JP2017103363A - Dimension correction method nanoimprint mold and method of manufacturing nanoimprint mold - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a nanoimprint mold in which a pattern dimension on the nanoimprint mold and transferred substrate is corrected in high accuracy, a dimension error occurred in a transferring step is calculated, and which has an excellent dimension accuracy and a high quality.SOLUTION: In a signal profile 40 of a line pattern of a nanoimprint mold, wave forms of an external side of both side edges are connected, and a virtual beam profile 41 is extracted. The beam profile 41 is differentiated, and a differentiated profile 42 is formed. A differentiated peak distance Wb of the differentiated profile is calculated as a correction width of a dimension.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a method for correcting dimensions of a nanoimprint mold used for manufacturing a semiconductor element, an optical element, and the like, and a method for manufacturing a nanoimprint mold.

  In recent years, with the miniaturization of semiconductor device patterns, high-resolution techniques such as shortening the wavelength of an exposure light source and the phase shift method have been used. Also, in order to cope with further miniaturization and higher integration, development of a direct drawing technique on a wafer with an electron beam and an exposure technique with extreme ultraviolet (EUV) light (wavelength: 13.5 nm) are in progress. However, these technologies have significant barriers due to the difficulty of solving the problems until practical use and the cost of an expensive exposure apparatus. On the other hand, in recent years, nanoimprint lithography (NIL) that can form patterns at low cost and high throughput has attracted attention.

  Nanoimprint lithography is expected as a technique capable of forming a fine pattern having a high resolution of about 10 nm, although the apparatus price and materials used are inexpensive. Nanoimprint technology creates a mold (also called a template or stamper) that has a pattern that is the same size as the pattern to be formed in advance, and presses the pattern of this mold directly on the resin film formed on the substrate to be transferred to dynamically deform it. Thus, the pattern to be formed on the resin film is transferred, the transferred substrate is etched using the resin film as a resist mask, and the uneven pattern is transferred. Since the mold can be used not only once but also repeatedly, the low-cost and high-throughput technology is being applied not only to the semiconductor device field but also to various fields.

  Nanoimprint lithography can be broadly classified into thermal imprint lithography in which a pattern is transferred by heat using a thermoreversible material and optical (UV) nanoimprint lithography in which a pattern is transferred by ultraviolet light using a photocurable material. Optical nanoimprints are capable of pattern transfer at room temperature and are free from concerns such as dimensional changes due to heat, and are therefore said to be superior in terms of productivity and resolution compared to thermal nanoimprints.

  The optical nanoimprint process includes a step of applying a photocurable resin on a substrate to be transferred, a step of aligning the substrate to be transferred and the mold (alignment), and lowering the mold onto the photocurable resin and pressing it directly to form an uneven pattern. Step of transferring (pressing), step of curing photocurable resin by light irradiation, step of raising and separating mold from photocurable resin (release), unnecessary photocurable resin (residual film) on transfer substrate It is comprised in the process of removing. Thereafter, using the photo-curable resin as a mask, the transferred substrate is etched to form an uneven pattern.

  For the mold, a light-transmitting substrate such as a quartz glass substrate is used. A metal layer such as chromium is formed on the surface of the quartz glass substrate, and a resist film is applied thereon. Electron beam drawing is performed on the resist film to form a fine resist pattern, and the metal layer and the quartz glass substrate are etched by dry etching. Thereafter, the resist film and the metal layer are peeled and removed, and foreign matter removal cleaning is performed to produce a mold.

JP 2012-183692 A

  The nanoimprint lithography is transferred at the same magnification, and the pattern size and dimensional variation of the master mold are directly transferred onto the transfer substrate. Therefore, dimensional accuracy and dimensional uniformity in the master mold are extremely important. However, even if the dimensional accuracy of the master mold is increased, there may be a dimensional error when removing the curing shrinkage of the photocurable resin during the transfer process and the residual film of the photocurable resin after mold release. .

  It is known that a photocurable resin contracts during curing. In general photocurable resins, linear shrinkage of several to ten and several percent occurs due to curing. Because the shrinkage varies depending on the material and pattern dimensions of the photo-curing resin, in application to semiconductor elements that require high dimensional accuracy, take into account the curing shrinkage due to the resin when designing the master mold, It is necessary to correct the effect of shrinkage.

  The residual film removal of the photocurable resin after mold release is performed by anisotropic etching with oxygen plasma. Although it is anisotropic etching, there is some etching effect in the lateral direction. Further, the residual film thickness of the resin varies depending on the pattern density. Basically, in order to remove all the residual film, the etching time is adjusted so that the thickest residual film can be removed. Therefore, excessive etching is applied in the vicinity of the pattern having a thin residual film thickness. For these reasons, there is a problem that the pattern dimension of the transferred substrate after etching has errors and variations, and the dimensional accuracy is lowered.

  In general, a scanning electron microscope (CD-SEM) is used to measure a pattern dimension on a master mold or a transferred substrate. CD-SEM uses a sample with a guaranteed pattern pitch size to determine the scale (nm / pixel) of the measurement image and guarantees the pitch dimension. However, the CD-SEM has a finite (significantly greater than 0) beam diameter, and as a result, the pattern dimension measurement value is measured with a value different from the actual pattern dimension. In the nanoimprint lithography which is the same size transfer, it is extremely important to accurately measure the actual pattern dimension.

  There has been proposed a method of designing the imprint mold by calculating the curing shrinkage of the resin in optical nanoimprint lithography by computer simulation (see, for example, Patent Document 1). However, since there is no description about the actual dimensional error between the actual pattern size and the measurement size measured by the measuring apparatus, it is difficult to produce an imprint mold with good dimensional accuracy.

  From the above, in nanoimprint lithography, it is necessary to take into account the influence in the transfer process and the influence of the dimension measuring apparatus in controlling the pattern dimension.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a high-quality nanoimprint mold with high dimensional accuracy by correcting the pattern dimensions of the nanoimprint mold with high accuracy.

  One aspect of the present invention is to imprint an imprint mold having an uneven pattern formed on one surface of a substrate onto a transfer material on a transfer substrate and photocure it, thereby transferring the pattern to the transfer material. In nanoimprint lithography, in which the transfer substrate is processed using the transfer material as a mask, and the imprint mold pattern is formed on the transfer substrate, the master mold is manufactured, the pattern dimensions on the master mold are measured, and the dimension correction value Calculating the pattern dimension on the master mold based on the correction value, transferring the pattern onto the transfer substrate, measuring the pattern dimension on the transfer substrate, and based on the correction value And a method of correcting the dimension of the pattern on the transfer substrate, and a method of correcting the dimension of the mold for nanoimprinting It is.

  In addition, in the step of calculating the dimension correction value, a signal profile of the line pattern included in the pattern is generated, and the waveform outside the signal peak position on both side edges is connected to extract a virtual beam profile, The distance between the differential peaks of the differential profile generated by differentiating the beam profile may be calculated as a dimension correction value.

  Further, the side wall angle of the line pattern on the master mold may be vertical or reverse taper.

  In the process of correcting the pattern dimension on the master mold, when the pattern is a concave pattern on the master mold, the correction value is added to the measured value of the pattern dimension on the master mold to correct the pattern. In the case of a convex pattern, the correction value may be subtracted from the measured value of the pattern dimension on the master mold for correction.

  In the process of correcting the pattern dimension on the transfer substrate, if the pattern is a convex pattern on the transfer substrate, the correction value is subtracted from the measured value of the pattern dimension on the transfer substrate to correct the pattern. In the case of a concave pattern, correction may be made by adding a correction value to the measured value of the pattern dimension on the transfer substrate.

  In another aspect of the present invention, an imprint mold having an uneven pattern formed on one surface of a substrate is embossed on a transfer material on a transfer substrate and photocured to transfer the pattern to the transfer material. In nanoimprint lithography, in which the transfer substrate is processed using the transfer material as a mask and the imprint mold pattern is formed on the transfer substrate, the dimensional error generated during the transfer process and the dimensional error in the design dimension of the master mold A method for producing a nanoimprint mold, comprising: a step of feeding back; and a step of producing a master mold based on a design dimension in which a dimensional error is fed back.

  Also, in the process of calculating the dimension error, the dimension error is calculated by taking the difference between the correction value of the pattern dimension on the imprint mold used in the transfer process and the correction value of the pattern dimension on the transfer substrate. Also good.

  In the process of feeding back dimension errors, a dimension correction table describing the dimension error information for each pattern dimension is created in advance, and the design dimension is corrected based on the dimension correction table so that the desired pattern dimension can be obtained after transfer. May be.

  As described above in detail, according to the present invention, the pattern dimension of the nanoimprint mold and the pattern dimension on the transferred substrate can be accurately corrected. Thereby, taking into account the dimensional change that occurs during the transfer process, a high-quality nanoimprint mold with good dimensional accuracy can be manufactured, and as a result, pattern transfer with good dimensional accuracy can be realized.

It is a figure which shows the pattern transfer by the imprint mold which concerns on one Embodiment of this invention. It is a figure which shows the principle of the pattern dimension measuring method by CD-SEM concerning one Embodiment of this invention. It is a figure which shows the relationship between the actual pattern dimension which concerns on one Embodiment of this invention, and the dimension measured by CD-SEM. It is a figure which shows the calculation method of the pattern dimension correction value which concerns on one Embodiment of this invention. It is a figure which shows the flow of the dimension correction method of the mold for nanoimprint which concerns on one Embodiment of this invention, and the manufacturing method of the mold for nanoimprint.

  Hereinafter, an embodiment will be described in detail with reference to the drawings regarding a method for correcting dimensions of a nanoimprint mold and a method for manufacturing a nanoimprint mold according to the present invention.

  The pattern dimension change due to curing shrinkage of the photocurable resin will be described with reference to the drawings. FIG. 1 is a diagram showing a process of transferring a concave pattern formed on a master mold 1 made of a quartz glass substrate onto a substrate 3 to be transferred.

  The photocurable resin 2 is applied on the substrate 3 to be transferred, and the mold 1 is placed thereon and aligned ((a) in FIG. 1). The mold 1 is lowered and pressed directly onto the photocurable resin 2 to transfer the pattern, and the photocurable resin 2 is cured by light irradiation on the upper surface of the mold 1 ((b) in FIG. 1). At this time, the photocurable resin 2 cures and shrinks, and shrinkage occurs in the height and lateral directions of the resin ((c) in FIG. 1). The mold 1 is lifted and pulled away from the photocurable resin 2 to remove an unnecessary remaining film of the photocurable resin 2 ((d) in FIG. 1). The transferred substrate 3 is etched using the photocurable resin 2 as a mask to form a convex pattern, and the photocurable resin 2 is peeled off ((e) in FIG. 1).

  The line width W3 of the convex pattern on the transferred substrate 3 shown in FIG. 1 is smaller than the line width W1 of the concave pattern of the mold 1. The line width of the upper surface portion of the convex pattern of the photocurable resin 2 becomes W2 (<W1) due to curing shrinkage. When the transfer substrate 3 is etched using the photo-curable resin 2 as a mask, the side wall of the transfer substrate is retreated in accordance with the width of the upper surface portion, so that the line width W3 of the transfer substrate is about the same as W2. Therefore, the pattern (W3) having a smaller dimension than the pattern dimension (W1) of the mold 1 is transferred by the curing shrinkage of the photocurable resin 2. As described above, a pattern having a dimension different from the pattern dimension of the master mold is transferred to the substrate to be transferred due to the effect of curing shrinkage of the photocurable resin.

  Next, it will be described with reference to the drawing that the measured value by the CD-SEM is output with a value different from the actual pattern dimension.

  The CD-SEM uses the difference in the amount of secondary electrons emitted according to the unevenness of the sample surface (tilting effect), outputs the luminance difference according to the amount of secondary electrons as an image, and the above image as a signal profile. Determine the pattern edges, and measure the dimensions.

  The principle at the time of measuring each uneven | corrugated pattern with CD-SEM in FIG. 2 is demonstrated. 2A is a principle diagram when measuring a convex pattern, and FIG. 2B is a principle diagram when measuring a concave pattern. When each pattern is scanned with a CD-SEM, a signal profile 20 is obtained. The waveform has a signal peak near the pattern edge. A differential profile 21 is generated by differentiating the signal profile 20.

  A threshold method, a differential gradient method, or the like is used for pattern dimension measurement. The pattern dimension measurement described in the present embodiment is a differential gradient method. As shown in FIG. 2, the differential gradient method detects the outer differential peak in the case of convex pattern length measurement, and detects the inner differential peak in the case of concave pattern length measurement as an edge position corresponding to the pattern bottom position. The distance is output as the measured value of the pattern dimension. In the convex pattern of FIG. 2A, W5 is output as the length measurement value, and in the concave pattern of FIG. 2B, W6 is output as the length measurement value.

  The electron beam of the CD-SEM has a finite beam diameter. Therefore, the waveform does not have a sharp peak at the pattern edge portion, but actually the peak shape of the secondary electron signal is a gentle waveform. Next, the relationship between the actual pattern dimension and the dimension measured by the CD-SEM when the side wall angle is tapered and vertical will be described.

  When the side wall angle of the pattern edge is tapered ((a) and (b) in FIG. 3), the influence of the beam diameter is canceled and a differential peak appears with high accuracy at the position of the pattern bottom. The actual pattern cross-sectional dimension and the dimension measured by CD-SEM (distance between differential peaks) are substantially the same. In the case of the convex pattern of FIG. 3A, the dimension (W7) of the pattern cross section and the CD-SEM measurement dimension (W8) are substantially the same. Similarly, in the case of the concave pattern of FIG. 3B, the dimension (W9) of the pattern cross section and the CD-SEM measurement dimension (W10) are substantially equal.

  On the other hand, when the side wall angle of the pattern edge is vertical (90 degrees) (FIGS. 3C and 3D), the differential peak appears at a position different from the actual pattern bottom due to the influence of the beam diameter. To do. This is because, when the electron beam is irradiated onto the surface, if the charge density of the beam has a spread in the radial direction, information from the electron beam irradiated in the vicinity is folded in addition to the information near the center of the beam diameter. Because it will be included. Therefore, as shown in FIGS. 3C and 3D, the position of the differential peak is located on the outer side in the convex pattern and on the inner side in the concave pattern. When the distance between each differential peak is measured as a line width, the convex pattern is larger than the actual dimension (W12> W11), and the concave pattern is smaller than the actual dimension (W14 <W13). The same applies to the case where the side wall angle of the pattern edge is a reverse taper. As described above in detail, as the side wall angle of the pattern becomes closer to the vertical, a measurement value different from the actual dimension is output.

In general, the beam diameter d of the SEM is expressed by the following equation.

ds is the electron source size, M is the overall magnification of the entire lens system, Cs is the spherical aberration coefficient, α is the probe beam opening angle on the sample surface, Cc is the chromatic aberration coefficient, ΔV is the energy width of the electron probe, and V is the acceleration voltage , Λ is the wavelength of the electron probe. The third term represents chromatic aberration and is inversely proportional to the acceleration voltage V. The fourth term represents an increase in the beam diameter due to the diffraction phenomenon and is proportional to the wavelength λ. The wavelength λ is inversely proportional to the acceleration voltage V. From the above, the acceleration voltage is a parameter that influences the beam diameter, and the beam diameter can be reduced by increasing the acceleration voltage, but it is difficult to reduce the beam diameter so as not to affect the pattern edge width.

  In the present invention, the beam diameter of the CD-SEM and its influence are measured in advance, and the pattern dimension is accurately measured by correcting the influence on the output value of the measurement value, and the result is fed back to the dimension design. Thus, a mold for nanoimprint with high dimensional accuracy is manufactured.

  A method for measuring the influence of the beam diameter will be described with reference to FIG. As described above, a signal profile 40 is obtained by scanning a pattern whose sidewall angle is vertical or inversely tapered with a CD-SEM. If the pattern is vertical or inversely tapered, the edge and shadow effects can be considered sufficiently small compared to the effect of the beam diameter, so the generated signal profile reflects the beam shape as it is. Can be considered. As described above, the width of the pattern edge of the CD-SEM is the peak width of the differential value of the signal profile. Ordinary dimension measurement is also performed with the differential maximum point as an edge. That is, it is the differential width of the beam shape that affects the actual length measurement value. Therefore, the differential width of the differential profile is defined as the beam diameter and the influence width due to the beam diameter.

  Of the signal profile 40 of the line pattern in FIG. 4, waveforms outside the both side edges are connected to extract a virtual beam profile 41 as a signal profile reflecting the beam shape. The beam profile 41 is differentiated to generate a differential profile 42. The differential peak-to-peak distance Wb of the differential profile 42 is defined as a correction value of the pattern dimension, which is an influence width due to the beam diameter.

  Although the effect is half the beam diameter at one edge of the pattern, when viewed as the pattern line width, it is necessary to consider the influence of both edges, that is, the line width is increased by an amount corresponding to the beam diameter (or Will be measured). The true pattern size can be corrected by adding or subtracting the above correction value to or from the measurement value.

  FIG. 5 is a process flow chart for determining the design line width of a nanoimprint master mold with good dimensional accuracy, showing an example of an embodiment of a method for correcting dimensions of a nanoimprint mold and a method for manufacturing a nanoimprint mold of the present invention. is there.

  First, a master mold is produced and a line pattern is measured by CD-SEM. A virtual beam profile is extracted by connecting the waveforms on the outer sides of both edges (S1). A differential profile is generated by differentiating the beam profile, which is a signal profile reflecting the beam shape. The distance between the differential peaks of the differential profile is calculated as the dimension correction width WR (S2). The calculation of the correction width WR corresponding to the beam diameter need only be performed for a representative pattern.

  Next, the pattern dimension WA of the master mold is measured (S3). Next, in the case of a concave pattern, the correction width WR is added to the measured pattern dimension value of the mold to correct the actual mold dimension WB (= WA + WR) (S4). In the case of a convex pattern, the correction width WR is subtracted from the measured pattern dimension value of the mold to correct it to the actual mold dimension WB (= WA−WR) (S5).

  Next, pattern transfer is performed on the photocurable resin applied on the transfer substrate using a master mold (S6). After the mold is released and unnecessary resin is processed to form a residual film, the pattern is transferred by etching the transferred substrate (S7).

  The pattern dimension WC on the transferred substrate is measured (S8). Next, in the case of a convex pattern, the correction width WR is subtracted from the measured pattern dimension value on the transferred substrate to correct it to the actual pattern dimension WD (= WC−WR) on the transferred substrate (S9). ). In the case of a concave pattern, the correction width WR is added to the pattern dimension measurement value on the transfer substrate to correct it to the actual pattern dimension WD (= WC + WR) on the transfer substrate (S10).

  Next, a dimensional error WE (= WB−WD) caused by the influence of curing shrinkage or the like during the transfer process is calculated from the difference between the pattern dimension WD on the transferred substrate and the pattern dimension WB of the mold (S11). The above steps (S3 and subsequent steps) are performed for each pattern dimension, and a dimension correction table is created (S12). The dimension correction table is fed back to the design line width of the master mold (S13). As described above, according to the present invention, a mold for nanoimprinting with small dimensional accuracy and dimensional variation can be manufactured.

  Hereinafter, specific examples of the present invention will be described in detail.

  A 6-inch square quartz glass substrate was prepared as a light-transmitting substrate for the master mold, and a chromium (Cr) film was formed to a thickness of 5 nm. Next, an electron beam resist was applied onto the Cr film, electron beam drawing was performed, and then development was performed to form a resist pattern. Next, using the resist pattern as a mask, the Cr film was dry etched (mixed gas of chlorine and oxygen) to form a pattern on the Cr film. Next, the resist pattern was stripped and removed using oxygen plasma, and dry etching (fluorine gas) was performed using the Cr film as a mask to form a pattern on the quartz glass substrate. Next, the Cr film was removed by etching to produce a master mold in which a pattern was formed on a quartz glass substrate. The pattern size was changed from 50 nm to 1000 nm, and various patterns such as isolated space, isolated line, line & space, hole, and dot pattern were produced.

  Next, the isolated line pattern on the master mold was measured with a CD-SEM to generate a signal profile. In the signal profile, waveforms outside the both edges were connected to extract a virtual beam shape. The signal profile reflecting the beam shape was differentiated to generate a differential profile. The distance between differential peaks of the differential profile was measured and calculated to be 11.2 nm. This value was defined as a dimension correction width WR.

  Next, various other pattern dimensions on the master mold were sequentially measured. As an example, the dimension of the isolated space pattern was measured as 122.1 nm. Next, dimension correction was performed on each pattern. In the case of the concave pattern, the correction width WR is added to the measurement value, and in the case of the convex pattern, the correction width WR is subtracted from the measurement value. The measurement values were corrected for all patterns on the master mold. As an example, the dimension of the isolated space pattern was corrected to 122.1 nm + 11.2 nm = 133.3 nm.

  Next, a silicon wafer was prepared as a transfer substrate. An ultraviolet curable resin PAK-01 (Toyo Gosei Co., Ltd.) was spin coated on a silicon wafer to form a transferred resin layer.

  Next, the master mold was placed so that the surface on which the pattern of the master mold was formed was opposed to the transferred resin layer on the silicon wafer, lowered, pressed directly against the transferred resin layer, and pressed.

  Next, the resin to be transferred was cured by irradiating UV light from the back surface of the master mold, and the pattern of the mold was transferred to the resin. At this time, due to the influence of the curing shrinkage rate of several percent, the resin shrinkage occurred in the height direction and the lateral direction in the transferred resin pattern.

  Next, the remaining film on the transfer substrate was removed by dry etching using oxygen plasma. At this time, the oxygen plasma ashing conditions were an oxygen flow rate of 50 sccm, a pressure of 30 Pa, and an RF power of 100 W.

  Next, the silicon wafer was etched using the resin pattern as an etching mask. At this time, the etching conditions of the silicon wafer were CF4 flow rate 30 sccm, C4F8 flow rate 20 sccm, pressure 30 Pa, ICP power 500 W, and RIE power 50 W.

  Next, the resin was peeled off by oxygen plasma ashing (conditions: oxygen flow rate 50 sccm, pressure 30 Pa, RF power 100 W) to form a pattern on the silicon wafer.

  Next, dimension measurement was performed using CD-SEM with respect to each pattern formed on the silicon wafer. As an example, the isolated space pattern on the master mold was transferred to the isolated line pattern on the silicon wafer, and the dimension was measured as 128.4 nm.

  Next, dimension correction was performed on the dimension measurement values of each pattern. In the case of the convex pattern, the correction width WR is subtracted from the measurement value, and in the case of the concave pattern, the correction width WR is added to the measurement value. The measurement values were corrected for all patterns on the silicon wafer. As an example, the dimension of the isolated line pattern was corrected to 128.4 nm−11.2 nm = 117.2 nm.

  Next, in order to calculate a dimensional error caused by curing shrinkage of the resin during the transfer process, a difference process between the pattern dimension on the master mold after the dimension correction and the pattern dimension on the silicon wafer was performed. As an example, in the case of the above isolated line pattern (on the transfer substrate), it was calculated as 133.3 nm-117.2 nm = 16.1 nm. Due to the curing shrinkage of the resin, it was transferred to the pattern on the silicon wafer with a dimension smaller than the pattern dimension on the master mold. This was performed for each pattern, and a dimension correction table for each pattern dimension was created.

  Next, feedback was given to the design dimensions of the master mold using a dimension correction table. Redesigned taking into account the dimensional variation due to resin shrinkage, and transferred the pattern onto the silicon wafer using the remanufactured master mold, the desired dimensional pattern was obtained and the dimensional accuracy was good. The pattern was transferred.

  Since the nanoimprint mold dimensional correction method and the nanoimprint mold manufacturing method of the present invention can manufacture a high-quality nanoimprint mold with high dimensional accuracy, it can be used in a wide range of fields using nanoimprint lithography. Be expected. For example, it is suitably used in manufacturing processes for imprint molds, optical elements (antireflection films), patterned media (such as hard disks and optical media), biochips, LEDs, microchannels, MEMS, and flexible electronics (organic EL). Is expected to do.

DESCRIPTION OF SYMBOLS 1 Master mold which consists of quartz glass substrates 2 Photocurable resin 3 Substrate to be transferred 20, 40 Signal profile 21, 42 Differential profile 41 Beam profile

Claims (8)

An imprint mold having a pattern with unevenness formed on one surface of a substrate is embossed on a transfer material on a transfer substrate and photocured to transfer the pattern to the transfer material, and the transfer material is masked In nanoimprint lithography for processing the transfer substrate as and forming the pattern of the imprint mold on the transfer substrate,
Producing a master mold; and
Measuring a pattern dimension on the master mold and calculating a correction value of the dimension;
Correcting a pattern dimension on the master mold based on the correction value;
Transferring the pattern onto a transfer substrate;
Measuring the pattern dimensions on the transfer substrate, and correcting the pattern dimensions on the transfer substrate based on the correction value;
A method for correcting dimensions of a nanoimprint mold, comprising: In the step of calculating the correction value of the dimension,
A signal profile of a line pattern included in the pattern is generated, a waveform outside the signal peak positions on both side edges is connected to each other, a virtual beam profile is extracted, and the differential profile generated by differentiating the beam profile The method for correcting a dimension of a nanoimprint mold according to claim 1, wherein the distance between differential peaks is calculated as a correction value of the dimension.   The method of claim 2, wherein a side wall angle of the line pattern on the master mold is vertical or reverse taper. In the step of correcting the pattern dimension on the master mold,
If the pattern is a concave pattern on the master mold, correct by adding the correction value to the measured value of the pattern dimensions on the master mold,
The method for correcting a dimension of a nanoimprint mold according to claim 1, wherein when the pattern is a convex pattern on the master mold, the correction value is subtracted from the measured value of the pattern dimension on the master mold for correction. In the step of correcting the pattern dimension on the transfer substrate,
If the pattern is a convex pattern on the transfer substrate, correct by subtracting the correction value from the measurement value of the pattern dimensions on the transfer substrate,
2. The dimension correction method for a nanoimprint mold according to claim 1, wherein when the pattern is a concave pattern on the transfer substrate, the correction value is added to the measured value of the pattern dimension on the transfer substrate for correction. An imprint mold having a pattern with unevenness formed on one surface of a substrate is embossed on a transfer material on a transfer substrate and photocured to transfer the pattern to the transfer material, and the transfer material is masked In nanoimprint lithography for processing the transfer substrate as and forming the pattern of the imprint mold on the transfer substrate,
Calculating the dimensional error that occurred during the transfer process;
A process of feeding back a dimensional error to the design dimension of the master mold;
Producing a master mold based on a design dimension obtained by feeding back the dimension error;
A method for producing a mold for nanoimprinting, comprising: In the step of calculating the dimensional error,
The nanoimprint use according to claim 6, wherein a dimensional error is calculated by taking a difference between a correction value of a pattern dimension on the imprint mold used in the transfer step and a correction value of the pattern dimension on the transfer substrate. Mold manufacturing method. The step of feeding back the dimensional error includes:
The dimension correction table which described the information of the said dimension error for every pattern dimension is created previously, and a design dimension is corrected so that a desired pattern dimension may be obtained after transfer based on the said dimension correction table. Manufacturing method of mold for nanoimprint.

Images