Patent Grant
10290498
This application is based upon and claims the benefit of priority from Japanese Patent Application No. 2017-176970, filed on Sep. 14, 2017; the entire contents of which are incorporated herein by reference.
Embodiments described herein relate generally to an imprint apparatus and an imprint method.
An imprint method has been proposed as a method for forming fine patterns. In the imprint method, a resist is applied onto a processing object matter. Then, the resist is pressed by a template provided with fine patterns, and recessed portions of the template are thereby filled with the resist. Then, the resist is irradiated with ultraviolet rays, and is thereby cured. The resist separated from the template is used as a mask for processing the processing object matter.
In an imprint process, a positioning process between the template and the processing object matter is performed. This positioning process is performed by using alignment marks provided on respective ones of the template and the processing object matter. The alignment marks have predetermined shapes and are arranged in Kerf regions, and thus the arrangement flexibility of the alignment marks is low.
In general, according to one embodiment, an imprint apparatus includes a template holder, a processing object holder, a monitor, and a first moving part. The template holder holds a template that includes a first alignment mark detecting displacement in a first direction. The processing object holder holds a processing object that includes a second alignment mark detecting displacement in the first direction. The monitor optically monitors a state where the first alignment mark and the second alignment mark are overlaid with each other. The first moving part moves at least one of the template holder and the processing object holder in the first direction, on a basis of a monitoring result obtained by the monitor. The first alignment mark includes a first template-side mark and a second template-side mark. The first template-side mark includes a first pattern in which a plurality of first portions are arranged with a first period in the first direction. The second template-side mark includes a second pattern in which a plurality of second portions are arranged with a second period in the first direction. The second alignment mark includes a first wafer-side mark and a second wafer-side mark. The first wafer-side mark includes a third pattern in which a plurality of third portions are arranged with a third period in the first direction. The second wafer-side mark includes a fourth pattern in which a plurality of fourth portions are arranged with a fourth period in the first direction. The first wafer-side mark and the first template-side mark are configured to be overlaid with each other to constitute a first moire mark. The second wafer-side mark and the second template-side mark are configured to be overlaid with each other to constitute a second moire mark. An average period of the first moire mark and an average period of the second moire mark are different from each other.
Exemplary embodiments of an imprint apparatus and an imprint method will be explained below in detail with reference to the accompanying drawings. The present invention is not limited to the following embodiments.
In the following description, an explanation will be first given of the size and arranging method of alignment marks according to a comparative example, and then an explanation will be given of an imprint apparatus, an imprint method, and a semiconductor device manufacturing method, which use alignment marks according to an embodiment.
The mesa part 211 includes a device formation pattern arrangement region RD, in which a device formation pattern for forming a device pattern on the wafer is arranged, and a mark arrangement region RM, in which a mark or the like to be used during the imprint process is arranged. The mark arrangement region RM is a frame-like region arranged at the peripheral side of the rectangular mesa part 211, for example. The device formation pattern arrangement region RD is a region of the mesa part 211 other than the mark arrangement region RM. For example, the device formation pattern includes a line-and-space pattern or the like, in which recessed patterns that extend are arranged at predetermined intervals in a direction intersecting with the extending direction.
The mark arrangement region RM is provided with an alignment mark or the like for performing positioning between the template 200 and the wafer.
Each of the alignment marks provided on the template 200 and the wafer 100 includes, for example, a diffraction grating pattern. For example, the diffraction grating pattern is composed of a so-called line-and-space pattern, in which a plurality of extending line patterns are arranged in parallel with each other and at predetermined intervals in a direction intersecting with the extending direction. Here, two directions orthogonal to each other provided on each of the template substrate 210 and the wafer 100 will be referred to as “X-direction” and “Y-direction”. In order to detect the X-direction component and Y-direction component of a positional deviation between the template 200 and the wafer 100, the alignment marks include a diffraction grating pattern extending in the X-direction and a diffraction grating pattern extending in the Y-direction. Each of the alignment marks may include both of a diffraction grating pattern extending in the X-direction and a diffraction grating pattern extending in the Y-direction, or may include only a diffraction grating pattern extending in either one of the X-direction and the Y-direction.
Next, an explanation will be given of positioning performed by using the alignment mark provided on the wafer 100 and the alignment mark provided on the template 200.
Then, as illustrated in
The moire mark 300 is composed of, for example, a so-called line-and-space pattern, in which line patterns are periodically arrayed in a direction intersecting with their extending direction. The line patterns are patterns provided on, for example, the template 200 or the wafer 100. The direction in which the line patterns are arrayed is a displacement detection direction. The structural period of the alignment mark 110 on the wafer 100 side and the structural period of the alignment mark 230 on the template 200 side are set to be slightly different from each other. With this arrangement, when the alignment mark 110 on the wafer 100 side and the alignment mark 230 on the template 200 side are overlaid with each other, a moire image is generated.
Where the period of the alignment mark 230 on the template 200 side and the period of the alignment mark 110 on the wafer 100 side are respectively denoted by PT and PW, the average period Pave of the two alignment marks 230 and 110 that generate a moire image is expressed by the following formula (1), and the moire period PM is expressed by the following formula (2).
Pave=(PT+PW)/2 (1)
Here, C denotes a coefficient that can change depending on the moire observation method, and the two-dimensional structures of the alignment marks 230 and 110. For example, where two alignment marks, each of which is composed of a one-dimensional pattern, are observed from directly above, C=1 is obtained. Alternatively, where one alignment mark is composed of a one-dimensional pattern, while the other alignment mark is composed of a checkered pattern, which is a two-dimensional pattern, and these alignment marks are observed from directly above, C=1/2 is obtained. This is because, when the checkered pattern is deviated by a half period from the one-dimensional pattern, the checkered pattern looks like the same pattern as the one-dimensional pattern with a different phase.
By observing the moire image, it is possible to perform detection by enlarging a displacement amount by a magnification ratio of PM/Pave, and thereby to obtain positional accuracy exceeding the optical resolution. As described above, the alignment marks 230 and 110 on the template 200 side and wafer 100 side are different in period (pitch) from each other to some extent. If the difference in period is too larger, the magnification ratio becomes smaller, or the number of periodic patterns composing one moire period becomes smaller, and the positional accuracy is thereby lowered. This is because a moire image is premised to be a smooth image substantially the same as a sine wave in theory, but looks blurred discrete patterns in practice; therefore, as the number of periodic patterns composing one moire period is reduced, the period of periodic patterns becomes closer to the moire period, and it becomes difficult to block off a false peak and/or fringe (an optical higher harmonic) by an optical system. For example, in order to enlarge the displacement amount by five times or more, it is necessary to set the ratio between the periods of the alignment marks 230 and 110 on the template 200 side and wafer 100 side to fall within a range of about 1.2 times or less. Accordingly, the periods of the alignment marks 230 and 110 on the template 200 side and wafer 100 side are preferably set to fall within a range of difference equal to or less than about 10% from the average period Pave.
Here, in the moire mark 300, there is clearly a lower limit of size in practical use. If the positional deviation amount Δx remaining from the rough detection stage is not less than about half the average period Pave, the periodic patterns may shift from the original position by a degree in units of just one period, and make it difficult to correctly perform position detection. Accordingly, the moire mark 300 needs to be composed with a structural period twice or more the positional error expected in the rough detection. Thus, the moire mark 300 is composed to satisfy the condition of the following formula (3). However, in a case where one of the alignment marks is a checkered pattern, the moire mark 300 is composed to satisfy the condition of the following formula (4).
Δx<Pave/2 (3)
Δx<Pave/4 (4)
Further, when the initial error is large while the moire mark 300 is small, the moire image shifts significantly, and the peak position of the moire image may go out from the moire mark 300, and make it impossible to detect the displacement. Accordingly, the moire mark 300 needs to have a size that can at least generate one period of the moire image. In practice, however, in consideration of influences given by a displacement amount detecting method for the moire image and by noises derived from stray light, the moire mark 300 may need to have a size for two to three periods of the moire image. Where the necessary moire period is denoted by N, the lower limit L of the size of the moire mark 300 is expressed by the following formula (5). Here, the periodic difference ΔP between the alignment marks 230 and 110 on the template 200 side and wafer 100 side is expressed by |PT−PW|.
The lower limit of the size of the moire mark 300 is defined by this formula (5). In a typical example, where the rough detection error is less than 0.5 μm, the alignment mark 230 on the template 200 side is composed of a pattern of lines, and the alignment mark 110 on the wafer 100 side is composed of a checkered pattern, the average period Pave of the moire mark 300 is preferably set to 2 μm or more. For example, where PT=2.06 μm and PW=1.94 μm are given, one period of the moire image is composed of periodic patterns of about 8.3 periods, and the minimum configuration size becomes 16.7 μm. If three periods of the moire image is required to perform position observation, the lower limit of the size of the moire mark 300 becomes 50 μm, below which the moire mark 300 cannot be formed.
In the direction of the moire mark 300 orthogonal to the displacement detection, no specific restriction is applied thereto, but, in practice, there is a preferable size in consideration of the resolution and/or SN (signal to noise ratio) of an optical system. Further, other than these, in practical use, in order to detect displacement in two-dimensional directions, it is required to provide the moire mark 300 in each of two directions, such as X- and Y-directions. On the other hand, in practice, an alignment mark needs to be contained in a suitable rectangular region, because of a technical request that the alignment mark should be recognizable as an alignment mark by an observation device. With these requirements, the lower limit of the area occupied by the moire mark 300 becomes L.
In practical use, in order to perform detection in one direction, there is a method that combines a single moire mark 300 with a pattern indicating a reference position, and a method that combines two regions designed to cause moire images to move in directions opposite to each other with respect to the displacement. In the latter method, the displacement can be detected only by measuring the relative positional relationship between the two moire images (differential detection) without consideration of the reference position, and the displacement amount can be enlarged by twice. Further, in ordinary situations, it is required to find displacement in directions on two-dimensional plane. Accordingly, in the latter method, in order to detect displacement in the X-direction, and the Y-direction intersecting therewith (or orthogonal thereto), and to set two regions in each direction, a moire mark including four regions are provided.
The structural periods of marks (template-side marks) arranged in respective ones of the XA region RXA,T, the XB region RXB,T, the YA region. RYA,T, and the YB region RYB,T of the template 200 are denoted by PXA,T, PXB,T, PYA,T, and PYB,T, respectively. Further, the structural periods of marks (wafer-side marks) arranged in respective ones of the XA region RXA,W, the XB region RXB,W, the YA region RYA,W, and the YB region RYB,W of the wafer 100 are denoted by PXA,W, PXB,W, PYA,W, and PYB,W, respectively.
Here, Pij,T≠Pij,W (i=X or Y, and j=A or B) holds, and the difference of each of the periods Pij,T and Pij,W from the average period Pij,ave falls within a range of 10% or less, as described above. In consideration of simplicity and symmetry of the design, PiA,T=PiB,W and PiA,W=PiB,T may hold, or PXj,k=PYj,k (k=T or W) may hold. The configuration described above is the basic configuration of the moire mark 300. Here, the average period of the structural periods in this case is denoted by Pave. In other words, in the comparative example, the moire mark 300 includes one combination of the alignment marks 230 and 110 on the template 200 side and wafer 100 side, and their average period is Pave.
As described above, in the comparative example, the area occupied by the moire mark 300 has a lower limit. Hereinafter, an explanation will be given of a moire mark 300 that can reduce the area of the moire mark 300 to be smaller than that of the comparative example while sustaining positioning accuracy at the same level as that of the comparative example. Further, an explanation will be given of an imprint apparatus, an imprint method, and a semiconductor device manufacturing method, which use the moire mark 300.
Again,
The structural periods of the alignment marks 230 arranged in respective ones of the XA region R1XA,T, the XB region R1XB,T, the YA region R1YA,T, and the YB region R1YB,T of the template 200, which have the first structure P1, are denoted by P1XA,T, P1XB,T, P1YA,T, and P1YB,T, respectively. Further, the structural periods of the alignment marks 110 arranged in respective ones of the XA region R1XA,W, the XB region R1XB,W, the YA region R1YA,W, and the YB region R1YB,W of the wafer 100, which have the first structure P1, are denoted by P1XA,W, P1XB,W, P1YA,W, and P1YB,W, respectively.
The structural periods of the alignment marks 230 arranged in respective ones of the XA region R2XA,T, the XB region R2XB,T, the YA region R2YA,T, and the YB region R2YB,T of the template 200, which have the second structure P2, are denoted by P2XA,T, P2XB,T, P2YA,T, and P2YB,T, respectively. Further, the structural periods of the alignment marks 110 arranged in respective ones of the XA region R2XA,W, the XB region R2XB,W, the YA region R2YA,W, and the YB region R2YB,W of the wafer 100, which have the second structure P2, are denoted by P2XA,W, P2XB,W, P2YA,W, and P2YB,W, respectively.
The average periods of the structural periods of the respective moire marks 300 having the first structure P1 and second structure P2 are denoted by P1ave and P2ave, respectively. Further, the relation with the initial error derived from the rough detection is assumed as follows: Where each of the alignment marks is composed of a one-dimensional pattern, the relation is expressed by the following formula (6). On the other hand, where one of the alignment marks is composed of a checkered pattern, the relation is expressed by the following formula (7).
2Δx<P2ave (6)
4Δx<P2ave (7)
Further, the periodic difference between the alignment marks 230 and 110 having the first structure P1 on the template 200 side and wafer 100 side is denoted by ΔP1, and the periodic difference between the alignment marks 230 and 110 having the second structure P2 on the template 200 side and wafer 100 side is denoted by ΔP2. In this case, the lower limits of the size sizes L1 and L2 of the respective moire marks 300 having the first structure P1 and second structure P2 are defined by the following formulas (8) and (9), respectively, on the basis of the formula (5).
Here, consideration will be given of a case to obtain capability equivalent to that of the moire mark according to the comparative example having the average period Pave and the periodic difference ΔP. For example, it is assumed that the first structure P1 has relations of P1ave=Pave/2 and ΔP1=ΔP/2, and the second structure P2 has relations of P2ave=Pave and ΔP2=2ΔP.
In the comparative example, the initial error caused by the rough detection can be absorbed if the relation of Δx<ΔP/2 is satisfied. In this respect, in the first structure P1 of
Further, the sizes L1 and L2 of the moire marks 300P1 and 300P2 having the first structure P1 and second structure P2 satisfy L1=L2=L/2, on the basis of the formulas (8) and (9). Accordingly, the areas of the first structure P1 and second structure P2 satisfy L12=L22=L2/4, and the total area of the moire marks 300P1 and 300P2 having the first structure P1 and second structure P2 come to be expressed by L2/4×2=L2/2. Thus, in the case of 2P1ave=P2ave, the total area is half the area of the moire mark according to the comparative example.
In the example described above, 2P1ave=P2ave holds; however, in order to retain the areal superiority over the comparative example, it is necessary to satisfy the following formula (10).
L12+L22≤L2 (10)
Where P1ave and P2ave satisfy the relation of the following formula (11), the area of the moire mark 300 becomes equal to the area of the comparative example.
√{square root over (2)}P1ave=P2ave (11)
Accordingly, in order to retain the areal superiority over the comparative example, P1ave and P2ave satisfy the relation of the following formula (12).
√{square root over (2)}P1ave≤P2ave (12)
Here, in the first embodiment, it is sufficient if the A region and the B region in each of the first structure P1 and the second structure P2 are arranged adjacent to each other; thus, the other arrangement can be set in any arrangement.
On the other hand, in
Further, in
As described above, the alignment mark 230 on the template 200 side is arranged in the mark arrangement region RM, and the alignment mark 110 on the wafer 100 side is arranged in each Kerf region RK. If a collective alignment mark arrangement area can not be ensured in each of the mark arrangement region RM and Kerf region RK, it may be adopted that marks having the first structure P1 and second structure P2, for detecting displacement in the X-direction, are arranged in a first region in each of the mark arrangement region RM and Kerf region RK, and marks having the first structure P1 and second structure P2, for detecting displacement in the Y-direction, are arranged in a second region other than the first region in each of the mark arrangement region RM and Kerf region RK, for example. In this way, the moire mark 300 according to the first embodiment is higher in arrangement flexibility.
Here, an explanation will be given of an example of a moire image to be formed by the moire mark 300. As illustrated in
Each alignment mark 230 on the template 200 side is composed of a line-and-space pattern, in which one-dimensional line patterns 231 and 232 or 233 and 234 are arranged in parallel with each other. Each alignment mark 110 on the wafer 100 side is composed of a checkered pattern. These alignment marks 230 and 110 are used to detect displacement in the X-direction or Y-direction. Here,
P1XA,T=P1XB,W=P1YA,T=P1YB,W=1,060 nm
P1XA,W=P1XB,T=P1YA,W=P1YB,T=1,000 nm
The second structure P2 is also provided with A regions R2XA,T and R2XA,W and B regions R2XB,T and R2XB,W for performing differential detection. The periods of periodic patterns formed in the respective regions are as follows:
P2XA,T=P2XB,W=P2YA,T=P2YB,W=2,240 nm
P2XA,W=P2XB,T=P2YA,W=P2YB,T=2,000 nm
In the above configuration, the average period P1ave of the first structure P1 is 1,030 nm, and the periodic difference ΔP1 between the alignment marks 230 and 110 on the template 200 side and wafer 100 side is 60 nm. Each of the periods of periodic patterns constituting the first structure P1 falls within a range of 10% or less from the average period P1ave. Further, the average period P2ave of the second structure P2 is 2,120 nm, and the periodic difference ΔP2 between the alignment marks 230 and 110 on the template 200 side and wafer 100 side is 240 nm. Each of the periods of periodic patterns constituting the second structure P2 falls within a range of 10% or less from the average period P2ave.
Here, the vertical direction period of the checkered pattern (the period in the direction orthogonal to the structural period of the alignment mark 230 on the template 200 side) is 4,500 nm. Further, noise cancelling patterns 241a, 241b, 242a, 242b, and 121a are provided around the line patterns 231 and 232 for the first structure P1 and the line patterns 233 and 234 for the second structure P2 in the template 200, and around rectangular patterns 111 and 112 for the first structure P1 in the wafer 100. In this example, it is premised that a dark field optical system is used to perform moire image monitoring; therefore, the noise cancelling patterns 241a, 241b, 242a, 242b, and 121a are provided to suppress scattered light (noise) to be generated at portions where the period structures break off. The shape and arrangement position of each of the noise cancelling patterns 241a, 241b, 242a, 242b, and 121a vary depending on the size and/or structure of the moire mark 300.
For example, as illustrated in
As illustrated in
The overall size of the moire mark 300 described above is 126 μm x 32 μm, which includes the noise cancelling patterns 241a, 241b, 242a, 242b, and 121a. On the other hand, a moire mark according to the scheme of the comparative example and having capability equivalent to that of the moire mark 300 described above comes to be about 120 μm×60 μm. Thus, the moire mark 300 according to the first embodiment has an area about half that of the moire mark according to the scheme of the comparative example and having capability equivalent thereto.
In
Next, an explanation will be given of an imprint apparatus for executing an imprint process that performs positioning by using the template 200 and the wafer 100, which include the moire mark 300 described above.
The wafer 100 includes a substrate, such as a semiconductor substrate, an underlying pattern formed on this substrate, and a processing target layer formed on this underlying pattern. When pattern transfer is performed, the wafer 100 further includes a resist formed on the processing target layer. As the processing target layer, an insulating film, metal film (conductive film), or semiconductor film may be cited.
The substrate stage 11 is provided to be movable on a stage bed 13. The substrate stage 11 is arranged to be movable along respective ones of two axes that extend along the upper surface 13a of the stage bed 13. Here, the two axes that extend along the upper surface 13a of the stage bed 13 will be referred to as “X-axis” and “Y-axis”. The substrate stage 11 is further arranged to be movable in the height direction that will be referred to as “Z-axis”, which is orthogonal to the X-axis and the Y-axis. The substrate stage 11 is preferably arranged to be rotatable about each of the X-axis, the Y-axis, and the Z-axis.
The substrate stage 11 is provided with a reference mark pedestal 14. A reference mark (not illustrated) is disposed at the top of the reference mark pedestal 14, and is used as a reference position for the imprint apparatus 10. For example, the reference mark is composed of a diffraction grating having a checkered pattern. The reference mark is used for performing calibration of alignment scopes 30 and positioning (attitude control and adjustment) of the template 200. The reference mark serves as the original point on the substrate stage 11. The X- and Y-coordinates of the wafer 100 placed on the substrate stage 11 are coordinates using the reference mark pedestal 14 as the original point.
The imprint apparatus 10 includes a template stage 21. The template stage 21 is configured to fix the template 200. The template stage 21 holds the peripheral portion of the template 200 by means of, for example, vacuum suction. The template stage 21 operates to position the template 200 with reference to the apparatus. The template stage 21 is attached to a base part 22.
A correction mechanism 23 and a pressurizing section 24 are mounted on the base part 22. The correction mechanism 23 includes an adjustment mechanism for slightly adjusting the position (attitude) of the template 200 in accordance with an instruction received from, for example, a controller 50. With this adjustment, the relative positions of the template 200 and the wafer 100 therebetween are corrected.
The pressurizing section 24 applies stress to the side surfaces of the template 200 to straighten distortion of the template 200. The pressurizing section 24 applies pressure to the template 200 from the four side surfaces of the template 200 toward the center. With this pressure application, the dimensions of a pattern to be transferred are corrected (magnification correction). The pressurizing section 24 applies pressure to the template 200 by a predetermined stress in accordance with an instruction received from, for example, the controller 50.
The base part 22 is attached to the alignment stage 25. The alignment stage 25 moves the base part 22 in the X-axis direction and the Y-axis direction to perform positioning between the template 200 and the wafer 100. The alignment stage 25 also has a function to rotate the base part 22 along an XY-plane. The rotational direction along the XY-plane will be referred to as “θ-direction”. Here, a template holder includes the template stage 21, and may further include the base part 22, the correction mechanism 23, the pressurizing section 24, and the alignment stage 25 in addition.
Each of the alignment scopes 30 serves as an optical monitoring unit for detecting alignment marks provided on the template 200 and alignment marks provided on the wafer 100. The alignment marks on the wafer 100 and the alignment marks on the template 200 are used to measure relative positional deviation between the template 200 and the wafer 100. Here, the respective alignment scopes 30 are preferably arranged at positions corresponding to the four corners of the mesa part 211 of the template 200, to simultaneously pick up images of the alignment marks arranged at the four corners of the mesa part 211.
The imprint apparatus 10 includes a light source 41 and a coating member 42. The light source 41 emits electromagnetic waves, for example, within the ultraviolet region. The light source 41 is arranged to be right above the template 200, for example. In another case, the light source 41 may be not arranged right above the template 200. In this case, an optical path is set by using an optical component, such as a mirror, so that light emitted from the light source 41 can be radiated from right above the template 200 toward the template 200. The light source 41 turns on or off the light irradiation to the template 200 in accordance with an instruction received from, for example, the controller 50.
The coating member 42 is a member for applying a resist onto the wafer 100. For example, the coating member 42 is formed of an inkjet head including a nozzle, and is configured to drop the resist from the nozzle onto the wafer 100. The resist used in the first embodiment may have a refractive index equivalent to the refractive index of the template 200. It should be noted that the “equivalent to” used here encompasses not only a state completely equal to each other but also a state slightly different from each other. The coating member 42 drops the resist onto a predetermined position on the wafer 100 in accordance with an instruction received from, for example, the controller 50.
The imprint apparatus 10 includes the controller 50. The controller 50 conducts overall control of the imprint apparatus 10. For example, the controller 50 executes a control process for the substrate stage 11, a control process for the light source 41, a positional deviation correcting process, a template height arithmetic process, a magnification correcting process, and so forth, in accordance with programs prescribing the contents of the respective processes.
The control process for the substrate stage 11 is a process of generating a signal for controlling the substrate stage 11 in the X-axis direction, the Y-axis direction, the Z-axis direction, and the θ-direction. With this process, the relative positions of the template 200 and the substrate stage 11 therebetween are controlled. The control process for the light source 41 is a process of controlling the light irradiation timing or irradiation amount used by the light source 41 when the resist is cured.
In the positional deviation correcting process, the alignment marks on the template 200, and the reference mark on the reference mark pedestal 14 or the alignment marks on the wafer 100 are used, to obtain a positional deviation of the template 200 relative to the reference mark, and to obtain a positional deviation of the wafer 100 relative to the template 200. Then, on the basis of these positional deviations, an arithmetic operation for achieving alignment between the template stage 21 and the substrate stage 11 is performed, and the positional deviations are thereby corrected.
In the template height arithmetic process, the alignment marks on the template 200, and the alignment marks on the wafer 100 or the reference mark on the reference mark pedestal 14 are used, to perform an arithmetic operation for calculating the template height at the alignment mark formation position of the template 200.
In the magnification correcting process, a predetermined arithmetic operation is performed on the basis of the template height, to calculate a stress for performing magnification correction to the template 200. Then, a signal for generating this stress is given to the pressurizing section 24.
Next, an explanation will be given of an imprint method including an alignment process between the template 200 and the wafer 100 in the imprint apparatus 10 described above.
First, the wafer 100 is loaded onto the substrate stage 11 of the imprint apparatus 10 (step S11). Then, a resist is dropped from the coating member 42 onto a shot region RS to be processed of the wafer 100 (step S12). Thereafter, rough detection is performed by using rough detection marks on the template 200 side and wafer 100 side (step S13). The rough detection is coarse positioning performed before the template 200 is brought closer to the wafer 100. The positional accuracy of this rough detection is Δx, and positioning error between the template 200 and the wafer 100 is Δx or less.
Thereafter, the template 200 is moved down and brought into contact with the resist on the wafer 100 to apply an impress (step S14). Further, in this impress process to the resist, a positioning process between the template 200 and the wafer 100 is performed by using the moire mark (step S15). In this positioning process, under monitoring by the alignment scopes 30, positioning with middle accuracy is performed by using the marks having the second structure P2 of the moire mark 300, and then positioning with higher accuracy is performed by using the marks having the first structure P1.
Specifically, in a state where the illumination (not illustrated) of the alignment scopes 30 is lit up, the alignment mark of a pattern of lines in the mark arrangement region RM of the template 200 is brought to be overlaid with the alignment mark of a checkered pattern in a Kerf region RK of the wafer 100. At this time, as the period of the alignment mark 110 on the wafer 100 is slightly different from the period of the alignment mark 230 of the template 200, a moire image is generated. The position of brightness bands in this moire reflects the positional deviation of the template 200 relative to the wafer 100 in an enlarged state. Accordingly, when the template 200 moves slightly with respect to the wafer 100, the position of brightness bands in the moire moves significantly. Thus, by utilizing the position of brightness bands in the moire it is possible to precisely adjust the position of the template 200 in the X-direction or Y-direction with respect to the wafer 100. Here, this positioning is performed for each of the X-direction and the Y-direction.
In the positioning utilizing the moire, even if the template 200 is deviated from the wafer 100 by one or more periods of the pattern, the deviation cannot be detected. However, in the rough detection, the positional deviation of the template 200 relative to the wafer 100 is set to be less than one period of the pattern. Thus, in the precise positioning utilizing the moire, there is no need to consider the possibility of the deviation being one or more periods.
Thereafter, the template 200 is kept in a state in contact with the resist for a predetermined time, so that the recessed patterns of the template 200 are filled with the resist (step S16). Then, the resist pattern is irradiated with ultraviolet rays through the template 200 (step S17). Consequently, the resist pattern is cured.
Thereafter, the template 200 is separated from the wafer 100 and the resist pattern (step S18). Then, it is determined whether the imprint process has been performed to all the shot regions RS on the wafer 100 (step S19). When the imprint process has not yet been performed to all the shot regions RS (No at step S19), a next shot region RS is selected (step S20), and the process sequence goes back to step S12. On the other hand, when the imprint process has been performed to all the shot regions RS (Yes at step S19), the imprint method ends.
After the imprint process is performed to all the shot regions RS, a subsequent process, for example, an etching process, such as a Reactive Ion Etching (RIE) method, is performed, on the basis of the resist pattern formed by the imprint process. The processes described above are repeated to manufacture semiconductor devices.
In the above description, a case is illustrated where one moire mark 300 includes alignment marks 230 and 110 for detecting displacement in the X-direction, and alignment marks 230 and 110 for detecting displacement in the Y-direction; however, the embodiment is not limited to this.
The mark arrangement region RM of the template 200 and each Kerf region RK of the wafer 100 may be provided with the marks M1X and M2X including only alignment marks for detecting displacement in the X-direction, or the marks M1Y and M2Y including only alignment marks for detecting displacement in the Y-direction. With this arrangement of moire marks, it is possible to detect distortion of the template 200 from results of positional deviation at respective positions.
In the first embodiment, the moire mark 300 is used in which the alignment mark 230 and the alignment mark 110 are arranged to be overlaid with each other. The alignment mark 230 has a periodic structure and is provided on the template 200. The alignment mark 110 has a periodic structure and is provided on the wafer 10, which is to be placed to face the template 200. The moire mark 300 includes the first structure P1 having an average period P1ave and the second structure P2 having an average period P2ave, which are set to satisfy the formula (12). Further, the moire mark 300 is set such that the relation with the initial error derived from the rough detection is as follows: Where each of the alignment marks is composed of a one-dimensional pattern, one of the alignment marks satisfies the formula (6). Where one of the alignment marks is composed of a checkered pattern, one of the alignment marks satisfies the formula (7). Consequently, it is possible to provide a moire mark 300 smaller in area as compared with the moire mark according to the comparative example, while sustaining positioning accuracy equivalent to that of the moire mark according to the comparative example.
Further, the alignment marks constituting each of the first structure P1 and the second structure P2 do not need to be arranged together in the mark arrangement region RM and each Kerf region RK. The alignment marks constituting the first structure P1 and second structure P2 may be arranged intricate with each other. Thus, the arrangement flexibility of the alignment marks becomes higher as compared with the comparative example. As a result, some of the alignment marks can be arranged dividedly into a dead space in the mark arrangement region RM and each Kerf region RK.
In the first embodiment, as seen in the simulation result of
Each alignment mark 230 on the template 200 side is composed of a line-and-space pattern, in which one-dimensional line patterns 231 and 232 or 233 and 234 are arranged in parallel with each other. Each alignment mark 110 on the wafer 100 side is composed of a checkered pattern, in which rectangular patterns 111 and 112 or 113 and 114 are periodically arranged in a two-dimensional plane. These alignment marks 230 and 110 are used to detect displacement in the X-direction or Y-direction. Here,
P1XA,T=P1XB,W=P1YA,T=P1YB,W=1,030 nm
P1XA,W=P1XB,T=P1YA,W=P1YB,T=1,000 nm
The second structure P2 is also provided with A regions R2XA,T and R2XA, and B regions R2XB,T and R2XB,W for performing differential detection. The periods of periodic patterns formed in the respective regions are as follows:
P2XA,T=P2XB,W=P2YA,T=P2YB,W=2,040 nm
P2XA,W=P2XB,T=P2YA,W=P2YB,T=1,800 nm
In the above configuration, the average period P1ave of the first structure P1 is 1,015 nm, and the periodic difference ΔP1 between the alignment marks 230 and 110 on the template 200 side and wafer 100 side is 30 nm. Each of the periods of periodic patterns constituting the first structure P1 falls within a range of 10% or less from the average period P1ave. Further, the average period P2ave of the second structure P2 is 1,920 nm, and the periodic difference ΔP2 between the alignment marks 230 and 110 on the template 200 side and wafer 100 side is 240 nm. Each of the periods of periodic patterns constituting the second structure P2 falls within a range of 10% or less from the average period P2ave.
Here, the vertical direction period of the checkered pattern (the period in the direction orthogonal to the structural period on the template 200 side) is 4,500 nm. Further, noise cancelling patterns 241a, 241b, 242a, 242b, 121a, 121b, and 122b are provided around the marks having the first structure P1 and the marks having the second structure P2, on the template 200 and wafer 100.
The overall size of the moire mark 300 described above is 158 μm×35 μm, which includes the noise cancelling patterns 241a, 241b, 242a, 242b, 121a, 121b, and 122b, and the rough detection mark MC.
Further, each alignment mark 110 on the wafer 100 side includes a phase inversion section 116 at and near the boundary between the A region R1XA,W or R2XA,W and the B region R1XB,W or R2XB,W. Each alignment mark 110 on the wafer 100 side is composed of a checkered pattern. Specifically, each alignment mark 110 on the wafer 100 side has a configuration in which the rectangular patterns 111 and 112 or 113 and 114 are arranged with predetermined periods in the X-direction and Y-direction. In the second embodiment, the phase inversion section 116 includes respective parts with phases inverted from those of the other regions, on the A region R1XA,W or R2XA,W side and the B region R1XB,W or R2XB,W side relative to the boundary between the A region R1XA,W or R2XA,W and the B region R1XB,W or R2XB,W.
As illustrated in
In the second embodiment, the phase inversion section 116 with inverted phases is provided between the A region R1XA,W or R2XA,W and the B region R1XB,W or R2XB,W, which are arranged to perform differential detection, on the wafer 100 side. Consequently, it is possible to observe the moire pattern of the A region R1XA,W or R2XA,W and the moire pattern of the B region R1XB,W or R2XB,W in a separated state when the moire mark 300 is monitored.
In the above description, a case is illustrated where each alignment mark is provided with the A region and the B region, i.e., two regions to perform differential detection. However, where a reference position is used, the two regions are not necessarily required to be provided.
A resist pattern transferred by an imprint method is used to perform a working process, such as etching or Chemical Mechanical Polishing (CMP), to a processing object. At this time, an alignment mark on a template has also been transferred onto the resist pattern, and the alignment mark is subjected to processing. Where a working process, such as etching or CMP, is performed in a state with the alignment mark transferred on the wafer side, stepped portions are formed on the processing object. If stepped portions are present on the processing object, a problem arises that positioning accuracy is lowered and/or pattern transfer becomes difficult in a subsequent imprint process. In the following embodiment made in consideration of the above problem, an explanation will be given of an imprint apparatus, an imprint method, and a semiconductor device manufacturing method, which can suppress generation of stepped portions due to an alignment mark.
Further, as illustrated in
An imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
In the third embodiment, each of the patterns constituting the alignment mark 230 on the template 200 side is composed of a plurality of first components 251 separated in the pattern width direction. Further, each of the patterns constituting the alignment mark 110 on the wafer 100 side is composed of a plurality of second components 151 separated in the pattern width direction. Consequently, when a process, such as CMP or etching, is performed in a state where an alignment mark has been transferred onto the wafer 100, as the pattern size in each Kerf region RK of the wafer 100 is almost the same as the pattern size in each pattern region RP, the polishing or etching are developed equivalently in the two regions. As a result, it is possible to suppress generation of stepped portions on the wafer 100.
In the above description, each of the line patterns 235 constituting the alignment mark 230 on the template 200 side is divided into a plurality of first components 251, and each of the rectangular patterns 115 constituting the alignment mark 110 on the wafer 100 side is divided into a plurality of second components 151. However, even if either one of the alignment mark 230 on the template 200 side and the alignment mark 110 on the wafer 100 side is composed of patterns each divided into a plurality of components, substantially the same effect can be obtained.
As illustrated in
As described above, by setting the long side of the first components 251 of the alignment mark 230 on the template 200 side to intersect with the long side of the second components 151 of the alignment mark 110 on the wafer 100 side, it is possible to make the signal waveform into a smoother shape.
An imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Further,
In the fourth embodiment, the first components 251 of the alignment mark 230 on the template 200 side and the second components 151 of the alignment mark 110 on the wafer 100 side are configured such that the long side direction of the first components 251 intersects with the long side direction of the second components 151. Further, the pitch of the first components 251 is set equal to the pitch of the second components 151. Consequently, it is possible to suppress generation of signal deformations in the waveform indicating signal intensity, which is obtained by monitoring the moire mark by using a dark field optical system. As a result, it is possible to perform the positioning without deteriorating the alignment accuracy, in addition to the effect of the third embodiment.
Further, also with a configuration in which the extending direction of the long side of the second components 151 intersects with the extending direction of the long side of the first components 251 by an angle other than 90°, and the pitch of the second components 151 is set different from the pitch of the first components 251, it is possible to obtain signal intensity substantially the same as that illustrated in
An imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Also in the fifth embodiment, an effect substantially the same as that of the fourth embodiment can be obtained.
Here, it is assumed that the second direction width of each line pattern 235 of the alignment mark 230 on the template 200 side is denoted by “a”, and the second direction width of each rectangular pattern 115 of the alignment mark 110 on the wafer 100 side is denoted by “b”. The example illustrated in
a>b (13)
a=b (14)
Here, an imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
In the sixth embodiment, the second direction width “a” of each line pattern 235 of the alignment mark 230 on the template 200 side is set equal to the second direction width “b” of each rectangular pattern 115 of the alignment mark 110 on the wafer 100 side. Consequently, it is possible to suppress generation of signal deformations, which are to be generated when the moire mark is monitored by a dark field optical system, and thereby to perform the positioning with high accuracy.
The third to sixth embodiments have taken as an example a case where one of the template-side alignment mark and the wafer-side alignment mark is divided into portions in the form of lines. The seventh and subsequent embodiments will take as an example a case where one of the template-side alignment mark and the wafer-side alignment mark is divided into portions in the form of contact holes.
An imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Also in the seventh embodiment, an effect substantially the same as that of the third embodiment can be obtained.
Alternatively, the configuration of each rectangular pattern 115 is defined as follows: Where one column of second components 152 arrayed in the first direction is referred to as “component column” 154, the second components 152 are arrayed such that the extending direction of the component columns 154 intersects with the extending direction of the first components 251 of the alignment mark 230 on the template 200 side. Here, the arrangement period of the first components 251 and the arrangement period of the second components 152 may be set the same as each other or different from each other.
Here, the moire mark illustrated in
Here, the arrangement period of the first components 251 and the arrangement period of the second components 152 may be set the same as each other or different from each other.
Here, an imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Also in the eighth embodiment, an effect substantially the same as that of the fourth embodiment can be obtained.
Alternatively, the configuration of each rectangular pattern 115 is defined as follows: Where one column of second components 152 arrayed in the first direction is referred to as “component column” 154, the component columns 154 extend in a zigzag state in the first direction, and are arranged in parallel with each other in the second direction. In this way, in the ninth embodiment, the component columns 154 of the second components 152 of the alignment mark 110 on the wafer 100 side are arrange not to be in parallel with the extending direction of the first components 251 of the alignment mark 230 on the template 200 side.
Here, an imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Also in the ninth embodiment, an effect substantially the same as that of the fourth embodiment can be obtained.
Here, an imprint method and a semiconductor device manufacturing method including positioning between the template 200 and the wafer 100 performed by using the moire mark described above are substantially the same as those described in the first embodiment.
Also in the tenth embodiment, an effect substantially the same as that of the fourth embodiment can be obtained.
In each of the third to tenth embodiments described above, the alignment mark 230 on the template 200 side and the alignment mark 110 on the wafer 100 side may be exchanged for each other.
Further, other than an imprint method, each of the moire marks described above may be applied to a transfer method, such as contact exposure or proximity exposure, in which patterning is performed by setting a transfer pattern (such as a template or mask) in contact with a transfer destination (such as a wafer or substrate) or in a similar state.
(Note)
[Note 1]
An imprint apparatus comprising:
a template holder that holds a template that includes a first alignment mark detecting displacement in a first direction;
a processing object holder that holds a processing object that includes a second alignment mark detecting displacement in the first direction;
a monitor that optically monitors a state where the first alignment mark and the second alignment mark are overlaid with each other; and
a first moving part that moves at least one of the template holder and the processing object holder in the first direction, on the basis of a monitoring result obtained by the monitor, wherein
the first alignment mark includes a plurality of first marks arranged with a first period in the first direction,
the second alignment mark includes a plurality of second marks arranged with a second period in the first direction,
the first alignment mark and the second alignment mark are configured to be overlaid with each other to constitute a moire mark, and
either one of each of the first marks and each of the second marks is composed of a plurality of components.
[Note 2]
The imprint apparatus according to Note 1, wherein
each of the first marks is composed of a plurality of first components, and
each of the second marks is composed of a plurality of second components.
[Note 3]
The imprint apparatus according to Note 2, wherein a long side direction of the first components differs from a long side direction of the second components.
[Note 4]
The imprint apparatus according to Note 2, wherein each of the first marks and each of the second marks are composed of a pattern having a line width smaller than a line width of a main body pattern that includes a device and a wiring line to be transferred to the processing object.
[Note 5]
The imprint apparatus according to Note 2, wherein the first period differs from the second period.
[Note 6]
The imprint apparatus according to Note 2, wherein
each of the first marks has a configuration in which the first components are periodically arranged with a third period, and
each of the second marks has a configuration in which the second components are arranged with the third period.
[Note 7]
The imprint apparatus according to Note 2, wherein
each of the first marks has a configuration in which the first components are periodically arranged with a third period, and
each of the second marks has a configuration in which the second components are arranged with a fourth period different from the third period.
[Note 8]
The imprint apparatus according to Note 2, wherein a width of each of the first marks in the first direction is equal to a width of each of the second marks in the first direction.
[Note 9]
The imprint apparatus according to Note 1, further comprising a second moving part that moves at least one of the template holder and the processing object holder in a second direction orthogonal to the first direction, on the basis of a monitoring result obtained by the monitor, wherein
the template includes a third alignment mark detecting displacement in the second direction,
the processing object includes a fourth alignment mark detecting displacement in the second direction, and
the third alignment mark and the fourth alignment mark are marks obtained by rotating the first alignment mark and the second alignment mark, respectively, by 90° in a plane defined by the first direction and the second direction.
[Note 10]
The imprint apparatus according to Note 2, wherein the first components and the second components are linear patterns.
[Note 11]
The imprint apparatus according to Note 2, wherein
each of the first marks has a configuration in which the first components, which are a plurality of linear patterns, are arranged in parallel with each other, and
each of the second marks has a configuration in which the second components, which are a plurality of contact hole-like patterns, are arranged in a two-dimensional state.
[Note 12]
The imprint apparatus according to Note 11, wherein
the first components are the linear patterns extending in a second direction orthogonal to the first direction, and
each of the second marks has a configuration in which component rows of the second components arrayed in the first direction are shifted from each other in the first direction depending on positions in the second direction.
[Note 13]
The imprint apparatus according to Note 11, wherein
the first components are the linear patterns extending in the first direction, and
each of the second marks has a configuration in which component rows of the second components arrayed in the first direction are shifted from each other in the first direction depending on positions in a second direction orthogonal to the first direction.
[Note 14]
The imprint apparatus according to Note 11, wherein
the first components are the linear patterns extending in a second direction orthogonal to the first direction, and
the second components have a shape in which a length in the first direction is larger than a length in the second direction.
[Note 15]
The imprint apparatus according to Note 11, wherein each of the first marks and each of the second marks are composed of a pattern having a line width smaller than a line width of a main body pattern that includes a device and a wiring line to be transferred to the processing object.
[Note 16]
The imprint apparatus according to Note 11, wherein
each of the first marks has a configuration in which the first components, which are a plurality of contact hole-like patterns, are arranged in a two-dimensional state, and
each of the second marks has a configuration in which the second components, which are a plurality of linear patterns, are arranged in parallel with each other.
[Note 17]
The imprint apparatus according to Note 11, wherein
the first alignment mark includes a line-and-space pattern in which a plurality line patterns extending in a second direction orthogonal to the first direction are arranged in parallel with each other, and
the second alignment mark includes a checkered pattern in which rectangular patterns are arranged in a two-dimensional state in the first direction and the second direction.
[Note 18]
The imprint apparatus according to Note 11, wherein
the first alignment mark includes a checkered pattern in which rectangular patterns are arranged in a two-dimensional state in the first direction and a second direction orthogonal to the second direction, and
the second alignment mark includes a line-and-space pattern in which a plurality line patterns extending in the second direction are arranged in parallel with each other.
[Note 19]
An imprint method comprising:
arranging a template and a processing object to face each other, the template including a first alignment mark detecting displacement in a first direction, the processing object including a second alignment mark detecting displacement in the first direction, to face each other;
applying a resist onto the processing object;
bringing the template into contact with the resist;
optically monitoring a state where the first alignment mark and the second alignment mark are overlaid with each other, under a state where the template is set in contact with the resist; and
performing positioning by moving at least one of the template and the processing object in the first direction, on the basis of a monitoring result, wherein
the first alignment mark includes a plurality of first marks arranged with a first period in the first direction,
the second alignment mark includes a plurality of second marks arranged with a second period in the first direction,
the first alignment mark and the second alignment mark are configured to be overlaid with each other to constitute a moire mark, and
either one of each of the first marks and each of the second marks is composed of a plurality of components.
[Note 20]
A semiconductor device manufacturing method comprising:
arranging a template and a processing object to face each other, the template including a first alignment mark detecting displacement in a first direction, the processing object including a second alignment mark detecting displacement in the first direction, to face each other;
applying a resist onto the processing object;
bringing the template into contact with the resist;
optically monitoring a state where the first alignment mark and the second alignment mark are overlaid with each other, under a state where the template is set in contact with the resist;
performing positioning by moving at least one of the template and the processing object in the first direction, on the basis of a monitoring result,
curing the resist, after recessed patterns of the template are filled with the resist;
separating the template from the resist; and
processing the processing object by using the resist thus cured, wherein
the first alignment mark includes a plurality of first marks arranged with a first period in the first direction,
the second alignment mark includes a plurality of second marks arranged with a second period in the first direction,
the first alignment mark and the second alignment mark are configured to be overlaid with each other to constitute a moire mark, and
either one of each of the first marks and each of the second marks is composed of a plurality of components.
While certain embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. Indeed, the novel embodiments described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions and changes in the form of the embodiments described herein may be made without departing from the spirit of the inventions. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the inventions.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2017-176970 | Sep 2017 | JP | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 5702567 | Mitsui et al. | Dec 1997 | A |
| 5917205 | Mitsui et al. | Jun 1999 | A |
| 8842294 | Minoda et al. | Sep 2014 | B2 |
| 9595447 | Inada et al. | Mar 2017 | B2 |
| 20020180067 | Hoshi | Dec 2002 | A1 |
| 20080094629 | Wu | Apr 2008 | A1 |
| 20120244319 | Wuister | Sep 2012 | A1 |
| 20160297117 | Sato | Oct 2016 | A1 |
| Number | Date | Country |
|---|---|---|
| 09-102457 | Apr 1997 | JP |
| 2002-64055 | Feb 2002 | JP |
| 2002-359171 | Dec 2002 | JP |
| 4848832 | Dec 2011 | JP |
| 5658271 | Dec 2014 | JP |
| 2015-90421 | May 2015 | JP |
| 5713961 | May 2015 | JP |
| 2015-138963 | Jul 2015 | JP |
| 5932859 | Jun 2016 | JP |
| 5943717 | Jul 2016 | JP |
| Number | Date | Country | |
|---|---|---|---|
| 20190080899 A1 | Mar 2019 | US |
