Method for identifying a defect affecting a control pattern carried by a microelectronic component; associated control method, instrumental system and computer program product.
The method uses small-angle X-ray scattering to quickly identify defects in microelectronic components by analyzing Bragg rods, simplifying defect extraction and optimizing manufacturing processes.
Patent Information
- Application Number
- FR2023004035
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-04-21
- Publication Date
- 2025-09-19
- Estimated Expiration
- 2043-04-21
AI Technical Summary
Existing methods for measuring defects in control patterns of microelectronic components are complex and time-consuming, particularly when considering multiple parameters, leading to prolonged data acquisition times.
A method utilizing small-angle X-ray scattering techniques to rapidly identify defects in control patterns by acquiring intensity measurements at various angles, reconstructing Bragg rods, and applying tests to determine defect types, including misalignment and edge deformation.
The method simplifies the extraction of geometric information from control patterns, enabling quick identification of defects and optimizing manufacturing conditions for microelectronic components.
Smart Images
Figure 00000017_0000 
Figure 00000018_0000 
Figure 00000018_0001
Abstract
Description
Title of the invention: Method for identifying a defect affecting a control pattern carried by a microelectronic component; associated control method, instrumental system and computer program product.
[0001] The present invention relates to the field of methods for measuring by small angle X-ray scattering - CD-SAXS (according to the English "Critical-Dimension Small Angle X-ray Scattering"), in particular methods for measuring by transmission - T-SAXS (for "transmission - Small Angle X-ray Scattering") or by reflection - GL SAXS (for "Grazing Incident - Small Angle X-ray Scattering").
[0002] More particularly, the present invention relates to a method for rapid identification of the type of fault affecting a control pattern resulting from the superposition of two line networks carried by a microelectronic component.
[0003] The manufacturing by stacking of a microelectronic component requires that the different levels constituting it be superimposed with great precision, so that the elements carried by these different levels are correctly positioned relative to each other in order to guarantee the correct functioning of the final component.
[0004] It is known to control the manufacturing precision by using a control pattern resulting from the superposition of a second network of lines produced on a second level of the component, above a first network of lines produced on a first level of the component. The first network of lines, respectively the second network of lines, is manufactured at the same time as the production of the elements of the first level, respectively of the second level.
[0005] This precision control is carried out by measuring the defects affecting the control pattern.
[0006] On this subject, the applicant filed patent application FR 22 02371, in which it was a question of measuring a covering defect, or "overlay" (abusively using the English term as a person skilled in the art would do). In this document, the overlay is defined as a translation of the second network of lines relative to the first network of lines. This translation is characterized by an angle α between first and second directions: the first direction being the direction normal to the lines of a network, in the plane of this network; and the second direction passing through the geometric centers of the cross sections of a line of the first network and a line of the second network, the pair of lines considered having to be perfectly superimposed when the angle α is equal to 90 0 .
[0007] However, the geometry of the control pattern can be altered in ways other than a simple translation shifting one array of lines relative to the other.
[0008] Thus, in patent application FR 23 01672, the measurement method makes it possible to evaluate not only a translation angle, to characterize a translation defect, but also a deformation angle of the edges of the network lines, to characterize a deformation defect.
[0009] The three-dimensional reconstruction of the shape of the lines from the measurements obtained by CD-SAXS is based on a modeling which is all the more complex as one considers a greater number of parameters for describing the potential defects. The extraction of measurements can therefore take a lot of time, in particular for the acquisition of the quantity of data necessary for the analysis in accordance with the modeling used.
[0010] The aim of the present invention is thus to propose a measurement method allowing a simplification of the extraction of information on the geometry of the control pattern.
[0011] For this purpose, the invention relates to a method for identifying a type to which a defect affecting a control pattern carried by a microelectronic component relates, the microelectronic component comprising at least a first level and a second level, the control pattern resulting from the superposition of a first network of lines, produced during the manufacture of the microelectronic component, on the first level and of a second network of lines, produced during the manufacture of the microelectronic component, on the second level, the identification method implementing a small-angle X-ray scattering technique, the method being characterized in that it comprises the steps consisting of: acquiring, by illuminating the control pattern, a plurality of intensity measurements of a transmitted or reflected X-ray beam for a plurality of angles of incidence of the X-ray beam;reconstructing, from the plurality of intensity measurements, at least two Bragg rods of order n and -n of a diffraction pattern; and, applying a plurality of tests on the diffraction pattern to determine the type of defect affecting the control pattern.;
[0012] According to particular embodiments, the identification method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:
[0013] - the fault results from a misalignment of the line networks with each other and / or from a deformation of the edges of the network lines.
[0014] - the acquisition step and the reconstruction step lead to a figure of diffraction comprising at least two Bragg rods of order n and -n and at least one Bragg rod of another order.
[0015] - the type of defect is identified among the following cases, the misalignment being characterized by a misalignment angle a and the deformation being characterized by a deformation angle pmoy: Case 1, corresponding to the ideal case in which the lines are perfectly superimposed and their edges are not deformed; Case 2, corresponding to the case where the lines are superimposed, but their edges are deformed in such a way that an average of these deformations is zero; Case 3, corresponding to the case where the lines are offset, but their edges are not deformed; Case 4, corresponding to the case where the lines are offset and their edges deformed by the same deformation angle, or the lines are offset and their edges deformed by the same average deformation angle; Case 5, corresponding to the case where the lines are offset and their edges deformed but according to different deformation angles; and Case 6, corresponding to all other defects.
[0016] - an orthonormal xyz reference frame being associated with the microelectronic component, the lines first and second line arrays being oriented along the y direction, and the first and second line arrays being superimposed along the z direction, along axes qx and qz, associated respectively with the x and z directions: a first test consists of verifying whether a position along the qz axis of at least one maximum of the Bragg rod of order -n is identical to that of the maximums of the Bragg rod of order n; a second test consists of verifying whether the Bragg rod of order n, or the Bragg rod of order -n, is symmetrical with respect to the qx axis;
[0017] - a third test consists of checking whether the Bragg rods of order -n and n are symmetrical with respect to an origin of the diffraction pattern with coordinates qz = 0 and qx = 0; and a fourth test consists of checking whether the position along the qz axis of an extremum evolves as a function of the order of the Bragg rod.
[0018] The invention also relates to a method for controlling the manufacture of a microelectronic component, the microelectronic component comprising at least a first level and a second level, as well as a control pattern resulting from the superposition of a first network of lines produced during manufacture on the first level and a second network of lines produced during manufacture on the second level, the method comprising the steps consisting of: identifying a type to which a defect affecting the control pattern relates by implementing the preceding identification method; and, adjusting manufacturing conditions of the microelectronic component according to the type of defect identified.
[0019] According to particular embodiments, the control method comprises one or more of the following characteristics, taken in isolation or in all technically possible combinations:
[0020] - optimal manufacturing conditions are gradually identified by a trial and error approach by iterating the identification and adjustment steps for a succession of microelectronic components.
[0021] - following identification of the type of fault, the control method includes a step of measuring a value of at least one characteristic parameter of the defect, taking into account the type of said defect in the choice of an instrumental method and / or an analysis method, the step of adjusting the manufacturing conditions also taking into account the measured value of each characteristic parameter.
[0022] The invention also relates to an instrumental system of the T-SAXS or GLSAXS type, comprising a detector, acquisition electronics and a computer, characterized in that the computer is suitably programmed so that said instrumental system implements the preceding method.
[0023] The invention also relates to a computer program product comprising software instructions which, when executed by the computer of the preceding instrumental system, allow the implementation of the preceding method.
[0024] The invention and its advantages will be better understood on reading the detailed description which follows of a particular embodiment, given solely by way of non-limiting example, this description being made with reference to the appended drawings in which:
[0025] [Fig-1] [Fig.l] is a schematic representation of an instrumental system identification of defects affecting the geometry of the control pattern by implementing a T-SAXS technique;
[0026] [Fig.2] [Fig.2] is a representation of the Fourier transform of a control pattern for which the geometry of the control pattern is defect-free;
[0027] [Fig.3] [Fig.3] is a schematic representation of the geometry of the control pattern associating deformation of the lines of the two superimposed networks and translation of the two superimposed networks;
[0028] [Fig.4] [Fig.4] is a representation in the form of blocks of an embodiment of the identification method according to the invention;
[0029] [Fig.5] [Fig.5] is a graph representing the intensity as a function of the qz coordinate for four Bragg rods of a diffraction pattern; and,
[0030] [Fig.6] [Fig.6] is a representation in block form of an embodiment of a measurement method integrating the identification method according to the invention.
[0031] [Fig.l] represents an instrumental system 1 adapted to a measurement by T-SAXS transmission.
[0032] An X-ray source S emits an X-ray beam in a direction z0, perpendicular to an observation plane P. The incident beam falls at an origin point O of the observation plane P. Directions x0 and y0 define an orthonormal reference frame of the plane P.
[0033] A microelectronic component C is interposed between the source S and the observation plane P.
[0034] The component C is provided with a control pattern Z making it possible to evaluate the manufacturing quality of the component C. The control pattern Z consists of a first level comprising a first network of lines and, superimposed on the first level, a second level comprising a second network of lines.
[0035] Preferably, the surface of the control pattern Z is flush with the rest of the surface of the component C.
[0036] The center A of the control pattern Z is placed on the axis of incidence of the X-ray beam.
[0037] An orthonormal xyz coordinate system is associated with component C. This coordinate system is attached to the center A of the control pattern Z so that the y direction corresponds to the y0 direction, and the z direction is normal to the surface of the control pattern Z.
[0038] As specified below, the lines of the line networks which make it possible to measure the overlay are arranged parallel to the y0 axis, so as to obtain Bragg peaks along the x0 axis.
[0039] The component C is placed on a support 30 allowing the component C to be rotated around the direction y so as to modify the angle of incidence of the X-ray beam on the control pattern Z, i.e. the angle between the direction z and the direction z0, which is also the angle between the direction x and the direction x0.
[0040] A detector 10 is placed in the observation plane P. It is for example composed of a strip of sensors arranged in the direction x0.
[0041] The intensity at point B of the x0 axis, measured by the sensor located at point B, depends on the angle 2d between the direction AO and the direction AB. This intensity is noted
[0042] As known per se, the image in the observation plane P is related to the spatial Fourier transform of the control pattern Z illuminated by the incident beam.
[0043] In reciprocal space, the coordinate associated with the x0 direction is the spatial frequency #0 defined by:
[0044] = 4sin(g) (1) ^0“ 2
[0045] where X is the wavelength of the X-rays used.
[0046] Let, still in the reciprocal space, but considering the spatial frequencies and Qz, respectively associated with the directions x and z of the reference frame linked to the control pattern Z: 100471 qx--q^r 100481
[0049] The detector 10 is connected to an electronic device, represented schematically in [Fig.l] by a cube bearing the reference 20.
[0050] The device 20 comprises control electronics making it possible to control the support 30 so that it positions the component C according to a set value of the angle of incidence V.
[0051] The device 20 also comprises acquisition electronics making it possible to carry out suitable pre-processing on the signals delivered by each of the sensors of the detector 10 and to digitize them.
[0052] The device 10 further comprises a computer for processing the pre-processed and digitized signals. The computer is a computer comprising calculation means, such as a processor, and storage means, such as a memory. The memory stores in particular the instructions of computer programs, in particular a program whose execution allows the implementation of a method for identifying defects affecting the control pattern.
[0053] From a theoretical point of view, it is possible to calculate exactly the spatial Fourier transform of the control pattern.
[0054] Thus, part A of [Fig.2] represents, in the direct space of the x and z coordinates (the xz plane being the plane transverse to the first and second line networks), a fraction of the Z control pattern carried by the component C. Here are represented the transverse sections of two lines 41 and 42 of the first line network 40 and two lines 51 and 52 of the second line network 50. The lines of each of these networks extend in the y direction, that is to say perpendicular to the plane of [Fig.2].
[0055] In the case of a defect-free Z control pattern (shown in FIG. 2A), i.e. when the Z control pattern complies with an ideal geometry, a line of a network has a rectangular cross-section, the side walls (or edges) of a line being perpendicular to the bottom of this line.
[0056] A line has a width of 1 and a depth of p.
[0057] Two lines of the same network are spaced by a step d.
[0058] The pitch between a line of the first grating and the corresponding line of the second grating is noted D. This pitch according to the thickness of the component is considered to be equal to p in the remainder of this description (the two gratings being superimposed directly on top of each other), because this pitch according to the thickness has no measurable effects on the diffraction pattern.
[0059] Still according to the ideal geometry, the first and second networks of lines are perfectly superimposed.
[0060] The exact calculation of the spatial Fourier transform of this pattern leads to the diffraction pattern (or diffraction map) shown in part B of figure 2. It is a representation in the reciprocal space of the spatial frequencies and which are respectively conjugate of the x and z coordinates.
[0061] For a zero value of the Fourier transform, there is, along the Qx direction, a succession of principal maxima, or Bragg peak. Each Bragg peak is identified by an integer n, called the Bragg order.
[0062] For a given value, that is to say for the Bragg peak of order n, the intensity along the direction forms what is called the Bragg rod of order n. A Bragg rod presents an alternation of local maxima and minima.
[0063] We can note q"'1 the position of the principal maximum of the Bragg rod of order n. q*° is equal to 0 for an ideal geometry control pattern.
[0064] For i positive integer, the ith secondary maximum of the upper part of the Bragg rod of order n is located at q^ (q1* positive) and the ith secondary maximum of the lower part of the Bragg rod of order n is located at q'^' (q^1 negative).
[0065] In part B of figure 2, the Bragg rod of order n is symmetrical with respect to the ^z axis. Thus, the position of the ith secondary maximum of the upper part of the Bragg rod of order n is equal (to the sign) to the position of the ith secondary maximum of the lower part of the Bragg rod of order n: q"'1 = -q"!.
[0066] Furthermore, it is noted that, when there is no overlay, the position along the ^z axis of the ith secondary maximum of a Bragg rod is constant, whatever the Bragg order n considered (q^ = q^.
[0067] As illustrated in [Fig.3], two types of defects can potentially affect the geometry of the Z control pattern such that it deviates from the ideal geometry of [Fig.2]:
[0068] - a translation defect: a translation of one network of lines relative to another network of lines, described by a translation angle a; and
[0069] - a deformation defect: a deformation of the edges of the lines of the two networks, described by a deformation angle
[0070] More precisely, the control pattern Z may have a defect resulting from the translation of one network of lines relative to the other in a direction normal to the lines of a network, in the plane of this network (direction x).
[0071] For example, in [Fig.3], the upper line 51 is translated in the x direction by a distance e relative to the lower line 41.
[0072] This translation defect is described by the angle a, or translation angle, between the normal direction x and the direction connecting the geometric centers Gi and G2 of a pair of lines.
[0073] The control pattern may also have a defect resulting from the deformation of the lines of the line networks, each line then having a diamond-shaped section (in other words, the side walls of a line are no longer at right angles to the bottom of this line).
[0074] For example, in [Fig.3], the edges between corners 44 and 46, on the one hand, and 43 and 47, on the other hand, of the lower line 41 are inclined, and the edges between the corners 54 and 56, on the one hand, and 53 and 57, on the other hand, of the upper line 51 are inclined.
[0075] To quantify this deformation, a first edge angle is defined as the angle between the x direction and the direction joining the lower right corner 43 of the first line 41 and the upper right corner 57 of the second line 51.
[0076] A second edge angle is defined as the angle between the x direction and the direction joining the lower left corner 44 of the first line 41 and the upper left corner 56 of the second line 51.
[0077] The deformation angle characterizing this deformation defect is then defined as the average of the tangents of the first and second edge angles: tan(PmJ = + tan^J
[0078] In the case where the control pattern is affected by a translation defect but not by a deformation defect, it is shown that = a, so that these two parameters then make it possible to extract the same information on the geometry of the pattern.
[0079] On the other hand, when the control pattern is affected by a translation defect and by a deformation defect, the equality — a is lost, and the parameters and a then make it possible to extract different information on the deviation from the ideal geometry.
[0080] The method according to the invention makes it possible to quickly identify the nature of the defect affecting the control pattern.
[0081] By automatically analyzing the existing symmetries (axial or central) on a diffraction figure, advantageously partial, brought back into space (qx, qz), it is possible to classify the profile of the control pattern having led to this diffraction figure.
[0082] More specifically, in the embodiment presented here in detail, the method 100 makes it possible to classify a defect according to six possible cases or classes. Each of these cases is represented schematically at the bottom of [Fig.4].
[0083] The different possible cases are:
[0084] Case 1: it corresponds to the ideal case where the pattern has no defects, the lines being perfectly superimposed and their edges not being deformed: a=pl=p2=pmoy = 90°;
[0085] Case 2: it corresponds to a deformation defect with compensation, the lines being superimposed, but deformed so that the average of the deformations is zero: a = 90°; 01= -02* 90°; 0moy = 9O°;
[0086] Case 3: it corresponds to a translation defect, the lines being shifted, but their edges not being deformed: a * 90J; 01= 02 = 0moy = 90 °;
[0087] Case 4: it corresponds to an equal translation defect and a deformation defect, the lines being shifted and their edges deformed by the same angle: a = 01 = 02 = 0moy * 90 ° or the lines are shifted and their edges deformed by the same average angle: a = 0moy * 90 u but 01 * 02;
[0088] Case 5: it corresponds to a translation defect and any deformation defect, the lines being shifted and their edges deformed, but according to different angles a * 90 °; a * 0moy; and 01 = 02 = 0nioy; and,
[0089] Case 6: it corresponds to a complex defect. This class includes all the other combinations leading to more complicated deformations of the pattern.
[0090] [Fig.4] is a block representation of the method 100.
[0091] In step 105, the measuring device of [Fig.l] is implemented to acquire a diffraction pattern of the control zone of a tested component. The diffraction pattern advantageously comprises only two Bragg rods. These two Bragg rods must be symmetrical to each other with respect to the qz axis. This is therefore the pair of Bragg rods of order -n and order n.
[0092] In step 110, a first test Q1 is carried out on the diffraction pattern acquired in step 105. This first test consists of verifying whether the positions along the qz axis of the maxima of the Bragg rod of order -n are identical to the positions along the qz axis of the maxima of the Bragg rod of order n. The maxima considered here comprise at least the main maximum ç”'°and (j^ of the Bragg rods of order n and -n and a secondary maximum ^'wet of the Bragg rods of order n and -n. If so, the method proceeds to step 120 and if not, the method proceeds to step 130.
[0093] In step 120, a second test Q2 is performed on the diffraction pattern acquired in step 105. This second test consists of verifying whether the nth-order Bragg rod (or the -nth-order Bragg rod) is symmetrical with respect to the qx axis (i.e., symmetrical with respect to the axis passing through qz = 0). For example, is the position along the qz axis of the ith maximum of the upper part of the nth-order Bragg rod identical to the position along the qz axis of the ith maximum of the lower part of the nth-order Bragg rod? Is q"1 = q™? If so, the method proceeds to step 125, and if not, the method is stopped and the defect is considered to correspond to case 3.
[0094] In step 125, the measuring device of [Fig.l] is activated to complete the diffraction pattern obtained in step 105 in order to obtain a third Bragg rod, for example the stem of order k. Step 125 is implemented only if, at step 105, this third stem has not already been acquired.
[0095] Then, in step 140, a fourth test Q4 is performed on the completed diffraction pattern. This fourth test consists of verifying whether the position along the qz axis of the ith extremum (minimum or maximum) evolves as a function of the order. For example, are q and q^1 in linear progression as a function of the order? If so, the method 100 stops and the pattern is considered to correspond to case 2. If not, the method stops and the defect is considered to correspond to case 1.
[0096] In step 130, a third test Q3 is applied to the diffraction pattern obtained in step 105 (which has at least the two Bragg rods of order -n and n). This third test consists of checking whether the Bragg rods of order -n and n are symmetrical with respect to the origin of the diffraction pattern, that is to say with respect to the point with coordinates qz = 0 and qx = 0. Do we have: q^ = q;^1? If so, the method goes to step 150. If not, the method stops and the defect corresponds to case 6.
[0097] In step 150, a fifth test Q5 is applied to the diffraction pattern. This fifth test consists of verifying whether the Bragg rod of order n is symmetrical with respect to a central point of this rod, but which has a non-zero coordinate along the qz axis. Do we have: qÿ - q'*'° = q^1 - q^? If so, the process stops and the defect corresponds to case 4. If not, the process stops and the defect is considered to fall within case 5.
[0098] It should be noted that the numeral adjective associated with each of the tests of the method 100 does not indicate the order in which this test is carried out. For example, the third test Q3 is applied only if the response to the first test Q1 is negative. This is then the second test which is actually applied to the diffraction pattern analyzed. In this alternative, the second and fourth tests, Q2 and Q4, are not applied. Depending on the response to this third test Q3, the fifth test Q5 will or will not be applied.
[0099] [Fig.5] represents the intensity profile of four Bragg rods recorded on a diffraction pattern obtained at the output of the device of [Fig.l]. Curve C 21 corresponds to the Bragg rod of order -2, curve C( 11 corresponds to the rod of order -1; curve C(1) corresponds to the rod of order 1 and curve C(2) corresponds to the rod of order 2.
[0100] It should be noted that, in [Fig.5], the profiles have been slightly shifted along the intensity direction to aid visualization. Profiles C 21 and C(2) on the one hand and profiles C( 11 and C(1) on the other hand would overlap perfectly without this slight shift for visualization.
[0101] Step 105 allows for example the acquisition of curves C(2) and C(2).
[0102] The application of the first test Q1 of step 110 shows that the positions of the maxima are identical for a pair of rods of opposite order: =
[0103] The application of the second test Q2 of step 120 shows that there is a symmetry along qz
[0104] Step 125 makes it possible to complete the diffraction figure with, for example, the acquisition of the curves C(0 and C(1).
[0105] The application of the fourth test Q4 of step 130 shows that the position of the maxima moves with the orders: q^"1
[0106] We therefore find ourselves with a defect falling under case 2.
[0107] As illustrated by the diagram in [Fig.6], the identification method 100 is advantageously integrated into a control method 200.
[0108] The method 200 is implemented along a mass production line of a microelectronic component. This is a multi-level component comprising at least a first level and a second level.
[0109] To control manufacturing accuracy, a control pattern is engraved on each component. This results from the superposition of a first network of lines produced during the manufacturing of the component on the first level and a second network of lines produced during manufacturing on the second level.
[0110] The method 200 is carried out on a first component.
[0111] It first consists, in a step 210, of identifying the type of fault possibly affecting the control pattern carried by the first component. To do this, the identification method 100 is implemented. This makes it possible to determine the case to which the fault relates.
[0112] Then, in a step 220, the manufacturing conditions are adjusted according to the result of step 210, i.e. the type of defect identified.
[0113] This adjustment of the manufacturing conditions may consist simply of modifying the operating parameters of the different devices along the manufacturing chain and of seeking to determine optimal manufacturing conditions progressively by a trial / error approach by iterating steps 210 and 220 on a succession of microelectronic components leaving the production chain.
[0114] Alternatively, step 210 of identifying the type of defect is followed by a step 215 of measuring a value of at least one parameter characteristic of the defect affecting the pattern of the microelectronic component being checked.
[0115] This measurement takes into account the result of step 210 in the choice of the most suitable instrumental method for acquiring measurement data with a view to extracting this characteristic parameter (if the data already acquired in steps 105 and 125 are not sufficient) and / or in the analysis method using suitable modeling to extract the value of the characteristic parameter considered from the measurement data.
[0116] The method of precise analysis of the angle a in the framework of the overlay of cases 2 and 3; a method of precise extraction of the angle [3^ in the framework of the overlay of cases 4 and 5; and a method of precise extraction of the three-dimensional depth profile of the lines for case 1 for example.
[0117] The instrumental method chosen is preferably a CD-SAXS technique, but other techniques can be used to carry out the acquisition of measurement data.
[0118] The choice of the instrumental method also covers the definition of the part of the diffraction pattern that it is necessary and sufficient to acquire to obtain the measurement data allowing a precise characterization of the defect.
[0119] The part of the diffraction pattern to be acquired can, in turn, impact the instrumental device to be used for the precise measurement.
[0120] For example, if the identification step 210 shows that the presence of overlay is characterized by an asymmetry along the qx axis, the measurement of a single Bragg rod is sufficient. Then, a point detector may be sufficient for a T-SAXS measurement. To detect the presence of the angle, the measurement of several Bragg orders is necessary. Then a one-dimensional detector is used for a T-SAXS measurement.
[0121] Finally, the adjustment step 220 takes into account the measured value of each characteristic parameter to adapt the manufacturing conditions and progressively lead to the manufacturing of a component whose control pattern does not contain any defects.
[0122] Those skilled in the art will note that the present method allows the identification of the nature of a defect affecting the control pattern in order to then adapt the method of analysis and extraction of the values of the parameters characterizing this defect.
[0123] If, in the embodiment presented above, an instrumental method by T-SAXS transmission was used, alternatively, an instrumental method by GLSAS reflection can be implemented.
[0124] Since CD-SAXS is a non-trivial characterization method, the method according to the invention makes it possible to direct towards the appropriate analysis method of the diffraction pattern. It is independent of the modeling. Once the defect has been classified, the adapted modeling (i.e. simplified compared to a complex modeling integrating all the characteristic parameters of all the potential defects) is applied to measurement data to extract precise information.
[0125] With the present method, the CD-SAXS control technique becomes particularly advantageous along a manufacturing line to, after each measurement of the overlay, realign the manufacturing equipment during the lithography process itself. This makes it possible to correct the manufacturing accuracy during the operation of the manufacturing line.
Claims
1. Claims A method (100) for identifying a type to which a defect affecting a control pattern (Z) carried by a microelectronic component (C) relates, the microelectronic component comprising at least a first level and a second level, the control pattern resulting from the superposition of a first network of lines, produced during the manufacture of the microelectronic component, on the first level and of a second network of lines, produced during the manufacture of the microelectronic component, on the second level, the defect resulting from a misalignment of the networks of lines with each other and from a deformation of the edges of the lines of the networks, the identification method implementing a small angle X-ray scattering technique, the method being characterized in that it comprises the steps consisting of: - acquiring, by illuminating the control pattern, a plurality of intensity measurements of a transmitted or reflected X-ray beam for a plurality of angles of incidence of the X-ray beam; - reconstructing, from the plurality of intensity measurements, at least two Bragg rods of order n and -n and at least one Bragg rod of another order of a diffraction pattern; and, - applying a plurality of tests to the diffraction pattern to determine the type of defect affecting the control pattern, an orthonormal xyz reference frame being associated with the microelectronic component, the lines of the first and second line networks being oriented in the y direction, and the first and second line networks being superimposed in the z direction, along axes qx and qz, associated respectively with the x and z directions: - a first test of the plurality of tests consists of verifying whether a position along the qz axis of at least one maximum of the Bragg rod of order -n is identical to that of the maxima of the Bragg rod of order n; - a second test of the plurality of tests consists of checking whether the Bragg rod of order n, or the Bragg rod of order -n, is symmetrical with respect to the qx axis; - a third test of the plurality of tests consists of verifying whether the Bragg rods of order -n and n are symmetrical with respect to an origin of the diffraction pattern of coordinates qz = 0 and qx = 0; and, - a fourth test of the plurality of tests consists of verifying whether the position along the qz axis of an extremum evolves as a function of the order of the Bragg rod.
2. The identification method according to claim 1, wherein the type of the defect is identified from among the following cases, the misalignment being characterized by a misalignment angle and the deformation being characterized by a deformation angle 0moy: - Case 1, corresponding to the ideal case in which the lines are perfectly superimposed and their edges are not deformed; - Case 2, corresponding to the case where the lines are superimposed, but their edges are deformed such that an average of these deformations is zero; - Case 3, corresponding to the case where the lines are offset, but their edges are not deformed; - Case 4, corresponding to the case where the lines are offset and their edges deformed by the same deformation angle, or the lines are offset and their edges deformed by the same average deformation angle; - Case 5, corresponding to the case where the lines are offset and their edges deformed but according to different deformation angles;and, - Case 6, corresponding to all other defects.;
3. A method of controlling (200) the manufacture of a microelectronic component, the microelectronic component comprising at least a first level and a second level, as well as a control pattern resulting from the superposition of a first network of lines produced during manufacture on the first level and a second network of lines produced during manufacture on the second level, the method comprising the steps of: - identifying (210) a type to which a defect affecting the control pattern relates by implementing an identification method (100) in accordance with any one of the preceding claims; and, - adjusting (220) manufacturing conditions of the microelectronic component as a function of the type of defect identified.
4. A control method according to claim 3, wherein optimal manufacturing conditions are progressively identified by a trial / error approach by iterating the identification and adjustment steps for a succession of microelectronic components.
5. Control method according to claim 3, in which, following the identification of the type of defect, the control method comprises a step of measuring (215) a value of at least one characteristic parameter of the defect taking into account the type of said defect in the choice of an instrumental method and / or an analysis method, the step of adjusting the manufacturing conditions also taking into account the measured value of each characteristic parameter.
6. Instrumental system (1) of the type by scattering of X-rays at small angles at grazing incidence - (GISAXS), comprising a detector, acquisition electronics and a computer, characterized in that the computer is suitably programmed so that said instrumental system implements an identification method according to any one of claims 1 to 2 to identify a type to which a defect affecting a control pattern carried by a microelectronic component analyzed with the instrumental system relates.
7. Computer program product comprising software instructions which, when executed by a computer of an instrumental system according to claim 9, allow the latter to implement an identification method according to any one of claims 1 to 2.