 |
Description  |
|
|
REFERENCE TO RELATED PATENTS
This application is related to and incorporates by reference the subject
matter of U.S. Pat. Nos. 4,710,642, 5,164,790, and 5,241,369.
BACKGROUND AND SUMMARY OF THE INVENTION
This invention relates generally to scatterometers and more particularly to
a lens scatterometer system that provides for illumination of a sample at
different angles of incidence without the necessity of rotating, tilting
or otherwise moving the sample during the course of a scatterometer
measurement.
Scatterometer arrangements, like those described in the prior art patents
cited above, have been used for characterizing the microstructure of
microelectonic and optoelectronic semiconductor materials, computer hard
disks, optical disks, finely polished optical components, and other
materials having lateral dimensions in the range of tens of microns to
less than one micron.
Exemplary of the prior art are two publications. The first is by Michael R.
Murnane, et. al., "Developed Photoresist Metrology Using Scatterometry",
Proceedings of the SPIE, Integrated Circuit Metrology, Inspection, and
Process Control VIII, Vol 2196, pp 47-59 (1994); the second is by Michael
R. Murnane, et. al., "Scatterometry for 0.24 .mu.m-0.70 .mu.m Developed
Photoresist Metrology", Proceedings of the SPIE, Integrated Circuit
Metrology, Inspection, and Process Control IX, Vol 2439, pp 427-436
(1995). This referenced prior art extends the capability of the
scatterometer measurements to enable characterization of structure having
lateral dimensions that are sub-tenth-micron. The prior art scatterometer
arrangement discussed in the literature is disadvantageous in that it
requires rotation of the sample while performing a scatterometer
measurement. This requirement precludes their use in applications in which
the sample must remain stationary. In addition, the two rotation stages
employed in this prior art scatterometer represents a mechanical
complexity, which can result in undesirable optical and mechanical
misalignment. Finally, the sample rotation required in this prior art
scatterometer necessitates increased sample handling, thus increasing the
risk of damage to the sample.
It is therefore the principle object of the present invention to provide a
scatterometer system that enables illumination of a sample at various
angles of incidence without rotating or otherwise moving the sample.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a pictorial diagram illustrating a prior art scatterometer system
employing a single detector and two rotation stages to move both the
sample and the detector.
FIG. 2 is a pictorial diagram of a lens in accordance with the present
invention, illustrating use of the lens to provide illumination of a
sample at different angles of incidence and to collect the light that is
diffracted from the sample, in accordance with the present invention.
FIG. 3a is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
rotating block, and light detection system for characterizing the light
that is diffracted from the sample.
FIG. 3b is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
rotating block, and light detection system for characterizing the light
that is diffracted from the sample.
FIG. 3c is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
rotating block, and two light detection systems for characterizing the
light that is diffracted from the sample.
FIG. 4 is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
mirror assembly, and light detection system for characterizing the light
that is diffracted from the sample.
FIG. 5a is a pictorial diagram of a portion of a lens scatterometer system
in accordance with the present invention, illustrating a block that is
rotated about two axes that is used for characterizing light that is
conically diffracted from the sample.
FIG. 5b is a pictorial diagram in accordance with the present invention,
illustrating the geometry involved in illuminating the sample under
separate control of the angle of incidence, .THETA., and .PHI., the angle
between the grating vector and the incident beam.
FIG. 6a is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
fiber optic assembly, and light detection assembly for characterizing
light that is diffracted from the sample.
FIG. 6b is a pictorial diagram of a portion of a lens scatterometer system
in accordance with the present invention, illustrating the use of a two
dimensional fiber optic assembly for characterizing light that is
conically diffracted from the sample.
FIG. 7a is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
laser array, and light detection assembly for characterizing light that is
diffracted from the sample.
FIG. 7b is a pictorial diagram of a portion of a lens scatterometer system
in accordance with the present invention, illustrating the use of a two
dimensional source array for characterizing light that is conically
diffracted from the sample.
FIG. 8a is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
linear beam, and light detection assembly for characterizing light that is
diffracted from a sample.
FIG. 8b is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, two beam
splitters, two linear beams, and light detection assembly for
characterizing light that is conically diffracted from the sample.
FIG. 8c is a pictorial diagram of a portion of a lens scatterometer system
in accordance With the present invention, illustrating the use of a lens,
beam splitter, two linear beams comprising a cross, and light detection
assembly for characterizing light that is conically diffracted from the
sample.
FIG. 9 is a pictorial diagram of a lens scatterometer system in accordance
with the present invention, illustrating the use of a lens, beam splitter,
rotating block, light detection system, and mirror for characterizing the
intensity and phase of light that is diffracted from the sample.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
The present invention may be understood by first referring to the prior art
scatterometer system illustrated in FIG. 1, hereafter referred to as the
2-.THETA. scatterometer arrangement. In this scatterometer arrangement two
rotational stages are incorporated. One stage, called the "sample stage"
is utilized to rotate the sample, and one stage, called the "detector
stage" is utilized to rotate a detector. Typically in this 2-.THETA.
scatterometer arrangement the rotation axes of the two stages are
coincident, although this is not required. The sample is illuminated with
a light beam that is incident on the sample at a point that is also on the
rotation axis of the sample; in other words the front surface of the
sample contains the axis of rotation of this sample stage. In this manner
the angle of incidence of the light illuminating the sample can be made to
vary over a range in a desired manner, and this can be controlled, for
example, by a computer that is connected to the sample stage. Further, as
the angle of incidence is changed by activating the sample stage, the
detector stage is activated to move the detector in a desired manner. The
two stages are activated either simultaneously, or practically
simultaneously.
As explained previously, the 2-.THETA. scatterometer arrangement is
especially useful for characterizing the light scattered and diffracted
from samples which are comprised of structure that is periodic. When
monochromatic, plane wave light is incident upon the periodic structure,
the light is diffracted into orders at angles governed by the simple
grating equation,
sin .THETA.+sin .THETA.'=n.lambda./d.
In this expression, .THETA. is the angle of incidence of the light,
.THETA.' is the angle made by the diffraction order, n is the order
number, .lambda., is the wavelength of the light, and d is the period or
pitch of the structure that is illuminated. This relationship is well
known and discussed in text books on optics.
The 2-.THETA. scatterometer thus monitors the intensity of a single
diffraction order as a function of the angle of incidence of the
illuminating light beam. The intensity variation of the 0-order as well as
higher diffraction orders from the sample can be monitored in this manner,
and this provides information which is useful for determining the
properties of the sample which is illuminated. Because the properties of a
sample are determined by the process used to fabricate the sample, the
information is also useful as an indirect monitor of the process. This
methodology is described in the literature of semiconductor processing.
Note that the light beam used to illuminate the sample might be the output
from a laser or it might be some other appropriate beam of radiation that
can be directed to illuminate the sample. Typically continuous, low power
lasers such as He-Ne, Ar-ion, He-Cd and semiconductor diodes are used for
the source of the light beam, although other sources of radiation might be
used equally well in the scatterometer arrangement described here. The
wavelength of the sources might range from x-ray through the visible and
microwave regions, to the long wavelength region which corresponds to
frequencies of just a few Hz. Generally, larger wavelengths provide for
characterizing samples that have structure of larger dimensions. The
following discussion will use the terminology "beam" or "light beam" to
refer to the radiation that illuminates the sample that is within this
wavelength region. Similarly, it is understood that the different
diffraction orders that result from illuminating the sample with the beam
will also be called "diffracted beams".
A shortcoming of the prior art 2-.THETA. scatterometer arrangement
illustrated in FIG. 1 is that the sample must be rotated in the process of
performing a scatterometer measurement. The angular range over which the
sample is rotated in this prior art configuration is typically 40 degrees
or more, and in some applications of the 2-.THETA. scatterometer
arrangement the sample must be rotated .+-.40 degrees or more (i.e. a
total of 80 degrees or more). Because the axis of rotation of the sample
is parallel to, and included in the surface of the sample, this rotation
precludes application of the prior art 2-.THETA. scatterometer arrangement
in situations in which the sample must necessarily be stationary. This
occurs practically at all steps in processing many materials, including
semiconductor materials, storage media, and the like. For example, in
processing semiconductor wafers in a vacuum environment, in which the
wafer can not be moved existing processing equipment and associated
processing techniques would require extensive modification to accommodate
wafer rotation. Such modifications would be impractical.
Additionally, the two rotation stages utilized in the prior art 2-.THETA.
scatterometer arrangement represent mechanical complexity. Eliminating one
or both of them would represent a significant simplification in
maintaining optical and mechanical alignment.
Another shortcoming of the sample rotation in implementing the prior art
2-.THETA. scatterometer arrangement illustrated in FIG. 1 is that the two
stages involve mechanical motion, and this generates particulate
contamination. Because the sample is located near to the stages,
contamination levels on the sample can increase because of this.
Finally, sample rotation in the prior art configuration of FIG. 1 requires
increased levels of sample handling, which in turn increases the risk of
damaging the sample. The sample must be fixed in a holder that will
sufficiently secure the sample for rotation, and this involves more
handling of the sample compared to an arrangement in which the sample is
stationary. Similarly, increased handling requires more time before the
sample can be examined.
FIG. 2 illustrates how a beam (100) can be directed to different points of
the entrance aperture of a lens (110) and be transmitted through the lens
(110) to illuminate a sample (120) at different angles of incidence. The
angle of incidence depends upon the radial location of the beam in the
entrance aperture of the lens. Only two beams (100) are shown in the FIG.
2 for purposes of illustration; these two beams (100) are labeled "1" and
"2" in FIG. 2. The invention would typically utilize many beams to
illuminate the sample (120). For simplicity of illustration, the beams
(100) are shown to travel in directions that are parallel to the axis of
the lens (110) prior to entering the lens (110). However, the beams are
not required to travel parallel to the axis of the lens (110) prior to
entering the lens (110), nor are the beams required to be parallel to each
other prior to entering the lens. The portion of the sample (120) which is
to be characterized is located in the image plane of the lens at the point
where the beams are imaged to a common point. In the special case in which
the beams travel parallel to each other prior to entering the lens, the
sample (120) is placed in the back focal plane of the lens; the back focal
plane is the same as the image plane in this situation. More specifically,
in the case in which the beams are all parallel to the lens axis, the
sample (120) is located at the back focal point of the lens.
It is understood that in application of the invention discussed herein,
many beams are directed to the entrance aperture of the lens (110) to
subsequently provide illumination of the sample (120) at many different
angles of incidence. In one embodiment of the present invention, the beams
are individually activated in sequence, such that only one beam
illuminates the sample (120) at one specific time and at one specific
angle of incidence. Alternatively, more than one of the beams can be
activated simultaneously, with each beam illuminating the sample (120) at
a corresponding angle of incidence. A third embodiment of the present
invention utilizes a single beam that is translated across the entrance
aperture of the lens (110). This achieves the effect of illuminating the
sample (120) with many beams at many different angles of incidence over a
period of time. A fourth embodiment of the present invention utilizes a
linear beam to illuminate the entrance aperture of the lens (110). This
achieves the effect of illuminating the sample (120) with a large number
of beams at a continuum of angles of incidence. The light detection
configuration, in part, determines the sample illumination arrangement
that is utilized.
Typically the diameter of the beam (100) is much smaller than the aperture
of the lens (110). For example, in one implementation the output of a
He-Ne laser is approximately 1 mm in diameter, and the lens entrance
aperture is in the range of 25 mm to 100 mm. Both the beam diameter and
the lens entrance aperture can be scaled larger or smaller by use of
appropriate optical elements. In this manner, the beam that exits the lens
illuminates the sample at substantially a single angle of incidence.
The lens (110) that is utilized in the present invention substantially
determines the range over which the angle of incidence of the beam can be
varied. Specifically, the f-number (f/#) of the lens will determine the
maximum angle of incidence the beam can have in illuminating the sample.
For example, in the case of beams that enter the lens parallel to the lens
axis, the maximum angle of incidence, .THETA., is given by sin.sup.-1
{1/(2f/#)}. Lenses are commercially available that have an f/# of 0.74 and
an aperture of 50 mm diameter. A beam that enters this lens traveling
parallel to the lens axis and 25 mm from the lens center illuminates the
sample with an angle of incidence of approximately 42.5 degrees. Beams
that pass through the lens at smaller radial positions, closer to the
center of the lens, exit the lens to illuminate the sample at smaller
angles of incidence. The relation between the location of the beam in the
lens entrance aperture and the angle of incidence of the beam at the
sample is determined by the lens design. Similar relations exist in the
case of the lens being cylindrical as opposed to spherical.
The light that illuminates the sample is diffracted by the sample into two
or more beams. There are two sets of diffracted beams: beams that are
transmitted into the sample and beams that are reflected from the sample.
The two so-called 0-order diffracted beams or orders, corresponding to n=0
in the simple grating equation, will always exist, with one transmitted
into the sample and the other reflected from the sample. Higher order
diffraction from the sample, e.g. the n=.+-.1, .+-.2, etc. orders that are
reflected and transmitted may or may not be present; the existence of
these higher orders is governed by the simple grating equation. The
intensity of the diffracted beams is extremely sensitive to the structure
comprising the sample. Specifically, the pitch of the lines comprising the
diffracting structure, as well as their width, height, and sidewall
curvature in the case of the sample being a relief grating, are
contributing factors that determine the diffraction characteristics of the
sample. If the sample is comprised of a phase grating, such as exposed,
but undeveloped photoresist, the pitch and width of the latent image
structure determine the diffraction characteristics. Details of the
diffraction characteristics are described in the literature.
One or more of the diffracted beams which are reflected from the sample
enter the bottom of the lens. The reflected 0-order beams will enter the
lens, as shown in FIG. 2; this is illustrated by beams 1-0 and 2-0 in the
figure for the two beams 1 and 2, respectively. The higher diffraction
orders reflected beams will enter the lens provided their diffraction
angle is within the acceptance angle of the lens. This is illustrated in
FIG. 2 by diffraction order 1-1 shown entering the lens, and diffraction
order 2-1 shown not entering the lens (110). Diffraction order 1-1
corresponds to one of the 1st-order diffracted beams from the incident
beam 1, and diffraction order 2-1 corresponds to one of the 1st-order
diffracted beams from the incident beam 2. The beams which enter the lens
transmit through the lens and exit the lens as illustrated in FIG. 2. For
simplicity, not all of the reflected diffraction orders, and none of the
transmitted diffraction orders are illustrated in FIG. 2.
It is understood that the construction details of the lens (110) of the
invention vary significantly, depending, for example, upon the performance
requirements of the lens (110). For example, the wavelength of the beam
(100) will determine the material properties of the elements which
comprise the lens (110). The lens (110) will typically be comprised of
transparent glass for wavelengths of the beam (100) which are within the
visible region. For wavelengths which are significantly shorter than those
of the visible region, some or all of the elements which comprise the lens
(110) will necessarily be reflecting to the beam (100). It is further
understood that other performance requirements of the lens (110) will
determine details of the construction and characteristics of the lens
(110).
FIGS. 3a and 3b illustrate scatterometer arrangements that utilize a
rotating block (150) to provide a means of translating the beam from the
source (105) to different points of the entrance aperture of the lens
(110), and thus to illuminate the sample (120) at different angles of
incidence, .THETA.. These arrangements comprise lens scatterometer systems
and represent an improvement over the 2-.THETA. scatterometer. The lens
scatterometer arrangements provide a means of changing the angle of
incidence, .THETA., of the beam at the sample (120) without moving the
sample (120).
The lens system scatterometer arrangements illustrated in FIGS. 3a and 3b
are comprised of the lens system and sample arrangement previously
described, together with a detection arrangement (130), a beam splitter
arrangement (140), and a rotating block (150) that rotates about a single
axis. The x-y axes of a coordinate system are illustrated in the figures.
The beam (100) is in the x-y plane as it originates from the source. The
rotating block is transparent at the wavelength of the beam. In the
arrangement the beam from the source (105) propagates through the rotating
block to different points on the beam splitter. At the beam splitter the
beam is partially reflected..A portion of the beam is directed to
different points of the entrance aperture of the lens (110) to illuminate
the sample (120) at different angles of incidence, .THETA..
In FIG. 3a the portion of the beam (100) that is reflected from the beam
splitter is directed toward the lens; the portion of the beam (100) that
is transmitted by the beam splitter is called the beam portion (300). In
general, the block (150) rotates about an axis that is not necessarily
parallel to the beam propagation direction. The specific arrangement of
the invention illustrated in FIG. 3a has the block rotation axis being
perpendicular to the x-y plane. In addition, the beam splitter (140) is
perpendicular to the x-y plane. The faces of the block (150) at which the
beam enters and exits are parallel, and they are also both parallel to the
axis of rotation. Additionally, these faces of the block as well as the
surfaces of the beam splitter (140) are perpendicular to the x-y plane.
Thus the beam (100) remains in the x-y plane after transmission through
the block (150) and after reflection from the beam splitter. The beam is
offset after transmission through the block, shown as OS in FIG. 3a, and
the amount of offset is dependent upon the rotation angle, .DELTA., of the
block about its axis of rotation. This relation is easily calculated and
is described in optics text books. In the specific arrangement illustrated
in FIG. 3a, the block rotation causes the beam to be offset from, and
parallel to, the beam prior to transmitting through the block, thus
remaining in the x-y plane. In addition, the axis of the lens (110) is in
the x-y plane in the arrangement illustrated in FIG. 3a. Thus the beam is
translated to different points along a line in the entrance aperture of
the lens, and the beam locations at different angles of incidence,
.THETA., at the sample (120) location also lie in the x-y plane. In the
arrangement illustrated in FIG. 3a, the beam portion (300) is directed to
a beam dump or other device and is not utilized.
It is understood that more generally the rotation axis of the block (150),
the axis of the lens (110), and the surfaces of the beam splitter (140)
are not required to be perpendicular or parallel to the x-y plane as
described above, in which case the beam locations at different angles of
incidence .THETA. define a surface that is not necessarily located in the
x-y plane. It is also understood that the block (150) can have a shape
that is not exactly as illustrated in FIG. 3a. For example, the block
(150) can be rectangular in two of its dimensions, as opposed to the
square shape illustrated in FIG. 3a. More generally, the block (150) can
be comprised of a shape that has a total number of faces different than
four, as illustrated in FIG. 3a.
The reflected diffraction orders pass through the lens (110) as previously
described to the beam splitter (140); for simplicity, only the reflected
0-order is illustrated in FIG. 3a. At the beam splitter the reflected
diffraction orders are partially transmitted by the beam splitter (140) to
the detection system (130) where their intensities are measured.
Measurements of the diffraction order intensifies are made for each of a
number of values of .DELTA. and corresponding beam (100) angles of
incidence .THETA.. The primary purpose of the invention illustrated in
FIG. 3a, namely to provide a means of illuminating the sample at different
angles of incidence and measuring the intensities of the reflected
diffraction orders, without requiring the sample to be moved is thereby
accomplished.
FIG. 3b illustrates essentially the same invention as illustrated in FIG.
3a. In the arrangement of FIG. 3b, the portion of the beam (100) from the
source that is transmitted by the beam splitter (140) is directed toward
the lens (110) to illuminate the sample (120); the portion of the beam
(100) that is reflected by the beam splitter is called the beam portion
(300). The reflected diffraction orders which pass through the lens to the
beam splitter are partially reflected to the detection system (130). In
the arrangement illustrated in FIG. 3b, the beam portion (300) is directed
to a beam dump or other device and is not utilized. Otherwise the
invention is essentially the same as that illustrated in FIG. 3a, with the
primary purpose of providing a means of illuminating the sample at
different angles of incidence .THETA. and measuring the intensities of the
reflected diffraction orders without moving the sample.
FIG. 3c illustrates how an additional detection system (135) can be
utilized with the invention of either FIG. 3a or 3b. The additional
detection system is shown used in the arrangement of FIG. 3a for
illustrative purposes. The additional detection system (135) is used on
the side of the sample (120) opposite to the side which is illuminated;
i.e., below the sample (120). In this manner the arrangement can
characterize the intensities of diffraction orders that are transmitted by
the sample (120). This measurement can be performed over a range of
incident angles of the beam without requiring the sample (120) to be
moved. Note that the additional detection system (135) below the sample
(120) can be utilized independently of the detection system (130) located
above the sample (120) that measures the intensities of the reflected
diffraction orders; the scatterometer arrangement can be configured to
have either or both of the detection systems. The detection system (135)
below the sample (120) is not required to be identical to the detection
system (130) above the sample (120). For example, the detection system
(135) below the sample (120) might include lower quality optical elements,
such as lenses of lessor optical quality than the lens (110).
The light detection systems (130) and (135) of FIGS. 3a, 3b, and 3c contain
a detector device. The detector device can be comprised of a simple,
single element such as a Si photodiode, a photomultiplier, or other
element appropriate for detecting the wavelength and intensifies of the
reflected or transmitted diffracted beams. A single element detector
provides an integrated measurement of all the diffraction order
intensities. Alternatively, the detector device can be comprised of a
one-dimensional or two-dimensional detector array, such as a ccd array, a
photodiode array, or other one-dimensional or two-dimensional detector
array appropriate for the wavelength and intensities of the diffracted
beams. Use of a detector array provides spatially resolved intensity
measurements of the individual diffraction orders and thus provides
additional information compared to that obtained in integrated
measurement. Similarly, the detector device can be a videcon, nuvecon, or
other similar detection element that provides spatially resolved intensity
measurements of the diffraction orders. The detection systems (130) and
(135) might also contain additional elements, such as lenses. In some
situations the detection systems (130) and (135) might be considered to be
a camera that utilizes either of the detector devices previously
mentioned.
FIG. 4 illustrates how the lens system scatterometer invention described in
FIG. 3a can utilize a set of mirrors (160) in place of the rotating block
(150) to direct the beam to different points of the entrance aperture of
the lens (110), and thus to illuminate the sample (120) at different
angles of incidence. In FIG. 4 the set of mirrors (160) is comprised of
two or more mirrors. In the arrangement, one or more mirrors of the set of
mirrors (160) translates in a manner to cause the beam at the beam
splitter (140) to be offset from the same beam prior to encountering the
set of mirrors. The amount of beam offset is dependent upon the amount of
mirror translation. This, in turn, causes the beam to pass through the
lens entrance aperture at different locations, and thus to illuminate the
sample (120) at different angles of incidence, similar to the description
of the invention of FIG. 3a, b, and c. FIG. 4 illustrates just one manner
of achieving this beam offset. In the arrangement illustrated in FIG. 4,
this is achieved by translating mirror 1 and keeping mirror 2 fixed in
position. In this configuration, the beam (100) is reflected from mirror 1
to different points on mirror 2, reflected from mirror 2 to different
points on the beam splitter (140), and it subsequently passes through the
lens (110) at different aperture locations, to thereby illuminate the
sample (120) at different angles of incidence. The angle of incidence,
.THETA., depends upon the position of mirror 1. The same effect can be
achieved by translating mirror 2 and keeping mirror 1 fixed in position.
In this manner the set of mirrors provides a similar function as the
rotating block (150) in the invention illustrated in FIGS. 3a, b, and c.
Otherwise the inventions of FIGS. 3a and 4 are essentially the same. A set
of mirrors can be similarly utilized in place of the rotating block of the
inventions described in FIGS. 3b and 3c. It is understood that more
generally some or all the mirrors of the set of mirrors (160) can be
non-planar, and that some or all can be made to rotate. Other manners of
translating or rotating mirrors which comprise a set of mirrors (160) can
be envisioned to provide a means of translating the beam to different
points in the entrance aperture of the lens (110), and thus to provide
sample illumination at different angles of incidence .THETA. without
requiring the sample (120) to be moved.
Other means can be envisioned of directing the beam to different points of
the entrance aperture of the lens (110). The two methods previously
discussed, which involve the rotating block (150) and the set of mirrors
(160) are but two means of achieving this.
FIG. 5a illustrates a modified version of the rotating block (150) of the
inventions illustrated in FIGS. 3a, 3b, and 3c. The rotating block (170)
is mounted in a manner that provides rotation about two axes, shown as
axis 1 and axis 2 in FIG. 5a. A typical application would include the two
axes being orthogonal. For example, mounting the rotating block (170) in a
gimbals arrangement would provide such orthogonal axes of rotation. Under
this biaxial rotation, the beam (100) that transmits through the block is
offset in two directions relative to the beam (100) which enters the
block. This is illustrated in FIG. 5a by the two beam offsets labeled OS1
and OS2, corresponding to offsets in the x-y plane and x-z plane,
respectively. The magnitudes of OS1 and OS2 are dependent upon the
respective rotation angles .DELTA.1 and .DELTA.2 of the rotating block
(170). The rotating block (170) replaces the block (150) of the
scatterometer configurations described previously in connection with FIGS.
3a, 3b, and 3c; otherwise these scatterometer configurations operate in
essentially the same manner as previously described.
FIG. 5b illustrates how the biaxial rotation of the block (170) causes the
beam to be directed to different points in the plane of the entrance
aperture of the lens (110). The plane of the entrance aperture of the lens
(110) is parallel to the plane defined by the y'-z' axes, and the lens
axis is in the x'-y' plane shown in FIG. 5b. In turn, the beam that
illuminates the sample lies within a cone that is determined by the f/# of
the lens (110). The incident beam makes an angle .THETA. with the normal
to the sample and angle .PHI. with the grating vector, k (175), as
illustrated in FIG. 5b. The grating vector, k (175), is in the plane of
the sample (120) and in the direction normal to the lines of one of the
sets of periodic structure comprising the sample (120). This is the
two-dimensional extension of the one-dimensional illumination arrangements
discussed in connection with FIGS. 3a, 3b, and 3c. More specifically, the
nonzero value of .DELTA.2 causes .PHI. to be nonzero. Diffraction in the
general case of .PHI. being nonzero is called "conical diffraction".
Measurements of the diffraction order intensities are made for one or more
combinations of .DELTA.1 and .DELTA.2 to provide diffraction data over a
range of values of .THETA. and .PHI.. Note that the x-y axes and the x'-y'
axes lie in the same plane. The arrangements illustrated in FIGS. 5a and
5b thus provide a means to investigate the conical diffraction
characteristics of a sample, with precise and independent control of
.THETA. and .PHI., that does not require moving the sample.
FIG. 6a illustrates a lens scatterometer arrangement which utilizes one or
more fiber optic elements (180) to comprise an array that provides beams
of light. The array of fiber optic elements (180) is appropriately
configured to provide beams (100) that are directed to different points of
the entrance aperture of the lens (110), such that the sample (120) is
illuminated at one or more desired angles of incidence, .THETA.. For
example, the lens scatterometer arrangement of FIG. 6a is similar to that
of FIG. 3a, except that nine fiber optic elements are arranged in a linear
array situated along the y-axis in place of the block (150). This will in
turn provide illumination of the sample (120) at nine different angles of
incidence, .THETA., and the beams which illuminate the sample (120) also
lie in the x-y plane, consistent with the discussion related to the lens
scatterometer arrangement of FIG. 3a. FIG. 6b illustrates how the fiber
optic elements (180) can be arranged in a two-dimensional array that is
utilized in the lens scatterometer arrangement; for convenience the other
elements of the lens scatterometer arrangement are not included in this
FIG. 6b. In the figure, the fiber optic elements (180) are arranged in two
lines contained in the y-z plane, with an included angle of .alpha.. This
arran | | |
