Error detection method and apparatus for asymmetric grating marks

Abstract

The disclosure provides an error detection method and device for an asymmetric grating mark, the method comprising: irradiating a to-be-detected asymmetric grating mark with spatially coherent light to obtain first diffraction beams of different diffraction orders, the spatially coherent light being a partial pupil beam; establishing an ideal grating mark according to the first diffraction beams; obtaining a first error feature in a prior asymmetric grating mark according to the prior asymmetric grating mark and the ideal grating mark; reconstructing the to-be-detected asymmetric grating mark according to the first error feature to obtain a plurality of reconstructed asymmetric grating marks; iteratively comparing the plurality of reconstructed asymmetric grating marks with the to-be-detected asymmetric grating mark to obtain a target reconstructed asymmetric grating mark, a difference between a characteristic value of the target reconstructed asymmetric grating mark and a characteristic value of the to-be-detected asymmetric grating mark being less than a first threshold; and obtaining an error of the to-be-detected asymmetric grating mark according to an error of the target reconstructed asymmetric grating mark.

Description

Technical Field

[0001] The present disclosure relates to the field of integrated circuit technology, and in particular to a method and device for detecting errors in asymmetric grating marks. Background Art

[0002] Phase grating position measurement systems are widely used in integrated circuit manufacturing, particularly in precision-critical steps such as photolithography alignment, motion stage positioning, and metrology and inspection systems. In these systems, precise position information is typically acquired by scanning grating marks. Due to the complexity of the integrated circuit manufacturing process, grating marks can deform after processes such as exposure, etching, and chemical-mechanical polishing. This asymmetric deformation can introduce additional measurement errors.

[0003] Currently, grating structure measurement methods are used to measure deviations. Scattering measurement methods typically utilize the zeroth and ±1st diffraction order signals in the pupil plane. To ensure sufficient interference between the zeroth and ±1st diffraction orders, the beam must cover the entire pupil, placing higher demands on the aberrations of the optical system. Furthermore, the high intensity of the zeroth-order signal significantly reduces signal contrast. This is especially true when other periodic structures are present around the structure being measured. Stray light further reduces the signal-to-noise ratio, potentially leading to measurement failures.

[0004] Among the methods for reconstructing grating structures, the nonlinear iterative method has high accuracy, but requires multiple iterations and has a high time cost; the library matching method requires the establishment of a huge amount of data, and the data interval setting directly affects the calculation accuracy. Summary of the Invention

[0005] (1) Technical issues to be resolved

[0006] In response to the existing technical problems, the present disclosure provides an error detection method and device for an asymmetric grating mark, which are used to at least partially solve the above technical problems.

[0007] (2) Technical solution

[0008] The disclosure provides an error detection method for an asymmetric grating mark, comprising: irradiating a to-be-detected asymmetric grating mark with spatially coherent light to obtain first diffraction beams of different diffraction orders, the spatially coherent light being a partial pupil beam; establishing an ideal grating mark according to the first diffraction beams; obtaining a first error feature in a prior asymmetric grating mark according to the prior asymmetric grating mark and the ideal grating mark; reconstructing the to-be-detected asymmetric grating mark according to the first error feature to obtain a plurality of reconstructed asymmetric grating marks; iteratively comparing the plurality of reconstructed asymmetric grating marks with the to-be-detected asymmetric grating mark to obtain a target reconstructed asymmetric grating mark, a difference between a feature value of the target reconstructed asymmetric grating mark and a feature value of the to-be-detected asymmetric grating mark being less than a first threshold; and obtaining an error of the to-be-detected asymmetric grating mark according to an error of the target reconstructed asymmetric grating mark.

[0009] Optionally, the establishing of the ideal grating mark according to the first diffraction beams comprises: adjusting the spacing between the spots of the first diffraction beams to obtain mutually separated diffraction order spots; obtaining a duty cycle and a groove depth of the to-be-detected asymmetric grating mark according to the intensities of the diffraction order spots; and establishing the ideal grating mark according to the duty cycle and the groove depth of the to-be-detected asymmetric grating mark.

[0010] Optionally, the obtaining of the first error feature in the prior asymmetric grating mark according to the prior asymmetric grating mark and the ideal grating mark comprises: establishing a rigorous electromagnetic wave simulation model; measuring the ideal grating mark in the rigorous electromagnetic wave simulation model to obtain a first measurement signal; measuring the prior asymmetric grating mark in the rigorous electromagnetic wave simulation model to obtain a second measurement signal; and calculating a feature vector difference between the first measurement signal and the second measurement signal to obtain the first error feature.

[0011] Optionally, the reconstructing of the to-be-detected asymmetric grating mark according to the first error feature to obtain the plurality of reconstructed asymmetric grating marks comprises: interfering diffraction beams of different diffraction orders to obtain an interference signal, wherein the diffraction orders are greater than ±1 order; measuring the intensity of the interference signal to obtain a third measurement signal; calculating a feature vector difference between the first measurement signal and the third measurement signal to obtain a second error feature in the to-be-detected asymmetric grating mark; determining an error weight of the second error feature in the first error feature; and reconstructing the to-be-detected asymmetric grating mark according to the error weight and the first error feature to obtain the plurality of reconstructed asymmetric grating marks.

[0012] Optionally, the determining of the error weight of the second error feature in the first error feature comprises: selecting part of the error features from the first error feature to obtain a third error feature, the weight of the third error feature in the first error feature being greater than a second threshold; and determining an error weight of the second error feature in the third error feature.

[0013] Optionally, the measuring the intensity of the interference signal to obtain the third measurement signal comprises: splitting the interference signal to obtain at least two split interference signals; and measuring the intensity of the at least two split interference signals to obtain the third measurement signal.

[0014] Optionally, the iteratively comparing the plurality of reconstructed asymmetric grating marks with the to-be-detected asymmetric grating mark to obtain the target reconstructed asymmetric grating mark comprises: iteratively comparing the plurality of reconstructed asymmetric grating marks with the to-be-detected asymmetric grating mark according to the error weight; and obtaining the target reconstructed asymmetric grating mark in a case where the error weight converges.

[0015] Optionally, the obtaining the error of the to-be-detected asymmetric grating mark according to the error of the target reconstructed asymmetric grating mark comprises: measuring the target reconstructed asymmetric grating mark in a rigorous electromagnetic wave simulation model to obtain second diffraction beams; determining phase differences between the second diffraction beams; obtaining the error of the target reconstructed asymmetric grating mark according to the phase differences; and obtaining the error of the to-be-detected asymmetric grating mark according to the error of the target reconstructed asymmetric grating mark and the error weight when the error converges.

[0016] Another aspect of the present disclosure provides an error detection device for an asymmetric grating mark, comprising: an aperture stop configured to partially block spatially coherent light to generate a partial pupil beam, the partial pupil beam being configured to irradiate a to-be-detected asymmetric grating to generate first diffraction beams of different diffraction orders; a beam splitter configured to split the first diffraction beams of different diffraction orders to obtain first split diffraction beams of different diffraction orders and second split diffraction beams of different diffraction orders; a first detector configured to detect the intensity of the first split diffraction beams of different diffraction orders; a cross diaphragm configured to adjust the overlapping area between the second split diffraction beams of different diffraction orders to cause the second split diffraction beams of different diffraction orders to interfere to obtain an interference signal; and a second detector configured to detect the intensity of the interference signal.

[0017] Optionally, the error detection device for the asymmetric grating mark further comprises: a splitting prism configured to split the interference signal to obtain at least two split interference signals.

[0018] (III) Beneficial Effects

[0019] Compared with the prior art, the error detection method and device for the asymmetric grating mark provided by the present disclosure have at least the following beneficial effects:

[0020] (1) The method of the present disclosure uses a partial pupil beam to irradiate a to-be-measured asymmetric grating mark, thereby reducing the influence of optical aberration. By establishing an ideal grating mark, the initial iterative grating parameters can be derived, providing an initial structure for reconstructing the asymmetric mark, reducing the number of iterations and accelerating convergence. By iterative calculation, a target reconstructed asymmetric grating mark identical or close to the structure of the to-be-measured asymmetric grating mark is found, and the error of the to-be-measured asymmetric grating mark can be calculated according to the error of the reconstructed asymmetric grating mark. In the case of ensuring measurement accuracy, the method is fast and non-destructive, and can be widely applied to online measurement feedback and detection in the manufacturing process.

[0021] (2) The method of the present disclosure uses high diffraction order signals with diffraction orders greater than ±1 to measure the intensity of the interference signal, which belongs to dark field measurement, has high signal-to-noise ratio, and improves the measurement accuracy.

[0022] (3) The present disclosure uses a beam splitting measurement method to measure the interference signal by splitting and contrasting, provides a reference signal, reduces the influence of environmental factors, and improves the measurement stability. BRIEF DESCRIPTION OF DRAWINGS

[0023] The above and other objects, features and advantages of the present disclosure will become more apparent from the following description of embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0024] Figure 1 A structural diagram of an error detection device for an asymmetric grating mark according to an embodiment of the present disclosure is schematically shown;

[0025] Figure 2 A flowchart of an error detection method for an asymmetric grating mark according to an embodiment of the present disclosure is schematically shown;

[0026] Figure 3 A diagram of separated diffraction order light spot patterns on the Fourier plane of a collection lens group 106 according to an embodiment of the present disclosure is schematically shown;

[0027] Figure 4 A structural diagram of an ideal grating mark according to an embodiment of the present disclosure is schematically shown;

[0028] Figure 5 a~5f A structural diagram of a prior asymmetric grating mark according to an embodiment of the present disclosure is schematically shown;

[0029] Figure 6 An interference light spot pattern after a cross diaphragm according to an embodiment of the present disclosure is schematically shown;

[0030] Figure 7 A result diagram of a separation detector measurement according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0031] In order to make the objects, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to specific embodiments and in conjunction with the drawings.

[0032] It should be noted that similar or identical parts are denoted by the same reference numerals in the drawings or the description. The technical features in the embodiments shown in the description can be freely combined without conflict to form new solutions. In addition, each claim can be an individual embodiment or the technical features in the claims can be combined to form new embodiments. In the drawings, the shape or thickness of the embodiments can be exaggerated and simplified for the purpose of convenience or clarity. Furthermore, elements or implementation manners not shown or described in the drawings are in forms known to those skilled in the art. In addition, although examples of parameters including specific values can be provided herein, it should be understood that the parameters do not need to be exactly equal to the corresponding values, but can be approximately equal to the corresponding values within an acceptable error tolerance or design constraint.

[0033] The various embodiments of the present disclosure described above can be freely combined to form additional embodiments, which are all within the scope of protection of the present disclosure, unless there is a technical obstacle or conflict.

[0034] Although the present disclosure is described in conjunction with the drawings, the embodiments disclosed in the drawings are intended to exemplarily illustrate the preferred embodiments of the present disclosure and cannot be understood as a limitation of the present disclosure. The dimensional proportions in the drawings are merely schematic and cannot be understood as a limitation of the present disclosure.

[0035] Although some embodiments of the general concept of the present disclosure have been shown and described, those of ordinary skill in the art will understand that changes can be made in these embodiments without departing from the principles and spirit of the general concept of the present disclosure, and the scope of the present disclosure is defined by the claims and their equivalents.

[0036] In one aspect, the present disclosure provides an error detection device for asymmetric grating marks, which is used to detect the asymmetric grating marks to obtain a measurement signal. According to the measurement signal, asymmetric feature information can be obtained, and then the error of the asymmetric grating marks to be measured can be calculated.

[0037] It can be understood that the error detection device for asymmetric grating marks can also be referred to as a phase grating position measurement device.

[0038] Figure 1 The structural diagram of the error detection device for asymmetric grating marks according to the embodiments of the present disclosure is schematically shown.

[0039] According to the embodiments of the present disclosure, as Figure 1As shown, the error detection device of the asymmetric grating mark, for example, comprises an aperture stop 103 for partially blocking the spatially coherent light to generate a partial pupil beam for illuminating the asymmetric grating to be measured, thereby generating first diffraction beams of different diffraction orders. A beam splitter 108 is used to split the first diffraction beams of different diffraction orders to obtain first split diffraction beams of different diffraction orders and second split diffraction beams of different diffraction orders. A first detector 109 is used to detect the intensity of the first split diffraction beams of different diffraction orders. A cross diaphragm 111 is used to adjust the overlapping area between the second split diffraction beams of different diffraction orders, so that the second split diffraction beams of different diffraction orders interfere with each other to generate an interference signal. A second detector 116 is used to detect the intensity of the interference signal. A radiation source 101 is used to generate a spatially coherent illumination beam. A collimating system 102 is used to generate a uniform and nearly parallel illumination beam. An illumination lens group 104, another beam splitter 105, and a collection lens group 106 are used to constitute a 4f optical system to make the spatially coherent illumination beam incident on a grating mark 107. The grating mark 107 is used to generate diffraction beams of different diffraction orders, which become parallel beams after passing through the collection lens group 106 again. After passing through the beam splitter 108, a part of the diffraction beams are directly imaged on the first detector 109. The first detector 109 is located at the focal plane position of the collection lens group 106 and is used to measure the light intensity of each diffraction beam. The light intensity information of each diffraction order is used to measure the groove depth and duty cycle of the grating mark 107, to provide an initial structure for measuring the asymmetry of the grating mark 107 for another part of the diffraction beams, to reduce the coupling between the structure parameters of the grating mark 107, and to reduce the difficulty of reconstructing the asymmetry of the grating. Another part of the diffraction beams passes through a 4f optical system composed of the collection lens group 106 and a focusing lens group 110 and converges at the cross diaphragm 111. The cross diaphragm 111 is used to make the other part of the diffraction beams overlap with each other to generate an interference signal. The interference signal passes through a 4f optical system composed of a first relay lens group 113 and a second lens group 115 and is detected by the second detector 116.

[0040] In order to reduce the influence of the environment and improve the measurement accuracy, the error detection device of the asymmetric grating mark further comprises a beam splitting prism 114 for splitting the interference signal into at least two split interference signals, and a third detector 117. The interference signal is split by the beam splitting prism 114 into two parts and converges on the second detector 116 and the third detector 117, respectively. The beam splitting prism 114 is used to split the interference signal after the cross diaphragm 111 to provide a reference signal, thereby reducing the influence of the environment and improving the signal-to-noise ratio of the measurement signal.

[0041] For example, by adjusting the size of the slit width a of the cross diaphragm 111, the diffraction order that interferes can be selected. Selecting a high diffraction order signal of a diffraction order greater than ±1 near the edge can enable the second detector 116 and the third detector 117 to form a hidden detection, further improving the measurement accuracy.

[0042] Based on the error detection device of the asymmetric grating mark described above, another aspect of the disclosure provides an error detection method of an asymmetric grating mark.

[0043] Figure 2 A flowchart of an error detection method of an asymmetric grating mark according to an embodiment of the disclosure is schematically shown.

[0044] According to an embodiment of the disclosure, as Figure 2 shown, the error detection method of an asymmetric grating mark, for example, includes:

[0045] S210, irradiating the to-be-measured asymmetric grating mark with spatially coherent light to obtain first diffracted beams of different diffraction orders, the spatially coherent light being a partial pupil beam.

[0046] According to an embodiment of the disclosure, the to-be-measured asymmetric grating mark is irradiated with a small-size partial pupil beam generated by the aperture diaphragm 103, for example, to obtain first diffracted beams of different diffraction orders. The first diffracted beams become parallel beams after passing through the collection lens group 106 and continue to propagate.

[0047] S220, establishing an ideal grating mark according to the first diffracted beams.

[0048] According to an embodiment of the disclosure, the ideal grating mark is established by steps S221-S223, for example.

[0049] Step S221, adjusting the spacing between the spots of the first diffracted beams to obtain mutually separated diffraction order spots.

[0050] Figure 3 A diffraction order spot diagram separated on the Fourier plane of the collection lens group 106 according to an embodiment of the disclosure is schematically shown.

[0051] To measure the intensity of each diffraction order spot, it is necessary to separate the spots of each diffraction order. When a parallel beam is incident on the grating mark 107, it can be known from the grating equation that the interval of each diffraction order spot on the Fourier plane of the collection lens group 106 is about:

[0052]

[0053] where d is the grating period, λ is the measurement wavelength, f 106The focal length of the collection lens group 106 is collected. When the incident light beam diameter is ε, the spot size on the Fourier plane of the collection lens group 106 is approximately:

[0054]

[0055] In order to separate the diffraction order spots from each other on the Fourier plane of the collection lens group 106, that is:

[0056] ω < Δξ, (3) It is required to measure the spot diameter ε ≥ 2.44d, at which the diffraction order spots are separated from each other on the Fourier plane of the collection lens group 106, as shown in Figure 3

[0057] In step S222, the duty cycle and groove depth of the asymmetric grating mark to be measured are obtained according to the intensity of the diffraction order spots.

[0058] For example, the first diffracted light beam in which the diffraction order spots are separated from each other is imaged on the first detector 109 after passing through the beam splitter 108. The light intensity of the part of the diffracted light beam is measured by the first detector 109. The detector 109 is located at the Fourier plane position of the collection lens group 106, and measures the light intensity of the diffraction spots on the Fourier plane to calculate the groove depth and duty cycle information of the grating mark 107, thereby providing the initial structure for reconstructing the asymmetry of the grating mark 107.

[0059] The duty cycle f can be obtained by comparing the light intensity of the second diffraction order spot with the light intensity of the first diffraction order spot, that is:

[0060]

[0061] Where I2 is the light intensity of the second diffraction order, and I1 is the light intensity of the first diffraction order. The groove depth h can be obtained by comparing the light intensity of the first diffraction order with the light intensity of the zeroth diffraction order, combined with the duty cycle f information, that is:

[0062]

[0063] When the duty cycle is 0.5, the groove depth h formula can also be simplified as:

[0064]

[0065] In step S223, an ideal grating mark is established according to the duty cycle and groove depth of the asymmetric grating mark to be measured.

[0066] Figure 4 The structural diagram of the ideal grating mark according to the embodiment of the present disclosure is schematically shown.

[0067] According to the embodiment of the present disclosure, as Figure 4 ​As shown, by calculating the duty cycle f and the groove depth h, an error-free ideal grating mark can be established. The initial structure for subsequent reconstruction of the asymmetric grating mark is selected from the actually measured structure information, considering the influence of the actual processing process, which is beneficial to reduce the iteration number, improve the convergence speed of reconstruction, and further improve the measurement speed.

[0068] In S230, a first error feature in the prior asymmetric grating mark is obtained according to the prior asymmetric grating mark and the ideal grating mark.

[0069] According to an embodiment of the present disclosure, the first error feature in the prior asymmetric grating mark is obtained, for example, by steps S231-S234.

[0070] In S231, a rigorous electromagnetic wave simulation model is established.

[0071] For example, according to the optical system as shown in Figure 1 , a rigorous electromagnetic wave simulation model is established.

[0072] In S232, the ideal grating mark is measured in the rigorous electromagnetic wave simulation model to obtain a first measurement signal.

[0073] For example, in the rigorous electromagnetic wave simulation model, the ideal grating mark is simulated to be irradiated by parallel spatial coherent light, different diffraction orders of diffraction beams are obtained, and the different diffraction orders of diffraction beams are interfered to obtain an interference signal. The intensity of the interference signal is measured, i.e., the first measurement signal I0 is obtained.

[0074] In S233, the prior asymmetric grating mark is measured in the rigorous electromagnetic wave simulation model to obtain a second measurement signal.

[0075] Figure 5 Figures a-5f schematically show structure diagrams of prior asymmetric grating marks according to embodiments of the present disclosure.

[0076] According to an embodiment of the present disclosure, the prior asymmetric grating mark is, for example, a prior asymmetric grating mark in an actual process, which has different types and sizes. The types of the asymmetric grating mark include, but are not limited to, bottom, top, sidewall, top round corner, bottom round corner, etc. asymmetric, and also includes mutual combinations of various types of asymmetry, for example, can be Figure 5 bottom asymmetry as shown in Figure a, Figure 5 sidewall asymmetry as shown in Figure b, Figure 5 top round corner asymmetry as shown in Figure c, Figure 5 top inclined asymmetry as shown in Figure d, Figure 5 bottom round corner asymmetry as shown in Figure e, and Figure 5a combination of top rounded corners, bottom rounded corners, and top asymmetry, as shown in FIG. 1f.

[0077] For example, in a rigorous electromagnetic wave simulation model, a parallel spatial coherent light is used to irradiate the above-mentioned various types and sizes of prior asymmetric grating marks, different diffraction orders of diffraction beams are obtained, and the different diffraction orders of diffraction beams are interfered to obtain an interference signal. The intensity of the interference signal is measured, that is, a second measurement signal I n .

[0078] Step S234, calculating the difference between the feature vectors of the first measurement signal and the second measurement signal to obtain a first error feature.

[0079] For example, by comparing the measurement signals I n of different types of asymmetric grating marks, the feature signals S n of different position and size asymmetric grating marks (i.e. the first error feature) are obtained, which are expressed as:

[0080] S n = I n -I0 (7)

[0081] For example, the first error feature S n is a database containing a plurality of error feature signals.

[0082] S240, according to the first error feature, reconstructing the asymmetric grating mark to be measured to obtain a plurality of reconstructed asymmetric grating marks.

[0083] According to an embodiment of the present disclosure, the asymmetric grating mark to be measured is reconstructed, for example, by steps S241-S245.

[0084] Step S241, interfering the diffraction beams of different diffraction orders to obtain an interference signal, wherein the diffraction orders are greater than ±1 order.

[0085] Figure 6 The interference spot diagram after the cross diaphragm according to an embodiment of the present disclosure is schematically shown.

[0086] For example, as Figure 6 shown, the diffraction beams of different diffraction orders are interfered by the cross diaphragm 111 to obtain an interference signal. And by adjusting the size of the slit width a of the cross diaphragm 111, the high diffraction order signal of the diffraction order close to the edge greater than ±1 order is selected, for example, the diffraction order signal of ±2 order is selected. For example, the smaller a is, the higher the selectable diffraction order is.

[0087] It can be understood that the diffraction order signal of ±2 level selected in the embodiment is only exemplary, and diffraction order signals of ±3, ±4 and the like can also be selected.

[0088] In step S242, the intensity of the interference signal is measured to obtain a third measurement signal.

[0089] For example, the intensity of the interference signal is measured by the second detector 116 to obtain a third measurement signal I. The third measurement signal is a measurement signal measured when actually measuring the asymmetric grating mark to be measured.

[0090] In order to improve the signal-to-noise ratio of the measurement signal, a reference signal can be generated to exclude the influence of environmental information.

[0091] For example, the third measurement signal is optimized by steps S2421-S2422.

[0092] In step S2421, the interference signal is split to obtain at least two split interference signals.

[0093] Figure 7 The result diagram of the separation detector measurement according to the embodiment of the present disclosure is schematically shown.

[0094] For example, the measurement signal is split into two parts by using the beam splitter prism 114, and is measured by the detector 116 and the detector 117 respectively, as shown in Figure 7 The detector 116 and the detector 117 measure the interference signal of the high diffraction order to form dark field detection, which reduces the influence of stray light and light source fluctuation and the like, and improves the measurement accuracy.

[0095] In step S2421, the intensity of the at least two split interference signals is measured to obtain a third measurement signal.

[0096] For example, the signal difference between the second detector 116 and the third detector 117 is compared and normalized as a measurement signal:

[0097]

[0098] Wherein, I 116 is the intensity of the interference signal detected by the second detector 116, and I 117 is the intensity of the interference signal detected by the third detector 117.

[0099] In step S243, the feature vector difference between the first measurement signal and the third measurement signal is calculated to obtain a second error feature in the asymmetric grating mark to be measured.

[0100] For example, by comparing the measurement signal I of the asymmetric grating mark to be measured with the measurement signal I0 of the initial structure, the feature signal S (i.e. the second error feature) of the asymmetric grating mark to be measured is obtained.

[0101] Step S244, determining the error weight of the second error feature in the first error feature.

[0102] For example, the actually measured feature signal S can be projected onto the feature vector S n to obtain the weight w.

[0103] In order to reduce the amount of data and improve the calculation speed, the feature vector S n can be compressed.

[0104] For example, the feature vector S n is compressed by steps S2441-S2442.

[0105] Step S2441, selecting part of the error features from the first error features to obtain the third error features, and the weight of the third error features in the first error features is greater than the second threshold.

[0106] For example, singular value decomposition (SVD) is used to select the feature vectors S j corresponding to the first j eigenvalues (i.e. the third error features), where j

[0107] Step S2442, determining the error weight of the second error feature in the third error feature.

[0108] For example, the actually measured feature signal S can be projected onto the feature vector S j to obtain the weight w j . By reducing the dimension compression calculation data amount, the orthogonality of the feature information is improved, which facilitates the projection of the feature signal S, and at the same time reduces the storage space of the database and improves the overall calculation speed.

[0109] Step S245, reconstructing the to-be-measured asymmetric grating mark according to the error weight and the first error feature to obtain a plurality of reconstructed asymmetric grating marks.

[0110] For example, the feature signal S' of the actual mark is reconstructed with w j and the feature vector S j .

[0111] S250, iteratively comparing the plurality of reconstructed asymmetric grating marks with the to-be-measured asymmetric grating mark to obtain a target reconstructed asymmetric grating mark, and the difference between the eigenvalue of the target reconstructed asymmetric grating mark and the eigenvalue of the to-be-measured asymmetric grating mark is less than the first threshold.

[0112] According to an embodiment of the present disclosure, the target reconstructed asymmetric grating mark is obtained, for example, by steps S251-S252.

[0113] At step S251, the plurality of reconstructed asymmetric grating marks and the to-be-measured asymmetric grating mark are iteratively compared according to the error weight.

[0114] For example, the difference between S’ and S is compared by using a parallel nonlinear iteration method, and the iteration w is repeated j until w j converges. The position deviation caused by the asymmetry of the grating structure can be reconstructed to improve the measurement accuracy.

[0115] At step S252, the target reconstructed asymmetric grating mark is obtained when the error weight converges.

[0116] The specific convergence requirement is determined according to the actual situation, for example, the difference between S’ and S is less than 0.1%.

[0117] At S260, the error of the to-be-measured asymmetric grating mark is obtained according to the error of the target reconstructed asymmetric grating mark.

[0118] According to an embodiment of the present disclosure, the error of the to-be-measured asymmetric grating mark is obtained, for example, by steps S261-S264.

[0119] At step S261, the target reconstructed asymmetric grating mark is measured in a strict electromagnetic wave simulation model to obtain a second diffracted light beam.

[0120] At step S262, the phase difference between the second diffracted light beams is determined.

[0121] At step S263, the error of the target reconstructed asymmetric grating mark is obtained according to the phase difference.

[0122] For example, the error of the target reconstructed asymmetric grating mark is calculated according to the calculation formula of the position error caused by the asymmetry of the mark:

[0123]

[0124] The error Δx j of the target reconstructed asymmetric grating mark is calculated. Wherein, is the phase difference caused by the ±m diffraction order of the asymmetric mark, and d is the grating period.

[0125] It can be understood that, in order to calculate the phase difference, the can be encoded into the interference signal. For example, the focusing lens group 110 is used to converge the diffraction spot of each diffraction order, and then the cross diaphragm 111 is used to diffract and expand the range of each diffracted light beam to generate an interference signal.

[0126] Step S264, according to the error of the target reconstructed asymmetric grating mark and the error weight at the time of convergence, the error of the asymmetric grating mark to be measured is obtained.

[0127] For example, the feature vector S is obtained based on optical model simulation j The asymmetric error Δx j of the asymmetric grating mark to be measured is calculated as Δx = Δx j × w j .

[0128] In summary, the error detection method of the asymmetric grating mark is proposed in the embodiments of the present disclosure. By using a spatially coherent narrow illumination beam to reduce the influence of optical aberration, by measuring the Fourier plane light intensity information, the grating parameters of the initial iteration are derived, the iteration number is reduced and the convergence is accelerated. The dark field measurement is used to obtain high diffraction order signals to improve the signal-to-noise ratio, improve the measurement accuracy, and use a beam splitter prism to provide a reference signal to reduce the influence of environmental factors and improve the measurement stability. The library search in the library matching method is combined with singular value decomposition to extract asymmetric feature information, the data amount is reduced by dimensionality reduction compression, the orthogonality and overall speed of the feature information are improved, and the asymmetric feature signal S to be measured is decomposed, and parallel nonlinear iteration is used to further improve the calculation accuracy.

[0129] The product embodiment part is similar to the method embodiment part, please refer to the method embodiment part, this place will not be repeated.

[0130] It should be understood that the specific order or hierarchy of steps in the disclosed processes should not be taken as a limitation of the example methods. Based on design preference, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of protection of the present disclosure. The attached method claims give the elements of various steps in an exemplary order, and are not intended to be limited to a specific order or hierarchy.

[0131] It should also be noted that the direction terms mentioned in the embodiments, such as "up", "down", "front", "back", "left", "right", etc. are only the direction of the drawings, and are not intended to limit the scope of protection of the present disclosure. Throughout the drawings, the same elements are represented by the same or similar reference numerals. When it may cause confusion in understanding the present disclosure, the conventional structure or configuration will be omitted. And the shape, size, positional relationship of the components in the drawings do not reflect the true size, proportion and actual positional relationship.

[0132] In the detailed description above, various features are grouped together in single embodiments for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting a necessity to more features than are expressly recited in each claim. Thus, the claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment. It is the intention of the inventor to encompass all such variations and / or permutations of the disclosure within the purview of the appended claims, which follow.

[0133] In addition, the terms "first", "second", and the like, do not denote any order, quantity, combination, or importance, but rather are used to nomenclature different components. Thus, such terms are used herein, for purposes of a specific example, and are in no way limiting. For example, a first element could be termed a second element; and, similarly, a second element could be termed a first element without departing from the scope of the disclosure. The singular forms "a", "an", and "the" include plural referents unless the context clearly dictates otherwise. The terms "comprises", "comprising", and "including" are to be construed as being inclusive and open ended and not exclusive or exhaustive. The terms "or" and "and" are to be construed as being inclusive and not exclusive or exhaustive. The terms "coupled" and "connected" are to be construed as permitting some form of interaction between or among the coupled or connected components.

[0134] The above detailed description merely describes specific embodiments of the disclosure, and is not intended to limit the scope of the disclosure. The description of the embodiments is merely provided for the purpose of illustration, and the intended scope of the disclosure is not limited to the embodiments described herein. Therefore, any modification, equivalent replacement, or improvement not departing from the spirit and principle of the disclosure should be included in the scope of the disclosure.

Claims

1. A method of error detection for asymmetric grating marks, characterized in that, include: Irradiating the asymmetric grating mark to be measured with spatially coherent light to obtain first diffraction beams of different diffraction orders, wherein the spatially coherent light is a partial pupil beam; Adjusting the spacing between the light spots of the first diffraction beam to obtain separate diffraction order light spots; obtaining the duty cycle and groove depth of the asymmetric grating mark to be measured based on the intensities of the diffraction order light spots; and establishing an ideal grating mark based on the duty cycle and groove depth of the asymmetric grating mark to be measured; Establishing a strict electromagnetic wave simulation model; measuring the ideal grating mark in the strict electromagnetic wave simulation model to obtain a first measurement signal; measuring a priori asymmetric grating mark in the strict electromagnetic wave simulation model to obtain a second measurement signal; calculating the difference between the eigenvectors of the first measurement signal and the second measurement signal to obtain a first error characteristic; Interfering the diffracted light beams of different diffraction orders to obtain an interference signal, wherein the diffraction order is greater than ±1 order; measuring the intensity of the interference signal to obtain a third measurement signal; calculating the difference between the characteristic vectors of the first measurement signal and the third measurement signal to obtain a second error feature in the asymmetric grating mark to be measured; determining an error weight of the second error feature in the first error feature; and reconstructing the asymmetric grating mark to be measured based on the error weight and the first error feature to obtain a plurality of reconstructed asymmetric grating marks; Iteratively comparing the multiple reconstructed asymmetric grating marks and the asymmetric grating mark to be measured according to the error weight; obtaining a target reconstructed asymmetric grating mark when the error weight converges, and a difference between a characteristic value of the target reconstructed asymmetric grating mark and a characteristic value of the asymmetric grating mark to be measured is less than a first threshold; The error of the asymmetric grating mark to be measured is obtained by reconstructing the error of the asymmetric grating mark according to the target.

2. The method of claim 1, wherein, Determining the error weight of the second error feature in the first error feature includes: Selecting part of the error features from the first error features to obtain a third error feature, wherein a weight of the third error feature in the first error feature is greater than a second threshold; An error weight of the second error feature in the third error feature is determined.

3. The method of claim 1, wherein, Measuring the intensity of the interference signal to obtain a third measurement signal includes: Splitting the interference signal to obtain at least two split interference signals; The intensities of the at least two branch interference signals are measured to obtain the third measurement signal.

4. The method of claim 1, wherein, The step of reconstructing the error of the asymmetric grating mark according to the target to obtain the error of the asymmetric grating mark to be measured comprises: measuring the target reconstructed asymmetric grating mark in the strict electromagnetic wave simulation model to obtain a second diffracted beam; determining a phase difference between the second diffracted beams; Obtaining an error of the target reconstructed asymmetric grating mark according to the phase difference; The error of the asymmetric grating mark to be measured is obtained according to the error of the target reconstructed asymmetric grating mark and the error weight at the time of convergence.

Images