |
Filed: |
July 10, 2008 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10824884 |
Apr 15, 2004 |
|
|
|
12170848 |
|
|
|
|
Current U.S.
Class: |
382/151 |
Current CPC
Class: |
G06T 7/74 20170101; G06T
2207/30148 20130101 |
Class at
Publication: |
382/151 |
International
Class: |
G06K 9/00 20060101
G06K009/00 |
Foreign Application Data
Date |
Code |
Application Number |
Apr 16, 2003 |
DE |
103 17 778.7 |
Claims
1. A method for increasing the accuracy of the positioning of a
first object relative to a second object by utilizing a recognition
of structures on the second object that have a minimum structure
width, the method comprising the steps of: (1) acquiring images of
an observation region that encompasses at least the first object
and a desired position on the second object; (2) at a first instant
T.sub.0, by means of a first recognition method having a resolution
accuracy that is higher or better than the minimum structure width,
determining the position of the first object relative to a second
object; and (3) repositioning the first object relative the second
object to the desired position at a second instant, wherein at
least one of the first and the second objects are movable using a
positioning device, wherein before about the second instant, by
means of a second recognition method, a relative displacement of
the first object with respect to the second object is determined
with respect to their positions at the first instant, and wherein
step (3) further comprises correcting for the relative displacement
of the first object with respect to the second object.
2. The method of claim 1 wherein a pattern recognition method is
used as the second recognition method.
3. The method of claim 2 wherein the resolution accuracy of the
pattern recognition method is lower or poorer than about the
minimum structure width.
4. The method of claim 2, further comprising: bringing the
positioning device to a basic position x.sub.0, y.sub.0,
.phi..sub.0 at about the first instant T.sub.0, and further in
temporal proximity to the first instant T.sub.0 using the pattern
recognition method to acquire a first image pattern from the
observation region that encompasses at least a portion of the
second object and a second image pattern from the observation
region that encompasses at least a portion of the first object;
bringing the positioning device to a basic position x.sub.0,
y.sub.0, .phi..sub.0 before about the second instant, and further
using the pattern recognition method to acquire a third image
pattern from the observation region that encompasses at least a
portion of the second object, and a fourth image pattern from the
observation region that encompasses at least a portion of the first
object; by means of the pattern recognition method, determining a
first pattern displacement from the first and third image patterns
and a second pattern displacement from the second and fourth image
patterns and further determining the relative displacement from the
first and second pattern displacements; and using the relative
displacement to correct the position x.sub.0, y.sub.0, .phi..sub.0
of the positioning device to a desired position at the second
instant.
5. The method of claim 4 wherein at least one of the first image
pattern and the third image pattern is respectively identical the
second image pattern and the fourth image pattern.
6. The method of claim 1 further comprising, after the second
object is processed, determining the relative displacements of the
first object and further objects that have minimum structure widths
using steps that are identical to steps (1)-(3) to correct the
relative positions of the further objects and the first object.
7. The method of claim 1 further comprising, after the second
instant, repeating in time the determination of the relative
displacement of the first and second objects so as to maintain a
desired position of the first object on the second object.
Description
BACKGROUND OF THE INVENTION
[0001] In many areas of technology it is necessary to position
objects relative to one another with high precision. This
requirement also exists in the field of semiconductor technology,
for example, during the testing of semiconductor components, which
may be fabricated on semiconductor wafers. A plurality of
semiconductor components of identical configuration are generally
situated on a semiconductor wafer. In this case, so-called probers
are used for testing the semiconductor components. For this
purpose, contact pads are arranged at different locations in the
semiconductor components (at the same location for each
semiconductor component on the semiconductor wafer). During the
testing with the probers, contact is made with the contact pads by
the tips of contact-making needles in the probers.
[0002] By means of such contact-making, an electrical contact with
the semiconductor component is produced, on the one hand to apply
specific signals to the semiconductor component and on the other
hand to measure the reaction to said signals.
[0003] According to setting methods in the prior art, the
positioning of contact-making needle tips of a prober (referred to
as a first object) relative to contact pads on the semiconductor
component (referred to as the second object) is performed under
optical or visual control. In such methods, the semiconductor
component is observed visually from above by means of a microscope
and the prober needle tips are then positioned onto corresponding
contact pads under visual observation. If the contact needles
happen to be set or positioned such that they lie on the contact
pads of the semiconductor component, the setting operation is
ended.
[0004] In some probers, it is also possible to mount the contact
needles with a corresponding setting on a so-called needle card. A
specific needle card is used for each type of a semiconductor
component. In such cases it is then necessary to bring a further
semiconductor component to be tested below the set contact-making
needles so that the contact-making needles again make contact with
the contact pads. If this has been done, the next test operation
can be performed.
[0005] The positioning of each semiconductor component below a
structure of contact needles may be done manually under visual
observation using manual probers commonly known in the prior art.
Automatic probers also are known. In the case of automatic probers,
it may be possible for each semiconductor component to be brought
automatically below the structure of contact-making needles if the
distances or orientation of the needles and the semiconductor
component is known (e.g., a rotation angle .phi. are known). In
this case, the displacement of the semiconductor wafer required in
order to perform an exact positioning may be calculated.
[0006] The visual or optical observation required for placing the
needle tips on the contact pads may be performed by means of
automatic image recognition systems. In this case, a pattern
recognition is performed on or by an image of an observed region of
the semiconductor component. The image may be recorded by a video
camera, a CCD linear array or matrix or other image recording
devices at a first instant. On account of the surface structure of
the semiconductor component, the latter has a pattern. This pattern
is significant for the component. If a further identical component
is now to be tested, the latter then exhibits the same pattern.
From the positional difference between the two patterns, the
pattern recognition system can then determine geometrical
correction values required to allow the semiconductor component
that is currently to be tested to be positioned precisely below the
needle structure by means of a positioning device.
[0007] In the context of the increasing miniaturization of the
structures on the semiconductor components, considerable
requirements or demands are made for the proper positioning of the
semiconductor component with respect to the tips of the
contact-making needles. The small widths of the miniaturized
structures may be of the order of the wavelength of the light
making it difficult or impossible to visually or optically resolve
them to a sufficient extent. Thus, complex AFM probers, which
operate according to the principle of atomic force microscopy
(AFM), may be used for making contact with semiconductor component
structures down in the range of 100 nm width. In this case, a
contact-making needle is moved at a small distance above the
surface of the semiconductor component, in particular in the region
in which the contact pad is situated. As a result of the movement,
the topography image of the region of the semiconductor surface is
scanned on account of an interaction force occurring between the
contact needle and the semiconductor surface. The exact position of
the contact pad is may thus determined without the need for a
visual observation.
[0008] The contact-making needle is referred to as a cantilever in
the case of AFM probers. A piezo-drive is available for moving the
cantilever, by means of which the cantilever executes a scan
movement in order to obtain an image of the surface situated
underneath, a scan. When the AFM prober is used, at a first
instant, the region where contact is subsequently made is scanned
by means of the cantilever. Once the scan is present, the tip of
the cantilever is brought to the desired position determined and
brought into contact at a second instant.
[0009] What is problematic in this case is that the semiconductor
wafer and cantilever are exposed to thermal influences. This leads
to a thermal drift in the time period between the first and second
instants, i.e., expressed generally, a relative displacement arises
between the first and second objects. In particular, this
phenomenon occurs during the testing of semiconductor components
under thermally controlled conditions. A so-called thermo-chuck is
used in this case, which, on the one hand, fixedly clamps the
semiconductor wafer during the test operation and, on the other
hand, sets a desired temperature in a higher or lower temperature
range in comparison with room temperature. The temperature
alteration of the semiconductor wafer, for example on account of
the thermal radiation, also influences the drive of the cantilever,
as a result of which the drift occurs, which can no longer be
disregarded in particular in the case of the small structure widths
since, when making contact, the drift means that the cantilever no
longer meets the position which was previously determined during
the scan operation.
[0010] As is rapidly apparent, even thermostatic regulation of the
surroundings cannot provide a remedy here since the drift is
generated by the method itself. This problem area may also occur in
other fields of application, in particular in the field of
semiconductor technology, for example during bonding operations.
Therefore, as a general proposition, a thermal drift or other drift
between two objects can be problematic.
[0011] An object of the present invention is to prevent the
disadvantageous influence of a thermal drift or other drift between
a first and a second object during a positioning of a first object
on a second object.
SUMMARY OF THE INVENTION
[0012] In accordance with the principles of the invention, a method
for increasing the accuracy of the positioning of a first object
relative to a second object is provided. The method may find
particular applications in the production or manufacturing of
semiconductor components. The method utilizes a recognition of
structures on the second object, which have a minimum structure
width, for positioning the first object. In the method at a first
instant, the position of the first object relative to a second
object is determined by use of a first recognition method or
procedure which has a resolution accuracy which is higher than the
minimum structure width. The first object is then at a second
instant positioned at a desired position on the second object.
Either the first or the second object, or both, may be movable by
means of a positioning device. Images of an observation region that
encompasses at least the first object and the desired position are
acquired during the method.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Further features of the invention, its nature, and various
advantages will be more apparent from the following detailed
description and the accompanying drawings, wherein like reference
characters represent like elements throughout, and in which:
[0014] FIG. 1 is a schematic representation of an image of an
observation region at a first instant in the process of placing a
first object on a second object, in accordance with the principles
of the present invention; and
[0015] FIG. 2 is a schematic representation of an image of an
observation region at a second instant in the process of placing a
first object on a second object, in accordance with the principles
of the present invention.
[0016] The following is a list of reference symbols used in the
FIGS. 1 and 2:
LIST OF REFERENCE SYMBOLS
[0017] 1 Semiconductor component [0018] 2 Cantilever [0019] 3
Contact pad [0020] 4 Tip of the cantilever [0021] 5 Observation
region [0022] 6 Structural element
DETAILED DESCRIPTION OF THE INVENTION
[0023] The present invention provides a method for overcoming the
influence of a thermal drift or other drift between a first and a
second object during a positioning of a first object on a second
object.
[0024] The inventive method overcomes the influence of thermal or
other drift between the two objects by virtue of the fact that in
the method, before or at a second instant, by means of a second
recognition method, a relative displacement of the first object
with respect to the second object is determined with regard to a
first instant, but at least with respect to the temporal proximity
thereof, and the position of the second object is corrected during
the positioning on the second object by means of correction values
which correspond to the relative displacement determined.
[0025] This method eliminates any temperature drift which may occur
between the first instant and the second instant by ascertaining
and correcting for the relative displacement of the two objects
between the first and second instants.
[0026] A favorable embodiment of the method provides for a pattern
recognition method to be used as the second recognition method.
Pattern recognition methods record images of the observation region
and acquire patterns contained in said images. By comparing two
identical patterns which are displaced or rotated relative to one
another in the imaging, it is possible to determine the coordinate
differences of each pixel of the two patterns. Computer-aided
calculation of the positional displacement of the two patterns with
respect to one another is thus possible. According to the
invention, a pattern recognition method may now be superimposed on
the controlled positioning of the first object on the second
object, thereby enabling the positional correction.
[0027] Since the pattern recognition methods recognize patterns
which need not necessarily represent sharp images, it is possible
that the resolution accuracy of the pattern recognition method is
lower (i.e. less fine) than the minimum structure width. As a
result, the method according to the invention can be realized very
simply and cost-effectively.
[0028] Although it is possible, in principle, to carry out the
second recognition method with a resolution accuracy which
corresponds to or is even greater (i.e., finer) than the minimum
structure width, it is also possible to use a method which makes
has less stringent requirements of the resolution accuracy. By way
of example, it is possible to use scanning electron microscopes for
the second recognition method in the cases of small structure
widths in semiconductor technology that lie in the wavelength range
of the light. These scanning electron microscopes would then
sharply image the observation region. However, this sharp imaging
would represent a pattern for a pattern recognition system in
exactly the same way as the image of an optical microscope which
would inevitably be unsharp on account of the proximity to the
wavelength of the light. However, since the unsharpness does not
adversely influence the characteristic of a pattern (in contrast to
an image), the pattern recognition method can thus operate in the
light range, i.e.--in this example--with a lower resolution
accuracy than the minimum structure width.
[0029] A particularly preferred embodiment of the method according
the invention may feature one or more of the following steps:
[0030] a step wherein the positioning device is brought into a
basic position x.sub.0, y.sub.0, .phi..sub.0 at the first instant
T.sub.0; [0031] a step wherein in temporal proximity to the first
instant T.sub.0, while the positioning device is situated in the
basic position x.sub.0, y.sub.0, .phi..sub.0, the pattern
recognition method is used to acquire a first image pattern from
the observation region, which encompasses at least the second
object; [0032] a step wherein in temporal proximity to the first
instant T.sub.0, while the positioning device is situated in the
basic position x.sub.0, y.sub.0, .phi..sub.0, the pattern
recognition method is used to acquire a second image pattern from
the observation region, which encompasses at least the first
object; [0033] a step wherein the positioning apparatus is brought
into the basic position x.sub.0, y.sub.0, .phi..sub.0 before the
second instant, the pattern recognition method is used to acquire a
third image pattern from the observation region, which encompasses
at least the second object, and the pattern recognition method is
used to acquire a fourth image pattern from the observation region,
which encompasses at least the first object, [0034] step wherein,
by means of the pattern recognition method, a first pattern
displacement of the first object is determined from the first and
third image patterns and a second pattern displacement is
determined from the second and fourth image patterns and the
relative displacement is calculated from the first and second
pattern displacements; and [0035] a step wherein the relative
displacement calculated is used to correct the desired position
x.sub.1, y.sub.1, .phi..sub.1 of the positioning device at the
second instant.
[0036] The drift both of the first and of the second object with
regard to the basic position of the positioning device is
determined by means of this embodiment. The displacements of both
objects are thus concomitantly included or determined. From the
difference between the respective two image patterns of the two
objects it is possible to determine the displacement of the image
patterns of the respective object and thus that of the object
itself. The relative displacement of the two objects with respect
to one another is then calculated from the displacements of the two
objects, which becomes possible since the displacements of the two
objects relate to a common basis, namely that of the basic
position.
[0037] In an expedient manner, in a pattern recognition method,
only in each case a common image pattern of the first and the
second object is recorded only in the basic position and in the
desired position of the positioning device. Since the pattern of
the first object can be assumed to be known, the first pattern (or
the second pattern if the configuration thereof is known) can then
already be established from the common image pattern by means of
the pattern recognition method. In this case, in one development of
the method, the first image pattern is identical to the second
image pattern and/or the third image pattern is identical to the
fourth image pattern.
[0038] In a further variant of the method according to the
invention, it is provided that, after the second object, the
relative displacements of further objects are determined in an
identical manner, from which, during the positioning of the further
objects on the first object, correction values for correcting their
desired positions are likewise determined. By way of example, if a
plurality of contact needles or cantilevers are used for testing
semiconductor components, it thus becomes possible to correct all
the drifts of all these objects.
[0039] In order to ensure that the first objects also actually
remain in the object desired position even if a drift occurs during
the further progression after the positioning of the first object
on the second object, it is provided that, after the second
instant, the determination of the relative displacement with regard
to the temporal proximity of the first instant is repeated and the
position of the positioned first object on the second object is
tracked such that the object desired position of the first object
on the second object is complied with.
[0040] The invention will now be explained in more detail below on
the basis of an exemplary embodiment with reference to FIGS. 1 and
2 which show images of an observation region at a first and second
instant, respectively. The exemplary embodiment relates to the
testing of semiconductor components 1 by means of a cantilever 2.
Electrically conductive connections (not specifically shown) are
connected to the cantilever 2 and serve for applying test signals
to the cantilever 2 and for recording and forwarding reaction
signals.
[0041] The cantilever is also connected to a positioning device
(not specifically shown). This positioning device is driven by a
piezo-crystal which may execute only very small movements as seen
macroscopically but, as seen microscopically, whose movements can
cover the entire observation region. These movements can be
executed very rapidly by means of the piezo-crystal, so that the
cantilever 2 can be scanned over the surface of the semiconductor
component 1. The surface may thus be sensed by means of the
principle of atomic force microscopy. Consequently, the position of
a contact pad 3 onto which the tip 4 of the cantilever 2 can be
positioned is also detected.
[0042] In the first position illustrated in FIG. 1, the sensing of
the surface of the semiconductor component 1 has already been
concluded. The tip 4 thus "knows" its desired position on the
contact pad 3.
[0043] The surface of the semiconductor component 1 is observed by
means of a CCD camera over the observation region 5. The image of
the observation region shown in FIG. 1 and FIG. 2 is only
figurative--since the minimum structure width is .about.100 nm and
the observation region is therefore imaged in an unsharp or diffuse
manner.
[0044] The image recorded by the CCD camera is processed further in
the further process, as is demonstrated below. The image may also
be displayed by means of a monitor for observing the operation.
[0045] Shortly before the scanning of the surface, the
semiconductor wafer on which the semiconductor component 1 is
situated (the semiconductor component being shown only partially in
FIG. 1 and FIG. 2,) may have been placed onto a thermo-chuck in
order to carry out the testing under elevated temperatures. The
semiconductor wafer is thus heated. The heating process still
persists at the first instant illustrated in FIG. 1.
[0046] The heating process may give rise to a thermal drift, which
becomes visible in FIG. 2. FIG. 2 illustrates the observation
region at the second instant. Dashed lines are used therein to
illustrate the position of the cantilever 2, of the contact pad 3
and of further structural elements 6 from FIG. 1. It thus becomes
possible to see the drift between the first and second instants in
the form of a displacement .DELTA.y.sub.obj2, .DELTA.x.sub.obj2 of
the contact pad 3 and of the structural elements 6 in the x and y
directions and a displacement .DELTA.y.sub.obj1 of the cantilever 2
in the y direction. The cantilever 2 has not experienced a drift in
the x direction and neither the cantilever 2 nor the semiconductor
component 1 has experienced an angular displacement by the rotation
angle .phi..
[0047] The exemplary embodiment is shown or described with only one
cantilever 2. In practice, however, a plurality of cantilevers may
be used, the method described below being employed
correspondingly.
[0048] Directly after the scan described above, shortly after the
first instant T.sub.0, the basic position x.sub.0, y.sub.0,
.phi..sub.0 is adopted by means of the positioning device. An image
pattern is taken there and is compared with the image pattern in
the basic position x.sub.0, y.sub.0, .phi..sub.0 at the second
instant T.sub.1. The pattern comparison is used to calculate the
relative displacement between the semiconductor component 1 and the
cantilever 2 from the displacements .DELTA.y'.sub.obj2,
.DELTA.x'.sub.obj1 and .DELTA.y'.sub.obj1 of the image patterns,
which correspond to the real displacements .DELTA.y.sub.obj2,
.DELTA.x.sub.obj2 and .DELTA.y.sub.obj1 of the two objects. In the
case of a setting of the desired position x.sub.1, y.sub.1,
.phi..sub.1 for achieving an object desired position in which the
cantilever 2 lies above the contact pad 3, the desired position
x.sub.1, y.sub.1, .phi..sub.1 is calculated correctively using the
relative displacement.
[0049] Usually, in each case only one image pattern of the first
and second instants are used for determining the displacements
.DELTA.y'.sub.obj2, .DELTA.x'.sub.obj1 and .DELTA.y'.sub.obj1 of
the image patterns. In this case, it is necessary for the system to
be taught the structure of the cantilever and the image pattern of
the semiconductor component 1.
[0050] The methods for the one cantilever case and the multiple
cantilever case may be subdivided into a number of steps or
substeps from a mathematical view point.
[0051] The following steps may be carried out for one cantilever:
[0052] 1. Learning of the cantilever models (as standard models,
only necessary in the case of a new type of cantilever or in the
case of another enlargement), [0053] 2. Calibration of the pattern
recognition system with respect to the positioning drive of the
respective cantilever 2 (only in the case of new installation or
change in the enlargement), [0054] 3. Movement to the observation
region 5, [0055] 4. Movement of the cantilever 2 out of the
observation region 5, [0056] 5. Automatic learning of the structure
of the semiconductor component 1, [0057] 6. Scanning and movement
into basic position, [0058] 7. Acquisition of the structure and
cantilever coordinates, [0059] 8. Readjustment of the cantilever
positions relative to the structure coordinates with the aid of the
positioning device, [0060] 9. Renewed acquisition of the structure
and cantilever coordinates and possible post-correction (successive
approximation), [0061] 10. Movement into object desired position.
The following steps may be carried out for a case with a plurality
of cantilevers: [0062] 1. Learning of the cantilever models (as
standard models, only necessary in the case of a new type of
cantilever or in the case of another enlargement), [0063] 2.
Calibration of the pattern recognition system with respect to the
positioning drive of the respective cantilever 2 (only in the case
of new installation or change in the enlargement), [0064] 3.
Movement to and learning of a specific test structure, [0065] 4.
Constant movement of all the cantilever tips 4 on the test
structure and acquisition of the structure and cantilever
coordinates, [0066] 5. Scanning with a cantilever, [0067] 6.
Movement into basic position of the respective contact position
(the contact position which corresponds to a cantilever) and
acquisition of the test structure and cantilever coordinates,
[0068] 7. Movement to the observation region 5, [0069] 8. Movement
of the cantilever 2 out of the observation region 5, [0070] 9.
Automatic learning of the structure of the semiconductor component
1, [0071] 10. Scanning and movement into basic position, [0072] 11.
Acquisition of the structure and cantilever coordinates, [0073] 12.
Readjustment of the cantilever positions relative to the structure
coordinates with the aid of the positioning device, [0074] 13.
Renewed acquisition of the structure and cantilever coordinates and
possible post-correction (successive approximation), [0075] 14.
Movement into object desired position.
[0076] It will be understood that the foregoing is only
illustrative of the principles of the invention, and that various
modifications can be made by those skilled in the art without
departing from the scope and spirit of the invention. For example,
the invention may be readily used in wireless data communication
systems using any of the variety of available or evolving wireless
data communication protocols.
* * * * *